Paper

MICRO-NANO
MECHATRONICS
NEW TRENDS IN MATERIAL,
MEASUREMENT, CONTROL,
MANUFACTURING AND
THEIR APPLICATIONS IN
BIOMEDICAL ENGINEERING
Editedby Toshio Fukuda,
Tomohide Niimi and Goro Obinata
Micro-Nano Mechatronics New Trends in Material, Measurement, Control,

Manufacturing and Their Applications in Biomedical Engineering
http://dx.doi.org/10.5772/55984
Edited by Toshio Fukuda, Tomohide Niimi and Goro Obinata
Contributors
Toshio Fukuda, Masahiro Nakajima, Masaru Kojima, Goro Obinata, Hitoshi Hirata,
Chikara Nagai, Shigeru Kurimoto, Shuichi Kato, Tomonori Nakano, Tomohide Niimi,
Eiji Shamoto, Norikazu Suzuki, Takashi Kato, Burak Sencer, Tomohiro Kawahara,
Fumihito Arai, Osamu Takai, Maria Antoaneta Bratescu, Tomonaga Ueno,
Nagahiro Saito, Minoru Ueda, Nobutada Ohno, Dai Okumura, Yusuke Kinoshita,
Kenji Fukuzawa, Ken-ichi Isobe, Naomi Nishio, Thanasegaran Suganya, Zhao Cheng,
Sachiko Ito, Noritsugu Umehara, Takayuki Tokoroyama, Hiroyuki Kousaka, Yang Ju,
Akihiro Sasoh, Kensuke Kuroda, Masazumi Okido, Masaru Takeuchi, Gauvain Haulot,
Chih-Ming Ho, Daniel T. Kamei, Hideharu Hibi, Mitsuhiro Shikida and Yajing Shen
Published by InTech
Janeza Trdine 9, 51000 Rijeka, Croatia
Copyright 2013 InTech
All chapters are Open Access distributed under the Creative Commons Attribution 3.0
license, which allows users to download, copy and build upon published articles even
for commercial purposes, as long as the author and publisher are properly credited,
which ensures maximum dissemination and a wider impact of our publications. After
this work has been published by InTech, authors have the right to republish it, in
whole or part, in any publication of which they are the author, and to make other
personal use of the work. Any republication, referencing or personal use of the work
must explicitly identify the original source.
Notice
Statements and opinions expressed in the chapters are these of the individual contributors
and not necessarily those of the editors or publisher. No responsibility is accepted for the
accuracy of information contained in the published chapters. The publisher assumes no
responsibility for any damage or injury to persons or property arising out of the use of any
materials, instructions, methods or ideas contained in the book.
Publishing Process Manager Danijela Duric
Typesetting www.pantype.com
Cover InTech Design Team
First published March, 2013
Printed in Croatia
A free online edition of this book is available at www.intechopen.com
Additional hard copies can be obtained from orders@intechopen.com
Micro-Nano Mechatronics New Trends in Material, Measurement, Control,
Manufacturing and Their Applications in Biomedical Engineering,
Edited by Toshio Fukuda, Tomohide Niimi and Goro Obinata
p. cm.
ISBN 978-953-51-1104-7
Contents
Preface
IX
Chapter 1
Research and Technology on Micro-Nano Mechatronics 1

Toshio Fukuda, Masahiro Nakajima and Masaru Kojima
Chapter 2
Neural Interfaces: Bilateral Communication

Between Peripheral Nerves and Electrical Control Devices 13
Goro Obinata, Hitoshi Hirata, Chikara Nagai,
Shigeru Kurimoto, Shuichi Kato and Tomonori Nakano
Chapter 3
High Knudsen Number Flow

Optical Diagnostic Techniques 33
Tomohide Niimi
Chapter 4
Precision Micro Machining

Methods and Mechanical Devices 49
Eiji Shamoto, Norikazu Suzuki, Takashi Kato and Burak Sencer
Chapter 5
Micro-Nano Robotics and

Mechatronics for Biomedical Applications 77
Tomohiro Kawahara and Fumihito Arai
Chapter 6
Synthesis of Nanomaterials
by Solution Plasma Processing 109
Osamu Takai, Maria Antoaneta Bratescu,
Tomonaga Ueno and Nagahiro Saito
Chapter 7
Tissue Engineering and Regenerative Medicine 123

Minoru Ueda
Chapter 8
Electronic Structure Calculations for Nano Materials 167

Nobutada Ohno, Dai Okumura and Yusuke Kinoshita
VI
Contents
Chapter 9
Measurement of Frictional
Properties on the Micro/Nanometer Scale 189
Kenji Fukuzawa
Chapter 10
Tissue Damage and Repair Caused by Immune System

and Personalized Therapy of Failed Organs by Stem Cells 207
Ken-ichi Isobe, Naomi Nishio,
Thanasegaran Suganya, Zhao Cheng and Sachiko Ito
Chapter 11
Tribology for Biological and Medical Applications 221

Noritsugu Umehara,
Takayuki Tokoroyama and Hiroyuki Kousaka
Chapter 12
Micro-Nano Materials Characterization and Inspection 241

Yang Ju
Chapter 13
Aerospace Application 271

Akihiro Sasoh
Chapter 14
Hydroxyapatite Coating on
Titanium Implants Using Hydroprocessing
and Evaluation of Their Osteoconductivity 287
Kensuke Kuroda and Masazumi Okido
Chapter 15
System Integration of
a Novel Cell Interrogation Platform 299
Masaru Takeuchi, Gauvain Haulot and Chih-Ming Ho
Chapter 16
Transferrin-Toxin Conjugates for Cancer 315

Daniel T. Kamei
Chapter 17
Tissue Engineering and

Regenerative Medicine for Bone Regeneration 321
Hideharu Hibi and Minoru Ueda
Chapter 18
MEMS Sensors and Their Applications 331

Mitsuhiro Shikida
Chapter 19
Single Cell Nanosurgery System 353

Toshio Fukuda, Masahiro Nakajima,
Yajing Shen and Masaru Kojima
Preface
Micro/Nano mechatronics is currently used in broader spectra, ranging from basic
applications in robotics, actuators, sensors, semiconductors, automobiles, and machine
tools. As a strategic technology highlighting the 21st century, this technology is
extended to new applications in bio-medical systems and life science, construction
machines, and aerospaceequipment, welfare/human life engineering, and other brandnew scopes. Basically, the miniaturizing technology is important to realize high
performance, low energy consumption, low cost performance, small space
instrumentation, light-weight, and so on.
In this book, the states of art of research progress are summarized through our project
COE for Education and Research of Micro-Nano Mechatronics and the R&D in
Center For Micro-Nano Mechatronics at Nagoya University. Our project strives to
foster young researchers who dare to challenge unexploited fields by building a
novel interdisciplinary field based on micro-nano mechatronics. This field is important
to promote the world-highest-level of micro-nano mechatronics research with an
emphasis on originality from a viewpoint of not only the acquisition of advanced
technology, but also social issues.
Our project implements a strategy to realize applications of micro-nano mechatronics,
which are based on mechanical engineering or materials science, control systems
engineering, and advanced medical engineering. As shown in Figure 1, the proposed
research teams include Nanocontrol engineering, Nano measurement engineering,
Nano design and manufacturing, and Nano materials science.
By establishing joint research and international collaborations between the above
research teams, we have created the most advanced micro-nano mechatronics. We
have also trained the researchers who can comprehend industrial circles and social
issues using an open cluster system as well as conduct research to solve problems
spanning these four basic fields. In particular, we initially focus on tasks in the bio- or
medical welfare technologies using a number of unexploited fields, which may
consequently produce venture enterprises.
Research fields and two-dimensional matrix structure

Treatment techniques based on live
organ transplant of cultured cell and
tissue
Mechatronic devices for regeneration &

induction of biomedical tissue
Advancedresearchfieldsof
bio/medicaltechnology

Mechatronics for cell
control
and biology

Mechatronics

Mechatronics

medical
operation

rehabilitation
for
medicine and
for
well-being and
SystemIntegrationwithMicroNanoMechatronics
Nano

control

engineering

Nano

measurement

engineering
Nano design

and

manufacturing
materials
Nano
science

TechnologyofMicroNanoMechatronics
Figure 1. Innovations by micro-nano mechatronics.
We acknowledge the excellent contributions of all people to contribute the chapters for
this book. We express our sincere appreciation for the publication of this book
supported by Nagoya University, the 21st COE program "Micro- and NanoMechatronics for Information-Based Society", and the global COE program COE for
Education and Research of Micro-Nano Mechatronics. This book would not have
been possible without these generous supports.
Nagoya, December 2012

Toshio Fukuda
Graduate School of Engineering, Nagoya University
Japan
Chapter 1
Research and Technology on Micro-Nano Mechatronics
Toshio Fukuda, Masahiro Nakajima and

Masaru Kojima
1.Introduction
Inourdailylife,variousdevicesareappliedforautomobiles,computerperipheries,printers,
cameras, amusements, robotics, automation, environmental monitoring, energy resource,
biological/medicaltreatments,andsoonMicrotechnologywascommonlyusedtorealize
highefficiency, highintegration, highfunctionality, lowenergy consumption, lowcost,
miniature, and so on By miniaturizing the elemental devices on sensors, actuators, and
computersinmicroscale,Micromechatronicscameupastheoneoftheimportanttechnol
ogyRecently,Nanotechnologyhasanimportantroleintheindustrialapplicationsasan
advancedfieldofmechatronicsnamedasNanomechatronicsThemicronanomechatronics
isbasicallydefinedtointegratemajorthreetechnologiesController,SensorandActua
torbasedontheelectronicsandmechanicalengineeringasdepictedinFigure1.
Figure 2 shows the demands of micronano mechatronics for various social and industrial
applicationsForvariousapplicationsforindustry,sometechniquesareimportant,especially
micro/nano fabrication, assembly, control, material, and evaluation techniques Micronano
mechatronics is based on various technologies, especially life science, medicine, sensing/
actuating, material science, energy/power, and design/control From social aspects, human
resource, environmental issue, saving energy, safety/security, medical/health, and aging
population are currently demanded From industrial aspects, service robots, dependable
products, tailormade products, alternative energy, technocare service, environmental
friendlyproducts,areparticularlydemandedThemicronanomechatronicsisakeytechnol
ogytosolvethoseproblems/issuesandleadingconventionaltechnologiesforfuture.
TheapplicationsofmicronanomechatronicsaremainlycategorizedintotheMechanical,
Electrical, and Biological/Medical applications The key point for the categorization is
inorganic (wet) and organic (dry) Mechanical applications are relatively based on the
inorganic materials or technologies, such as lithography technique On the other hand,
Biological/Medicalapplications,theorganicmaterialsortechnologiesareused,suchasself
2013 Fukuda et al.; licensee InTech. This is an open access article distributed under the terms of the
Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits
unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Micro-Nano Mechatronics New Trends in Material, Measurement, Control, Manufacturing and Their Applications in
Biomedical Engineering
Figure 1. Micro-Nano mechatronics
Figure 2. Micro-Nano mechatronics for social and industrial demands
assemblytechniqueInbetweenthem,Electricalapplicationsareplacedfordeliveringor
calculating information and so on Since the micronano mechatronics is the composite
academic fields, the required technologies are mainly categorized in to basic/middle/high
integrationlevelsasdepictedinFigure3.
High Integration Level
Basic Level
MicroNano
Material
Technology
MicroNano
Actuation
Technology
MicroNano
Computing/
Information
Technology
MicroNano
Assembly
Technology
MicroNano
Energy
Generation/
Storage
Technology
MicroNano
Fabrication
Technology
MicroNano
Sensing
Technology
Integration
Figure 3. Required technologies for micro-nano mechatronics
Basically,nanotechnologyisplacedinthecombinationsofthetopdownandbottomup
approachesThepossibilitytocontrolthestructureofmatteratombyatomwasfirstdiscussed
byRichardFeynmanin1959seriously[1]Oneoftheapproachestofillthegapbetweentop
down and bottomup approaches is Nanomanipulation, which realizes controlling the
positionatthemicro/nanometerscale,isconsideredtobeoneofthepromisingwaysItmightbe
a key technology to lead the appearance of replicationbased assemblers The topdown
fabricationprocess,ormicromachining,providesnumbersofnanometerstructuresatonceOn
theotherhand,thebottomupfabricationprocess,orchemicalsynthesissuchasselfassembly
[2],alsoprovidesnumerousnanometerstructuresInfact,bothapproachesreachnanometer
scalewiththelimitationsofphysical/chemicalaspectsatpresentHence,thetechnologytofillits
gapisconsideredtobeoneoftheimportantatthismomentformicronanomechatronics
Especially,currentresearchdirectionsaremainlytwoflows,greeninnovationandlife
innovationasdepictedinFigure4Theseinnovationswillbeachievedinvariousresearchand
developmentsTable1and2showthechallengingissuesbycategorizedfields.
2.Micromechatronicsforindustrialandresearchapplications
In micro scale, the important technologies are Micro Machine, Micro Mechatronics, Micro
Fabrication and Assembly for micromechatronics Recently, borderless applications are
investigatedsuchasMicroBiology,WetMechatronics,MicroTotalAnalysisSystem,Micro
MedicalEngineering,andRegenerativeMedicalEngineeringSomeexamplesofmicrodevices
mainlyinresearchfieldaremicroactuator[3],microinkjethead[4],microforcesensor[5],
microtactilesensor[6],microfuelbattery[7],microfluidicsdevice[8],bloodvesselsimulator
[9],andsoon.
Figure 4. Green and life innovations based on micro-nano mechatronics
Green innovation
Technological Challenges
Natural Resources
Micro/nano devices for Discovery of oil resource, Management of water

resources, Prevention of forest destruction, etc
Environmental Pollution
Monitor, control and management of Environment, Distributed sensing &

control, Pollution control, Green vehicle, etc
Energy Development
Energy saving, harvesting and alternatives , Energy grid and

management, Power control and green electronics, etc
Food and Agriculture
Safety, testing and tracing, Efficient harvesting, nutritious products,

geneticallymodified products, etc
Table 1. Challenges for green innovation by micro-nano mechatronics
Life innovation
TechnologicalChallenges
Medicine for life
Inspection and diagnosis , Regenerative medicine, Gene therapy and life

science, monitoring diseases, Neuro Science, Insitu diagnostics, Cell
diagnosis and surgery, New drug and medicine, DDS, Minimally invasive
surgery, Rehabilitation, Technocare, Wearable robots, Cyborg, QoL, etc
BiologyAnalysis and Synthesis
Sensing , manipulation and automation, New species, DNA diagnosis &

manipulation, Cell screening, transport, cultivation, and function and
differentiation control, Artificial cell, Life in chip, Cloning of stem cells, etc
Table 2. Challenges for life innovation by micro-nano mechatronics
Figure 5. Blood vessel simulator and surgical operation system
In medical applications, surgical supporting or simulating technologies are important We

developedthePatientbloodvesselsimulatorEVE(EndoVascularEducator)aschallenging
thefrontierofthesurgicalsimulationsince1989Thissimulatorisfabricatedandassembled
basedonthemicrofabricationtechnologiesofrapidprototypingtechnique[9].
Recently,wehaveintereststorealizetheInvitrorealizationofInvivoenvironmentFor
the purpose, it is needed to understand our biological system in detail The investigation
approachusingmicro/nanodevicescanmeasureandcontrolinsideInvivobiologicalsystem
Moreover,Invitroconstructionsandmeasurementsofbiologicalsystemcanrevealitclearly
insingle/multiplecellItisquiteimportantforthesignificantimprovementofregeneration
engineering/biology,inducingtheevaluationofclinicalconditionofbloodvesselWehave
beenworkedbloodvesselsimulatorandsurgicaloperationsystemtoimproveandintegrate
fortheInvivorealizationBasedontheEVEsystem,wearetryingtodevelopnoveltypeof
surgicalsimulationsystemwiththeInvitrorealizationofInvivoenvironmentThehuman
cellscanbeusedtoconstructasbloodvesselorotherorgansincludingdiseasesby3Dcell
assembly techniqueThesurgical simulator canbe integrated with micronano sensors and
devices to evaluate the surgical operations and other applications such as drug delivery
system.
3.Nanomechatronicsforindustrialandresearchapplication
Innanoscale,theimportanttechnologiesareNanomeasurement,NanofabricationandNano
assemblyfornanomechatronicsTheadvancedapplicationsareinvestigatedforquantumdots
[10],quantumprocessing[11],photoniccrystal[12],drugdeliverysystem(DDS)[13],field
emissiondisplay[14],nanofieldemissionelectronsource[15],nanoXraysources[16],nano
actuator[17] , nanotemperature sensor[18] , nanoIR sensor[19] , supermolecules for solar
energyconversion[20],andsoon.
Thewidescalecontrolleddevicesfromatomicscaletometerscaleisexpectedtorealizeinthe
near future For the high integrated, miniaturized, and functionalized NEMS, one of the
effectivewaysistousethebottomupfabricatednanostructuresornanomaterialsdirectlyAs
typicalexample,thenanodevicesareinvestigatedbasedonthecarbonnanotubes(CNTs)It
has interesting mechanical, electronic and chemical properties which have been under
investigationinvariousstudies[21]ThereispossibilitytousetheirfinestructuresdirectlyFor
example,telescopingcarbonnanotube,whichisfabricatedbypeelingofftheouterlayersof
multiwalled carbon nanotubes (MWNTs), is one of the most interest nanostructures As
previousworks,thepullingoutoftheinnercorewaspulledoutmechanicallyinsideaTEM
[22]TheMWNTswereusedthemastherotationaxisofsiliconchipasrotationalactuators
[23].WereportedonthedirectmeasurementsofelectrostaticactuationofatelescopingMWNT
insideSEMandTEM[17][24][25]Thoseapplicationsarenewlydevelopedusingthebottom
upstructuresofCNTs.
4.ResearchoncenterformicronanomechatronicsinGraduateSchoolof
Engineering,NagoyaUniversity
WeestablishedaCenterforMicronanoMechatronicsatGraduateSchoolofEngineering,
NagoyaUniversityin2008withtheaimofapplyingnanotechnologytopracticalsystemsin
micronanoscalefromasystemapproachviewpoint[26]OurCenterstrivestofosteryoung
researcherswhodaretochallengeunexploitedfieldsbybuildinganovelinterdisciplinary
fieldbasedonmicronanomechatronicsThisfieldwillpromotetheworldhighestlevelof
micronanomechatronicsresearchwithanemphasisonoriginalityfromaviewpointofnot
onlytheacquisitionofadvancedtechnology,butalsosocialissues.
Ourcenteraimsnotonlytocreatenovelfunctionalmaterialsandadvancedmechatronics,but
alsotodiscoverbreakthroughsinnextgenerationmedicinewepromoteresearchesinfour
basic fields, Nano control engineering, Nano measurement engineering, Nano design and
manufacturing,andNanomaterialsscienceandconductanappliedresearchencompassing
allthesebasicresearchfieldstoattendtotheneedsoftheadvancedmedicalengineering.
Ourcenterimplementsastrategytorealizeapplicationsofmicronanomechatronics,which
arebasedonmechanicalengineeringormaterialsscience,controlsystemsengineering,and
advanced medical engineering As shown in Figure 6, by establishing joint research and
international collaborations between the above research fields, we are creating the most
advanced micronano mechatronics and train researchers who can comprehend industrial
circles and social issues using an open cluster system as well as conduct research to solve
problemsspanningthesefourbasicfieldsInparticular,wewillinitiallyfocusontasksinthe
bio or medical welfare technologies using a number of unexploited fields, which may
consequently produce venture enterprises Some research results are figured as shown in
Figure7inthefourbasicresearchfieldsDetailinformationormorerecentresultswillbegiven
bythefollowingchaptersinthisbook.
Figure 6. Milestone of Micro-Nano mechatronics research fields
Figure 7. Examples of research results of micro-nano mechatronics from the center for micro-nano mechatronics of
Nagoya University
5.Conclusion
This chapter presents the brief introduction of research and technology on micronano
mechatronicsForindustrialapplications,variousdevicesaredevelopedandavailable,such
as applications to the automobiles, computer peripheries, amusements, printers, cameras,
robotics automation, environmental monitoring, biological/medical treatments, energy
resource,andsoonThosedevicesareinvestigatedbasedonthemicroandnanomechatronics
technologies to realize highefficiency, highintegration, highfunctionality, lowenergy
consumption,lowcost,miniature,andsoonMicronanomechatronicstechnologiescanbe
appliedtobreakthoughtheadvancedindustrialfieldincludingthenanobiologyandmedical
applications.
Authordetails
ToshioFukuda1,MasahiroNakajima2andMasaruKojima1
1DepartmentofMicroNanoSystemsEngineering,NagoyaUniversity,Nagoya,Japan
2GraduateSchoolofEngineeringCenterForMicronanoMechatronics,NagoyaUniversity,
Nagoya,Japan
References
[1] FeynmanR.P.TheresPlentyofRoomattheBottom.CaltechsEngineeringandSci
ence1960;232236.
[2] Whitesides G. M, Grzybowski B. SelfAssembly at All Scales. Science 2002;295
24072409.
[3] KannoI,TazawaY,SuzukiT,KoteraH.Piezoelectricunimorphmiroactuatorswith
XshapedstructurecomposedofPZTthinfilm.MicrosystemTechnologies2007;13(8)
825829.
[4] IshidaY,HakiaiK,BabaA,AsanoT.ElectrostaticInkjetPatterningUsingSiNeedle
Prepared by Anodization. Japanese Journal of Applied Physics 2005;44(7B)
57865790.
[5] Sun Y, Nelson B. J. MEMS Capacitive Force Sensors for Cellular and Flight Biome
chanics.BiomedicalMaterials2007;2(1)1622.
[6] Motoo K, Arai F, Fukuda T, Katsuragi T, Itoigawa K. High sensitive touch sensor
with piezoelectric thin film for pipetting works under microscope. Sensors and Ac
tuatorsA2006;12616.
10
[7] Yoshida K, Hagihara Y, Tanaka S, Esashi M. Normallyclosed Electrostatic Micro

ValvewithPressureBalanceMechanismforPortableFuelCellApplication.In:pro
ceedings of 19th IEEE International Conference on Micro Electro Mechanical Sys
tems,MEMS2006,2226January2006,Istanbul,Turkey.
[8] MotooK,TodaN,AraiF,FukudaT,SekiyamaK,NakajimaM.GenerationofCon
centration Gradient from TwoLayered Flow with HighSpeed Switching Micro
valve.BiomedicalMicrodevices2008;10329335.
[9] IkedaS,AraiF,FukudaT,NegoroM.AnInVitroSoftMembranousModelofIndi
vidualHumanCerebralArteryReproducedwithViscoElasticBehavior.In:proceed
ings of the 2004 IEEE International Conference on Robotics and Automation,
ICRA2004,1822April2005,Barcelona,Spain.
[10] InvitrogenCorporation.http://www.invitrogen.com/(accessed26July2012).
[11] Niskanen A. O, Harrabi K, Yoshihara F, Nakamura Y, Lloyd S, Tsai J. S. Quantum
CoherentTunableCouplingofSuperconductingQubits.Science2007;316723726.
[12] VlasovY.A,OBoyleM,HamannH.F,McNabS.J.Activecontrolofslowlightwith
photoniccrystalwaveguides.Nature2005;4386569.
[13] KakudoT.S,FutakiS.TransferrinmodifiedliposomesequippedwithapHsensitive
fusogenic peptide: an artificial viral like delivery system. Biochemistry 2004;43(19)
56185628.
[14] Saito Y, Uemura S. Field Emission from Carbon Nanotues and its Application to
ElectronSources.Carbon2000;38169182.
[15] YabushitaR,HataK,SatoH,SaitoY.Developmentofcompactfieldemissionscan
ningelectronmicroscopeequippedwithmultiwalledcarbonnanotubebundlecath
ode.JournalofVacuumScienceandTechnologyB2007;25640642.
[16] ZhangJ,ChengY,LeeY.Z,GaoB,QiuQ,LinW.L.,LalushD.,LuJ.P.,ZhouO.A
nanotubebasedfieldemissionxraysourceformicrocomputedtomography.Review
ofScientificInstruments2005;76094301.
[17] NakajimaM,AraiS,SaitoY,AraiF,FukudaT.NanoactuationofTelescopingMulti
walled Carbon Nanotubes inside Transmission Electron Microscope. Japanese Jour
nalofAppliedPhysics2007;42L10351038.
[18] ElshimyH.M,NakajimaM,ImaizumiY,AraiF,FukudaT.FabricationofFIBCVD
Nano Temperature Sensors for Local Temperature Sensing in Water Environments.
JournalofRoboticsandMechatronics2007;19(5)512518.
[19] LeeJ.U.Photovoltaiceffectinidealcarbonnanotubediodes.AppliedPhysics.Let
ters2005;87073101.
[20] MandaliaH.C,JainV.K,PattanaikB.N.ApplicationofSupermoleculesinSolarEn
ergyConversionAReview.ResearchJournalofChemicalSciences2012;289102.
[21] IijimaS.HelicalMicrotubulesofGraphiticCarbon.Nature1991;3545658.
[22] CumingsJ,ZettlA.LowFrictionNanoscaleLinearBearingRealizedfromMultiwall
CarbonNanotubes.Science2000;289602604.
[23] FennimoreA.M,YuzvinsskyT.D,HanW.Q,FuhrerM.S,CummingsJ,ZettlA.Ro
tationalActuatorsBasedonCarbonNanotubes.Nature2003;424408410.
[24] Dong L. X, Nelson B. J, Arai F, Fukuda T. Towards Nanotube Linear Servomotors.
IEEETransactiononAutomationScienceandEngineering2006;3228235.
[25] NakajimaM,AraiF,FukudaT.InsituMeasurementofYoungsModulusofCarbon
Nanotube inside TEM through Hybrid Nanorobotic Manipulation System. IEEE
TransactiononNanotechnology2006;5243248.
[26] CenterforMicronanoMechatronics,GraduateSchoolofEngineering,NagoyaUni
versity.http://www.mech.nagoyau.ac.jp/cmm/index_e.html(accessed26July2012)
11
Chapter 2
Neural Interfaces: Bilateral Communication Between

Peripheral Nerves and Electrical Control Devices
Goro Obinata, Hitoshi Hirata, Chikara Nagai,

Shigeru Kurimoto, Shuichi Kato and
Tomonori Nakano
1.Introduction
NeuralInterfacesaredatalinksbetweenhumannervoussystemandanexternaldevice,and
allowthetransmissionofinformationtoandfromthehumannervoussystemtotheexternal
device.Bioelectricsignalscanbeobtainedfromthenervoussystem,and/ortransmittedtothe
nervoussystemviaimplantedorsurfaceelectrodes.Onebeneficialapplicationofsuchneural
interfacesistoobtaincontrolsignalsfromundamagedsensorymotorareasofpatientsbrain
forcontrollingneuroprostheticdevicessuchasartificiallimbsorwheelchairs.Paralyzedor
amputated people can reconstruct certain motor functions by using such neuroprosthetic
devices. Brainmachine interfaces, which are direct data links between human brain and
machines,areonekindofneuralinterfaces[1].Suchinterfaceshavebeenproposedduringthis
decade to control prosthetic limbs, or to control machines such as wheelchairs or robotic
manipulators[2].However,ourpresentknowledgeonbrainfunctionsissolimitedthatwedo
notfullyunderstandthecodingofinformationexpressingthebehaviorinmotions;specially,
wedonotknowthevariationofthecodinginindividualdifferencesorinrelatedthoughtsor
emotions.
Inordertodesignpracticalneuroprostheticdevices,wefocusonneuralinterfaceswhichlink
peripheralnervoussystemandexternaldevicesinthispaper(PNI,PeripheralNeuralInter
face).Sinceperipheralnervoussystemismuchsimplerthancentralnervoussystem,wemay
avoidthedifficultproblemswhichcomefromourlimitedknowledgeonbrainfunctionswhile
theneuroprostheticdevicesinteractjustwiththeperipheralnerves.InSection2,wedescribe
interfacestoperipheralmotorneuronsbyusingreinnervationtypeelectrodes.Theelectrodes
areconstructedbyimplantingembryonicneuronsintoperipheralmotorunits.Theendplates
of the implanted neurons which grow into muscles make biological interfaces for motor
2013 Obinata et al.; licensee InTech. This is an open access article distributed under the terms of the
14
commandstothemuscles.Theotherendsoftheimplantedneuronsareconnectedbyspecial
typesofelectrodestocommunicatewithelectricdevicesofmotorcontrollers.Thediameterof
mammalian motor neurons is from 0.5 m to 20 m. There is a reason why our research
requires biocompatible micronano technologies for achieving such interfaces of new
reinnervationtypeelectrodes.
Neuroprostheticdeviceswithinterfacesdetectingelectromyogram(EMG)areinpracticaluse
[3].SuchEMGinterfacesrequiretoplacetheelectrodesattheendplateofamotorneuron;
thus,theparalyzedoramputateduserscannotcontroltheparalyzedoramputatedmuscles
withthesamepassesofmotorneuronsasbeforeparalyzedoramputated.Moreover,EMG
interfacespickupthesignalsatendplateofamotorneuron;sensorysignalsfromreceptors
arenotobtainableviaEMG.ThismeansthattwowaycommunicationbyEMGinterfacesis
impossible.Wecanassemblesensorsandstimulatorsintoaneuroprostheticdevice.Ifwehave
awaytosendthesignalsfromtheassembledstimulatorstosensorynervoussystem,ideal
neuroprostheticdeviceswithtwowaycommunicationwillbeachievable.Thecombinationof
EMG pickup and electrical stimulation with surface electrodes was proposed for twoway
communication[4].However,thereisnoclearresultontheamountoftransmissiveinforma
tionthroughtheafferentpasses.InSection3,weshowapreliminarystudyonthepossibility
ofsensoryfeedbackviaaxialfibersofperipheralsensoryneurons.Inotherwords,wetryto
achieveanartificialafferentpassforfeedingsignalsbacktothebrain.Amethodforimproving
ontheamountoftransmissiveinformationhasbeenproposedbyusingaconfigurationwith
multichannelelectrodes.
One application of PNIs is to achieve unconscious muscular movements such as walking,
writing,dancing,playingmusicalinstruments,andsoon.Foranexample,inwalkingmotions
someneuralnetworksgeneratethepatternofwalking,andthesteadybehaviorisclosedwithin
theperipheralnervoussystemandthepatterngeneratorsinspinalcord.Theuppercentral
nervoussystemprovidesthetriggersofwalkingsuchasstart/stop,speedup/slowdown,turn
right/turnleft. The understanding on pattern generators for walking is now enough to
simulatehumanwalkingorsomeanimalswalking.InSection4,wegiveawalkingsimulation
ofrattoshowthepossibilityofpracticalusagesofPNIs.
2.Transplantationofembryonicneuronsintoperipheralnerveforms
functionalmotorunits
2.1.Backgroundandpurpose
Therehasbeenarapidsurgeinclinicaltrialsinvolvingstemcelltherapiesoverthelastthree
tofouryears[5].Thosetrialsareestablishingtheclinicalpathwaysforregenerativemedicine,
especiallyinnervoussystem.Sincederivationofhumanembryonicstemcells(hESCs)in1998
[6], hESCs have been thought a promising source of replacement cells for regenerative
medicine.AlthoughGeronCorporationhasbeennolongerenrollingpatientsforthetrial,they
conductedthefirstclinicaltrialintheUnitedStatestoevaluatethesafetyofoligodendrocyte
Neural Interfaces: Bilateral Communication Between Peripheral Nerves and Electrical Control Devices
precursor cells derived from hESCs in patients with thoracic spinal cord injuries [7]. The
companyAdvancedCellTechnologyhasreportedpromisingresultsfromtheclinicaltrialson
Stargardts Macular Dystrophy with retinal pigment epithelium derived from hESCs [8].
However,neuronalreplacementincentralnervoussystemisstilllimitedduetoanumberof
cells required to reconstruct complex structures and narrow therapeutic time window [9].
There are still several hurdles to be overcome in order to establish clinical application of
neuronalreplacementtherapiesincentralnervoussystems.
One experimental approach to rescue denervated muscles from damage to lower motor
neuronsoraxonaldisconnectionistransplantationofmotoneuronsintoaperipheralnerveas
asourceofneuronsformusclereinnervation.SinceErbreportedthereinnervationofdener
vated muscle by embryonic motoneurons transplanted into peripheral nerve in 1993 [10],
severalstudieshaveinvestigatedthefactorsthatimprovemotoneuronsurvivalinperipheral
nerve[1113].Consideringthesimplicityofneuralnetworkandwidewindowsofopportunity
forthetreatment[14],peripheralnervesystemcanbeanidealtargetforneuronalreplacement
therapy. This transplantation strategy may provide the potential to excite these muscles
artificiallywithelectricalstimulation.Theaimofthisstudywastoevaluatewhethertrans
planted rat embryonic motorneurons into adult rat peripheral nerve would survive, and
whethertransplantedmotorneuronswouldformfunctionalmotorunits.
2.2.Methods
ThesciaticnervesofadultFischer344ratsweretransected;1weeklater,dissociatedembryonic
spinal neurons were transplanted into the distal stump of the tibial and peroneal nerves.
Surgical controls underwent the same surgeries but had only medium injected into the
peripheralnerves.Tissueanalysisandmeasurementofankleangleswereperformedtwelve
weeksafterneuraltransplantation.
2.2.1.Cellpreparation
VentralspinalcordcellwereobtainedfromFischer344ratembryos(JapanSLC,Inc.,Shizuoka,
Japan).AfterFischerratsonday14ofpregnancywereanesthetizedwithisoflurane,ventral
spinalcordswereresectedfromthefetusesusingasurgicalmicroscopeandwerecutintosmall
piecesinicecoldHanksbalancedsaltsolution.Ventralspinalneuronsweredissociatedusing
papaincontainingseparationsolution(MBX9901;SumitomoBakeliteCo.Ltd,Tokyo,Japan).
For implantation, dissociated neurons were suspended in Neurobasal medium (Gibco)
containingB27supplement(Gibco),Glutamax(Gibco),andN2supplement(Gibco).
2.2.2.Surgicalproceduresandtransplantation
Allprocedureswereperformedon8weekoldmaleFischer344rats(JapanSLC,Inc.)under
isofluraneanesthesia.ThesciaticnervesofFischerratswerecompletelytransectedatmidthigh.
Thenerveswereligatedonbothendsandtheproximalnervestumpwassuturedintohip
musclestopreventreinnervation.Approximately1x106neuronscontainedwithin10lof
mediumwereinjectedintothedistalstumpsofthetibialandperonealnervesafter1weekof
15
16
nervetransectionusingaHamiltonsyringe.Theinjectionsitewas20mmproximaltothose
entriesintothelateralgastrocnemiusandanteriortibialismuscles.Surgicalcontrolsunder
wentthesamesurgeriesbuthadonlymediuminjectedintotheperipheralnerves.
2.2.3.Reproductionofanklemotions
Thestainlesssteelwireelectrodeswereplacedontheperonealandtibialnervesoftheneural
transplantationgroupandthosewerecoveredwithsiliconegelforinsulationandimmobili
zation.Theotherendsofwireswerepassedthroughthedorsalneckskinandconnectedtothe
generator(NeuropackMEB5504;NihonKohden,Tokyo,Japan).Usingvideotechnique,ankle
anglesoftheneuraltransplantationgroupweremeasuredwithelectricalstimulationofthe
peronealnervesat60Hz,0.4mA.Theankleangleismeasuredfromthelineconnectingthe
kneeandanklejoints,andthelineconnectingtheanklejointandthemetatarsalhead[15].
2.2.4.Tissueanalysis
Afterthemeasurementofankleangles,theratswereperfusedthroughtheleftventriclewith
4% paraformaldehyde. The tibial nerves, gastrocnemius and anterior tibialis muscles were
removedforimmunohistochemicalandhistochemicalanalysis.Thetibialnervedistaltothe
transplantwasthenfixedin4%glutaraldehydeinphosphatebuffer.Thenerveswereembed
ded in Epon and 1m thick sections were stained with Toluidine blue (SigmaAldrich,
Germany) for light microscopic examination. The total number of myelinated axons was
measured.Thegastrocnemiusmusclesweredissectedfreefromtheoriginandinsertionand
weighedimmediately.Thenthegastrocnemiusmuscleswereembeddedinparaffin,sectioned
into10mcrosssectionsandstainedwithhematoxylinandeosin.Themeanmusclefiberarea
wasmeasuredatthemiddleofthelateralgastrocnemiusmusclebelly.
Neuronsurvivalwasexaminedinthecelltransplantationgroup.Thetransplantsitesoftibial
nerveswerecryoprotectedinsucrose,andthenfrozenindryicecooledisopentane.Thetibial
nerve sections of 10m thichness were stained with the antibodies against choline acetyl
transferase (ChAT, 1:50; Millipore, Billerica, MA, USA) in order to assess the survival of
motoneuronsandthoseaxonsinthetibialnerve.Theanteriortibialismuscleweresectioned
at 50m to assess neuromuscular junctions using antineurofilament H antibody (1:500;
Millipore,Billerica,MA)andbungarotoxin(1:200,MolecularProbes,Eugene,OR)which
revealsterminalmuscleinnervationbystainingmotorendplates.Sectionswerestainedusing
a full automatic Immunohistochemical staining system: Ventana HX system (Ventana,
Yokohama,Japan)accordingtothemanufacturersinstruction.
2.2.5.Statistics
Forstatisticalanalysis,StudentttestorANOVAwithTukeyposthoccomparisonswasused,
asappropriate.AllstatisticalanalyseswereconductedusingtheStatisticalPackageforSocial
Scienceversion19.0software(SPSSInc,Chicago,IL,USA).Thesignificancelevelwassetat
0.05.Alldataareexpressedasmeanstandarddeviation.
2.3.Result
2.3.1.Motoneuronsurvivalandneuromuscularjunctionformations
Transplantedmotoneuronsurvivalwasobservedupto12weeksinallofthecelltransplanted
tibialnerves.Neuromuscularjunctionformationswerealsopresentinthelateralgastrocne
miusmuscleoftheratsthatreceivedcelltransplants(Figure1).
2.3.2.Axonregeneration
No myelinated axons were present in the surgical control group. The mean number of
myelinatedaxonsinthetibialnervesofanimalsthatreceivedcelltransplantionwaslessthan
halfofuninjuredanimals(11981218versus2510802,p=0.079)(Figure2).
2.3.3.Analysisofmuscleanatomy
Nosignificantdifferenceswerefoundinmusclefibercrosssectionalareabetweentransplan
tationandsurgicalcontrolgroups(326m291versus176m2105,p=0.28).Thenaivegroup
showedthelargestgastrocnemiusmusclefiberarea(2873233m2,p<0.001)
2.3.4.Reproductionofanklemotions
Thereinnervatedmuscleoftransplantationgroupcouldreproduceankledorsiflexionartificial
lythroughelectricalstimulation.Videobasedanalysisrevealedthattheankleanglesinthe
neuraltransplantationgroupsignificantlydecreasedwithelectricalstimulationofthepero
nealnerveascomparedtosurgicalcontrol(30.47.7versus106.79.7,p<0.001)(Figure3).
Figure 1. Twelve weeks after transplantation, motoneurons survived in the tibial nerve. The ChAT (choline acetyltrans
ferase) positive axons and motoneurons (arrow) were observed in the tibial nerve.
17
18
(a)Naive
(b)Neurontransplantation
(c)Surgicalcontrol
Figure 2. Myelinated axons in tibial nerve were assessed with toluidine blue stain.
Figure 3. Controlled reproduction of ankle motion and measurements of dorsiflexed angles. The ankle motions were
restored with electrical stimulation of the peroneal nerve in neural transplantation group (A) as compared to surgical
control group (B).
2.4.Discussion
Thetransplantedratembryonicmotoneuronscouldsurviveintheratperipheralnerve,and
transplantedmotoneuronsformedfunctionalmotorunits.Transplantationofmotoneurons
intoperipheralnerveprovidedthereproductionofankledorsiflexion.Thetransplantationof
embryonic neurons into peripheral nerve could restore ankle motions in conjunction with
electricalstimulation,eventhoughnoneuralconnectionbetweencentralnervoussystemand
musclewerepresent.
Althoughallratswereabletoflextheirankle,thesmallnumberofaxonswaspresentinthetibial
nervesofcelltransplantationgroupascomparedwithnaivegroup.Endoneurialenvironment
in peripheral nerve may not provide suitable condition for motoneurons. Transplanting
motoneuronwithglialcellsinthecentralnervesystems,suchasastrocytes,maypromotethe
survivalofmotoneurons.Furtherstudiesareneededtoclarifythenecessityofcentralnervous
systemderivedcellstoimprovesurvivalrateoftransplantedneuronsinperipheralnerve.
No significant differences were found in muscle fiber area between transplantation and
surgicalcontrolgroups.Andthemusclefiberareawasmuchsmallerforthecelltransplanta
tion group than for the naive group. Before neural connection between motoneurons and
musclewasestablished,othertreatmentstrategyhadbeenneededtopreventmuscleatrophy.
After the neural connection, continuation of electrical stimulation and exercise beyond the
periodofthisstudymayimprovemusclefiberareaandthequalityofgait.
Inclinicalsetting,severalinvestigatorshavestatedthatperformingreconstructionofdener
vated muscle with nerve transfer technique in peripheral nerve within 9 months notably
improvesfunctionaloutcome[16,17].Motoneuronscanbeplacedverycloselytoneuromus
cularjunctionusingtheneuraltransplantationintoperipheralnerve.Targetmusclescanbe
reinnervatedinshortterm.Consideringthesefacts,usefulfunctionalrecoveryofdenervated
musclecanbeexpectedwiththismethodevenifthetreatmentwasdelayeduntilmorethan6
monthsafterWalleriandegeneration.Consideringthatgeneratingneuralcellsfrominduced
pluripotentstemcells(iPSCs)takesafewmonths,thewidewindowsofopportunityforthe
treatmentcanbeadvantageinpropagatingsafemotoneuronsderivedfromiPSCsmadefrom
theinjuredpatientsowncells.
Another advantage of transplantation into peripheral nerve is that much less number of
motoneuronswillberequired,comparedwithneuraltransplantationintothecentralnervous
system.TheiPSCsarelikelytocarryahigherriskoftumorigenicitythanEScells,duetothe
inappropriatereprogrammingofthesesomaticcells,theactivationofexogenoustranscription
factors,orotherreasons[18,19].Alargernumberoftransplantingcellsareneeded,thehigher
risk of tumorigenicity remains an obstacle for neural transplantation therapy. Presumably
muchlessnumberofmotoneuronsaresupposedtoberequiredforneuraltransplantationinto
peripheralnerve,becausethenumberofmotorunitisusuallyaboutseveralhundred[20].
Consideringtheseissues,transplantationofmotoneuronsintoperipheralnervecanbeanideal
targetforregenerativetherapyfromaclinicalviewpoint.
Transplantationofratembryonicmotoneuronsintoperipheralnerveprovidesreproduction
ofsimplebehaviorsinconjunctionwithelectricalstimulation.Recentbreakthroughsinstem
cellbiologyhaveraisedpossibilitiesoftheclinicalapplicationofthistreatmentstrategy.Our
resultsleadustobelievethattransplantationofiPSorESderivedmotoneuronsintoperiph
eralnervechangestreatmentstrategyforthediminishedvoluntarymusclefunction.
3.Sensoryfeedbackbyelectricalstimulation
3.1.Purpose
Thepurposeofthissectionistoshowthepossibilityofsensoryfeedbackinnueroprosthetic
devices.Ifwehaveawaytosendthesignalsfromtheassembledsensorstosensorynervous
19
20
system,leadingneuroprostheticdeviceswithtwowaycommunicationwillbeachievable.In
thisSection,weshowapreliminarystudyonthepossibilityofsensoryfeedbackviaaxialfibers
ofperipheralsensoryneurons.Wehavetestedseveralpatternsofelectricalsignalsandthe
placementsofsurfaceelectrodesforcarryingsensoryinformationtobrainviaaxialfibersof
neurons.Inapreliminaryexperimentwithonechannelelectricalstimulation,theamplitude
modulationofthestimulationvoltageachievesonly2bitresolution.Then,weincreasethe
channelsandtakeamethodoffrequencymodulationofthestimulationelectricalsignal.
3.2.Methods
Onepairofsurfaceelectrodes,oneanodeandonecathode,isusuallyusedforonechannel
electricalstimulation.Thetwopairsofsurfaceelectrodesinourexperimentsareshownin
Figure4a.Theplacementsoftheelectrodeswerewellcontrolledforeachexperimenttocertify
therepeatabilitybasedonthemarkersonskinshowninFigure4b.Wesetthetwopairsof
electrodes(A,B)and(C,D),andmadetheelectricalcurrentflowfromAtoBorfromCtoD
ineachpair.Thedutyratioofstimulationvoltageswas5%,andthefrequencywassetat2000
Hzforthechannel(A,B).Thedutyratio5%wasdefinedbecausewefounditlooksgoodfor
all five subjects to identify the stimulation frequency in several trials with one channel
stimulation.
Figure 4. The electrodes placement
Theappliedvoltagesweretunedforeachsubjectinsimilarwayforthegoodidentifiabil
ity.Theelectricalstimulationofsurfaceelectrodeswithabout2000Hzgivessomefeeling
ofvibrationaroundattheelectrodeswithoutanypain.Thestimulationfrequencyofthe
channel(C,D)wassetatfivefrequencies:2000Hz,2001Hz,2002Hz,2004Hzand2010
Hz. The change of stimulation frequency on the channel (C, D) causes different sensa
tions to the subjects. More specifically, the subjects sense the frequency difference be
tweenthetwochannelsinadditiontothebasefrequency2000Hz.Inotherwords,the
subjectssensetheinterferencepotentialofthepairelectricalpotentials.Foranexample,1
Hz sensation will be felt by the subject if the frequencies of 2000 Hz and 2001 Hz are
selected.Weappliedelectricalstimulationsof2000Hztoasubjectbyusingthepairofthe
electrodes(A,B),andsetthesameanddifferentfrequencies forthepairoftheelectro
des(C,D)forthediscriminationexperiments.Thephaseshiftwaszerodegreewhenthe
both frequency were set at 2000 Hz. After enough trials for each combination of the
stimulationfrequencies,eachsubjectcarriedout40trialsofthediscrimination.Inthetrials,
allcombinationsoffrequencieswerepresentedinrandomorder.Fivesubjectswereasked
toidentifythecombinationsoftheelectricalstimulations.
3.3.Results
The average rate of the correct answers over 40 tests per subject was 90%. All errors were
occurredinthecasesofnearestfrequenciesdifferenceofthetwopairselectrodes.Thisresults
show that the difference of the two channel stimulation frequencies is key point for the
identifiabilityofstimulations.Thebiggerdifferenceinstimulationfrequencies,thehigherrate
ofthecorrectanswerswillbeachieved.
3.4.Discussion
Thesimpleamplitudemodulationofonechannelelectricalstimulationachievesonly2bits
resolution for transferring sensory information to brain. On the other hand, frequency
modulation with interferencepotential ofthe pair electrical potentialshas achieved a high
resolution which corresponds to the number of combination of given different stimulation
frequencies.Foranexample,wecanobtain3.5bits(1digitindecadalsystem)resolutionifwe
use five different frequencies. We can increase number of the channels; thus, interference
potentialofthepairelectricalsignalswithmorethantwopairsofelectrodesholdsforththe
possibility that we can achieve a high resolution for sensory feedback via axial fibers of
peripheralsensoryneurons.
Furtherresearchonmultichannelelectrodes,speciallymorethanthreechannels,willbe
required to clarify the effect of multiple interference of stimulation frequencies on the
sensory resolution of transferring the information. The frequency range which is most
appropriatefortransferringsensoryinformationshouldbealsoclarified.Itiscommonthat
electrical stimulations with a certain frequency band involve a pain of the subject.
Appropriatefrequencybandsshouldbeidentifiednotonlytoachieveahighresolution,
buttoavoidsubjectspain.
21
22
4.Restorationofwalkingbyartificialperipheralnervesystem
Simulationstudywithneuromusuculoskeletalmodelofrat
4.1.Purpose
OneapplicationofPNIsistoreplacethepartofneuralnetworksforaparticularmovement
by artificial electrical networks or artificial digital networks. In this section, we propose a
methodofPNIsforpartiallyimpairedneuralnetworks.Whenwelookatinjuredpersons,it
is common that some muscles do not work because the corresponding motor neurons are
damagedbutthemusclesdonotreceiveanydamagethemselves.Insuchcases,themuscles
canworkifthemotorcommandsareprovided.
Inthissection,wemakethefollowingassumptionsasconditionsforoursimulationofinjured
ratswalking.First,motorcommandstomusclesforwalkingaregeneratedbyspecifiedneural
networks.Thenetworksiscalledcentralpatterngenerator(CPG),whichhasbeenidentified
inspinalcordsofmammalian.Theuppercentralnervoussystemisassumedtogiveonlythe
triggerofwalking.Second,somesensorysystems,whichmaybemusclespindlesoranother
somatosensory organs, are damaged but we are able to obtain the necessary signals for
constructingthemotorcommandsbyusingsomeartificialsensorsinsteadofthedamaged
sensorysystems.Third,weareabletoprovidetheconstructedmotorcommandsforthetarget
musclesthroughreinnervationtypeelectrodes,whicharedescribedinSection2.Basedon
theseassumptions,wegiveawalkingsimulationofrattoshowthepossibilityofpractical
usagesofPNIs.Simulationsofhumanbipedalwalkinghavebeensuccessfullyproposed[21,
22];therefore,wemayrestorethewalkingmotionforpatientswithspiralcordinjuryifour
proposedmethodworkswellforratwalking.
AnexampleoftheexperimentalsetupcorrespondingtothesimulationisgiveninFigure5.
Thecamerasystemincludingimageprocessingistoobtainthenecessarysignalssuchasjoint
angles,jointangularvelocities,andothersignalsforthepositionsandorientationsofbody
: Marker for measuring positions of body

: Reinnervation type electrodes
CPG
Controller
Injured limb
Electric
Stimulator
Figure 5. Proposed experimental setup for reconstructing walking motion.
Stimulator
for
movement
segments.Itispossibletoextendthistypeofneuroprostheticsystemwithportablesensors
andotherelectricaldevicestomorepracticalonefordisabledpersons.
Thepurposeofthissectionistoconductwalkingsimulationwhichwillplayanimportantrole
ofoursuccedentneuroprostheticsystem.Thebasicconceptoftheneuroprostheticsystemis
showninFigure5.TheroleistocheckwhetherthedesignedartificialCPGcontrollerworks
adequatelyornot,andwhetherthesensingsystemisabletoprovidenecessaryinformation
forthecontrollersadaptingtotheenvironmentalchanges.Thus,thewalkingsimulationgiven
inthissectionmeansthepreliminarystudyofmotionplanningandcontrolforoursuccedent
neuroprostheticsystem.
4.2.Neuromusuculoskeletalmodel
The neuromusculoskeletal model of rat is composed of a rigid link system, joint torque
generators,neuralcontrollersofCPGswiththeupperleveltriggers.Theinertialpropertiesof
theentirebodyarerepresentedbyathreedimensional17rigidlinksystemwith16joints.All
thejointsareonedegreeoffreedom,andtheaxesofthejointsareperpendiculartosagittal
plane.TheschematicviewsinsagittalandhorizontalplanesaregiveninFigure6.Thedappled
circlesindicatethecenterofmassofeachsegment.
Aviscoelasticpassivemomentinducedbysofttissuesandapassivemomentofnonlinear
elasticityinducedbytendonsandbonesassumedtoactoneachjoint.Thecharacteristicis
givenby
passivei i , i kiJ1 exp kiJ2
J

J
J
ki 3 i i ki 4 exp ki 5
J
J
i i ki 6 ci i
(1)
.
whereiistheithjointanglevariable, i istheithjointangularvelocityvariable,andanother
parameters are constants representing the joint characteristics. These parameters were
determinedfromtheestimationsforhumanjoints[23].Theinertiapropertiesofeachbody
segment,suchasthemassandthemomentofinertia,wereestimatedfromthesizeofanato
mizedeachbodysegmentwiththeaveragedensityofratbody[24].
60mm
y
z
50mm
110mm
(a)viewinsagittalplane
Figure 6. Assumed skeletal system of rat.
(b)viewinhorizontalplane
23
24
Thefunctionsoftheneuralcontrolsystemaredividedintotwotypes.Thefirstfunctionisa
rhythmpatterngeneratorcorrespondingtothespinalcordlevel.Thissubsystemcomposes
CPGswhichworkascommandgeneratorsforthejoints.Thecommandgenerationofeach
CPGstartsbyreceivingastimulusfromthehighercenter.EachunitofCPGisasecondorder
nonlineardynamicalsystemwiththeinputsfromthehighercenterandthesensorysystems.
Duringmovementofthelinksystem,somaticsensorysignalsarefedbacktotheCPGssothat
the musculoskeletal system is able to interact with the environment. This is the second
functionoftheneuralcontrolsystem.
TheneuralcontrolsystemofCPGisanonlinearfeedbackcontroller:
1
Tr
xi + xi = aij y j bzi + ui + si
(2)
j=1
1
T a zi
(3)
+ zi = yi
yi = max (0, xi )
(4)
Here,iandjaretheoscillatornumbersthatarecoupledintheantagonisticrelation,xiisthe
membranepotentialoftheneuron,Tr,Tdandbarethecoefficientscorrespondingtoaging
variation,yiisthefiringrateoftheneuron,aijistheweightcoefficient,ziisavariablecorre
spondingtotheadaptationlevel,siisthestimulationinputfromthebrain,anduiisthefeedback
signalfromthesomaticsensation.Theoscillatorneuronsrepresentedbyequations(2),(3)and
(4) are coupled to one another in a depressive manner, and the entire walking behavior
undergoes an adaptation process by causing an oscillation in each joint movement [25].
Movementofajointiscoupledwiththatofotherjoints.Inthissimulation,onlythefourjoints
atshouldersandhipswereassumedtogenerateactivelythetorques.Onepairoftheoscillator
unitisattachedatoneofthefourjoints.TheextraconnectionsbetweenCPGsofleftandright
shoulders, and also between CPGs of left and right hips are assumed for coordinating the
motionsofleftandrightlimbs.Theseextraconnectionscausemutuallyinhibitionsbetween
leftandrightCPGsandthengeneratealternatemotionsbetweentheleftandrightlimbs.The
structureofCPGsisgiveninFigure7.
u0
u0
LF
w21
LH
Figure 7. Structure of CPGs.
RF
w43
RH
Thejointtorqueisdeterminedasfollow:
= pF yF pE yE
(5)
whereyFandyEaretheoutputsofonepairofCPGsinwhichoneCPGgeneratesflexiontorque
andtheotherCPGgeneratesextensiontorqueateachactivejoint.Foranotherjointsexcept
theseactivejoints,therotationalmotionswillbeinducedaccordingtothepassivecharacter
isticsdefinedby(1).
Theinteractionbetweenthefootandthegroundwasmodelledasacombinationofsprings
anddampers.Thegroundreactionforcesproducedbythespringsanddamperswereassumed
toactonfourpointsofeachfoot:twointheheelandtwointhetoe.
In this simulation, we did not take the muscle model into considerations. However, the
generatedmuscleforcescorrespondingtothegeneratedtorquesbythisneuralcontrollercan
beestimatedwithanappropriateprocedurebythemethodin[23].
WesearchedforsuitableparametersoftheCPGbasedcontrollersuchasTr,Td,aij,b,etc.in
eachjointtoachievearealisticwalkingpattern,usingaperformanceindexthatisaweighted
linearcombinationofenergyconsumptionperunitwalkingdistance,musclefatigueindex
andfootclearance.
4.3.Simulationresults
Figure8showsthewalkingsimulationresultsinsagittalplanewhicharecalculatedby0.1
second.Mammalianswithfourlimbsareusuallyabletomoveinsomepatterns,suchaswalk,
trot, and gallop. Each pattern is appeared around the corresponding optimal velocity. In
mammaliansmoving,optimalmeansenergyefficientwayingatepatternsforgettingunit
movingdistance.FromFigure8,thepatternofwalkisobservedandthemovingvelocityis
matchedapproximatelytotheoptimalvelocityofratwalking.Figure9showsthattheoutputs
ofCPGssuccessfullygeneratedrhythmicpatternsasfunctionsoftimeandindicatedappro
priate phase shifts each other for coordinated motions in four limbs. The corresponding
generatedtorquesaregiveninFigure10.ThesimilarpatternsastheCPGoutputsareobserved.
WecandefinetheperiodofwalkingfromFigure10,anditis0.2second.Thephaseshiftis
180degreefromtheleftforelimbtotherightforelimb,andalso180degreefromtheleft
forelimbtothelefthindlimb.Ontheotherhand,thephaseshiftbetweentheleftforelimb
andhindlimbis97degreeandintherightsideitisthesame.Thesephaseshiftsmatchtothe
patternofwalk[26].Figure10showstheactivejointsanglescontrolledbytheCPGcontrol
lers.Itseemsthatthedifferencebetweenleftandrightangelsbothinforelimbsandhind
limbscomesfromtheeffectofinitialconditionsonthejointangelsandjointangularvelocities.
4.4.Discussion
We observed a large rolling motion of the trunk in the walking simulation. This can be
explainedbythefactthatnojointofrollingmotionisintroducedintheskeletalmodel.Atthis
stage, we do not have any knowledge about rolling joints of rat. The placements, passive
25
characteristicssuchasrangeofjointangleandviscoelasticpropertyshouldbeclarifiedbefore
therollingjointsareintroducedintheskeletalmodel.Theproperintroductionsuchrolling
jointsreducesthemagnitudeoftrunkrollingmotionandmayimprovethesimulationquality
incomparisonwiththeactualwalkingmotionofrat.
t=0.0[sec]
10
20
40 [mm]
30
t=0.5[sec]
10
20
30
10
20
30
t=0.1[sec]
10
20
40 [mm]
30
10
20
10
20
30
40 [mm]
t=0.3[sec]
40 [mm]
30
10
20
30
10
20
10
20
30
40 [mm]
30
Figure 8. Walking simulation results.
Forefoot Left
Hindfoot Left
Forefoot Right
Hindfoot Right
cycle
2
1
0
100
200
300
-1
-2
40 [mm]
40 [mm]
t=0.8[sec]
t=0.4[sec]
40 [mm]
t=0.6[sec]
t=0.7[sec]
t=0.2[sec]
CPG output
26
Time [msec]
Figure 9. CPGs outputs as commands for joint torques in steady state.
400
500
40 [mm]
Joint Torque [Nm]
Forefoot Left
Hindfoot Left
Forefoot Right
Hindfoot Right
cycle
100
-1
200
300
400
500
Time [msec]
Figure 10. Generated joint torques.
Joint Angle [degree]
80
60
40
20
0
0
-20
-40
-60
-80
Forefoot Left
Hindfoot Left
Forefoot Right
Hindfoot Right
100
200
300
400
500
Time [msec]
Figure 11. Joint angles at shoulders and hips.
Weextendedthemodeltothecaseofmoreactivejointscontrolledbytheneuralcontrollers.
Otherfouractivejointswereintroducedatelbowsinforelimbsandatkneesinhindlimbs.
Thesimulationresultsshowedthatthegeneratedmotionwasapatternofgallop.Fromthis
simulation,itwillberequiredthatweconsiderthemethodfordistinguishpatternsofmoving
whenwesearchtheparametersintheneuralcontrollers.
Although there exist several problems to be solved, the simulation results presented here
indicate the possibility of functional restoration for the cases of partially impaired neural
networks.Wewillbeabletousethissimulationtechnologyforcheckingwhetherthedesigned
artificialCPGcontrollerworksadequatelyornot,andforprovidingnecessaryinformationto
the controllers adapting to the environmental changes. The usage of this kind of online
simulatorisalsoillustratedinFigure5.
27
28
5.Conclusion
Inthischapter,wepresentedthreetechnologiesforachievingPNIs.ThePNIswillbedesigned
forpatientsofpartiallyimpairedneuralnetworks.InSection2,weshowthatrestorationof
motoneuronsispossiblebyusingreinnervationtypeelectrodeswhichareconstructedby
implantingembryonicneuronsintoperipheralmotorunits.Severalmethodsforrestoring
motoneuronfunctionshavebeenproposedduringthesethreedecadesbecausethemusclescan
generatetheforcesifsomeelectricpotentialsareprovidedatthemotorpoints.Theseelectric
potentials may be provided by artificial ones with some electrodes. Functional electrical
stimulation(FES)hasbeenappliedtohumanpatientswithspinalcordinjuryandhasshown
higherpotentialforrestoringfunctionalmovements[27].UsuallyinFES,somemetalwiresare
usedfortransmittingtheelectricalsignalsasthemotorcommandsforthemuscles.However,
themechanicalpropertiesofthemetalwiressuchaselasticityaredifferentfrombiologicaltissues:
then,themismatchingbetweenthewiresandthetissueswillappearandresultinmalfunc
tionsoftheFESsystems.Ifweusereinnervationtypeelectrodesconstructedfromembryonic
neurons,whicharealmostsameastheoriginalmotorneurons,thedemandingproblemof
mechanicalpropertymismatchingwithmetalwiretransmissionlineswillbesolved.Thisisthe
reasonwhyweseektoestablishthetechnologyaroundreinnervationtypeelectrodes.
Sensoryfeedbackincontroloffunctionalmovementsplaysanimportantrolefortherobust
nessandtheadaptationtoenvironmentalchangesinthetaskexecutions.Afewstudieshave
beencarriedouttointroducesensoryfeedbackfortherestorationofmotorfunctions[28].We
canimprovegreatlythequalityoffunctionalrestorationbysensoryfeedbackforthecasesof
partiallyimpairedneuralnetworks.InSection3,weshowedthepossibilityofsensoryfeedback
via axial fibers of neurons in nueroprosthetic devices. A method for passing the feedback
signalsviaaxialfibersofneuronswithsurfaceelectrodeshasbeentested.Itisshownthatthree
orfourbitsresolutionforsensoryfeedbackcanbeachievedwiththeproposedmethod.
InSection4,weshowedthepossibilityofpracticalusagesofPNIsbytakingwalkingsimulation
of rat. In the concept of restoring motor functions, controlling joint torques with sensory
feedback is a key point of the technology. The reinnervation type electrodes described in
Section2andthesensoryfeedbackviaaxialfibersofneuronsdescribedinSection3willbe
includedinsuchanewtypeofnueroprostheticdevice.
Authordetails
GoroObinata1,HitoshiHirata2,ChikaraNagai3,ShigeruKurimoto2,ShuichiKato2and
TomonoriNakano2
1EcoTopiaScienceInstitute,NagoyaUniversity,Nagoya,Japan
2GraduateSchoolofMedicine,NagoyaUniversity,Nagoya,Japan
3GraduateSchoolofEngineering,NagoyaUniversity,Nagoya,Japan
References
[1] Wolpaw J. R, Birbaumer N, McFarland D. J, Pfurtscheller G, Vaughan T. M. Brain
computer interfaces for communication and control. Clinical Neurophysiology
2002;113(6)767791.
[2] PatilP.G,TurnerD.A.Thedevelopmentofbrainmachineinterfaceneuroprosthetic
devices. The Journal of the American Society for Experimental NeuroTherapeutics
2008;5137146.
[3] Patterson P. E, Katz J. A. Design and evaluation of a sensory feedback system that
providesgraspingpressureinamyoelectrichand.JournalofRehabilitationResearch
andDevelopment1992;29(1)18.
[4] ScottR.N,BrittainR.H,CaldwellR.R,CameronA.B,DunfieldV.A.Sensoryfeed
back system compatible with myoelectric control. Medical and Biological Engineer
ingandComputing1980;18(1)6569.
[5] TrounsonA,ThakarR.G,LomaxG,GibbonsD.Clinicaltrialsforstemcelltherapies.
BMCmedicine2011;952.
[6] ThomsonJ.A,ItskovitzEldorJ,ShapiroS.S,WaknitzM.A,SwiergielJ.J,MarshallV.S,
Jones J.M. Embryonic stem cell lines derived from human blastocysts. Science
1998;28211451147.
[7] AlperJ.GerongetsgreenlightforhumantrialofEScellderivedproduct.Naturebi
otechnology2009;27213214.
[8] SchwartzS.D,HubschmanJ.P,HeilwellG,FrancoCardenasV,PanC.K,OstrickR.
M,MickunasE,GayR,KlimanskayaI,LanzaR.Embryonicstemcelltrialsformacu
lardegeneration:apreliminaryreport.Lancet2012;379(9817)713720.
[9] Okano H. Stem cell biology of the central nervous system. Journal of Neuroscience
Research2002;69698707.
[10] ErbD.E,MoraR.J,BungeR.P.Reinnervationofadultratgastrocnemiusmuscleby
embryonicmotoneuronstransplantedintotheaxotomizedtibialnerve.Experimental
Neurology1993;124372376.
[11] GrumblesR.M,CasellaG.T,RudinskyM.J,GodfreyS,WoodP.M,ThomasC.K.
The immunophilin ligand FK506, but not the P38 kinase inhibitor SB203580, im
provesfunctionofadultratmusclereinnervatedfromtransplantsofembryonicneu
rons.Neuroscience2005;130619630.
[12] GrumblesR.M,SesodiaS,WoodP.M,ThomasC.K.Neurotrophicfactorsimprove
motoneuron survival and function of muscle reinnervated by embryonic neurons.
JournalofNeuropathology&ExperimentalNeurology2009;68736746.
29
30
[13] GrumblesR.M,WoodP,RudinskyM,GomezA.M,ThomasC.K.Musclereinnerva
tionwithdelayedorimmediatetransplantofembryonicventralspinalcordcellsinto
adultratperipheralnerve.CellTransplantation2002;11241250.
[14] MerrellG.A,BarrieK.A,KatzD.L,WolfeS.W.Resultsofnervetransfertechniques
forrestorationofshoulderandelbowfunctioninthecontextofametaanalysisofthe
Englishliterature.JournalofHandSurgeryAmericanVolume2011;26303314.
[15] YuP,MatloubH.S,SangerJ.R,NariniP.Gaitanalysisinratswithperipheralnerve
injury.Muscle&Nerve2001;24231239.
[16] MillesiH.Surgicalmanagementofbrachialplexusinjuries.JournalofHandSurgery
AmericanVolume1977;2367378.
[17] NaganoA,TsuyamaN,OchiaiN,HaraT,TakahashiM.Directnervecrossingwith
theintercostalnervetotreatavulsioninjuriesofthebrachialplexus.JournalofHand
SurgeryAmericanVolume1989;14980985.
[18] MiuraK,OkadaY,AoiT,OkadaA,TakahashiK,OkitaK,NakagawaM,Koyanagi
M,TanabeK,OhnukiM,OgawaD,IkedaE,OkanoH,YamanakaS.Variationinthe
safetyofinducedpluripotentstemcelllines.Naturebiotechnology2009;27743745.
[19] Yamanaka S. Strategies and new developments in the generation of patientspecific
pluripotentstemcells.Cellstemcell2007;13949.
[20] Bromberg M. B. Updating motor unit number estimation (MUNE). Clinical neuro
physiology:officialjournaloftheInternationalFederationofClinicalNeurophysiol
ogy2007;11818.
[21] Taga G, Yamaguchi Y, Shimizu H. Selforganized control of bipedal locomotion by
neural oscillators in unpredictable environment. Biological Cybernetics 1991;65(3)
147159.
[22] HaseK,YamazakiN.Computersimulationstudyofhumanlocomotionwithathree
dimentionalentirebodyneuromusculoskeletalmodel.JSMEInternationalJournal.
SeriesC2002;45(4)10401050.
[23] HaseK,YamazakiN.Developmentofthreedimensionalmusculoskeletalmodelfor
variousmotionanalyses.TransactionsoftheJapanSocietyofMechanicalEngineers.
C1995;61(591)44174412.
[24] HayakawaT,IwakiT.Acoloratlasofsectionalanatomyoftherat.Tokyo:Adthree;
2008.
[25] Matsuoka K. Sustained oscillations generated by mutually inhibiting neurons with
adaptation.BiologicalCybernetics1985;52(6)367376.
[26] NanuaP,WaldronK.J.Energycomparisonbetweentrot,bound,andgallopusinga
simplemodel.JournalofBiomechanicalEngineering1995;117(4)466473.
[27] ShimadaY,SatoK,KagayaH,etal.Closedloopcontrolforfunctionalelectricalstimu
lationwithpercutaneouselectrodesinparaplegics.In:proceedingsofthe5thVienna
InternationalWorkshoponFES,1719August1995,Vienna,Austria.AustrianSociety
forArtificialOrgans,BiomaterialsandMedicalReplacementDevices;1995.
[28] HofferJ.A,BaruM,BedardS,CalderonE,DesmoulinG,DhawanP,JenneG,KerrJ,
Whittaker M, Zwimpfer T. J. Initial results with fully implanted NeurostepTM FES
systemforfootdrop.In:proceedingsof10thAnnualConferenceoftheInternational
FESSociety,Montreal,Canada:2005.
31
Chapter 3
High Knudsen Number Flow

Optical Diagnostic Techniques
Tomohide Niimi
1.Introduction
Alongwiththedevelopmentinmicroandnanotechnologywithsmallcharacteristiclength
and high technology under high vacuum environments, it has been strongly desired to
understandthermofluidphenomenaaroundamicroandnanosystemsuchasamagnetic
head slider of hard disk drive, micro thruster for micro satellite, semiconductor thin film
fabricationsystemandsoon.Incaseswherethecharacteristicdimensionoftheflowisofthe
orderofthemeanfreepath(about60nmattheatmosphericcondition),wehavetoanalyze
theflowsattheatomicormolecularlevel,becausethecontinuumapproachbecomesinvalid.
Theatomicormoleculargasflowhadbeenusedsynonymouslywiththerarefiedgasflowso
far,buthastobeappliedalsototheflowaroundthemicroandnanodevicesworkingunder
atmosphericconditionsasmentionedabove.WecalltheseflowswithlargeKnHighKnudsen
numberflows.Asanindicatorofrarefactionofgasflows,theKnudsennumber(Kn)isdefined
bytheratioofthemeanfreepath()dividedbythecharacteristicdimension(L)oftheflow.
TheflowregimesareclassifiedonascaleofKnnumber,asfollows.
Kn<103:Continuumflowregime
CompressibleNavierStokesequationwithnoslipboundarycondition
103<Kn<101:Slipflowregime
Compressible NavierStokes equation with velocity slip and temperature jump at the
boundary
101<Kn<10:Transitionflowregime
Boltzmannequation,whereintermolecularcollisionsshouldbetakenintoaccount
2013 Niimi; licensee InTech. This is an open access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use,
distribution, and reproduction in any medium, provided the original work is properly cited.
34
10<Kn:Freemolecularflowregime
Boltzmannequation,whereintermolecularcollisionsarenegligiblecomparedwithinter
actionsbetweenthegasmoleculesandthewalls
ForhighKnudsennumberflows,wehavetotakeintoaccountthefollowingsthatmaybe
neglected for the continuum flow regime. In the case of large , there appear the strong
nonequilibriumphenomenabecauseoffewintermolecularcollisions.ForextremelysmallL,
ontheotherhand,theflowfieldisstronglyinfluencedbyinteractionofmoleculeswithasolid
boundaryratherthanintermolecularcollisions.
ExperimentalanalysesofthermofluidphenomenarelatedtothehighKnudsennumberflows
needtheopticaldiagnostictechniquesbasedonatomsormolecules,suchastheiremission
andabsorptionofphotons.However,theexperimentaltechniquesarebehindondevelopment
comparedwiththemolecularsimulationtechniquessuchasthemoleculardynamicmethod
(MD),thedirectsimulationMonteCarlomethod(DSMC)andsoon.
In this chapter, the optical diagnostic techniques for the high Knudsen number flows are
mainlydescribed,suchaslaserinducedfluorescence(LIF),resonantlyenhancedmultiphoton
ionization (REMPI) and pressuresensitive molecular film (PSMF), and some experimental
resultsobtainedbytheuseofthetechniques,i.e.,applicationsofLIFtovisualizationofrarefied
gas flows including complicated shock wave system and to measurement of rotational
temperature,establishmentofaREMPIsystemanditsapplicationtodetectionofrotational
nonequilibriuminhighlyrarefiedgasflows,anddevelopmentofthePSMFformicrogasflow
measurements.
2.LaserInducedFluorescence(LIF)
Someatomsandmoleculesemitfluorescencewhenirradiatedbyalaserbeamwhosewave
lengthcorrespondstoanabsorptionlineofmolecules,asshowninFigure1.InFigure2[1],
LIFofiodinemolecules(I2)seededinargongaswasappliedtovisualizationofaflowfieldof
asupersonicfreejetissuedfromasonicnozzlewithexitdiameterof0.5mmintoavacuum
chamber,clearlyshowingthebarrelshockwavesandMachdisk.Figure3showsanexample
of the visualization of complicated flowfield structure and shock wave systems of two
interactingsupersonicfreejets[2]byusingI2LIF.NOLIFhasbeenalsoemployedtovisualize
thelowdensityhighspeedflows[3].
A method for planar measurement of rotational temperature using twoline I2LIF was
proposedfortherarefiedgasflows[4].Ifthefluorescenceintensitiesatapointintheflowfield
aredesignatedbyF1andF2whentheiodinemoleculesintherotationallevelsJ1andJ2are
excited,respectively,theratioF1/F2dependsonrotationaltemperatureandtworotational
quantumnumbers,J1andJ2.Therefore,oncetwoabsorptionlinesareselected,therotational
temperaturecanbededucedfromtheratioofthefluorescenceintensities.
Figures4aretypicalimagesofasupersonicfreejetvisualizedbytheuseofirradiationoflaser
beams with different wavelengths corresponding to the respective absorption lines. These
High Knudsen Number Flow Optical Diagnostic Techniques
Figure 1. Principle of LIF
Figure 2. Supersonic free jet visualized by I2-LIF [1]
Figure 3. Interacting supersonic free jets visualized by I2-LIF [2]
35
36
images are obtained at the same pressure condition: source pressure Ps = 16kPa and back
groundpressurePb=100Pa.Itcanbeseenthatthefluorescenceintensitydistributionsarevery
different,dependingontheabsorptionlines.Ifusingtwoimagesamongthem,wecandeduce
thetemperaturedistributiontwodimensionallyasshowninFigure5.
Figure 4. Supersonic free jets visualized by I2-LIF with different wavelengths. P and R means P- and R-branches, respec
tively, and the number the rotational energy level [4].
Figure 5. Two-dimensional temperature distribution of a supersonic free jet [4].
3.ResonantlyEnhancedMultiPhotonIonization(REMPI)
Tomeasuretherotationalpopulationinrarefiedsupersonicnitrogenfreejetsandfinallyto
confirm the nonBoltzmann distribution of the rotational levels experimentally, a REMPI
(Resonantly Enhanced MultiPhoton Ionization) method has been applied to detection of
nitrogenionsdirectlyasanioncurrent.REMPIisknowntohavehighdetectionsensitivity,
whichallowsobtainingthesignalundertheverylownumberdensitycondition.Herethe
principleofREMPIisintroducedandthe(2+2)N2REMPIisappliedtoasupersonicnitrogen
freejettodetectthenonBoltzmanndistributionoftherotationallevels.
Figure 6 depicts the schematic energy level diagram for (2+2) N2REMPI, illustrating the
relevantprocesses.Inthisprocess,nitrogenmoleculesatthegroundstate(X1g+)areexcited
totheresonantstate(a1g)bytwophotonabsorption.Thentheexcitedmoleculesareionized
byadditionaltwophotonenergy.Sincefourphotonsparticipateinthisprocess,theioncurrent
isproportionaltothefourthpoweroflaserfluxinprinciple.However,whenthelaserfluxis
sufficiently high so that almost all the excited molecules are ionized, the ion current is
proportional to square of laser flux, because the REMPI process reflects the twophoton
transitionprocessfromthegroundtotheresonantstate.Inotherwords,theREMPIspectrum
reflects directly the rotational population in the ground state of neutral molecules, so the
rotationaltemperaturecanbededucedfromtheREMPIspectra,providedthattheflowisin
equilibrium,thatis,therotationalenergydistributionfollowstheBoltzmanndistribution.
TheREMPItechniqueisappliedtodetectionoftherotationalnonequilibriuminanitrogen
freemolecularflow.Allexperimentsarecarriedoutinavacuumchamberevacuatedbya
turbomolecular pump and a dry pump as a backing pump, allowing an oilfree vacuum
environment. Nitrogen gas is issued from a sonic nozzle with D = 0.50mm diameter, and
expandedintothechamber.Stagnationtemperatureiskeptat293KandsourcepressuresP0
issetat30Torr(3.9103Pa)to1Torr(1.3102Pa).ANd:YAGpumpeddyelaseroperated
Figure 6. (2+2) N2-REMPI process.
37
38
withRhodamine6Gdyeisusedasalasersource,andtheoutputisfrequencydoubledbya
BBOcrystal.Thewavelengthofthelaserbeamisrangedfrom283to284.1nm.Thebeamis
focused with a quartz lens on a centerline of a nitrogen freemolecular flow. The ionized
nitrogenmoleculesaredetectedbyasecondaryelectronmultiplieroratungstenLangmuir
probe.Awavelengthisscannedbyastepof0.001nm,andthesignalintensityisintegrated
for100laserpulsespereachstep.
Figure7representsanexperimental(2+2)N2REMPIspectrumofthe(v,v)=(1,0)vibrational
band,measuredforP0D=15Torrmmandatafocalpointofx=2.5mmdownstreamfromthe
nozzleexit(x/D=5)alongthecenterlineofthejet.Inthisfigure,thehorizontalaxisindicates
thewavelengthofthelaser,andtheverticalonethesignalintensitynormalizedbyapeak.
NumbersonrulersinthefigurecorrespondtospectralpositionsforO,P,Q,RandSbranches.
WhenthepopulationnumberisdesignatedbyN(J),therotationallineintensityIJ,Jin(2+2)
N2REMPIspectraisgivenby
I J ,J = Ag ( J )S ( J , J ) N ( J ) / (2J + 1)
(1)
whereJandJaretherotationalquantumnumberoftheresonantandgroundstate,respec
tively,Aisaproportionalconstantindependentoftherotationalquantumnumber,andg(J)
isthenuclearspinstatisticalweightofnitrogenmoleculesformedbyN14atoms.S(J,J)isthe
twophotonHnlLondonfactor[5]forthea1g X1g+transition.N(J)isproportionalto
(2J+1) exp (Erot/kTrot) (Erot: rotational energy, Trot: rotational temperature, k: Boltzmanns
constant),providedthattherotationalenergydistributionfollowstheBoltzmanndistribution.
Inthiscase,IJ,Jisgivenby[6].
I J ,J = Ag ( J )S ( J , J )exp(E rot / kT rot)
(2)
and the rotational temperature can be deduced from a slope of Boltzmann plot of ln(IJ,J /
gS(J,J)) versus Erot/k. If there appears nonlinearity in the Boltzmann plot, therefore, the
rotational energy distribution deviates from the Boltzmann distribution and the rotational
temperaturecannotbedefined.
Figure8showsaresultofBoltzmannplotsatseveralx/DsforP0D=15Torrmmbyusing
the measured REMPI spectra. In this figure, the horizontal axis indicates theErot/k and the
vertical one the ln(IJ,J / gS(J,J)). Numbers attached to the data points are the rotational
quantumnumbersofthegroundstate.ItisfoundfromFigure8thatalltheBoltzmannplots
demonstratethenonlinearityevenatx/D=1.0forP0D=15Torrmmandespeciallythedata
inhigherrotationallevelsdeviatefromaline,confirmingthenonBoltzmanndistributionof
therotationallevels.
Figure 7. N2-REMPI spectrum [6].
Figure 8. Boltzmann plots [6].
39
40
TherotationaltemperatureisdeducedfromtheBoltzmannplotsofFigure8byusingonlythe
linearportionoftheplotslyingatsmallerrotationalquantumnumbers.Figure9showsthe
rotational temperature distribution along the centerline of a supersonic free jet, where the
horizontalaxisindicatesx/Dandtheverticalonetherotationaltemperature.Solidcircles
indicatethemeasuredrotationaltemperatures,andthebrokenlinesthetranslationaltemper
ature distribution calculated from isentropic relations using the Mach numbers [7]. The
rotationaltemperaturedistributionmeasuredbyREMPIiscomparedwithotherdatameas
uredbydifferentways,i.e.therotationaltemperaturesmeasuredbyusingonlysomerotational
linesconstitutingalinearportionoftheBoltzmannplotsobtainedbytheelectronbeammethod
[8] and calculated by using an equation of the rotational temperature distribution for the
rotationalrelaxation[9].Asyoucansee,allresultsagreewitheachother,ifusingonlythedata
atsmallerrotationalquantumnumbersfortheREMPIandelectronbeamtechniques.
Figure 9. Rotational temperature distribution along the centerline of a supersonic free jet [6].
To clearly reveal the nonBoltzmann distribution, the rotational population obtained by

REMPIforP0D=15TorrmmispresentedinFigure10,inwhichthehorizontalaxisindicates
therotationalquantumnumberandtheverticalonethepopulationratioofeachrotational
leveltothetotal.Symbolssuchassolidcirclescorrespondtothepopulationratioforeach
measurementpointandeachcurvetoaBoltzmanndistributionattemperaturededucedfrom
aBoltzmannplotusingonlythedataforsmallerrotationalquantumnumbersasmentioned
above.IntheconditionofP0D=15Torrmm,therotationalpopulationatx/D=1.0follows
almosttheBoltzmanndistribution,whereasatx/D=7.0theexperimentaldatadeviatefrom
theBoltzmanndistributionevidently.Comparingbetweenthedistributionsofx/D=7.0and
20.0,forJ7thebothshowthesametendency,suggestingpartialfreezingoftherotational
Figure 10. Rotational population obtained by REMPI [6].
population.Generallythehigherrotationalquantumnumberthemoleculeshave,thelower
thetransitionprobabilitybecomes,resultinginthedeviationfromtheBoltzmanndistribution
particularlyforthehigherrotationallevels.Thisisthereasonwhythefreezingoftherotational
populationstartspartiallyatthehigherrotationallevels.
4.PressureSensitivePaints(PSP)andPressureSensitiveMolecularFilm
(PSMF)
Therehavebeennoappropriatetechniquesforthemeasurementofgaspressureonasolid
surfaceinsidemicrosystems.Tomeasurepressuredistributionsinalowdensitygasflows
with high Knudsen number, a pressure sensitive paint (PSP) technique [10, 11] have been
adopted.
ThepressuremeasurementtechniqueusingPSPisbasedontheoxygenquenchingoflumi
nescentmolecules.PSPiscomposedofluminescentmoleculesandabindermaterialtofixthe
luminescentmoleculestoasolidsurface.Figure11depictstheschematicenergyleveldiagram
forPSPandoxygenmolecules.WhenPSPlayerappliedtothesurfaceisilluminatedbyUV
light, the luminescent molecules are excited by absorption of photon energy, and then the
moleculesemitluminescence.Ontheotherhand,oxygenmoleculewithtripletgroundstate
actsasaquencheroftheluminescence.Asaresult,thephosphorescenceintensitydecreases
asanincreaseinpartialpressureofoxygen.Pressureonthesolidsurfacecanbededucedfrom
therelationshipbetweenpressureandtheluminescenceintensity(SternVolmerplot[11]).
41
42
I ref
I
= An
n=1
( )
P n
Pref
(3)
whereIistheluminescenceintensityandPistheoxygenpressure.Irefistheluminescence
intensityataknownreferencepressurePref.ThecoefficientsAnaretheconstantscalledasthe
SternVolmercoefficientsdeterminedbycalibrationtests.TheluminescenceintensityIofthe
idealPSP(N=1)dependsinverselyonPfollowingtoEq.3,butactualPSPshavenonlinear
dependenceofI1onP.Inpractice,asecondorderpolynomial(N=2)iscommonlyused.
BecausePSPworksasasocalledmolecularsensor,itseemsalsosuitableforanalysesofhigh
Knudsennumberflows,whichrequirediagnostictoolsinthemolecularlevel.However,an
applicationofPSPtomicrodevicesisverydifficult,becauseconventionalPSPsareverythick
comparedtothedimensionofmicrodevicesowingtotheuseofpolymerbinders.Moreover,
theydonothavesufficientspatialresolutionforthepressuremeasurementofmicroflowsdue
totheaggregationofluminescentmoleculesinpolymerbinders[12].
Pressuresensitive molecular films (PSMFs) have been developed by using the Langmuir
Blodgett(LB)technique[13]tofabricatethinfilms,andtestedtoconfirmthefeasibilityofthe
pressuremeasurementaroundmicrodevices[14].PSMFwithnanometerorderthicknessand
highspatialresolutionissuitableforanalysesofmicroflows.
LBfilmsarefabricatedaccordingtothefollowingprocedure.First,adropofadilutesolution
ofamphiphilicmoleculesinavolatilesolventisspreadontheinterfacebetweenairandsub
phase.Afterthesolventisevaporated,amonolayerofthemoleculesremainsontheinterface.
Themonolayeristransferredtoasubstratewithcompressingthemonolayersoastocontrol
theorderofthemolecules.
In some cases, arachidic acid (AA) has been adopted as a spacer molecule to control the
intermolecularspacingofluminescentmolecules,andtoformastableLBfilmwithhighly
orderedstructure.However,itisdesirableforhighpressuresensitivitytoadjustthemolar
ratioofluminophoreandAA[14].
Figure 11. Schematic energy level diagram. S: Singlet state, T: Triplet state.
ThePSMFsamplewasappliedtoaglassslide,andthesamplewassetinsideavacuumchamber
evacuated by vacuum pumps to examine the pressure sensitivity. Pure oxygen gas was
suppliedintothechamber,andthepressureinthechamberwasmonitoredbyacapacitance
manometer.ThetemperatureofthesamplewascontrolledbyaPeltierelement.Axenonarc
lampwithabandpassfilter(40020nm)wasusedasanexcitationlightsourceilluminating
thesampleviaanopticalfiber.Theluminescencewasfilteredbyalongpassfilter(600nm)to
eliminatethelightfromthexenonarclamp,andwasdetectedbyanimageintensifiedCCD
camera(512512pixels,14bit).
ThePSMFbasedonPd(II)MesoporphyrinIX(PdMP),whichisusedasluminescentmolecules,
hasbeenfabricatedanditisclarifiedthatthepressuresensitivityofPSMFofPdMPissuffi
cientlyhigh[14].However,thePSMFcomposedofPdMPcannotbeapplicabletothepressure
rangehigherthan130PaduetosaturationofoxygenquenchingofPdMP;thisiscausedby
thelonglifetimeofPdMP,whichissolong(1.0msec[15])thatmostofluminescentmolecules
arequenched.ItisdesiredforPSMFtoworkaroundanatmosphericpressure,becausethe
mostmicrodevicesareusedinanatmosphericpressure.InordertofabricateausefulPSMF
foranatmosphericpressurerange,foursamplesofPSMFsareprepared,composedofPt(II)
MesoporphyrinIX(PtMP),Pt(II)MesoporphyrinIXdimethylester(PtMPDME),Pt(II)Proto
porphyrinIX(PtPP)andCu(II)MesoporphyrinIXdimethylester(CuMPDME).Thoselumi
nescentmoleculeshaveshorterlifetimecomparedwithPdMP(e.g.thelifetimeofPtMPDME
andCuMPDMEare0.14msecand0.1msec,respectively[15]).Figure12showsthepressure
sensitivitiesoffourPSMFsbelow21kPa(equaltothepartialpressureofoxygeninatmospheric
pressure)[16].ThehorizontalaxisoftheSternVolmerplotisthenormalizedpressureP/Pref
andtheverticalaxisistheinverseluminescentintensityratioIref/I,whereIrefisthereference
luminescentintensityatthereferencepressurePref=1.0102Pa.Itisclarifiedthatthepressure
sensitivity of PtMP is highest among the four tested PSMFs and is comparable to that of
conventionalPSPs.ThetemperaturedependencyofPSMFalsohastobestudied,becausethat
ofPSPisthemainfactorofthemeasurementerror.PSMFwasappliedtothemeasurementof
microgasflows[17].
Figure 12. Stern-Volmer plots of PSMFs [16]
43
44
Theflowinaconvergingdivergingnozzlewasexaminedforthedemonstration,becausethe
pressure considerably changes in a nozzle and it was easy to compare the result with a
numericalone.
Figure 13 shows the geometry of a convergingdiverging micronozzle, which was two
dimensionalgeometry,andthethicknessofthemicronozzleswas1.0mm,andthewidthof
thethroatwas103m,thediverginglengthwas492mandthediverginganglewas23.5
degrees.Themicronozzlewasputonaholderwithaninletandoutletport,andthenitwas
coveredwiththePSMFdepositedglass(seeFigure13b).Airwassuppliedfromtheinletport,
andtheoutletportwasevacuatedbyascrollpump.PtMPwasemployedasaluminophoreof
PSMF.LuminescentintensityofPSMFwasdetectedthroughafluorescentmicroscope.The
imagesweretakenwiththeexposuretimeof8.0s,and32pictureswereaveragedtoreduce
thenoise.
Figure 13. Micro-nozzle.
Figure 14. Pressure distribution inside the micro-nozzle [17].
ThemicronozzleflowwasanalyzedbythedirectsimulationMonteCarlo(DSMC)method
[18], which is a popular method for high Knudsen number flow analysis. The numerical
conditionswereadjustedtothePSMFexperimentalconditions.
Figure14showsthepressuredistributionofthemicronozzlewiththeinletpressurePin=10.0
kPaandtheoutletpressurePout=1.0kPa,wherethewindoffimagesweretakenatPin=Pout=
1.0kPa.ThenumericalresultbyDSMCisalsoshowninthelowersideofFigure14.Moreover,
thepressureprofilesalongthecenterlineareshowninFigure14b.Thoughthepressureprofile
containssomenoiseasshowninFigure14b,itisclearlyobservedthatthepressuredistribution
obtainedbyPSMFisquantitativelyingoodagreementwiththatbyDSMC.Especially,the
sharppressuredropnearthethroatwascapturedbyPSMF,andthisresultindicatesthatthe
spatialresolutionofPSMFishighenoughformicroscalemeasurement.Thepressurevalue
at300mfromthethroatisabout8kPaasshowninFigure14bandissmallerthanPin=10.0
kPa,whichwasmeasuredat10mm.Thispressuredropmaybecausedbywallfrictionand
theaccelerationattheconvergingareaofthenozzle.
Authordetails
TomohideNiimi
DepartmentofMicroNanoSystemsEngineering,NagoyaUniversity,Nagoya,Japan
References
[1] Fujimoto T, Niimi T. Three Dimensional Structures of Interacting Freejets. Rarefied
GasDynamics1988,SpaceRelatedStudies,AIAA;391406.
[2] NiimiT,FujimotoT,TaoiN.FlowFieldsofInteractingParallelSupersonicFreeJets.
JSMEInternationalJournalSeriesB1996;39(1)95100.
[3] Mori H, Taniguchi M, Nishihira R, Niimi T. Experimental Analyses of LinearType
AerospikeNozzleswithandwithoutSidewalls.AIAApaper2005;1350.
[4] Niimi T, Fujimoto T, Shimizu N. Planar Measurement of Temperature in Rarefied
GasFlowbyLIFImages.Proceedingsof17thInternationalSymposiumonRarefied
GasDynamics1991:14821489.
[5] Bray R. G, Hochstrasser R. M. (1976). Twophoton absroption by rotating diatomic
molecules.MolecularPhysics1976;31(4)11991211.
45
46
[6] MoriH,NiimiH,AkiyamaI,TsuzukiT.ExperimentalDetectionofRotationalNon
Boltzmann Distribution in Supersonic Free Molecular Nitrogen Flows. Physics of
Fluids2005;17(11)117103.
[7] AshkenasH,ShermanF.S.Thestructureandutilizationofsupersonicfreejetsinlow
densitywindtunnels.Proceedingsof4thInternationalSymposiumonRarefiedGas
Dynamics1966;284105.
[8] MarroneP.V.Temperatureanddensitymeasurementsinfreejetsandshockwaves.
PhysicsofFluids1967;10(3)521538.
[9] Gallagher R. J, Fenn J. B. Relaxation rates from time of flight analysis of molecular
beams.JournalofChemicalPhysics1974;60(9)34873491.
[10] Liu T, Champbell T, Burns S. P, Sullivan J. P. Temperature and pressuresensitive
paintsinaerodynamics.AppliedMechanicsReviews1997;50227246.
[11] BellJ.H,SchairerE.T,HandL.A,MahtaR.D.Surfacepressuremeasurementsusing
luminescentcoatings.AnnualReviewofFluidMechanics2001;33155206.
[12] MoriH,NiimiT,HirakoM,UenishiH.Molecularnumberfluxdetectionusingoxy
gensensitiveluminophore.PhysicsofFluids2005;17100610.
[13] Ulman A. An Introduction to Ultrathin Organic Films From LangmuirBlodgett to
SelfAssembly.SanDiego:AcademicPress;1991.
[14] MatsudaY,MoriH,NiimiT,UenishiH,HirakoM.Developmentofpressuresensi
tive molecular film applicable to pressure measurement for high Knudsen number
flows.ExperimentsinFluids2007;42543550.
[15] Becker R. S. Metalloporphyrins. Electronic spectra and nature of perturbations. I.
Transition metal ion derivatives. The Journal of Physical Chemistry 1963;67(12)
26622669.
[16] MatsudaY,MoriH,SakazakiY,UchidaT,SuzukiS,YamaguchiH,NiimiT.Exten
sionandcharacterizationofpressuresensitivemolecularfilm.ExperimentsinFluids
2009;47(6)10251032.
[17] MatsudaY,UchidaT,SuzukiS,MisakiR,YamaguchiH,NiimiT.Pressuresensitive
molecular film for investigation of micro gas flows. Microfluidics and Nanofluidics
2011;10(1)165171.
[18] Bird G. A. Molecular Gas Dynamics and the Direct Simulation of Gas Flows, New
York:OxfordUniversityPress;1994.
47
Chapter 4
Precision Micro Machining Methods and

Mechanical Devices
Eiji Shamoto, Norikazu Suzuki, Takashi Kato and

Burak Sencer
1.Ultraprecision/micromachiningofdiesteelbyellipticalvibration
cutting[1]
Anovelmethodtoattainultraprecisionsculpturinginmicro/nanoscalefordifficulttocut
materialsisintroduced.Ellipticalvibrationcuttingtechnologyiswellknownforitsexcellent
performance in achieving ultraprecision machining of steel materials with single crystal
diamond tools. Elliptical vibration locus is generally controlled and held to a constant in
practice.Onthecontrary,theproposedmethodutilizesthevariationsoftheellipticalvibration
locusinapositivemanner.Depthofcutcanbeactivelycontrolledinellipticalvibrationcutting
bycontrollingvibrationamplitudeinthethrustdirection.Byutilizingthisasafasttoolservo
function in elliptical vibration cutting, high performance micro/nano sculpturing can be
attainedwithoutusingconventionalfasttoolservotechnology.Followingsectionsdescribe
thedevelopmentofthehighperformancemicro/nanosculpturingsystemandultraprecision/
micromachiningapplications.
1.1.Introduction
Theauthorshavedevelopedellipticalvibrationcuttingtechnology[2],andhavedemon
stratedthatultraprecisionmachiningofdifficulttocutmaterials,suchashardenedsteeland
hard/brittlematerials,canbeattainedpracticallybyapplyingultrasonicellipticalvibrationto
singlecrystaldiamondtools[3].Severalultrasonicellipticalvibrationtoolsandtheircontrol
systemswerealsodevelopedinthepaststudies[4].Sincevariationofvibrationamplitudes
causesdeteriorationofmachiningaccuracyandsurfacequality,mostresearcheffortswere
dedicatedtokeepingtheellipticalvibrationlocusultrapreciselyconstant.Otherwise,ultra
precisioncuttingcannotbeachievedinpractice.
2013 Shamoto et al.; licensee InTech. This is an open access article distributed under the terms of the
50
Ontheotherhand,theauthorsfocusonutilizingthevariationinvibrationamplitudeina
positivemanner,incontrastwithconventionalstudies.Itisconsideredasauniquefunction,
i.e.,thedepthofcutcanbeactivelycontrolledbycontrollingthevibrationamplitudeinthe
depthofcutdirectionwhilemachining.Byutilizingthisfunctiontoserveasasortoffasttool
servo(FTS),theultraprecisionsculpturingofdifficulttocutmaterialsinmicro/nanoscaleis
achievedefficiently.Notethatitisredundantanddisadvantageoustocombinetheelliptical
vibrationtoolwiththeconventionalFTS,sincebothdeviceshaveactuatorsandthevibration
toolistooheavytobeactuatedathighfrequencybytheFTS.
Followingsectionsintroducetheproposedmachiningmethodandavibrationcontrolsystem
ofatwodegreeoffreedom(2DOF)ellipticalvibrationtool,whichenablespreciseamplitude
controlofultrasonicellipticalvibrationathighspeed.Subsequently,experimentalverifica
tionsintexturedgroovingandinnanosculpturingaredescribed.
1.2.Ultraprecisionmicromachiningmethodfordifficulttocutmaterials
Figure1showstheproposedmachiningwithdepthofcutcontrolinellipticalvibrationcutting.
Thetoolisfedatanominalcuttingspeedandvibratedellipticallyatthesametime.Because
ofthisintermittentprocessatanultrasonicfrequency,toolwearandadhesionarerestricted,
andtheultraprecisioncuttingofhardenedsteelcanbeattainedwithsinglecrystaldiamond
tools.Thevibrationamplitudeinthedepthofcutdirectioniscontrolledsimultaneouslyinthe
proposedmachiningprocessasshowninthefigure.Thetrajectoryofthecuttingedge,then,
changesdynamically,anditsenvelopeistransferredtothefinishedsurface.
Figure 1. Machining by controlling amplitude in elliptical vibration cutting
Bycontrollingtheamplitudeultrapreciselyathighspeed,theultraprecisionsculpturingof
the difficulttocut materials can be achieved efficiently without using conventional FTS
technology[5].Inotherwords,theellipticalvibrationcuttingtechnologyisequippedwitha
FTSfunctionbyitself.Althoughamplitudecontrolcommandisnotidenticalwiththeenvelope
ofthecuttingedgetrajectory,asshowninFigure1,theirdifferenceisnotcrucialtothepresent
Precision Micro Machining Methods and Mechanical Devices
study.Thedifferenceisinsignificantinpracticewhentheslopeisnotsteep.Thedepthofcut
canbecontrolledwithinhalfofthemaximumamplitudeinthedepthofcutdirection,and
availablefrequencyrangeoftheamplitudecontrolislimitedtothatwhichisrelativelylower
than the elliptical vibration frequency. Therefore, performance in the role as FTS strongly
dependsonthespecificationsofthevibrator.
1.3.Ellipticalvibrationtoolandhighspeedamplitudecontrolsystem
Figure2showsthedevelopedsystemofthehighspeedamplitudecontrolofellipticalvibration
atafrequencyofabout36kHz.Atwodegreeoffreedom(2DOF)ellipticalvibrationtool[6],
whichwasdesignedtogeneratearbitraryellipticalvibration,isutilizedinthepresentstudy.
The vibrator is actuated by using some PZT actuators, which are sandwiched with metal
cylindricalparts,namelyaboltclampedLangevintypetransducer(BLT).Asthevibratoris
designedtohavesameresonantfrequenciesinsecondresonantmodeoflongitudinalvibration
andfifthresonantmodeofbendingvibration,itcangeneratelargelongitudinalandbending
vibrationssimultaneouslyatthesameultrasonicfrequencybyapplyingexcitingvoltagesto
theactuators.Thus,anarbitrary2DOFellipticalvibrationcanbeobtainedatthediamondtool
tipattachedtothevibratorbycombiningbothresonantvibrationmodeswithsomephaseshift.
Figure 2. Elliptical vibration control system and frequency response of amplitude control
Gainoftheamplifiercanbecontrolledbyexternalinputinthedevelopedsystem,andthus
the exciting voltage supplied to the actuator is changed. The amplitude is, consequently,
controlledbytheexternalinput.Asthemaximumamplitudeinthedepthofcutdirectionis4
mpp,thevibrationamplitudecanbecontrolledtochangethedepthofcutwithin2mby
this system. Measured frequency response of amplitude control is shown in Figure 2. The
developedsystemisabletocontrolthevibrationamplitudewithafrequencybandwidthof
morethan300Hz.Thisfrequencybandwidthisrelativelynarrowascomparedwiththatof
51
52
conventional FTS. It might not, however, be a big problem because the elliptical vibration
cuttingtechnologyisavailableonlyatrelativelylowcuttingspeed.
1.4.Ultraprecisionmicro/nanosculpturingoftexturedgrooves
Thedevelopedcontrolsystemwasappliedtogroovingexperiments.Theultrasonicelliptical
vibrationtoolwasmountedonanultraprecisionplaningmachine,NIC300(NagaseIntegrex
Co.,Ltd.),andwasfedstraightforgrooving.Thevibrationamplitudewascontrolledwith
sinusoidalandzigzagwavecommandsatthesametime,andthen,microtexturedgrooves
were formed on the surface of a hardened steel workpiece. The vibration amplitude was
controlledtochangefrom2mppto4mpp.Thiscorrespondstothedepthofcutvariationof
1m.Ontheotherhand,theamplitudeinthecuttingdirectionissettobeconstant,4mpp.
Figure3showsphotographsofgroovesmachinedatthecuttingspeedof0.2m/minbythe
singlecrystaldiamondtoolwithanoseradiusof1mmandtheirsurfaceprofilesmeasuredby
alasermicroscope,VK9500(KeyenceCorp.).Itwasconfirmedthatthegrooveswithvarious
ultraprecisionmicrotexturescanbemachinedsuccessfullyonthehardenedsteel,andmirror
surfacequalitycanbeobtainedonallgrooves.Measuredsurfaceprofilesofsinusoidalgrooves
agreedpreciselywiththecommandwaves.Ontheotherhand,measuredcornersofzigzag
groovesarerounded,andtheirstepheightswererelativelysmallerthanthevariationwidth
ofcommandwavesof1m.Thisisconsideredtobecausedbycuttingoffhighfrequency
componentsintheamplitudecontrolcommandsbytheLPF,whichthezigzagwavesinclude
attheirsharpcorners.
(a)texturedgroovewithsinusoidalcommand
(b)texturedgroovewithzigzagcommand
Figure 3. Microphotographs and surface profiles (cutting speed: 0.2 m/min, freq.: 100 Hz)
1.5.Ultraprecisionmicro/nanosculpturingonplanesurfaces
Inordertoattainarbitrarysculpturing,anultraprecisionmicro/nanosculpturingsystemwas
developedbyusingthedevelopedvibrationcontrolsystemandtheultraprecisionplaning
machine.Figure 4 shows the developed sculpturing system, where the planing machineis
simplycontrolledtomachineaplanesurfaceatconstantcuttingspeed.Theultrasonicelliptical
vibrationtoolisattachedonaZaxistableofthemachinetool.Thevibrationamplitudeinthe
depth of cut direction along the Zaxis is controlled in synchronization with cutting feed
motionintheXaxis,andthen,arbitrarymicro/nanosculpturingcanbeattainedonaflattop
surfaceofasteelworkpiecebymerelycombiningsimpleplaningoperationsatconstantcutting
speedwithhighspeeddepthofcutcontrol.Anindustrialcomputerisutilizedtodetectthe
coordinatepositionsandgeneratethevoltagesignalforcontrollingvibrationamplitude.The
coordinatevaluesareconstantlymonitoredinthedevelopedsystembydirectlycommunicat
ingwithNCcontrolsystemorbyusinganexternalopticalsensor.Thedynamiccommand
signalforvibrationamplitudecontrolisgeneratedtochangethedepthofcutinaccordance
withtheCADdata.
The developed machining system was applied to nano sculpturing experiments of picture
images. CAD data for sculpturing were produced from gray scale images, where the gray
valuesof8bits(256gradations)inpixelswereconvertedintotheamplitudecommands.As
the vibration amplitude in the experiments was changed within a range from 2 mpp to 4
mpp,thedepthofcutwaschangedwithin1m.Theresolutionofthedepthofcutcontrolis,
therefore,about4nm.Hardenedsteelworkpieces(64x48mm,JIS:SUS420J2,HRC53)were
machinedwithsinglecrystaldiamondtoolswithanoseradiusof1mm.Thesizeoforiginal
imagesissetto3200x2400pixels,andthus,1pixelcorrespondsto20x20m.
Figure 4. Developed micro/nano sculpturing system using elliptical vibration cutting
Figure5showsanexampleofthenanosculpturingofpictureimages.Thedepthofcutwas
controlledinnanoscaleinaccordancewiththeimagedata.AsshowninFigure5,thegray
scalepictureimage,themicrophotographofthemachinedsurfaceanditsprofile,whichwas
measuredbyopticalsurfaceprofiler(ZYGONewView6200),correspondwelltoeachother.
53
54
Theresultshowsthatpictureimagescanbeprintedsuccessfullyonhardenedsteelasnano
scalesculptures.
(a)grayscaleimage (b)microphotograph
(c)measuredprofile
Figure 5. Gray scale image for amplitude command, machined surface and measured surface profile
Nano sculpturing experiments involving dimple patterns were also carried out with the
developedsculpturingsystem.Sinusoidalcommandstocontrolthevibrationamplitudewere
input to the elliptical vibration control system during machining, and the phase of the
sinusoidal commands was changed by 180 degrees in every cutting feed, so that precisely
alignedpatternsweresculptured.
Figure6showsmicrophotographs.Thehexagonaldimplepatterns,whosebordersaresharp,
canbeobservedontheleft.Ontheotherhand,isolatedcirculardimplepatternswerealso
sculpturedsuccessfullyontheright,asthemaximumdepthofcutwasconsiderablysmaller
thantheamplitudevariation.Theresultsshowthatavarietyofdimplepatternscanbeobtained
ultrapreciselyonthesteelmaterialsbyusingthedevelopedsculpturingsystem.
1.6.Conclusion
Novelultraprecisionsculpturingtechnologyfordifficulttocutmaterialsatmicro/nanoscale
wasproposedbyutilizingellipticalvibrationcuttingtechnology.Intheproposedmethod,the
depthofcutiscontrolledwithouttheconventionalFTStechnologybyactivelymanipulating
thevibrationamplitudeinthedepthofcutdirection.Inordertoverifytheproposedmethod,
the vibration amplitude control system and a high performance micro/nano sculpturing
systemweredevelopedandappliedtosculpturingexperimentsonhardenedsteel.Conse
quently, micro textured grooves, an image of a picture and various dimple patterns were
manufacturedonthehardenedsteelworkpiecesuccessfullyasnanoscalesculptures.These
weredonebymerelycombiningasimplefeedmotionataconstantspeedwithhighspeed
depthofcutcontrol.
Figure 6. Microphotographs of sculptured dimples
2.Analyticalpredictionofchatterstabilityinballendmillingwithtool
inclination[7]
Anewanalyticalmodeltopredictchattervibrationinballendmillingwithconsiderationof
toolinclinationisintroducedinthissection.
2.1.Introduction
Theballendmillingisanimportantprecisionmachiningprocess,whichisusedinproduction
ofdies,molds,impellers,screwpropellersandpartswithfreeformsurfaces.Astheslender
ballendmillsorthethinworkpiecestructuresareflexible,theselfexcitedchattervibration
oftenoccursandcausessevereproblemssuchasshorttoollifeanddeteriorationofsurface
quality.Manyresearchersinvestigatedpredictionoftheselfexcitedchattervibrationinthe
millingprocess.AltintasandBudak[8]developedananalyticalmodeltosolvethechatter
stabilityforthecylindricalendmills.Altintas,Shamoto,LeeandBudakextendedtheanalytical
modelforthecylindricalendmillstopredictthestabilitylimitsinballendmilling[9],butthe
toolinclinationhasnotbeenconsideredduetogeometriccomplexityoftheballendmilling
process.
On the other hand, 5axis machining technology has been widely spread recently, and it
enablesarbitrarytoolinclinationforbettermachiningefficiency,accuracyandstability.
Therefore, an analytical model of the ball end milling process with the selfexcited chatter
vibrationisdevelopedandverifiedwithconsiderationofthetoolinclination.
2.2.Outlineofanalyticalmodeltopredictchatterstabilityinballendmillingwithtool
inclination
TheballendmillingprocesswiththetoolinclinationisillustratedinFigure7.TheCartesian
coordinatesxyzarefixedtotheballendmillandalignedwiththeworkpiececoordinatesuvw
beforetheinclination.Theoriginsofboththecoordinatesystemsareplacedatthepresentball
55
56
center.u,vandwarethecuttingfeeddirection,thepickfeeddirectionandthenormaldirection
tothefinishedsurfacerespectively.Thetoolisinclinedaroundxbyixandthenaroundyby
iy,asshowninthefigure.Theskyblueregionshowsengagementoftheballendmillwiththe
workpiece.Ifvibrationoccursintheprocess,itchangestheuncutchipthicknessandgenerates
thedynamiccuttingforceinthisengagementregion.Thisdynamiccuttingforceexcitesthe
mechanicalstructureandgeneratesvibrationagain.
Figure 7. Ball end milling process with tool inclination.
This closed loop system can be expressed by the block diagram shown in Figure 8. If the
vibrationgrowsupthroughthisclosedloop,thesystemisunstableandgeneratestheself
excitedchattervibration.Thediagramcontainstwoselfexcitationmechanisms,i.e.regener
ation and mode coupling. Note that the original or static milling forces and the forced
vibrations, which change in synchronization with the tooth passage, are neglected in the
presentstudy,becausetheydonotbasicallyaffecttheselfexcitedchattervibration.
Intheballendmillingprocess,mostofthesurfaceremovedbythepresenttoothhasbeencut
bytheprevioustooth,asshowninthexycrosssectioninFigure7,andthusthepresentuncut
chipthicknessfluctuatesbynotonlythepresentvibrationbutalsothepreviousvibration.This
regenerativeeffectisrepresentedgenerallybythedelaytermeTs,whereTisthetoothpassing
period.Theoverlapfactorcanbeapproximatedby1,sincethefeedrateisusuallysmall
enoughcomparedwiththeballradiusr.
Themultidirectionalvibrationscanbecoupledwhentheirmodesindifferentdirectionshave
closeresonantfrequencies.Thepresentstudy,therefore,dealswiththexyzvibrationsandtheir
modecouplingeffectasshowninthediagram.
The transfer functions are supposed to be measured, while the directional milling force
coefficients,whicharecalledprocessgainsinthepresentpaper,needtobecalculatedforthe
predictionofthechatterstability.Theprocessgainsaretimedependentinthemillingprocess,
but it is known that they can be approximated by their averages, i.e. DC components [8].
Therefore,theaverageprocessgainsarederivednumericallybyassumingsmalldisplacement
ineachdirection[7].
Regenerative x
displacement y
z
Ball end milling

process gain
Pxx Pxy Pxz
Pyx Pyy Pyz
Pzx Pzy Pzz
Present dynamic
displacement
Transfer function
Gxx (s) Gxy (s) Gxz (s)

Gyx (s) Gyy (s) Gyz (s)
Gzx (s) Gzy (s) Gzz (s)
Delay
e Ts
Previous dynamic
displacement
Dynamic
cutting force
fx
fy
fz
Figure 8. Block diagram of ball end milling process with regenerative and mode-coupling chatter vibrations.
2.3.Experimentalsetupandmeasurement
Themachiningexperimentswereconductedontheinclinedworkpiecesurfacewithaslender
ballendmillandaverticalmachiningcenter(Millac3VADS,OkumaCorp.),asshownin
Figure9.Thevibrationwasmeasuredatthetoolholderwiththedisplacementsensor.The
transfer function matrix was measured by the impulse response method, and the specific
cutting forces, which are required to compute the process gains, are identified using a
dynamometerandanobliquecuttingmodel[10].
2.4.Analyticalandexperimentalresults
Figure10showsthepredictedgainmarginsgm,thepredictedstabilitylimits,i.e.contourlines
of gm=1, and experimental results of chatter vibrations. The experimental vibrations were
classifiedbyusingthechatterfrequencycomponentofthemeasureddisplacements0,asshown
57
Spindle
Ball end mill

Workpiece
Displacement
sensor
z
y
Machining
center
Figure 9. Ball end milling experiment with tool inclination.
underthefigure.Forexample,severechatterwasdetectedatd>2mmandix=10deg,wherethe
measureddisplacementhasalargechatterfrequencycomponentofs0>1.3m.Ontheother
hand, only the spindle speed harmonics were observed at d=2.5 mm and ix=45 deg. The
analyticalandexperimentalresultsareallinagoodagreementasshowninthefigure.There
werenochattervibrationsatgm>1.2,whileseverechattervibrationsweredetectedatgm<0.6.
4
3
Predicted
stability limit
gm=1
10
1
0
-80
-60
-40 -20
0
20
40
60
Tool inclination angle ix deg
80
Predicted gain margin gm
100
Depth of cut d mm
58
0.1
Figure 10. Predicted gain margins and chatter stability limits, and experimental results at varied tool inclination angle
ix. Workpiece: aluminum alloy (JIS:A5052); Cutter: HSS ball end mill (EBD80820, OSG Corp.), Cutter: number of flutes nf
= 2, radius r = 10 mm; Cutting conditions: iy = 0 deg, pick feed p = 1 mm, n = 6240 min-1, feed rate of 0.01 mm/tooth;
Cutting fluid: soluble. Chatter vibrations were classified as follows; : no chatter (s0 0.12 m), : slight chatter (0.12
m < s0 0.24 m), : chatter (0.24 m < s0 1.2 m), *: severe chatter (1.2 m < s0).
2.5.Conclusion
Theanalyticalmodeloftheballendmillingprocesswiththeselfexcitedchattervibrationwas
developedwithconsiderationofthetoolinclination,anditwasappliedtopredictthechatter
stabilityatvariedspindlespeedandtoolinclinationinthepresentresearch.Thepredictedchatter
stabilityagreedwellwiththeexperimentalresults,anditisexpectedthatthepresentanalysis
willcontributetoimprovemachiningefficiency,surfacequalityandtoollifeinthefreeform
machiningwiththeflexiblestructureslikeslenderballendmillsandthinworkpieces.
3.Anewfluidbearingutilizingtravelingwaves[11]
A new noncontact fluid bearing utilizing traveling waves is proposed in this research.
Conventional hydrostatic bearings utilize externally compressed fluid, which requires
plumbing and compressors. In contrast, on proposed bearing system the moving part is
supportedwithathinfluidfilmcompressedbythewavestravelingradiallyonthebearing
surface. The proposed bearing realizes noncontact smooth motion without such a large
apparatus, and furthermore it has a capability to electrically control the bearing force or
clearance.
3.1.Introduction
Bearingsareoneofthemostfundamentalandnecessaryelementsformachinestogenerate
motionbetweentosurfacessmoothlywithlowfriction.Therehavebeenconstantdemands
forlowfrictionornoncontactbearingsformanyyears[1214].
Asfornoncontactbearings,L.D.Girardinventedahydrostaticbearingin1865[15].Anactive
magneticbearingwasinventedaroundthesametime,anditsactualsystemwasdeveloped
asearlyas1950[16].Thefeasibilityofasqueezebearingwasfirstdemonstratedandreported
in1964[17].Thesenoncontactbearingshavebeenimprovedandutilizedinpracticeaccording
totheirdifferentcharacteristics.However,itseemsthatnofundamentalprinciplesofnon
contactbearingshavebeenproposedafterthoseinventions.
Anewprincipleofnoncontactfluidbearings,whichutilizestravelingwaves,isproposedin
thisresearch,andaprototypedeviceisdevelopedonthebasisoftheproposedprinciple.
3.2.Workingprincipleofthenewfluidbearingsystem
Accordingtothisnewprinciplethenoncontactfluidbearingisrealizedbygeneratingthe
travelingwavesradiallyonthebearingsurfaceasshowninFigure11.
TheworkingprincipleofthefirstprototypeisillustratedinFigure11a.Threesetsofpiezo
actuatorsareplacedradiallyaroundthebearingsurface,andsinusoidalvoltageisappliedto
theactuatorswithaphaseshifttogeneratetravelingwavesontheflexiblesurface(Figure
11b).Asthewavesaregeneratedradially,thefluidistransportedfromoutsidetothecenter,
which in return generates pressure and a floating force to support an object. When the
amplitudeandfrequencyofthevoltageareincreased,morefluidispumpedunderthebearing
surfaceallowinghigherloadstobesupported.
Thefollowingkeypropertiesshowthatthisnewbearingdesignispromisingforhighprecision
applications:
59
60
i)Itisverycompact,doesnotrequireanypumporbulkytubestosupplyair,ii)thebearing
gapcanbecontrolledbyadjustingdrivingvoltages,iii)ithasanaturalresistancetomoment
offorcewithasinglepad,andiv)canbeusedduringhighspeedmotion.
Figure 11. Principle of proposed fluid bearing a) Working Principle b) Traveling Wave
3.3.Evaluationofthedevelopedbearingsystem
In order to study the properties of the proposed bearing system a prototype bearing is
developedwithrespecttothebearingstructureillustratedinFigure11.Figure12showsthe
prototypebearingstructurefloatingontheguidesurface.Atmosphericairisutilizedasthe
bearingfluidinthisprototype.
Developed device
Guide surface
Figure 12. Prototype fluid bearing system
Thedevelopedfluidbearingiscontrolledwithapiezodrivingapparatus,anditisconfirmed
thatthedevicecanbefloatedsuccessfullyandcanmovesmoothlywithalmostnofrictionlike
ahydrostaticairbearing.Theperformanceofthedevelopedbearingisevaluatedexperimen
tallyinthefollowing.
3.3.1.Travelingwaveonthebearingsurface
Displacementdistributiononthebearingsurface,anditstransientchangearemeasuredwith
anopticalfibersensorinthecirculardirection.Themeasureddisplacementsaresynchronized
Displacement Pm
Displacement Pm
byconsideringphaseshiftsfromtheappliedvoltages,andonecycleoftheirtransientchange
isshowninFigure13.Itshowsthatthetravelingwaveisgeneratedwithamplitudeofabout
6manditisabsorbednearthecenter.Asshown,anearlyperfectwavetotransportthefluid
couldbegeneratedonthebearingsurface.
Distance
from center
mm
Distance
from center
mm
Distance
from center
mm
b) 450
Displacement
Pm
Displacement Pm
a) 00
Distance
from center
mm
Distance
from center
mm
Distance
from center
mm
Distance
from center
mm
d) 1350
Displacement
Pm
Displacement Pm
c) 900
Distance
from center
mm
Distance
from center
mm
Distance
from center
mm
Distance
from center
mm
f) 2250
Displacement
Pm
Displacement Pm
e) 1800
Distance
from center
mm
Distance
from center
mm
g) 2700
Distance
from center
mm
Distance
from center
mm
h) 3150
Figure 13. One cycle of traveling wave (Conditions: Freq.: 100Hz, Voltage: 80Vp-p)
Distance
from center
mm
61
62
3.3.2.Maximumloadcapacity
Utilizingtheabovetravelingwavemethod,maximumloadcapacityofthebearingismeasured
ontheprototypebearing.Weightsarestackedonthecenterofthedevice.Criticalweightload
betweenthecontactandthenoncontactisrecordedasthemaximumloadcapacity.Figure
14showsthemaximumloadcapacitymeasuredatvariousamplitudesandfrequenciesofthe
drivingvoltage.Themaximumloadcapacityisincreasedastheamplitudeandthefrequency
areincreased.Thistendencyisthesameasthatofthefloatingdisplacement,buttheeffectof
thefrequencyisrelativelysmallinthiscase.Themaximumloadcapacityofabout100Ncan
beobtainedat80Vpp.
Figure 14. Maximum load capacity under various driving conditions.
3.3.3.Evaluationofvariousproperties
Other properties of the bearing system are also studied experimentally. An experiment to
producetheattractiveforceisalsocarriedoutbyreversingthetravelingwaves.Itisconfirmed
thatthedevicecanbeattractedtotheguidesurface.Forexample,thenegativegaugepressure
measuredwiththepressuresensorisabout3.6kPaatthecenterunderconditionsof80Vp
pand50Hz.
Moment of force is applied to the device, and it is confirmed that the present device has
resistance to the moment of force without using plural bearing pads. For example, the
developeddeviceisfloatedevenwhenaweightloadof30Nisappliedonitataneccentric
position,whichis40mmawayfromthecenter.Thisisconsideredtocausedbyviscosityof
theair,i.e.theairpressuredistributioncanbeasymmetrictoproducetheresistanceforceto
themoment.
Vibrationofthesupportedobjectisevaluatedbyusingtheopticalsensor.Forexample,the
supportedobjectisvibratedverticallywithanamplitudeof0.66matthecenter,whenthe
deviceisdrivenat80Vppand100Hz.Itisconsideredthatthisvibrationismainlycausedby
theincompletenessofthetravelingwave,whosevalueisroughlythesameasthisvibration
amplitude.
3.4.Conclusion
Anewfluidbearingisdevelopedwhichutilizestravelingwaves.Evaluatingtheprototype
devicethefollowingremarkscanbeconcluded:
Itispossibletorealizeanoncontactairbearingbytheproposedmethod.
Floatingdisplacementorloadcapacityisincreasedwithanincreaseinthedrivingvoltageand
frequency,i.e.thedeviceiselectricallycontrollable.Maximumloadcapacityisrecordedas
100N.
Theproposeddevicecanalsoproduceattractiveforceinsteadofthefloatingforcebyreversing
thetravelingwave.
Thebearingalsohasresistancetomomentofforceevenwithasinglepad.
4.MassproduciblerapidmixerbasedonBakerstransformation
Wedevelopedanovelmethodologytofabricatethreedimensionalpassivetypemixerbased
onthebakerstransformation(BT).BTisthebesttransformationformixingfluidsoflaminar
flow.WenewlydesignedtheBTstructurewithisovolumetricchangewithoutanyseparation/
joiningprocessoftwochannels.ItisasuitablesolutionformassproducingBTmoldstructures
byutilizingprecisioncuttingtechniques.TwoscalesofBTmixerswithsimilarstructuresare
introducedherein.Theoneisformicrofluidicanalyticalsystemstoaccomplishwellmixed
solutionsinashortchannellength,andtheotheroneinminiaturescaleaimsathighperform
ancemixingofhighviscosityfluidsinfoodprocessingorresinblending.Anultraprecision
fiveaxisplaningmachineanddiamondcuttingtoolswereusedforamicrofluidicBTmixer
mold on a oxygenfree copper block, in which the flow passage area was 3.2E9 m2. For a
miniatureBTmixermoldonanaluminiumblock,aprecisionmachiningcenterandanend
millwitha1mmradiuswereused.Theflowpassageareawas3.2E5m2.Westudiedtheir
mixingperformancesbynumericalanalysesandobtainedtheBTmixingresultsshowinggood
similaritieswiththatofnumericalanalyses.Moreover,themixingperformanceofthemicro
BTmixerwasquantitativelyexaminedtoaccomplishcompletemixingoverawiderangeof
flowrates.
4.1.Backgroundofmicromixersforbioinformatics
Inthepastdecadesmicrofluidicsystemshavebeenwidelyusedinchemistry,biology,and
nanobiotechnology,includingDNA[18]orproteinanalysis[19],cellsorting[20],andchemical
63
64
reactions [21]. Mixing inside microchannels plays an important role in those microfluidic
analysis systems, and many researchers have made efforts to develop innovative mixing
techniques inside microchannels [22]. In particular, mixing of solutes with a low diffusion
coefficientinsidemicrochannelsisimportantandusefulforavarietyofapplicationsincluding
immunoassay[23],anditisdesirabletogetthemostefficientandrapidmixingpossibleinside
them.
Mostpassivetypemixers[2427]dependonsimplemixingtechniqueswithoutanyexternal
powersources,unlikeactivetypemixers,whichusesuchsourcesasultrasonic[28],magnetic
stirring[29],andbubbleinducedacousticactuation[30].Inthepassivetypemixers,onlythe
structural design induces the mixing of fluids affected by the convection flow and large
interfaceoffluids.
Comparisonofactivetypemicromixerswiththepassivetypeshowsthattheformercanrealize
excellentmixingperformance,butwithsomedisadvantages:1)difficultyinintegratingother
microfluidiccomponents;2)highcostfromthestandpointofbeingadisposabledevice;3)
complexcontrolunitsoranexternalpowersource.Ontheotherhand,passive(static)type
micromixershavethreeadvantages:1)easyintegrationwithothermicrofluidiccomponents;
2)lowcost;3)noexternalpowersourceisneeded.Thesepassivestructures,however,didnot
haveallthedesiredfeaturesofprovidingrapidmixing,highmixingefficiency,awiderange
offlowrates,andhavingnodeadvolume.
4.2.MicrofluidicmixerutilizingBakerstransformation
Inthispaper,weproposeanewdesignforthe3Dmicromixer,whichisbasedonthebakers
transformation(BTinshort),whichprovidesthehighestmixingperformanceasdemonstrated
inchaotictheories[31].
4.2.1.Bakerstransformation
TheschematicillustrationoftheBTprocess(stretch,cutandfuse)isshowninFigure15.The
BTprocesstransformfluidlayersfromoneintotwo,andthereforethetransformationofn
timesproduces2nfluidlayers.Consequently,thebakerstransformationcanexponentially
shortenthediffusionlength.
Figure16aillustratestheBTmixingprocess.Twotypesoffluidsareindicatedinblueand
yellowcolors.Thesuccessive3Dconfigurationchangesfoldthebluefluidontotheyellow
fluidgradually.Aftercompletingthefolding,thecombinedfluidsarestretchedandthenhalf
ofthefluidsarebent90,whichthesestepsmakeagreatdifferencebetweenbakerstransfor
mationandparallellamination.Thebentpartofthefluidsaremovedtotheoppositeside.
Thenthemovedpartisgraduallyfoldedontotheoriginalfluidsagain.
4.2.2.NumericalCFDanalysis
Themicrofluidicdistributionofverticalcrosssectionalviewswasinvestigatednumerically,
byusingCFD(computationalfluiddynamics)software(ANSYSCFX).Theboundarycondi
tionsofthenumericalsimulationwere:20mm/sflowratefortheinlets;0Pastaticpressure
fortheoutlets;and0mm/sflowrateforthechannelwall(thewallwasregardedashavingno
roughnessandskiddingdidnotoccur).ThenumericalsimulationresultsinFigure16b.
4.2.3.Fabricationmethod
We have fabricated bakers transformation structures with isovolumetric change (without
dead volume) on an oxygenfree copper block. We selected a planing process by using an
ultraprecisionplaningmachine,NIC300(NagaseIntegrexCo.,Ltd.),whichconsistsofthree
feedtableswithdoublehydrostaticoilguidewaysontheXYZaxes,tworotaryindextables
ontheBCaxesandafiveaxiscontrolsystem.
Figure 15. Schema of Bakers transformation (BT)
Acustomorderedultraprecisiondiamondcuttingtool(A.L.M.T.DiamondCorp.),whichhad
a1mmnoseradius,wasusedtofinishthetopsurfaceofthemold.Theproposedmicrochannels
werethenmachinedwithanotherultraprecisiondiamondtool(A.L.M.T.DiamondCorp.).A
schematicofthemoldisshowninFigure17a.Onanoxygenfreecopperblock(3060mm),
theBTdeviceisdesignedtohavetwoinlets,injectionmicrochannelswith10mmlengthand
20mheight,10.4mmlengthBTstructures,andoutletchannelwith33mmlengthand20m
height.
PDMStypemicrofluidicBTmixerwasthendeveloped.PDMS(DowCorningInc.)andcuring
agent(DowCorningInc.)weremixedataratioof10to1,andthenthemixturewaspoured
ontothemold,curedat65Cfor2h,peeledfromthemold,andbakedat120Cfor30min.
Because the baked PDMS was sequentially used for the second mold, it was soaked in a
commercialdetergentsolution(5%)for5min.Then,itwasrinsedwithdoubledistilledwater
anddriedinavacuumchamber.Afterdrying,thePDMSmixturewaspouredontothePDMS
mold,cured,andpeeledfromthemoldasabove.Beforebondingofthesecondreplication
PDMSandacommercialslideglass,accessholeswerepunchedintothePDMSandthenthe
PDMSandslideglasswerebothtreatedunderoxygenplasma.
The3DconfigurationofthestructuresisshowninFigure17b.Thesestructuresweredesigned
to have transverse (xy) and longitudinal (xz) movement of solutions at the same time to
realizetheBTlikethatofFigure16.VerticalcrosssectionsoftheBTdevicealongtheyaxis
areillustratedinFigure17c.ThedottedlinesindicatethereplicationpointinFigure17b.One
cycleoftheBTdeviceis1040mandthereare10cyclesinall.SEMimageoftheBTdeviceis
givenasacrosssectionalview(Figure17d).Thescalebaris100m.
65
66
Figure 16. Schematics of (a)mixing process of BT device and (b)microfluidic distributions derived from numerical
simulations.
Figure 17. (a) Schematic of the BT device. (b) Schematic 3D diagram of the mold for the BT device. (c) Schematic illus
trations of vertical cross-sections of the BT device. (d) SEM image of the BT device. (e)&(f) Series of confocal micro
graphs for one cycle in a microchannel.
4.2.4.AnalyticalevaluationofBTmixingperformance[Yasuietal.,2011]
Wequantitativelystudieditsmixingperformancetoattaincompletemixingoverawiderange
offlowrates.ThemixingperformanceoftheBTdeviceisevaluatedusingFITC(fluorescein
isothiocyanate;diffusioncoefficient:2.61010m2s1)[32]andgoatantimouseIgG(immuno
globulinG;diffusioncoefficient:4.61011m2s1)[33]dissolvedinphosphatebufferedsaline
(PBS).
4.2.4.1.Confocalmicroscopy
Aconfocalmicroscope(FV1000,Olympus)wasusedtoobservethemixingbehaviourthrough
afocuslens40x/0.90(UPLSAPO,Olympus).Excitationlaserwas488nm.Confocalimages
werecapturedatadataacquisitionrateof0.90framespersecond.Stacksofeachconfocalxy
scanof512512pixelswerecollectedwithastepof0.5minthezdirection.Scanspeedwas
200s/pixel.Zseriesimagesweremergedintoverticalcrosssectionalimagesandanalyzed.
The flames in Figure 17e and Figure 17f show confocal micrographs of the vertical cross
sectionsofachannelsimilartothoseinFigure17cforonecycle;themixingofFITCsolution
andPBS(Figure17e)andthemixingofIgGsolutionandPBS(Figure17f).Flowratewas100
mm/s.Thescalebaris100mandthewhitedottedlinesindicatechanneloutlines.Themixing
behaviourshowedagoodagreementwiththefluiddistributionsofnumericalsimulations.
To evaluate the mixing performance, the mixing ratio was calculated using the following
formula[34];
67
68
1 N
N
i=1
(I i I i perf .mix )2 / N1 (I i0 I i perf .mix )2

N
i=1
100
(1)
whereN,Ii,Ii0,andIiperf.mixarethetotalnumberofpixels,thefluorescenceintensityatpixeli,
thefluorescenceintensityatpixeliwithoutmixingordiffusion,andthefluorescenceintensity
ofthecompletelymixedsolutionatpixeli,respectively.Generally,the90%mixingratiowas
regardedascompletemixing.
4.2.4.2.Analysisofmixingratio
ThemixingratiosofFITCsolutionvs.PBSandIgGsolutionvs.PBSareshowninFigures18a
and18b,respectively.Themixingratiowascalculatedusingformula(1).Circles,squares,and
trianglesshowy=1.04,5.02,and10.4mm(i.e.after1,5,and10cyclesofmixing),respectively.
Thedottedlineindicates90%mixingratio.TheBTdeviceprovidedcompletemixingofFITC
solutionandPBSatflowvelocitiesupto400mm/sfor10cycles,andcompletemixingofIgG
solutionandPBSwasattainedatvelocitiesupto50mm/sfor10cycles;thesewerecharacter
ized by the values of low Reynolds number (Re = Ul/ < 100, where U is the average flow
velocity,listhetypicalcrosssectionaldimension,andisthekinematicviscosityofthefluid).
ThedifferenceinmaximumflowratetoattaincompletemixingbetweenFITC(400mm/s)and
IgGsolution(50mm/s)wasattributedtothedifferentdiffusioncoefficients.Fromthesefigures,
wesawthemixingperformanceincreasedasthenumberofcyclesincreased,e.g.whenmixing
IgGsolutionandPBSat4mm/sflowvelocity,themixingratioswere44,72,and97%after1,
5and10cycles,respectively.
Figure 18. (a) Mixing ratio of FITC solution and PBS vs. flow rate. (b) Mixing ratio of IgG solution and PBS vs. flow rate.
(c) Schematic diagrams of the BT device. (d) Confocal micrographs of vertical cross-sections of a microchannel.
Figure18dshowsconfocalmicrographsoftheverticalcrosssectionsofthemoststretched
channel(160m20m,widthandheight,respectively)fordifferentcyclesasindicatedin
Figure18c.Thescalebaris100m.
Inthesemicrographs,FITCsolutionandPBSintheupperpartofFigure18dorIgGsolution
andPBSinthelowerpartofFigure18dwereintroducedtoobservethemixingbehaviour
insidetheBTdevice;asyringepumpoperatedatconstantflowvelocityof100mm/swasused.
Asthesimulationresultsindicated,therightedgeofthechannelwasnotfullymixedforboth
ofFITCandIgGafter5cycles.Atthisflowvelocity,FITCsolutionandPBSwerecompletely
mixedafter10cycles,butIgGsolutionandPBSwerenot.Thisinsufficientmixingcomesfrom
the10foldsmallerdiffusioncoefficientofIgGcomparedtoFITC.
To determine residence time to attain complete mixing, we calculated the time to achieve
completemixingfromFigure18aandFigure18bbydividingthelengthbyflowvelocity.Figure
19showsthemixingratiovs.residencetimeafter10cycleswiththeBTdevice.Filledandopen
symbolsshowFITCandIgGmixing,respectively.Thedottedlineindicates90%mixingratio.
The logarithmic fitting curves are expressed as Y=98.35+6.49logX (red filled circles),
Y=94.50+8.74logX (red open circles), Y=52.59+62.71logX (black filled triangles), and
Y=6.12+34.12logX(blackopentriangles);whereYispercentageofmixingandXisresidence
time.
TheresidencetimeforFITCintheBTdevicewas51msandforIgG306ms.Consideringthe
residencetimesinthemicrochannelwithoutBTstructuresof4.0sforFITCand297sforIgG
(thatcouldnotbeattainedinthismixinglengthbutwascalculatedfromthefittingcurve),our
BT device showed significant potential for mixing FITC solution and IgG solution more
efficiently and rapidly in a 10.4 mm mixing length microchannel than in a microchannel
withoutBTstructures,andmixingratewasmorethan70foldfasterforFITCsolutionand900
foldfasterforIgGsolution.
Figure 19. The mixing ratio vs. residence time in the BT device (circles) and microchannel without BT structures
(triangles).
69
70
4.3.MiniatureBTdeviceforhighviscosityfluidmixing
WehavealsodevelopedtheminiatureBTmixer,whichwasascaleupmodelofthemass
produciblemicroBTmixer,tomakeitpossiblethatthemixingmethodologyweproposedwas
availableformultiscalestructures[35].Withthemassproduciblemolddesign,itsshortlength
andhighmixingratiocharacteristics,theminiatureBTmixerwouldfeasibleforcommercial
useoftreatinghighviscositysolutionsinmixing/heating/chemistry.Herein,wedevelopedthe
miniaturefluidchannelstructurealongwiththebasicguidelinejustlikeasthedevelopment
ofmicroBTmixer.
High viscosity solution forms the laminar flow when flowing inside the miniature fluid
channelwithtensofmm2passagearea.Inthefoodindustry,oilsorflavouringmaterialsneed
tobemixedsmoothlyandhomogeneously.Thesameneedisformaterialindustrieslikemixing
resinsorotherchemicals.Theviscosityoffoodshasthedistributionof0.1100Pas,whichis
100100,000timesbiggerthanthatofwater.Justlikethewatersolutionsinasimplemicro
channel,asimplemacrochannelcannotmixfoodsolutions,thereforebakerstransformation
(BT)isfeasible.
4.3.1.Prototype
AminiatureBTdeviceformixinghighviscosityliquidswasprototypedasamixeritself,not
asamoldherein.Thealuminiumalloyplatewasfabricatedbyusingthemachiningprocess
witha2mmdiameterendmill,alongwiththedesignillustratedinFigure20.Thefabricated
structureisshowninFigure21.
Figure 20. Illustration of three-dimensional BT mixer:
Figure 21. Prototype BT mixer made of hard aluminium alloy
InFigure20,theside,top,andcrosssectionviewareshowninupper,middle,andlowerlevel,
respectively.Thedottedlinesindicatethereplicationpointsintheflowchannel.Thefarleft
ofthechannelillustrationisconnectedtotwoseparateinlets.Twocolorsrepresentthedifferent
fluidsmixedbyBTideally:(1)1stBTforming2folds;(2)2ndBTforming4folds;(3)reprise
ofthe2ndBT.ThentimesofBTforms2nfolds,meaningthediffusiondistanceamongevery
layerresultsin1/2ntobetheexponentialaccelerationoffluidmixing.
4.3.2.Preliminaryexperiment
Inordertoconfirmtheflowdistributiondirectly,weintroducedcureliquidsiliconerubbers
tocutthemwhentheytotallycuredin8hours,andobservedcrosssections.ThefabricatedBT
structurewassealedwiththeflataluminumplate,andtwoinletswereconnectedtosyringes,
providingthesiliconerubbers.Rotatingahandlemovessyringestranslationaltopushthe
siliconerubbersout.WeusedthewhitesiliconerubberTSE3504(MomentivePerformance
MaterialsInc.;viscosity:10Pasat23C,specificgravity:1.22at23C,demoldtime:8hoursat
25C)foroneinlet,andfortheother,thebluecoloredmasterME50M(MomentivePerform
ance Materials Inc.; viscosity: 800 Pa s at 23C) was blended into TSE3504 with 1 wt% for
distinction.
Thedistributionofverticalcrosssectionalviewswasalsoinvestigatednumericallybyusing
ANSYSCFX.Acontrolvolumewasthetetrahedralshapeandthelengthofeachsidewas0.2
mmatthemaximum.Theboundaryconditionswere:6mm/sflowratefortheinlets;0Pastatic
pressurefortheoutlets;and0mm/sflowrateforthechannelwall.Thesolutionviscosityand
specificgravityweresetto10Pasand1.0,respectively.
4.3.3.Results
BoththeexperimentalresultsandnumericalsimulationresultsaredisplayedinFigure22.The
toplineisthecuredsiliconerubber.Themiddleandthebottomlinearetheexperimentaland
thesimulationresultsascrosssectionalviewsintheflowdirection,respectively.
71
72
Figure 22. Results of four cycles of BT procedures
InFigure23,theexperimentalfoldingresultsinthefirstthreecycles(centercolumn)were
compared with the ideal BT illustration (left column) and the numerical analysis (right
column).
Theexperimentalresultsshowrelativelygoodsimilaritieswiththenumericalanalyticalresults
oftwo,four,andeightlayersfoldedinsideflowpassages.Meanwhile,comparedwiththeideal
BTillustration,therearenotsmalldifferences,especiallyinitsmarginalregions.Theevery
foldededgewasntprocessedperfectlybecauseoftheshearresistancedecreasingtheflow
rate.Themixingperformance,however,reliesmainlyoninterlayerdistancesnearthecenter
of flow path. Therefore, the proposed miniature BT mixer is able to achieve high mixing
performanceeventhoughitdoesntshowtheidealBTmixing.
Figure 23. Comparison among cross-sections after each BT cycle
4.4.Conclusion
Inordertomeetthedemandofmixingfluidsformultiscalemixers,Wehavedevelopeda
novelmethodologytofabricatethreedimensionalpassivetypemixersbasedonthebakers
transformation.WeaimedatfabricatingBTstructurewithisovolumetricchangewithoutany
separation/joiningprocessoftwochannelsforthesakeofmassprocucingBTmoldstructures
by utilizing precision cutting techniques. Two scales of BT mixers, one for microfluidic
analyticalsystemandtheotherformixinghighviscosityfluidsinfoodprocessingorresin
blending,weredeveloped.Weperformedlaboratoryexperimentsforevaluatingthemixing
performancesoftwoprototypes.Theconfocalmicroscopyfor10cycles(10.4mminlength)of
microfluidicBTshowedthatthesignificantpotentialformixingFITCsolutionandIgGsolution
moreefficientlyandrapidlythaninamicrochannelwithoutBTstructures,andmixingrate
was more than 70fold faster for FITC solution and 900fold faster for IgG solution. The
comparisonofnumericalanalysisandtheexperimentalresultofmixingtwocoloredsilicone
rubbersbyusingprototypedminiatureBTmixer(3cycles)showedgoodsimilaritiesintheir
crosssectionalphases.
Authordetails
EijiShamoto,NorikazuSuzuki,TakashiKatoandBurakSencer
DepartmentofMechanicalScienceandEngineering,NagoyaUniversity,Nagoya,Japan
References
[1] SuzukiN,YokoiH,ShamotoE.Micro/nanosculpturingofhardenedsteelbycontrol
lingvibrationamplitudeinellipticalvibrationcutting.Precisionengineering2011;35
4450.
[2] Shamoto E, Moriwaki T. Study on Elliptical Vibration Cutting. Annals of the CIRP
1994;43(1)3538.
[3] ShamotoE,MoriwakiT.UltraprecisionDiamondCuttingofHardenedSteelbyAp
plyingEllipticalVibrationCutting.AnnalsoftheCIRP1999;48(1)441444.
[4] ShamotoE,SuzukiN,TsuchiyaE,HoriY,InagakiH,YoshinoK.Developmentof3
DOF ultrasonic vibration tool for elliptical vibration cutting of sculptured surfaces.
AnnalsoftheCIRP2005;54(1)321324.
[5] Dow T. A, Miller M. H, Falter P. J. Application of a Fast Tool Servo for Diamond
Turning of Nonrotationally Symmetric Surfaces. Precision Engineering 1991;13(4)
243250.
73
74
[6] Suzuki N, Haritani M, Yang J, Hino R, Shamoto E. Elliptical Vibration Cutting of

TungstenAlloyMoldsforOpticalGlassParts.AnnalsoftheCIRP2007;56(1)127130.
[7] ShamotoE,AkazawaK.AnalyticalPredictionofChatterStabilityinBallEndMilling
withToolInclination.AnnalsoftheCIRP2009;58(1)351354.
[8] AltintasY,BudakE.AnalyticalPredictionofStabilityLobesinMilling.Annalsofthe
CIRP1995;44(1)357362.
[9] AltintasY,ShamotoE,LeeP,BudakE.AnalyticalPredictionofStabilityLobesinBall
End Milling. ASME Journal of Manufacturing Science and Engineering 1999;121
586592.
[10] ShamotoE,AltintasY.PredictionofShearAngleinObliqueCuttingwithMaximum
ShearStressandMinimumEnergyPrinciples.ASMEJournalofManufacturingSci
enceandEngineering1999;121399407.
[11] Shamoto E, Suzuki N, Hamaguchi A. A New Fluid Bearing Utilizing Traveling
Waves.AnnalsoftheCIRP2006;55(1)411414.
[12] Kawai T, Ebihara K, Takeuchi Y. Improvement of Machining Accuracy of 5Axis
Control Ultraprecision Machining by Means of Laminarization and Mirror Surface
Finishing.AnnalsoftheCIRP2005;54(1)329332.
[13] ShamotoE,MuraseH,MoriwakiT.Ultraprecision6AxisTableDrivenbyMeansof
WalkingDrive.AnnalsoftheCIRP2000;49(1)299302.
[14] ShamotoE,ParkC,MoriwakiT.AnalysisandImprovementofMotionAccuracyof
HydrostaticFeedTable.AnnalsoftheCIRP2001;50(1)285290.
[15] CameronA.BasicLubricationTheory.EllisHorwoodLimited;1976.
[16] HabermannH,LiardG.L.PracticalMagneticBearings.IEEESpectrum1979,Septem
ber:2630.
[17] SalbuE.O.J.CompressibleSqueezeFilmsandSqueezeBearings.JournalofBasicEn
gineering.TransactionsoftheASME,SeriesD1964;86(2)355364.
[18] BurnsM.A,JohnsonB.N,BrahmasandraS.N,HandiqueK,WebsterJ.R,Krishnan
M,SammarcoT.S,ManP.M,JonesD,HeldsingerD,MastrangeloC.H,BurkeD.T.
AnIntegratedNanoliterDNAAnalysisDevice.Science1998;282(5388)484487.
[19] IkamiM,KawakamiA,KakutaM,OkamotoY,KajiN,TokeshiM,BabaY.Immuno
piller Chip: A New Platform for Rapid and Easytouse Immunoassay. Lab Chip
2010;10(24)33353340.
[20] OzakiK,SuginoH,ShirasakiY,AokiT,ArakawaT,FunatsuT,ShojiS.Microfluidic
CellSorterwithFlowSwitchingTriggeredbyaSolgelTransitionofaThermorever
sibleGelationPolymer.,SensorsandActuatorsB2010;150(1)449455.
[21] LoseyM.W,SchmidtM.A,JensenK.F.MicrofabricatedMultiphasePackedbedRe
actors: Characterization of Mass Transfer and Reactions. Industrial & Engineering
ChemistryResearch2001;40(12)25552562.
[22] Nguyen N. T. Micromixers; Fundamentals, Design and Fabrication. Norwich, NY:
WilliamAndrewInc.;2008.
[23] Tachi T, Kaji N, Tokeshi M, Baba Y. Microchipbased Homogeneous Immunoassay
usingFluorescencePolarizationSpectroscopy.LabChip2009;9(7)966971.
[24] LiuR.H,StemlerM.A,SharpK.V,OlsenM.G,SantiagoJ.G,AdrianR.J,ArefH,
BeebeD.J.PassiveMixinginaThreedimensionalSerpentineMicrochannel.Journal
ofMicroelectromechanicalSystems2000;9(2)190197.
[25] StroockA.D,DertingerS.K,AjdariA,MezicI,StoneH.A,WhitesidesG.M.Chaotic
MixerforMicrochannels.Science2002;295(5555)647651.
[26] HongC.C,ChoiJ.W,AhnC.H.Anovelinplanepassivemicrofluidicmixerwith
modifiedTeslastructures.LabChip2004;4(2)109113.
[27] Kim D. S, Lee S. H, Kwon T. H, Ahn C. H. A Serpentine Laminating Micromixer
CombiningSplitting/RecombinationandAdvection.LabChip2005;5(7)739747.
[28] Yang Z, Goto H, Matsumoto M, Maeda R. Active Micromixer for Microfluidic Sys
tems using Leadzirconatetitanate (PZT)generated Ultrasonic Vibration. Electro
phoresis2000;21(1)116119.
[29] LuL.H,RyuK.S,LiuC.AMagneticMicrostirrerandArrayforMicrofluidicMixing.
JournalofMicroelectromechanicalSystems;11(5)462469.
[30] LiuR.H,YangJ,PinderaM.Z,AthavaleM,GrodzinskiP.BubbleinducedAcoustic
Micromixing.LabChip2002;2(3)151157.
[31] OttinoJ.M.TheKinematicsofMixing:Stretching,Chaos,andTransport.Cambridge
TextsinAppliedMathematics,CambridgeUniversityPress;1989.
[32] PeriasamyN,VerkmanA.S.AnalysisofFluorophoreDiffusionbyContinuousDis
tributionsofDiffusionCoefficients:ApplicationtoPhotobleachingMeasurementsof
MulticomponentandAnomalousDiffusion.BiophysicalJournal1998;75(1)557567.
[33] deBeerD,StoodleyP,LewandowskiZ.MeasurementofLocalDiffusionCoefficients
inBiofilmsbyMicroinjectionandConfocalMicroscopy.BiotechnologyandBioengin
eering1997;53(2)151158.
[34] JhonsonT.J,RossD,LocascioL.E.RapidMicrofluidicMixing.AnalyticalChemistry
2002;74(1)4551.
[35] YasuiT,OmotoY,OsatoK,KajiN,SuzukiN,NaitoT,WatanabeM,OkamotoY,To
keshiM,ShamotoE,BabaY.MicrofluidicBakersTransformationDeviceforThree
dimensionalRapidMixing.LabChip2011;11(19)33563360.
75
Chapter 5
Micro-Nano Robotics and

Mechatronics for Biomedical Applications
Tomohiro Kawahara and Fumihito Arai
1.Introduction
Inmedicalfield,onerecenttrendinclinicaloperationisminimallyinvasivesurgerywhichis
usedanendoscopeandsmallinstrumentstoavoidadamageforpatients[1,2].Thissurgical
technique has been well introduced to many diagnosis and treatment department such as
internalmedicine,ophthalmology,urology,especiallysurgicaldepartment.Aswellknown,
surgicalrobotshavebeenintroducedandoperatedinthisfieldtosupportsurgicaloperations
[3,4].Generally,effectofminimallyinvasivesurgeryforpatientisdefinedasfollows:
PatientValue =
Effective
Invasiveness 2
Therefore, in order to reduce invasiveness such as scars, it is important to miniaturize the

endoscope and surgical tools, while maintaining the performance of devices. In addition,
doctorswouldliketoaddsomefunctionsforimprovementofsafetyandusabilityofsurgical
toolsduringtreatments.Therefore,smallandthininstrumentwithhighfunctionarehighly
requiredandmicromechatronicsmakesasubstantialcontributiontofabricateandintegrate
suchnoveltoolsandsystems.
Ontheotherhand,targetofscientistsinbiosciencefiledhasbeenchangedfromtissuetosingle
cell level for specific investigation and understand of biological function. Especially, since
investigationofmechanicalcharacteristicsofcellsisreallyimportanttoknowfunctionsand
growthfactorsofcell,micronanoactuatorsarehighlyrequiredtostimulateandmeasurefor
aninvestigationofsinglecell[5,6].Sinceconventionalcommercialmanipulatorsarequite
largeandslow,itisdesiredtodevelophighperformancetoolswhichincludemicronanoro
botscapableofmanipulationandsensingofcellswithahighspeedandahighthroughput.
ByapplyingmicrofabricationtechniquesuchasMEMStechnology,wecanmakeandintegrate
tiny mechanical parts as a robot about similar size as cells with diameter of 10100 m.
Therefore,forbioapplication,micronanofabricationandroboticsarealsoimportantinorder
torealizethenovelrobotsandsystems.
2013 Kawahara and Arai; licensee InTech. This is an open access article distributed under the terms of the
78
Inthischapter,weintroducehowtodesign/fabricatethemicrodeviceandrobotusingfor
actualmedicalandbioscienceapplications.Currentproblemandfuturedirectionofmicro
nano mechatronics for biomedical applications are also discussed to encourage vigorous
researchactivityinthisfield.
2.Medicalapplication
2.1.Background
Minimallyinvasivesurgeryisawellestablishedmethodinmodernmedicine[7].Inparticular,
Endoscopic Submucosal Dissection (ESD) is well known for availability of 10 mmorder
cancerexcision[8].InESD,adoctorinsertsanoralendoscopeintoastomach,andperforms
surgicalproceduresbyusingnarrowtoolsforremovingtumortissues.Therefore,sincelarge
incisionsarenotrequired,thisoperationisaminimallyinvasiveproceduretoapplyaquick
recoveryforpatients.
However,thedisadvantagesofESDarethenarrowoperativefieldprovidedbyanendoscope
andthelowoperabilitybyasimplemechanismforceps.Itusesahookknifeandaninsulation
tippedelectrosurgicalknife(ITknife)tocutthelesion.ITknifeisdesignedtopreventthe
penetrationofthegastricwallbycoveringthetipoftheelectricalknifewithaceramicball.
However,perforationisstillencountered,leadingtoprolongedresectiontime.Therefore,the
ESDforgastriccancerrequiresadoctorwithtechnicalskillshigherthanthoserequiredforthe
other endoscopic procedures. A basic technical principle of the surgical resection is the
resection of appropriate tissues, which is forced to standout by injecting saline under the
tumor. However, this endoscopic surgery is onearm surgery that it can only insert one
endoscopetocutandexfoliatethelesion.Ideally,onearmispullingthelesionandanother
armtocutwouldimprovetheefficiencyandreducetheriskofcomplications,suchasbleeding
andperforationfromcuttingtheunconfirmedbloodvessels.Theexfoliatingprocedurewith
ahookorITknifeisespeciallydifficultduringinthesurgery.Ittakesthemostofthesurgery
time. A couple of studies have attempted to address this issue. In clinical field, several
techniquesforassistingendoscopicsurgeryareproposedandconducted[9,10].However,
these approaches are not enough to solve the timeconsuming and the skilldependent
problemsonESD,therefore,anewapproachhasbeendesiredforyears.
Therehavebeenmanyworksdiscussingsurgicalassistmicroarms[11].Catheterisagoodtool
fordiagnosinginnersurfaceofstomach,largeintestine,andsoforth[12].Mostcathetershave
a gripper, so that they can treat the surface of internal organs. However, when a catheter
contactstothestomachwall,iteasilyhasdeformation.Itisdifficulttokeeptheliftingposture
ofthecatheterforsurgicalassist.Haradaetal.developedthe2DOFmicromanipulatorwith
thediameterof2.4mmforintrauterinefetalsurgery,eventhoughithasnogripper[13].On
theotherhand,therobotictoolsincludedendoscopesystemsaredevelopedbyacoupleof
companies[14,15].Sincetheserobotsareembeddedinthespeciallydevelopedendoscope,
wecannotapplytheserobotsforanormaloralendoscope.
Micro-Nano Robotics and Mechatronics for Biomedical Applications
Considering this background, we have proposed a surgical concept for ESD by utilizing a
coupleofmicroarms,asshowninFigure1[16,17].Inthisapproach,adoctorcaneasilyoperate
thesurgicalprocedureswithouttimeconsumingmannerbythewiredrivenmicroarms,which
havemultiDOFforthewiderangemovement.Theseconventionalmicroarm,however,have
thecouplingproblembetweenthewireforthearmjointandthewireforthegripper,asshown
inFigure2.Whenweopenthegripper,thejointisalsomovedbythewirecoupling.Therefore,
wecouldnotpreciselycontrolthemovementofthemicroarm,andwemayapplydamageto
tissueoforgansbythisproblem.Mostpopularmethodfordecouplingmechanisminrobotics
iscomposedofseveralgears.Inthismethod,itisdifficulttodesignacompactmicroarmfor
endoscopeinsertion.Ikutaetal.havedevelopedthe5DOFwiredrivenmicroarmwiththe
diameterof3.0mmformicrosurgeryindeepareaoforgans[18].Inthisdevice,anumberof
wires are required to control the microarm, even though the decoupled wire drive was
achievedbythefinemicrojointdesign.
We have also proposed the decoupling wire driven exoskeletal microarm with gripper
enablingthesurgicaloperationofacancerintissuesduringESD.Basedonbothaphotoli
thographyandawireelectricdischargetechnique,wefabricatethedecouplingstructurefor
themetalframeofthemicroarm[19].Byusingthisapproach,thedevelopedextrathin2DOF
microarmcanbeinsertedtotheendoscopechannelwiththediameteroflessthan3mm.The
developedmicroarmiscapableofliftingupatissuelayeroforgansduringESD.
(a)ConventionalESD
Figure 1. Conceptual image of robotic endoscope surgery
Figure 2. Wires coupling problem to drive a microarm
(b)Proposedroboticsurgery
79
80
2.2.Basicconcept
InactualESD,therequiredspecificationsforamicroarmmountedtoanoralendoscopeareas
follows:
i.
Diameterofmicroarm:lessthan3.0mm
ii.
Movableangleofmicroarm:+/6080degree
iii.
Thicknessoftissuelayertohandle:2.05.0mm
iv.
Weightoftissuelayertohandle:approx.30g.
Figure3(a)showsaconceptualimageoftheproposedmicroarmwhichiscomposedofabase
link,aforelink,anarmjoint,andagripper.Sincethesizelimitationofendoscopechannelis
approximately3.0mm,wehavetodesignthemicroarmwiththediameteroflessthan3.0mm.
Weapplyanexoskeletalstructureforthearmdesigntoreducethesizeofthemicroarm.The
exoskeletaliswellknownforaminiaturizedstructuresuchasaninsectarm.Moreover,since
anexoskeletalstructurehasarotationaljoint,wecanestimatetheexactpostureofthemicroarm
bymeasuringthejointanglefromtheoutsideofthebody.
Atthefirst,wedesignthearmjointforbendingthemicroarmandthegripperforgrasping/
liftingofstomachtissuesbasedonananalyticalapproach.Fromtheseresults,wealsodesign
thedecouplingstructuretoassemblingthemechanicalelementsofthemicroarm.
2.3.Design
2.3.1.JointDesign
Figure3(b)showsthebasicstructureofthearmjoint,whichhastheshapeoftrapezoidframe.
Toincreasethemovableareaofthearmjoint,weactivelyapplytheelasticdeformationofthe
frametothejointdesign.Figure4showsthetrapezoidframemodelforthearmjointwhere
W,Ii,li,b,,MC,RB,andRCareatensionbywire,asecondmomentofarea,lengthofbeams,
ahalflengthofbottomoftrapezoid,anglefrombottomoftrapezoid,amomentatapointC,
anormalforceatpointB,andanormalforceatpointD,respectively(i=1,2).Fromthismodel,
wecalculatetheappropriateparametersofliandforarmdesign.Here,
b = l1cos + l2
(1)
Sincethemicroarmhasthesizelimitation,weuseb=0.49mmonthisanalysis.Wecanobtain
theequationaboutthedeformationofthebeamBDasfollows:
EI
d 2y
dx2
) (
) (
= R B x M C H x l2 W x l2 H x l2
RB =
W (l2 d )
2l2
(2)
(3)
(a)
(b)
Figure 3. Design of a robotic microarm with a microgripper
whereE,H(x),anddareYoungsmodulus,Heavisidefunction,andadistancebetweenthe
pointCandthepointofapplicationoftensileforcebywire,respectively.Byusingtheequations
(1)(3),wesimulatethedeformationanglecatCasfollows:
C =
W (b l1cos )(b l1cos + d )

6EI
+ 2(b l1cos ) {B cos ( B ) + D cos ( D )}
(4)
whereB,B,D,andDarethedisplacementofthebeamABatpointB,theangleofthebeam
ABatpointB,thedisplacementofthebeamDFatpointD,andtheangleofthebeamDFat
pointD,respectively.
Forthenumericalsimulation,weuseW=1.0Nandd=0.63mm.Fromtheresultofthesimulation
asshowninFigure5,wedeterminetheparameterssuchasl1=0.48mm,l2=0.00mm,and=0
deg.Toconfirmtheavailabilityofthedesignedparameters,wecarriedouttheFEManalysis
byusingtheCOMSOLMultiphysicssoftware.Figure6showsanexampleofthesimulation
result.Fromthisfigure,wecanseethattherotationcenterexistsinthecenterofthejointarm
andthearmjointhaslargedeformationbythewiretension.Thismeansthatwecanrealize
81
82
theextrathinmicroarmwhichhasboththewiredecouplingandthelargemovableangleof
joint.Therequiredaccuracyofthedesignedparameters(50morder)isenoughtofabricate
theactualmicroarmbyawireelectricdischargemachining.
Figure 4. Mechanical model of a trapezoid frame for the arm joint
Figure 5. Evaluation of the joint flexibility
Figure 6. FEM analysis of the designed joint
2.3.2.Gripperdesign
Inordertoimplementagrippertothemicroarm,itisimportantthatgripperhasasimple
mechanismandalargegripforce.Fromthesepoints,weapplyMuramatsuetal.proposed
principlewhichisbasedonabehavioroflongpillarbucklingtothegripperdesign,asshown
inFigure7[20].Byutilizingthisprinciple,wecandevelopthesimplegripperwithoutlarge
mechanicalelements.Todeterminetheshapeofthegripper,wecarriedoutFEManalysisby
theCOMSOLsimulator.Astheresultoftrialanderror,wedecidedtheshapeofthegripper
whichisnormallyopentype,asshowninFigure8.Theappropriatenessofshapedesignis
confirmedthroughbasicexperiments,insection2.5.
Figure 7. Concept of a buckling mechanism for the gripper design
2.3.3.Decouplingdesign
Fromthediscusseddesignsabove,wedescribethewiredecouplingdesignformicroarmand
gripper.Figure3(b)showsthewireimplementedmicroarm,wheretheredlineshowsthe1
83
84
wireforthegripperandthebluelineshowsthe2wiresforthearmjoint.Toachievethewire
decoupling,thewireforgripperispreciselypassedthroughthecenterpositionofthearm
joint.Therefore,evenifthearmjointisbentbywires,thegripperiscompletelyunaffected
withthedecouplingdesign.Asaresultofthedesign,wedecidetheparametersofthemicroarm
suchasthewidthoftheopenedgripperWg=4.5mm,theheightofthegripperHg=1.0mm,the
lengthoftheforelinkLa=10.0mm,andtheheightofthejointHa=1.8mm,respectively.
Figure 8. Stress distribution of the gripper
2.4.Fabrication
Figure9showsthemechanicalpartsofmicroarmwhichshouldbefabricatedbymicrofabri
cation.Inordertofabricateandassemblethemicroparts,weusedthecombinationofawire
electric discharge machining technique and a photolithography based electroplating
technique.
Figure 9. Mechanical parts to construct the microarm
Atfirst,toincreasethestiffnessofthemicroarm,thearmjointandgripperwerefabricatedby
using a wireelectric discharge, as shown in Figure 10, where the material of each part is
phosphorbronze,andthethicknessis0.2mm.
Figure 10. Fabricated arm joint and gripper machined by electrical discharge machining
Ontheotherhand,toassembleofthejointandgripper,werequirethelinkpartswiththe
accuracyof10morder.ByapplyingaphotolithographytothisfabricationasshowninFigure
11, we developed the extra thin mechanical parts for the microarm. For a first step of the
fabrication,thesacrificiallayer(LOR5B,TokyoOhkaKogyoCo.Ltd.)wascoatedonSiwafer.
ThenAuCrwassputteredonthewafer(thickness=300nm).Next,thephotoresist(KMPR3035,
KayakuMicroChemCo.Ltd.)wascoatedonthesubstrate.Afterremovingphotoresistand
sacrificiallayer(RemoverPG,KayakuMicroChemCo.Ltd.),ultrasoniccleaningwascon
ductedtopeelofftheremainingAulayer.Furthermore,byintroducingtheproposedStacking
MicroassemblyProcesscalledSTAMPasshowninFigure12[16],wefabricatedthelayered
linkpartstoincreasethepartsthickness.Figure13showsthefabricatedandassembledpart
(baselink)forthemicroarm.
Figure 11. Electroplating based on photolithographic micropattern
85
86
Figure 12. Fabrication method of base links using STAMP (Stacked Micro-assembly Process)
Finally, we completely developed the microarm with the diameter of 2.62 mm (maximum
height:1.9mm,maximumwidth:1.8mm),asshowninFigure14.
Figure 13. Fabricated and assembled single link part for the microarm
2.5.Experiments
2.5.1.Basicexperiments
Toconfirmtheperformanceofthedevelopedmicroarm,weperformedthebasicexperiments.
Effectofthedecouplingdesign:ThemicroarmwaspulledbyBolfurwires(diameter=100
m)withmanualoperation.Then,wemeasuredtherotationangleofthemicroarmfromthe
capturedimagebycamera(Figure15(a)(d)).Fromthisexperiment,weconfirmedthatthe
maximumrotationangleis60deg.with27.0Ntensionforce.Furthermore,wealsoconfirmed
Figure 14. Fabricated and assembled 2DOF microarm which is driven by wires
thatthearmjointmovementisnotmostlyaffectedbythatofthegripper,asshowninFigure
16(a)and(b).
Figure 15. Bending demonstration of the developed microarm
87
88
(a) Verification experiment to confirm the effectiveness off the decoupling design.
(b) When the joint was bent by wires with deg., the gripp
per was closed by pulling the another wire.
Then, we measured the to confirm the effect of the decou
upling mechanism.
Figure 16. Evaluation of the decoupling design
Gripforceofthemicroarm:First,thegripforceofthegripperwasmeasuredbythesensor
madebyametalflatplatewithastraingauge.Asaresult,weconfirmedthatthegripforceof
thegripperis0.52Natamaximum,asshowninFigure17.Next,asiliconerubberimitateda
humantissuewasgrippedandliftedbythedevelopedmicroarm,asshowninFigure18.In
thisexperiment,wechangedthethicknessofthe100.0x50.0mmsizesiliconerubber.From
this experiment, it was confirmed that the microarm can lift up the rubber sheet with the
thicknessof3.7mm(21.0g).InactualESD,wehavetohandleamucousmembranewitha
thicknessof2.05.0mmandaweightofapproximately30g.Throughthisexperiment,the
usabilityofthedevelopedmicroarmwasconfirmed.
Figure 17. Grip force according to grip position
Figure 18. Overview of the lifting experiment
2.5.2.Animalexperiments
Basedontheexperimentalresults,weappliedthedevelopedmicroarmtotheanimalexperi
ments.Themicroarmcouldinserttheactualendoscopechannel(2.7mm),asshowninFigure
19(a),andFigure19(b)showstheoverviewoftheexperimentconductedinthepigstomach.
Inthisexperiment,wesucceededininsertinganddrivingthemicroarminactualstomach.
However,sincethelimitationofthewireelectricdischargemachining,thegrippercouldnot
closecompletely.Therefore,wethinkthatthegripforceofthemicroarm(0.52N)isnotenough
tograspthetissuewithoutslippingduringESD.Inthefuturework,weshouldredesignand
fabricatethegrippertoimprovethegripforce.Thespecificevaluationofthegripforcewillbe
conducted(e.g.therelationshipbetweenthegripperangleandthegripforce).Ontheother
89
90
hand,thedevelopedmicroarmwasmadefromphosphorbronzeandnickelwhichmaycause
allergicreaction.Foraclinicalapplication,themicroarmshouldbecoatedbybiocompatible
material.
Webelievethatthedevelopedmicroarmcanbeappliedtootherendoscopicsurgeries,for
example NOTES (natural orifice translumenal endoscopic surgery), SPS (single port endo
scopicsurgery)andsoon[21,22].
(a)
(b)
Figure 19. Overview of the animal experiment
2.6.Discussionandsummary
Here, we described a developed decoupling wire driven exoskeletal microarm that can be
appliedtoESD.Themainresultsofourstudyareasfollows:
1.
The1.8mmthicknessmicroarmiscomposedofagripperwith0.52Ngripforce,anarm
jointwith+/60deg.rotatableangle,andtwolinkparts.Thesemechanicalelementsare
fabricatedbyawireelectricdischargeandaphotolithographytechnique.
2.
Through our basic experiments, we have confirmed the developed microarm can be
installedtoanendoscopechannelwiththediameterofthe2.7mmandthewiredecoupling
ofthemicroarmisachieved.
3.
Availablehandlingsizeofthedevelopedmicroarmis100.0x50.0x3.7mm(21.0g),we
applythemicroarmtoactualendoscopeduringESD.
Foroneofthefuturedirectionofmicrosurgicalrobot,newdesignapproachisalsorequired
todrasticallyimprovetheperformanceofthemicroarm.Currently,sincetheseconventional
instrumentsareconsistedofrigidmaterials,itisdifficulttoachieveboththeflexibilityandthe
biocompatibilitytoavoiddamageoftissues.Therefore,moreversatileinstrumentsareneeded
forapplicationofactualendoscopicsurgery.Incontrast,biologicalstructuresuchaslegsof
insectorcrustaceanscanmoveflexiblyandexertgoodstrengthbecauseoftheirexoskeletal
structure.Exoskeletalstructurehastheadvantagetominiaturizethewholestructurewitha
rotationaljoint,whichmakeiteasytoestimatetheexactpostureofthejointbymeasuringthe
joint angle from the outside of the structure. Moreover, joint has an anisotropic stiffness.
Consideredbythesemechanisms,bionicdesignisthemechanismandthematerialintroduced
torealizesuchasalivingstructure.Byusingthisapproach,bionicjointhasapotentialtosolve
theconventionalproblemsonmechanicaljointmadebyrigidmaterials.Forinstance,wehave
developedabionicjointwhichismadebydifferentstiffnessmaterialsthataresiliconeand
polymer[23].Thejointissmall(<2mm),anditcaninserttothechanneloforalendoscopeto
assistofthesurgicalprocedure,asshowninFig.20.Byusingthisapproach,webelievethata
doctorwillbeabletoeasilyaccessandoperateaffectedareasduringendoscopicsurgery.
Figure 20. Design concept of bionic joint and fabricated microarm
3.Bioscienceapplication
3.1.Background
Recently,inbiosciencefield,mechanicalstimulationtoasinglecellishighlyrequiredtofigure
outfunctionsandmechanicalcharacteristicsofcells.Especially,thisapproachisveryuseful
forunderstandingthemechanismofaquaticmicroorganismsintermsofneurologyandbio
fueltechnology[24].Sincethemechanicalcharacteristicsofanumberofaquaticmicroorgan
isms have not been fully understood yet, mechanical approach is really important to
understandthecellinteraction,growthfactorofcells,andfunctionofmechanoreceptor.In
ordertocopewiththisissue,conventionalinvestigationformicroorganismswascarriedout
manually or by using micromanipulators under microscope. However, throughput and
dexterityofmeasurementwereverylow,becausethemanipulatorhadtohandletheslender
glasstools.Additionally,sincepetridishwasusedtheseoperation,wecannotmaintainthe
stableconditionofmeasurementsystemwithoutcontaminationanddisturbance.
Therearetwoimportantpointstoevaluatetherelationshipbetweenthemechanicalstimula
tionandtheresponseofmicroorganism.Thefirstthingisthatweshouldmeasuretheamount
ofappliedstimulationtocellasanappliedforce.Theotherthingisthat,weshouldkeepthe
experimentalconditiontoobservethechemicalreactionafterstimulationofcellprecisely.
91
92
Ontheotherhand,therearemanyworksofthemicronanorobotsandsensorsbasedon
MEMS/NEMStechnologiestomanipulateand/ormeasurethemicroorganisms[2529].Sun
etal.havebeendevelopedtheSOIbased2DOFforcesensorwhichisconsistedofthecomb
structurestodetectthechangeofthecapacitance[30,31].Thissensorwasappliedtothe
evaluationofthemechanicalpropertyofoocyte.Nakajimaetal.havebeendevelopedthe
nanoprobestomeasurethestiffnessoftheC.elegansbytheAtomicForceMicroscope(AFM),
whichisthecantileverbasedforcesensing[32].Cappellerietal.haveproposedtheNforce
sensormadebyPolydimethylsiloxane(PDMS)formicrorobotics[33,34].Thissensorcan
measuretheforceoftwodirectionsbyusingthevisionsensor.Thesesensors,however,
requiretheoperationbymicromanipulatorstomeasuretheobjectsinthenarrowspace.This
meansthatitisdifficulttoclosethetotalmeasurementsystemtopreventacontamination
ofcellandliquid.
Incontrast,Abbottetal.havebeencontinuouslyresearchingmagneticactuatorsoperatedby
pairsofHelmholtzcoils.Theyfabricatedspiralshapemicrorobotandthemicrorobotswimin
liquidenvironmentanditcanachieve6DOFpreciseactuation[35].Pawasheetal.alsoapplied
pairsofHelmholtzcoilsformicrorobotactuation.Theyactivelyemployedstickslipeffectson
theslidingmagneticobjectbyswitchingthemagneticfieldsofverticaldirectionandhorizontal
direction[36].Sakaretal.alsodevelopedaHelmholtzcoilsdrivenmicrotransporter[37].They
integratedthemagneticactuationwithvisionfeedbackcontrol,andtheyachievedtodeliver
themicrogeltoneuron.Wethinkthattheserobotswithboththesubmicronorderpositioning
accuracyandthemillimeterorderdrivespeedareveryusefultoinvestigatemicroorganisms.
However,theydidnothaveanysensorstoevaluatethecharacteristicsofcellinaclosedspace.
Recently,opticalforcedrivennanorobotswithnmpositioningaccuracyhavebeendeveloped.
Ikuta et al. developed 6DOFnanorobot made by light curing resin. They succeeded in the
measurementofappliedforcetoYeastcellswithapprox.5mdiametersinafluidicenviron
ment[38].Araietal.alsodevelopedtheforcefeedbacksystembyusingthemultibeamoptical
tweezers[39].ThedisadvantageoftheopticaldrivemethodisfromnNtopNordergenerative
force. It is not enough to push 100 m size microorganisms which have a stiff structure.
Therefore, onchip manipulation/sensing of microorganisms by functionalized microrobots
arehighlyrequited.
3.2.Basicconcept
Intheconventionalworksofourgroup,themagneticallydrivenmicrotools(MMT)havebeen
proposedforautomationofcellmanipulation,asshowninFigure21[40].TheMMT,whichis
placedinamicrofluidicchip,iscomposedofamagneticmaterial(Ni),andcanbeactuatedby
apermanentmagnetmountedonamotorizedstage(anXYlinearstageandarotationmotor)
outsidethemicrofluidicchip.UsingfourmagnetstodriveasingleMMT,wecanachievestable
movement of the MMT with 3DOF (Xdirection, Ydirection and rotation). This allows the
manipulationofcellsinacompletelyclosedbiochipwithgenerativeforcesoftheorderofmN.
Since a closed microchip is used, we can avoid any contamination and maintain stable
conditions during the measurements. This system contributes to our ability to observe
continuouslytheresponseofthechemicalsofmicroorganismsafterstimulation.Further,our
approachhastheadvantagesofbeingadownsizedactuationsystemandhavingadisposable
chipstructure.Conventionally,wesucceededinincreasingthepositioningaccuracyofa3DOF
MMT from several hundred m to 1 m by using both the new method of arranging the
permanentmagnets[41]andavibrationbasedmethodofreducingfrictionbetweentheMMT
andthesubstrate[42].
Figure 21. Basic concept of on-chip microtools and fabricated microfluidic chip with four microtools made of Ni.
Here, we consider that the beam structure is mounted on the tip of the MMT and the
deformationofthebeamismeasuredbythemicroscopetoachievetheonchipforcesensing,
as shown in Figure 22(a). However, there is a large friction between the MMT and the
substrateofthebiochip(normallyglass)becausetheMMTisattractedbythemagnet.Itis
bigissueforonchipaccurateforcesensingbecausetheeffectofthefrictionisincludedto
the measured force data. Furthermore, since the beam structure becomes very thin, we
concernthatitisdifficulttoassembletheMMTtothemicrochipwithoutanydamage.To
avoidtheseissues,wenewlyproposethelayertypeMMTwhichincludestheMMTwith
thebeamstructuresupportedbythespring,themagneticmaterial,andthemicrochannel,
asshowninFigure22(b).ThespacebetweentheMMTandthesubstrateismaintainedby
themicrospacer.Therefore,thereisnofrictionontheforcesensingstructureoftheMMT.
Byusingthelayerstructure,themicropatternoftheMMTisprotectedandeasytoassem
bletheseveralmicroparts
93
94
(a) 2DOF MMT with a force sensing structure
(b) Layer fabrication method to avoid friction and damage of MMT
Figure 22. Concept of force sensing microtool and layer fabrication method
3.3.Basicsensingprinciple
Inthefirststepoftheonchipforcesensing,wedesigntheframeshapedbeamtotheMMT.
ThisstructurenotacantileverstructureisimportanttokeepstraightthepostureoftheMMT
whenwepushcells.Inaddition,toincreasethesensitivityoftheforcesensingbythecamera,
themagnificationmechanismofthebeamdeformationisplacedintotheframe,asshownin
Figure23.
Todesigntheparameteroftheforcesensingstructure,wesetthemechanicalmodelofthe
frame,asshowninFigure23,whereF,,l,h,d,r,I1=b1h13/12,I2=b2h23/12,andEaretheforceto
bemeasured,thedisplacementofthebeam,thewidthoftheframe,theheightoftheframe,
the width of the magnification mechanism, the height of the magnification mechanism,
Youngsmodulus,thesecondmomentofareaofthebeamBC,andthesecondmomentofarea
ofthebeamCD,respectively.Here,thedisplacementofthemagnificationmechanismis
X =
(5)
whereisthemagnificationratio.Therefore,fromthisstaticmodel,wecanobtaintheapplied
forceasfollows[43]:
F = kX
=
4E I 1
r
d ( L l )( I 1 H + I 2l ) 1
( I 1 H + 2I 2l )
(6)
Figure 23. Frame model of force sensing structure and designed parameters
3.4.Fabrication
By considering the results of the previous works which applied the force to the cells, we
selectedanSU8negativephotoresistsforamaterialoftheMMTtopreventthedamageto
cellsormicroorganisms.Aswellknown,SU8isflexible(approx.2to10GPa),andwecan
easilymakethemicropatternbyusingthephotolithographytechnique[44,45].TheYoungs
modulusofSU8dependsonthethicknessoftheSU8andtheexposureamount.Byconsid
eringthediameteroftargetedcellwithadiameterofapprox.0.050.1mm,wesetthatthe
thicknessoftheMMTwasb1=b2=0.05mmandtheexposureamount(UVDose)was500mJ/
cm2.Then,weused3.2GPaYoungsmodulusfortheSU8layerdesign.Ontheotherhand,
fromthelimitationofthefieldofviewofthemicroscopewiththe10Xobjectivelens(1.5x1.0
mm),wesetl=0.92mm,h=0.6mm,r=0.5mm,andd=0.5mmtoobservethewholeforcesensing
structureduringmeasurements.Finally,fromtheresolutionofthephotolithographyofthe
SU8withtheemulsionmask,wedecidedash1=0.02mmandh2=0.04mm.
Figure 24 shows the results of the FEM analysis based on the designed parameters. To
determinetheoptimalconditionofthemovableareaoftheMMT,wevariedthewidthofthe
supportspring.Finally,wedeterminedthataspringwidthof50mwassuitable.Fromthe
results of the simulation, we confirmed that the support spring can be manipulated by
magneticactuation,andthattheforcesensingstructureworkswell,eventhoughtherewere
somedifferencesbetweentheanalysisandthesimulationbecauseofmodelingerror.
95
96
Figure 24. FEM simulation of support springs and force sensing structure
Figure 25 shows the fabrication process for the proposed layer type MMT with the frame
structure.Thebiochipwascomposedoftheglasssubstratewiththethicknessof100m,the
microspacerwiththethicknessof15m,theSU8layerwiththethicknessof50mincludes
theMMTsandthemicrochannel,thehemisphereswiththediameterof1mmtoactuatethe
MMTs,andthePDMScoverwiththethicknessof5mm.Figure26(a)showsthefabricated
SU8layer,andFigure26(b)showstheenlargedviewofthefabricatedMMT.Topushthelocal
pointofasmallcell,atriangleshapedprobewasattachedtothetipoftheMMT.Figure27
showsthesideviewoftheMMT.Thespacebetweentheforcesensingstructureandtheglass
substrateismaintainedbythespringstructure.
Figure 25. Fabrication process of MMT with a force sensing structure
Figure 26. Fabricated MMT with force sensing structure
Figure 27. Assembled MMT layer (without Fe hemisphere) and microspacer layer
3.5.Experiments
Next,weevaluatedtheforcemeasurementbythedevelopedMMT.Inordertomeasureforce
dataaccurately,theMMTwaspushedagainstacommercialforcesensor(KYOWA,LVS5GA)
withanaccuracyof10Nbyusingalinearstagewithapositionaccuracyof2m,asshown
inFigure28.Then,wemeasuredthedeformationusingamicroscopeequippedwithaCCD
camera (resolution: 1 m/pixel). In this experiment, we confirmed that there was a high
linearity between the displacement of frame and the displacement of the magnification
mechanismX,andthatthemagnificationratiowas2.72.Therefore,Xcanbeusedtoestimate
theappliedforce.
97
98
Figure29(a)showstherelationshipbetweenthedisplacementXandtheforcemeasuredby
thecommercialsensor.Fromthisresult,weconfirmedthatthereisahighlinearitybetween
thedisplacementandtheX.Therefore,themagnificationratioisapproximately2.7andX
canbeusedtoestimatetheappliedforce.
Figure 29(b) shows the relationship between the X and the force data measured by the
commercial sensor. From these results, we can see that there is a hysteresis in different
conditions.Furthermore,whentheappliedforceislargerthan3mN,theshapeoftheframe
doesnotreturntotheinitialstate(offset).Basedonthesecharacteristics,wehavetoonlyuse
thedisplacementXinthepushingphase.And,itisbettertolimittheforcerangetolessthan
3mNtoestimatetheforcedataproperly.
Fig. 29(c) left and center show the measured force data by the commercial sensor and the
estimatedforcedatacalculatedfrom(6).Fromthisresult,wecanseethatthereisadifference
betweenthemeasuredandtheestimatedforces.Wethinkthatthiswascausedbyerrorsinthe
fabricationandinthenumericalanalysis.Bycorrectingtheestimateddata,weconsideredthe
fabricationerror,andwecanmeasureaccurateforcedata,asshownintherightsideofFig.
29(c).
Next,Fig.29(d)showstheindividualdifferencebetweenMMTsintheSU8layer.Fromthese
results,weconfirmedthatthevariationamongtheMMTsislessthan10%.Thismeansthat
bycalibratingoneoftheMMTsfromthefabricatedlayer,wecanuseotherMMTs(included
inthelayer)withoutcalibration.
Through these experiments, we obtained the minimum resolution of the force sensing as
approximately100N.Theresolutionforthistypetheforcesensorwasdeterminedbythe
resolutionofthecamerausedandthesensitivityoftheframedeformation.Alargedeformable
frame caused a large hysteresis and offset as discussed above. Therefore, by using a high
resolutioncamera,wecanincreasethesensorperformancewithcomparativeease.Asafirst
step,webelievethatthisperformanceisreasonabletodeterminetherangeofforcerequired
topusha100msizesinglecell.
Figure 28. Overview of force calibration experiment
Figure 29. Evaluation of calibration of fabricated force sensing structure
3.6.Application
WeappliedthedevelopedMMTtoactualmicroorganism.Inthisexperiment,weusedoneof
centricdiatomcalledPleurosiralaevis.Thiscellhasauniquebehavior,thatis,itisstimulated,
the chloroplasts within it agglomerate around the nucleus, as shown in Figure 30 [46].
Furthermore, this agglomeration phenomenon is transmitted to other connected cells, as
showninFigure31.Interestingly,thephenomenoneventransmitstounconnectedcells.The
reason for this spreading of chloroplast agglomeration is to be because of the release of
unknowncomponentsfromthestimulatedcelltostimulateothercellstoincreasegrowth.This
99
100
phenomenon is considered to be similar to information transmission in neurons. From the

perspectiveofneurologyandalgaebasedbiofuel,theinvestigationofthisphenomenoniswill
beusefulforunderstandingthemechanismofcommunicationbetweenaquaticmicroorgan
ism.Althoughitiswellknownthatthesecellshaveareceptors,andthus,thattheagglomer
ation phenomenon is caused by optical and electrical stimulations, the mechanism of this
phenomenon is still not fully understood [47, 48]. Conventional studies with a mechanical
approach, however, could not clarify the relationship between the applied force and the
responseofP.laevisduetoforcesensingproblem[49].Therefore,quantitativeinvestigationof
theappliedforcemustbeperformedinordertoelucidatetheagglomerationphenomenonof
chloroplastsofP.laevisbymechanicalstimulation.
(a) Overview of single cell
(b) After optical/electrical/mechanical/chemical stimulation
Figure 30. A single cell of Pleurosira laevis (P. laevis)
Figure 31. Propagation of agglomeration in P. laevis
Fig.32showstheoverviewoftheexperimentbydevelopedMMT.Asaresult,wesucceeded
inrotating,capturing,stimulationofmicroorganismsbyusingdualMMTwithforcesensing
structure.Then,theappliedforcewasestimatedasshowninFigure33.Theagglomeration
propagationphenomenonofP.laeviswasalsoobservedbyMMTstimulation,asshownin
Figure34.Infuturework,therelationshipbetweentheappliedforceandtheagglomeration
phenomenonisinvestigatedbytheproposedsystem.Theoptimizedmicrochanneldesignis
alsorequiredtoreducethewholemeasurementtimeandtheobservations.
Figure 32. On-chip P. laevis stimulation by MMT
Figure 33. Measured force data by force sensing structure
3.7.Discussionandsummary
Inthissection,wediscussedanonchipforcesensingbymagneticallydrivenmicrotool(MMT)
whichenablesmeasurementofstimulantpropertyofsinglecell:
101
102
Figure 34. Measured agglomeration propagation phenomenon
1.
Themicrochipiscomposedofamicrospaceer,anMMTlayer,anFehemisphereanda
PDMScover.Thesepartsareassembledbynewlyproposedlayerfabricationtechnique.
2.
Theframebasedforcesensingstructurewiththedisplacementmagnificationmechanism
isdesignedandfabricatedbasedonanalyticalapproach.
3.
Throughourbasicexperiments,wehaveconfirmedtheeffectivenessofthefabrication
methodandtheperformanceoftheforcesensing.
4.
ThroughourexperimentsforP.laevisstimulation,weconfirmedthattheproposedforce
sensing mechanism works well with accuracy of 100N, and the MMT has a enough
powertopushP.laevis.
Currently,theforcesensingstructureoftheMMTwasdesignedtomeasureaonedirectional
force, taking into account both the specifications of the P. laevis measurement that were
requestedandthedifficultiesinvolvedinthefabricationprocess.TheMMTwasfoundtobe
satisfactory for investigating the stimulant property of P. laevis. For wider applications,
however,itwouldbebettertomountamultiDOFforcesensingstructureontothetipofthe
MMT.Additionally,inordertomeasurethemicroorganismprecisely,weshouldconsidera
fullyautomated3DOFpositioncontrolofMMT.Automatedvisionbasedforceestimationand
MMTpositioncontrolcanbeintroducedintoourdevelopedsystembyexploitingrobotics
technologies. We believe that our approach is useful for investigating of living cells and
microorganismsinaclosedmicrofluidicchip.Thecombinationofthesimpleanddisposable
SU8basedstructureandthecamerabasedmeasurementachievesverylowcostforcesensing
system.
Inourfutureplan,thedevelopedsystemwillbeappliedtomeasureothermicroorganisms
andtheMMThasapotentialtocaptureandmeasuremotilemicroorganismswhichcontaina
large proportion of aquatic microorganisms [50]. In parallel, to improve the force sensing
resolutionfromNordertonNorder,electricalsensorsshouldbeimplementedtotipofthe
MMT,althoughthereisacoupleofproblemonfabricationprocess.Inaddition,wearenow
addingmorefunctionsforMMTtoexpandapplicationofmicrorobots[51].Theperformance
oftheMMTsuchaspositioningaccuracyanddrivingspeedmaybeincreasedbyintroducing
asubmicroandananoorderfabricationapproach[52,53].
4.Conclusion
Weshowedtwospecificapproachesbasedonmicromechatronicsandfabricationformedical
andbioscienceapplications.Theresultsindicatedthatmicroroboticsisalsoreallyimportant
todrive,controlandmeasurethedevelopeddevices.Currently,manipulationandsensingof
smaller object are strongly requited in biomedical field. In order to realize a nanoscale
manipulationandmeasurement,appropriateintegrationofmicroandnanomechatronicsare
mostimportantfactor,becausetherearedifficultiestoseamlesslyintegratethefunctioningof
microandnanoobjects.Inaddition,handlingandsensingofmicroscalewetobjectsuchcells
are also important as mentioned above. Therefore, it is required to establish and develop
innovativemethodologiesregardingcontrol,sensingandfabricationinthefieldsofmicro
nanomechatronicsandrobotics.Forthepurposeofensuringtheappropriateandeffective
integrationofthetechnologyandscience,weshouldbecontinuedtheongoingpromotionof
corroborationinabroadrangeofareas.
Acknowledgements
This work has been supported by the Ministry of Education, Culture, Sports, Science and
Technology GreatinAid for Scientific Research (21656064, 22860030) and the Nagoya
UniversityGlobalCOEProgram,``COEforEducationandResearchofMicroNanoMecha
tronics,.
Authordetails
TomohiroKawaharaandFumihitoArai
References
[1] Hasegawa H, Kabeshima Y, Watanabe M, Yamamoto S, Kitajima M. Randomized
Controlled Trial of Laparoscopic versus Open Colectomy for Advanced Colorectal
Cancer.SurgicalEndoscopy2003;17(4)636640.
103
104
[2] TheClinicalOutcomesofSurgicalTherapyStudyGroup.AComparisonofLaparo
scopicAssistedandOpenColectomyforColonCancer.TheNewEnglandJournalof
Mdicene2004;350(20)20502059.
[3] ntuitive Surgical Inc. da Vinci Surgical System. http://www.intuitivesurgical.com/
(accessed13May2012)
[4] SpaunG.O,ZhengB,SwanstromL.L.AMultitaskingPlatformforNaturalOrifice
TranslumenalEndoscopicSurgery(NOTES):aBenchtopComparisonofaNewDe
viceforFlexibleEndoscopicSurgeryandaStandardDualChannelEndoscope.Sur
gicalEndoscopy2009;2327202727.
[5] DischerD.E,JanmeyP,WangY.L.TissueCellsFeelandRespondtotheStiffnessof
TheirSubstrate.Science2005;31011391143.
[6] EnglerA.J,SweeneyH.L,DischerD.E.MatrixElasticityDirectsStemCellLineage
Specification.Cell2006;126677689.
[7] KondoH,GotodaT,OnoH,OdaI,KozuT,FujishiroM,SaitoD,YoshidaS.Percuta
neous TractionAssisted EMR by using an InsulationTipped Electrosurgical Knife
forEarlyStageGastricCancer.GastrointestinalEndoscopy2004;59(2)284288.
[8] MiyamotoS,MutoM,HamamotoY,BokuN,OhtsuA,BabaS,YoshidaM,Ohkuma
M,HosokawaK,TajiriH,YoshidaS.ANewTechniqueforEndoscopicMucosalRe
section with an InsulatedTip Electrosurgical Knife Improves the Completeness of
ResectionofIntramucosalGastricNeoplasms.GastrointestinalEndoscopy2002;55(4)
576581.
[9] MinamiS,GotodaT,OnoH,OdaI,HamanakaH.CompleteEndoscopicClosureof
Gastric Perforation Induced by Endoscopic Resection of Early Gastric Cancer using
EndoclipscanPreventSurgery.GastrointestinalEndoscopy2006;63(4)596601.
[10] Sakamoto N, Osada T, Shibuya T, Beppu K, Matsumoto K, Shimada Y, Konno A,
Kurosawa A, Nagahara A, Ohkusa T, Ogihara T, Watanabe S. The Facilitation of a
NewTractionDevice(SOclip)AssistingEndoscopicSubmucosalDissectionforSu
perficialColorectalNeoplasms.Endoscopy2008;40E9495.
[11] MartinC,ChapelleF,LemaireJ.J,GoguG.NeurosurgicalRobotDesignandInterac
tiveMotionPlanningforResectionTask.In:ProceedingsoftheIEEE/RSJInternation
alConferenceonIntelligentRobotsandSystems,1115October2009,St.Louis,USA;
2009.
[12] AraiF,FujiR,FukudaT,NegoroM.RobotAssistedActiveCatheterInsertion:Algo
rithms and Experiments. The International Journal of Robotics Research 2009;28(9)
11011117.
[13] HaradaK,TsubouchiK,FujieM.G,ChibaT.Micromanipulatorsforintrauterinefe
talsurgeryinanOpenMRI.In:ProceedingsoftheIEEEInternationalConferenceon
RoboticsandAutomation.1822April2005,Barcelona,Spain;2005.
[14] HattoriA,SuzukiN,HayashibeM,SuzukiS,OtakeY,TajiriH,KobayashiS.Devel
opmentofaNavigationFunctionforanEndosocopicRobotSurgerySystem.Studies
inHealthTechnologyandInformatics2005;111167171.
[15] RothsteinR.I,SwanstromL.L.UseoftheDirectDriveEndoscopicSystem(DDES)
for InVivo Mucosal Resection in a Porcine Model. Gastrointestinal Endos
copy2008;67AB146.
[16] NarumiK,AzumaD,AraiF.FabricationofArticulatedMicroarmforEndoscopyby
Stacked Microassembly Process (STAMP). Journal of Robotics and Mechatronics
2009;21(3)396401.
[17] AzumaD,LeeJ,NarumiK,AraiF.FabricationandFeedbackControlofanArticulat
edMicroarm.In:ProceedingsoftheIEEEInternationalConferenceonRoboticsand
Automation.1217May2009,Kobe,Japan;2009.
[18] Ikuta K, Yamamoto K, Sasaki K. Development of Remote Microsurgery Robot and
NewSurgicalProcedureforDeepandNarrowSpace.In:ProceedingsoftheIEEEIn
ternational Conference on Robotics and Automation. 1419 September 2003, Taipei,
Taiwan;2003.
[19] Kawahara T, Matsumoto T, Muramatsu T, Osada T, Sakamoto N, Arai F. Develop
mentofaDecouplingWireDrivenExoskeletalMicroarmforEndoscopicSubmucosal
Dissection. In: Proceedings of the 3rd IEEE/RASEMBS International Conference on
BiomedicalRoboticsandBiomechatronics,2629September2010,Tokyo,Japan;2010.
[20] MuramatsuN,AndoH.EfficiencyofGrippingMechanismusingBucklingPhenom
enonofLongColumn.In:ProceedingsoftheIEEEInternationalConferenceonRo
boticsandBiomimetics.1822December2009,Guilin,Guangxi,China;2009.
[21] KoC.W,KalloA.N.PerOralTransgastricAbdominalSurgery.ChineseJournalof
DigestiveDiseases2006;76770.
[22] BaronT.H.NaturalOrificeTransluminalEndoscopicSurgery.BritishJournalofSur
gery2007;9412.
[23] Owaki H, Kawahara T, Arai F. Bionic Design of Microjoint for Minimally Invasive
Surgical Instrument. In: Proceedings of IEEE International Symposium on Micro
NanoMechatronicsandHumanScience.69November2011,Nagoya,Japan;2011.
[24] Yadugiri V. T. Milking Diatoms a New Route to Sustainable Energy. Current Sci
ence2009;97(6)748750.
[25] FauverM.E,DunawayD.L,LilienfeldD.H,CraigheadH.G,PollackG.H.Micro
fabricated Cantilevers for Measurement of Subcellular and Molecular Forces. IEEE
TransactionsonBiomedicalEngineering1998;45(7)891898.
[26] PapiM,SyllaL,ParasassiT,BrunelliR,MonaciM,MaulucciG,MissoriM,Arcovito
G,UrsiniF,SpiritoM.D.EvidenceofElastictoPlasticTransitionintheZonaPelluci
105
106
da of Oocytes using Atomic Force Spectroscopy. Applied Physics Letters 2009;95

15390211539023.
[27] JeongK.H,KellerC,LeeL.DirectForceMeasurementsofBiomolecularInteractions
byNanomechanicalForceGauge.AppliedPhysicsLetters2005;8619390111939013.
[28] Koch S, Thayer G, Corwin A, de Boer M. Micromachined Piconewton Force Sensor
forBiophysicsInvestigations.AppliedPhysicsLetters2006;8917390111739013.
[29] WacogneB,PieralliC,RouxC,GharbiT.MeasuringtheMechanicalBehaviorofHu
manOocyteswithaVerySimpleSU8MicroTool.BiomedicalMicrodevices2008;10
411419.
[30] SunY,NelsonB.J,PotasekD.P,EnikovE.ABulkmicrofabricatedMultiAxisCapac
itiveCellularForceSensorusingTransverseCombDrive.JournalofMicromechanics
andMicroengineering2002;12832840.
[31] SunY,WanK.T,RobertsK.P,BischofJ.C,NelsonB.J.MechanicalPropertyCharac
terizationofMouseZonaPellucida.IEEETransactionsonNanobioscience2003;2(4)
279286.
[32] NakajimaM,AhmadM.R,KojimaS,HonmmaM,FukudaT.LocalStiffnessMeas
urements of C. elegans by Buckling Nanoprobes Inside an Environmental SEM. In:
ProceedingsoftheIEEE/RSJInternationalConferenceonIntelligentRobotsandSys
tems,1115October2009,St.Louis,USA;2009.
[33] CappelleriD.J,PiazzaG,KumarV.TwoDimensional,VisionbasedNforceSensor
forMicrorobotics.In:ProceedingsoftheIEEEInternationalConferenceonRobotics
andAutomation,1217May2009,Kobe,Japan;2009.
[34] CappelleriD.J,KrishnanG,KimC,KumarV,KotaS.TowardtheDesignofaDecou
pled,TwoDimensional,VisionbasedNForceSensor.JournalofMechanismsand
Robotics2010;202101010210109.
[35] AbbottJ.J,PeyerK.E,LagomarsinoM.C,ZhangL,DongL,KaliakatsosI.K,Nelson
B.J.HowShouldMicrorobotsSwim?TheInternationalJournalofRoboticsResearch
2009;28(1112)14341447.
[36] Pawashe C, Floyd C. S, Sitti M. Modeling and Experimental Characterization of an
Untethered Magnetic MicroRobot. The International Journal of Robotics Research
2009;28(8)10771094.
[37] SakarM.S,SteagerE.B,KimD.H,KimM.J,PappasG.J,KumarV.SingleCellMa
nipulationusing Ferromagnetic Composite Microtransporters, AppliedPhysics Let
ters2010;9604370510437053.
[38] Ikuta K, Sato F, Kadoguchi K, Itoh S. Optical Driven MasterSlave Controllable
NanoManipulatorwithRealTimeForceSensing.In:ProceedingsoftheIEEEInter
nationalConferenceonMicroElectroMechanicalSystems,1317January2008,Tuc
son,Arizona,USA;2008.
[39] AraiF,IchikawaA,OgawaM,FukudaT,HorioK,ItoigawaK.HighSpeedSepara
tionSystemofRandomlySuspendedSingleLivingCellsbyLaserTrapandDielec
trophoresis. In: Proceedings of the IEEE International Conference on Robotics and
Automation,2126May2001,Seoul,Korea;2001.
[40] YamanishiY,SakumaS,OndaK,AraiF.PowerfulActuationofMagnetizedMicro
toolsbyFocusedMagneticFieldforParticleSortinginaChip.BiomedicalMicrodevi
ces2008;10411419.
[41] Hagiwara M, Kawahara T, Yamanishi Y, Arai F. Driving Method of Microtool by
Horizontally Arranged Permanent Magnets for Single Cell Manipulation. Applied
PhysicsLetters2010;9701370110137013.
[42] Hagiwara M, Kawahara T, Yamanishi Y, Masuda T, Feng L, Arai F. OnChip Mag
netically Actuated Robot with Ultrasonic Vibration for Single Cell Manipulations.
LabonaChip2011;1120492054.
[43] KawaharaT,SugitaM,HagiwaraM,YamanishiY,AraiF,KawanoH,ShihiraIshika
wa I, Miyawaki A. OnChip Force Sensing by Magnetically Driven Microtool for
MeasurementofStimulantPropertyofP.laevis.In:ProceedingsoftheIEEEInterna
tional Conference on Robotics and Automation, 913 May 2011, Shanghai, China;
2011.
[44] KhooH.S,LiuK.K,TsengF.G.MechanicalStrengthandInterfacialFailureAnalysis
of Cantilevered SU8 Microposts. Journal of Micromechanics and Microengineering
2003;13822831.
[45] BachmannD,SchoberleB,KuhneS,LeinerY,HieroldC.FabricationandCharacteri
zationofFoldedSU8SuspensionsforMEMSApplications.SensorsandActuatorsA
2006;130131379386.
[46] Makita N, ShihiraIshikawa I. Chloroplast Assemblage by Mechanical Stimulation
andItsIntercellularTransmissioninDiatomCells.Protoplasma1997;1978695.
[47] FurukawaT,WatanabeM,ShihiraIshikawaI.Greenandbluelightmediatedchlor
oplast migration in the centric diatom Pleurosira laevis. Protoplasma 1998;203
214220.
[48] ShihiraIshikawa I, Nakamura T, Higashi S, Watanabe M. Distinct Responses of
ChloroplaststoBlueandGreenLaserMicrobeamIrradiationsintheCentricDiatom
Pleurosiralaevis.PhotochemistryandPhotobiology2007;8311011109.
[49] Hanada Y, Sugioka K, Kawano H, ShihiraIshikawa I, Miyawaki A, Midorikawa K.
NanoAquarium for Dynamic Observation of Living Cells Fabricated by Femtosec
ond Laser Direct Writing of Photostructurable Glass. Biomedical Microdevices
2008;10403410.
107
108
[50] KawaharaT,SugitaM,HagiwaraM,YamanishiY,AraiF.UltraHighSpeedRobot
HandandEyeforInvestigationofMicroorganismsinaChip.In:Proceedingsofthe
15thInternationalConferenceonMiniaturizedSystemsforChemistryandLifeScien
ces,26October2011,Seattle,USA;2011.
[51] IchikawaA,AraiF.OnChipNoncontactActuationofaMicroPipetteDrivenbyPer
manent Magnets. In: Proceedings of the International Conference on Micro Electro
MechanicalSystems,29January2February2012,Paris,France;2012.
[52] SakumaS,AraiF.NoncontactNanometricPositioningofProbeTipforMeasurement
ofMechanicalParametersofCell.In:ProceedingsoftheInternationalConferenceon
MicroElectroMechanicalSystems,29January2February2012,Paris,France;2012.
[53] Hagiwara M, Kawahara T, Iijima T, Arai F. High Speed Microrobot Actuation in a
MicrofluidicChipbyLevitatedStructurewithRibletSurface.In:Proceedingsofthe
IEEE International Conference on Robotics and Automation, 1418 May 2012, St.
Paul,USA;2012.
Chapter 6
Synthesis of Nanomaterials by
Solution Plasma Processing
Osamu Takai, Maria Antoaneta Bratescu,
Tomonaga Ueno and Nagahiro Saito
1.Introduction
Recently,plasmasintheliquidphasehaveattractedagreatattentionbecauseofitsapplications
to industrial materials processing [1]. In particular, glow discharge in the liquid phase
(solutionplasmaSP)isausefultoolforthesynthesisofnanomaterials.Wenamedthe
synthesisprocessusingplasmainliquidsasSolutionPlasmaProcessing(SPP).
Plasmainwaterhasbeenproducedin1899byDr.Wilsingbetweendifferenttypesofmetal
electrodes, in order to study the spectral properties in connection with the astronomical
observations[2].Theearliestworksondischargeinliquidshavestudiedthearcandspark
dischargeinwaterandsaltsolutions[3].
The detailed structure of the solution plasma is still unclear at present. Currently, SPs are
generatedbynanosecondspulsedordcvoltages.Also,acexcitationswithfrequenciesraging
from50HzuptoMHzwereused[47].
Inourresearchgroup,wefocusedonthefundamentalaspectsofthesolutionplasmadiag
nosticsrelatedwiththesynthesisofnovelnanomaterislsprocessedbySP.Figure1showsa
modelofthesolutionplasma.ThecharacteristicregionsofSParetheplasmagasphase,the
liquidphase,theinterfacebetweenplasmagasandliquid,andtheinterfacebetweenelectrode
surfaceandgasplasma.Theemissioncenterofplasmaislocatedinthegasphasewhichis
surroundedbytheliquidphase.Nearthegas/liquidinterfaceanionsheathisformed.The
plasma is confined by a condensed phase, which produces unique features of the solution
plasma. This solution plasma provides extremely rapid reactions using activated chemical
speciesandradicalsunderhighpressure[1].
SPPisanewusefulandsimplemethodforthemetalnanoparticles(NPs)synthesissincethis
nonequilibriumplasmacanprovideextremelyrapidreactionsduetothereactivechemical
2013 Takai et al.; licensee InTech. This is an open access article distributed under the terms of the Creative
110
species,radicals,andUVradiationproducedinanatmosphericpressureplasma[1].Themost
importantmeritsoftheSPPfortheNPssynthesis,ascomparedwithchemicalmethods,consist
in the short processing time (in the range from few minutes to several tens of minutes),
preparation in room temperature and pressure conditions, and low energy of plasma. The
noveltyoftheSPPmethodusedinourlaboratoryconsistsinthefactthatplasmaoperatesin
glowdischargelimits,offeringasuitablemediumtocontrolthechemicalreactionsinsidethe
solutions [10, 11]. This is possible because plasma offers a new reaction medium, where
hydrogen,hydroxyl,andoxygenradicalsareproduced,wherethehydrogenradicalisthe
mostresponsibleforthereductionreactionofthegoldiontotheneutralatom,andtherefore
areductionagentisnotnecessary.TheSPPmethodseemswellsuitedfortheNPssynthesis
offeringthepossibilitytocontrolthesizebycontrollingthesurroundingchemistryofthegold
NPs,addingthusanotherlevelofutilityofthisproceduretomaterialscience[12,13].
SPPwassuccessfullyusedforloadingmetalNPsoncarbonmaterialstopreparecomposite
structureswhichcanimprovethecatalyticactivityoffuelcells[14].
ManyotherapplicationsofSPPinnanomaterialtechnologyhavebeenperformedinourgroup
relatedwithtemplateremovalinmesoporoussilicasynthesisprocess[15],decompositionof
organicdyesorcompounds[16,17],surfacemodificationofmetals[19],andsterilizationof
water[1].
Figure 1. Model of Solution Plasma. The main regions are: plasma gas phase, gas phase, liquid phase and the interfa
ces between plasma and gas, and gas and liquid.
InthefollowingswewilldiscussabouttheconditionsforSPgeneration[16,1921],theroleof
radicals,ionsandsurroundingchemistryinNPssynthesis[11,12,17,22],andsynthesisof
nanomaterialswithenhancedcatalyticactivity[1,14].
Synthesis of Nanomaterials by Solution Plasma Processing
2.ConditionsforSPgeneration
AtypicalexperimentalsetupforthegenerationofSP,withtimedependentelectricaland
optical diagnostics of plasma and synthesis of nanomaterials is shown schematically in
Figure2.AtypicalpowersupplyforSPgenerationhasthefollowingcharacteristics:bipolar
pulsetypewithamaximumvoltageandcurrentof5kVand10A,respectively,avariable
repetitionfrequencyintherangefrom5to60kHz,andavariablepulsewidthfrom500
nsto3s[10,16,1820].
Figure 2. Typical experimental set-up for SP generation, with time-dependent electrical and optical diagnostics of
plasma, and synthesis of nanomaterials.
WeinvestigatedthedependenceofSPcharacteristicsontheinterelectrodedistance.InSP,the
formationofdifferentradicals,theexcitedatomsandmoleculesarestronglyinfluencedbythe
geometryandtheinputelectricalpowerinthesystem.Figure3showsvariousregionsofSP
dependingontheinterelectrodedistanceandtheappliedpulsedhighvoltage.Typicalregions
aretheglowdischargeregime,whentheinterelectrodespaceislessthan2mmandthepeak
voltage is more than 2 kV, the corona discharge regime, when the interelectrode distance
increases and the peak high voltage is also high, and the prebreakdown regime where
electrochemicalreactionsdominate[20].
Innanomaterialsprocessing,animportantfactorconsistsincontrollingofthesolutionplasma
stability.ThevalueofpHandtheconductivityofthesolutionalsodeterminetheoperation
regime of SP. In the diagram from Figure 4 various conditions of plasma determined by
solutionconductivity,bipolarpulsewidthandfrequency,generateagloworanarcdischarge.
Ifthesolutionconductivityishigh,morethan1mS/cm,theioniccurrentthroughtheliquid
ishigh,andatthesameinputelectricalpower,plasmaisinstable,ifthepulsewidthissmaller
111
112
than 2 s. If the pulse width increases, the input electrical power increases and even the
solutionconductivityishigh,astableglowdischargecanbeobtained.
Figure 3. Dependence of SP characteristics on the inter-electrode distance and the applied peak high voltage.
Figure 4. Influence of solution conductivity and input electrical parameters of the pulsed power supply on SP stability.
Different operation regimes of SP: glow discharge and arc discharge. (a) Stability conditions of SPP in a solution of LiCl
with 4 mM concentration. (b) Stability conditions of SPP in a solution of LiCl with 40 mM concentration.
Independentlyofthesolutionconductivity,ifthepulsewidthishigherthan2.5sandthe
pulsedpowersupplyfrequencyis30kHz,theplasmageneratedinsolutionswitchesfastto
arcdischargeregime.
The optical emission spectra strongly depend on SP regime working. Corona discharge is
characterizedbyastrongemissionoftheOHradicalascomparedwiththeglowdischarge
regime,whentheinterelectrodedistanceisaround1mm[20].
Figure 5. (a) Optical absorption spectra of the OH radical measured in HCl, KCl, and KOH solutions, on the positive
applied voltage, under the same experimental conditions (discharge voltage of 700 V, frequency of 25 kHz, and pulse
width of 4 s). (b) The time evolution of the OH radical density measured in HCl, KCl, and KOH solutions, under the
same experimental conditions above mentioned. Lines are used as guides for the eyes.
TheopticalabsorptionspectraoftheOHradicalmeasuredindifferentsolutionwithHCl,
KCl,andKOH,underapositivevalueofthehighvoltage,andthecorrespondingtime
dependentsignalsoftheemissionlineoftheOHradical,arerepresentedinFigure5(a)
and (b), respectively. The OH radical number density measured by broad band absorp
tion spectroscopy was highest for the HCl solution plasma (2x1017 cm3) when positive
voltagepulseswereappliedtotheelectrodes.KOHishighlybasicandcanbeanimpor
tantsourceofhydroxylradicals,butinthisexperimentthedensitywasthelowestforthis
typeofsolution(~5x1016 cm3)[24].
ThemainchemicalreactionsresponsibleforthegenerationofthereactivespeciesinSPP
are[4,5]:
H2O H + OH
2H2O O2 + 2H2
O2 2 O
O2 2 H
H + e H* + e
O + e O* + e
(1)
113
114
3.Roleofradicals,ionsandsurroundingchemistryinnanomaterials
synthesisbySPP
InmetalNPssynthesis,SPPisusedwithoutanychemicalreductionagent,sincehydro
genradicalproducedinplasmagasphaseistransferredinliquidphasewherethemetal
ionisreducedtotheneutralform.Theroleofhydrogenradicalinthereductionprocess
of Au ions from HAuCl4 to Au neutral atoms, which nucleate to produce gold NPs, is
illustratedinFigure6.
Inthisexperiment,solutionswithdifferentionconcentrationsandconductivitieswereused
tosynthesizegoldNPs.Fromtheopticalemissionspectra,wecanobservethatinasolution
withhighconcentrationofions(40mMLiCl),therelativenumberdensityofhydrogenradical
wassmallerthaninthesolutionwith4mMLiClconcentration(Figure6(a)).Thereforeinthese
conditions,asmallernumberofAuionswerereducedtotheneutralformandtherelative
numberdensityofgoldNPswasless,ascanbeobservedfromtheUVvisspectra(Figure6
(b))andphotos(Figure6(c))[10].
Figure 6. Role of hydrogen radical in the reduction reaction of Au ion to neutral form, in the synthesis process of gold
NPs. (a) Optical emission spectra of SP generated in LiCl solution with different molar concentrations. (b) Time evolu
tion of the UV-vis spectra of solutions resulted after SPP, showing gold NPs formation. (c) Photos of the solutions con
taining gold NPs.
Plasmahastheroletoprovidethereactivespecies,especiallytheHradicalsnecessarilyforthe
reductionprocess.InordertoconfirmmoretheroleofHradical,weattemptedtosynthesize
the gold nanoparticles from the solution containing PBN (ntertbuthylphenylnitrone),
whichworksasaspintrapagent.
WhenPBNreactswithHandOH,PBNHandPBNOHadductsareproduced,respectively.
InanESRmeasurement,theseadductscanbedetected.DuringSPP,thePBNHadductwas
Figure 7. Dependence of chemical surrounding on Au NPs prepared in solutions with different pH values. (a) ToF-SIMS
mass spectra of the negative ions fragments Au, AuOH and CNAu, Au2OH, and Au2Cl3. (b) HRTEM images of Au NPs
synthesized in solutions with pH 3 and 12.
moredetectedthanPBNOHadduct.ThisindicatedthatHradicalsaresuppliedfromplasma
tothesolution.Moreover,theformationofgoldnanoparticleswasnotobservedinthesolution
containingPBNsincetheHradicalsproducedinthesolutionwhichrepresentthereducing
agentforthegoldions,weretrappedbyPBNmolecules[11].
SizecontrolledgoldNPshavebeensynthesizedusingSPP.ThegoldNPsexhibitsizesfrom
12nmto10nmwhenthesolutionpHwasadjustedintherangefrom12to3,respective
ly. The chemical environment surrounding the gold NPs depends on the preparation
conditions and determines the electrostatic interaction among the nanoparticles, which
alterstheirfinalsize.InformationobtainedfromXPSanalysis,ToFSIMSmassspectra,and
UVvisibleabsorptionspectroscopywereconsistentanddemonstratethatthegoldNPsare
partially oxidized on the surface, when synthesized in a pH 12 solution, and remain
surrounded by gold chloride compounds when synthesized in a pH 3 solution. Plasma
diagnosticsshowsthatahighelectrondensitycontributestogeneratealargernumberof
hydrogenradicals,whichrepresentthemaincomponentinthereductionprocessofthe
goldionintotheneutralform[18].
Figure7showsthedependenceofchemicalsurroundingonAuNPspreparedinsolutions
withdifferentpHvalues.TheToFSIMSmassspectraofthenegativeionsfragmentswereAu,
AuOHandCNAu,Au2OH,andAu2Cl3.ThechemicalsurroundingofgoldNPssynthesized
inpH3solutionismainlyformedbyAuClspecies,whilethechemicalsurroundingofgold
NPsproducedinpH12solutioniscomposedbyoxidizedformofAuO.Figure7(b)displays
the HRTEM imagesofAuNPs synthesized insolutionswith pH 3and12. We canclearly
observethatsmallsizegoldNPsweresynthesizedinsolutionwithpH12.
ThemainroleofSPPduringthesynthesisofthegoldNPsconsistsintheproductionoftheH
radicalsintheplasmagasphasewhicharenecessarytoreducethegoldionAu3+toatomic
Au0,intheliquidphase.
115
116
Beforeplasma,insolution,thehydrolysisofHAuCl4occurs:
AuCl4 + jOH AuCl4 j (OH)j + jCl
(2)
where0<j<4andthereplacementofClbyOHdependsonsolutionpH,aswasfoundinthe
UVvisibleresults.
Intheliquidphaseofplasma,thereductionofgoldionsoccursindifferentways:
AuCl4 + 3H 3HCl + Au0 + Cl
(3)
Au(OH)4 + 3H 3H2O + Au0 + OH
(4)
wherethehydrogenradicals(H)areprovidedbythegasphaseplasmaandthereaction(3)
and (4) takes place in the pH 3 and pH 12 solution, respectively. The pH 6 solution is an
intermediatecasetotheothertwosolutions,wherethehydroxocomplexesasAuCl3(OH)and
AuCl2(OH)2representamolarfractionofabout0.6inthesolution[17].
BychangingthesolutionpHinthepreparationcondition,thesizeofthegoldNPscanbe
controlled.TheconnectionbetweenthegoldNPssizeandthesolutionpHcanbeunderstood
by:(i)theeffectoftheredoxstandardpotentialinthereactions(3)and(4),(ii)theelectrostatic
repulsionforcebetweenAuOions,and(iii)theprotectivelayerofthesurfactant(Figure7).
WealsoattemptedtoregulatethesizeofthegoldNPsbyamethodbasedondischargein
reverse micelle solutions [12] (Figure 8). The reactive species generated by the discharge
reduced[AuCl4]onlyinsidewaterdropletsinthereversemicellesolutions.Atthelowervalues
ofwatertosurfactantratio(W),theaveragediameterissmallerandthesizedistributionis
narrower.
Thesizeofgoldnanoparticlesvariedfrom4.0to11.4nm.SizeofgoldNPsformedinsidethe
waterdropletswasregulatedbythesizeofreversemicelles.ThissuggeststhatSPPinglow
discharge regime in reverse micelle solutions can be applied as a plasma nanoreactor for
nanomaterialfabrication.
We analyzed the gold NPs synthesized by chemical reduction process and SPP and we
investigatethemicrostructuralcharacteristicsoftheseinSPP[24].Microstructuralcharacter
isticsofgoldnanoparticles(AuNPs)fabricatedbySPPinreversemicellesolutionshavebeen
studiedbyhighresolutiontransmissionelectronmicroscopy(HRTEM).
ThesynthesizedAuNPs,withanaveragesizeof6.31.4nm,havedifferentcrystalcharac
teristics:fccsinglecrystallineparticles,multiplytwinnedparticles(MTPs),andincomplete
MTPs(singlenanotwinnedfccconfiguration).ThecrystalstructurecharacteristicsoftheAu
NPs synthesized by SPP method were analyzed and compared with similarsize Au NPs
obtainedbytheconventionalchemicalreductionsynthesis(CRS)method.TheTEManalysis
results show that the Au NPs synthesized by the CRS method have shapes and crystal
Figure 8. Synthesis of gold NPs in reverse micelle solutions in SPP. (a) Mechanism of reverse micelle solution with dodec
ane as solvent and sodium bis(2-ethylhexyl)sulfosuccinate (AOT), as surfactant. (b) Photos of solutions obtained by SPP,
with different processing times. (c) TEM images of gold NPs synthesized in reverse micelle solutions for two ratios W.
structures similar to those nanoparticles obtained by the SPP method. However, from the
detailedHRTEManalysis,therelativeamountoftheAuMTPsandincompleteMTPstothe
totalamountoftheAuNPssynthesizedbytheSPPmethodwasobservedtobearound94%,
whereastherelativeamountofthesekindsofcrystalstructuresfabricatedbytheCRSmethod
wasabout63%.ItismostlikelythattheenhancedformationoftheAuMTPsisduetothefact
that the SPP method generates highly reactionactivated species under low environmental
temperatureconditions.Figure9showsmicrostructuralcharacteristicsofgoldNPssynthe
sizedinSPP.
Figure 9. Microstructural characteristics of gold NPs synthesized in SPP. The synthesized Au NPs, with an average size
of 6.3 1.4 nm, have different crystal characteristics; fcc single-crystalline particles, multiply-twinned particles (MTPs),
and incomplete MTPs (single-nanotwinned fcc configuration).
117
118
4.Synthesisofnanomaterialswithenhancedcatalyticactivity
Inordertoimprovetheenergyconversionefficiencyinfuelcells,weloadedPtnanoparticles
on carbon nanoballs (CNBs) by SPP by reducing Pt ion on CNB surface via a surfactant
molecule.Inthisstudy,weemployedpoly(vinylpyrrolidone)(PVP)orsodiumdodecylsulfate
(SDS)topreparePtnanoparticlessupportedonCNB(Pt/CNB)bytheSPP,andtheelectro
chemicalpropertiesascatalystwereevaluatedbycyclicvoltammetry(CV)[14].
CNBs were prepared by thermal decomposition process of ethylene and hydrogen gases.
Duringthesynthesisprocess,thecolorofthesolutionchangedfromyellowtodarkbrown
indicatingtheimprovementofthedispersibilityofCNBinsolution.Moreover,TEMimages
andelementalmappingimagesshowedthePtNPssupportedonCNB.
ThecatalyticactivityofthePt/CNBusingSDSassurfactantwasshowntobehigherthanthe
Pt/CNB prepared with PVP system. The SDScontaining Pt/CNB also showed the higher
activitythanthatobtainedbytheconventionalchemicalmethod.
WealsostudiedtheinfluenceofthesolutionpHonthesizeofPtNPssupportedonCNBsand
thecatalyticactivityofthecompositenanomaterial.Figure10showstheinfluenceofsolution
pHonthesizeandcatalyticactivityofsynthesizedPtNPssupportedonCNBs.FromTEM
imagesofPtNPsonCNBsforsolutionswithvariouspHwecanobservethattheaveragesize
ofPtNPsisdecreasingasthesolutionpHincreases.TheHRTEMimageofPtNPoncarbon
showsthecrystallinestructureofPtNPandtheorganicsurroundingsurfactantPVP.Thecyclic
voltammetry measurements of Pt@CNBs show that small size Pt NPs have high catalytic
activity(Figure10(d)).
Figure 10. Influence of solution pH on the size and catalytic activity of synthesized Pt NPs supported on CNBs. (a) TEM
images of Pt NPs on CNBs for solutions with various pH, indicating the average size of Pt NPs. (b) HRTEM image of Pt
NP on carbon, with the organic surrounding PVP. (c) EDS mapping of Pt NPs. (d) Cyclic voltammetry measurements of
Pt@CNBs.
5.Conclusions
SPPisarapid,usefulmethodfornanomaterialssynthesis.Thecomplexityofthephenomena
makesthismethodafieldoffundamentalinvestigations.Abetterknowledgeofthechemical
reactions inside the plasma gas phase, liquid phase, and interfaces can make SPP a more
powerfultoolinnanotechnology.
Acknowledgements
ThisworkwaspartiallysupportedbyTokaiRegionNanotechnologyManufacturingCluster
sponsoredbyMinistryofEducation,Culture,Sports,Science,andTechnology(MEXT),Core
ResearchforEvolutionalScienceandTechnology(CREST)ofJapanScienceandTechnology
(JST)Agency,andGlobalCOEProject,sponsoredbyMinistryofEducation,Culture,Sports,
Science,andTechnology(MEXT)
Authordetails
OsamuTakai1,MariaAntoanetaBratescu1,TomonagaUeno2andNagahiroSaito1
1EcoTopiaScienceInstitute,NagoyaUniversity,Nagoya,Japan
2GreenMobilityCollaborativeResearchCenter,NagoyaUniversity,Nagoya,Japan
References
[1] TakaiO.Solutionplasmaprocessing(SPP).PureandAppliedChemistry2008;80(9)
20032011.
[2] ilsingJ.AstrophysicalJournal1899;10113125.
[3] ockyerN.AstrophysicalJournal1902;15(3)190198
[4] oshiA.A,LockeB.R,ArceP,FinneyW.C.Formationofhydroxylradicals,hydro
genperoxideandaqueouselectronsbypulsedstreamercoronadischargeinaqueous
solution.JournalofHazardousMater1995;41(1)330.
[5] un B, Sato M, Clements J. S. Optical study of active species produced by a pulsed
streamercoronadischargeinwater.JournalofElectrostatics1997;39(3)189202.
[6] unB,SatoM,HaranoA,ClementsJ.S.Nonuniformpulsedischargeinducedradical
productionindistilledwater.JournalofElectrostatics1998;43(2)115126.
119
120
[7] unka P. Pulse electrical discharges in water and their applications. Physics of Plas
mas2001;8(5)2587.
[8] olbJ.F,JoshiR.P,XiaoS,SchoenbachK.H.Streamersinwaterandotherdielectric
liquids.JournalofPhysicsD:AppliedPhysics2008;41(23)234007
[9] ruggemanP,LeysC.Nonthermalplasmasinandincontactwithliquids.Journalof
PhysicsD:AppliedPhysics2009;42(5)053001.
[10] HiedaJ,SaitoN,TakaiO.Sizeregulatedgoldnanoparticlesfabricatedbyadischarge
in reverse micelle solutions. Surface & Coatings Technology 2008;202(2223)
53435346.
[11] SaitoN,HiedaJ,TakaiO.Synthesisprocessofgoldnanoparticlesinsolutionplasma.
ThinSolidFilms2009;518(3)912917.
[12] Hieda J, Saito N, Takai O. Exotic shapes of gold nanoparticles synthesized using
plasmainaqueoussolution.JournalofVacuumScienceandTechnologyA2008;26(4)
854856.
[13] BratescuM.A,SaitoN,TakaiO.Redoxreactionsinliquidplasmaduringironoxide
and oxidehydroxide nanoparticles synthesis. Current Applied Physics 2011;11(5)
S30S34.
[14] Ichin Y, Mitamura K, Saito N, Takai O. Characterization of platinum catalyst sup
portedoncarbonnanoballspreparedbysolutionplasmaprocessing.JournalofVac
uumScienceandTechnologyA2009;27(4)826830.
[15] Pootawang P, Saito N, Takai O. Solution Plasma Process for Template Removal in
Mesoporous Silica Synthesis. Japanese Journal of Applied Physics 2010;49(12)
1262021.
[16] BarochP,AnitaV,SaitoN,TakaiO.Bipolarpulsedelectricaldischargefordecompo
sitionoforganiccompoundsinwater.JournalofElectrostatics2008;66(56)294299.
[17] BratescuM.A,ChoS.P,TakaiO,SaitoN.SizeControlledGoldNanoparticlesSyn
thesized in Solution Plasma. Journal of Physical Chemistry C 2011;115(50) 24569
24576.
[18] BratescuM.A,HiedaJ,UmemuraJ,SaitoN,TakaiO.Analysisoforganicpollutant
degradationinpulsedplasmabycoherentantiStokesRamanspectroscopy.Journal
ofVacuumScienceandTechnologyA2011;29(3)03130217.
[19] AnitaV,SaitoN,TakaiO.Microarcplasmatreatmentoftitaniumandaluminumsur
facesinelectrolytes.ThinSolidFilms2006;506507364368.
[20] Potocky S, Saito N, Takai O. Needle electrode erosion in water plasma discharge.
ThinSolidFilms;518(3)918923.
[21] MironC,BratescuM.A,SaitoN,TakaiO.TimeresolvedOpticalEmissionSpectro
scopy in Water Electrical Discharges. Plasma Chemistry Plasma Processing
2010;30(5)619631.
[22] MironC,BratescuM.A,SaitoN,TakaiO.Opticaldiagnosticofbipolarelectricaldis
charges in HCl, KCl, and KOH solutions, Journal of Applied Physics 2011;109(12)
123301.
[23] Cho S. P, Bratescu M. A, Saito N, Takai O. Microstructural characterization of gold
nanoparticles synthesized by solution plasma processing. Nanotechnology
2011;22(45)4557017.
[24] Miron C,Bratescu M.A,Saito N, Takai O. Effect ofthe electrode work functionon
thewaterplasmabreakdownvoltage.CurrentAppliedPhysics2011;11(5)S154S158.
121
Chapter 7
Tissue Engineering and Regenerative Medicine

Minoru Ueda
1.Introduction
Tissuespecificadultstemcellsarethemosthopefulcellatthismomentforclinicalusebecause
itmayrepresentMotherNaturesrepaircells.Suchcellsarepotentiallypresentwithinallof
thetissueofthebodyandmayremaindormantuntiltheyareactivatedinresponsetotissue
injury.Initially,thechemicalenvironmentatthesiteofanyinjuryisveryhostile.Theseadult
stem cells, having a low oxygen requirement, appear to have the ability to survive this
environment.Whenadequatenumbersofcellshavebeenachievedbymultiplication,theyare
thenprogrammedtomatureandrepairtissuedamageofacertainmagnitude.Ifthisisthe
case,withthedevelopmentofappropriatetissuespecificscaffoldsandtheuseoftheoptimal
celltype,Ibelievethatphysiciansandscientistswillultimatelybeabletorepairorreplaceany
tissueinthehumanbodythatisinjuredordamagedasaresultofdiseaseortrauma.Studies
involvingtheuseofstemcellsandmaturecells,incombinationwithgeneticmanipulation
anddeterminationoftheefficacycellulardeliverysystemsandscaffoldings,shouldbeenable
rapidprogressiontohumantreatments.Itismybeliefthatexploringtheuseofappropriate
vehiclesandcelltypeswillultimatelyleadtoresolutionofstrokesymptoms,suchasparalysis,
and may help reverse symptoms associated with such central nervous system diseases as
ParkinsonsdiseaseandAlzheimersdisease.
2.Researchprojects
Thischapterwaseditedbycollectingalltheachievementperformedinthelaboratoryoforal
and maxillofacial surgery and it brings together the specific experiences of the scientific
communityintheseexperiencesofourscientificcommunityinthisfieldaswellastheclinical
experiencesofthemostrenownedexpertsinthefieldsfromalloverNagoyaUniversity.The
editorsareespeciallyproudofbringingtogethertheleadingbiologistsandmaterialscientists
togetherwithdentist,plasticsurgeonsandsurgeonsofallspecialitiesfromalldepartmentof
themedicalschoolofNagoyaUniversity.Takentogether,thisuniquecollectionofworldwide
2013 Ueda; licensee InTech. This is an open access article distributed under the terms of the Creative
124
expertachievementandexperiencesrepresentsthecurrentspectrumofpossibilitiesintissue
engineeredsubstitution.
2.1.Boneregenerationwithselfassemblingpeptidenanofiberscaffoldsintissue
engineeringforosseointegrationofdentalimplants
Theaimofthisstudywastoevaluatethecorrelationbetweentheosseointegrationofdental
implantsandtissueengineeredboneusingananofiberscaffold,PuraMatrix(PM).Thefirst
molarandallpremolarsinthemandibleregionofdogswereextracted,andthreebonedefects
werepreparedwithatrephinebaronbothsidesofthemandibleafter4weeks.Theexperi
mentalgroupswereasfollows:1)PM,2)PManddogmesenchymalstemcells(MSCs),3)PM,
dogMSCs,andplateletrichplasma(PRP),and4)acontrol(defectonly).Implantswereplaced
withinthepreparedareas8weekslater,andassessedbyhistologicalandhistomorphometric
analyses(boneimplantcontact(BIC)).TheBICsforgroups1,2,3,and4were40.77%,50.35%,
55.64%and30.57%,respectively.ThefindingsindicatethatPMmaybeusefulasascaffoldfor
boneregenerationarounddentalimplants.
Figure 1. Molecular structure of PM. Molecular models of amiphiphilic self-complementary peptide have 16 amino
acids with an alternating polar and nanopolar pattern. A= alanine; R= arginine; D= aspartic acid; + and refer to the
positively and negatively charged residues, respectively (From Kohgo et al. [1]. Reprinted with permission from Quin
tessence Publishing Co, Inc, Chicago).
Figure 2. PM is an injectable scaffold and shows good plasticity. Bar = 5 mm. (From Kohgo et al. [1]. Reprinted with
permission from Quintessence Publishing Co, Inc, Chicago).
Figure 3. Photographs of histologic sections, as seen on light microscopy, 8 weeks after implant placement. (a to c) In
the control group, the buccal and lingual walls were not sufficiently for dental implants. (d to f) In the PM group, slight
bone regeneration in the lingual wall was observed. (g to i) However, slightly more could be seen in the PM/dog MSCs
group. (j to l) On the other hand, the amount of regenerated bone was greatest in the PM/dog MSCs/PRP group (a, d,
g and j, magnification tion rat, e, h, and k, magnification 200; c, f, i, and l, maginification 200) (From Kohgo et al. [1].
Reprinted with permission from Quintessence Publishing Co, Inc, Chicago).
BIC = bone-to-implant contact; SD = standard deviation; PM = PuraMatrix; dMSCs = dog mesenchymal stem cells; PRP =
Platelet-rich plasma * P < 0.01 **P < 0.05 (From Kohgo et al. 2011. Reprinted with permission from Quintessence
Publishing Co, Inc, Chicago).
Table 1. Results of BIC (mean SD)
125
126
Theresultssuggestthattissueengineeredbonecanintegratewellarounddentalimplants.PM
isa3Dstructuremayhavethepotentialtobeascaffoldapplicableinbonetissueengineering.
2.2.Selfassemblingpeptidenanofiberscaffolds,plateletrichplasma,andmesenchymal
stemcellsforinjectableboneregenerationwithtissueengineering
The ideal biomaterial scaffolds, particularly in craniomaxillofacial, plastic, or orthopedic
fieldswithcomplicatedbonedefectshapes,shouldhaveexcellentplasticityforfittinginto
complexdefectshapesandtherateofabsorptionneedstobefastinordertoavoidinfection.
Moreover, these materials and their internal structures may not provide a particularly
favorable environment for cell survival and bone regeneration, and internal, microscale
environment of these materials needs to serve as an extracellular matrix (ECM). ECM is a
dynamicorganizednanocompositethatnotonlyprovidesmechanicalsupportforembedded
cells but also interacts with cells and promotes and regulates cellular functions such as
adhesion,migration,proliferation,anddifferentiationandisconsequentlyinvolvedinthree
dimensionalmorphogenesis.ThisstudyconsideredtheuseofthematrixmaterialPM,which
issynthesizedbychemicalpeptidemethodsandhassimilaritiestothefibersandporesizes
foundintheECM.WeinvestigatedacapabilityofPMasascaffoldforboneregenerationin
combinationwithdogMSCsand/orPRPusingtissueengineeringandregenerativemedicine
technology.First,teethwereextractedfromanadulthybriddogsmandibleregion.After4
weeks,bonedefectswerepreparedonbothsidesofthemandiblewithatrephinebar.The
listedgraftmaterialswereimplantedintothesedefects:1)control(defectonly),2)PM,3)PM/
PRP,4)PM/dogMSCs,and5)PM/dogMSCs/PRP.At2,4,and8weeksafterimplantation,
eachsamplewascollectedfromthegraftareawithatrephinebarandassessedbyhistological
andhistomorphometricanalyses.
From histological evaluation, It was observed that the bone regenerated by PM/dog
MSCs/PRP was excellent quality, and it was found that mature bone had been formed.
Histometrically,at8weeksnewlyformedboneareascomprised12.391.29%(control),25.28
3.92%(PM),27.723.15%(PM/PRP),50.073.97%(PM/dogMSCs),and58.435.06%(PM/
dog MSCs/PRP). The PM/dog MSCs and PM/dog MSCs/PRP groups showed a significant
increaseatallweekscomparedwiththecontrol,PM,orPM/PRP.Theseresultsshowedthat
MSCs might keep their own potential and promote new bone regeneration in the three
dimensionalstructurebyPMscaffolds.Takentogether,itissuggestedthatPMmightbeuseful
asascaffoldofboneregenerationincelltherapy,andtheseresultsmightleadtoaneffective
treatmentmethodforbonedefects.Inthefuturethistissueengineeredbonegraftingmaterial
couldbeusedasaminimallyinvasivemethodinsteadoftraditionalgraftingprocedures.
2.3.Effectsofselfassemblingpeptidehydrogelscaffoldonboneregenerationwith
recombinanthumanbonemorphogeneticprotein2
Various biomaterials have been tested as scaffolds for bone regeneration, such as beta
tricalciumphosphate,hydroxyapatite,andpolymers.However,ascaffoldhasstillnotbeen
foundthathasthecharacteristicsofbiologicsafety,absorbability,cellinteraction,andbone
inductivity. A selfassembling peptide hydrogel scaffold is made of artificial synthetic
Figure 4. Schema of the experimental protocol (From Yoshimi et al. [2, 3]. Reprinted with permission).
127
128
Figure 5. Histological evaluation of control, PM, PM/PRP, PM/dog MSCs, and PM/dog MSCs/PRP implantations at
each time point (center figure:lower magnification, right figure: higher magnification). Sections of representative im
plants are shown from the respective group. The sections were stained with hematoxylin and eosin. Original magnifi
cation, 25 for all photographs. (A) 2 weeks Control group, (B) 4 weeks Control group, (C) 8 weeks Control group, (D)
2 weeks PM group, (E) 4 weeks PM group, (F) 8 weeks PM group, (G) 2 weeks PM/PRP group, (H) 4 weeks PM/PRP
group, (I) 8 weeks PM/PRP group, (J) 2 weeks PM/dog MSCs group, (K) 4 weeks PM/dog MSCs group, (L) 8 weeks
PM/dog MSCs group, (M) 2 weeks PM/dog MSCs/PRP group, (N) 4 weeks PM/dog MSCs/PRP group, and (O) 8 weeks
PM/dog MSCs/PRP group (From Yoshimi et al. [4, 5]. Reprinted with permission).
materialsfeaturingbiologicsafetyandabsorbability.PMisexpectedtobeacandidateasa
scaffoldforboneregeneration.Recombinanthumanbonemorphogeneticprotein2(rhBMP2)
hasexhibitedhighosteogenicactivityinexperimentalstudies.Theobjectiveofthispilotstudy
was to histologically evaluate bone regeneration using a selfassembling peptide hydrogel
scaffoldwithrhBMP2ontheboneaugmentationinarabbitcalvariamodel(Figure6).
Figure 6. Photographs of surgical procedure. In the left and right parietal and frontal bones, 4 circular slits were pre
pared. Five holes were prepared in outer cortical bone inside this circle (a). Apical end of the 4 titanium cylinders was
pressed into each slit, and primary fixation was obtained (b). Four titanium cylinders were filled with respective materi
als (c). The top of the cylinders was closed with a titanium lid (d) (From Ikeno et al. [4]. Reprinted with permission from
Quintessence Publishing Co, Inc, Chicago).
Figure 7. Higher magnification of a histological section Bar = 100m. NB, newly formed bone; CT, connective tissue;
BV, blood vessel; OB, osteoblast-like cells (From Ikeno et al. [4]. Reprinted with permission from Quintessence Publish
ing Co, Inc, Chicago).
Newboneformationseemedtooccurfromthecalvarialbonethroughtheperforationsinthe
outer cortical bone; newly formed bone was also observed in all groups. Under higher
magnification,newlyformedbone,includingcellsandsomebloodvessels,wasobservedin
theconnectivetissueandtheedgesofbonewerelinedwithosteoblastlikecellsinthecylinder
(Figure8).RegeneratedtissueinthesampletreatedwithPM/rhBMP2wasobservedinabout
twothirdsofthecylinders(Figure9).
HistomorphometricanalysisshowedthatregeneratedtissueinthecylinderwithPM/rhBMP2
wassignificantlyincreasedcomparedtotheemptycontrol(Fig.8).Themeanareavaluesof
regeneratedtissueinthecylinderswere35.80%10.35%(control),47.94%5.65%(rhBMP2),
48.94%11.33%(PM),and58.06%14.84%(PM/rhBMP2).Themeanareavaluesofnewly
formedboneinthecylinderswere9.39%4.34%(control),14.03%2.25%(rhBMP2),13.99%
2.15%(PM),and16.61%3.79%(PM/rhBMP2).NeitherrhBMP2norPMalonesignificantly
enhancedboneregenerationcomparedtotheemptycontrolcylinder.
129
130
Figure 8. Histological sections of cylinders with empty control (a), rhBMP-2 (b), PM (c), and PM/rhBMP-2 (d) (toluidine
blue stain) (From Ikeno et al. [4]. Reprinted with permission from Quintessence Publishing Co, Inc, Chicago).
Figure 9. Regenerated tissue area measured by image analysis. The percentage of area of regenerated tissue (left),
maximum height of newly-formed bone (middle), and area of newly-formed bone (right) in the cylinder with each
group was displayed. Bar = SD. Analysis of variance (ANOVA) (From Ikeno et al. [4]. Reprinted with permission from
Quintessence Publishing Co, Inc, Chicago).
PM,asyntheticselfassemblingpeptide,usedincombinationwithrecombinanthumanbone
morphogeneticprotein2significantlyenhancedboneregenerationinaboneaugmentation
modelinrabbits.PMpromisestobeanalternativesyntheticmaterialasausefulcarrierfor
recombinanthumanbonemorphogeneticproteinforboneregeneration.
2.4.Astudyofbonehealingaroundthetitaniumscrewimplantsinosteoporosis:Can
sandblastedsurfacecontributetotheimplantstability?
Dental implants are widely performed as a prosthetic treatment in edentulous patients.
However,intheagedpatients,therearevarioussystemicriskfactorssuchaosteoporosis.The
aim of this study was to investigate whether estrogen deficiency interrupts bone healing
aroundtitaniumimplantsandtoevaluatewhetherbonehealingaroundimplantsunderthe
conditionofestrogendeficiencyisaffectedbyimplantsurfacevariance.FortyeightfemaleSD
ratsweredividedintotwogroups:ovariectomizedrats(OVX;n=24)andshamoperatedrats
(SHAM;n=24).Eachgroupwasfurtherdividedintotwogroups:amachinepolishedimplants
placedgroupandasandblastedimplantsplacedgroup.Bothimplantswereplacedintherat
leftfemurcentrifugalsite84daysafterOVXorshamsurgery.After28,56days,theratswere
killed,andnondecalcifiedsectionwasobtained.BICandbonearea(BA)aroundtheimplants
wereassessedwithcorticalboneandcancellousbone.Furthermore,bonedensity(BD)was
evaluatedina500mmwidezoneofcancellousbonelateraltotheimplants(Figure10).At28
and 56 days after implantation, no significant difference was found between the OVX and
SHAMgroupsforBICandBAincorticalbone.BIC,BA,andBDwithcancellousbonewas
lowerinOVXgroupthaninSHAMgroup.However,BICandBAtendedtoimprovebythe
varianceofimplantsurface(Figure10.11.12.13).
Figure 10. Histomorphological observation and morphological evaluations (From Tateishi et al. [5, 6]. Reprinted with
permission from Quintessence Publishing Co, Inc, Chicago).
131
132
Figure 11. Bone to implant contact (BIC) at 28 days after implantation (From Tateishi et al. [5, 6]. Reprinted with per
mission from Quintessence Publishing Co, Inc, Chicago).
Figure 12. Bone area (BA) inside the implant thread at 28 days after implantation (From Tateishi et al. [5, 6]. Reprinted
with permission from Quintessence Publishing Co, Inc, Chicago).
Figure 13. Bone density (BD) around the implants at 28 days after implantation (From Tateishi et al. [5, 6]. Reprinted
with permission from Quintessence Publishing Co, Inc, Chicago).
Estrogen deficiency affected bone healing and bone density around the titanium implants,
especiallyincancellousbone,butsandblastedsurfacepropertyhasthepossibilitytoimprove
osseointegration. However, the positive effect by rough surface property is limited on the
implantsurface.
2.5.Osteogenicinductionofbonemarrowderivedstromalcellsonsimvastatinreleasing,
biodegradable,nanotomicroscalefiberscaffolds
Tissueengineeringisaneffectiveapproachforthetreatmentofbonedefects.Statinshavebeen
demonstratedtopromoteosteoblasticdifferentiationofbonemarrowderivedMSCs.Electro
spunbiodegradablefibershavealsoshownapplicabilitytodrugdeliveryintheformofbone
tissueengineeredscaffoldswithnanotomicroscaletopographyandhighporositysimilarto
the natural ECM. The aim of this study was to investigate the feasibility of a simvastatin
releasing,biodegradable,nanotomicroscalefiberscaffold(SRBFS)forbonetissueengineer
ingwithMSCs.SimvastatinwasreleasedfromSRBFSslowly(Figure14).MSCswereobserved
tospreadactivelyandrigidlyadheretoSRBFS.MSCsonSRBFSshowedanincreaseinalkaline
phosphataseactivity2weeksaftercellculture(Figure15).Furthermore,osteoclastogenesis
wassuppressedbySRBFSinvitro(Figure16ac).Thenewboneformationandmineralization
intheSRBFSgroupweresignificantlybetterthaninthebiodegradablefiberscaffold(BFS)
withoutsimvastatin12weeksafterimplantationofthecellscaffoldconstructintoanectopic
siteonthemurineback(Figure17ac).TheseresultssuggestthatSRBFSpromotedosteoblastic
differentiationofMSCsinvitroandinvivo,anddemonstratefeasibilityasaboneengineering
scaffold.
Figure 14. Cumulative release of SRBFS in vitro (From Wadagaki et al. [7, 8]. Reprinted with permission).
133
134
Figure 15. ALP activity of MSCs on SRBFS and BFS measured on days 7 and 14 (From Wadagaki et al. [7, 8]. Reprinted
with permission).
Figure 16. Simvastatin released from SRBFS inhibits RANKL-induced osteoclastogenesis 5 days after stimulation. Cells
were cultured for 5 days with (a) SRBFS and (b) BFS after RANKL treatment and stained for TRAP expression. Bar = 500
m. (c) The total number of TRAP-positive multinucleated osteoclasts (i.e., those containing three nuclei) per well were
counted (From Wadagaki et al. [7, 8]. Reprinted with permission).
Figure 17. (a) Bone formation area analyzed by H&E staining 12 weeks after implantation. (b) Percent BSP (bone sialo
protein) -positive cells as determined by immunohistochemical staining with BSP 12 weeks after implantation. (c) Ac
cumulated calcium content per sample 12 weeks after implantation (From Wadagaki et al. [7, 8]. Reprinted with per
mission).
2.6.Conditionedmediafrommesenchymalstemcellsenhancedboneregenerationinrat
calvarialbonedefects
Recentlytissueengineeringhasbecomeavailableasatreatmentprocedureforboneaugmen
tation.However,thisprocedurehasseveralproblemssuchasanexpensivecostforcapital
investmentandcellculture,complicatedsafetyandqualitymanagementofcellhandlingand
invasivenessofcellcollectionforpatients.Ontheotherhand,itwasreportedthatthestem
cellssecretedmanygrowthfactorsandchemokinesduringtheircultivationandthatcould
affectonthecellularcharacteristicsandbehavior.Thisstudyinvestigatedtheeffectofstem
cellculturedconditionedmediaonboneregeneration(Fig.18).Culturedconditionedmedia
fromhumanbonemarrowderivedmesenchymalstemcells(MSCCM)enhancedthemigra
tion,proliferationandexpressionofosteogeneticmarkergenes,suchasosteocalcinandRunx2,
ofratMSCsinvitro.MSCCMincludedcytokinessuchasinsulinlikegrowthfactor(IGF)1
andvascularendothelialgrowthfactor(VEGF).
Invivo,apreparedbonedefectofaratcalvarialmodelwasimplantedinfivedifferentrat
groupsusingoneofthefollowinggraftmaterials:humanMSCs/agarose(MSCs),MSCCM/
agarose (MSCCM), and defect only (Defect). After 4 and 8 weeks, implant sections were
evaluatedusingmicrocomputedtomography(microCT)andhistologicalanalysis.MicroCT
analysis indicated that the MSCCM group had a greater area of newly regenerated bone
comparedwiththeothergroups(P<0.05)(Figure19)andhistologicalanalysisat8weeks
135
136
indicatedthatthenewlyregeneratedbonebridgealmostcoveredthedefect.Interestingly,the
effectsofMSCCMwerestrongerthanthoseoftheMSCsgroup.
Figure 18. Outline of the experimental protocol.
Figure 19. Micro-CT analysis of bone regeneration following implantation of MSC-CM or controls into a bone defect.
InvivoimagingalsoshowedthatmigrationofinjectedrMSCstothebonedefectintheMSC
CMgroupwasgreaterthanintheothergroups(Figure20).
Figure 20. In vivo imaging of infected rat MSCs migration to implants.
These results demonstrated that MSCCM can regenerate bone through mobilization of
endogenousstemcells.Theuseofstemcellculturedconditionedmediaforboneregeneration
willbeauniqueconceptthatutilizesparaclinefactorsofstemcellswithoutcelltransplantation.
2.7.RecoveryofneocallusformationintheHDOgapbytheangiogenicactivitiesofSDF1
Distractionosteogenesis(DO)isauniquetherapythatinducesskeletaltissueregeneration
withoutstem/progenitorcelltransplantation.AlthoughtheselfregenerationpropertyofDO
providesmanyclinicalbenefits,thelongtreatmentperiodrequiredisamajordrawback.A
highspeedDOmousemodel(HDO),inwhichthedistractionwasdonetwotimesfasterthan
incontrolDO(CDO)mice,failedtogeneratenewbonecallusintheDOgap.Wefoundthat
thiswascausedbytheunsuccessfulrecruitmentofbonemarrowendothelialcells(BMECs)/
endothelialprogenitorcells(EPCs)intothegap.Wethentestedtheabilityofalocalapplication
of stromal cellderived factor1 (SDF1), a major chemoattractant for BMECs/EPCs, to
acceleratetheboneregenerationinHDO.Ourdatashowedthat,inHDO,SDF1induced
callusformationinthegapthroughtherecruitmentofBMECs/EPCs,thematurationofneo
bloodvessels,andincreasedbloodflow.Theseresultsindicatethattheactiverecruitmentof
endogenous BMECs/EPCs may provide a substantial clinical benefit for shortening the
treatmentperiodofDO.
137
138
Figure 21. Traction speed affects new bone callus formation in mouse DO models.(A, B) Surgical procedure for the
mouse DO model. An anterior longitudinal incision was made on the right leg of the animal. Needles were inserted
through the skin into the proximal and distal metaphysis of the tibia (A). Subsequently, sets of needles were fixed to
the custom-made fixator with acrylic resin (B). After polymerization of the resin, osteotomy was carried out at the mid
dle of the diaphysis (arrow). (C) Distraction protocols and experimental design. After a 5-day latency period, distrac
tion was started at a rate of 0.2 mm/12 h (C-DO) or 0.4 mm/12 h (H-DO). The lengthening was continued for 8 days in
the C-DO and 4 days in the H-DO model, resulting in a length increase of 3.2 mm. Black arrowheads indicate the time
points of sacrifice. White arrowheads indicate the time points for injecting 200 ng SDF-1 protein (+SDF-1). LA, latency
period; AD, active distraction period; CO, consolidation period. (DI) Representative micrographs of sections displaying
the DO gap stained with HematoxylinEosin (HE) (D, F and H) and Alcian BlueFast red (E, G and I) (n = 8). The left and
right of each figure correspond to the end of the distal and proximal bone fragment, respectively. Neo-callus forma
tion was evident within the C-DO gap at the end of the consolidation period (D). A little cartilage was observed in the
periosteal but not in the endosteal region (E). The H-DO gap was filled with fibrotic tissues and periosteum-derived
cartilages (F, G). The local administration SDF-1 rescued the callus formation in the H-DO gap (H, I). Bar = 300 m (DI).
(From Fujio et al. [9]. Reprinted with permission).
WetestedwhetherthelocaladministrationofSDF1,achemoattractantforBMECs/EPCs,
wouldrescuethedisruptedcallusformationandintegrationofBMECs/EPCsintheHDO
gap.Acollagengelmatrixcontaining200ngofSDF1proteinwasinjectedintotheHDOgap
every other day (Figure 21C). We found that the high levels of SDF1 rescued the callus
formationandincreasedthenumberofBMECs/EPCsintheHDOgap(Figures21H,Iand
22C).Althoughabout80%oftheCD31+cellscoexpressedSca1intheCDOgap,intheH
DOgaptreatedwithSDF1,50%oftheCD31+cellswerenegativeforSca1,suggestingthat
someoftheBMECs/EPCsrecruitedbySDF1hadalreadydifferentiatedintomatureendo
thelialcells(Figure22E).
Insummary,ourstudydemonstratedthatlocallyadministeredSDF1promotestherecruit
mentofendogenousBMECs/EPCsandneocallusformationintheDOgap.Weproposethat
the regulation of endogenous stem/progenitor cell trafficking is a powerful therapeutic
strategyinskeletalregeneration.
Figure 22. Contribution of BM-ECs/EPCs to DO healing. Mice were sacrificed at the middle of the active distraction
period in each group: day 9 and day 7, respectively, for the C-DO and H-DO group. (A, B, and C) The recruitment of
BM-ECs/EPCs was evaluated by immunofluorescence staining for CD31 (red). CD31+ BM-ECs/EPCs accumulated in the
C-DO (A), but not in the H-DO gap (B). Local administration of SDF- 1 rescued the recruitment of BM-ECs/EPCs to the
H-DO gap (C) (n = 8). (D) Boxed area in (C) is shown in higher-magnification micrographs, in which the CD31 signals
are seen together with Sca-1 (green). Note that SDF-1 treatment increased the number of CD31+Sca-1 cells in the gap.
(E) CD31-single-positive and CD31+Sca-1+ cells were counted by Image J software. The number of CD31+Sca-1+ BMECs/EPCs in the C-DO gap was significantly higher than that in intact bone marrow, whereas that in the H-DO gap was
significantly lower. SDF-1 treatment rescued the number of BM-ECs/EPCs in the H-DO gap. The dotted line represents
native bone. Data represent the mean SD. **P > 0.01 and *P > 0.05. Intact: intact bone marrow. Bar = 100 m (A, B,
C) and 50 m (D) (From Fujio et al. [9]. Reprinted with permission).
2.8.EffectofGDF5andBMP2ontheexpressionoftendo/ligamentogenesisrelated
markersinhumanPDLderivedcells
The effect of growth differentiation factor 5 and bone morphogenetic protein 2 on human
periodontalligamentderivedcellswasinvestigatedwithspecialreferencetotendo/ligamen
togenesisrelatedmarkers.TheresultsfromthisstudyshowedthatbothGDF5andBMP2
affectthedifferentiationofPDLderivedcellsinvitro.However,theeffectonthosedifferential
ligamentmakerswasnotidenticalandtheunderlyingmechanismsmightbecomplex.This
study focused on the tendo/ligamentogenesis related markers and confirmed the effect of
thosefactorsnotonlyoncrudePDLderivedcellsbutalsoonSTRO1+andSTRO1PDL
derivedcells.TheresultsfromourstudyshowedsomepotentialbeneficialeffectofGDF5on
periodontaltissueregeneration.However,theunderlyingmechanismsappeartobecompli
cated,andtheoverallbenefitoftheclinicalapplicationofthefactorrequiresfurtheranalyses.
139
140
Figure 23. ALP activity of crude PDL-derived cells in passage 2. ALP activity of the Dex group was significantly greater
compared with control, BMP-2 and GDF-5 groups. *P < 0.05. Values are the mean standard deviation of five experi
ments (From Inoue, M. et al. [10]. Reprinted with permission).
Figure 24. Quantitative RT-PCR analysis for scleraxis (a) and tenomodulin (b) gene expression of crude PDL-derived
cells in passage 2. Cells were cultured with culture medium with or without BMP-2, GDF-5 for 1, 3 and 7 days. There
were no significant differences among control, BMP-2 and GDF-5 groups on any time points. Values are the mean
standard deviation of five experiments (From Inoue, M. et al. [10]. Reprinted with permission).
Figure 25. Western blot analyses of scleraxis crude PDL-derived cells (a) and STRO-1-PDL-derived cells (b) at passage 2.
Expression of scleraxis was detected in all samples. An experiment representative of five similar studies is shown. (c)
GDF-5 treated crude PDL-derived cells had significantly higher scleraxis expression than the other groups. (d) There
was a similar tendency to crude PDL-derived cells, but no significant difference in STRO-1- PDL-derived cells among all
groups. Values are the mean standard deviation of five experiments (From Inoue, M. et al. [10]. Reprinted with per
mission).
Figure 26. The results from western blot analyses of scleraxis in crude PDL-derived cells,STRO-1+ and STRO-1- PDL-de
rived cells at passage 2. All samples were treated with rmGDF-5. (a) The expression of scleraxis proteins was detected in all
groups for 7 days. An experiment representative of six similar studies is shown. (b) Reduced scleraxis expression in
STRO-1+PDL-derived cells was statistically significant compared to crude P2 and STRO-1- PDL-derived cells. Values are the
mean standard deviation of six experiments (From Inoue, M. et al. [10]. Reprinted with permission).
141
142
2.9.TransientTWEAKoverexpressionleadstoageneralsalivaryepithelialcellproliferation
Tumornecrosisfactorlikeweakinducerofapoptosis(TWEAK)isamultifunctionalcytokine
thathasproapoptotic,proangiogenicandproinflammatoryeffects.Inliver,TWEAKleads
toproliferationofprogenitorovalcells,butnotofmaturehepatocytes.Thisstudyevaluated
thehypothesisthatTWEAKoverexpressioninsalivaryglandswouldleadtotheproliferation
ofasalivaryprogenitorcell.Arecombinant,serotype5adenoviralvectorencodinghuman
TWEAK,AdhTWEAK,wasconstructed,initiallytestedinvitro,andthenadministeredtomale
Balb/cmiceviacannulationofWhartonsduct.TWEAKexpressioninvivowasmonitoredas
proteinsecretedintosalivaandserumbyenzymelinkedimmunosorbentassays.Salivarycell
proliferationwasmonitoredbyproliferatingcellnuclearantigenstainingandapoptosiswas
monitored using TUNEL staining. AdhTWEAK administration led to a dosedependent,
transient TWEAK protein expression (Figure 27), detected primarily in saliva. Salivary
epithelialcellproliferationwasgeneralized,peakingonapproximatelydays2and3(Figure
28,29).TWEAKexpressionhadnodetectableeffectonapoptosisofsalivaryepithelialcells.
TransientoverexpressionofTWEAKinmurinesalivaryglandsleadstoageneralproliferation
ofepithelialcellsvsaselectivestimulationofasalivaryprogenitorcell.
Figure 27. Effect of the AdhTWEAK dose administered on the detection of hTWEAK in murine saliva and serum (From
Sugito et al. [11]. Reprinted with permission).
Figure 28. Detection of proliferating cell nuclear antigen (PCNA) staining in submandibular glands of mice following
AdhTWEAK administration. AdhTWEAK (109 particles per gland), or saline, was delivered to both submandibular
glands (n = 4 mice per group) and PCNA staining performed on gland sections as described in Materials and methods
to evaluate cell proliferation. Brown staining represents PCNA-positive nuclei. Sections are counterstained with hema
toxylin. (a) Day 0 after saline administration; (b) Day 1 after AdhTWEAK administration; (c) Day 2 after-AdhTWEAK
administration; (d) Day 3 after AdhTWEAK administration. Bar = 20 m (From Sugito et al. [11]. Reprinted with permis
sion).
143
144
Figure 29. Quantification of AdTWEAK-induced salivary epithelial cell proliferation (From Sugito et al. [11]. Reprinted
with permission).
2.10.Injectablesofttissueaugmentationbytissueengineeringandregenerativemedicine
withhumanmesenchymalstemcells,plateletrichplasma,andhyaluronicacidscaffolds
Therestorationofsofttissuebyadequateimplantmaterialisneededincaseoffunctionaland
aestheticimpairmentsfromlossofsoftconnectivetissue.Theimplantsvary,andrepeated
injectionsaregenerallyrequiredforvaryingdegreesofmaterialresorption.Recentlyitwas
reportedthatautologouscellinjectionwasusefultoimprovesofttissue.Theaimofthisstudy
wastoevaluatethepossibilityofsofttissueaugmentationadoptingtissueengineeringand
regenerativemedicine(TERM)technologyforalongerdurationofinjectedimplants.TERM
isthecombinationandreorganizationofthreetypesofinjectionmaterialstoregenerateorgans
or tissues: 1) living cells, including cultured human MSCs or human fibroblasts (Fibro); 2)
scaffoldsofhyaluronicacid(HA);and3)growthfactorsofPRP.Theexperimentalcombina
tionswereasfollows:HA,HA/Fibro,HA/MSCs,HA/PRP,HA/PRP/FibroandHA/PRP/MSCs.
Thesewereintradermallyinjectedintoimmunodeficientratsandevaluatedbyhistological
analysis,thepercentageoforiginalvolumeandthemaintenancevolume(Figure30).
Thepercentageoforiginalvolumevaluesat14daysshowedsignificantdifferencesbetween
groupswithandwithoutPRPuponcomparison(Table2).Asforthemaintenancevolume
values,HA/PRP/Fibro,andHA/PRP/MSCsfrom7to14dayswerehigherthanothers(Figure
31). HA/PRP/MSCs groups maintained the shape and dimensions of the injected implant,
indicatingthattheinjectedcellsproducedtypeIcollagen(Figure32).Thefindingssuggest
thatasofttissueengineeredprocedurewithMSCsmaybeusefulforlongerlastingsofttissue
augmentation.
Figure 30. Schema of experimental protocol (a) and dermal mound formed after sample injection and the measure
ment diagram (b) (From Okabe et al. [12]. Reprinted with permission).
145
146
Figure 31. The percentage of maintenance volume values (%) in HA/PRP, HA/PRP/Fibro and HA/PRP/MSCs. Bar = SD.
*P < 0.05 (From Okabe et al. [12]. Reprinted with permission).
Figure 32. Distribution of type I collagen produced by MSCs (a) or Fibro (b) in the HA/PRP gels. Bar = 50 m (From
Okabe et al. [12]. Reprinted with permission).
2.11.Potentialityofnewcelltherapyforskinregenerationinwoundhealing
Theprocessofwoundhealingdependsuponavarietyofinteractionsbetweencellsandthe
extracellularmatrix.Thereishyaluronicacidinoneoftheextracellularmatrix.Itiswellknown
thathyaluronicacidnotonlysupportstissuearchitectureasapassivestructuralcomponent
ofthematrixinvariousconnectivetissuesbutisalsoinvolvedindynamiccellularprocesses
duringwoundhealing.Recently,celltherapieswhichisalowaggressionispaidattentionin
placeofsurgicaloperationandmedicaltreatmentofwoundhealing.Itiswellknownthat
human dental pulp cell shows the property that resembled MSCs. In this research, we
examinedthebenefitofstemcellsfromhumanexfoliateddeciduousteeth(SHEDs)inwound
healing(Figure33).
Figure 33. Schema of experimental protocol (From Nishino et al. [13]. Reprinted with permission).
Figure 34. Wound measurement (From Nishino et al. [13]. Reprinted with permission).
SHEDsandMSCssignificantlyacceleratedwoundclosurecomparedwithFibroandcontrol
treatment(Figure34).Atday7and14,theevaluationbyfluorescencemicroscopeshowedthat
PKH 26 positive cells (Fibro, MSCs, SHEDs) were surrounded by hyaluronic acid binding
proteinproductionofhyaluronicacid.Theplasmamembraneoftransplantedcellsshowed
redfluorescencebyPKH26.Hyaluronicacidwasvisualizedwithgreenfluorescence.DAPI
wasusedtovisualizethenuclei(bluefluorescence)(Figure35).
147
148
Figure 35. Histological evaluation (From Nishino et al. [13]. Reprinted with permission).
Table 2. Hyaluronic Acid Contents at Day 7 and 14 (From Nishino et al. [13]. Reprinted with permission).
Quantitate of hyaluronic acid was determined by measuring with ELISA. Significantly

increasedamountsofhyaluronicacidinwoundedtissueswereobservedatday7and14in
MSCs,SHEDs,andFibroascomparedwithcontrol(P<0.05).Thisstudydemonstratedthat
deciduousteeth,consideredasmedicalwaste,wouldbenoveltherapeuticapproachesinthe
treatmentofwoundsandnewstemcellsourceforwoundhealing.
2.12.UmbilicalcordWhartonsJelly:Anewpotentialcellsourceofmesenchymalstemcells
forwoundhealing
Neonatalcongenitaldisease,suchascleftandlippalate,involvessofttissuedefectaswellas
skeletalabnormality.Thusthedevelopmentofthetherapeuticapproachesacceleratingboth
skeletalregenerationandwoundhealingisvaluableforthetreatmentofneonatalcongenital
abnormality(Figure36).Umbilicalcordsareroutinelydiscardedasmedicalwastesinclinic.
WehavesucceededinisolatingstemcellsfromtheWhartonsJellyinumbilicalcords(socalled
umbilicalcordmatrixstemcells:UCMSCs).UCMSCsexhibitedmultipotentialdifferentiation
activities.Inthisresearch,wefocusedontheisolationandidentificationofMSCsfromthe
Whartonsjellyofumbilaicalcord.TheeffectoflocalinjectionofUCMSCsoncutenaouswound
healingwasexaminedusingexcisionalwoundmodelinmice.
Figure 36. New concept of treatment of cleft lip & plate (From Shohara et al. [14, 15]. Reprinted with permission).
Wehavesucceededinisolatingstemcellsfromumbilicalcords(Figure37A,B).Proliferation
of UCMSCs was significantly greater than that of bone marrow stromal cells. UCMSCs
proliferatedmuchfasterthandidbonemarrowstromalcells.UCMSCsexhibitedmultipoten
tialdifferentiationactivitiestowardosteogenic,adipogenic.WefoundthatUCMSCsshared
most of their immunophenotype with bone marrow stromal cells, including positivity for
CD90, CD73, CD105, but negativity for CD11b, CD34(endothelial progenitor cell marker),
CD45(hematopoieticmarkers),andHLADR.ItshowsthatUCMSCsexpressedhighlevelsof
mesenchymal stem cell markers. To investigate the wound repair activity of UCMSCs, we
transplanted them in mouse excisional wound splinting model and the acceleration of the
woundclosurewasevaluated(Figure38).WefoundthatthewoundreceivingtheUCMSCs
exhibitsignificantlyfasterhealingcomparedwithPBSinjectedcontrol(Figure39A).After14
daysfromoperation,theclosedwoundareawas99.720.17%intheUCMSCsgroup,while
thatinthecontrolwas82.135.85%(Figure39B).Histologicalanalysisofwoundsonday14
149
150
indicated that granulation tissue of UCMSCsgroup appeared to be thicker and larger as

compared with the untreated group. Thus, these results demonstrated that the engrafted
UCMSCsacceleratedwoundhealingprocess.
Figure 37. (A) Morphology and cross sectional image of umbilical cord. (B) UCMSCs exhibited fibroblastic morphology
with bipolar spindles shape (From Shohara et al. [14, 15]. Reprinted with permission).
Figure 38. 6 mm full-thickness excicional wound splinted with silicone plate (From Shohara et al. [14, 15]. Reprinted
with permission).
Figure 39. (A) Representative photograph of the wound at day 0, 7, 14 after UCMSCs transplantation. (B) Measure
ment of wound closure at different time points (From Shohara et al. [14, 15]. Reprinted with permission).
Insummary,thepresentstudydescribestheisolationandprimarycharacterizationofstem
cellsfrommedicalwastes,suchasumbilicalcord.Furthermore,localinjectionofUCMSCs
acceleratedcutenaouswoundhealingprocess.Thisstemcellscanbeisolatedwithoutinvasive
surgicalprocedures,providinguniquecellresourcesforregenerativemedicine.Togetherwith
the distinct advantages of UCMSCs, such as accessibility, painless procedures to donors,
possiblesourceforautologouscelltherapyandlowerriskofviralcontamination,wesuggest
thatUCMSCsshouldbeconsideredapromisingcellresourceforcelltherapy.
2.13.Accelerationofwoundhealingwithstemcellderivedgrowthfactors
Recently,ithasbeenrevealedthatbonemarrowderivedMSCsaccelerateskinwoundhealing,
anditisattractingattentionasanewcelltherapy.However,MSCsareisolatedfrombone
marrowfrompatient,andcollectionofthebonemarrowisconsiderablyinvasive.Inaddition,
thereisaproblemthatproliferativecapacityofMSCsdecreaseswithaging.Thus,application
ofMSCsforelderlypatientsandfreshcaseswasdifficultbecauseittakesalongtimetocultivate
thecells.Ontheotherhand,ithasbeenknownthatMSCssecretemanygrowthfactors.We
presumedthatgrowthfactorssecretedbyMSCsplayamainroleinwoundhealingeffect,and
examinedeffectofMSCCMonwoundhealing.
Figure 40. Scheme of the experiment(From Tamari et al. [16-18]. Reprinted with permission from Quintessence Pub
lishing Co, Inc, Chicago).
Weobservedwoundhealingprocessmacroscopicallyandhistologicallyusinganexcisional
woundsplintingmousemodel,andexaminedexpressionlevelofhyaluronicacidrelatedto
thewoundhealingprocesstoevaluatewoundhealingeffectofMSCs,MSCCM,andcontrol
(PBS)(Figure39).TheMSCsandMSCCMgroupsacceleratedwoundhealingascompared
withthecontrolgroup.TheareaofwoundintheMSCsandMSCCMgroupsat5daysand
laterindicatedstatisticallysignificantdifferenceascomparedwiththecontrolgroup.(Figure
151
152
40,41)At7daysafteradministrationofMSCsorMSCCM,epithelializationwasaccelerated,
thick connective tissue was formed in the skin defective area, and the area of wound was
reducedintheMSCsandMSCCMgroupsascomparedwiththecontrolgroup.Hyaluronic
acidwasexpressedinthemarginalpartofthewoundsignificantly.At14daysafteroperation,
infiltrationofinflammatorycellswasdecreasedascomparedwith7days,andthewoundwas
closedintheMSCsandMSCCMgroups,whiledefectivepartofepitheliumwasobservedin
thecontrolgroup.Expressionofhyaluronicacidwasdecreasedfrom7days.At7and14days
afteroperation,expressionlevelsofhyaluronicacidwere1512.084.10ng/mgand683.756.4
ng/mgincontrolgroup,2450.3225.7ng/mgand1690.2170.2ng/mginMSCsgroup,and
2360.6230.0ng/mgand1570.1142.5ng/mginMSCCMgroup.At7and14days,theMSCs
andMSCCMgroupsexpressedsignificantlyhighlevelofhyaluronicacidascomparedwith
thecontrolgroup(P<0.05).Theexpressionlevelofhyaluronicacidwaslowerat14daysthan
thatat7daysinallthreegroups.
Figure 41. Macroscopic findings at 7 and 14 days after injection. Left panels (A, C, and E) at 7 days after administra
tion; Right panels (B, D, and F) at 14 days after administration. (A and B) control group; (C and D) MSCs group; (E and
F) MSC-CM group. Bar = 3 mm (From Tamari et al. [16-18]. Reprinted with permission from Quintessence Publishing
Co, Inc, Chicago).
These experimental results indicated that both MSCs and MSCCM groups have wound
healingaccelerationeffectascomparedwiththecontrolgroup.Thewoundhealingaccelera
tioneffectsoftheMSCCMgroupandtheMSCsgroupwereequivalent.Accordingly,itis
suggested that the MSCCM contains growth factor derived from stem cells and is able to
providewoundhealingaccelerationeffectequivalenttostemcelltransplantation,andmay
becomenewtherapeuticmethodforwoundhealinginthefuture.
Figure 42. Time-dependent changes in the area of wound area determined with image analysis software. area of actual
wound/area of original wound 100. Test for significant difference (ANOVA). MSC-CM group vs. control group, *P < 0.05.
**P < 0.01 (From Tamari et al. [16-18]. Reprinted with permission from Quintessence Publishing Co, Inc, Chicago).
2.14.Humandentalpulpderivedstemcellspromotelocomotorrecoveryaftercomplete
transectionoftheratspinalcordbymultipleneuroregenerativemechanisms
Spinalcordinjuryresultedinseverefunctionaldisability.Inworldwide,2.5millionpeople
livewithspinalcordinjuryMorethan200,000newinjuriesreportedeachyear.Currentlyno
curativetherapyisavailable.Recentstudyhasdemonstratedthattransplantationofstemcells
ininjuredspinalcordwouldsupportfunctionalrecovery.Thecellularresourcesofstemcell
areHumanembryonicstemcell,Inducedpluripotentstemcells,Humanembryonicneural
stemcell,andadultmesenchymalstemcellsfrombonemarrow.Thesestemcellshavebeen
showntobevaluablecellularresourcesfortreatmentofanimalmodelofSCI.However,touse
153
154
themforpatient,therearesignificantethnical,safetyandinvasiveproblems.Thus,anideal
stemcellresourceforSCItreatmentisstillanelusivesubject.Herewehaveexaminedaneuro
regenerativeactivityoftoothderivedstemcellsinaratmodelofSCI.Meritsofteethderived
stem cells for SCI treatment are 1, They can be easily isolated from medical wast such as
wisdomteethandmilkteeth.2,AStheyareoriginatedfromNeuralcrest,theycanbeavaluable
cellular resources for treatment of neurodegenerative disease. 3, as they are autologous
cellularresources,mustbesafeforcelltherapy.
Figure 43. Merits of teeth derived stem cells for SCI treatment.
FlowcytometryanalysisshowedthatSHEDsandDPSCsexpressedasetofmesechymalstem
cellmarker,CD90,CD73,CD105andCD90,butnotendothelial/hematopoieticmarker,CD54,
CD34, CD45, CD11b/c or HLADR, as describe previously. Majority of these stem cells
uniquely coexpressed several neural linage markers, including Nestin and Doublecortin
(neural stem cell marker), GFAP (astrocyte marker), CNPase (immature oligodendrocyte
maeker)andA2B5(oligodendrocyteprecursormarker)altogether.Nextwehaveexamined
expressionofneurotropicfactorsbyrealtimePCR.Comparingwithskinderivedfibroblasts
SHEDs expressed NT3 (Neurotrophin3) and BDNF (Brain Derived Neurotrophic Factor),
morethan3and5times,respectively.TheseresultsdemonstratethatSHEDandDPSC,stem
cells derived dental pulps, are valuable cellular resources for neuroregeneration therapy.
SHEDs or DPSCs were transplanted into the transected spinal cord as described in the
MaterialsandMethods.At8weekspostengraftment,BBBscoringsuggestedarecoveryof
hindlimblocomotorfunctionbothinSHEDsorDPSCstransplantedrats(n=11forSHEDs
andn=10forDPSCs)incomparisonwithvehiclecontrols(n=10)analysisrevealedahigher
frequencyoflocomotionrecovery(BBBscore>7forSHEDsand>6forDPSCs)inSHEDsor
DPSCs vs. vehicle controls. Thus, both SHED and DPSC promote functional recovery of
completely transected spinal cord. As axonal myelination promotes functional recovery of
injuredspinalcord,westudiedwhethertransplantedSHEDspreservedmyelinatedareain
injured spinal cord. In the PBStreated control group, the transverse area of the lesion site
exhibited no or little staining of Fluoromyelin. In SHEDstransplanted, however, myelin
positiveareacovered15%and80%ofentirespinalcord,atepicenterand5mmcaudallesion
site,respectively,showingthattransplantedSHEDsplaysignificantroleinpreservationof
axonalmyelinationininjuredspinalcord.
Our study revealed that engrafted SHEDs exhibited three major therapeutic benefits for
recoveryafterSCI,includinginhibitionoftheSCIinducedapoptosisofneurons,astrocytes,
andoligodendrocytes,whichpromotedthepreservationofneuralfibersandmyelinsheaths,
regenerationofthetransectedaxonthroughthedirectinhibitionofmultipleAGIsignals,such
aschondroitinsulfateproteoglycansandMAG,byparacrinemechanisms,andreplacement
of lost or damaged oligodendrocytes after SCI through specific differentiation into mature
oligodendrocytesundertheextremeconditionsofSCI.Toourknowledge,theneuroregen
erativeactivitiesandareuniquetotoothderivedstemcells,andarenotexhibitedbyanyother
previouslydescribedstemcells.Thus,ourdatademonstratethattoothderivedstemcellsmay
providesignificanttherapeuticbenefitsfortreatingtheacutephaseofSCIthroughbothcell
autonomous and paracrine/trophic regenerative activities. We demonstrated multifaceted
neuroregenerative activities of toothderived stem cells that fulfill many requirements for
functionalrecoveryafterSCI.Inadditiontotheirremarkableneuroregenerativeactivities,we
didnotobservethemalignanttransformationofengraftedSHEDs8weeksaftertheirimplan
tation (data not shown). Furthermore, SHEDs and DPSCs can be obtained from exfoliated
deciduousandimpactedadultwisdomteethwithoutadversehealtheffects.Thus,thereare
fewethicalconcernsregardingtheirclinicaluse.Weproposethattoothderivedstemcellsmay
beanexcellentandpracticalcellularresourceforthetreatmentofSCI.
2.15.Growthfactorsderivedfromdentalpulpstemcells:Anewpotentialclinicalbenefits
forCNSregenerationtherapy
ThereisnoeffectivetreatmentforSCI.IntheacutephaseofSCI,damagedneuronsaregoing
todiewithinadayaftertheinjurybyinflammatoryreaction.Untilaweaklater,multipleaxon
growth inhibitors (AGIs) are produced by the astroglial scar and degenerated myelin sur
rounding the injured CNS (Figure 44). We have been reported the clinical benefits of the
engraftedhumanDPSCs(hDPSCs)andstemcellsfromhumanSHEDsinthetreatmentofacute
phaseratSCImodel.Recently,growthfactorssecretedfromstemcellswereimportanttostem
155
156
celltherapy.SowefocusedontheparacrineeffectofgrowthfactorsderivedfromSHEDs.
Here,weshowthatlocaladministrationofaserumfreeconditionedmediaofSHEDs(SHED
CM)intotheratSCIresultedinremarkablerecoveryofhindlimblocomotorfunctions.We
foundthatthiseffectisassociatedwithseveralparacrineeffectofSHEDCM.
Figure 44. Acute phase of SCI.
We examined the effects of the treatment with SHEDCM, MSCCM or FibroCM on the
functionalrecoveryaftercontusionSCI.CMswerecontinuouslydeliveredintrathecallyby
infusion pump into the SCI epicenter. The level of recovery of hindlimb locomotion was
evaluatedusingtheBasso,Beattie,Bresnahanlocomotorratingscale(BBBscale).Afterthe
recovery period, the rats that had received SHEDCM were able to support their weight
throughtheplantarsurfaceofthepawandstepwithforehindlimbcoordinatedmannerIn
contrast,althoughtheMSCCMorFibroCMtreatedratsmove3jointsofhindlimb,theywere
notabletowalkwithweightsupport.TheseresultsdemonstratethatSHEDCM,butnotMSC
CMandFibroCM,providessignificanttherapeuticbenefitforthetreatmentoftheacutephase
ofSCI.IntheacutephaseofSCI,muchinflammatorycytokinesarereleased.Theseinflam
matorycytokinesactivatemicrogliaandleadstoexpansionofinflammatoryreaction.Wegot
ratspinalcordmRNAafterSCI,andanalyzedexpressionofinflammatoryandantiinflam
matorycytokines.WefoundthatinflammatorycytokinesexpressiondownregulatedMSC
CM,FibroCM,andSHEDCMgroupsratherthancontrolgroupsatthreetimepoints.Onthe
otherhand,importantly,antiinflammatorycytokinesexpressionupregulatedonlySHEDCM
groups. We got mouse primary microglia mRNA after LPS stimulation and analyzed the
expression.Similarlyinvivoresults,inflammatorycytokinesexpressiondownregulatedMSC
CM, FibroCM, and SHEDCM culture groups rather than control groups. But,uniquely,
glutamate and NO level in microglia culture medium reduced only SHEDCM groups.
Damagedneuronsandglialcellsaregoingtodiewithinadayafterinjury,andexpandcell
deathinoneweek.Thesecelldeathleadstodegenerationofmyelinandchronicatrophyof
spinalcord.SowetriedTUNELstaining.Comparedtothecontrolgroups,SHEDCMgroups
weredownregulatedTUNELpositivecellssignificantlyat24hourand1weekafterSCI.And
westainedmyelinbyfluoromyelinateightweeksafterSCI.Atcontrolgroups,spinalcord
wasgoingtoatrophyfromepicenter,butontheotherhandSHEDCMgroupsweremaintained
myelinandkeptformofspinalcordfromepicenter.UntilaweekafterSCI,multipleAGIsare
produced by astroglial scar and degenerated myelin surrounding injury site. These AGIs
acceleratetheneuronalapoptosisandinhibitaxonalregrowth.Wegotprimarycerebellum
granuleneurons(CGNs)andtriedneuriteoutgrowthassayonAGIscoatingdish.Andwe
foundthatCGNsonlySHEDsandDPSCCMculturegroupsinhibitedAGIsactivityandstrong
promotedtheirneurite.SowestainedbyNFMand5HTeightweeksafterSCI.Andwefound
thatneuronalaxonextendedbeyondtheinjuryepicentertocaudal.Ourstudiesdemonstrate
potentialclinicalbenefitsofSHEDCMforthetreatmentoftheacutephaseofSCI,providing
anovel,safeandeffectiveneuroregenerativetherapythatprotectspatientsCNSfromthe
traumaticandischemicCNSinjury.
Figure 45. Multifaceted treatment strategy.
157
158
2.16.SHEDCMrnhancesrecoveryoffocalcerebralischemiainrats
Regenerative therapy using stem cells is a promising approach for treatment of stroke.
Recently,wereportedthatDPSCsamelioratedischemictissueinjuryinratbrainandacceler
ated functional recovery after middle cerebral artery occlusion (MCAO). In this study, we
investigated the effects of SHEDCM after pMCAO (permanent middle cerebral artery
occlusion).AdultmaleSpragueDawleyratsweresubjectedtopMCAO.SHEDCMwasthen
administered intranasally and motor function and infarct volume evaluated. SHEDs were
culturedinDMEMserumfreemedium.ConditionedmediumofSHEDswascollectedafter
48hofcultureandcentrifugedat1500rpmfor5min.Thesupernatantwasrecentrifugedat
3000 rpm for 3 min followed by collection of the second supernatant, named SHEDCM.
SeventytwohoursafterpMCAO,theratswereanesthetizedagainwith1.5%isofluraneina
mixtureof70%N2Oand30%O2.Atotalof100LofSHEDCMwasadministeredtoratvia
theolfactorypathwayusingaHamiltonmicrosyringe(Figure46).TheSHEDCMpreparation
wasadministeredin10Latatime,withanintervalof2minbetweeneachadministration.
Intranasaladministrationwasperformedeverydayfromdays3to15.
Figure 46. Intranasal administration.
The rats were blindly examined on days 1, 3, 6, 9, 12, and 15 using a standardized motor
disabilityscalewithslightmodifications.Thecryosectionsobtainedfromsamplesonday16
werestainedwithhematoxylinandeosin.ImageJwasusedtodetermineeachinfarctareain
12coronalsectionsat1.00mmintervals.Theentireinfarctionareawascoveredbythese12
coronalsections.Regionalinfarctvolumeswerecalculatedbysummingtheinfarctareasand
multiplyingtheseareasbythedistancebetweensections(1.00mm).Thetwogroupsdisplayed
almostthesamehighscoreformotorfunctionintheearlystagesasshowninFigure47right.
Differencesinthescoreappearedgraduallybetweenthetwogroupsduringthemiddlestage.
On day 15, progressive improvement in motor disability in the SHEDCM group became
significantcomparedtothePBSgroup.AsshowninFigure47left,therewasasignificant
decreaseininfarctvolumeonday16intheSHEDCMgroupcomparedtothePBSgroup.
These results suggest that SHEDCM promoted regeneration. Recently, we reported the
characteristicsofSHEDscomparedwithDPSCsandMSCs.TheresultsindicatedthatSHEDs
possessedhighproliferationabilityandwereenrichedwithextracellularmatrix,suggesting
itmaybeausefulsourceforstemcellbasedtherapy.Inaddition,usingmicroarrayanalysis,
we showed that SHEDs had higher expression levels of several growth factors, such as
fibroblastgrowthfactor,transforminggrowthfactor,connectivetissuegrowthfactor,nerve
growth factor, and bone morphogenetic protein. Taken together, these findings indicate
SHEDsareamorepotentiallyusefulsourceofstemcellsforcelltherapythanDPSCsandMSCs.
Figure 47. Evaluation of motor function and Reduction in infarct volume.
However, cell therapy is always associated with problems, such as canceration, immune
rejection, and ethical issues. Therefore, it is necessary to find alternative treatments to cell
therapy.Studiesinrecentyearshaveresultedintherecognitionofaparacrinefunctionin
factors,andhavesuggestedthatstemcelltransplantationmayalsoberegardedascellbased
cytokinetherapy.Accordingly,weinvestigatedtwostepsfordevelopinganewtreatmentfor
cerebralischemia.Inthefirststep,weusedSHEDCMasanewsourcefortreatmentofcerebral
ischemia.Wehavereportedpreviouslyinaratmodelofstrokethatneedleadministrationof
DPSCsinducedrecoveryofmotordisabilityandreductionininfarctvolume,proliferationof
their presumptive progeny in SVZ, migration to the infarct, and differentiation into the
appropriateneurons.Asseveralgrowthfactorsinvolvedinneuralregenerationaresecreted
fromDPSCs,wehypothesizedinthecurrentstudythatSHEDCMmayimproverecoveryof
motor disability and reduce infarct volume. In the second step, we investigated intranasal
administrationofSHEDCM.Usingthisadministration,therapeuticmoleculestraversethe
BBB through the olfactory pathway and the lessstudied trigeminal neural pathway. An
important advantage of intranasal administration is that it is less invasive with the factors
beingdelivereddirectlytothebrain.
159
160
OurresultssuggestedthatSHEDCMincludingsomegrowthfactorsmayproduceeffectsin
strokemodel.Thesuccessofthetwostepsinvestigatedsuggestitmaybepossibletousea
shortcut for clinical application. Administration of SHEDCM resolves the ethical issues
involvedwithcelltherapies,asSHEDCMisnotacellbutratheraconjugateofmanygrowth
factors.AsSHEDCMcanbestocked,itispossibletouseitforacutestagesofstroke,either
alone or with readymade treatment, such as recombinant tissue plasminogen activator,
anticoagulation,andantiplatelettherapy.Thisstudysuggestedthatintranasaladministration
ofSHEDCMmayhelprecoveryinacutestrokepatientsinthefuture.Inconclusion,regen
erationtherapyusingSHEDCMisaverysafemethodwithnoassociatedproblemsandis
thereforeapotentialcandidateforinnovativetreatmentofcerebralischemia.
3.Conclusion
Theendpointofallourstudiesistoestablishsuccessfulnoveltherapiesforpatientssuffering
from diseases or disorders. These studies suggest that engineered tissues may have an
expandedclinicalapplicabilityinthefutureandmayrepresentaviabletherapeuticoptionfor
thosewhorequiretissuereplacementorrepair.Thefutureoutlookispositivewiththerecent
discoveryofiPSCsandstemcellsfromhumanSHEDs.Weespeciallyfocusonthepotentialof
SHEDs,whichwereidentifiedasapopulationofhighlyproliferative,clonogeniccellscapable
ofdifferentiatingintoavarietyofcelltypesincludingneuralcells,adipocytes,andodonto
blasts.Thestemcellfieldisalsoadvancingrapidly,providingnewtherapeuticoptions.Inthe
future,bankingthesestemcellsmayprovideaconvenientsourceforautologoustherapyand
formatchingrecipientswithhistocompatibledonors.Tissueengineeringholdsgreatpromise
forthefutureofmedicine,asexperimentaleffortsarecurrentlyunderwayforvirtuallyevery
typeoftissueandorganwithinthehumanbody.
Authordetails
MinoruUeda
DepartmentofOralandMaxillofacialSurgery,GraduateSchoolofMedicine,NagoyaUni
versity,Nagoya,Japan
References
[1] Kohgo T, Yamada Y, Ito K, Okabe K, Yoshimi R, Yajima A, Baba S, Ueda M. Bone
regenerationfordentalimplantsusingtissueengineeredbonewithanovelscaffold.
In: 25th Annual Meeting of Academy of Osseointegration. 46 March 2010, Orland,
FL,USA;2010.
[2] Yoshimi R, Katagiri W, Osugi M, Inukai T, Yamamoto A, Hibi H, Ueda M. Novel

xenografttechniqueusingthestemcellsculturedconditionedmediaforboneregen
eration.In:20thAnnualScientificMeetingoftheEuropeanAssociationofOsseointe
gration,1315October2011,Athens,Greece;2011.
[3] YoshimiR,KatagiriW,OsugiM,InukaiT,HibiH,UedaM.Stemcellculturedcondi
tioned media of stem cells from exfoliated deciduous teeth enhanced cell prolifera
tion: association with Wnt/betacatenin pathway. In: The 56th Congress of the
JapaneseSocietyofOralandMaxillofacialSurgeons.2123October2011,Osaka,Ja
pan;2011.
[4] Ikeno M, Hibi H, Kinoshita K, Hattori H, Ueda M. Effects of the permeability of
shieldswithautologousbonegraftsonboneaugmentation.Oral&CraniofacialTis
sueEngineering2011;1(3)198204.
[5] TateishiH,OkamotoY,KinoshitaK,HibiH,UedaM.EffectsofImplantSurfaceon
Bone Healing AroundTitanium Screw Implants in Ovariectomized Rats. Oral and
CraniofacialTissueEngineering2011;1(1)4249.
[6] Tateishi H, Okamoto Y, Kinoshita K, Hibi H, Ueda M. A Study of Bone Healing
Around the Titanium Screw Implants in Ovariectomized Rats.: Can Rough Surface
Contribute to The Implant Stability? In: 26th Annual Meeting of Academy of Os
seointegration,35March2011,WashingtonD.C.,USA;2011.
[7] WadagakiR,MizunoD,YamawakiOgataA,SatakeM,KanekoH,HagiwaraS,Ya
mamoto N, Narita Y, Hibi H, Ueda M. Osteogenic Induction of Bone MarrowDe
rived Stromal Cells on SimvastatinReleasing, Biodegradable, Nano to Microscale
FiberScaffolds.AnnalsofBiomedicalEngineering2011;39(7)18721881.
[8] Wadagaki R, Mizuno D, YamawakiOgata A, Satake M, Kaneko H, Yamamoto N,
NaritaY,HibiH,UedaM.Osteoinductionofbonemarrowderivedstromalcellson
simvastatinreleasing fiber scaffolds. In: The 10th Congress of Japanese Society for
RegenerativeMedicine.12Mar2011,Tokyo,Japan;2011.
[9] FujioM,YamamotoA,AndoY,ShoharaR,KinoshitaK,KanekoK,HibiH,UedaM.
Stromalcellderivedfactor1enhancesdistractionosteogenesismediatedskeletaltis
sueregenerationthroughtherecruitmentofendothelialprecursors.Bone2011;49(4)
693700.
[10] InoueM,EbisawaK,ItayaT,SugitoT,YamawakiOgataA,SumitaY,WadagakiR,
NaritaY,AgataH,KagamiH,UedaM.EffectofGDF5andBMP2ontheexpression
oftendo/ligamentogenesisrelatedmarkersinhumanPDLderivedcells.OralDiseas
es2011;
[11] SugitoT,MineshibaF,ZhengC,CotrimA.P,GoldsmithC.M,BaumB.J.Transient
TWEAKoverexpressionleadstoageneralsalivaryepithelialcellproliferation.Oral
Diseases2008;15(1)7681.
[12] OkabeK,YamadaY,ItoK,KohgoT,YoshimiR,UedaM.Injectablesofttissueaug
mentationbytissueengineeringandrebeneratibemedicinewithhumanmesenchy
161
162
mal stromal cells, plateletrich plasma and hyaluronic acid scaffolds. Cytotherapy
2009;11(3)307316.
[13] Nishino Y, Ebisawa K, Yamada Y, Okabe K, Kamei Y, Ueda M. Human Deciduous
TeethDentalPulpCells(hDPC)withBasicFibroblastGrowthFactor(bFGF)Enhance
WoundHealingofSkinDefect.JournalofCraniofacialSurgery2011;22(2)438442.
[14] ShoharaR,YamamotoA,TakikawaS,IwaseA,HibiH,KikkawaF,UedaM.Umbili
calcordWhartonsjelly:Anewpotentialcellsourceofmesenchymalstemcellsfor
wound healing. In: Proceedings of 9th International Society Stem Cell Research.
1518June2011,Toronto,Canada;2011.
[15] ShoharaR,YamamotoA,TakikawaS,IwaseA,HibiH,KikkawaF,UedaM.Umbili
calcordWhartonsjelly:Anewpotentialcellsourceofmesenchymalstemcellsfor
woundhealing.In:ProceedingsofInternationalSymposiumonMicroNanoMecha
tronicsandHumanScience.7November2011,Nagoya,Japan;2011.
[16] TamariM,NishinoY,YamamotoN,UedaM.AccelerationofWoundHealingwith
StemCellDerivedGrowthFactors.Oral&CraniofacialTissueEngineering2011;1(3)
181187.
[17] TamariM,NishinoY,YamamotoN,FurueH,ShigetomiT,UedaM.Effectofmesen
chymal stem cell derived paracrine factor on wound healing in mice skin. In: The
56thcongressoftheJapaneseSocietyofOralandMaxillofacialSurgeons.2123Octo
ber2011,Osaka,Japan;2011.
[18] TamariM,NishinoY,YamamotoN,UedaM.WoundHealingAccelerationbyStem
CellDerivedGrowthFactors.In:ProceedingsofInternationalSymposiumonMicro
NanoMechatronicsandHumanScience,7November2011,Nagoya,Japan;2011.
[19] IkenoM,HibiH,KinoshitaK,HattoriH,UedaM.Effectsofselfassemblingpeptide
hydrogel scaffold on bone regeneration with recombinant human bone morphoge
neticprotein2.Oral&CraniofacialTissueEngineering2011;1(2)9197.
[20] IkenoM,HibiH,KinoshitaK,YajimaA,HattoriH,UedaM.Theeffectofshieldson
the bone augmentation model in rabbit calvariae. In: The 63th Annual Meeting of
TheJapaneseStomatologicalSociety.1617April2009,Hamamatsu,Japan;2009.
theboneaugmentationmodelinrabbitcalvariae.In:XIIIBiennialInternationalCon
gressoftheInternationalSocietyofCraniofacialSurgery.2630Sepember2009,Ox
ford,UK;2009.
the bone augmentation model in rabbit calvariae. In: The 9th Congress of the The
Japanese Society of Regenerative Medicine. 1819 March 2010, Hiroshima, Japan;
2010.
[23] Ikeno M, Hibi H, Kinoshita K, Hattori H, Shibuya H, Yajima A, Ueda M. Effects of
selfassemblingpeptidehydrogelscaffoldontheboneaugmentationinrabbitcalvar
iae. In: The 55th Annual Meeting of The Japanese Society of Oral and Maxillofacial
Surgeons,1618October2010,Chiba,Japan;2010.
[24] Ikeno M. The effect of selfassembling peptide hydrogel scaffold on the bone aug
mentationmodelinrabbitcalvariae.In:ProceedingsofInternationalSymposiumon
MicroNanoMechatronicsandHumanScience,710November2010,Nagoya,Japan;
2010.
[25] InoueT,SugiyamaM,HattoriH,YamamotoA,HibiH,UedaM.SHCMEnhances
Recovery of Focal Cerebral Ischemia in Rats. In: 32nd Annual Meeting of the Japa
neseSocietyofInflammationandRegeneration,23June2011,Kyoto,Japan;2011.
[26] InoueT,SugiyamaM,HattoriH,HibiH,UedaM.SHCMEnhancesRecoveryofFo
calCerebralIschemiainRats.In:The56thCongressoftheJapaneseSocietyofOral
andMaxillofacialSurgeons,2123October2011,Osaka,Japan;2011.
[27] InoueT,SugiyamaM,HattoriH,HibiH,UedaM.SHCMEnhancesRecoveryofFo
cal Cerebral Ischemia in Rats. In: 22th IEEE International Symposium on Micro
NanoMehatronicsandHumanScience,7November2011,Nagoya,Japan;2011.
[28] Kohgo T, Yamada Y, Ito K, Yajima A, Yoshimi R, Okabe K, Baba S, Ueda M. Bone
Regeneration with Selfassembling Peptide Nanofiber Scaffolds in Tissue Engineer
ingforOsseointegrationofDentalImplants.InternationalJournalofPeriodontics&
RestorativeDentistry2011;31(4)e9e16.
[29] Matsubara K, Yamamoto A, Sakai K, Ueda M. The potential clinical benefits of the
serumfreeconditionedmediaderivedfromdentalpulpstemcellsinCNSregenera
tiontherapy.In:3thNagoyaglobalretreatNagoyaUniversityGlobalCOEProgram
IntegratedFunctionalMolecularMedicineforNeuronalandNeoplasticDisorders,
2526February2011,Nagoya,Japan;2011.
[30] Matsubara K, Yamamoto A, Sakai K, Ueda M. The potential clinical benefits of the
serumfreeconditionedmediaderivedfromdentalpulpstemcellsinCNSregenera
tion therapy. In: The 10th Congress of the Japanese Society for Regenerative Medi
cine,12March2011,Tokyo,Japan;2011.
[31] Matsubara K, Yamamoto A, Sakai K, Matsushita Y, Ueda M. The potential clinical
benefitsoftheserumfreeconditionedmediaderivedfromdentalpulpstemcellsin
CNS regeneration therapy. In: 2th Symposium of Nagoya University and National
InstituteforPhysiologicalSciences,20August2011,Nagoya,Japan;2011.
[32] Matsubara K, Yamamoto A, Sakai K, Matsushita Y, Ueda M. The potential clinical
benefitsoftheserumfreeconditionedmediaderivedfromdentalpulpstemcellsin
CNSregenerationtherapy,In:GlobalCOEprogramThe3rdInternationalSymposi
um,89December2011,Nagoya,Japan;2011.
[33] NishinoY,YamadaY,EbisawaK,NakamuraS,OkabeK,UmemuraE,HaraK,Ueda
M.Stemcellsfromhumanexfoliateddeciduousteeth(SHED)enhancewoundHeal
ingandpossibilityofnovelcelltherapy.Cytotherapy2011;13(5)598605.
163
164
[34] Nishino Y, Ebisawa K, Yamada Y, Okabe K, Kamei Y, Ueda M. Human Deciduous

Teeth Dental Pulp Cells (hDPC) with Basic Fibroblast Growth Factor (bFGF) En
hanceWoundHealingofSkinDefect.In:8thAnnualMeetingoftheInternationalSo
cietyforStemCellResearch,1619June2010,SanFrancisco,CA,USA;2010.
[35] NishinoY,YamadaY,EbisawaK,NakamuraS,OkabeK,UmemuraE,HaraK,Ueda
M.Stemcellsfromhumanexfoliateddeciduousteeth(SHED)enhancewoundheal
ingandpossibilityofnovelcelltherapy.In:54thcongressoftheJapaneseSocietyof
OralandMaxillofacialSurgeons,911October2010,Sapporo,Japan;2010.
[36] OkabeK,YamadaY,ItoK,YoshimiR,KohgoT,YajimaA,HibiH,UedaM.Experi
mental study for softtissue regeneration with tissue engineering and regenerative
medicine. In: The 52nd Congress of the Japanese Society of Oral and Maxillofacial
Surgeons,2930September2007,Nagoya,Japan;2007.
[37] OsugiM,KatagiriW,YoshimiR,InoueM,HaraK,InukaiT,HibiH,UedaM.The
stemcellsculturedconditionedmediaenhancedboneregeneration.In:20thAnnual
Scientific Meeting of the European Association of Osseointegration, 1315 October
2011,Athens,Greece;2011.
stem cells cultured conditioned media enhanced bone regeneration. In: The 56th
CongressoftheJapaneseSocietyofOralandMaxillofacialSurgeons,2123October
2011,Osaka,Japan;2011.
stem cells cultured conditioned media enhanced bone regeneration. In: The 32nd
Congress of the Japanese Society of Inflammation and Regeneration, 23 June 2011,
Kyoto,Japan;2011.
[40] SakaiK,YamamotoA,MatusbaraK,NakamuraS,TauchiR,WakaoN,ImagamaS,
Hibi H, Ueda M. Multifaceted neuroregenerative activities of toothderived stem
cellspromotelocomotorrecoveryaftercompletetransectionofthespinalcord.The
JournalofClinicalInvestigation2012;122(1)8090.
[41] SakaiK,YamamotoA,HibiH,YamadaY,FujioF,YamagataM,UedaM.Transplan
tationofdentalpulpstemcellsinspinalcordinjury.In:88thInternationalAssocia
tionforDentalResearch,417July2010,Barcelona,Spain;2010.
[42] SakaiK,YamamotoA,HibiH,YamadaY,FujioF,UedaM.Transplantationofhu
mandentalpulpstemcellsincompletetransactionoftheratspinalcord.In:2ndNa
goyaUniversityGlobalCOEProgram,IntegratedFunctionalMolecularMedicinefor
NeuronalandNeoplasticDisorders,2627February2010,Nagoya,Japan;2010.
[43] Sakai K, Matsubara K, Yamamoto A, Ueda M. Transplantation of dental pulp stem
cellsinspinalcordinjury.In:The3rdSymposiumofYoungResearchersInnovative
MEMSdesignandthebiomedicalapplicationMicro/NanoGlobalCOE,6December
2010,Nagoya,Japan;2010.
[44] Sakai K, Matsubara K, Yamamoto A, Hibi H, Ueda M. Engrafted dental pulp stem
cells promoted functional recovery of completely transected rat spinal cord. In: 9th
InternationalSocietyStemCellResearch,1518June2011,Toronto,Canada;2011.
[45] Shohara R. Is it possible? Body repairing with stem cells from medical wastes. In:
ProceedingsofInternationalcrossdisciplinarysymposiumonMicroNanoSystems,
12November2009,Nagoya,Japan;2009.
[46] ShoharaR,TakikawaS,YamamotoA,HibiH,FujioM,SakaiK,YamagataM,Goto
M,IwaseA,KikkawaF,UedaM.Developmentofbonetissueengineeringusinghu
manumbilicalcordderivedtissueregenerativecells.In:ProceedingsofInternational
Symposium on MicroNanoMechatronics and Human Science, 9 November 2010,
Nagoya,Japan;2010.
[47] TateishiH.AnExperimentalStudyofBoneHealingAroundtheTitaniumScrewIm
plants in Ovariectomized Rats: Enhancrement of Bone Healing by Bone Marrow
Stromal Cells. In: Proceedings of International Symposium on MicroNanoMecha
tronicsandHumanScience,9November2010,Nagoya,Japan;2010.
[48] Tateishi H, Okamoto Y, Kinoshita K, Hibi H, Ueda M. A study of bone healing
around the titanium screw implants in ovariectomized rats. Effects of bone healing
byvarianceofimplantsurfaceproperty.In:30thAnnualMeetingofJapaneseSociety
of Oral Implantology Kinki Hokuriku Section, 1920 November 2010, Fukui, Japan;
2010.
[49] Ueda M, Nishino Y. Cell based cytokine therapy for skin rejuvenation. Journal of
CraniofacialSurgery2010;21(6)18611866.
[50] Wadagaki R. Osteogenic Induction of Bone MarrowDerived Stromal Cells on Sim
vastatinReleasing,Biodegradable,NanoMicroFiberScaffolds.In:The3rdSymposi
umofYoungResearchersInnovativeMEMSdesignandthebiomedicalapplication,
6December2010,Nagoya,Japan;2010.
[51] YoshimiR,YamadaY,ItoK,NakamuraS,AbeA,NagasakaT,OkabeK,KohgoT,
Baba S, Ueda M. Selfassembling peptide nanofiber scaffolds, plateletrich plasma,
andmesenchymalstemcellsforinjectableboneregenerationwithtissueengineering.
JournalofCraniofacialSurgery2009;20(5)15231530.
[52] YoshimiR,YamadaY,ItoK,KohgoT,OkabeK,YajimaA,HibiH,UedaM.Applica
tionforboneregenerationbyregenerativemedicinewithnewthreedimensionma
trixusingnanotechnology.In:52ndAnnualMeetingofJapaneseSocietyofOraland
MaxillofacialSurgeons,29September2007,Nagoya,Japan;2007.
165
Chapter 8
Electronic Structure Calculations for Nano Materials
Nobutada Ohno, Dai Okumura and

Yusuke Kinoshita
1.Introduction
Thisisachapteronelectronicstructurecalculationsfornanomaterialsbasedonfirstprinciples
densityfunctionaltheory(DFT)[1,2].TheDFThasbecometheprimarytoolforelectronic
structurecalculationsforsolidsandhasalsobecomepopularforatomsandmolecules.There
aremanyreviewsandbooksontheDFT[35].Inthischapter,theworksofOhnoandcoworkers
onelectronicstructurecalculationsfordeformedboronnitridenanotubes(BNNTs)usingthe
DFTaredescribed[6,7].
TheexistenceofBNNTswastheoreticallypredictedbyRubioetal.[8,9]andthenmultiwalled
(MW)BNNTswerefirstsynthesizedbyChopraetal.[10].Sincethen,BNNTshaveattracted
theattentionofmanyresearchersowingtotheirimportantproperties[11].Themechanical
strength[12,13]andthermochemicalstability[14]ofBNNTsarecomparabletothoseofcarbon
nanotubes (CNTs) [15]. For instance, experiments (using a thermal vibrational amplitude
technique[12]andanelectricfieldinducedresonancemethod[16])andatomisticsimulations
(firstprinciples [1720], tightbinding [21, 22], and classical molecular mechanics [23, 27]
calculations)measuredtheYoungsmodulusofBNNTstobeintherange0.71.2TPa,which
isclosetothatofCNTs(e.g.,theaveragevalueis1.8TPa[28]and1.25TPa[29]).Incontrast,
theelectricalconductivityofBNNTsiscompletelydissimilartothatofCNTs.WhileCNTs
becomeeithermetallicorsemiconductivedependingonthechirality,BNNTsareelectrically
insulating regardless of the diameter and chirality [11]. This is a notable characteristic of
BNNTsthatisdifferentfromCNTs.Therefore,BNNTsareexpectedtobeusedaselectrical
insulationcoatingsforconductingorsemiconductingnanochains,nanowires,andnanotubes
insevereconditionssuchashightemperaturesandchemicallyhazardousenvironments.
However, a recent experimental study indicated that a bent MWBNNT was electrically
conductive[30],andatheoreticalstudyshowedthatflatteningdecreasedtheenergygapofa
zigzagsinglewalled(SW)BNNT[31].TheseresultsindicatethattheusefulnessofBNNTsas
nanocoatings might be lost under certain conditions (e.g., deformation caused by thermal
2013 Ohno et al.; licensee InTech. This is an open access article distributed under the terms of the Creative
168
Figure 1. Simulation model of (8,0) SWBNNT.
stressorbyasubstrateconstraint).Alternatively,BNNTscanbeusedinnanoelectronicdevices
byintroducingdeformation.Inanycase,ouraimistoelucidatetheelectronicstructuresof
deformedBNNTs.Inthefollowingsections,electronicstructurecalculationsforSWBNNTs
undertension,torsion,andflattening(Section2)andforMWBNNTsunderflattening(Section
3)willbediscussed.
2.SWBNNTsundertension,torsionandflattening
2.1.Simulationprocedure
Thissectionfocuseson(n,0)SWBNNTswithn=6,8,10,where(n,m)isthechiralindex.The(n,
m) tube has a diameter of 3a n2 + nm + n 2 / , where a is the nearest interatomic distance
between boron and nitrogen atoms. Figure 1 shows the simulation model of the (8, 0)
SWBNNT.TheBNNTislocatedatthecenteroftheunitcellsothattheaxialdirectionisparallel
to the z direction. The cell size is ( 3an / + 2LV) ( 3an / + 2LV) 3a with a=0.145 nm and
LV=0.5nm,whereLVisthelengthofvacuumregionintheunitcell.Althoughathreedimen
sionalperiodicboundaryconditionisused,thecellsizesinthexandydirectionsaresuffi
cientlylargetoavoidanyinteractionwithneighboringimagecells.Itisconfirmedthatthey
have little effect (less than 1%) on the total energy, charge distribution, and energyband
structureofadeformedBNNTwhentheyarelargerthanthetubediameterby1.0nm.
Atomicpositionsandthecellsizeinthezdirectionarefirstrelaxedusingtheconjugategradient
methoduntilatomicforcesandthestresscomponent,zz,becomelessthan0.01eV/and0.01
GPa, respectively. After obtaining the equilibrium structure, tension, torsion, or flattening
deformationisappliedwhereatomicconfigurationsarerelaxeduntiltheirforcesbecomeless
than0.01eV/.
Figure 2. Schematics of tension, torsion, and flattening of SWBNNT.
Intension(Figure2(a)),theaxialstrain,zz,isdefinedas
zz
Lz LZ 0
LZ 0
(1)
where Lz0 and Lz are the cell sizes in the z direction of unstretched and stretched BNNTs,
respectively.Inthisstudy,zzisintherange0.000.10withanincrementof0.02.
Intorsion(Figure2(b)),atomiisrotatedidegreesaboutthezaxis.iisdefinedas
i = z i
(2)
169
170
360 N
nNz Lz 0
(3)
whereisthespecificangleoftwist,ziisthezcoordinateofatomi,Nisaninteger,andNzis
thenumberofprimitiveunitcellsinthezdirection.Thetorsionanglemustbeanintegral
multiple(N)of360/ntofulfilltheperiodicboundarycondition.Inthisstudy,thevalueofN
is1andthatofNzisintherange35.
Inflattening(Figure2(c)),compressioninthexdirectionisappliedbyreducingthedistance
betweenimaginarywalls.Onceanatomcontactsawall,theatomisallowedtomoveonlyon
thewall.Theflatteningratio,,isdefinedas
=
D0 D
D0
(4)
whereD0isthetubediameteratequilibriumandDisthedistancebetweentheimaginary
walls.Inthisstudy,isintherange0.000.50withanincrementof0.05.
FirstprinciplesDFTcalculationsareconductedusingtheViennaabinitioSimulationPackage
(VASP)[32,33].Thewavefunctionsareexpandedinaplanewavebasissetwithacutoff
energyof350eV.TheultrasoftpseudopotentialproposedbyVanderbilt[34]isusedandthe
exchangecorrelation energy is evaluated by the generalized gradient approximation of
PerdewandWang[35].TheBrillouinzoneintegrationisperformedbytheMonkhorstPack
scheme [36] using a 1 1 4 kpoint mesh for atomic and electronic relaxations. After the
relaxation, the energyband structure is obtained by calculating energy eigenvalues of 30
pointsonlineintheBrillouinzone.
2.2.Resultsanddiscussion
2.2.1.Energybandstructures
ItiswellknownthattheDFTunderestimatestheenergygap.Foraquantitativediscussionof
theenergygap,amodifiedtheorysuchastheGWapproximation(GWA)isnecessary[3740].
Nonetheless,previousstudiesonbulkhexagonalBNandanisolatedBNsheetshowedthat
theshapeoftheenergybandsbytheDFTisquitesimilartothatbytheGWAexceptforthe
magnitudeoftheenergygap[37,38].Thus,theDFTcanqualitativelypredictenergyband
structuresofBNNTs.
Figure3showsthechangeintheenergybandstructuresofthe(8,0)SWBNNTundertension,
torsion,andflattening.The(6,0)and(10,0)showachangingtrendsimilartothe(8,0)band
structure.Acommonfeatureamongthethreedeformationmodesisthatboththevalence
bandmaximum(VBM)andconductionbandminimum(CBM)arelocatedatthepoint(k=0)
duringthedeformations.AnothercommonfeatureisthatthechangeintheenergyoftheVBM,
EVBM,isalmostzero.Notethatwhilethetensionandflatteningobviouslydecreasetheenergy
of the CBM, ECBM, the torsion hardly decreases ECBM. The results suggest that all the three
deformationmodesdecreaseenergygaps,Eg=ECBMEVBM,ofSWBNNTs,buttorsionhasless
ofaneffectontheenergygapthantensionandflattening.
Figure 3. Change in the band structure of an (8,0) SWBNNT. The origin of the energy scale is set at the Fermi level.
171
172
Figure 4. Energy gaps of the SWBNNTs as a function of (a) axial strain, (b) specific angle of twist, and (c) flattening
ratio.
Figure4showstheenergygapsofthe(6,0),(8,0),and(10,0)SWBNNTsasafunctionofthe
axialstrain,specificangleoftwist,andflatteningratio.Theenergygapofthe(10,0)with=41.7
deg/nm(Nz=5)isnotshowninthefigurebecauseitcollapsed.Undertensionandtorsionexcept
for=20.827.8deg/nminthe(10,0),theenergygapdecreasesalmostlinearlyandtherateof
decrease hardly depends on the diameter. In contrast, under flattening, the energy gap
decreasesquadraticallyorexponentiallyandtheamountofdecreasesignificantlydependson
thediameter;aSWBNNTwiththesmallerdiametershowsalargerdecreaseintheenergygap.
Itisalsoshownthatflatteningresultsinafewtimeslargerdecreaseintheenergygapthan
tensionandtorsion.
AlthoughthediscussionsofarinthissectionhasdealtwiththeSWBNNTsunderthethree
simple deformation modes, BNNTs would be subjected to combined deformation in their
practicaluse.Therefore,theenergygapoftheSWBNNTssubjectedtoflatteningfollowing
axialtensionisfurtheranalyzed(Figure5).ItisfoundthatprecedingtensionshiftsanEg
curvedownwardwithoutdramaticchangesinitsshape,andthattheextentoftheshiftalmost
correspondstotheenergygapdecreaseinducedbysimpletension(Figure4(a)).Thisresult
suggeststhattheenergygapoftheSWBNNTsunderacombinationofthethreedeformation
modescanbededucedfromFigure4.Intherestofthissection,therefore,onlythesimple
deformationmodeswillbediscussed.
2.2.2.ChargedensitiesattheCBM
HerethemechanismofdeformationinducedelectronicchangesintheSWBNNTsisdiscussed
intermsofchargedensitiesattheCBM(Figure6).TheCBMiscomposedofboronderived
states.Infact,CBMchargedensitiesarehigharoundboronatoms,whiletheyarelowaround
nitrogenatoms.Itisfoundthatthe*state(pzorbitalsofboronatoms)hybridizeswiththe*
Figure 5. Change in the energy gap of (8,0) SWBNNT under flattening following axial tension.}
state along a circumference passing through boron atoms under no deformation (Figure
6(a),zz=0.00),andthatthetension,torsion,andflatteninginducethechangeintheCBMstate.
Withincreasingflatteningdeformation(Figure6(d)),chargesaretransferredfromtheflattened
to the curved regions, resulting in an overlap of the charge densities and formations of
electronicbondsbetweenneighboringboronatomsinthecurvedregions.Itisthismechanism
thatresultsinthedecreaseinECBMintheflattenedSWBNNTs.ComparingthethreeSWBNNTs
with=0.45(Figures6(d)(f)),theelectronicbondsbecomestrongerasthediameterbecomes
smaller.Therefore,aflattenedSWBNNTwithasmallerdiametershowsalargerdecreasein
theenergygap.
Under tension (Figure 6(a)), the tube curvature increases because of Poisson contraction,
leadingtotheenhancementof**hybridizationsandthedecreaseinECBM.Figure6(a)shows
thenarrowingwhitecenterareaofzerochargedensitiesandthespreadinggrayareaof*
*hybridizations.Thesameistrueforthetorsion(Figure6(b)),butitinduceslesschangein
chargedensitiesthantension(thesizeofthewhitecenterareachangeslittleinFigure6(b)),
resultinginasmallerdecreaseintheenergygapundertorsionthanundertension(Figures
4(a),(b)).Itshouldbenotedthatelasticbucklingoccurredatabetween20.8and27.8deg/nm
in the (10,0), leading to local flattening (Figure 6(c)). Therefore, the relation of Eg versus
deviatesfromthelineardecreaseatof20.827.8deg/nminthe(10,0)(Figure4(b)).Itisobvious
thattheoverlapofchargedensitiesismuchstrongerunderflatteningthanundertensionor
torsion.Therefore,thedecreaseintheenergygapintheformerismuchlargerthaninthelatter.
173
174
Figure 6. Change in the CBM charge density. Cross sections passing through boron atoms are shown.
2.2.3.Deformationforces
Figure7showsthedeformationenergyasafunctionofaxialstrain,specificangleoftwist,and
flatteningratio.Thecurvesintension,torsion,andflatteningarefittedbycubic,quadratic,
andquarticpolynomials,respectively.Thefirstandsecondderivativesofeachcurveprovide
thedeformationforce(Figure8)andtheelasticmodulus,respectively.Youngsmoduliofthe
(6,0),(8,0),and(10,0)arethuscalculatedtobe0.759,0.794,and0.811TPa,respectively.They
areingoodagreementwiththosemeasuredinexperiments(1.220.24TPa[10]and0.722TPa
[16])andotherfirstprinciplescalculations(0.762,0.785,and0.803TPafor(6,0),(8,0),and(10,0),
respectively[18]).ItisfoundinFigure8thatforcesunderflatteningaresmallerthanunder
tensionandtorsion,becausestronginplaneBNcovalentbondspreventinplanetensionand
torsion.Itisalsofoundthatforcesrapidlyincreaselaterunderflattening.Therapidincrease
startsfromaround=0.3and0.4inthe(6,0)and(8,0),respectively,wheretheimaginarywall
distancesare0.35and0.38nm,respectively.BecausetheinterlayerdistanceofhexagonalBNs
andMWBNNTsisaround0.34nm,therapidincreasewouldbeattributedtotherepulsive
forcebetweenthetwoflattenedregions.
Figure 7. Deformation energy of the SWBNNTs as a function of (a) axial strain, (b) specific angle of twist, (c) flattening
ratio.
Figure 8. Forces required to deform (6,0), (8,0) and (10,0) SWBNNTs.
Figure9showstherelationshipbetweenenergygapanddeformationforce.Thethreebands
totherightillustratetheobtainablerangeoftheenergygapbyintroducingtension,torsion,
andflattening.Intensionandtorsion,alargerforceisrequiredforalargertubetoinducethe
sameamountofenergygapdecrease.Theoppositeistrueinflattening,i.e.,alargerforcefor
asmallertube.ThekeyfindingsfromFigure9arethat(i)theflatteningwithaforcesmaller
thanthatappliedfortensionortorsionleadstothelargerdecreaseintheenergygap,and(ii)
flatteningoffersalargerobtainablerangeoftheenergygapthantensionandtorsion:1.44.0
eVunderflattening,2.52.8eVand3.14.0eVundertension,2.72.8eV,3.43.6eV,and3.84.0
eVundertorsion.ThesefindingsindicatethatflatteninghasthepotentialtoenableBNNTsto
beusedasnanoelectronicdevices.However,avalidquestioniswhetherflatteningBNNTsis
experimentallyfeasible.
175
176
Figure 9. Relationship between energy gap and force of (6,0), (8,0) and (10,0) SWBNNTs.
Inordertoanswerthisquestion,theestimatedflatteningforcesarecomparedwiththoseof
SWCNTsthatBarbozaetal.havealreadyexperimentallysucceededinflatteningbymeansof
anatomicforcemicroscopy(AFM)tip[41].Althoughtheydidnotactuallymeasureflattening
forces of (n,0) SWCNTs with n 10, they proposed and validated a universal relationship
amongtheappliedforce,SWCNTdiameter,AFMtipradius,andflatteningratio:
F D03/2
(2R )1/2
(1 )3/2
2 + 2 + t g 1
(5)
whereRistheAFMtipradiusandisaconstant(=1.21018J).Equation(5)indicatesthatthe
( )1/2shouldbeuniversaltoanySWCNT.Theyshowedthatallexperimental
quantityFD3/2
0 2R
datafallonasinglecurveobtainedbyEquation(5)upto 0.4.FromEquation(5)andthe
geometriccontactconditionsbetweenatubeandanAFMtip,theflatteningforceperunit
lengthofa(6,0)SWCNT(D0=0.470nm)iscalculatedtobe15.4N/mwhen=0.4andR=30nm.
Incontrast,fromFigure8,thatofthe(6,0)SWBNNT(=F/Lz0)isestimatedtobe16.8N/mat
=0.4. The results demonstrate that the flattening force is almost equal in SWCNTs and
SWBNNTs, indicating that the same experiments as Barboza et al. would be feasible for
SWBNNTs.ThefactthatCNTsandBNNTsalmosthavethesametubeshapeandsizewhen
theirchiralindexesarethesame(a 0.142nminCNTsanda 0.145nminBNNTs)also
encouragesthefeasibilityofflatteningBNNTs.Itisthereforeconcludedthattheflattening
forcesestimatedarenotunrealisticandstronglyexpectedthatthesameorsimilarexperimental
techniquealsoappliestoBNNTs.
3.MWBNNTsunderflattening
3.1.Simulationprocedure
Thissectionfocuseson(5,0),(13,0),and(21,0)SW,(5,0)@(13,0)and(13,0)@(21,0)DW,and
(5,0)@(13,0)@(21,0)TWBNNTs.Figure10showsthesimulationmodelofthe(13,0)@(21,0)
DWBNNT.Theinitialnearestinteratomicdistancebetweenboronandnitrogenatomsisset
as0.145nm.Boron(nitrogen)atomsintheoutertubearestackedabovenitrogen(boron)
atomsintheinnertube[42].TheaxialdirectionoftheBNNTisparalleltothezdirection.
TheBNNTislocatedatthecenteroftheunitcellwithasizeof3.637nm3.637nm0.435
nm.Eventhoughathreedimensionalperiodicboundaryconditionisemployed,thecellsizes
inthexandydirectionsarelargeenoughtoavoidinteractionwithneighboringimagecells,
becausetheyhavelittleeffect(lessthan1%)ontheenergy,chargedistribution,andenergy
bandstructureofaflattenedBNNT,whentheyaregreaterthanthediameteroftheBNNT
plus1.0nm.
Figure 10. Simulation model of (13,0)@(21,0) DWBNNT.
177
178
Figure 11. Schematic illustration explaining flattening compression of BNNTs.
Atomic positions and the cell size in the zdirection are first relaxed using the conjugate
gradient method until atomic forces and the stress component, zz, become less than 0.01
eV/ and 0.01 GPa, respectively. After obtaining the equilibrium structure, a flattening
compressioninthexdirectionisappliedbyreducingthedistancebetweenimaginarywalls
untiltheBNNTcollapses(Figure11).Onceanatomcontactsawall,theatomisallowedto
moveonlyonthewall.Duringcompression,thecellsizesarefixedandatomicconfigurations
arerelaxeduntiltheirforcesbecomelessthan0.01eV/.Toinvestigatethedegreeofdefor
mation,theflatteningratio,,isdefinedas
=
D0 D
D0
(6)
whereD0isthediameteroftheoutermosttubeatequilibrium,andDisthedistancebetween
theimaginarywalls.
FirstprinciplesDFTcalculationsareconductedusingtheViennaAbInitioSimulationPackage
(VASP)[32,33].Thewavefunctionsareexpandedinaplanewavebasissetwithacutoff
energyof350eV.TheultrasoftpseudopotentialproposedbyVanderbilt[34]isused,andthe
exchangecorrelation energy is evaluated by the generalized gradient approximation of
PerdewandWang[35].TheBrillouinzoneintegrationisperformedbytheMonkhorstPack
scheme[36]usinga114kpointmesh.
3.2.Resultsanddiscussion
3.2.1.Energybandstructures
Figure 12 shows the change in the energy band structures of the (13,0) SWBNNT and
(13,0)@(21,0)and(5,0)@(13,0)DWBNNTsduringflatteningdeformation.TheotherSWBNNTs
andthe(5,0)@(13,0)@(21,0)TWBNNTshowasimilarchangingtrendofthebandstructureto
the (13,0) SWBNNT and the (5,0)@(13,0) DWBNNT, respectively. Both the valence band
maximum (VBM) and the conduction band minimum (CBM) of the (13,0) SWBNNT and
(13,0)@(21,0)DWBNNTarelocatedatthepoint(k=0)duringthedeformation,butthoseof
the(5,0)@(13,0)DWBNNTmovetok0midwayduringthedeformationandthenreturntothe
point.IneachBNNT,theenergyoftheVBM,EVBM,hardlychanges,whilethatoftheCBM,
ECBM,changes,indicatingthatthechangeintheenergygap,Eg,ismainlycausedbyachange
in ECBM. In the (13,0) SW and (13,0)@(21,0) DWBNNTs (Figure 12(a), (b)), ECBM decreases
monotonically.Incontrast,inthe(5,0)@(13,0)DWBNNT(Figure12(c)),ECBMfirstincreasesand
thendecreases.
Figure 12. Change in the band structure of (13,0) SWBNNT and (13,0)@(21,0) and (5,0)@(13,0) DWBNNTs in flatten
ing deformation. The origin of the energy scale is set at the Fermi level.
179

180

Figure 13. Relationship between energy gap, Eg, and flattening ratio, , of (5,0), (13,0), and (21,0) SWBNNTs,
(5,0)@(13,0) and (13,0)@(21,0) DWBNNTs, and (5,0)@(13,0)@(21,0) TWBNNT.
Figure13showstheenergygapsofthe(5,0),(13,0),and(21,0)SWBNNTs,(5,0)@(13,0)and
(13,0)@(21,0)DWBNNTs,and(5,0)@(13,0)@(21,0)TWBNNTasafunctionoftheflatteningratio.
TheenergygapofthethreeSWBNNTsdecreasesalmostmonotonically,andtheamountof
decreasebecomessmallerwithincreasingtubediameter.Theenergygapofthe(13,0)@(21,0)
DWBNNTalsodecreasesmonotonically,butitexhibitsamorerapiddecreasethanthe(13,0)
and(21,0)SWBNNTs.Itshouldbenotedthattheenergygapsofthe(5,0)@(13,0)DWBNNT
and(5,0)@(13,0)@(21,0)TWBNNTincreaseduringtheearlystageandthendecrease.Thisshift
occursearlierinthelatter(=0.34)thanintheformer(=0.48).Thefactthatthe(5,0)@(13,0)
DWBNNTand(5,0)@(13,0)@(21,0)TWBNNTshowdifferentchangingtrendsofEgfromthe
SWBNNTsprovesthatinterwallinteractionssignificantlyaffecttheelectronicstructuresofthe
flattened(5,0)@(13,0)DWBNNTand(5,0)@(13,0)@(21,0)TWBNNT.
3.2.2.ChargedensitiesattheCBM
Figure14showschargedensitiesattheCBMoftheflattened(5,0),(13,0),and(21,0)SWBNNTs
atacrosssectionpassingthroughboronatoms.Thecharacteristicsofthenearlyfreeelectron
(NFE)stateareobservedintheBNNTswithasmallcurvature(Figure14(b):=0.00,(c):=0.00,
0.20),while**hybridizationsappearintheothers.ThereasontheCBMofthe(13,0)and
(21,0)SWBNNTschangesfromaNFElikestatetoa**hybridizedstateisthatthetube
curvatureincreaseslocallyastheflatteningdeformationincreases(in(n,0)SWBNNTsunder
nodeformation,theCBMisaNFElikestatewhenn13anda**hybridizedstatewhenn
<13,andthehybridizationbecomesstrongerwithincreasingtubecurvature)[8].Theenergy
gapofthe(5,0)SWBNNTismuchsmallerthanthoseofthe(13,0)and(21,0)SWBNNTsbecause
ofitsstrong**hybridization.Withincreasingflatteningdeformation,chargeistransferred
fromflattenedregionstocurvedones,leadingtoanoverlapofthechargedensities.TheECBM
of the SWBNNTs decreases under flattening because of the formation of electronic bonds
betweenneighboringboronatomsinthecurvedregions.Thechargedensitydistributionin
curvedregionsofthe(13,0)SWBNNTat=0.21issimilartothatofthe(21,0)SWBNNTat
=0.52,whichresultsinthemhavingalmostthesameenergygapof4.2eV.Thisisbecause
theyhavealmostthesamevalueofD,namelythesamecurvatureofthecurvedregion.Figure
15showstherelationshipbetweentheenergygapandimaginarywalldistanceofthe(13,0)
and(21,0)SWBNNTs.Theirenergygapsarealmostequalunderasamewalldistance.

Figure 14. Change in the CBM charge density of (5,0), (13,0), and (21,0) SWBNNTs in flattening deformation.
181

182

Figure 15. Energy gap of (13,0) and (21,0) SWBNNTs as a function of imaginary wall distance.

Figure 16. Change in the CBM charge density of (13,0)@(21,0) DWBNNT during flattening deformation.
TheCBMchargedensitydistributionofthe(13,0)@(21,0)DWBNNTissimilartothatofthe
SWBNNTs(Figure16).Intheinnertube,chargetransferfromflattenedtocurvedregionsis
observedandanoverlapofthechargedensitiesisinducedinthecurvedregions.Thedecrease
inECBMofthe(13,0)@(21,0)DWBNNTiscausedbythesamemechanismasintheSWBNNTs
mentionedabove.Becausethechargedensitiesaredistributedalmostentirelyintheinnertube
during deformation, one might expect that the Eg curve of the (13,0)@(21,0) DWBNNT
coincideswiththatofthe(13,0)SWBNNT.However,Egoftheformerisinfactsmallerthan
thatofthelatterunderthesame.AsshowninFigure16,theflatteningratiooftheinnermost
tube,in,mustbelargerthantomaintaintheinterwallspacingconstant.Thismeansthatthe
BBbondsintheflattened(13,0)@(21,0)DWBNNTarestrongerthanthoseintheflattened(13,0)
SWBNNTunderthesame,resultinginalargerdecreaseinECBMintheformerthaninthe
latter.

!

#

"
Figure 17. Change in the CBM charge density of (5,0)@(13,0) DWBNNT and (5,0)@(13,0)@(21,0) TWBNNT during
flattening deformation.
183
184
It should be noted that the (5,0)@(13,0) DWBNNT and (5,0)@(13,0)@(21,0) TWBNNT show
different changes in CBM charge densities from those of the SWBNNTs and (13,0)@(21,0)
DWBNNT(Figure17).First,theCBMchargedensitiesgraduallytransferfromboronatomsin
theinnermosttubetoboronatomsinthesecondinnermosttube.Thischargedelocalization
and spreading account for the increase in ECBM in the (5,0)@(13,0) DWBNNT and
(5,0)@(13,0)@(21,0)TWBNNTduringtheearlystageofthedeformation.Then,overlapofthe
chargedensitiesincurvedregionsinthesecondinnermosttubeisinducedwithincreasing
deformation. Consequently, ECBM of the (5,0)@(13,0) DWBNNT and (5,0)@(13,0)@(21,0)
TWBNNTdecreaseslaterinthedeformation.Itisthismechanismthatresultsintheinitial
increaseandsubsequentdecreaseinEginthe(5,0)@(13,0)DWBNNTand(5,0)@(13,0)@(21,0)
TWBNNT.Becausethelatterhasalargerinthantheformerunderthesame(seetheatomic
positionsintheinnermosttubeatof0.43inFigure17),chargespreadingfromthefirsttothe
secondinnermosttubeiscompletedearlierinthelatter.Therefore,thelattershowsanearlier
shiftfromincreasetodecreaseinEg.
3.2.3.Criticaldiameteroftheinnermosttube
Itisevidentthatacriticalchiralindex,(nc,0),existsforflattenedzigzagMWBNNTs.Consider
theinnermosttubeofazigzagMWBNNT,denotedas(nin,0).Inthecaseofnin>nc,Egdecreases
monotonicallyastheflatteningdeformationincreases.Inthereversecase(nin<nc),Egfirst
increasesandthendecreases.Fromtheresultsobtainedinthisstudy,ncisproventobean
integerbetween5and13.Furthermore,ifzigzagBNNTshavethesamenin,azigzagBNNT
containing more walls shows a more rapid change in Eg (compared with the change in Eg
between (13,0) SW and (13,0)@(21,0) DWBNNTs and between (5,0)@(13,0) DW and
(5,0)@(13,0)@(21,0)TWBNNTsinFigure13).
IntheaforementionedexperimentalstudyonabentMWBNNT,anotabletendencyforazigzag
atomicarrangementandlocalflatteninghavebeenobserved[30].JudgingfromtheHRTEM
images,ninoftheMWBNNTislargerthannc.Therefore,itcanbesaidthatapossiblereason
forthechangefrominsulatingtosemiconductinginthebentMWBNNTiselectronicchanges,
asshowninFigure16.Tothebestofourknowledge,therehasbeennoexperimentalstudyon
deformed BNNTs with nin< nc, but the results for the (5,0)@(13,0) DWBNNT and
(5,0)@(13,0)@(21,0) TWBNNT obtained in this study are expected to be good predictions.
BendingexperimentsonBNNTswithnin<ncaregreatlyanticipated.
4.Summary
InSection2,theelectronicstructuresof(n,0)zigzagSWBNNTssubjectedtotension,torsion,
andflatteningwereinvestigatedbyfirstprinciplesDFTcalculations.Theresultsrevealedthat
the three deformation modes decrease the energy gaps of the SWBNNTs because of the
decreaseintheCBMenergycausedbyanoverlapofCBMchargedensitiesbetweencircum
ferentiallyneighboringboronatoms.ThekeyfindingsofSection2arethatflatteningwitha
forcesmallerthanthatappliedfortensionortorsioncausesalargerdecreaseintheenergygap
andthattheforcerequiredforflatteningSWBNNTsisnotunrealistic.
InSection3,electronicstructuresofflattenedMWBNNTswiththezigzagchiralindex(n,0)
wereinvestigatedbyfirstprinciplesDFTcalculations.ThekeyfindingsobtainedinSection3
aresummarizedasfollows:
1.
Whenthechiralindexoftheinnermosttube,nin,ofazigzagMWBNNTislargerthanthe
critical one, nc, the energy gap decreases monotonically with increasing flattening
compression.
2.
Whennin<nc,theenergygapfirstincreasesandthendecreaseswithincreasingflattening
compression. This initial increase and subsequent decrease are caused by a charge
spreadingfromthefirsttothesecondinnermosttubeandbybondformationincurved
regionsinthesecondinnermosttube,respectively.
3.
Thencisfoundtobeanintegerbetween5and13.
Authordetails
NobutadaOhno,DaiOkumuraandYusukeKinoshita
Department of Computational Science and Engineering, Graduate School of Engineering,
NagoyaUniversity,Nagoya,Japan
References
[1] HohenbergP,KohnW.InhomogeneousElectronGas.PhysicalReview1964;136(3B)
B864B871.
[2] KohnW,ShamLJ.SelfConsistentEquationsIncludingExchangeandCorrelationEf
fects.PhysicalReview1965;140(4A)A1133A1138.
[3] Jones RO, Gunnarsson O. The density functional formalism, its applications and
prospects.ReviewsofModernPhysics1989;61(3)689746.
[4] LundqvistS,MarchN.TheoryoftheInhomogeneousElectronGas.NewYork:Ple
numPress;1983.
[5] MartinR.M.ElectronicStructure.Cambridge:CambridgeUniversityPress;2004.
[6] Kinoshita Y, Hase N, Ohno N. Flatteninginduced electronic changes in zigzag sin
gleandmultiwalledboronnitridenanotubes:AfirstprinciplesDFTstudy.Physical
ReviewB2009;80(12)125114.
185
186
[7] Kinoshita Y, Ohno N. Electronic structures of boron nitride nanotubes subjected to

tension, torsion, and flattening: A firstprinciples DFT study. Physical Review B
2010;82(8)085433.
[8] BlaseX,RubioA,LouieS.G,CohenM.L.StabilityandBandGapConstancyofBor
onNitrideNanotubes.EurophysicsLetters1994;28(5)335340.
[9] RubioA,CorkillJ.L,CohenM.L.Quasiparticlebandstructuresofshortperiodsu
perlattices and ordered alloys of AlN and GaN, Physical Review B 1994;49(3)
19521956.
[10] ChopraN.G,LuykenR.J,CherreyK,CrespiV.H,CohenM.L,LouieS.G,ZettlA.
Boronnitridenanotubes.Science1995;269(5226):966967.
[11] GolbergD,BandoY,TangC,ZhiC.BoronNitrideNanotubes.AdvancedMaterials
2007;19(18)24132432.
[12] ChopraN.G,ZettlA.Measurementoftheelasticmodulusofamultiwallboronni
tridenanotube.SolidStateCommunications1998;105(5)297300.
[13] DumitricaT,BettingerH.F,ScuseriaG.E,YakobsonB.I.Thermodynamicsofyieldin
boronnitridenanotubes.PhysicalReviewB2003;68(8),085412.
[14] Golberg D, Bando Y, Kurashima K, Sato T. Synthesis and characterization of ropes
madeofBNmultiwallednanotubes.ScriptaMaterialia2001;44(89)15611565.
[15] IijimaS.HelicalmicrotubulesofgraphiticcarbonNature1991;3545658.
[16] SuryavanshiA.P,YuM.F,WenJ,TangC,BandoY,Elasticmodulusandresonancebe
haviorofboronnitridenanotubes.AppliedPhysicsLetters2004;84(14)25272529.
[17] AkdimB,PachterR,DuanX,AdamsW.W.Comparativetheoreticalstudyofsingle
wallcarbonandboronnitridenanotubes.PhysicalReviewB2003;67(24)245404.
[18] BaumeierB,KrugerP,PollmanJ.Structural,elastic,andelectronicpropertiesofSiC,
BN,andBeOnanotubesPhysicalReviewB2007;76(8)085407.
[19] KudinK.N,ScuseriaG.E,YakobsonB.I.C2F,BN,andCnanoshellelasticityfrom
abinitiocomputations.PhysicalReviewB2001;64(23)235406
[20] PengY.J,ZhangL.Y,JinQ.H,LiB.H,DingD.T.Abinitiostudiesofelasticproper
ties and electronic structures of C and BN nanotubes. Physica E: Lowdimensional
SystemsandNanostructures2006;33(1)155159
[21] HernandezE,GozeC,BernierP,RubioA.ElasticPropertiesofCandBxCyNzCom
positeNanotubes.A.PhysicalReviewLetters1998;80(20).
[22] HernandezE,GozeC,BernierP,RubioA.AppliedPhysicsA1999;68(3)287292.
[23] GriebelM,HamaekersJ,HeberF.Amoleculardynamicsstudyontheimpactofde
fectsandfunctionalizationontheYoungmodulusofboronnitridenanotubes.Com
putationalMaterialsScience2009;45(4):10971103.
[24] LiC,ChoT.W.JournalofNanoscienceandNanotechnology;2006.
[25] SantoshM,MaitiP.K,SoodA.K.Elasticpropertiesofboronnitridenanotubesand
theircomparisonwithcarbonnanotubes.JournalofNanoscienceandNanotechnology
2009;9(9):54255430.
[26] SongJ,WuJ,HuangY,HwangK.C,JiangH.Stiffnessandthicknessofboronnitride
nanotubes.JournalofNanoscienceandNanotechnology2008;8(7)37743780.
[27] Verma V, Jindal V. K, Dharamvir K. Elastic moduli of a boron nitride nanotube.
Nanotechnology2007;18(43),435711.
[28] TreacyM.M.J,EbbesenT.W,GibsonJ.M.ExceptionallyhighYoungsmodulusob
servedforindividualcarbonnanotubes.Nature1996;381(678680).
[29] KrishnanA,DujardinE,EbbesenT.W,YianilosP.N,TreacyM.M.J.Youngsmodu
lusofsinglewallednanotubes.PhysicalReviewB1998;58(20)1401314019.
[30] BaiX,GolbergD,BandoY,ZhiC,TangC,MitomeM,KurashimaK.Deformation
driven electrical transport of individual boron nitride nanotubes. Nano Letters
2007;7(3)632637.
[31] KimY.H,ChangK.J,LouieS.G.ElectronicstructureofradiallydeformedBNand
BC3nanotubes.PhysicalReviewB2001;63(20)205408.
[32] KresseG,FurthmullerJ.Efficientiterativeschemesforabinitiototalenergycalcula
tionsusingaplanewavebasisset.PhysicalReviewB1996;54(16)1116911186.
[33] KresseG,FurthmullerJ.AbinitiomoleculardynamicsforliquidmetalsPhysicalRe
viewB1993;47(1)558561.
[34] Vanderbilt D. Soft selfconsistent pseudopotentials in a generalized eigenvalue for
malism.PhysicalReviewB1990;41(11)78927895.
[35] PerdewJ.P,WangY.Accurateandsimpleanalyticrepresentationoftheelectrongas
correlationenergy.PhysicalReviewB1992;45(23)1324413249.
[36] MonkhorstH.D,ParkJ.D.OnSpecialPointsforBrillouinZoneIntegrations.Physi
calReviewB1976;13(12)51885192.
[37] ArnaudB,LebegueS,RabillerP,Alouani,M.Hugeexcitoniceffectsinlayeredhex
agonalboronnitride.PhysicalReviewLetters2006;96(2)026402.
[38] BlaseX,RubioA,LouieS.G,CohenM.L.Quasiparticlebandstructureofbulkhex
agonalboronnitrideandrelatedsystems.PhysicalReviewB1995;51(11)68686875.
187
188
[39] ParkC.H,SpataruC.D,LouieS.G.ExcitonsandManyElectronEffectsintheOpti
cal Response of SingleWalled Boron Nitride Nanotubes. Physical Review Letters
2006;96(12)126105.
[40] WirtzL,MariniA,RubioA.Excitonsinboronnitridenanotubes:dimensionalityef
fects.PhysicalReviewLetters2006;96126104.
[41] BarbozaA.P.M,ChachamH,NevesB.R.A.UniversalResponseofSingleWallCar
bonNanotubestoRadialCompression.PhysicalReviewLetters2009;102(2)025501.
[42] JhiS.H,RoundyD.J,LouieS.G,CohenM.L.Formationandelectronicpropertiesof
doublewalled boron nitride nanotubes. Solid State Communications 2005;134(6)
397402.
Chapter 9
Measurement of Frictional Properties on

the Micro/Nanometer Scale
Kenji Fukuzawa
1.Introduction
Scanningprobemicroscopes(SPMs)wereinventedbyBinnigandRohrerin1980sandhave
beenoneofthekeymethodsformeasurementsandmanipulationsinnanotechnologiessince
then [1, 2]. The first SPMs were a scanning tunnelling microscope (STM), which uses a
tunnellingcurrent.Theverticalresolutionofsubnmcomesfromtheextremelyshortlength
ofinteractionbetweentheprobeandsamplesurface.Ifasharptipcanbeprepared,thelateral
resolutionofsubnmispossible.STMcanseesamplesurfacesatanatomicresolutionand
manipulateasingleatom.TheconceptofSTMcanbeextendedtovariousfieldsandwide
varietiesofpropertiescanbemeasuredonthenanoscalebychangingtheinteractionbetween
theprobeandsample.Itshouldbenotedthatmicrofabricationtechniquesbasedonmicro
machining techniques has been indispensable to the development of SPMs. Atomic force
microscopes(AFM)areoneofthemostsuccessfulmicroscopesinSPMs[3].Asaprobe,they
use micro cantilevers with a tiny tip, whose apex radius is around several nm. Typical
dimensionsofthecantileverare100mlongandafewmthick.Microstructuresatthese
dimensionsarenoteasytofabricateforconventionalmachining.Therefore,microfabrication
methodshavebeendeveloped.MicromachingnotonlyimprovedtheperformanceofSPMs
butalsomadethemapopulartechnologybecausethefabricationmethodsweresuitablefor
massproduction.
SPMhavebeenextendingitsapplicationtomechanicalfields.Oneofthesuccessfulapplica
tionsisafrictionforcemicroscope(FFM),whichhasbeenakeytechnologyofmicro/nano
tribology[4].FFMcanmeasurethelocalfrictionalorrheologicalpropertiesatamicro/nano
scalealthoughAFMislimitedtomeasurementofthetopography.Theperformanceofmicro/
nano mechanical devices, such as micro/nano electro mechanical systems (MEMS/NEMS),
computerharddiskdrives(HDDs),isdominatedbysurfaceforcessuchasfrictionandviscous
forcesratherthanvolumeforcessuchasgravitationalforces.Therefore,themeasurementof
thelocalfrictionalpropertiesisveryimportantfortheperformanceanalysisanddesignof
2013 Fukuzawa; licensee InTech. This is an open access article distributed under the terms of the Creative
190
micro/nanomechanicaldevices.FFMisalsoappliedtolocalcharacterizationofthesample
surface by using that different materials usually have different frictional properties. If the
sampleconsistsofcompositematerialssuchaslaminatedmaterialsorcarbonfiberreinforced
materials, the local composition can be evaluated at a nanometer scale by measuring the
frictionalpropertydistribution.Recently,thelocalchemicalcompositionandpropertiessuch
ashydrophobicitycanbemappedbymeasuringfrictionforcedistributionwithachemically
functionalisedprobe.Inthischapter,thebasisanddrawbacksofFFMarebrieflyreviewedand
recenttrialstoovercomethedrawbacksaredescribed.
Movement
by deflection
Laser for
optical lever method
Laser spot
Movement D
C
by twist
Detection
plan of PD
Susbstrate
(Probe base)
Micro-cantilever
Four segmented
photodiode (PD)
Sample
Friction force
(Lateral force)
Tip
Load
(vertical froce)
Figure 1. Schematic of conventional micro-cantilever-based FFM.
2.ConventionalFFM
Conventional FFMs use a micro cantilever as a probe, which is used in AFM. By slightly
modifyinganAFMsetup,replacingatwosegmentedphotodiode(PD)asapositionsensitive
detector(PSD)withafoursegmentedone,FFMmeasurementbecomepossible.Becauseof
thisconvenience,manyofthecommercialAFMshaveafoursegmentedphotodiodeandcan
make a FFM measurement with a micro cantilever probe for AFM. However, probes are
designedforAFM,notforFFM.ThislimitstheperformanceofFFM.InSPMs,thedevelopment
of the probe is significant as described above. Although many trials have been done to
overcome the drawbacks, a standard probe for FFM has not been established yet. Here,
measurementprincipleofconventionalFFMsandthentheirdrawbacksareexplained.
Measurement of Frictional Properties on the Micro/Nanometer Scale
2.1.MeasurementprincipleofconventionalFFM
WhentheprobeisscannedunderaconstantloadconditionasshowninFigure1,themicro
cantilever twists along its long axis due to the friction force applied to the probe tip. FFM
measuresthefrictionforcesbydetectingthistorsionofthecantilever.Inadditiontothefriction
force,theload,theforceintheverticaldirection,issimultaneouslyappliedtothetip.This
causestheverticaldeflectionofthemicrocantilever.Letteringandzbetorsionangleand
verticaldeflection,respectively.Thefrictionforceinlateraldirection,FLandtheloadinthe
verticaldirection,FVarerespectivelygivenby
kl
(1)
F V = kv z
(2)
FL =
whereklandkvaretothetensionalrigidityandspringconstantofthecantilever.Inaddition,
listhelengthofthecantilever.Thefrictionforceandloadcanbesimultaneouslyobtained
frommeasuredtorsionangleandverticaldeflectionzbyusingEqs.(1)and(2).Ifthesharp
tipwhoseapexradiusisaroundonenanometerispreparedandtheprobeisscannedonthe
samplesurfacewithapiezoscanner,thefrictioncoefficient,whichisthemostfundamental
frictionalproperties,canbemappedonananometerscale.Next,amethodformeasuringthe
torsion angle and vertical deflection z is explained. This is based on the method for
measuringtheverticaldeflectioninAFM,whichiscalledopticallevermethod.Aschematic
setupisshowninFigure1.Thelightfromthelaserdiodeisfocusedontothemicrocantilever
surfaceandthereflectedlightformsthelightspotonthefoursegmentedPD.ThefourPDs
detectsthelightintensitiesseparatelyandthedifferenceinPDsignals,(IA+IB)(IC+ID)and(IA
+IC)(IB+ID) are obtained. Here, IA, IB, IC, and ID are the signals from PD of A, B, C, and D,
respectively.Whentheloadisappliedtotheprobeandmicrocantileverdeflectsvertically,the
lightspotmovesintheverticaldirection.Thiscausesthechangeinthedifferentialsignal(IA
+IB)(IC+ID).Inasimilarfashion,thefrictionforceisapplied,thedifferentialsignal(IA+IC)(IB
+ID)changesduetothelateraldisplacementofthelightspot,whichisgeneratedbythetorsion
ofthemicrocantilever.ThemodeofprobescanninginFFMisthesameasthecontactAFM
mode.Inthismode,theprobeisscannedunderaconstantloadcondition,whichisachieved
bycontrollingtheprobeheightwithapiezoactuatorsothattheverticaldeflectionsignal(IA
+IB)(IC+ID)keepsasetvalue.TypicalprobesforFFMareIandVshapedmicrocantileversas
showninFigure2.Usually,IshapedcantileverismadeofsinglecrystalsiliconandVshaped
one is made of silicon nitride. Typical spring constant in vertical direction, KV is around 1
N/m.BothprobesarebasicallydesignedforcontactmodeAFM.Thus,onlytheintroduction
ofthefoursegmentedPDcanprovideFFMmeasurementwithusualAFMsetups.
2.2.DrawbacksofconventionalFFMprobes
ConventionalFFMprobeshavebeenveryusefulformappingfrictionpropertiesandproduc
ingfruitfulresultsinvariousfields,however,ithasafundamentalproblemtobesolved.It
191
192
usesthetorsionangleandverticaldeflectionzfordetectingthefrictionforceFLandload
FV.Twodeformationshavetobeindependentofeachotherforaccuratemeasurementofthe
friction force and load. Considering that both deformations occur at the end of the micro
cantilever,theyarepossibletointerferewitheachother.Theindependentdeformationsare
validonlywhenthedeformationsaresmallenough,whichmeansthatthefrictionforceand
loadaresmallenough.ThismechanicalinterferencecausesignificantdrawbacksinFFM.
(a)
(b)
Microcantilever
Microcantilever
Tip
Probe base
Tip
Probe base
Figure 2. Typical types of micro cantilever for FFM. (a) I-shaped and (b) V-shaped cantilevers. The figures are viewed
from the tip side.
Thefirstdrawbackisthedecreaseofthemeasurementaccuracy[5,6].AsdescribedinSection
2.1,inEqs.(1)and(2),thetensionalrigidityklandspringconstantkvareassumedtobeconstant.
However,whenthefrictionforceislarge,thetorsionalrigidityklandthespringconstantkv
dependsontheloadandfrictionforce.TheselimitFFMapplicationsforconventionalprobe.
Therefore,conventionalFFMsremainatthelevelofqualitativeevaluationmeansinmany
applications.
The second drawback is the decrease of the measurement performance. The fundamental
equations,Eqs.(1)and(2)indicatethattheimprovementoftheforcesensitivityrequiresthe
decrease of the rigidity, kl or kv. In contrast, the reduction of the mechanical interference
requiresthehighrigidityforthecounterpartdeformation.Forexample,thetorsionalrigidity
hastobesethighinordertomeasuretheloadaccurately.SincetheVshapedcantileversare
designedforacontactAFMmode,crossedcantileverstructureisadoptedinordertoincrease
klandreducethetorsionofthecantilever.Thisleadstothedecreaseofthesensitivityofthe
frictionforce.Inordertomeasurethefrictionforceaccurately,theverticalspringconstantkv
has to be higher, which decreases the load sensitivity. Thus, the design for the high force
sensitivitiesofboththefrictionforceandloadisdifficultforconventionalcantileverprobedue
tothemechanicalinterference.InconventionalFFMs,sinceprobesaredesignedforAFM,most
oftheprobeissetsoastoreducetorsionofthecantilever.Therefore,frictionforcesensitivity
ismadevictimofthatoftheload.
In addition to the drawbacks due to the mechanical interference, conventional FFMs have
anotherfundamentaldrawback.AsdescribedinSection2.1,thetorsionofthemicrocantilever
isdetectedtoobtainthefrictionforce.Thisisimplementedbytheopticallevermethod.The
optical lever method can provide the differential signal (IA+IC)(IB+ID) corresponding to the
torsionangleinEq.(1).Forquantitativemeasurement,therelationshipbetweenthesignal
(IA+IC)(IB+ID)andanglehastobeknown.Forthispurpose,thetorsionangleofthemicro
cantileverhastobemeasured.Inaddition,thismeasurementhastobedoneforthemicro
cantileverinstalledinAFMsetupbecausethesignal(IA+IC)(IB+ID)maychangeforthesetup
tosetup.Thistorsionanglemeasurementisnoteasyinusualsetups.Therefore,thestandard
calibrationmethodforthefrictionforcehasnotbeenestablishedyet.Thisisoneofthereasons
whyFFMremainsatthelevelofqualitativeevaluationmeans[7].
3.DualaxisFFMprobes
Manytrialshavebeendoneinordertoovercometheabovedrawbacksoftheconventional
FFM.Onepromisingcandidateisadualaxisprobe.Themechanicalinterferenceproblemin
theconventionalprobeiscausedbythefactthattheprobeusesthetwodeformationsofthe
samepart,theprobeend.Ifthedifferentpartsoftheprobedeformduetothefrictionforce
andload,themechanicalinterferencecanbereducedoreliminated.Dualaxisprobeisbased
onthisidea.Chuietal.presentedtheprobethatseriallyconnectedanarrayofcantileversto
aVshapedcantileverwithatip[8].TheschematicstructureisshowninFigure3.Thesurface
ofthecantileverarrayissetverticallytotheVshapedcantileverone.Thedoublecantilever
arraylaterallydeflectsforthefrictionforceandVshapedcantileververticallydeformsforthe
load.Thefrictionforceandloadisobtainedfromtheselateralandverticaldeflections.Ando
etal.presentedtheconceptofanothertypeofdualaxisprobe,wherethetwodoublecantile
versthatrotateby90degreestoeachotherareseriallyconnectedasshowninFigure4[9].The
doublecantileverwithitssurfacenormaltothesamplesurfacedeflectsforthefrictionforce
andanotherdoublecantileverdoesfortheload.
ThedetectionmethodforverticaldeflectionoftheprobeforconventionalFFMcanbeapplied
tothedualaxisprobewhereasthatforlateralonecannotbecausetheprobedoesnottwistand
deflectslaterally.Intheopticallevermethod,whichiswidelyusedinconventionalFFMs,the
lightdoesnotmoveonthePDeveniftheprobedeflectslaterally.Therefore,theopticallever
methodcannotbeappliedtodualaxisprobesdirectly.Moreover,inthebothdualaxisprobes,
thedeformationpartsareseriallyconnectedasshowninFigures3and4.Thedisplacementof
193
194
theprobeendisthesumofthedeformationsoftwodeformationpartssuchasthecantilever
arrayandVshapedcantilever.Thismeansthatthedeformationsofthetwopartscannotbe
determinedbymeasuringthedisplacementoftheprobeendascanbedoneinconventional
FFMs.IntheprobethatpresentedbyChuietal.,inordertosolvethisproblem,theyfabricated
averyspecialtypeoftheprobethathaspiezoresistivesensorsbothonthelateralandvertical
cantileverstodetectlateralandverticaldeflections.Ingeneral,thepiezoresistivesensorisnot
easytofabricateonthedoublecantileverarraybecausetheirsurfaceisnormaltothebase
substrate.IntheprobethatpresentedbyAndoetal.,thesolutionforthisproblemwasona
conceptuallevelandnotdemonstratedexperimentally.Thus,thestandardprobeforthedual
axisFFMhasnotbeenestablishedyet.Recently,anewtypeofdualaxisprobecalledmicro
mechanicalprobe(MMP)hasbeenpresented[10].Thisprobecanovercomethesignificant
problemofthedualaxisprobeasmentionedabove.ThedetailsoftheMMParedescribedbelow.
V-shaped cantilever
for load detection
Silicon substrate
Tip
Double cantilever
for friction force detection
Figure 3. Schematic of dual-axis FFM probe with an array of double cantilevers for the friction force detection and Vshaped cantilever for the load detection.
Silicon
substrate
Double cantilever
for load detection
Tip
Double cantilever
for friction force detection
Figure 4. Schematic of a dual-axis FFM probe with serially connected double cantilevers for the friction and load
detections
3.1.ConceptofMicroMechanicalProbe(MMP)
ThestructureandsetupofMMPareschematicallyshowninFigure5.Thisprobeconsistsof
thedoublecantileverforthefrictionforcedetectionandthetorsionbeamfortheloaddetection.
Thetipisfabricatedattheendofthedoublecantilever.Thedoublecantileverissupportedby
thetorsionbeamandrotatesalongthetorsionbeamwhentheloadisappliedtothetip.Inthis
probe,thedeformationpartsarenotconnectedseriallyasdoneinthedualaxisprobesthat
abovementioned.Thiscanprovidethatboththedeformationsofthedoublecantileverand
torsionbeamscanbemeasuredbydetectingthedisplacementoftheprobeend.Bydetecting
thelateraldeflectionofthedoublecantileverxandthetorsionangleofthetorsionbeam,,
thefrictionforceFLandloadFVcanbeobtainedby
F L = kdl x
(3)
ktv
lt t
(4)
FV =
wherekLdandkVtarethespringconstantofthedoublecantileverandtorsionalrigidityofthe
torsionbeam,respectively.IntheMMP,thelateraldeflectionxandthetorsionanglecanbe
measuredbydetectingthelateralandverticaldisplacementsoftheprobeend.Asdescribed
above,thelateraldisplacementofthedualaxisprobecannotbedetectedbytheopticallever
method.IntheMMP,thelowreflectionpatterncalledopticalleverpatternisfabricatedonthe
probeheadsurface.Asoneoftheopticalleverpattern,amicroroofshapedstructurewas
presented[11].Thelightthatfocusedontotheroofshapedstructurescatterwhereasthelight
focusedontotheflatsurfacereflectsandgoestothePDfortheopticallevermethod.Therefore,
thereflectionlightformsthelightspotwithadarkpatternonthePD.Whentheloadisapplied,
thelightspotonthePDmovesintheverticaldirectionasdoesinconventionalFFM.Incontrast,
whenthefrictionforceisapplied,thespotdoesnotmovebutthedarkpatterngeneratedby
theopticalleverpatternmoveslaterally.Therefore,thedifferentialsignal(IA+IB)(IC+ID)gives
theverticaldisplacementoftheprobeheadorthetorsionangleofthetorsionbeamwhereas
(IA+IC)(IB+ID)givesthelateraldeflectionofthedoublecantileverx.Thismethodformeasur
ing the vertical and lateral displacements are the same as that in conventional FFMs as
explainedinSection2.1.Thisindicatesthattheopticallevermethod,whichiswidelyusedin
conventionalFFMs,canbeappliedtotheMMPbaseddualaxisFFM.Therefore,theMMPcan
be easily installed to conventional FFM or AFM setups with FFM options and can change
conventionalFFMtodualaxisFFM.
3.2.MechanicaldesignofMMP
Let us think about the necessary structure for the MMP [12]. In general, the needed force
resolution for the friction force and load is of the order of 1 nN. The typical displacement
resolutionoftheopticallevermethodisoftheorderof0.1nm.Therefore,arequiredspring
constantofthedoublecantileverandtorsionbeamisoftheorderof1N/m.Inconventional
FFMprobes,thetypicallateralandverticalspringconstantsofVshapedcantileverisaround
195
196
Torsion beam
for load detection
Movement of optical lever

pattern image
Optical lever
pattern image
Laser for
optical lever
C
Laser spot
Silicon
substrate
Four-segmented
PD
Optical lever pattern
Double cantilever
for friction force
detection
Tip
Friction force
Load
Sample
Figure 5. Schematic of the micro mechanical probe (MMP)-based FFM. The MMP consists of a double cantilever for
the friction force detection and a torsion beam for the load detection.
100and0.1N/m,respectively,andthetypicallateralandverticalspringconstantsofIshaped
cantileverisaround100and1N/m,respectively.Theseindicatethatthefrictionforcesensi
tivityistwoorthreeorderofmagnitudelowerthanthatoftheload.Thisisbecausethese
probesaredesignedforAFMandthemechanicalinterferenceisreducedbysacrificingthe
frictionsensitivityasdescribedabove.
Thespringconstantofthedoublecantileverbeaminthelateraldirectionisgivenby
kdl = 2E
bd h d3
ld3
(5)
whereEisYoungsmodulus.thespringconstantofthedoublecantileverbeaminthevertical
directionisgivenby
kdv =
3
1 bd h d
E 3
2
ld
(6)
Theeffectivespringconstantofthetorsionbeamintheverticaldirectionisgivenby
ktv = 2 f
Gbt h t3
lc2lt
(7)
wherethetorsionangleisconvertedtotheverticaldisplacementoftheprobeend.Thelateral
springconstantofthetorsionbeamktlisgivenby
ktl =
3
2 bt h t
E 2
3 l l
c t
(8)
ByusingEqs.(5)to(8),thespringconstantsareobtainedaskdl=4.3N/m,kdv=1.5x103N/m,
ktv=13.1N/m,ktl=4.9x102N/mwhenld=1200m,bd=185m,hd=5m,lt=370m,bt=70
m,andht=20m.Here,l,b,andhdenotethelength,width,andthickness,respectivelyand
the suffixes d and t mean the double cantilever and torsion beam, respectively. The pitch
betweenthedoublecantileverbeamsissettowh=400mandthetipheighttissetto10m.
Thelateralrigidityofthetorsionbeamktlis100timeslargerthanthatofthedoublecantilever
beamkdl,andtheverticalrigidityofthedoublecantileverbeamkdvis100timeslargerthanthat
ofthetorsionbeamktv.Theprobewiththesedimensionscanmeasurethelateralandvertical
forcesindependentlywithaforceresolutionof1nNwithoutthemechanicalinterference.
3.3.FabricationmethodofMMP
Here,thefabricationmethodoftheMMPisdescribed.Asdescribedintheabovesection,the
doublecantileveroftheMMPrequiresthestructurewithaveryhighaspectratioofaround
40:1.Intheaboveexample,acantileverwithawidthof185mandthethicknessof5m.The
structure at these small dimensions is difficult to fabricate for conventional machining
techniques. Moreover, the high aspect structure is not easy for even the micromachining
techniquesalthoughtheyaresuitableforfabricationofthestructureonthemicrometerscale.
Among the current micromachining techniques, anisotropic chemical etching and deep
reactiveionetching(DRIE)arecandidatesthatcanfabricatemicrostructurewithahighaspect
ratio.AlthoughtheDRIEcanprovideheightaspectmicrostructure,ittakeseffortstomakethe
fabricatedsurfaceflat.Incontrast,anisotropicetchingcanproduceflatsurfacemoreeasily
becauseacrystalplaneappearsbyetching.Intheanisotropicchemicaletchingtheshapesof
structuresthatcanbefabricatedarelimitedtothecrystalstructureofthesubstratewhereas
theyarenotintheDIE.Thecasewhereanisotropicetchingisselectedisexplainedbelow.
(a)
Slit shaped opening

[001]
Etching mask
(b)
Groove formed by etching

W
D
[010]
[010]
[001]
Silicon substrate
Figure 6. Anisotropic chemical etching through a slit-shaped opening.(a) A {100} silicon substrate. (b) Groove forma
tion by the etching.
197
198
Anisotropicchemicaletchingusesthatetchspeeddependsonthecrystalorientation.Figure
6explainshowtofabricateaverticallyorientedsurfaceforthedoublecantilever.Insingle
crystalsilicon,theetchspeedfor{100}planeishigherthanthoseofotherorientationplanes
whenaqueoussolutionofpotassiumhydrate(KOH)isusedasanetchant.Whena{100}silicon
waferisusedasasubstrate,{100}planeisparallel(e.g.(100)plane)ornormal(e.g.(010)plane)
tothewafersurface.ByKOHetchingthe{100}siliconwafer,(010)and(01 0)planesappearand
formthenormaltothewafersurfacewithacrystallographicaccuracy.First,anetchingmask
film,whichistypicallymadeofsiliconoxideorsiliconnitrideforKOHetching,isformedon
thewafer.Thewaferisdippedintheetchantaftertheslitshapedopeningisformedinthe
etchingmaskfilmasshowninFigure6.Thesiliconsubstrateinthisopeningisexposedtothe
etchantandetchedduringtheetchingwhereasthesiliconsubstratecoveredwiththeetching
maskfilmisprotectedfromtheetchant.Ifwechooseasiliconwaferwhosecrystalplaneisthe
[001]directionandorientationflatofthelowerendis 1 00 ,thedirectionoftheopeningshould
besettobeparalleltothecrystaldirectionof[100]or[010].InthecaseshowninFig.6,the
openingisalongthe[100]direction.Astheprogressoftheetching,thegrooveisformed,whose
bottom face is the (100) plane and the sidewalls are (010) and (01 0) planes. When the two
parallelslitshapedopeningsareformedintheetchingmask,averticallyorientedthinplate
is formed between the openings. By arranging the three openings so that two thin films
betweenthethreegroovesremainwhenthesubstrateisfullyetched,adoublecantileverwith
amicrometerorderthicknesscanbeformed.Thus,thethicknessofthedoublecantileveris
determinedbytheslitpitchandwidthbythethicknessofthesiliconwafer.Sincetypicalwafer
thicknessisaround100to500m,thedoublecantileverwithahighaspectratioofseveral
tenstooneispossible.
It should be noted that the dimensions of target devices are partially constrained in this
fabricationmethod.Theetchratesofthecrystalplaneinthebottomsurfaceandsidewallsare
equalbecausealltheplanesbelongtothe{100}planes.Inaddition,the(010)and(01 0)planes
areetchedinthelateraldirectionwhereasonlythe(100)planeisetchedintheverticaldirection
(Figure6).Therefore,thegrowthspeedofthegrooveinthelateraldirectionistwiceaslarge
asthatintheverticaldirection.Thatistosay,thewidthisalwaystwiceaslargeasdepthin
theetchedgroove.Inordertomakethedevicefreestanding,allthesiliconsubstratehastobe
etched.Inthiscase,theetchdepthisequaltothewaferthickness.Therefore,thewidthofthe
etchedgrooveislargerthantwicethewaferthickness.Thethicknessofthedoublecantilever
isdeterminedbythepitchoftheslitshapedopeningsandthewidthoftheetchedgroove,that
is,thewaferthickness.Whentheslitsarearrangedinthephotomask,theserelationsshould
be consideredcarefully. Inaddition, the pitchofthe cantilevers ofthe double cantileveris
largerthantwicethewaferthicknessduetotheabovereason.
AnexampleofthefabricationmethodoftheMMPisschematicallyinshowninFigure7[11].
Thedirectionsofthewaferplaneandorientationflataresettobe(001)and(1 00),respectively.
Consideringtherelationshipbetweenthewidthanddepthofthegroove,thewaferthickness
shouldbeselected,forexample,around200m.Thelargerwaferthicknesscausesthelarger
pitchofthecantilevers,whichleadstothelargerprobeheadthatconnectsthecantilever.This
decreasesthenaturalfrequencyoftheprobeandreducestherobustnessagainstenvironmental
(a)
(b)
A
Roofshaped
optical lever
pattern
A
Ridge for
torsion
beam
{100}
Silicon
wafer
Tip
(c)
(d)
Silicon
oxide
D
(010)
plane
(100)
plane
Slit-shaped
opening
Double
cantilever
beam
Torsion
D beam C
B
(010)
plane
Silicon
oxide
A
B
Tip
Figure 7. Fabrication process of the micro mechanical probe (MMP). (a) Formation of the optical lever pattern and
ridge for the torsion beam. (b) Formation of the tip on the backside of the silicon substrate. (c) Formation of etching
mask for the substrate etching. (d) Free standing probe after the substrate etching.
mechanical noises. Usually, the thickness of commercial silicon wafers has a variation of
around 10 m. Prior to the fabrication, the substrates should be uniformly etched so as to
reducethethicknessvariation.Thefabricationprocessesmainlyconsistsofthreeanisotropic
chemicaletching.TheetchantisanaqueoussolutionofKOHforalltheetchings.Siliconoxide
filmsforetchingmaskcanbeformedbythermaloxidationmethod.Inthefirstetching,the
torsionbeamandopticalleverpatternareformed(Figure7(a)).Theopticalleverpatternis
199
200
formedbyusingtheundercutetchingunderarectangleetchingmask.Inthesecondetching,
thetipattheprobeendisformed(Figure7(b)).Thetipisalsofabricatedbyundercutetching.
Inthethirdetching,thedoublecantileverbeamisformed(Figure7(c)).Throughaslitshaped
openingintheetchingmaskontherearsideofthewafer,thesubstratesiliconisetched.Finally,
the freestanding double cantilever beam is formed (Figure 7(d)). The electron microscope
imageofafabricatedprobeisshowninFigure8.Amagnifiedviewoftipandaroofshaped
opticalleverpatternarealsoshown.Inthisexample,thethickness,widthandlengthofthe
doublecantileverwereabout8m,167m,and1410m,respectively.Usingtheanisotropic
chemicaletching,theMMPwithahighaspectratiodoublecantilevercanbeformed.
Roof-shaped
optical lever pattern
Toision beam
10 m
Double
cantilever
300 m
Tip
10 m
Figure 8. Fabricated micro mechanical probe (MMP).
3.4.Forcecalibration
TheforcecalibrationisindispensableforquantitativeFFMmeasurements.IntheMMP,the
differentialsignalsfromthefoursegmentedPD,(IA+IB)(IC+ID)and(IA+IC)(IB+ID)havetobe
convertedtothelateraldeflectionofthedoublecantileverandtorsionangleofthetorsion
beamorverticaldisplacementoftheprobehead,respectively.Afterthesesensitivitycalibra
tion,thedeflectionandtorsionanglecanbeconvertedtothefrictionforceandload,respec
tively.
Thelattercanbeimplementedasisdoneforthecalibrationoftheverticaldeflectioninusual
AFMsorFFMs.Usingforcecurvemeasurement,theconversioncoefficientcanbeobtained
fromtherelationshipbetweenthepiezodisplacementanddifferentialsignal(IA+IB)(IC+ID).In
conventionalFFMs,asdescribedinSection2.2,thecalibrationofthefrictionforceisnoteasy
becausethetorsionangleofthemicrocantileverhastobemeasured.Incontrast,thecalibration
ofthefrictionforceintheMMPissimilartousualcalibrationmethodfortheverticaldeflection
inAFMs.ThatisisbecausethecalibrationofthefrictionforceintheMMPcanbeimplemented
by obtaining the relationship between the differential signal, (IA+IC)(IB+ID) and the lateral
deflectionofthedoublecantilever.
OnemethodofthefrictionforcecalibrationisshowninFigure9.Inthismethod,astepstructure
isusedasastandardsample.Forexample,thefractionofthesiliconwafer,whichisglued
ontotheflatsubstrate,canbeusedasthestepstructure.Thestepheightissettobehigherthan
thedoublecantileverwidth.Duringtheprobescanningonthesubstrate,theprobestopsat
thestepstructureasshowninFigure9(a).Incontrast,theprobebase,whichisattachedtothe
piezo scanner, continues to move. When the optical lever system of an FFM apparatus is
implementedsothatitmovestogetherwiththepiezoscannerorprobebase,theopticallever
patternappearstomoveintheinversedirectionalthoughthepatternstopsatthestepstructure.
ThiscausethelateralmovementofthedarkopticalleverpatternonthePDoftheopticallever
system as showin in Figure 9(b). Since the lateral movement of the pattern is equal to the
displacementoftheprobebasei.e.piezoscanner,therelationshipbetweenthesignal(IA+IC)
(IB+ID)andthelateraldeflectioncanbeobtained.
Figure 9. Calibration of friction force with a step-structure. (a) Arrangement of the micro mechanical probe and step
structure. (b) Typical relationship between the differential signal for probe lateral deflection and probe displacement.
201
202
Figure 10. Calibration of the friction force by using the probe sticking due to the static friction at the beginning of the
probe scan.
Intheabovemethod,thestepstructurethatspeciallypreparedisused.Iftheprobesticksto
thesubstrateatthebeginningoftheprobescanbythestaticfrictionalforce,thestepstructure
isnotnecessary.Inmanycases,atthebeginningofthescan,theproberemainsstillonthe
samplesubstrateduetostaticfrictionalthoughtheprobebaseorpiezoscannermoves.After
theelasticforceovercomesthestaticfriction,theprobestarttoslideonthesubstrateandthe
frictionmodechangesfromthestatictothekineticfriction.Theconversioncoefficientcanbe
obtainedfromtherelationshipbetweenthedifferentialsignalandpiezodisplacementbefore
theprobeslidingasshowninFig.10.Althoughthismethodiseasilyimplemented,theerror
ispossibletooccurwhenthesampledoesnotsticktheprobeenough.Therefore,thesample
forthecalibrationhastobecarefullyselected.
Inthepreviousreport[13],theconversioncoefficientofsignaltothedeflection,wherethe
MMPwasinstalledtothecommercialAFMsetup(NanoScope,Bruker),wasoftheorderof
severalmV/nm.Here,thiswasdeterminedbythestepstructuremethod.Fromthisresult,the
minimumdetectionlimitofthelateraldeflectionwasobtainedastheorderof0.1nmbecause
the typical noise level was around 1 mV. These results indicate that sub nN force can be
detectedwhenthelateralspringconstantofthedoublecantileverisoftheorderof1N/m,
3.5.ExampleofMMPbasedFFMmeasurement
Figure11showsanexampleoftheFFMmeasurement.Thesamplewasanmthicklubricant
film on a substrate. In this example, the lubricant was pepfluoropolyehter (PFPE) and the
substratewasamagneticdisk.Bothofthelubricantanddiskareusedinthelubricationofthe
HDDs.Inaddition,thelubricantfilmwaspatternedbytheultraviolet(UV)lighttreatment
[14].TheUVlightilluminationbondsthelubricantmoleculeontothedisksurfacebyphoto
chemicalreaction.Ifsomepartsofthelubricateddiskisirradiatedandwholediskisrinsed
bythesolvent,theUVirradiatedlubricantfilmremainsandtheunirradiatedlubricantsnot.
Bycontrollingtheirradiationareawithaphotomask,thepatternednmthicklubricantfilm
canbeobtained.Here,alineandspacepatternwhoselinewidthandpitchwereabout5m
and15m,respectively.Thelubricatedregionshowedthelowerfrictionwhereasthedisk
exposed region showed higher friction. In this measurement, the contact mode was used,
whichmeansthattheloadwasconstant.Therefore,thefrictionforcedistributionshownin
Figure11showsthedistributionofthefrictioncoefficientbecausethecoefficientisobtained
bydividingthefrictionforcebytheload.
25 m
Figure 11. Example measured by a micro mechanical probe. The sample was a nanometer thick lubricant film with
line-and-space-patterned by the ultra violet light radiation. The lower region is lubricant remained region and the
higher is substrate magnetic disk exposed region.
Thus,theMMPcanelevateFFMfromthequalitativevisualizationmethodtothequantitative
measurementmethod.Inaddition,theoptimumconditionsasaFFMprobecanbeselected
becausetheMMPbaseddualaxisFFMbecomefreefromthelimitationduetothereduction
ofthemechanicalinterference.Thiscanprovidehighsensitivemeasurementofbothfriction
force and load, which is difficult in the conventional probe. The MMP is expected to be a
standardprobeforFFMbyovercomingtheproblemoftheconventionalFFMprobes.
3.6.Futuredirections
Asdescribedabove,thesignificantproblemsofthedualaxisprobecanbesolvedbytheMMP
basedFFMandtheMMPcanprovidethefrictionpropertiesonthenanometerscale.Thisis
expectedtoopenupthenewfieldofFFMasaquantitativeevaluationmethod.Forfurther
advancements,thevariousslidingspeedsareexpected.Inthecurrentdualaxisprobes,the
slidingspeedislimitedtothescanspeed,typicallyaround1m/s,whichisdeterminedby
203
204
thescannerpiezo.Inmanyapplications,especiallymechanicalapplicationssuchasMEMS/
NEMSorHDDs,thefrictionpropertiesatahigherslidingspeedisrequired.
4.Summary
Inthischapter,thebasicsandlatestresearchresultsofFFM,whichisoneofthekeymeasure
mentmethodsformicro/nanomechatronics,isdescribed.FFMcanprovidethelocalfriction
propertiesonthemicro/nanoscale.FFMiswidelyusedintribologyandmaterialcharacteri
zationaschemicalforcemicroscopes.DualaxisprobesarenewtypesofFFMprobes,which
canovercomethesignificantdrawbacksoftheconventionalcantileverprobes,especially,the
reductionofthemechanicalinterferencebetweenthelateralfrictionforceandverticalload.
However,thedualaxisprobeshaveproblemstobesolvedsuchasthelateraldisplacement
detectionmethod.Themicromechanicalprobecanovercometheproblemsandispossibleto
elevate the FFM from a visualization tool to the quantitative evaluation method for local
frictionpropertiesonthemicro/nanoscale.Thisisexpectedtobeausefulmethodfordesigning
micro/nanomechatronicdevicessuchasMEMS/NEMSortheheaddisklubricationofHDDs
andimprovetheirperformancedrastically.
Acknowledgements
This work was supported in part by the Japanese Ministry of Education, Culture, Sports,
Science and Technology through Global Center of Excellence (GCOE) project, Grant No.
20360077,theNEDOGreenITProject,andtheStorageResearchConsortium.
Authordetails
KenjiFukuzawa
References
[1] BinnigG,RohrerH.Scanningtunnelingmicroscopecombinedwithascanningelec
tronmicroscope.ReviewofScientificInstruments1986;57(2)221224.
[2] BinnigG,RohrerH.Scanningtunnelingmicroscopy,Scanningtunnelingmicroscopy
frombirthtoadolescence.ReviewsofModernPhysics1987;59(3)615625.
[3] Binnig G, Quate C. F, Gerber C. Atomic Force Microscope, Physical Review Let
ters1986;56(9)930933.
[4] MateC.M,McCellandG.M,ErlandssonR,ChiangS.Atomicscalefrictionofatung
stentiponagraphitesurface.PhysicalReviewLetters1987;5919421945.
[5] MitsuyaY,OhshimaY,NonogakiT.Couplingandnonlineareffectsofcantileverde
flectionandtorsionencounteredwhensimultaneouslymeasuringverticalandlateral
forcesusingthescanningprobemethod.Wear1997;211198202.
[6] AmakawaH,FukzuawaK,ShikidaM,ZhangH,ItohS.DualAxisMicroMechanical
ProbeforHighlySensitiveFrictionForceMicroscope.In:ProceedingsofWorldTri
bologyCongress2009,WTC2009;2009.
[7] MunzM.Forcecalibrationinlateralforcemicroscopy:areviewoftheexperimental
methods.JournalofPhysicsD2010;43(6)063001.
[8] ChuiB.W,KennyT.W,MaminH.J,TerrisB.D,RugarD.Independentdetectionof
vertical andlateralforces with asidewallimplanteddualaxispeizoresistive cantile
ver.AppliedPhysicsLetters1998;72(11)13881390.
[9] AndoY,NagashimaT,KakutaK.UsingFIBprocessedAFMcantileverstodetemine
microtribologycharacteristics.TribologyLetters2004;91523;
[10] FukuzawaK,TeradaS,ShikidaM,AmakawaH,ZhangH,MitsuyaY.DualAxisMi
croMechanicalProbeforIndependentDetectionofLateralandVerticalForces.Ap
pliedPhysicsLetters2006;89173120.
[11] AmakawaH,FukzuawaK,ShikidaM,ZhangH,ItohS.ImprovedSensitivityofDu
alAxis MicroMechanical Probe for Friction Force Microscope. Tribology Online
2008;3(7)356360.
[12] Fukuzawa K, Terada S, Shikida M, Amakawa H, Zhang H, Mitsuya Y. Mechanical
Design and Force Calibration of DualAxis MicroMechanical Probe for Friction
ForceMicroscopy.JournalofAppliedPhysics2007;101(3)034308.
[13] Amakawa H, Fukzuawa K, Shikida M, Zhang H, Itoh S. Quantification of friction
force on dualaxis micromechanical probe for friction force microscopy. Tribology
Online2010;5(3)(2010),144149.
[14] ZhangH,MitsuyaY,FukuokaN,ImamuraM,FukuzawaK.Selforganizedpattern
ingofmolecularlythinliquidpolymerfilmsutilizingmolecularflowinducedbyul
travioletirradiation.AppliedPhysicsLetters2007;90123119.
205
Chapter 10
Tissue Damage and Repair Caused by

Immune System and Personalized
Therapy of Failed Organs by Stem Cells
Ken-ichi Isobe, Naomi Nishio,

Thanasegaran Suganya, Zhao Cheng and
Sachiko Ito
1.Introduction
Our immune system has developed to fight against invaders. Pathogenic invaders are
microorganismsincludingvirus,bacteria,fungusandamoeba.
Inspeciesdevelopment,innateimmunesystemappearedfirst,thenafterjowedfishacquired
immunesystemdeveloped.Oncemicroorganismsinfectsinorgans,cellsbelongingtoinnate
immunesystemsuchasneutrophilsandmacrophagesmovetothesiteofinfection.Microor
ganisms are sensed by patternrecognition receptors (PRRs) of the innate immune system
includingdendriticcells(DC).Thesecellsphagocytizeinfectiousagentsandproducecytokines
andchemokines,whichinducetissueinflammation.InphagolysozomeofDC,theproteins,
whichconstitutethemicroorganisms,aredegraded.TheseDCmigratestolymphnodeand
present the microorganisms antigens to T cells, which are central of adaptive immune
responses.
Similartotheeradicationofpathogens,theinflammatoryresponseisalsocrucialfortissue
and wound repair and called sterile inflammation. Hostderived nonmicrobial stimuli are
releasedfollowingtissueinjuryorcelldeath.Theseendogenousmoleculeshavebeentermed
damageassociatedmolecularpatterns(DAMPs).DAMPsarenormallypresentintracellular
andarethereforehiddenfromrecognitionbytheimmunesystem.Oncetissuesareinjured
these molecules are released into the extracellular environment by dying cells and trigger
inflammationundersterileconditions[1](Figure1).
2013 Isobe et al.; licensee InTech. This is an open access article distributed under the terms of the Creative
208
Microorganisms
External stimuli
damages
Internal stimuli
DAMPs
Macrophages
Dendritic cells
PRRs
Pro-inflammatory cytokines
Figure 1. Damaged cells by various stimuli produce DAMPs, which activate tissue resident macrophages and dendritic
cells. Activated innate cells produce pro-inflammatory cytokines.
2.Woundrepairandinnateandacquiredimmunity
2.1.Skinwoundhealing
Woundhealingisoneofthemostimportantsubjectsinmedicineandbiology.Forexample
accidentalinjuriesormedicalsurgeryinitiatewoundrepairprocesses.Fromthepracticaland
fundamentalimportancewoundhealinghasbeenthesubjectofintenseinvestigationformany
years[2].Recentlyincreaseofelderlypersonnecessitatestheresearchofwoundrepairfurther.
The ability to wound repair declines with age, which induces incurable pressure ulcer in
elderlypatients[3].Acuteinflammatoryreactionsplayintegralrolesinnormalwoundhealing
andtissuerepair.Onceparenchymaandbloodvesselaredamaged,thecoagulationsystemis
activated,whichreleasechemicalmediatorsthatpromotevascularpermeabilityandleukocyte
Tissue Damage and Repair Caused by Immune System and Personalized Therapy of Failed Organs by Stem Cells
adhesionandrecruitment.Within30min.neutrophilsmigratetothesiteofwound,which
phagocytizedeadcellsandcellulardebrisandcleartissue.Thenneutrophilsarethemselves
phagocytized by tissue macrophages. Although M1 macrophages damage tissue cells pro
ducingtissuedamagingproteases,whichareprimarilykillmicroorganisms,M1macrophages
alsoengulfdamagedtissuecells.ThenM2macrophagesmigratefrombloodvesseltorepair
tissue.Neutrophilshavespecificcellsurfaceantigenscalled,Gr1.Inmurinemodelofwound
repair(Figure2),wedepletedneutrophilsbyantiGR1antibody.Althoughinyoungmicethe
kineticsofwoundhealingwasnotdifferentbythedepletionofneutrophils,thedepletionof
neutrophilsbyantiGr1antibodydramaticallydelayedwoundhealinginagedmice[4].Bone
marrowhasbothmaturedneutrophilsandmacrophages.First,wefoundthatinsplenectom
izedmice,bonemarrow,spleenandthymusinjectionacceratedwoundhealing(Figure3).We
transplantedbonemarrowcellsfromGFPmice(C57BL/6background)toC57BL/6mice.We
foundthatbonemarrowcellsinjectionacceleratedwoundhealingandGFPpositiveneutro
philsandmacrophagesmigratedtothewoundtissue.Becausewoundhealingofagedmiceis
relativelyinefficient,wetransplantedyoungbonemarrowcellstoagedmice.Wefoundthat
youngbonemarrowcellsenhancedwoundhealingofagedmice[5,6].Incaseofmicroorgan
ismsacquiredimmunityhelptophagocytizeinvader.Bymicroorganismsinfections,immu
noglobulin specific for the microorganisms bind to the surface and induce phagocytosis
mediatedbyFcreceptors.WeexaminedwhetherBcells,whichproduceantibodiestodamaged
tissues,mightbeengagedintheprocessofwoundhealing.Wefoundthatwoundhealingin
splenectomizednudemicewasdelayedandthetransferofBcellsacceleratedwoundhealing
in splenectomized mice. Further we detected several autoantibodies binding to wounded
tissues[7](Fig.4).AdvancingagegraduallydecreasedthestrengthofIgG1autoantibodies,
whichbindtowoundedtissues,althoughthestrengthofIgMautoantibodieswasrelatively
stablebyadvancingage[5,6].
2.2.DSSinducedcolitis
Asamucosalwoundrepairmodelweuseddextransulfatesodium(DSS).Colitismayresult
from DSS toxicity to colonic epithelial cells. DSS is commonly used in rodent models to
chemicallyinduceacuteintestinalinflammation,andtheDSSinducedcolitisischaracterized
byweightloss,bloodydiarrhea,epithelialcelldamage,andimmunecellinfiltration,aswell
as an increased production of inflammatory mediators including TNF, IL6, IL12, and
interferons.ColitiswasinducedbyoraladministrationofDSStotwomonthsoldC57BL/6
miceat2%(w/v)indrinkingwateradlibitumforfivedaysfollowedbynormaldrinkingwater.
Bythisschedule,micealmostcompletelyrecovered.InterestinglyatrecoveryphaseofDSS
inducedcolitis,weobservedstrongupregulationofinnateimmunecellshavingGr1+CD11b
+
cell surface maker in spleen and bone marrow. Transplantation of splenic DSSderived
Gr1+CD11b+ cells into DSStreated mice improved colitis and promoted efficient colonic
mucosalhealing[8](Figure5).AntiGr1antibodytreatmentworsenedtheDSSadministered
colitis.TheseresultsindicatethatGr1+CD11b+cellsinducedbyDSSworkedtorepaircolon
woundhealingandrepaircolitis[9].
209
210
Experimental design
Skin wounds by 3mm

punch biopsy
Sample preparation
I.V. injection of immune cells

control
Take samples by 6mm

punch biopsy
FACS
analysis
Figure 2. Murine experimental model of wound healing. The dorsal skin was punched through with a sterile disposa
ble biopsy punch (diameter 3 mm). Separated immune cells were injected intravenously. The wounds and their sur
rounding areas were cut for analyses with biopsy punch with a diameter of 6 mm at each day after wounding.
Immune cell injection

~ immune cells ~
Control
24 h
48 h
96 h
PBS
12
BM
10
PBS
8
Spleen
6
Spleen
Bone
Marrow
Thymus
2
Thymus
0h
24h
48h
96h
Figure 3. Examples of analysis. After 3 weeks of the removal of spleen, bone marrow, spleen or thymus cells were
injected at the same time of punch biopsy.
< Immunoprecipitation >

Sham Splenectomy
250 KDa
226KDa Myosin, heavy polypeptide 9, non-muscle isoform 1
150 KDa
164KDa Carbamoyl-phosphate synthase [ammonia],

mitochondrial precursor
100 KDa
75 KDa
50 KDa
42KDa Actin, alpha cardiac muscle
37 KDa
25 KDa
B6 mice (2.5M)
Figure 4. Immunoprecipitation of autoantibodies, which were bound to tissue cells. After separation each band was
analyzed by LC/MS.
Control
DSS only
Day 11 after
DSS
DSS+ GR-1+CD11+ transplantation
Day 5 after
DSS treatment
Figure 5. Colitis was induced by oral administration of DSSat 2 % (w/v) in drinking water ad libitum for five days fol
lowed by normal drinking water. Upper show the FACS analysis of spleen of DSS-treated mice. Increased GR-1+CD11b
+ cells were transplanted to mice of at the last day of DSS administration.
211
212
3.Sterileinflammationtotissuedegeneration
3.1.Overview
Neutrophilsandmacrophagesproduceproinflammatorycytokines,includinginterleukins1
alphaandbeta(IL1and)andtumornecrosisfactoralpha(TNF),chemokines,matrix
proteases. These materials help to destroy invading bacteria and virus. However, invaded
tissuealsobedamagedbytheseinflammatoryreactions.Oncemicroorganismsareclearedby
immunesystem,theinvadedtissueisrepairedtohealthystate.However,somemicroorgan
ismssuchashepatitisCviruskeepstayinginhosttissueandrepeatedlyattackhosttissue,
whichinducesfataltissuedamagesfinally.Insterileinflammationsimilarphenomenaoccur.
Externalstimuli,whichdestroytissuecells,produceDAMPsandinducesterileinflammation.
Becauseimmunecellssecretetoxicproteasestokillmicroorganisms,thesameimmunecells
destroytissuecells.Ifexternalstimuliarelimited,repairsystemsuchasM2macrophageswork
torepairtissuedamagesandreturnstonormalcondition.However,repeatedstimulimay
inducenextinflammationbeforetissuerepairandwillinducechronictissuedamages,which
cannotberecovered.Intheseprocesses,autoreactiveTcellsandBcellsareengaged.
3.2.Bleomycin(BLM)inducedmurinemodelofexperimentalsystemicsclerosis
BLMisaglycopeptideproducedbythebacteriumStreptomycesverticillus,whichcancleave
DNA,anditiswidelyusedasanantitumoragentforvarioustypesofmalignancies[10].Thus
BLMadministrationdestroystissuecellsandinducessterileinflammation.Weexaminedthe
effectsofrepeatedinjectionofBLMtoadultmice.WhenBLMwasinjectedintotheshaved
backsofadultC3HorBALB/cmice(100g/mouse)5daysperweekfor3weeks,notonlyskin
fibrosis,butalsoesophagealandgastricdamagerelatedtofibrosiswereobserved.Injection
ofBLMinducedinnateimmuneinflammation,whichdidnotresolve.Dendriticcells,which
engulfeddeadcells,transferredautoantigenstoTcells.TransferofCD4+TcellsfromBLM
treated BALB/c mice induced the same pathological changes and antibody production in
untreatedBALB/cnudemice.Th17cellssecretIL17,whichstimulatedinnateimmunecellsto
damagetissues[11].
3.3.Amyloidbetapeptide(A)worksasaDAMP
AaccumulationisthoughttobecentraltothepathogenesisofAlzheimersdisease(AD)[12].
A is produced by proteolysis processing of Amyloid precursor protein (APP). A balance
between amyloidogenic processing of APP and the removal of soluble A by clearance
pathwaysandenzymemediateddegradationmaintainsAlevels.Aisnowconsideredto
beoneofDAMPs,whichactivateinnateimmunecellsespeciallymicroglia.Putativesensors
ofAareNLRP3,CD36andRAGE[1315].Byusingmicroglialcellsorcellline,wehaveshown
thatAproduceseveralcytokinesincludingMCSFbyPI3K/AKTandNFBpathway,which
stimulatemicroglialproliferationandmigration[16].FurtherAinducesseveralchemokines
(CCL7,CCL2,CCL3,CCL4andCXCL2)byPI3K/AKTandErkpathwayinthemicroglia,which
inducethemigrationofmicrogliaormacrophages[17].Ainducedsterileinflammationalso
produced matrix metalloproteinase (MMP3, MMP12 and MMP13), which might degrade
damagedtissuefurtherormightworktodegradeAtoprotectagainstdangerousstimuli[18].
Ainducedmicroglialmigrationmayinducescytoskeletalsystem.WehaveshownthatA
bindstoIQGAP1andactin,whicharecytoskeletalcomponents,andmaystimulateRho/Rac
signaling[19].
3.4.Dgalactose(Dgal)inducedthymusdegeneration
Dgalathighlevelsinducestheproductionofreactiveoxygenspecies(ROS)andadvanced
glycationendproducts(AGEs;[20])andinducessterileinflammation.Weadministered50
mgDgalfor60daysto2monthsoldmice.Wefoundunorganizeddistributionsofkeratin5
andkeratin8proteinsinthethymusofthesehosts,whichresembledtoagedmice[21].
4.Towardpersonalizedcelltherapy
4.1.Overview
Ourbodyhasthecapacitytoregeneratefrominjury.However,damagedtissuesinduced
bycontinuousmicrobialorsterilestimulicannotberecovered.Continuousstimulationby
TGF produced by macrophages induces fibroblasts to secrete matrix proteins such as
fibrinogen. Damaged tissue cells are replaced by fibrous component. Many patients are
suffering from these conditions. About 90% of all deaths from chronic obstructive lung
diseases (COPD) are attributable to cigarette smoking. Long term Cigarette smoking
induces continuous sterile inflammation. Patients who have low blood oxygen levels in
theirbloodaregivensupplementaloxygen.However,oxygensupplycannotresumelung
function.Longtermheavyalcoholicstimuliinducecontinuoussterileinflammationinliver
andeventuallydevelopcirrhosis.Adultonsetdiabetesmellitusisnowfoundtobecaused
by repeated sterile inflammation, which is derived from obesity. Renal failure has
numerouscauses.Themostcommonisdiabetesmellitus.Lung,liver,andkidneyfailure
hasnotreatmenttocurediseasesexcepttransplantation.Transplantationtothesediseas
eshaslimitationfromorganshortageandimmunerejectioncausedbyHLAdiscrepancy.
Another common degenerative diseases are Alzheimer and Parkinson disease. These
diseasesalsoarecausedbycontinuoussterileinflammation.Becausethesediseasesaffect
singleorgan,wemayusehealthyothertissuecellsforregenerativemedicine.Severaltrials
arenowongoing[22].
4.2.MuscleregenerationfrompluripotentEmbryonicStem(ES)cells
Agedperson,whohaslongbeeninbed,hasseriousproblemofmuscleatrophy.Wesetup
murinemodelofmuscleregeneration.EScellshavingLacZmarkerwereculturedonTypeIV
collagen dishes with 10% FCS. After 5 days of culture PDGFR+ mesodermal progenitor
populationwassortedandtransplantedtoinjuredmuscleofKSNnudemice.Weobserved
LacZpositiveCD34+Pax7+cells(satellitecells)intransplantedmuscle.Becauseoftransplant
213
214
edcellsaresatellitecells;theymayautonomouslyproliferateinvivo[23].Thenwetriedto
differentiateskeletalmusclecellswithoutserum.WeculturedEScellswithBMP4without
serum.BythisconditionEScellsdifferentiatedintoosteochondrogeniccellsinvitroandin
vivo.EarlyremovalofBMP4followedbylithiumchloride(LiCl)promotedthedifferentiation
tomyogenicprogenitorcellsandfinallydifferentiatedtoskeletalmusclecellsinvitro[24].
4.3.Thymicepithelialcellsfrominducedpluripotentstemcells
One of the most serious problems of aged person is infection caused by ageassociated
immunefunctionaldecline.Theageassociatedatrophyofthethymusandthedeclineof
naveTcellsoutputaremainlycausedbydegenerationofthymicepithelialcells(TEC).
WetriedtodifferentiateiPSCstoTEC.Thethymusinitiallydevelopsasanendodermal
epithelialcellatembryonicday10encapsulatedbymesenchyme.Wefollowthisembryon
icdevelopmenttoinduceTECfromiPSCs.First,wetrytodifferentiateiPSCstomesendo
dermtodefinitiveendoderm.WeaddedactivinAandLithiumchloride(LiCl)toiPSCsin
collagen IV coated dishes. We found that endodermal epithelial cell cluster were sur
roundedbymesodermcellssameasembryonicdevelopment.ByaddingFgF8thenFgF7
and FgF10, the population of endodermal epithelial cells enlarged and matured to TEC
progenitor.BytheadditionofRANKligandTECprogenitordifferentiatedtomedullary
TEClikecells.
E9.5
E11
d pharyngeal
h
l pouch
Third
iPSCs
Fgf7,Fgf10
Definitive endoderm
0HVHQGRGHUP
'
$FWLYLQ $QJPO
/L&O
P0
'
'
)*) QJPO
/L&OP0
'
)*)QJPO
)*)QJPO
%03QJPO
/L&OP0
7(&
'
)*)QJPO
)*)QJPO
%03QJPO
5$1./QJPO
/L&OP0
Figure 6. Upper figure shows the fetal development of TEC. Lower figure shows the protocol of TEC differentiation
from iPSCs.
4.4.EstablishmentofinducedPluripotentStemCells(iPSCs)fromagedmice
Thegenerationofinducedpluripotentstemcells(iPSCs)frommurineandhumansomatic
cellsopensthe possibility ofpersonalized celltherapies fortreatinghumandiseaseand/or
repairing the damaged tissues of elderly patients [25]. By the lifespan extension many
developedcountrieshaslargenumberofelderlypatients,whosufferfromincurablediseases.
Forpersonalizedcelltherapyofelderlyperson,technologytoestablishiPSCsfromagedperson
isnecessary.FurtheriPSCsfromagedpersonneedtodifferentiatetotargettissuecells.Inorder
toestablishmodelsystemtotreatelderlypatientbyiPSCs,wedesignedmurinemodel.We
foundthatefficiencytoestablishiPSCsfromagedmicewaslow.However,wecouldsucceed
toestablish2clonesofiPSCsfrombonemarrowderiveddendriticcellsof21montholdC57BL/
6micecarryingGFPmarkerbyretrovirusencodingfourfactors(Oct3/4,Sox2,Klf4andcMyc).
EstablishediPSCshavepluripotentmakerssuchasSSEA1,Oct4andPou5f1andalkaline
phosphataseactivitywaspositive.OurAgediPSCsmadeteratomacontainedtissuesofthree
germlayers,whentheyweretransplantedtothedorsalflankofC57BL/6(syngeneic)mice.
Theyalsodifferentiatetovarioustissuecellsincludingmuscle,liverandneuronbyhanging
dropfollowedbytwodimensionalculture[26].
ByusingiPSCswetriedtodifferentiateintocardiovascularcellsforthetreatmentofischemic
tissue, which might be caused by atherosclerosis. We simply cultured iPSCs in Type IV
collagendishandseparatedFlk1+cellsbymagneticactivatedcellsorting(MACS)orFACS
sorting. Flk1+ cells from iPSCs cocultured with HUVECs on Matrigel made the network
structures.Flk1+cellsweretransplantedadductormusclesintheischemiclimbofnudemice.
Revascularizationoftheischemichindlimbwasacceleratedinmicethatweretransplanted
withFlk1+cells[27].
Aged C57BL/6 with GFP

iPSCs
Chimera
formation
Figure 7. Establishment of aged-iPSC from GFP positive C57BL/6 mice. Chimera mice production from these iPSCs.
215
216
5.Conclusionsandperspective
Humanhealthdependsonimmunesystem.Herewepresentthedatabyusingmurinemodels,
whichindicatethatimmunecellsreacttoselfantigensreleasedfromdamagedcellsbyexternal
stresses.Oncemicrobesareclearedbyimmunesystem,invadedtissueisregeneratedbytissue
stemcells.Tissuerecoveryalsooccursinsingleorshorttermstimuli.However,continued
infection or stimuli induces permanent tissue damage. Aging is one of continued stimuli
caused by ROS produced from mitochondria. Recent progress of regenerative medicine
especiallyiPStechnologywillenabletoregeneratefailedorgansbyusingpatientowntissue
cells.Toconstructpatientspecific3DstructuredorgansbydifferentiatediPSCswiththehelp
ofcomputerprogramwillopennewpersonalizedtherapy.
Acknowledgements
Wethanktothefollowingresearchersworkinginourdepartment;MohammadNizamUddin,
RongZhang,HirohikoSuzuki,YutaInami,TohruYoshikai,HidetoshiSakuraiandHideaki
Ishikawa.WethankMinoruTanaka,IkuyoMizuguchiandYushikazuFujitaintheDivision
for Medical Research Engineering, Nagoya University Graduate School of Medicine for
assistancewithflowcytometry,confocalmicroscopyandelectronmicroscopyrespectively.
WealsothenktothesecretaryworksofNanaeOiwa.ThisworkwassupportedbytheGCOE
program and research grant provided by the Ministry of Education, Science, Technology,
SportsandCulture,Japan.
Authordetails
KenichiIsobe,NaomiNishio,ThanasegaranSuganya,ZhaoChengandSachikoIto
DepartmentofImmunology,GraduateSchoolofMedicine,NagoyaUniversity,Nagoya,Ja
pan
References
[1] ChenG.Y,NuezG.Sterileinflammation:sensingandreactingtodamage.Nature
ReviewsImmunology2010;10(12)82637.
[2] Sonnemann K. J, Bement W. M. Wound repair: toward understanding and integra
tion of singlecell and multicellular wound responses. Annual Review of Cell and
DevelopmentalBiology2011;27237263
[3] Swift M. E, Burns A. L, Gray K. L, DiPietro L. A. Agerelated alterations in the in

flammatory response to dermal injury. Journal of Investigative Dermatology
2001;117(5)10271035.
[4] NishioN,OkawaY,SakuraiH,IsobeK.Neutrophildepletiondelayswoundrepair
inagedmice.Age(Dordr).2008;30(1)1119.
[5] NishioN,ItoS,OkawaY,IsobeK.Thewoundrepairiscontrolbymonocytelinage
cells.ProceedingsofInternationalSymposiumonMicroNanoMechatronicsandHu
manScience,MHS2010,710November2010;Nagoya,Japan;2010.
[6] NishioN,ItoS,SuzukiH,OkawaY,IsobeK.EffectofAgingonAutoAntibodiesto
WoundedTissues.TheOpenGeriatricMedicineJournal2010;31724.
[7] NishioN,ItoS,SuzukiH,IsobeK.Antibodiestowoundedtissueenhancecutaneous
woundhealing.Immunology2009;128(3)369380.
[8] ZhangR,ItoS,NishioN,ChengZ,SuzukiH,IsobeK.UpregulationofGr1+CD11b+
population in spleen of dextran sulfate sodium administered mice works to repair
colitis.InflammAllergyDrugTargets.2011;10(1)3946.
[9] Zhang R, Ito S, Nishio N, Cheng Z, Suzuki H, Isobe K. I. Dextran sulphate sodium
increasessplenicGr1(+)CD11b(+)cellswhichacceleraterecoveryfromcolitisfollow
ing intravenous transplantation. Clinical & Experimental Immunology 2011;164(3)
417427.
[10] Povirk L. F, Austin M. J. Genotoxicity of bleomycin. Mutation Research 1991;257(2)
127143.
[11] IshikawaH,TakedaK,OkamotoA,MatsuoS,IsobeK.Inductionofautoimmunityin
ableomycininducedmurinemodelofexperimentalsystemicsclerosis:animportant
roleforCD4+Tcells.JournalofInvestigativeDermatology2009;129(7)16881695.
[12] HardyJ.Amyloid,thepresenilinsandAlzheimersdisease.TrendsinNeurosciences
1997;20(4)154159.
[13] YanS.D,ChenX,FuJ,ChenM,ZhuH,RoherA,SlatteryT,ZhaoL,NagashimaM,
Morser J, Migheli A, Nawroth P, Stern D, Schmidt A. M. RAGE and amyloidbeta
peptideneurotoxicityinAlzheimersdisease.Nature1996;382(6593)685691.
[14] HalleA,HornungV,PetzoldG.C,StewartC.R,MonksB.G,ReinheckelT,Fitzger
aldK.A,LatzE,MooreK.J,GolenbockD.T.TheNALP3inflammasomeisinvolved
in the innate immune response to amyloidbeta. Nature Immunology 2008;9(8)
857865.
[15] StewartC.R,StuartL.M,WilkinsonK,vanGilsJ.M,DengJ,HalleA,RaynerK.J,
Boyer L, Zhong R, Frazier W. A, LacyHulbert A, El Khoury J, Golenbock D. T,
MooreK.J.CD36ligandspromotesterileinflammationthroughassemblyofaToll
likereceptor4and6heterodimer,NatureImmunology2010;11(2)155161.
217
218
[16] Ito S, Sawada M, Haneda M, Fujii S, OhHashi K, Kiuchi K, Takahashi M, Isobe K.

Amyloidbetapeptidesinducecellproliferationandmacrophagecolonystimulating
factor expression via the PI3kinase/Akt pathway in cultured Ra2 microglial cells.
FEBSLetters2005;579(9)19952000.
[17] ItoS,SawadaM,HanedaM,IshidaY,IsobeK.Amyloidbetapeptidesinduceseveral
chemokine mRNA expressions in the primary microglia and Ra2 cell line via the
PI3K/Aktand/orERKpathway.NeuroscienceResearch2006;56(3)294299.
[18] ItoS,KimuraK,HanedaM,IshidaY,SawadaM,IsobeK.Inductionofmatrixmetal
loproteinases(MMP3,MMP12andMMP13)expressioninthemicrogliabyamyloid
beta stimulation via the PI3K/Akt pathway. Experimental Gerontology 2007;42(6):
532537.
[19] ItoS,NishioN,IsobeK.AnalysisofAmyloidPeptideBindingProteinsinMicroglial
Cells.TheOpenGeriatricMedicineJournal2012;516.
[20] LuJ,ZhengY.L,WuD.M,LuoL,SunD.X,ShanQ.Ursolicacidamelioratescogni
tiondeficitsandattenuatesoxidativedamageinthebrainofsenescentmiceinduced
byDgalactose.BiochemicalPharmacology2007;74(7):10781090.
[21] UddinM.N,NishioN,ItoS,SuzukiH,IsobeK.ToxiceffectsofDgalactoseonthy
musandspleenthatresembleaging.JournalofImmunotoxicology2010;7(3)165173.
[22] Isobe K, Cheung H. S, Wu J. Toward Personalized Cell Therapies by Using Stem
Cells.JournalofBiomedicineandBiotechnology2012;2012172564.
[23] SakuraiH,OkawaY,InamiY,NishioN,IsobeK.Paraxialmesodermalprogenitors
derivedfrommouseembryonicstemcellscontributetomuscleregenerationviadif
ferentiationintomusclesatellitecells.StemCells.2008;26(7):18651873.
[24] Sakurai H, Inami Y, Tamamura Y, Yoshikai T, SeharaFujisawa A, Isobe K. Bidirec
tional induction toward paraxial mesodermal derivatives from mouse ES cells in
chemicallydefinedmedium.StemCellResearch2009;3(23)157169.
[25] TakahashiK,YamanakaS.Inductionofpluripotentstemcellsfrommouseembryon
icandadultfibroblastculturesbydefinedfactors.Cell2006;126(4):663676.
[26] ChengZ,ItoS,NishioN,XiaoH,ZhangR,SuzukiH,OkawaY,MuroharaT,Isobe
K.Establishmentofinducedpluripotentstemcellsfromagedmiceusingbonemar
rowderivedmyeloidcells.JournalofMolecularCellBiology2011;3(2):9198.
[27] SuzukiH,ShibataR,KitoT,IshiiM,LiP,YoshikaiT,NishioN,ItoS,NumaguchiY,
YamashitaJ.K,MuroharaT,IsobeK.Therapeuticangiogenesisbytransplantationof
induced pluripotent stem cellderived Flk1 positive cells. BMC Cell Biology
2010;11(72).
[28] TokuhashiY,KikkawaF,TamakoshiK,SuganumaN,KuzuyaK,AriiY,KawaiM,
HattoriS,KobayashiI,FuruhashiY,NakashimaN,TomodaY.Arandomizedtrialof
cisplatin, vinblastine, and bleomycin versus cyclophosphamide, aclacinomycin, and

cisplatininepithelialovariancancer.Oncology1997;54(4)281286.
[29] InamiY,YoshikaiT,ItoS,NishioN,SuzukiH,SakuraiH,IsobeK.Differentiationof
inducedpluripotentstemcellstothymicepithelialcellsbyphenotype.Immunology
&CellBiology2011;89(2):314321.
219
Chapter 11
Tribology for Biological and Medical Applications

Noritsugu Umehara, Takayuki Tokoroyama and
Hiroyuki Kousaka
1.Introduction
Tribologyisthescienceandtechnologyoftwointeractingsurfacesinrelativemotionandof
relatedsubjectsandpractices.Thefieldoftribologyisinterdisciplinary.Untilrecently,ithas
beendominatedbymechanicalandchemicalengineerswhoconductedmacroteststopredict
frictionandwearinmachinecomponentsanddevelopednewlubricants.Afterthedevelop
ment of biological engineering and medical engineering, tribology for various interacting
surfaces of human, creatures, cells and so on has keen attention in biotechnology. Also
interaction of surfaces between medical devices and biocells as a blood is becoming to be
interested.Onthebasisontheoldtribology,wecannotsolveissuesforbiomedicalapplica
tions,becausebioandmedicalsurfacesarebasicallysoftandchangedtheirmechanicaland
chemicalpropertieswithtemperaturedrastically.Soweshouldproposeanddevelopnovel
technologyforbiomedicalapplications.
In this chapter, we introduce several new technologies to reduce adhesion of thermally
coagulatedblood[1,2]andfishproteins[3]tometalsurfaceforaradioknifeandgrillrods
respectively. Also ion beam irradiation increased the bonding strength of a probe DNA to
polymersubstrateforaDNPsequencetipinlowprice[4].Andinordertoreducetheadhesion
and friction of rubber, plasma and UV irradiation methods were proposed for the rubber
plunger of disposal plastic syringe [5] and rubber plug of medicine bottle [6] respectively.
Thoseproposedsurfacetechnologiesforbiologicalandmedicalapplicationswillbeexplained
inthefollowingsections.
2.Reductionofadhesionofcoagulatedbloodonatipofradioknife
Radioknifeisacommonoperationtoolforsurgerytoavoidintensebleeding.However,there
is severe trouble of severe adhesion of coagulated blood on the tip of radio knife during
operation.Thenonconductivecoagulatedbloodstopselectriccurrentandtheradioknifecan
2013 Umehara et al.; licensee InTech. This is an open access article distributed under the terms of the
222
notwork.Inordertoavoidthisissue,nurseshavetoremovethecoagulatedbloodfromthe
tipoftheradioknifefrequentlywithabrasiveclothes.
Inordertoovercomethisissue,weconfirmedtheadhesionmechanismofbloodduringtheusage
ofaradioknifeandproposedtwotypesofnewtipsforradioknifeasvibrationcooledtypeand
watercooledtype.Watercooledtypeshowedgoodresults[1,2].Figure1showstheexperimen
talsetuptoobtainsurfacetemperatureofradioknifebythermocamera.Highfrequencyelectric
voltagewasappliedfromelectricpowersourceasshowninFigure1aftertheradioknifetipwas
dippedintobovineblood.Duringoperationwithelectricpower,temperaturedistribution
aroundthetipwasmeasuredwithinfraredthermocamera.Alsoaftertheoperation,theadhesion
forceofcoagulatedbloodonthetipwasmeasuredwithsscratchtest.Atfirstweobservedthe
coagulatedbloodonthetipwithvibrationbyabimorphtypedpiezoelectricactuator.Figure2
showstheobservationresultsofcoagulatedbloodonthetipwithoutandwithvibrationafter
test.Itcanbeseenthatamountofcoagulationofblooddecreasedwithincreasingofampli
tude.Aftertheobservation,adhesionforcewasmeasuredforeachcoagulatedbloodonthetip
byascratchtestofapinstylus.Themaximumscratchingforcewasdefinedasanadhesionforce.
Figure3showstheeffectofamplitudeofvibrationontheadhesionforce.Itcanbeseenfrom
Figure3thattheadhesionforcedecreasedwiththeincreasingofvibrationamplitudegradual
ly.Whentheamplitudewasreachedto19m,theadhesionforcewithvibrationwas1/3ofthat
withoutvibration.Ontheviewpointoftemperaturesurroundingaradiotipwithelectriccurrent,
thevibrationdecreasedtemperatureofsurfaceofblood.Wesupposedthatthetemperature
reductionistheessencetoreducethecoagulationofblood.Thereforewemadeaprototypeof
atipwithwatercooling.Figure4showstheobservationresultsofcoagulatedbloodonthetip
withoutandwithwatercoolingaftertest.Waterwascirculatedinsidearadioknifetipunder
variousflowrateofcoolingwater.Itcanbeseenthatamountofcoagulationofbloodde
creasedwithtemperature.Figure5showstherelationshipbetweenadhesionforceofcoagulat
edbloodandflowrateofwater.Itcanbeseenthatwatercoolingwith120ml/minofflowrate
provided1/4ofadhesionforcecomparingtotheconventionaltip.Onthebasisofthewell
achievementofproposedtip,surfacetemperatureshouldbeessentialtocontroltheadhesionof
coagulatedblood.
PIEZO Driver
Function generator
(Amplifier)
Specimen
holder
Vibrator
Radio knife power source
Specimen
Stainless
steel case
(High-frequency power source)
Blood
Z stage
Figure 1. Schematic diagram of experimental setup for blood coagulation to a radio knife tip
(b)
(a)
5 mm
5 mm
(d)
(c)
5 mm
5 mm
Figure 2. Observation results of coagulated blood on the tip without and with vibration (a) No vibration as conven
tional, (b) vibration amplitude:7 m, (c) vibration amplitude:14 m and (d) vibration amplitude:19 m
Adhesion force F , N
5
Conventional
Vibrated
4
3
Frequency : 6.8 kHz
Amplitude: 7 - 19 m
Power : 50W
Cutting time : 20 s
Sample : bovine blood
Specimen : SUS304
2
1
0
0
5
10
15
Vibration amplitude a , mm
20
Figure 3. The relationship between vibration amplitude and adhesion force of conventional and vibrated tips
223
(b)
(a)
5 mm
5 mm
(d)
(c)
5 mm
5 mm
Figure 4. Observation results of coagulated blood on the tip without and with water-cooling (a) No water-cooling as
conventional, (b) flow rate: 0 ml/min, (c) flow rate: 60 ml/min and (d) flow rate: 120 ml/min
5
Conventional
224
Cooled
3
2
Cooling temp. : 23
Power : 50 W
Cutting time : 20 s
Specimen : SUS304
1
0
0
20
40
60
80
100
Flow rate Q , ml/min
120
140
Figure 5. The relationship between flow rate and adhesion force of conventional and water cooled tips
T0
T1
Heated specimen
Radio knife tip (Conventional)
ional)
Radio knife tip (Vibrated)
d)
Radio knife tip (Cooled)
Cooling temp : 23
Flow rate: 0 ~ 120 ml / min
Vibration frequency : 6.8 kHz
Vibration amplitude : 7 ~ 19 Pm
Power : 50W
Cutting time : 20 s
0
30
40
50
60
70
80
Temperature T ,
90
100
110
Figure 6. Relationship between adhesion force of coagulate blood and surface temparature on the tip of convention
al, vibrated and water-cooled tips.
Thereforeweobtainedtherelationshipbetweensurfacetemperatureoftiponadhesionforce
asshowninFigure6.Itcanbeseenfromthisfigurethatadhesionforceofbloodwasconstant
under65oC,roseupandreachedtothelargevalueifthetemperatureofbloodwasmorethan
80oCduringtemperaturerising.Itwasclarifiedthatthistemperaturedependenceofadhesion
forceofbloodwassimilartothetemperaturedependenceofblooddeteriorationattempera
turerising.Alsoitwasfoundthatwecanrealizehightemperatureatthetissuesurfaceeven
ifthetemperatureofatipislowerthatseveradhesiontemperaturewithbothproposedtips.
3.ReductionofadhesionbetweensteelandgrilledfishproteinwithUltra
HydrophobicDLC
Whenyougrillfish,youprobablyhaveexperiencesthatskinandmeatofthefishhasadhered
tosomemetalgrids.Thisissueisstillunsolvedontheadhesionofthedenaturedfishbodyby
heating.ConventionalgridsofgrillarecoatedwithFluorineresintoavoidthesevereadhe
sionofgrilledfishtothegrids.Ifthemaximumtemperatureofgrillingroomislessthat300oC,
thecoatedgridwithFluorineresincanworkwell.Howeverwhenthetemperatureinthegrilling
roomrisesupto500oC,conventionalgrillrodswithFluorineresincoatingcannotworkwell,
becauseFluorineresinshoulddisappearbyheatdecomposition.Therecentgrillingequip
mentheatsbothfromtopsideandbottomside,andextraheatingisappliedtothegrillingrods
thatriseitupto500oCevenifthetemperatureoncontactpartbetweengridandfishisatabout
150oC.
225
226
Thereforewehavetriedtodevelopnewgrillsthathaveweakadhesiontogrilledfishand
temperatureenduranceathightemperature.WeconsiderthatthereasonwhyFluorineresin
coatingcanreducetheadhesionistheFluorineresinhaslowsurfaceenergyandhydopho
bicity.Ifwecanintroducethenewcoatingthathasultrahydrophobicityandtemperature
enduranceupto500oC,itcanworkwell.SowefocusedonUltraHydrophobicDiamondlike
CarboncoatingnamedasUHDLCcoating[7].Ifitcankeeptheultrahydrophobicproperties
uptothe500oC,ithasthepossibilitytouseforthenewgrillrods.
Inthisstudy,weintroducedUHDLCcoatingtogrillrodsandevaluatethepossibilityofits
usage.InordertoknowthetemperatureenduranceofUHDLCcoatings,weevaluatedthe
effect of annealing temperature on the morphology and contact angle of water. Also we
evaluatedtheeffectofUHDLCcoatingontheadhesionofrealfishaftergrilling[3].
2 m
Figure 7. FE-SEM image of the needle-like structure of UH-DLC surface
Grill
rod
22.5
3
10
(b)
(a)
Jig for grill rods
Jig for grill rods
Figure 8. (a); The schematic of 2.45 GHz surface wave-excited plasma (SWP) apparatus, (b); Jig for grill rods
123 oC
2100 oC, 9minutes
3200 oC, 9 minutes
4300 oC, 9 minutes
5400 oC, 9 minutes
6500 oC, 9 minutes
Figure 9. The status of UH-DLC film after heating at each temperature for 9 minutes
UltraHydrophobicDiamondlikeCarbonwerecoatedbysurfacewaveexcitedplasmaCVD
thatusedArgasandTMS(TetraMethylsilane,[Si(CH3)4])gasasprecursorgasesonstainless
steelblockandgrillingrods.Figure7showsthecrosssectionalSEMimageoftheUHDLC
coatingsofonspecimen.
HereweexplainhowtocoattheUHDLCfilm.Figure8showstheschematicofthePECVD
equipmentwithastainlesssteelchamberthatis400mmininnerdiameterand300mmin
height,presentlyemployedforsomegrillrods.Figure8(b)showsajigwhichkeepgrillrods
paralleltothequartztube.Wesetitontheinsulatorandattachedsomegrillrodsonitsjigs
3.0mmhole.Andincaseofblockspecimens,weonlyplacethemonthejig.Priortodeposition,
gaspressureinthechamberwasdecreasedtodegreesof104Pawithturbomolecularpump.
Thebottomendofthechamberwasconnectedtoacoaxialwaveguidethroughwhich2.45
GHzmicrowaveswereintroducedviaaquartzplatecoupledwithaquartztubethatis20mm
indiameterand2mminthickness.
2.45GHz microwaves introduced into the chamber propagate as surface wave along the
interface between the surface of quartz tube and plasmas, generating highdensity plasma
columnalongthetubeasshowninFigure8.Asprecursorgases,CH4andTetraMethylsilane
(TMS,Si(CH3)4)areintroducedintothechambertogetherwithArgas.
For ultrahydrophobic structure of DLC coating on the sample surface, an azimuthally
symmetric2.45GHzsurfacewaveexcitedplasma(SWP)wasproposed.Whenhighworking
pressureat220PaandTMSgasmixture,ithasaprofoundeffectonnanoscalepostgeneration
becauseoflargeTMSparticleisflowinnerchamber.Andalsoetchingwithglowdischargeof
Argaswasanothereffectofroughsurface.
Atfirst,weobservedtheblockspecimencoatedUltraHydrophobicDiamondlikeCarbonby
SEM.AndthenweheatedtheflatspecimencoatedUHDLCinanelectricfurnacefor9minutes
227
228
at100 oC.Afterthat,weobservedUHDLCcoatingbySEM.Next,weheatedthisspecimen
whichwasheatedat100oCintheelectricfurnacefor1hourat200oC.Thesameasdescribed
above,wealsoheatedat300oC,400oC,and500oC.AndthenobserveditsspecimenbySEM
afterheatingateachtemperature.Figure9showssixSEMpicturesoftheUHDLCcoating
afterheatingintheelectricfurnaceeveryfor9minutesateachtemperature.Wecouldconfirm
that surface of the UHDLC coating looked a little rough as the heating with the rise of
temperature,butthiscoatingdidnotdisappearandshowbigchange.
Pulling up
(a)
(b)
Figure 10. (a) The exclusive jig for pulling fish, (b) The schematic of pulling a fish
InordertoknowthereductioneffectofUHDLCandFluorinecoatingsonadhesionofgrilled
fishagainstgrillingrods,theadhesionwasevaluatedfromthemaximumpillingforceofthe
fishaftergrillingwithgasgrillequipment.
WeusedanexclusivejigforpullingafishasshowninFigure10(a).Weusedahorsemackerel
asstandardfish,becauseahorsemackerelisoneofthestandardfishforthetest.Weputan
exclusive jig and then a horse mackerel on the center of grids of grill as shown by Figure
10(b),andputtheminthegrillingequipment.Weusedonlythreegrillrodsforthisexperiment
asshowninFigure10(b).Weheatthemfor9minutes.Whileweheatthem,wecheckedthe
temperatureswithtwothermocouplesinthegrillequipment.Certainly,weconfirmedthatthe
temperatureinthegrillequipmentwasabout500oCandthecontacttemperaturebetweenfish
andthegrillrodwas150 oC.Afterthat,wetakethemoutfromthegrillingroom,andleave
themfor1minuteinair.Andthenwepullthefishupwithapullinggaugeandmeasurethe
maximumadhesionforce.
Wedidtheadhesiontestofarealfishinthegrillingequipmentfordifferentrodsasnocoating,
UHDLC coating and Fluorine resin coating. Figure 11 shows the adhesion result of this
experiment. These results were the average adhesion force of four time experiments. The
adhesionforcesofUHDLCcoatingandFluorineresincoatinggrillrodswereasabouthalfas
thatofnocoatinggrills.AlsoitwasconfirmedthattheadhesionforceoftheUHDLCcoating
grillrodswasasmuchasthoseoftheFluorineresincoatinggrillrods.
Inthissection,weproposednewgrillrodsthatwerecoatedwithUltraHydrophobicDia
mondlikeCarbon.TheadhesionofgrilledfishtoUltraHydrophobicDLCcoatinggrillrods
wereashalfasthatofnocoatinggrillrodsandasmuchasthatofFluorineresincoatinggrill
rods.Moreover,UltraHydrophobicDiamondlikeCarbonhasheatenduranceat500 oCof
grillingroom.Ontheotherhand,thehydrophobicpropertyofFluorineresincoatingdisap
pearedat500 oCofthegrillingroom.So,webelievethatUltraHydrophobicDiamondlike
Carboncoatinghasapossibilityfortheusageasnewgrillinggridsathightemperature.
Fish; horse mackerel

Cooking time; 9 minutes
Position; Center of grill
In a grilling equipment
Adhesion force F, N
3.5
3
2.5
2
1.5
1
0.5
0
No coating
UH-DLC Fluorine resin

Grill specimen
Figure 11. Experimental result of grilling fish in the gas grill equipment
4.IncreasingofbondingstrengthbetweenprobeDNAandpolymer
substrateforDNAchipwithionbeamirradiation
MappingofallDNAsequenceofhumanhadfinishedbyTheInternationalHumanGenome
Projectin2000.Thusdifferenceofgeneofeachhumancanbemeasured.Ifdistinctivesequence
in DNA sequence of human having disease caused by gene can be found, observation of
mechanismofsuchdiseaseanddevelopmentofnewmedicinemovedaheadwell.DNAchip
isanapparatustoexamineDNA,onwhichDNAisfixed.ExistingDNAchipisforresearch
anddevelopment,andexpensive.So,forclinic,inexpensive,morereliableandsensitive
DNAchipisneededasshowninFigure12.
ThereforeDNAchipmadefromaplasticwassuggested[8].Forhighreliabilityandsensitivity,
affinitybetweenDNAandplasticmustbehigher.Butplasticishydrophobic,affinitytoDNA
229
230
isnotwell.Therefore,itisnecessarytoimproveaffinitybetweenDNAandplasticbymodi
fyingsurfaceofplastichydrophilicityusingsurfacemodification.Inthisstudy,inorderto
achieve DNA chip made from a plastic, it is purpose to develop technology to improve
adhesionforceofDNA[4].Inparticular,influenceofionspeciesandiondoseonwettability
andsurfaceroughnessofPC(polycarbonate)includedTiO2ationirradiationisrevealed.And
in order to keep hydrophilicity after irradiation, influence of atmosphere in storage and
irradiationtimesonkeepingtimeofhydrophilicityisrevealed.Inaddition,influenceofion
irradiationonadhesionofDNAisrevealed.
Atfirst,PCincludedTiO2wassetinECRionirradiationapparatus,anditwasexhaustedto
5.01033.0104Painvacuumchamber.Thenionizedgasinplasmasourcewasaccelerated,
irradiatedonthesurfaceofspecimen.Inthisstudy,Ar,N2andO2ionswereirradiatedonthe
surfaceofspecimencontrollingiondose(cm2).
Beforeionirradiation,0.25ldistillatedwaterwasdroppedonthesurfaceofspecimen,contact
angledegwasmeasuredbyopticalmicroscope.Contactanglewasaveragevalueof7times
measurementperaspecimen.Andthespecimenafterirradiationwasstoredinairorwater,
contactanglewasmeasuredeveryseveralhours.
Then, in order to know influence of ion irradiation on wettability of the surface of plastic
substrate,specimenswereanalyzedbySEM,AFM,andXPS.
InordertoknowinfluenceofionirradiationonadhesionforceofDNA,itwasmeasuredby
forcecurvemeasurementusingAFMinwater[9].0.05MDNAsolution(SSCbuffered)was
droppedonthesubstrate,andDNAwasfixedbyUVcrosslinkerafterdrying.DNAwas400
baselength,ofwhichmolecularlengthwas290nm.Afterfixing,thesubstrateheldinwater,
AFM force curve measurement was conducted. AFM cantilever was made from silicon, of
whichconstantofspringwas0.03nN/m,andpeakradiuswasabout10nm.Andcontactforce
betweencantileverandsubstratewasabout5nN.
Figure 13 shows relationship between ion dose D and contact angle of water on the PC
includedTiO2.thevalueofwasabout75degatuntreated,andthenrapidlydecreased,finally
becamelessthan8degatmorethan41019cm2.FromanalysisbyAFMandXPS,itisseemed
thatincreaseofTiO2,whichishydrophilicity,onsurfaceofspecimenandincreaseofsurface
roughnessleadtodecreaseofcontactangle.
DNA adhesionarea
Figure 12. The proposed DNA chip made from plastic substrate
Contact angle , deg
90
Specimen : PC included TiO2
Ion species : Ar, N2, O2
Acceleration voltage : 600 V
Microwave power : 60W
MeanSDn=7
Ar
N2
O2
60
30
0
0
12
19
15
-2
Ion dose D , x10 cm
Figure 13. Relationship between ion dose and contact angle of water on the PC included TiO2.
Contact angle , deg
120
O2 x 1
O2 x 2
Untreated
90
60
Specimen : PC included TiO2
Ion species : N2
Acceleration voltage : 600 V
Microwave power : 60W
Liquid sample : Water
MeanSDn=7
30
0
0
480
960
1440
1920
2400
Time , hour
Figure 14. Relationship between contact angle and time under air after several times irradiation of O2 ion, Number
is ion irradiation times.
231
InordertotreatDNAchipsubstrateeasily,itisnecessarythathydrophilicityofsubstrateis
keepforalongtimeinair.Figure14showsrelationshipbetweencontactangleandtimeunder
airafterseveraltimesirradiationofO2ion.Hydrophilicityofsubstratewaskeptforabout200
hours at 1 time irradiation of O2 ion, but for 1440 hours at 2 times irradiation. Ikada et al.
reportedthathydrophilicgroupsofsurfaceofpolymermodifiedbyplasmaturnaround(hide
under)astimepasses,thereforehydrophilicitydescend[10,11].Inthisstudy,similarphe
nomenonoccurred,anditisseemedthathydrophilicgroupsbecamedifficulttoturnaround
byrepeatedirradiation.
Figure15showsrelationshipbetweenadhesionforceofDNAprobetothesurfaceofspecimen
andseveraltimesirradiationofions.AdhesionforceofDNAwasabout0.09nNatuntreated,
andthenincreasedbyrepeatedirradiationofN2ionfromabout0.13nNat1timeirradiation
toabout0.62nN,about7timesaslargeasatuntreated,at4timesirradiation.Incontrast,
adhesionforcedidnotincreaseatirradiationofO2ion.FromanalysisbyXPS,itisseemedthat
electrostaticbondbetweenaminogrouponsurfaceofspecimenformedbyionirradiationand
phosphoricacidinDNAleadtoincreaseofadhesionforceofDNA.
In conclusion, in order to development of DNA chip made from plastic, ion beam was
irradiatedtoPCincludeTiO2andcontactangleofspecimenwasmeasured.Theresultsclearly
showedthatcontactangledecreasedwithincreasingiondose.
Hydrophilicityofsurfaceofspecimenformedbyionirradiationcouldbekeptforalongtime
byrepeatedirradiation.
InordertoknowinfluenceofionirradiationonadhesionforceofDNA,itwasmeasuredby
forcecurvemeasurementusingAFMinwater.AdhesionforceofDNAincreasedbyrepeated
ionirradiationofN2,butdidnotbyionirradiationofO2ion.
Adhesion force, nN
232
Adhesion force of substrate

Adhesion force of DNA
Specimen: PC induced TiO2

Ion species: N2, O2
Acceleration voltage: 600 eV
Microwave power: 60 W
Number of irradiation: 1~4 times
MeanSD n=57
DNA solution: 0.05 M
0
Untreated
N2 1
N2 2
N2 3
N2 4
O2 1
O2 2
Specimens
Figure 15. Relationship between adhesion force of DNA probe to the surface of specimen and several times irradia
tion of ions.
5.Reductionoffrictionbetweenthermoplasticelastomersandplasticswith
photochemicalfluorination
Inmedicalfield,plasticandglasssyringesarewidelyusedtoinsertmedicinesintohuman
bodiesdirectly.Fromahygienestandpoint,theyaredisposedaftersingleuse.Generally,glass
syringesareinferiorintheaccuracyofdimensionandtheproducecostandtheyalsorequire
greatcareatthetimeofdisposal.Thus,replacingglasssyringeswithplasticonesisdesired.
Plasticsyringesaregenerallyusedwithsiliconeoillubricatingtheslidingareabetweenbarrels
andgaskets,wherethebarrelsaretypicallymadeofplasticssuchasPP(Polypropylene)and
thegasketsareusuallymadeofeithervulcanizedrubberorTPE(Thermoplasticelastomer)as
showninFigure16.Siliconeoilisbiologicallyandchemicallyinertbutconsideredtohave
somedemerits:possibilityofaccumulationinhumanbodiesanddecreaseofefficacybecause
of adsorption of medicines constituent [12]. These demerits are especially pronounced in
prefilledtypesyringes.Therefore,thedevelopmentofunlubricatedplasticsyringesisdesired
formedicaluse.Inthisstudy,inordertodecreasethefrictionforcebetweenbarrelsandgaskets
underunlubricatedcondition,wetriedtofluorinatethesurfacesofPPandTPEspecimensby
using PFPE (Perfluoropolyether) and VUV (vacuum ultraviolet) irradiation with excimer
lamp.ThismethodhasbeenalreadytriedtoPP[13].
First,wedroppedPFPEonaspecimenandputafusedsilicaglassonittomakethinandflat
PFPE layer. Then we irradiated them with VUV to make specimens surfaces react photo
chemicallywithPFPEasshowninFigure17.Aftertheirradiation,wecleanedthespecimen
withHFE(Hydrofluoroether)byusingultrasoniccleanertoremoveresidualPFPE.Theeffect
ofthephotochemicaltreatmentwasevaluatedbyfrictiontest,measurementofsurfacefree
energies, and FTIR (Fourier transform infrared) analysis where ATR (Attenuated total
reflection)methodwasadopted.Inthefrictionmeasurements,treatedPPwasslidagainstnon
treated TPE. And treated TPE was slid against nontreated PP. Surface free energies of a
specimenwerecalculatedfromthemeasuredcontactanglesofwaterandCH2I2dropletson
thespecimen.
Intheexperimentalresults,itwasconfirmedthatthefrictioncoefficientbetweentreatedTPE
andnontreatedPPwasdecreasedbyupto77%asshowninFigure18.Moreover,CFpeak,
which indicates fluorination of surfaces of specimens, was detected in FTIR spectra. And
surfacefreeenergiesofthemdecreased.ItissuggestedthatthesurfacesofbothPPandTPE
were fluorinated and fluorination of TPE was predominantly effective for decreasing the
friction coefficient between PP and TPE. We expected that the photochemical fluorination
changedonlychemicalproperty.Thus,tovalidatepossibilityofotherchanges,weinvestigated
thechangeinrealcontactareasbetweenfluorinatedTPEspecimensandaglassplate(BK7)
with contact microscope. Contrary to our expectation, decreases in real contact area of
fluorinatedTPEwereobservedwithdecreasingfrictioncoefficient,indicatingthechangein
surfacemechanicalpropertiesofthefluorinatedTPEspecimens.Accordingly,itisindicated
that the decreasing of friction coefficient between fluorinated TPE and nontreated PP is
attributedtodecreasingofrealcontactareaandadhesionarisenfromthereductionofsurface
freeenergy.
233
Barrel (PP etc.)
Plunger (PP etc.)

PP Polypropylene
TPE Thermoplastic
elastomer
Gasket (TPE etc.)
Silicone oil
Figure 16. Schematic view of plastic syringe
In vacuum (0.1 Pa)
UV 172 nm
Rotary pump
Fused silica glass

PFPE
Specimen
Figure 17. Schematic of photochemical treating process with UV and PFPE.in vacuum
6.0
Friction coefficient
234
Normal load W: 1 N , Sliding speed V: 2 mm/s

Wavelength of UV: 172 nm
Dropped PFPE & Cleaned by HFE
TPE vs. Non-treated PP
in air (30-50 %RH)
5.0
4.0
3.0
2.0
1.0
0.0
0
20
40
60
Irradiation time to TPE t , min
Figure 18. Relation between irradiation time to TPE and friction coefficient.
80
100
6.ReductionofadhesionofCIIRrubbertosteelplatewithplasma
irradiation
Adhesion,orthestickingofdifferentmaterialsattheirinterface,isofgeneralinterestinmany
branchesoftechnology,includingmicroelectronicdevices,medicalproductsandmanufac
turing.Adhesionbetweenrubbersandmetalsisoftenthemainsourceoftroubleinamachine.
Thus,ifmoldedrubberproductseasilysticktomolds,rollers,andpickuphandsmadeof
metals,theproductivityoftheirmanufacturinglinebecomeslow.Therefore,efficientutiliza
tionofrubbersheetsdemandsmodificationsonsomedesirablepropertiesoftherubbersurface
without affecting the bulk characteristics. Plasma treatment is one of the most employed
methodstoattainthisgoal.Oneofthemostsignificantbenefitsoftheplasmaprocessisitoffers
and additional advantage that the surface modification does not affect the desirable bulk
propertiesoftherubber.
Inapreviousstudy,wehavedemonstratedthatthesurfacewaveexcitedplasmatreatment
reduced the adhesion force between a medical rubber, chlorideisobuteneisoprene rubber
(CIIR)andstainlesssteelball(SUS440C)byusingoxygenandargongases.
Wehavealsoshownadecreaseintherealcontactareawithincreasingtimeandmicrowave
power,andasimilartrendintheresidualratesoftheadhesionforceandtherealcontactarea
ofCIIRrubber.Therefore,itisassumedthattheadhesionforceisstronglysubjectedtothereal
contactarea[14].However,themainreasonforthereductionsintherealcontactarearemains
unknown.Recentworkshaveshownthatplasmatreatmentincreasestherougheningofrubber
surfaces[15].Thesurfaceroughnessmayaffecttheadhesionforce,whichislargelydependent
onthecontactgeometryandsurfacetopography[16].
Theobjectiveofthisresearch,wereportonourattemptstoclarifythefactortoreducethe
adhesionforceduringthesurfacewaveexcitedplasmatreatmentprocess.Itisalsoattempted
tofigureouttheYoungsmodulusbehaviorinmicroscaletomeasurewithoutbulkproperty
by using micro slicer and surface roughness changes are measured by 3D laser scanning
microscope.
ResultsofadhesionforcesbetweenCIIRrubberandstainlesssteelballasafunctionofplasma
treatmenttimeat200WareshowninFigure19.Itisapparentfromthefiguresthattheadhesion
force dramatically decreased with oxygen plasma treatment according to treatment time.
Similardecreasingtrendwasalsoobservedwithargonplasmatreatment.However,at1min
treatment time, the adhesion force was higher with argon treatment than oxygen plasma
treatmentofCIIRrubber.After1min,argonplasmatreatmentwasmoreeffectivethanoxygen
plasmatreatmentindecreasingtheadhesionforce.Theadhesionforcecouldntbemeasured
after10minbecauseitwaslowerthanthemeasurablevalueof0.001N.Insummary,thefigures
showedthatplasmatreatmenttimeisaveryimportantfactorthatdecreasestheadhesionforce.
Loadpenetrationdepthcurvesbyusingthenanoindenterwith50Nmaximumloadswere
obtained.ThethicknessofthepreparedCIIRrubberwasabout3mmforthegeneralthickness
andabout50mforthecuttingthicknessbyusingthemicrowaveslicer.Despitethesame
conditions,thepenetrationdepthobtainedfromthe3mmthicknessCIIRrubberwasclearly
235
236
different from that obtained from the 50 m thickness. In particular, the unloading curve
coincided with the loading curve at a penetration depth of more than 290 nm due to the
thicknessdifference.Thisindicatesthatnotonlythebulkpropertyhasgreatlyaffectedthe
Youngsmodulus,butalsoitisaffecteditsaccuratemeasurement.Inotherwords,asurface
waveexcitedplasmatreatmentincreasedtheYoungsmodulusinmthicknessscale.Asa
result,wedeterminedthatitispossibletodevelopacleardifferencebetween3mmand50
mYoungsmodulusoftheCIIRrubberwithoutanyinfluencefromthebulkpropertyby
usingthemicrowaveslicer.
Figure 19. Adhesion force between CIIR sheet and stainless-steel ball after oxygen and argon plasma treatments at a
microwave power of 200 W and a gas pressure of 30 Pa, for 0, 1, 5, 10, and 15 minutes.
Figure 20. Effect of plasma treatment time on the elastic modulus of CIIR rubber 50 m in thickness which was cut
from the top surface of treated CIIR sheet. The elastic modulus was measured by nano-indenter for oxygen (black) and
argon (red) plasma treated CIIR sheets.
Figure20showstheYoungsmodulusprofileof50mthicknessCIIRrubbermeasuredby
nanoindenter after oxygen and argon plasma treatment with increasing time. The results
treatmenttime(206.1MPa),andfollowedbyasteadystate.AhigherYoungsmodulus(236.4
MPa)wasobtainedwithargonplasmatreatment,butonlyafter15mintreatmentwithargon
gas.Asaresult,thisimprovedYoungsmodulusbytheoxygenandargonaredepictedwith
squares (oxygen plasma treatment) and circles (argon). For the oxygen plasma treatment,
Youngsmoduluswasslightlyhigher(39.8MPa)thantheuntreatedCIIRrubber(38.0MPa).
However,itincreasedsignificantlybetween5min(53.8MPa)and10minsurfacewaveexcited
plasmatreatmentisanimportantfactorthatreducestheadhesionforce.
Figure 21. Surface roughness changes of CIIR sheet as a function of plasma treatment time. The measurements were
done for oxygen (dotted line) and argon (solid line) plasma-treated CIIR sheets. The surface roughness parameters, Rz
(maximum peak height roughness), Rq (Root mean square roughness) are defined by JIS B 0601 2001.
Figure 21 shows surface roughness as a function of plasma treatment time. The surface
roughnessdeviationsofCIIRrubberbyoxygenandargonplasmatreatmenthaschanged.The
entiresurfaceroughnessfactorsincreasedwithincreasingtreatmenttime.
Especially,theRzofargonplasmatreatedCIIRrubber(25.78m)isrougherthanthatofoxygen
plasmatreatedrubber(12.88m).Andtheseresultsimplythatchangeinmorphologydueto
surfaceroughnessreducedtherealcontactareaagainsttheSUS440Cball.
Figure22shows3DlaserscanningmicroscopephotographsofargonplasmatreatedtoCIIR
rubber.Intheabsenceofargonplasmatreatmentat1min,thesubsurfaceoftheCIIRrubber
lookslessgranularandgenerallyhasasmoothershape.However,with5minargonplasma
treatmentat200W,changesinthesurfacewerevisible.Nevertheless,thesurfaceofCIIRrubber
patternwaschangedcomparedtotheuntreatedCIIRrubber.ThesubsurfaceoftheCIIRrubber
wasgrowingrougherwithincreasingtreatmenttime(Figure4c).Inthisstudy,the200W,15
treatment conditions resulted in the roughest surface. As a result, this surface roughness
237
238
changebyetchingeffectmighthaveaffectedtheadhesionforce,whichislargelydependent
onthecontactgeometryandsurfacetopography[16].
Figure 22. laser scanning microscope images of CIIR sheets after argon plasma treatment with 200 W at a gas pres
sure of 30 Pa for treatment times of (a) 1, (b) 5, (c) 10, and (d)15 min.
Inthiswork,wetriedtoclarifythechangeinsurfacemechanicalpropertiesofCIIRsheetafter
the plasma treatments. In order to evaluate the change in Youngs modulus of CIIR sheet
surface,thetop50mthicknessofaplasmatreatedCIIRsheetwascutawaytoavoidthebulk
property.TheYoungsmodulusmeasurementswithnanoindentershowedthecleardiffer
ence between the surface and bulk elastic modulus of CIIR rubber after plasma treatment,
indicatingthesuccessofsurfacemodificationwithoutchangingbulkproperty.Inaddition,it
wasshownthattheplasmatreatmentwithArgasincreasedtheYoungsmodulusofCIIRsheet
surfacefrom38MPato236.4MPa.Andalso,surfaceroughnessofCIIRrubberhaschanged
to rougher with both oxygen and argon gas plasma treatments. These changes in Youngs
modulusandroughnessatthesurfaceofCIIRsheetareconsideredtobethemainreasonsfor
theplasmaassistedreductionofadhesionforcebetweenstainlesssteelball(SUS440C,JIS)
andCIIRsheet.
Authordetails
NoritsuguUmehara,TakayukiTokoroyamaandHiroyukiKousaka
DepartmentofMechanicalScienceandEngineering,NagoyaUniversity,Japan
References
[1] Nakashima Y, Tokoroyama T, Umehara N, Akagami Y. Reduction of Adhesion of
CoagulationtoaRadioKnifeTipwithFilledMicroGrooves.TransactionoftheJSME
C2011;77(779)29162921.
[2] Nakashima Y, Tokoroyama T, Umehara N, Akagami Y. Development of a Radio
Knife Suppression the Adhesion of Coagulated Blood. Transaction of the JSME C
2012;78(786)605614.
[3] Honda N, Kajiya M, Jang YJ, Kousaka H, Umehara H. Reduction of adhesion be
tween steel and grilled fish protein with UltraHydrophobic DLC. Journal of Ad
vancedMechanicalDesign,Systems,andManufacturing2010;4(1)356364.
[4] Demizu Y, Umehara N, Muraoka M, Imai J, Akagami Y, Kanda S. Improvement of
adhering strength of DNA for polycarbonate substrate including Titaniumu Oxide
withionbeamirradiation.TransactionoftheJSMEC2010;76(768)21422147.
[5] UsamiK,KousakaH,UmeharaN,TokoroyamaT.Frictionreductionbetweenther
moplasticelastomersandplasticsunderunlubricatedcondition.Transsactionofthe
JSMEC2010;76(767)18331837.
[6] Kim J. H, Umehara N, Kousaka H, Shimada M, Hasegawa M. Effect of microscale
Youngs modulus and surface roughness on adhesion property to plasmatreated
rubbersurface.JournalofMechanicalScienceandTechnology2010;24119122.
[7] JangYJ,UmeharaN.WettingandAdhesionBehaviorsofaC:HFilmDepositedon
NanoScaleCopperDotedSurfaces.TribologyOnline2008;3(5)294297.
[8] Yoshino M, Matsumura T, Umehara N, Akagami Y, Aravindan S, Ohno T. Wear
2006;274260.
[9] KatoM,MuraokaM,AkagamiY.In:PrceedingofJSMEM&M2005MechanicsofMa
terialsConference;2005.
[10] IkadaY,MatsunagaT,SuzukiM.JournaloftheChemicalSocietyofJapan1985;1079
6.
[11] McCarthy T. J. In: Proceedings ACS Organic coatings and applied polymer science
proceedings1983;48520.
239
240
[12] YagoK,KuroyamaM,OtoriK,HirayamaT,OgawaY,AotoH.AStudyonAdsorp
tionofCalcitoninPreparationonSyringes.JournaloftheNipponHospitalPharma
cistsAssociation2000;273279.
[13] Sato Y, Anai H, Murahara M. Functional Substitution on Polypropylene Film for
Glaucoma Implant by Using Excimer Lamp. In: Proceedings of the School of Infor
mationTechnologyandElectronics2003,TokaiUniversity;2003.
[14] Kim J. H, Nitta I, Umehara N, Kousaka H, Shimada M, Hasegawa H. Relationship
between Real Contact Area and Adhesion Force of PlasmaTreated Rubber Sheets
AgainstStainlessSteelBall.TribologyOnline2008;4(1)361365.
[15] Grythe K. F, Hansen F. K. Surface Modification of EPDM Rubber by Plasma Treat
ment.Langmuir2006;22(14)61096124.
[16] BurnhamN.A,KulikA.J.HandbookofMicro/Nanotribology,2ndedition.BocaRa
ton,FL,USA:CRCPress;1998.
Chapter 12
Micro-Nano Materials Characterization and Inspection
Yang Ju
1.Introduction
The individual surface atoms of flat samples could be made visible in real space until the
introductionoftheScanningTunnelingMicroscope(STM)in1981byBinnig,Rohrer,Gerber,
andWeibel[1].Thispowerfulinstrumenthasprovidedabreakthroughinourpossibilitiesto
investigatematterontheatomicscale.Withinoneyearofitsinvention,theSTMhashelped
tosolveoneofthemostintriguingproblemsinsurfacescience:thestructureoftheSisurface.
Becauseoftheirfabulouscontribution,G.BinnigandH.RohrerwererewardedwiththeNobel
Prize in physics in 1986. A huge number of conductors and semiconductors have been
investigated on the atomic scale and marvelous images of this world of atoms have been
created within the first few years after the inception of the STM. Today, the STM is an
invaluableassetinthesurfacescientiststoolbox.
DespitethegreatsuccessoftheSTM,ithasaseriouslimitation.TheSTMrequireselectrical
conductionofthesamplematerial,becausetheSTMneedsthetunnelingcurrentwhichis
flowingbetweenapincontactwithorverynearingthesample.Thus,theSTMcanonly
imageelectricallyconductivesamples,whichlimitsitsapplicationtoimagingmetalsand
semiconductors.Butevenconductorsexceptforafewspecialmaterials,likehighlyoriented
pyrolytic graphite can not be studied in ambient conditions by STM but have to be
investigatedinanultrahighvacuum(UHV).Inambientconditions,thesurfacelayerof
solidsconstantlychangesbyadsorptionanddesorptionofatomsandmolecules.UHVis
required for clean and well dened surfaces. Taking the above condition into account,
Binnigspeculatedtheatomicforcebetweenthetipandsample,theAtomicForceMicro
scope (AFM) [2, 3] was invented by him in 1986. Because electrical conductivity of the
sampleisnotrequiredinAFM,theAFMcanimagevirtuallyanysolidsurfacewithout
theneedforsurfacepreparation.Consequently,thousandsofAFMsareinuseinuniversi
ties,publicandindustrialresearchlaboratoriesallovertheworld.
2013 Ju; licensee InTech. This is an open access article distributed under the terms of the Creative
242
1.1.Principleofatomicforcemicroscope
The AFM consists of a cantilever with a sharp probetip at its end that is used to scan the
specimensurface(seeFigure1).Thecantileveristypicallysiliconorsiliconnitridewithatip
radiusofcurvatureontheorderofnanometers.Whenthetipisbroughtintoproximityofa
samplesurface,forcesbetweenthetipandthesampleleadtoadeflectionofthecantilever
according to Hookes law. Depending on the situation, forces that are measured in AFM
includemechanicalcontactforce,vanderWaalsforces,capillaryforces,chemicalbonding,
electrostaticforces,magneticforces,etc.Alongwithforce,additionalquantitiesmaysimulta
neouslybemeasuredthroughtheuseofspecializedtypesofprobe.Thedeflectionismeasured
usingalaserspotreflectedfromthetopsurfaceofthecantileverintoanarrayofphotodiodes.
Figure 1. Atomic Force Microscope.
1.2.Microwavetechniqueformaterialscharacterization
Themicrowavemethodsformaterialscharacterizationgenerallyfallintoresonantmethods
and nonresonant methods. Resonant methods are used to get knowledge of dielectric
propertiesatsinglefrequencyorseveraldiscretefrequencies,whilenonresonantmethodsare
oftenusedtogetageneralknowledgeofelectromagneticpropertiesoverafrequencyrange.
Bymodifyingthegeneralknowledgeofelectricalpropertiesoveracertainfrequencyrange
obtainedfromnonresonantmethodswiththeaccurateknowledgeofelectricalpropertiesat
severaldiscretefrequenciesobtainedfromresonantmethods,accurateknowledgeofmaterials
propertiesoverafrequencyrangecanbeobtained.
1.2.1.Reflectionmethod
Inareflectionmethod,thepropertiesofasampleareobtainedfromthereflectionduetothe
impedancediscontinuitycausedbythepresenceofthesampleinatransmissionstructure.The
reflectionmethodisatypeofnonresonantmethod.Fromtheviewoftransmissionline,ina
reflectionmethod,thesampleundertestisintroducedintoacertainpositionofatransmission
line,andsotheimpedanceloadingtothetransmissionlineischanged.Thepropertiesofthe
sample are derived from the reflection due to the impedance discontinuity caused by the
sampleloading.
Inareflectionmethod(seeFigure2),themeasurementfixturemadefromatransmissionline
isusuallycalledmeasurementprobeorsensor.Inordertoincreasethemeasurementaccuracy
andsensitivity,ortosatisfyspecialmeasurementrequirements,themeasurementprobesare
oftenspeciallydesigned.
Incident
wave
Sample
Reflected
wave
Transmitted
wave
Figure 2. Boundary condition for material characterization using a non-resonant method.
1.2.2.Nearfieldscanningprobe
Inthereflectionmethods,thereisaspecialmethodnamednearfieldscanningprobeshould
beintroduced.Scanningtechniquesforlocalcharacterizationofconductingandinsulating
filmsareattractingmuchinterest.Manyeffortshavebeenmadeondevelopingmicrowave
nearfieldscanningtechniques,andvarioustypesofnearfiledmicrowavemicroscopeshave
beendevelopedfordifferentpurposed.Underthereflectionmethod,thepropertiesofasample
areobtainedfromthereflectivityduetothepresenceofthesample.
In principle, any type of transmission lines can be used to develop nearfield microwave
microscopes. In a nearfield microwave microscope developed form parallelboard wave
guide, the most important part is an aperture in the form of a narrow slit (the following
243
244
mentionednanoslitplaysthisroleinMAFMprobe).Whenasamplesurfaceisinthenear
filedzoneoftheslit,themicrowaveisreflectedmostlyfromtheregionundertheslit.Since
reflectionfromasamplesurfaceisdeterminedbytheresistivity,bymeasuringtheamplitude
andphaseofthereflectedwavewhilerasterscanningthesurface,itispossibletomapthe
microwaveresistivityofthesurface.Forconductivelayerwiththicknessesmuchlargerthat
theskindepth,wecangetsurfaceimpedance,whileforthinlayer,wecangetsheetresistance.
In the determination of microwave resistivity, it is necessary to measure layer thickness
independently.
1.3.DevelopedAFMbasedandmicrowavetechniqueformeasuringtheelectrical
properties
Electrical properties are the most significant intrinsic characteristics of substances; they
strongly affect the work functions of different materials, especially in nanometerscale
materialsanddevices.Thus,measuringelectricalpropertieshasbecomeanurgentneedin
manyareasofmoderntechnology.Forinstance,intheelectronicsindustry,criticalfeature
sizes are becoming smaller, and it is necessary to evaluate the electrical properties of the
materialswiththespatialresolutiononananometerscaletoestablishtheknowledgetopredict
the behavior of materials in real devices. In addition, newly developed materials, such as
conducting plastic thin films and biomaterials, which may possibly have some uncertain
physicalproperties,willbeimportantinthefieldofsurfacescienceandbiologicalapplications.
Despitebeingintenselystudiedforyears,theirelectricalproperties,especiallytheirconduc
tivityandpermittivity,arestillpoorlyunderstood.
As the first section of this chapter saying, atomic force microscopy (AFM) has played an
importantroleinnanoscalescienceandtechnologybecauseitisoneofthemostversatile
instrumentsavailableforimagingandmanipulatingstructuresonthenanometerscale[47].
Several attempts based on atomic force microscopy have been made to characterize the
electricalinformationofmaterialsonthenanometerscale,suchasconductingatomicforce
microscopy(CAFM)[8,9],scanningcapacitancemicroscopy(SCM)[10,11]andelectrostatic
force microscopy (EFM) [12, 13]. Although CAFM can produce a nanoscale electrical
characterizationofthinfilms,theAFMtipmustcontacttheconductingsubstratetoapplya
current,soduringtheprobingprocess,theAFMtipwillscratchthesurfaceofthesample.SCM
cancharacterizeelectricalinformationbymeasuringthecapacitancebetweenthetipofthe
probeandthesample.However,itsuffersfromalimitedspatialresolutionandissensitiveto
the thickness of the specimen. EFM, including Kelvin probe force microscopy (KFM) [14],
scanningsurfacepotentialmicroscopy(SSPM)[15]andscanningMaxwellstressmicroscope
(SMM)[16],canmeasurethesurfaceelectricalpotentialofmaterialsbydetectingtheelectro
staticforcebetweentheprobetipandthesample.However,thevanderWaalsforcesand
chemicalbondingforces,aswellastheelectrostaticforcesareincludedinthemeasureddata.
Hence,thesamplesurfacechemistryandatmosphericconditionsgreatlyimpactthemeasured
electricalpotential.
Ontheotherhand,microwavemeasurementshavebeenofgreatinteresttomanyresearchers
becausemicrowavescanpropagateeasilyinair,andthesampleresponseisdirectlyrelatedto
theelectricalpropertiesofthematerial[17].Thus,toobtainthemicroscopicelectricalinfor
mation,avarietyofmicrowavemicroscopeshavebeendeveloped[18,19].Steinhaueretal.
developed a nondestructive and noninvasive nearfield scanning microwave microscope
(NSMM),whichcanimagethelocalpermittivityandtenabilityofadielectricthinfilmwitha
spatialresolutionof1m[20].ZhangandcoauthorsimprovedtheNSMMtoinvestigatethe
localperpendiculardielectricinformationofsinglephasemultiferroicthinfilmsandsingle
crystalmaterials[21].FerdDueweretal.introducedscanningevanescentmicrowavemicro
scopy (SEMM) [22, 23], which measures the changes of the tipsample capacitance at the
resonant frequency and the quality factor of microwave absorption. They succeeded in
imagingthetopographyandsurfaceresistanceofmetallicsamples.However,toevaluatethe
electrical properties of materials using microwaves, it is necessary to keep the standoff
distancebetweenthemicrowaveprobeandthesampleconstantbecausemicrowavesignals
inthenearfieldareextremelysensitivetothisdistance.Otherwise,itwouldbedifficultto
distinguishwhetherthechangesinthesignalareduetothedifferenceofthematerialprop
ertiesorthevariationofthestandoffdistance.Inparticular,toevaluatetheelectricalproper
tiesofmaterialswithhighresolutiononthenanometerscale,itisindispensabletocontrolthe
standoffdistancepreciselyontheorderofnanometers.
Recently,tosolvetheproblemofhowmicrowavesignalsareaffectedbythestandoffdistance,
atechniqueofcombiningAFMwithmicrowavemicroscopyhasbeenstudied[2427].K.Lai
etal.inventedamicrowaveimpedancemicroscope(MIM)[24,25],whichfedamicrowave
signaltoasiliconnitridecantileverwithaPttipthatwasusedtoinvestigatethenanoscale
dielectricinhomogeneityinanoninvasivemanner.TheWeidegroupcombinedanNSMM
withanAFM(NSMMAFM)[26,27]byaddingamicrowavesignaltoacommercialprobe.
TheNSMMAFMcanmeasurethetopographyanddielectricconstantofthinfilmsimultane
ously. However, it is noted that MIM and NSMMAFM do not use matched probes or
cantileversasthemicrowaveguideconnectedwiththesourceofmicrowavesignals.Thus,the
microwavesignalsmaynotpropagatealongtheprobeandemitfromthetipapexoftheprobe.
Therefore, these techniques can only measure the changes of the probesample system
impedancebutnottheintrinsicelectricalpropertiesofthemeasuredmaterials.
Tosummarize,theseAFMbasedmethodologiesandmicrowavemicroscopytechniquescan
only image relative electrical properties, rather than the absolute values of the intrinsic
electricalproperties,suchastheconductivity,permittivity,andpermeability.Thus,theneed
remainsforamicroscopytechniquethatcanprovideasimultaneousmeasurementoftopog
raphyandelectricalpropertiesonthenanometerscale.
2.Theprinciple,fabricationandevaluationofmicrowaveAFM
Themicrowaveatomicforcemicroscope(MAFM)isacombinationoftheprinciplesofthe
scanningprobemicroscopeandthemicrowavemeasurementtechnique[2832].MAFMcan
maintaintheconstantstandoffdistancebetweentheMAFMprobetipandscannedsample
surface, by detecting the deflection of the atomic force between them, and measure the
electricalpropertiesofmaterialswithnanometerscalespatialresolution.
245
246
Figure3showstheschematicdiagramoftheMAFMprobethatwasusedtomeasurethe
electricalpropertiesofmaterialsinthisstudy.DifferentwiththenormallycommercialAFM
probes, the MAFM probe is having a special structure. That is, a pair of metal films was
fabricatedonsurfacesoftheMAFMcantileverbyEB(electronbeam)vapormethod.Themetal
filmsplayaroleofparallelboardmicrowavesignalwaveguide,whichcanmakethemicro
wavesignalpropagateinthebodyofMAFMcantileverandemitattheprobetipapex.Then,
bydetectingtheresponseofmicrowavesignalreflectedfromthevicinityofthesample,the
electricalpropertiescanbeevaluatedonthenanometerscale.
Figure 3. Schematic diagram of the M-AFM probe that was used to measure the electrical properties of materials in
this study.
2.1.FabricationofMAFMprobe
2.1.1.FabricatingthetipofMAFMprobe
TorestraintheattenuationofmicrowaveintheMAFMprobe,GaAswasusedasthesubstrate
of the probe. On the other hand, to obtain the desired structure, wet etching was used to
fabricatethetipoftheprobe.Differentwiththedryetching,asideetchingwilloccurunder
theetchingmask.Utilizingthisproperty,amicrotipcanbefabricatedbyetchingawafer,of
whichasmallmaskwasintroducedonthesurfaceinadvance.Inthecaseofsinglecrystalline
wafer,suchasSiandGaAs,thechemicalactivitiesaredifferentfordifferentcrystallineplanes,
thereby,theetchratesarealsodifferent.Therefore,thesideplaneobtainedatthesideofthe
maskpatternisthemostinactiveplane(thatistheplanehavingthemostlowetchingspeed)
whichisparalleltothesideofthemaskpattern.Consequently,theresultofetchingisstrong
affectedbythedirectionofmaskpattern.Ontheotherhand,GaAshasasphaleritestructure
thatismorecomplexthanthatofSi,whichhasasimilarstructureasdiamond.Therefore,the
predictionoftheetcheffectsisverydifficult[28,29].
Figure 4. Fabricated tip of GaAs probe.
Figure 5. GaAs probe tip with high magnification of apex part. The key component in an AFM is the tip, which should
be as very sharp as possible.
Inourstudy,itwasfoundthatonlythesquareresistpatterncanformasharptip(Figure4
andFigure5).Inthecaseofhexagonalpattern,thereasonthattipwasnotformedwellmay
beduetothesideoftheetchingmasktobetooshort.Thereasonfortriangularpatternmay
duetothatthereisnocrystallineplaneparalleltothesideoftheetchingmask.Inaddition,it
wasalsofoundthatonesideofthesquaremaskbeing45tothe<011>directioncanformatip
withahigheraspectratiocomparingwiththecaseofonesideoftheresistpatternbeingparallel
to<011>direction.
247
248
2.1.2.FabricationofMAFMprobe
TheprocessofprobefabricationisshowninFigure6indetails:(a)Patterningtheetchingmask
fortipgeneration;(b)Formingthetipsbywetetching;(c)Patterningthestencilmaskforthe
waveguideandevaporatingthemetalfilm;(d)Removingresistandfilm;(e)Patterningthe
etchingmaskforthebeamofcantilever;(f)Formingthebeamofcantileverbywetetching;(g)
Patterningtheetchingmaskonbacksideforthefabricationoftheholder;(h)Formingthe
holder;(i)Evaporationofmetalfilmonthebackside;(j)Introducingslitapertureatthetipof
theprobe.
Intheexperiment,nodopedsemiinsultedGaAswaferhaving(100)orientedsurfaceand350
mthicknesswasused.Atfirst,thetipswereformedbyetchingthewaferfor100secondsto
reachtheetchingdepthof7.7m.Afterthat,Aufilmusedtoconstructthewaveguidewas
evaporated on the substrate. The film thickness was about 50 nm (Figure 6(c)). After the
deposition,thepatternofthewaveguidewasformedbyliftoffprocess,wherethefilmonthe
resist mask corresponding to the area without waveguide pattern was removed (Figure
6(d)).Then,inordertoformthebeamofthecantilever,thebeametchingmaskwaspatterned.
Here,byconsideringthechemicalactivitiesatdifferentcrystallineplanes,thelengthdirection
oftheetchingmaskwaspatternedalongthe<011>direction.Inconsequence,thesideetching
occurredundertheresistmask,andmesatypeplanesappearedatthebothsidesofthebeam
(45inclinedplane).Ontheotherhand,inversemesatypeplanewasformedattheendofthe
beam(6075inclinedplane).Etchingdepthofthebeamwasabout20m.
Inthesameconditionsasthebeamfabricationprocess,holderwasformedbybacksideetching
(Figure6(f)).Here,theetchingmaskwaspatternedonthebottomsurface,andetchingwas
carriedoutuntilthesubstratewaspenetrated.Thestirringwasperformedbymagneticstirrer
inordertoetchthesampleuniformly.Inthestep(i)asshowninFigure6,Aufilmwasdeposited
onbottomsurfaceoftheprobetopropagateamicrowavesignalintheprobe.Thethickness
ofthefilmwas50nm,whichisthesameasthatonthetopsurfaceoftheprobe.Bothplane
surfacesofthewaveguidewhichwereevaporatedAufilmareconnectedattheendofthe
beam.However,thereisnoAufilmonthesidesofthebeam,sincetheformedinclinedplanes
at the beam sides are not face to the direction of the evaporation. Finally, by using FIB
fabrication,aslitatthetipofprobewasformedtoopentheconnectionoftheAufilmonthe
twosurfacesoftheprobe.Consequently,ahomogeneousparallelplatewaveguidewasformed
andmicrowavesareabletopropagatealongtheprobeandemitatthetipapexoftheMAFM
probe.
ItshouldbementionedthatdimensionsoftheGaAssubstrateandtheAufilmsoftheM
AFM probe decide the characteristic impedance of the waveguide, in order to make
certainlythatmicrowavesignalscanpropagateproperlyintheMAFMprobeformaxi
mumsensitivity,thewaveguideshouldhaveacharacteristicimpedanceof50(tomatch
thecharacteristicimpedanceofacoaxialtransmissionline).Thus,thecantileverandthe
body of the MAFM probe were designed with the dimensions of 2503015 m and
2740720340mrespectively.
Figure 6. Fabrication processes of the M-AFM probe. (a) Patterning the etching mask for the generation of tip. (b)
Forming the tip by wet etching. (c) Patterning the resist mask for the waveguide. (d) Evaporating the metal film. (e)
Removing the resist and metal film. (f) Patterning the etching mask for the beam of cantilever. (g) Forming the beam
of cantilever by wet etching. (h) Patterning the etching mask on back side for fabrication of the holder. (i) Forming the
holder. (j) Evaporation of metal film on the back side. (k) Introducing the micro slit at the tip of probe.
2.2.SEMobservationforfabricatedMAFMprobes
TheSEMimagesofthefabricatedMAFMprobesaredepictedinFigure7toFigure10.Figure
7showstheSEMphotographofthefabricatedMAFMprobes.There44probeswerefabricated
inoneprocessforonesubstrate.Figure8showstheasfabricatedcantileveroftheMAFM
probe.ThedimensionsoftheMAFMprobedependonseveralsmallvariationsofexperi
mentalparameters,includingthedevelopingtimeoftheresistpattern,thewetetchingrate,
andtheEBevaporationrate.TheaveragedimensionsofthecantileverandthebodyoftheM
AFM probes are typically 2523114 m and 2742723339 m, respectively. Thus, the
characteristicimpedanceoftheMAFMprobesis,onaverage,49.3.Figure9depictsanSEM
photograph of the FIBfabricated nanoslit that has been patterned across the cantilever
through the center of the probe tip. The observed tip is located near the front edge of the
cantilever.AscanbeobservedinFigure10,thetipisapproximately7mhigh,andthenano
slitisapproximately100nminwidth.
249
250
Figure 7. The fabricated M-AFM probes.
Figure 8. The cantilever of the M-AFM probe.
Figure 9. The 100-nm-wide-FIB-fabricated nano-slit that is across the cantilever and through the center of the tip.
Figure 10. The high-magnification image of the M-AFM-probe tip.
251
252
2.3.MeasuringtopographybyMAFMprobe
InordertoconfirmthespatialresolutionofthefabricatedMAFMprobes,theAFMtopogra
phyoftwogratingsampleshaving2000line/mmand17.9nmstepheightweremeasuredby
acommercialSiAFMprobe,aGaAsprobewithoutnanoslitandaMAFMprobewithnano
slit,respectively.
Probe
A commercial Si probe
A GaAs probe without the

nano-slit
A M-AFM probe with the

nano-slit
The resonance frequency
Q-value
Spring constant (N/m)
262
370
Typical value: 42
185
510
Typical value: 134
201
333
Typical value: 134
(kHz)
Table 1. The properties of AFM probes in the atmosphere.
A JSPM5400 was used for measurement of the sample under the noncontact mode
(frequencymodulation(FM)mode).Thepropertiesofthreekindsofprobearegivenin
Table1,theresonancefrequencywassweptandtheQvaluewasdenedbythefollow
ingrelation,Q=f0/(f+f),wheref0 isthepeakfrequency,f+ andf theshiftedfrequencyfrom
f0 at70.7%ofpeakintensity.TheQvalueindicatesaresonancesharpnessofthecantile
ver;thehighertheQvalue,thebetterstabilizationoftheoscillation.
Figures11to13showthetopographiesofthestandardsamplehaving2000lines/mmobtained
bythecommercialSiprobe,GaAsprobewithoutthenanoslit,andMAFMprobewiththe
nanoslitunderthenoncontactmode,respectively.Themeasurementswereperformedinthe
air,andtheAFMworkedinnoncontactmode,withaworkingenvironmenttemperatureof
25.0Candarelativehumidityof50%.TheresonancefrequencyofSiprobeandMAFMprobes
(withoutnanoslitandwithnanoslit)were262kHz,185kHzand201kHz,respectively,and
theQvalueofthemwere370,510and333.Thescanareawas22m2,scanningspeedwas3
m/s,andthewhitespotsinthesefiguresareduetomicrodustonthesamplesurface.Even
thoughtheQvalueislowerthanthatoftheGaAsprobewithoutthenanoslit,thecommercial
Siprobestillcanobtainalittlehigherresolutiontopographyduetothehigheraspectratioof
thetip.
ComparingtheobtainedimagesofFigure12,Figure13withtheonesinFigure11,theresults
illustrate that MAFM probe has a similar capability for sensing surface topography of
materialsasthatofcommercialAFMprobes.
Figure 11. Surface topography of the grating sample obtained by the commercial Si probe.
Figure 12. Surface topography of the grating sample obtained by the GaAs probe without the nano-slit.
253
254
Figure 13. Surface topography of the grating sample obtained by the M-AFM probe with the nano-slit.
Figure 14. Topography of the grating sample obtained by the commercial Si probe.
Figure 15. Topography of the grating sample obtained by the commercial Si probe.
Inordertoevaluatetheaccuracyofheightmeasurement,agratingsamplehaving17.9nm
1nmstepheightwasmeasuredbyusingthecommercialSiprobeandtheMAFMprobe,
respectively.Figures14and15showtheAFMtopographiesandthecrosssectionprofilesof
thegratingsampleobtainedbythecommercialSiprobeandthefabricatedMAFMprobe,
respectively.Fromtheslopeofthestepofthecrosssectionprofileinthefigures,itisconfirmed
thatthefabricatedMAFMprobeshavethecapabilitytocatchtheAFMtopographywiththe
resolutionofnanometerorder.Theheightofthestepofthegratingsampleobtainedbyeach
probewas19.17nmand19.67nm,respectively.ThefabricatedMAFMprobehasalsohigh
resolutionalthoughtheresolutionwasinferiorascomparedtothecommercialSiprobe.The
reasonwhytheresolutiondegradedwiththefabricatedprobeisthatthetipoftheprobeswas
cut by FIB fabrication. From these results, it is considered that the control of the standoff
distancebetweentheprobeandthesamplewithhighprecisionwasachievedbythefabricated
MAFMprobe.
255
256
3.Microwaveimagingformaterialsonnanometerscale
3.1.Experimentalsetup
Figure16schematicallydepictstheintegratedtestsystemoftheMAFM[31].InourMAFM
system,theinitialmicrowavesignals,whichareworkingatafrequencyf=16.66GHz,are
generatedbyamicrowavegenerator.Next,thefrequencyofthemicrowavesignalsisextended
by a sixfrequency multiplier, which results in a stable testing frequency f = 94 GHz. The
microwavesignalspropagatethroughanisolatorandacirculatorandthenpropagateintothe
MAFMprobe.Thetransmissionlinethatconnectsthecirculatorandtheprobechangesfrom
a rectangular waveguide into a coaxial line, which then changes into the parallelplate
waveguide(intheMAFMprobe).Adetectorisconnectedtothecirculator,tomeasurethe
microwavesignalsthatarereceivedbythetipoftheprobeandindicatethevoltagedatathat
areconvertedfromthereflectedmicrowavesignals.Themeasuredsignalsaresynchronized
withpositionalinformationthatisobtainedfromtheAFMscanner,whichisthenusedtocreate
amicrowaveimage.Atthesametime,byevaluatingtheoutputvoltagedata,theelectrical
propertiesofthemeasuredmaterialscanbedetermined.
Figure 16. Diagram of the M-AFM system.
Wepreparedasampleforthescanningtestofsurfacetopographyandmicrowaveimaging.
At first, a resist mask was patterned onto the glass substrate wafer by lithography. After
developingtheresistpattern,a200nmthickAulayerwasdepositedontheglasssubstrateby
electronbeam(EB)evaporation.Finally,theunexposedphotoresistswereliftedoffinacetone.
TheresultingAuandglassstepstructureisdepictedinSEMimageofFigure17.
Figure 17. SEM image of measured sample.
3.2.MicrowaveimageofAu/glassstepsample
Figures18and19depicttheMAFMscanningresultsofthesampleatthestepareabetween
theAucoatingfilmandtheglasswafersubstrate.Themeasurementswereperformedinthe
air,andtheMAFMworkedinnoncontactmode,withaworkingenvironmenttemperature
of24.5Candarelativehumidityof38.4%.TheresonancefrequencyofMAFMprobewas
133kHzandtheQvalueofitwas295.Thescanareawas1010m2,scanningspeedwas5
m/s.Figure18depictsthesurfacetopographyoftheMAFMmeasuredsample.Inthisimage,
theleftsiderepresentstheAufilm,whereastherightsideistheglasssubstrate.Ascanbeseen
inthescanningprofiledepictedinFigure18,thethicknessoftheAufilmwasapproximately
200nmonaverage.
Figure19depictsthemicrowaveimageofthevoltagethatwasconvertedfromthemeasured
microwavesignals,whichweresimultaneouslyacquiredbytheMAFMprobeatthecorre
sponding position depicted in Figure 18. This experimental result demonstrates that the
microwaveimagehastwospatialphases.Becausethestandoffdistancebetweenthetipofthe
MAFMprobeandthesurfacesoftheAufilmandglasssubstrateisconstantandcontrolled
bytheatomicforce,thus,theresponseofthemicrowavesignalswereobservedtochangebased
onthedifferentelectricalcharacteristicsofthemeasuredmaterials.AsperFigure19,theoutput
voltageovertheglassareaislargerthanthatovertheAuarea,becausethescanningstarted
fromtheAuareawiththeinitialoffsetfromthenullingoperation.
257
258
Figure 18. AFM topography image of the Au/Glass step sample.
Figure 19. Microwave image of the output voltage that was converted from the measured microwave signals.
An analysis of the scanning profile depicted in Figure 19 demonstrates that the spatial
resolutionishigherthan120nm,andthattheoutputvoltagemeasuredovertheAuandglass
areas were 262.5 mV and 281.7 mV, respectively. The difference between the measured
voltagesbetweentheAuandglassareasis19.2mV.Since,thestabilityofthemeasurementis
high,thisvalueislargeenoughforevaluatingtheelectricalpropertiesofothermaterialshaving
theconductivitybetweenAuandglass.Astheresultspresented,theMAFMshouldallowus
to scan the electrical conductivities of other conductor materials on a nanometer scale. In
addition, based on the same principle, it can also be used to measure the permittivities of
dielectricmaterialsonananometerscale.
4.Quantitativemeasurementoftheelectricalpropertiesofmaterialsonthe
nanometerscale
Forquantitativemeasurement,theoperatingfrequencyofMAFMissetat94GHz.Thehigh
frequencymicrowavesareeasytopropagateinthewaveguideandemitfromthenanosliton
the probe tip. Since the width of the nanoslit is around 100 nm, the field of microwave
interactingwiththemeasuredmaterialscanbeconsideredtobein100nmorder.Thus,ifthe
thicknessofmeasuredmaterialsislargerthan100nm,thereflectionfromthebottomsurface
ofthesamplecanbeneglected.Therefore,onlythereflectionfromthetopsurfaceneedstobe
considered.
Moreover,thediodedetectorworksinasmallsignalrange,whereitisconsideredtobea
squarelawdetector.Therefore,whilekeepingthestandoffdistancebetweenthetipoftheM
AFMprobeandsamplesconstant,theoutputreflectedvoltageV,whichvariesonlywiththe
conductivityofthesample,hasarelationshipwiththesquaredabsolutevalueofthetopsurface
reflectioncoefficient, | s | 2as
2
V k0 s b0
(1)
Thetwoundeterminedconstantsk0andb0canbecalibratedwithtwosampleswhoseconduc
tivities are known. For good conductors, which are used in this experiment, the surface
reflectioncoefficient | s | canbewrittenas[34]
1 / j 0
1 / j 0
(2)
where 0 and represent permittivity of free space and the conductivity of the measured
material,respectively,andistheangularfrequencyofthemicrowave.Forsemiconductor
259
260
orisolatingmaterials,similarequationscanalsobeconstructed.Then,theconductivitycanbe
determinedfromEq.(47)as
2
2
2
0 4 s s 1 / s 1

4 s s 1
2
4 s
(3)
Afterk0andb0inEq.(1)arecalibratedusingtworeferencesampleswithknownconductivities,
theconductivitiesofanysamplescanbecalculatedfromthemeasuredvoltage.Therefore,the
MAFMallowsustoquantitativelyevaluatetheelectricalconductivitiesofmaterialsonthe
nanometerscale.
ItshouldbenotedthatEq.(2)and(3)arederivedundertheplanewavecondition,whilethe
probe works in nearfield mode. Although nearfield analysis may further improve the
precision of evaluation results, it requires more reference samples, which will increase the
complexityofthemeasurement.Sincethetestedmaterialwasveryclosetotheopenendof
theprobetip(thestandoffdistanceofseveralnanometerswasextremelysmallascompared
withthewaveguidewidth(~100nm)andthewavelength),thisproblemcanbeequivalentto
thecasethatthematerialsurfaceisterminatedattheendofthewaveguide,whichcanbe
representedbytheplanewavemodel.Therefore,theplanewaveapproximationisusedinthis
study.
ThereisalimitationofMAFMtechniquewehavetoface.Inthecaseofthatthethicknessof
measuredsamplesmallerthan100nm,thereflectionofmicrowavesignalfromthebottom
surfaceofthesampleandthesubstrateshouldbeconsidered.Therefore,theMAFMcannot
usethepreviousmentionedequationstoobtaintheelectricalpropertiesofmeasuredsample
quantitatively.
Fivedifferentmetallicfilms(Cu,Pb,Al,CoandZn)withEBfabricationwerepreparedforthe
quantitativemeasurement.Thetestedelectricalconductivitiesbythefourpointprobevander
PauwmethodwereobtainedasthestandardvaluesforcalibrationandevaluationofMAFM
results.Thetestedelectricalconductivitiesofthesemetalfilmsareintherangeof4.46106
S/mto5.68107S/m.Themeasurementswereperformedintheair,andtheAFMworkedin
noncontactmode,withaworkingenvironmenttemperatureof23.0Candarelativehumidity
of50%.TheresonancefrequencyofMAFMprobewas107kHzandtheQvalueofitwas675.
Thescanareawas22m2,scanningspeedwas1m/s.Beforescanning,wesettheoriginal
voltagetobezerowhilemaintainingaconstantdistanceof2.6mbetweentheprobetipand
the sample. During the scanning process, the standoff distance between the probe tip and
sampleswasfixedatseveralnanometersbytheatomicforce,andthevoltagecorresponding
totheinspectedsamplewasmeasured.
Figures 20 to 24 show the topographies and microwave images of the five samples. The
variationsofthemeasuredvoltagesforthefivesamplesarelessthan0.46mV,whichismuch
smaller than the dynamic range of the MAFM. The signaltonoise ratio of the MAFM
measurementswasevaluatedtobe20.14dBonaverage.
Figure 20. Topography and microwave image of measured Cu sample.
Figure 21. Topography and microwave image of measured Pb sample.
Figure 22. Topography and microwave image of measured Co sample.
261
262
Figure 23. Topography and microwave image of measured Al sample.
Figure 24. Topography and microwave image of measured Zn sample.
The Figure 25 shows the variation margins of measured local voltages for samples with
differentconductivities.UsingthemeasuredvoltagesoftwosamplesobtainedfromFigures
20and21(4.89mVforCuand18.01mVforPbonaverage)andtheirtestedconductivities
(5.68107S/mforCuand4.46106S/mforPb)forcalibration,thetwoundeterminedconstants
inEq.(1)werecalculatedtobek0=5.9632andb0=5.9629.Then,theconductivitiesofAl,Co
and Zn samples were evaluated with Eq. (1) and Eq. (2) by using the measured voltages
obtainedfromFigures20to24.
Figure 25. Variation margins of measured local voltages for samples with the different conductivities.
Figure26showsthevariationmarginsofmeasuredlocalvoltagesforsampleswiththesquare
ofsurfacereflectioncoefficientofthem.Aspreviousmentioned,twoundeterminedconstants
inEq.(1)werecalculatedtobek0=5.9632andb0=5.9629.Thatmeansthemethodinthiswork
wasbasedonapremiseofthatthesurfacereflectioncoefficientandmeasuredvoltageshould
bekeptinalinearrelationship.Itisnotedthatthesurfacereflectioncoefficientandmeasured
voltagecouldbeprovidedinalinearrelationship(seethefittingstraightlineinFigure26).
Thus,theapplicabilityoftheevaluationmethodinmyworkcanbeproved.
Figure 26. Variation margins of measured local voltages for samples with the square of surface reflection coefficient
of them.
263
264
Figure27showstheevaluatedresultsversusthetestedvaluesofAl,CoandZnsamples.Itis
notedfromFigures20to24thatnocorrelationcanbeobservedbetweenthemicrowaveimages
andtheircorrespondinggeometryimages.Inotherwords,thevariationsofthemeasuredlocal
voltages are not caused by the surface morphology. The main causes of error bars of the
evaluatedconductivitiesareasfollows.Firstly,thefilmsamplespreparedbyEBevaporation
were not homogenous in the microscopic view, and the distribution of conductivity was
locationdependent(localconductivity).
Figure 27. Evaluated conductivities of the samples in comparison with the tested conductivities of them.
Itisbelievedthatthevariationmarginsofmeasuredlocalvoltages(seeFigure25)bytheM
AFMcausedtheerrorbarsoftheevaluatedconductivities.Secondly,themicrowavesignalfor
conductivity measurement was very small, which might be affected by the measurement
environment.Therefore,theuncertaintyofthemicrowavemeasurementmaycontributetothe
errorbars.ItisalsonotedfromFigure27thatthedeviationofevaluatedconductivitiesfrom
thevaluestestedbytheVanderPauwmethodis2.03%,7.24%and11.6%fortheZn,Coand
Al, respectively. One of the causes of this deviation is that the standoff distance variation
betweendifferentmaterialsmayaffectthemeasuredvoltage,therebyinducingdeviationof
evaluatedconductivity,especiallyforhighconductivitymaterialssuchasAl.Anothercause
ofthedeviationmaybetheevaluationequationwhichwasderivedundertheplanewave
approximationratherthanthemuchmorecomplicatednearfieldanalysis.Thequantitative
evaluation was performed three times, and the similar results as shown in Figure 25 were
obtained.Ontheotherhand,theevaluatedresistivitiesofthefivesamplescanbepresented
outasshowninFigure28.
Figure 28. Evaluated resistivities of the samples in comparison with tested c resistivities of them.
5.Summary
a.
Weinventedoutanoveldevicenamedofmicrowaveatomicforcemicroscope(MAFM),
which is a combination of the principles of the scanning probe microscope and the
microwavemeasurementtechnique.MAFMcanmaintaintheconstantstandoffdistance
betweentheMAFMprobetipandscannedsamplesurface,bydetectingthedeflection
oftheatomicforcebetweenthem,andmeasuretheelectricalpropertiesofmaterialswith
nanometerscalespatialresolution.
b.
MicrowaveAFM probes were fabricated on the GaAs wafer by using the wet etching
process.AwaveguidewasintroducingontheprobebyevaporatingAufilmontheboth
surfacesoftheprobes.Theopenstructure(thenanoslit)ofthewaveguideatthetipapex
oftheMAFMprobewasobtainedbyusingFIBfabrication.SEMwasusedtoobservethe
fabricatedMAFMprobes.Astheresults,theaveragedimensionsofthecantileverand
the body of the MAFM probes are typically 2523114 m and 2742723339 m,
respectively. Based on these dimensions, the characteristic impedance of the MAFM
probesis,onaverage,49.3.Inthisway,theMAFMprobecouldmatchwellwiththe
coaxialline,whichhasanimpedanceof50.Theobservedtipislocatednearthefront
edge of the cantilever and the tip is approximately 7 m high, and the nanoslit is
approximately100nminwidth.
c.
TheAFMtopographyofthegratingsamplehaving2000line/mmand18nmstepheight
wasmeasuredbythefabricatedMAFMprobe.AFMmeasurementswereperformedby
comparingwiththecommercialSiAFMprobe.TheresultsindicatedthatGaAsmicro
waveprobehasacapabilitytocatchAFMtopographyofgratingsamplesandhavinga
highaccuracyforlateralandheightevaluation,similarasthecommercialAFMprobe.
265
266
d.
We have created an MAFMobtained microwave image using a compact microwave

instrumentthatwasoptimallysynchronizedwithanAFMscanner.Thedistinguishing
featuresofMAFMareitsabilitytomaintainaconstantstandoffdistancebetweenthe
probetipandthesamplesurfaceandtomeasurethemicrowavesignalinteractedwith
thesample.Therein,boththetopographyandelectricalpropertyimagesofthesample
canbesimultaneouslycharacterized.Therefore,MAFMisabletomeasure,insitu,the
distributionofelectricalpropertiesonananometerscale.Asshownintheexperimental
results,wesuccessfullygeneratedamicrowaveimageofa200nmAufilmcoatingona
glasswafersubstratewithaspatialresolutionof120nm,and,moreover,wemeasured
thevoltagedifferencebetweenthesetwomaterialstobe19.2mV.Webelievethatthehigh
spatialresolutionandsimultaneousmeasurementcapabilityofthisMAFMsystemwill
haveimportantimplicationstonanotechnologycharacterizationintheimmediatefuture.
e.
We also demonstrated a novel evaluation equation and calibration technique for the
quantitativemeasurementofthelocalconductivity.Basedontheanalyticalandexplicit
expressions proposed, using two reference samples with known conductivities, the
conductivitiesofanysamplescanbecalculatedfromthemeasuredvoltage.Ourresults
demonstrate that MAFM is able to quantitatively measure, in situ, the distribution of
electricalpropertiesonthenanometerscale.
Authordetails
YangJu
DepartmentofMechanicalScienceandEngineering,NagoyaUniversity,Japan
References
[1] BinnigG,RohrerH,GerberC,WeibelE.SurfaceStudiesbyScanningTunnelingMi
croscopy.PhysicalReviewLetters1982;495761.
[2] Binnig G, Quate C. F, Gerber C. Atomic Force Microscope. Physical Review Letters
1986;56930933.
[3] RugarD,HansmaP.AtomicForceMicroscopy.PhysicsToday1990;432330.
[4] MeliM.V,BadiaA,GrtterP,LennoxR.B.SelfAssembledMasksfortheTransferof
NanometerScale Patterns into Surfaces: Characterization by AFM and LFM. Nano
letters2002;2131135.
[5] ChenL.W,CheungC.L,AshbyP.D,LieberC.M.SingleWalledCarbonNanotube
AFMProbes:OptimalImagingResolutionofNanoclusterandBiomoleculesinAm
bientandFluidEnvironments.Nanoletters2004;417251731.
[6] WenC.K,GohM.C.AFMNanodissectionRevealsInternalStructureDetailsofSin
gleCollagenFibrils.Nanoletters2004;4129132.
[7] WadeL.A,ShapiroI.R,MaZ.Y,QuakeS.R,CollierC.P.CorrelatingAFMProbe
Morphology to Image Resolution for SingleWall Carbon Nanotube Tips. Nano let
ters2004;4725731.
[8] Olbrich A, Ebersberger B, Boit C. Conducting Atomic Force Microscopy for Nano
scale Electrical Characterization of Thin SiO2. Applied Physics Letters 1998;73
31143116.
[9] XuD.G,WattG.D,HarbJ.N,DavisR.C.ElectricalConductivityofFerritinProteins
byConductiveAFM.NanoLetters2005;5571577.
[10] KopanskiJ.J,MarchiandoJ.F,LowneyJ.R.ScanningCapacitanceMicroscopyMeas
urements and Modeling: Progress towards Dopant Profiling of Silicon. Journal of
VacuumScienceandTechnologyB1996;14242247.
[11] SmolinerJ,BreznaW,KlangP,AndrewsA.M,StrasserG.QuantitativeScanningCa
pacitance Microscopy on Single Subsurface InAs Quantum Dots. Applied Physics
Letters2008;92092112.
[12] Stern J. E, Terris B. D, Mamin H. J, Rugar D. Deposition and Imaging of Localized
Chargeoninsulatorsurfacesusingaforcemicroscope.AppliedPhysicsLetters1988;
5327172719.
[13] Hu Z. H, Fischbein M. D, Drndi M. Local Charge Transport in Twodimensional
PbSe Nanocrystal Arrays Studied by Electrostatic Force Microscopy. Nano Letters
2005;514631468.
[14] NonnenmacherM,OBoyleM.P,WickramasingheH.K.KelvinProbeForceMicro
scopy.AppliedPhysicsLetters1991;5829212923.
[15] FujiharaM,KawateH.ScanningSurfacePotentialMicroscopeforCharacterizationof
LangmuirBlodgettFilms.ThinSolidFilms1994;242163169.
[16] Yokoyama H, Inoue T. Scanning Maxwell Stress Microscope for Nanometerscale
SurfaceElectrostaticImagingofThinFilms.ThinSolidFilms1994;242,3339.
[17] JuY,InoueK,SakaM,AbH.ContactlessMeasurementofElectricalConductivityof
Semiconductor Wafers Using the Reflection of Millimeter Waves. Applied Physics
Letters2002;8135853587.
[18] Ash E. A, Nicholls G. Superresolution Aperture Scanning Microscope. Nature
1972;237510512.
[19] RosnerB.T,VanderWeideD.W.HighfrequencyNearfieldMicroscopy.Reviewof
ScientificInstruments2002,73,25052525.
[20] SteinhauerD.E,VlahacosC.P,WellstoodF.C,AnlageS.M,CanedyC,RameshR,
Stanishevsky A, Melngailis J. Imaging of Microwave Permittivity, Tunability, and
267
268
Damage Recovery in (Ba, Sr)TiO3 Thin Films. Applied Physics Letters 1999;75
31803182.
[21] ZhangX.Y,WangX.C,XuF,MaY.G,OngC.K.HighFrequencyDielectricProper
tiesDistributionofBiFeO3ThinUsingNearfiledMicrowaveMicroscopy.Reviewof
ScientificInstruments2009;80114701.
[22] DuewerF,GaoC,TakeuchiI,XiangX.D.TipsampleDistanceFeedbackControlina
Scanning Evanescent Microwave Microscope. Applied Physics Letters 1999;74
26962698.
[23] GaoC,HuB,ZhangP,HuangM.M,LiuW.H,TakeuchiI.QuantitativeMicrowave
EvanescentMicroscopyofDielectricThinFilmsUsingaRecursiveImageChargeAp
proach.AppliedPhysicsLetters2004;8446474649.
[24] LaiK,KundhikanjanaW,PengH,CuiY,KellyM.A,ShenZ.X.TappingModeMi
crowaveImpedanceMicroscopy.ReviewofScientificInstruments2009;80043707.
[25] KundhikanjanaW,LaiK.J,WangH.L,DaiH.J,KellyM.A,ShenZ.X.Hierarchyof
ElectronicPropertiesofChemicallyDerivedandPristineGrapheneProbedbyMicro
waveImaging.NanoLetters2009;937623765.
[26] VanderWeideD.W.LocalizedPicosecondResolutionwithaNearfieldMicrowave/
ScanningforceMicroscope.AppliedPhysicsLetters1996;70677679.
[27] KarbassiA,RufD,BettermannA.D,PaulsonC.A,VanderWeideD.W,Tanbakuchi
H, Stancliff R. Quantitative Scanning Nearfield Microwave Microscopy for Thin
Film Dielectric Constant Measurement. Review of Scientific Instruments 2008;79
094706.
[28] Ju Y, Kobayashi T, Soyama H. Development of a Nanostructural Microwave Probe
BasedonGaAs.MicrosystemTechnologies2008;1410211025.
[29] Ju Y, Hamada M, Kobayashi T, Soyama H. A Microwave Probe Nanostructure for
AtomicForceMicroscopy.MicrosystemTechnologies2009;1511951199.
[30] Hosoi A, Hamada M, Fujimoto A, Ju Y. Properties of MAFM Probe Affected by
NanostructuralMetalCoatings.MicrosystemTechnologies2010;1612331237.
[31] ZhangL,JuY,HosoiA,FujimotoA.MicrowaveAtomicForceMicroscopyImaging
forNanometerscaleElectricalPropertyCharacterization.ReviewofScientificInstru
ments2010;81123708.
[32] FujimotoA,ZhangL,HosoiA,JuY.StructureModificationofMAFMProbeforthe
MeasurementofLocalConductivity.MicrosystemTechnologies2011;17715720.
[33] Giessibl F. J. Advances in Atomic Force Microscopy. Reviews of Modern Physics
2003;75949.
[34] PozerD.M.MicrowaveEngineering2nded.NewYork:JoneWileyandSons;1998.
[35] LinC.C,JuoT.J,ChenY.J,ChiouC.H,WangH.W,LiuY.L.EnhancedCyclicVol
tammetryUsing1DGoldNanorodsSynthesizedviaAAOTemplateElectrochemical
Deposition.Desalination2008;233113119.
[36] YueY.M,ChenM.J,JuY,ZhangL.StressinducedGrowthofWellAlignedCu2O
NanowireArraysandTheirPhotovoltaicEffect.ScriptaMaterialia2012;668184.
269
Chapter 13
Aerospace Application
Akihiro Sasoh
1.Introduction
Usually,Micronanotechnologyreferstomechatronicsinmicrometertonanometerlength
scale,inotherwords,tospatialmicronanometertechnology.Itisalsopossiblethattempo
rallylocalizedmechatronicsiseffectivetoimprovemacroscalemechatronics.Thisconcept
canbereferredtoastemporalmicronanotechnology,inwhichmechanical/electricalinputs
are done during a short time of the order of mirco or nanosecond. In this chapter, we
demonstratethistechnologyinaerospaceapplication.
2.Improvementofsupersonicaerodynamicsusingrepetitivenanosecond
laserpulses
Intheaspectofthesupersonicaerodynamicperformance,shockwaveformationleadsinevita
blytoseriousproblemspreventingthedevelopmentofhighperformancesupersonicvehicle.
Arepresentativeproblemissonicboomwhichisanimpulsivenoiseinducedbyasupersonic
aircraft.Wavedragforceinducedbyshockwaveisanotherseriousproblemagainstimprove
mentoftheaerodynamicperformance.Thisstudyconsidersfurthertechnologytoreduceawave
dragforce.Wavedragreductionovera20mmdia.cylinderwithatruncatedconenoseina
Mach1.94flowisdonebydepositinglaserpulseenergiesatrepetitionfrequenciesupto80kHz
andaverageinputpowerof400Watamaximum.Inactualapplicationofenergydeposition
scheme[18]toreducethedrag,itshouldbetakenintoaccountbothadragcoefficientand
efficiencyofenergydeposition.Thepurposeofthischapteristoinvestigatetheimpactsofnose
shapeonsupersonicdragreductionperformancewithrepetitiveenergydepositions.
2.1.Tradeoffintruncatedconeshape
Figure1indicateswhytruncatedconeshapeisimportanttoapplyenergydepositionscheme.
Inordertorealizetheenergydepositionsforimprovementofdragreduction,twoconditions
shouldbesatisfied.Thefirstconditionisthatthemagnitudeofadragforceoradragcoefficient
2013 Sasoh; licensee InTech. This is an open access article distributed under the terms of the Creative
272
hasalowervaluethantheoneofabaselinevalue,forexampleobtainedbyaconicalbody.
Anotherconditionistoobtainalargevalueofanefficiencyofenergydeposition[1](orpower
gain),whichisdefinedbytheratioofasavedpropulsionpowertoapowerdepositedinto
theflow.AsshowninFigure1,withaflatfacedcylinderbody,adecrementindragislargest
although its baseline drag is largest. Without energy deposition, a conical nose leads to a
smallestdrag,yettheefficiencyofenergydepositionisworst.Bytruncatingthecone,this
tradeoffproblemcanbesolved.Althoughthedragreductionperformancewithtruncatedcone
modelisweaker,itiseasilyaccomplishedthetargetdragforcebecauseitsbasedragforcehas
smallervaluecomparedtobluntbody.Inthiscase,theeffectivetime,forwhichtheeffectof
theenergydepositiononabluntbodylasts,isakeyparametertoknowwhatshapeofabody
issuitableforobtainingalowerdrag.
Sakai[8]reportedthattheeffectivetimeisevaluatedduringtheinteractionofthelowdensity
corecreatedbyasinglelaserpulseusinganNd:YAGlaserwiththebowshockwaveovera
bluntbody.Theevaluatedeffectivetimeonaflatfacedcylinderislongerthanonahemisphere
under the same energy deposition condition. The longer effective time for the flatfaced
cylinderisduetothefactthattherecirculationzonewithvortices,whichareproduceddueto
baroclinicinteraction,keepsforlongertimeintheforebodyregion.Thisbehaviorresultsin
thereductionofthedragforalongerperiod.Healsopresentedthatthemodulateddragfor
theflatfacedcylinderisnearlythesamewiththatforthehemisphereandthatthedragvalue
ishigherthanthatforasharpconewiththesamebasediameter.Itshouldbenotedthatthe
efficienciesfortheflatfacedcylinderaretypicallyhigherthanforthehemisphereunderthe
sameenergydepositioncondition.Thus,itisbelievedthataflatfacedgeometryhasapotential
advantagetobeusedinthedragreductionwithenergydeposition.
Figure 1. Trade-off in truncated cone model for improving the drag reduction performance.
Sakai [9] then proposed to employ a truncated cone and estimated its drag reduction per
formanceusingcomputationalfluiddynamicsmethodwithEulerequations.Inaccordance
withhisresults,whilethemagnitudeofdragforceisreducedwithfrontfaceareaoftruncat
edconemodeldecreasing,theefficiencyofenergydepositionishigherasthefrontfacearea
is increased. A quasisteady state flowfield over truncated conical geometry is established
with the higher repetitive frequency of pulse energy, typically higher than 50 kHz. In the
quasisteadystateflowfield,recirculationzonecomposedofseveralvorticesmakesavirtual
spikeinfrontofthetruncatedconebody.
Figure 2. Schematic and photographs of experimental apparatus.
273
274
Wind tunnel
wall
f = 60 mm
BK-7 window
45 Mirror
focus
15mm
f = -30 mm
Laser
f = 60 mm
45 Mirror
Figure 3. Schematic of laser optics system
2.2.Apparatus
Overall experimental apparatus is the same as of used in Ref. 6. The experimental facility
comprisesofasupersonicwindtunnel,laseropticsystemanddiagnosticsystem.Whilethe
supersonicwindtunnelisoperating,measurementsystemofstagnationpressure,dragforce
and visualization system is operated. Figure 2 shows the schematic and photographs of
apparatusincludingthesupersonicwindtunnel.Windowdiameterforvisualizationis90mm.
Dragreductionperformancewithconstantpulseenergyisestimatedasafunctionoflaser
frequency. Nd:YVO4 is used only to deposit the repetitive pulse energy. In our laser optic
system(Figure3),laserfrequencycanbeappliedupto50kHzunderallowablepulseenergy,
E,is7.2mJ.Laserpulsesupto8kHzisdepositedwithE=5.0mJ.
ThedragforceismeasuredbyusingforcebalancesystemintroducedinRef.5(Figure4).The
flowfiledoverthemodelisvisualizedviaschlierensystemincludingtwo300mmdia.concave
mirrorsandacircularknifeedge.Inasinglerunofthewindtunnelonehundredframesof
schlierenimagesarecapturedintoahighspeedframingcamera(ShimadzuHPV1)witha
framingintervalof4s,1/4ofwhichisanexposureperiod.Tomeasurethetimedependent
stagnationpressure,apiezoelectricpressuretransducer(H112A21,PCBInc.,risetimeof1s,
sensitivityof7.015mV/Pa)isflushmountedatheadofthemodel.
ThedataacquisitionsystemisshowninFigure5.
Figure 4. Drag measurement system and example of calibration.
275
276
Pressure
transducer 2
Pressure
transducer 1
Sub-chamber
Vacuum
pump
Laser
HPV-1
High Speed Camera
Delay
generator
Flash ramp
Oscilloscope 1
Signal
conditoner
Amplifier
Oscilloscope 2
Function
generator
Figure 5. Data acquisition system.
2.3.Dragreductionperformanceofflatfacedtruncatedconemodel
First,thedragreductionperformancewithflatfacedtruncatedconicalbodyisinvestigated.
Figures6showstheschematicillustrationoftruncatedconemodelsusedintheexperiments.
Thebasediameterofbody,d,is20mm.Thefrontfacediameterisdefinedasdf.Thediameter
ratio, df/d, is varied from 1.0 (cylinder model) to 0(cone model) with half apex angle of 15
degree.Thestagnationpressureismeasuredonlyon3types(df/d=1.0,0.75and0.5)flatfaced
truncatedconmodel,becausediameterofpressuretransducer,5.56mm,iscomparabletothe
nosedimension.
2.25d
df / d = 1.0
df
df / d = 0.75
df / d = 0.5
df / d = 0
Figure 6. Truncated cone models for improving the drag reduction performance.
df/ d
E [mJ]
f [kHz]
1.0
5.0
10 ~ 80
0.75
5.0
10 ~ 80
0.75
7.2
10 ~ 50
0.5
5.0
10 ~ 80
0.5
7.2
10 ~ 50
5.0
10 ~ 80
7.2
10 ~ 50
0.5
5.0
10 ~ 80
0.5
7.2
10 ~ 50
0.5
5.0
10 ~ 80
0.5
7.2
10 ~ 50
Shape of front face
Flat face
Concave face
Table 1. Experimental conditions.
277
278
(a) df/d=1.0
(b) df/d=0.75
(c) df/d=0.5
(d) df/d=0
Figure 7. Shock layer without laser pulses.
Experimental conditions are shown in Table 1. Drag reduction performance with constant
pulse energy is estimated as a function of laser frequency. In our laser optic system, laser
frequencycanbeappliedupto50kHzunderallowablepulseenergy,E,is7.2mJ.Ontheother
hand,laserpulsesupto80kHzisdepositedwithE=5.0mJ.
Fortheabsenceoflaserpulses,theshocklayeroverthetruncatedconemodeliscompared
withadifferentdf/dvalueinFigure7.Althoughallofshocklayerhavebowshockshape
except for the case of the conical model(df / d = 0.0), those shock standoff distances are
decreasedwithdecreasingfrontfacearea;shockstandoffdistanceis0.45dfordf/d=1.0,
0.31dfordf/d=0.75and0.25dfordf/d=0.5.
Figures8presentsschlierenimageswithlaserpulseenergydepositions(f=80kHz,E=5.0mJ).
Withenergydepositions,theeffectiveapexangleofdistortedshocklayerbecomessmaller
withdf/dincreasing.Inparticular,shocklayershapeofdf/d=1.0issimilartoobliqueshock.As
the residence timeofvortex rings is longer, the virtualspike composedofseveral vortices
becomesmoresharply.Forthedf/d=0.0,thebaroclinicalvortexringisnotobservedbecause
laserheatedgasinteractswithattachedobliqueshockwave.
The effective residence time of vortex ring [10, 11] can be seen from stagnation pressure
historiesinFigure9.Withdf/d=1.0,pulsetopulseinteractionissignificantatf=10kHz.Hence,
stagnationpressurehistoryshowsthealmostquasisteadystatebehavior.However,stagna
tionpressuredecrementcausedbyvortexringcanbefoundfordf/d=0.75.Inthecaseofdf/
d=0.5,stagnationpressureisalmostrecoveredintotheformerstate,andthenaffectedbyblast
wave. Even if pulsetopulse interaction is somewhat occurred, the effect on stagnation
pressureisveryweak.
Figure10showsexamplesofthetimevariationofthedrag.Figures11and12showthedrag
reductionperformanceofflatfacedtruncatedconemodelasafunctionoff.Whenthepulse
energyof5.0mJisdeposited,dragforcealmostlinearlyreduceswithf.Fordf/d=1.0,D/D0
isobtainedupto21%,andtheefficiencyofenergydepositionisnearlyconstantvalueofabout
7.Underthesamecondition,amountofdragreductionandefficiencyofenergydeposition
aredecreasedwithdecreasingofdf/d.WithE=5.0mJ,propulsionenergysavingisnotrealized
ifdf/dissmallerthan0.5.
(a) df / d = 1.0
(c) df / d = 0.5
(b) df / d = 0.75
Figure 8. Instantaneous schlieren images at f= 80kHz, E=5.0mJ.
(d) df / d = 0.0
279
1.1
pst / pst,0
1.0
0.9
100 s
0.8
t
(a)df/d=1.0
pst / pst,0
1.1
1.0
0.9
100 s
0.8
t
1.1
pst / pst,0
280
(b)df/d=0.75
1.0
0.9
100 s
0.8
t
(c)df/d=0.5
Figure 9. Stagnation pressure histories, f=10kHz, E=5.0mJ.
1.1
Nd:YLF Laser (E=6.6 mJ )
Nd:YVO4 Laser (E=6.2mJ)
1.0
D / D0
f = 1kHz
0.9
6kHz
8kHz
2kHz
4kHz
25kHz
10kHz
50kHz
0.8
Laser irradiation
0.7
-1
t[s]
Figure 10. Example of drag variation
10
E=5.0mJ
df / d = 0.0
df / d = 0.5
df / d = 0.75
df / d = 1.0
20
40
60
80
100
f [ kHz ]
Figure 11. Efficiency of energy deposition of flat-faced truncated cone model, E=5.0mJ/pulse.
281
2.0
E=5.0mJ
df / d =1.0
1.5
CD
282
0.75
1.0
0.50
0.0
0.5
0.0
20
40
60
80
100
f [ kHz ]
Figure 12. Variation of drag coefficient with df/d, E=5.0mJ/pulse.
The drag coefficient, CD, is plotted against f in Figure 12. As repetitive laser frequency is
increased,thedragcoefficientisdecreased.Withoutenergydepositions,dragcoefficientofdf/
d=1.0is1.8.Althoughthedragcoefficientofdf/d=1.0isdecreaseddownto1.46withf=80kHz
andE=5.0mJ,thatisstillhigherthanthebasedrag(=0.55)ofconicalmodel(df/d=0.0).
2.4.Dragreductionperformanceofconcavefacedtruncatedconemodel
Inordertoimprovethedragreductionperformance,experimentalstudiesareconductedon
concavefacedtruncatedconemodelswithdf/d=0.5asseeninFigure13.Theradiuscurvatures
offrontfaceareR/d=0.5and1.0,respectively.Alloftruncatedconemodelshavesamelength,
andlocationofdepositingpulseenergyis2daheadofthemodel.
df / d = 0.5
R / d = 0.5
df / d = 0.5
R / d = 1.0
Figure 13. Schematic diagram of concave-faced truncated cone model with df/d = 0.5.
3
E=5.0mJ, df / d = 0.5
Flat face
Concave face (R/d=1.0)
Concave face (R/d=0.5)
20
40
60
80
100
f [ kHz ]
Figure 14. Power gain of concave-faced truncated cone model, df/d=0.50, E=5.0mJ/pulse.
Fromtheaboveresults,itwasconcludedthatdragreductionperformanceoftruncatedcone
becomepoorsincetheeffectiveresidencetimeofvorticesbecomeshorterwithdf/ddecreasing.
Therefore,concavefacedtruncatedconeshapeisconsideredtoimprovethedragreduction
performanceoftruncatedconeinthissection.Truncatedconemodelwithdf/d=0.5isused,
andradiuscurvature(R/d)ofconcavefaceisvariedfrom0.5to1.0.
Figure14presentstheeffectofconcaveradiuscurvatureonthedragreductionperformance.
ItisinterestingthatD/D0withconcavefacedtruncatedconebecomeshighercomparingwith
flatfacedtruncatedcone.D/D0isincreasedwithR/d=1.0forE=5.0mJandf=80kHz.Asshown
in Figure 10, the efficiency of energy deposition is slightly enhanced with concavefaced
truncatedcone.Thisimpliesthatconcavefacedtruncatedconeisusefultoimprovethedrag
reductionperformance.Fromtheseresults,itisconfirmedthatdragcoefficientofconcave
facedtruncatedconeisslightlydecreased.InthecaseofR/d=0.5,dragreductionperformance
isalmostsamewithoneofR/d=1.0.However,dragcoefficientofR/d=0.5becomeshigherthan
thatofR/d=1.0becausesmallradiuscurvatureleadstoincreaseofbasedragforce.
2.5.Summaryofthischapter
Although a blunt body shape brings the better drag reduction performance due to energy
depositions, a truncated cone shape has considerable advantage to satisfy the necessary
283
284
conditions in actual application of energy deposition scheme; the magnitude of drag force
shouldbelowerthanthebasedragforceofasharpconicalbodyandtheefficiencyofenergy
depositionsshouldbehigherthanunity.Fromthesedemands,dragreductionperformance
overtruncatedconemodelwasexperimentallyestimated.
Intheexperiments,atruncatedconemodelwithhalfangleof15degreeisused.Thediameter
ratioofflatfacedtruncatedconeisvariedfrom1.0to0.0.Fromthepresentresultsonflatfaced
truncatedcone,theeffectiveresidencetimeofvorticesresultedbylaserheatedgasesinterac
tionwithshockwaveplaysanimportantroleindragreductionperformanceoftruncatedcone
model.Inordertoimprovethedragreductionperformanceofthetruncatedcones,thedrag
reductionperformanceofaconcavefacedtruncatedconemodelisevaluated.Intheconcave
facedtruncatedconeexperiments,theradiuscurvatureofconcavefaceisvariedfrom0.5to
1.0withdiameterratioof0.5.Fromthecomparisonwithflatfacedtruncatedcone,theconcave
facedtruncatedconehasmoreeffectivedragreductionperformance.
3.Conclusion
Asisdemonstratedinthischapter,micronanotechnologyisapplicablealsotomechanical
manipulation in the temporal manner. An example of micronanosecond mechanics is
demonstratedbyrepetitivelydepositingnanosecondlaserpulsesoverasupersonicobject.
Theseshortperiodenergydepositioninducenonlinearflowmechanics,therebyimproving
the supersonic aerodynamics performance. It follows from these results that miconano
technologyshouldrefertospatiotemporalmechatronics,whichwillleadtofurthertechno
logicaladvancements.
Acknowledgements
TheauthorswouldliketothankMessrs.A.Saito,N.ShirakiandK.Kumazawa,Technical
Division, Nagoya University for their valuable technical assistances. This research was
supportedbyJapanSocietyforPromotionofScienceasGrantinAidforScientificResearch
(S),No.22226014.
Authordetails
AkihiroSasoh
DepartmentofAerospaceEngineering,NagoyaUniversity,Nagoya,Japan
References
[1] KnightD.SurveyofAerodynamicDragReductionatHighSpeedbyEnergyDeposi
tion.JournalofPropulsionandPower2008;2411531167.
[2] TretyakovP.K,GaraninA.F,GrachevG.N,KrainevV.L,PonomarenkoA.G,Tish
chenkoV.N,YakovlevV.I.ControlofSupersonicFlowaroundBodiesbyMeansof
HighPowerRecurrentOpticalBreakdown.PhysicsDoklady1996;41566567.
[3] Sasoh A. Is Fly By Light Power possible ? In: Proceedings of 46th Autumn Joint
ConferenceofKansai&ChubuBranch,JapanSocietyforAeronauticalandSpaceSci
ences(inJapanese);2009.
[4] Takaki R, Liou M.S. Parametric Study of Heat Release Preceding a Blunt Body in
HypersonicFlow.AIAAJournal2002;40501509.
[5] SasohA,SekiyaY,SakaiT,KimJH,MatsudaA.WaveDragReductionoveraBlunt
NosewithRepetitiveLaserEnergyDepositions.AIAAJournal2010;4828112817.
[6] JaeHyungKimJ.H,MatsudaA,SakaiT,SasohA.WaveDragReductionwithAct
ingSpikeInducedbyLaserPulseEnergyDepositions,AIAAJournal;inpress.
[7] AdelgrenR.A,YanH,ElliottG.S,KnightD.D,BeutnerT.J,ZheltovodovA.A.Con
trol of Edney IV Interaction by Pulsed Laser Energy Deposition. AIAA Journal
2005;43256269.
[8] SakaiT,SekiyaY,MoriK,SasohA.InteractionBetweenLaserInducedPlasmaand
ShockWaveOveraBluntBodyinasupersonicflow.In:ProceedingsofInstitutionof
MechanicalEngineersJournalofAerospaceEngineeringPartG2008;222605617.
[9] Sakai T. Supersonic Drag Performance of Truncated Cones with Repetitive Energy
Depositions.InternationalJournalofAerospaceInnovation2009;13143.
[10] OginoY,OhishiN,TaguchiS,SawadaK.BaroclinicVortexInfluenceonWaveDrag
ReductioninducedbyPulseEnergyDeposition.PhysicsofFluids2009;21066102.
[11] Kim J.H, Matsuda A, Sasoh A. Formation of a virtual spike builtup with vortex
ringsgeneratedbyrepetitiveenergydepositionsoverabowshocklayer.Physicsof
Fluids2011;23(1)021703.
285
Chapter 14
Hydroxyapatite Coating on Titanium Implants Using

Hydroprocessing and Evaluation of Their
Osteoconductivity
Kensuke Kuroda and Masazumi Okido
1.Introduction
Titanium(Ti)anditsalloysareusedasartificialjointsandteethrootsinorthopedicanddental
settingsbecausetheyhavetheadvantagethattheirmechanicalpropertiesareclosertothose
ofbonethanarethoseofstainlesssteelorcobaltchromiumalloys.However,thedifference
inmechanicalpropertiesbetweenTiandnaturalboneleadstonegativeeffects,suchasstress
shielding.Tomitigatetheseeffects,manynewTialloyshavebeendevelopedforhardtissue
implants, with a focus on controlling the alloy element and its content, phase, and other
characteristics.
Whenimplantsdonotundergosurfacemodificationtoenhancetheosteoconductivity,ittakes
arelativelylongtimetofixthemetallicimplanttobonesuchthatitisstable.Therearemany
approachesforimprovingtheosteoconductivityofTianditsalloys.Theseapproachescanbe
classified into the following two techniques: (1) bioactive compounds that accelerate bone
formationarecoatedonmetallicimplants,and(2)aroughsurfaceatthemacrolevelisformed
onthemetallicimplants,andtheingrowthofboneresultsinanchorageoftheimplants.These
techniqueshaveachievedacertainlevelofsuccess,andthesurfacemodifiedimplantshave
beenusedclinically.However,therearestillweaknesseswiththecoatingthatneedresolution,
aswellasunclearpointsregardingtheeffectofthesurfacepropertiesontheosteoconductivity.
Sincehydroprocessingcanbeusedtopreparethecoatingoncomplexshapedsubstrateswith
complextopography,whichmanyimplantshave,wefocusontheuseofhydroprocessingin
manytechniquesforcoatingthebioactivecompound,especiallyhydroxyapatite(HAp),and
expoundonthecharacteristicsofthetechniquesandissues.Moreover,wedescribeindetail
theevaluationoftheosteoconductivityofimplantscoatedwithHAp,usinginvivotestingin
rattibiae.
2013 Kuroda and Okido; licensee InTech. This is an open access article distributed under the terms of the
288
2.HApcoating
HAp(Ca10(PO4)6(OH)2),whichisthemaininorganiccomponentinthemammalboneortooth
[1],hasattractedattentionasasurfacecoatingcompoundbecauseofitshighosteoconductiv
ity.ManypyromethodsofformingHApandothercalciumphosphatecoatingsonmetallic
substrateshavebeenreported(e.g.,plasmaspraying[2,3],solgelmethod[4,5],electronbeam
sputteringmethod[6],andionbeamsputteringmethod[7]).However,allhaveweakpoints
inrelationtocoatingwithHAponcomplexshapedimplants.Plasmasprayingremainsthe
mostcommonlyusedtechniqueforHApcoatingonaTiorTialloysubstrateinthefabrication
ofartificialjointreplacements[1]andinendosseousdentalimplants[2].Ontheotherhand,
many hydro coating techniques (e.g., cathodic electrolysis method [810], electrophoretic
method[11,12],andthermalsubstratemethod[1318])havebeenproposedasapproachesto
forming thin film coatings on metallic substrates. The cathodic electrolysis and thermal
substratemethodsaresinglestepcoatingtechniquesinanaqueoussolution,andtheycoatthe
HAp directly from the solution. The electrophoretic method is omitted from this review
because it uses HAp formed by other methods in advance, despite the hydroprocessing.
Therefore,inthispaper,wedescribethecathodicelectrolysisandthermalsubstratemethods.
2.1.TheoryofHApcoatingusinghydroprocessing
It is known that the solubility of HAp in an aqueous solution decreases with increasing
temperatureandthattherelationshipbetweentheHApsolubilityproduct,KSP/(moldm3)9,
andthetemperature,T/K,isgivenby[19]:
logKSP 8219.41 / T 1.6657 0.098215T .
(1)
Therefore,heatinganaqueoussolutioncontainingCa2+andPO43ionsresultsintheprecipi
tationofcalciumphosphates,suchasHAp,inthesolution.
TheionicproductofHAp,KIP/(molL1)9,isexpressedasfollows:
5
K IP Ca 2+ PO 4 3 OH ,
(2)
where[X]indicatesthemolarconcentration(molL1)ofionicspeciesX.Theincreasein[Ca2+]
or[PO43]contentorpHvalueinthesolutioninitiatestheprecipitationofHApbecauseKIP
achievesKSP.Moreover,[PO43]increaseswithincreasingpHvalue(Figure1).Therefore,the
increaseinpHdirectlyacceleratestheprecipitationofHAp,whichindirectlyincreasesthe
[PO43]content.
Figure2showsthesolubilitycurvesofvariouscompoundsonacalciumorthophosphate[19];
as shown, there are many compounds other than HAp. This figure indicates that CaHPO4
(DCPA)isthemoststablecompoundatpH<5,withHApthemoststableatpH>5.Therefore,
Hydroxyapatite Coating on Titanium Implants Using Hydroprocessing and Evaluation of their Osteoconductivity
Content,log(C HxPO4(3x)/molL1)
HAp can be easily obtained in a solution where pH > 5 and where the ion content and
temperature are controlled. However, HAp cannot precipitate in the pH < 5 solution, and
hydroprocessingusingtheprecipitationphenomenonintheaqueoussolutioncannotgive
Ca3(PO4)2(TCP),abioactivecompound.
0
-2
H3PO4
H2PO4-
HPO42-
PO43-
-4
-6
-8
-10
0
10
12
14
pH
Calciumcontent,log(CCa/molL1)
Figure 1. Logarithmic concentration diagram for orthophosphoric acid.
0
-1
DCPD
-2
-3
DCPA
-4
OCP
HAp
-5
TCP
10
pH
Figure 2. Solubility curves of calcium orthophosphoric compounds at 37 oC, depending on pH in aqueous solution.
HAp: hydroxyapatite (Ca10(PO4)6(OH)2), TCP: calcium phosphate (Ca3(PO4)2), OCP: octacalcium phosphate (Ca8H2(PO4)6
5H2O), DCPA: dicalcium phospate anhydrous (CaHPO4), DCPD: dicalcium phospate dihydrate (CaHPO4 2H2O).
289
290
2.2.Thermalsubstratemethodinaqueoussolution[13]
Thisprocessinvolvespassinganalternatingcurrentthroughametallicsampleimmersedin
anaqueoussolution.Theimmersedmetallicsampleheatsuptomorethan100CbyJoule
heating,eventhoughthehydroprocessingoccursatatmosphericpressure(Figure3).There
fore,thismethodcanproducethespecialreactionconditions(>100Cinanaqueoussolution)
bynotusingthepressurevessel.
Thermo-meter
AC power supply
Powerstat
Ammeter
Coolant
TC supporter
Thermocouple
Copper lead rod
(coated with epoxy resin)
Cooling tube
Solution (0.2 dm33)
Pyrex beaker
Ti sample (in vivo)
(2x5 mm)
Ti sample (in vitro)
(t0.3 mm)
Figure 3. Experimental apparatus for HAp coating.
Whenthethermalsubstratemethodinanaqueoussolutionisused,thefactthatthesolubility
ofHApdecreaseswithincreasingtemperaturemeansthatHApprecipitationoccursonlyon
thesubstrate.ItisimportanttocoatHApwhilecontrollingtheconcentrationofthesoluteand
thepHvalueofthesolutionandtemperature,becausetheyaffectthedegreeofsupersaturation
of HAp in the solution (Eqs (1) and (2)). Figure 4 (a)(d) shows the change in the surface
morphologyofthesamplescoatedwithHApundercontrolledpHandtemperature,whose
foctorsdeterminedthedegreeofthesupersaturationwithrespecttoHAp[14,15,17,18].The
precipitateatpH=4.0(Figure4(a))appearedtopileuplikebricksandwasidentifiedasDCPA,
whichisastablecompound.Ontheotherhand,inthesolutionatpH=8.0,theprecipitatewas
Figure 4. SEM photographs of the surface of the samples treated by various methods.
composedofHAp(Figure4(b)(d)).FromtheEDXanalysis,themolarratioofcalciumto
phosphorous(Ca/P)ofHApwas1.411.43.ThisshowsthecoatedHApwascalciumdeficient.
ThepHdependenceofthesolubilityofthecalciumphosphatecompoundsexplainswhythe
precipitatechangedwithincreasingpHofthesolution,i.e.,thesolubilitycurvesofDCPAand
HApcrossatapproximatelypH=5forvariouscompoundsofcalciumphosphate[19].The
surfacemorphologyoftheprecipitatedHApstronglydependsoncoatingtemperature:low
temperature(40C)gavenetlikeHAp(Figure4(b));hightemperature(140C)gaveneedle
likeHAp(Figure4(d));andmidtemperature(60C)gaveplatelikeHAp(Figure4(c)).That
is,byusinghydroprocessing,wecancontrolthecrystallineform,whichcouldnothavebeen
achievedusingtraditionalmethods.Figure5showsthescanningelectronmicroscopy(SEM)
photographs of the HApcoated samples on porous Ti alloy surfaces formed by sintering
Ti6Al4Vparticles(ca.100mindiameter)oncpTisubstrates[16].Heatingat100Cfor15
min.inapH=7solutionledtoHApprecipitationovertheentiresurfaceoftheTi6Al4Vsintered
particles(onbothfrontandbackfaces)andonthebasecpTisubstrateoftheexperimental
samples.Inparticular,itwasfoundthatHApprecipitatewasalsodetectedatthesinterneck
regionsofadjacentparticlesandonthebasesubstrate,whiletheoriginalopenporedgeometry
wasmaintained.Therefore,thismethodcanbeusedtoapplytheHApcoatingtoasubstrate
withcomplextopography.
291
292
(a)
(b)
(c)
100 m
0.5mm
(d)
50 m
(e)
100 m
50 m
Figure 5. and/or cross-sectional views of the samples with surface roughness. (a)-(c) beads-sintered porous samples
(as sintered), (d)(e) HAp coated on beads-sintered porous samples (thermal substrate method, 0.7 mM CaCl2+0.3 mM
Ca(H2PO4)2, pH 7, 100 oC, 15 min.).
BiologicalapatiteinnaturalbonedoesnotappearintheformofpureHApanditcontainsa
considerableamountofcarbonateions[20](about7.4mass%withrespecttototalboneand
11.4mass%withrespecttotheinorganiccomponentinnaturalbone[21]).Carbonateapatite
(CO3Ap),whichreplacesPO43and/orOHionswithCO32ions,issimilartotheinorganic
componentofbone,anditseemstobeamorepromisingbioactivematerialthanstoichiometric
HAp, because CO3Ap has greater solubility than pure HAp [20]. In addition, it has been
reportedthatCaCO3displaysbioactivity,suchascellcompatibilityandhardtissuecompati
bility[22,23];thatis,CO 32isexpectedtoinfluencebiologicalreactivityandosteoconductive
properties.ItisalsowellknownthatthesolubilityofCaCO3inanaqueoussolutiondecreases
withincreasingtemperature[24].Inthesolution,whenCO32ionsareadded,CO3AporCO3
Ap/CaCO3compositefilmsareeasilyobtainedonasubstrate.TypicalSEMphotographsofthe
surfaceofthesamplesareshowninFigure4(e)(f),coatedinthepH=8solutionwith<0.5
mMNaHCO3addedat140Cforaperiodof15min.,andafterthesteamautoclavingtreatment
(5mass%CO3inthisfilm)[25,26].Theprecipitatescoatedfromthesolutionwith>0.5mM
NaHCO3addedcontainedCO3ApandCaCO3atalltemperatures,andtheXraydiffraction
spectrashowedamixtureofcalcite,vaterite,andaragonite.ThecrystallineformofCO3Ap
waschanged,dependingontheaddedNaHCO3content,aswellascoatingtemperature.In
particular,addingasignificantamountofNaHCO3(>5mM)broughtaboutspherelikeshaped
CO3Ap(Figure4(f))inthe140Ccoating.InCO3Apfilms,FTIRanalysisrevealedthat
CO32wassubstitutedforPO43(TypeBCO3Ap)inadvance,whichwassimilartobiological
apatite[27],andaddingmoreCO32tothesolutiongavethesubstitutionforOH(TypeA).
Therefore,inthesampleswith<0.5mMNaHCO3added,TypeBCO3Apwasobtained,and
inthesampleswith>5mMNaHCO3added(i.e.,havingthebinaryphaseofCO3Ap/CaCO3),
TypeABCO3Apwasformed.
NaturalbonecontainsCO3Apandaconsiderableamountoforganiccomponents,suchas
collagen(about23mass%[21]).Itisknownthatthehybridorganicinorganicstructureinitiates
pliable bone. Some researchers have reported the preparation of nanocomposites of HAp/
collagenandHAp/gelatin[2830],asnaturalboneisconsideredananocompositeofmineral
andproteins.Moreover, immobilization ofcollagen onimplants displaysa tighter fixation
with the surrounding tissue, since the collagen behaves as an adhesive protein with cells
because of the amino groups in the collagen molecules [31, 32]. From the viewpoint of
osteoconductivity,weexpectedthatpreparingtheHAp/collagencompositecoatingwouldbe
amorepromisingapproachthanusinganindividualcoatingofeitherCO3AporHAp.Inthe
solution to which acidsoluble collagen is added, HAp/collagen or HAp/gelatin composite
films are easily obtained on a substrate, depending on coating temperature. In general, as
mammaliancollagenrapidlydenaturestogelatinat>45C,HAp/collagencompositecanbe
obtainedat<40CandHAp/gelatincompositeat>50C.Figure4(g)(h)showsthesurfaceof
thesamplescoatedinthepH=8solutionwith72mgL1collagen,derivedfromcalf,at140C
and40C(1015mass%collagenorgelatininthefilm)[33].ThesurfacemorphologiesofHAp/
collagenandHAp/gelatinsignificantlydependonthecoatingtemperatureandarenotaffected
bywhetherthecompositefilmcontainscollagenorgelatin.Thatis,collagenandgelatinhave
onlyasmalleffectontheHApcrystalgrowthoftheadsorptionontoHAp.Hydroprocessing
can be used to form HAp/collagen and HAp/gelatin composite films, which could not be
formedusinghightemperatureprocessing,andthecontentofcollagenandgelatininthefilms
canbecontrolledupto60mass%.
2.3.Cathodicelectrolysisinaqueoussolution[810]
Intheelectrochemicaltechnique,aredoxreactionproducessupersaturationofOHionsnear
theelectrodeintheaqueoussolutioncontainingCa2+andPO43inthesamemannerasinthe
thermal substrate method. This local effect induces heterogeneous nucleation on the metal
substrateservingastheelectrode.Theadditionofhydrogenperoxidetothesolutionprevents
H2gasgenerationatthecathodicelectrodeandpromotesnucleationandgrowthoftheHAp
coating. Adding H2O2 to electrolytes enhances the formation of OH ions at the solution
electrodeinterfaceatalowercathodicpotential,asdescribedinthefollowingreaction(3)[10]:
H 2O 2 +2e =2OH .
(3)
Inthismethod,thesurfacemorphologyoftheprecipitatedHApgreatlydependsoncoating
temperature [34] in the same manner as in the thermal substrate method. The effect of
temperatureonthesurfacemorphologyofcoatedsamplesisshowninFigure4(i)(j).The
HAp crystals had a similar shape to those formed using the thermal substrate method,
althoughthesizeofHApcrystalsdifferedbetweenthecathodicelectrolysisandthethermal
substratemethods.Themolarratio(Ca/P)ofHApwasalmostsameasthatusingthethermal
substratemethod.Thecoatingsat>100Cwereconductedinthepressurevessel.Whenusing
the electrolysis solution, to which CO32 or collagen are added, CO3Ap, HAp/collagen, or
HAp/gelatincompositefilmsareformedonasubstrate,dependingoncoatingtemperature.
293
3.Evaluationofosteoconductivity
Theevaluationmethodsforthebioactivityoftheimplantsareclassifiedintoinvitroandin
vivomethods.Inthisreview,theinvivoevaluationmethodisdescribed.Ininvivoevaluation,
manytypesofanimalsatdifferentageswereusedinvariousstudies,anddifferentresearchers
usedadifferentimplantedpartoftheanimals.Moreover,aunifiedquantificationcriterionhas
notyetbeenestablished,andthecriteriausedinvariousstudiesarenotcompatiblewithone
another.Therefore,weusetheboneimplantcontactratio,RBI,asanosteoconductiveindex
basedontheobservationofbodytissueontheimplants.Boneimplantcontactwasdetermined
bylinearmeasurementofdirectbonecontactwiththeimplantsurface.Thesumofthelength
oftheboneformationontheimplantsurfacewasmeasuredandexpressedasapercentageof
thetotalimplantlength(boneimplantcontactratio)inthecancellousboneandthecortical
boneparts[17,18,25,26,33].
sumofthelengthofthepartofboneformationontheimplantsurface
100
totalimplantlength
RBI (%)
(A-1)
n=6 6
(1)
(2)
(B-1)
66 6
(C-1)
64 5
(D-1)
65
46
(E)
4
(A-2)
6
(B-2)
6 6
(4)
(C-2)
6 6
80
BoneImplantcontactratio,RBI (%)
294
60
* *
40
20
0
60
*
* *
innaturalbone
40
20
0

0 10 20 0 10 20 0 10 20 0 10 20 30
0
20
40 0
20
40
20
40 0
TotalCO3 contentinthecoating,CCO3/mass%
Collagenorgelatincontentinthecoating,Cc,gF/mass%
Figure 6. Bone-implant contact ratio, RB-I, for the various surface coated samples (rats tibia, 14 days). *p < 0.05, (1)
cortical bone part, (2) cancellous bone part. (A) needle-like, (B) plate-like, (C) net-like, (D) spherical-like, (E) as-polished
Ti. Hap CO3Ap (A-1, B-1, C-1) or HAp/gel. composite (A-2, B-2) CO3Ap/CaCO3 composite (A-1, B-1, C-1, D-1) or
HAp/col. composite (C-2)
Figure6showstheboneimplantcontactratios,RBI,ofthesamplescoatedunderthevarious
conditionsmentionedaboveandclassifiedbasedonthefollowingfoursurfacemorphologies:
(A)needlelike,(B)platelike,(C)netlike,and(D)spherelike.Thesamplesarethencompared
withthecontrolimplant((E)noncoatedTi).InFig.6,thesamplesaredistinguishedaccording
tocolorbasedonwhetherornotthecoatingcontainedCaCO3andcollagenorgelatin(white:
HAp;gray:CO3AporHAp/gelatin;black:CO3Ap/CaCO3orHAp/collagen).TheRBIvalue
ofHApcoatedsamples(whitebar)isthesameasorhigherthanthatoftheaspolishedone
(E). In particular, RBI in the cancellous bone part is highest in the sample coated with the
needlelikeHAp(A1).TheinfluenceofthedifferentsurfacemorphologiesonRBIisapparent
[17,18].AsmallamountofCO3includedinCO3Apdoesnotinfluenceosteoconductivity,and
anincreasedamountofCO3(>15mass%CO3),includingthatinCO3Ap/CaCO3,hasanegative
effecton(blackbarin(A1),(B1),(C1),and(D1))[25,26].TheRBIvalueofHAp/gelatin
coatedsamplesisthesameasthatofHAp(grayandwhitebarsin(A2)and(B2)),andwe
didnotfindapositiveeffectoftheadditiontoHApontheosteoconductivity,oranynegative
effectswithinthelimitofgelatincontentused.IntheHAp/collagenfilms(C2),osteoconduc
tivitywasimproved,andmaximumRBIwasobtainedwhenthecollagencontentwasthesame
asthatinnaturalbone.Theadditionoftoomuchcollagen,exceedingthatamountofcollagen
contentinnaturalbone,inhibitedtheimprovementoftheosteoconductivity[33].
4.Conclusion
Theinsideofthehumanbodyisequivalenttoawaterenvironmentatroomtemperature,since
thewatercontentinthebodyisabout60%.ItisthoughtthathydroformedHAphasgreater
osteoconductivitythanHApsynthesizedusingpyroprocessing,becausesynthesizedHApin
theaqueoussolutionatneutralpHandroomtemperatureissimilartothatformedinthebody.
Inaddition,titaniumdioxide,TiO2,whichdoesnotexistinthehumanbody,isaremarkable
compoundwithrespecttoitsosteoconductivity.Itisimportanttoresearchandimprovethe
osteoconductivityofsubstancessuchasHAp,TiO2,andCaTiO3.However,weneedtopay
attentiontothepropertiesoftheircompounds,suchassurfaceroughness[35],crystallinity,
andcorrosivity,allofwhichinfluenceosteoconductivity.Furthermore,theevaluationcriterion
forosteoconductivityhasnotbeenadequatelyestablished.
Thedevelopmentofimplantswithhighfunctionalityisanimportantproblemthaturgently
needstobesolved,insteadofmerelymakingprogressinmedicaltechnology.Itisthoughtthat
nothingcancompetewithsuchimplantsintheprogressanddevelopmentoftheindividual
technology.Wehopethattheseimportantproblemscanbesolvedusingthecombinationof
thediscoveryofnewbioactivecompounds(organicandinorganic)andtheircoatingtechni
ques,alloydesignsfortheimplants,and/orthegrowthofrelatedsurroundingtechniquesfor
them.
295
296
Authordetails
KensukeKurodaandMasazumiOkido
DepartmentofMaterialsScienceandEngineering,NagoyaUniversity,Nagoya,Japan
References
[1] Hench L. L, Wilson J. An introduction to bioceramics. Singapore, World Scientific
Publishing;1993.
[2] PilliarR.M,DeporterD.A,WatsonP.A,PharoahM,ChipmanM,ValiquetteN,Car
terS,DeGrootK.TheEffectofPartialCoatingwithHydroxyapatiteonBoneRemod
eling in Relation to PorousCoated TitaniumAlloy Dental Implants in the Dog.
JournalofDentalResearch1991;70(10)13381345.
[3] KwehS.W.K,KhorX.A,CheangP.Plasmasprayedhydroxyapatite(HA)coatings
with flamespheroidized feedstock: microstructure and mechanical properties. Bio
materials2000;21(12)12231234.
[4] Mavis B, Ta A. C. Dip coating of calcium hydroxyapatite on Ti6Al4V substrate.
JournalofAmericanCeramicSociety2000;83(4)989991.
[5] LangstaffS,SayerM,SmithT.J.N,PughS.M,HespS.A.M,ThompsonW.T.Re
sorbablebioceramicsbasedonstabilizedcalciumphosphates.PartI:rationaldesign,
sample preparation and material characterization. Biomaterials 1999;20(18)
17271741.
[6] KimD.H,KongY.M,LeeS.H,LeeI.S,KimH.E,HeoS.J,KoakJ.Y.Com
positionandcrystallizationofhydroxyapatitecoatinglayerformedbyelectronbeam
deposition.JournalofAmericanCeramicSociety2003;86(1)186188.
[7] ChenT.S,LacefieldW.R.Crystallizationofionbeamdepositedcalciumphosphate
coatings.JournalofMaterialsResearch1994;9(5)12841290.
[8] Momma H. Thin hydroxyapatite layers formed on porous titanium using electro
chemicalandhydrothermalreaction.JournalofMaterialsScience1994;29(4)949953.
[9] OkidoM,KurodaK,IshikawaM,IchinoR,TakaiO.Hydroxyapatitecoatingontita
niumbymeansofthermalsubstratemethodinaqueoussolutions.SolidStateIonics
2002;151(14)4257.
[10] OkidoM,NishikawaK,KurodaK,IchinoR,ZhaoZ,TakaiO.Evaluationofthehy
droxyapatite film coating on titanium cathode by QCM. Materials Transactions
2002;43(12)30103014.
[11] Nie X, Leyland A, Matthews A. Deposition of layered bioceramic hydroxyapatite/

TiO2coatingsontitaniumalloysusingahybridtechniqueofmicroarcoxidationand
electrophoresis.SurfaceandCoatingsTechnology2000;125(13)407414.
[12] Sena L. A, Andrade M. C, Rossi A. M, Soares G. A. Hydroxyapatite deposition by
electrophoresis on titanium sheets with different surface finishing. Journal of Bio
medicalMaterialsResearch2002;60(1)17.
[13] KurodaK,IchinoR,OkidoM,TakaiO.Hydroxyapatitecoatingontitaniumbyther
malsubstratemethodinaqueoussolution.JournalofBiomedicalMaterialsResearch
2002;59(2)390397.
[14] KurodaK,IchinoR,OkidoM,TakaiO.EffectsofionconcentrationandpHonhy
droxyapatite deposition from aqueous solution onto titanium by the thermal sub
stratemethod.JournalofBiomedicalMaterialsResearch2002;61(3)354359.
[15] Kuroda K, Miyashita Y, Ichino R, Okido M, Takai O. Preparation of calcium phos
phatecoatingontitaniumbythethermalsubstratemethodandtheirinvitroevalua
tion.MaterialsTransactions2002;43(2)30153019.
[16] KurodaK,NakamotoS,IchinoR,OkidoM,PilliarR.M.Hydroxyapatitecoatingson
a 3D porous surface using thermal substrate method. Materials Transactions
2005;46(7)16331635.
[17] Kuroda K, Nakamoto S, Miyashita Y, Ichino R, Okido M. Osteoinductivity of HAp
filmswithdifferentsurfacemorphologycoatedbythethermalsubstratemethodin
aqueoussolutions.MaterialsTransactions2006;47(5)19311934.
[18] Kuroda K, Nakamoto S, Miyashita Y, Ichino R, Okido M. Osteoinductivity of HAp
filmswithdifferentsurfacemorphologycoatedbythethermalsubstratemethodin
aqueoussolutions.JournalofJapanInstituteofMetals2007;71(3)342345.
[19] ElliotJ.C.Studiedininorganicchemistry18,Structureandchemistryoftheapatites
andothercalciumorthophosphates.NewYork:Elsevier;1994.
[20] TangR,HennemanZ.J,NancollasG.H.Constantcompositionkineticsstudyofcar
bonatedapatitedissolution.JournalofCrystalGrowth2003;249(34)614624.
[21] AokiH.MarvelousBiomaterial,Apatite(inJapanese).Tokyo:ISHIYAKUPUBLISH
ERS,INC.;1999.
[22] HaneinD,SabanaryH,AddadiL,GeigerB.Selectiveinteractionsofcellswithcrystal
surfaces Implications for the mechanism of cell adhesion. Journal of Cell Science
1993;104(2)275288.
[23] OhgushiH,OkumuraM,YoshikawaT,InoueK,SenpukuN,ShorsE.C.Boneforma
tionprocessinporouscalciumcarbonateandhydroxyapatite.JournalofBiomedical
MaterialsResearch1992;26(7)885895.
[24] NihonKagakuKai.KagakuBinran,KisoHen(inJapanese).Tokyo:Maruzen;2004.
297
298
[25] KurodaK,MoriyamaM,IchinoR,OkidoM,SekiA.Formationandosteoconductivi
tyofcarbonateapatiteandcarbonateapatite/CaCO3compositefilmsbythethermal
substratemethodinaqueoussolutions.MaterialsTransactions2008;49(6)14341440.
tyofcarbonateapatiteandcarbonateapatite/CaCO3compositefilmsbythethermal
substratemethodinaqueoussolutions.JournalofJapanInstituteofMetals2009;73(5)
346353.
[27] Zhang Q, Chen J, Feng J, Cao Y, Deng C, Zhang X. Dissolution and mineralization
behaviorsofHAcoatings.Biomaterials2003;24(26)47414748.
[28] ChangM.C,IkomaT,KikuchiM,TanakaJ.Preparationofaporoushydroxyapatite/
collagennanocompositeusingglutaraldehydeasacrosslinkageagent.JournalofMa
terialsScienceLetters2001;20(13)11991201.
[29] WangR.Z,CuiF.Z,LuH.B,WenH.B,MaC.L,LiH.D.Synthesisofnanophase
hydroxyapatite/collagen composite. Journal of Materials Science Letters 1995;14(7)
490492.
[30] Chang M. C, Ko C, Douglas W. Preparation of hydroxyapatitegelatin nanocompo
site.Biomaterials2003;24(17)28532862.
[31] IshiiT,KoishiM,TsunodaT.Nuregijutsuhandbook(inJapanese).Tokyo:Techno
system;2001.
[32] Okada T, Ikada Y. Surface modification of silicone for percutaneous implantation.
JournalofBiomaterialsScience,PolymerEdition1995;7(171180).
tyofhydroxyapatite/collagencompositefilmsbythethermalsubstratemethodin
aqueoussolutions.MaterialsTransactions2009;50(5)11901995.
[34] BanS,MarunoS.Hydrothermalelectrochemicaldepositionofhydroxyapatite.Jour
nalofBiomedicalMaterialsResearch1998;42(3)387395.
[35] YamamotoD,KawaiI,KurodaK,IchinoR,OkidoM,SekiA.Osteoconductivityof
Anodized Titanium with Controlled MicronLevel Surface Roughness. Materials
Transactions2011;52(8)16501654
Chapter 15
System Integration of
a Novel Cell Interrogation Platform
Masaru Takeuchi, Gauvain Haulot and
Chih-Ming Ho
1.Introduction
Microfluidics[1,2]becomesthebackbonetechnologyforbiomedicaldiagnosesandthera
peuticsduetothelengthscalematchingbetweencellsandfluidicdevices.Themicroelectro
mechanicalsystem(MEMS)basedmolecularsensorshavereachedthesensitivityofdetecting
a few nucleic acid molecules and several pg/ml of proteins [3, 4]. During the last decade,
integrationofmicrofluidicdevicesintoasystemforinteractingwithcellularsystembecomes
possibleandgreatlyadvancesthecapabilityofcontrollingthecellfates.
Whilethenumberofmicrofluidicdevicesincreasingwiththedesiredfunctionalitiesofthe
microfluidicsystem,largenumberofinterconnectionsarenotjustdifficulttomanufactureand
canbeamajorroadblock.TheOptoelectronicReconfigurableMicrochannel(OERM)[5]has
basicallyalleviatedthedifficultdesignandfabricationproblem.OERMisanewtechnology
providinglightcontrolledcreation,annihilationandreconfigurationofmicrochannelswithin
seconds.Alightpatternisprojectedonareconfigurablechipresultingintheformationofa
correspondingmicrochannelnetworkthroughalowpoweroptoelectroniceffect.Whenthe
lightpatternismodified,themicrochannelnetworkreconfiguresaccordingly.Manufacturing
and reconfiguration of microchannels have been demonstrated in ice, frozen cyclohexane,
frozenDMSOandfrozenhexadecane.Asaconsequence,itisexpectedthatthetechnologycan
beusedwithmanymorematerialswithreasonablefreezingpointandlatentheatoffusion.
Hexadecaneisimmisciblewithwaterandiswidelyusedinmicrofluidicemulsions.Itsuseas
a reconfiguration template paves the way towards aqueous droplet manipulation on the
platform.Fabricationandreconfigurationhappeninseconds.
Thereconfigurabilitybydefreezingandrefreezingfluidcertainlyprovideaninnovativeway
ofgreatlysimplifyingtheinterconnectproblem.Ontheotherhand,mostcellsneedtobein
fluidsabovefrozenpointforproperphysiologicalfunctions.Forexample,humancellsneed
2013 Takeuchi et al.; licensee InTech. This is an open access article distributed under the terms of the
300
tobein37oCenvironment.Therefore,anadditionalmicrofluidicsystemoperatedat37oCneeds
to be designed and integrated seamlessly with the OERM system for accomplishing the
biologicalstudies.Inthisarticle,wewillintroducethedesignandmanufacturingmethodsof
thetwotemperaturefluidicinterrogativeplatform.
2.Optoelectroreconfigurablemicrochannel(OERM)
ThefirstissuewithdesigningOERMaretheprocessesofmeltingandfreezingchannelsof
micron size. The problem of phase transition belongs to the Stefan problem category. It
involvesamovingboundarythatrepresentsthephasechangelocation[6].Suchproblemshave
been extensively studied [7]. Their resolution is often strenuous and requires numerical
methods[8,9].Itisbeyondthescopeofthischapter.Nevertheless,phasetransitioncanstill
beinvestigatedwithouttheneedofacompletesolution.Phasetransitionmainlyaffectsthe
timedependenceofthetemperatureprofiles.Ithas,however,anindirectimpactonthesteady
statethatstemsfromthedifferencebetweenwaterandicethermalconductivities.Itcreatesa
discontinuityinthetemperaturegradient.Moreprecisely,thefirstlawofthermodynamics
imposes that the heat flux J(x, t) be continuous. The boundary conditions at the water/ice
interface located in x0(t) with water and ice respectively on the left and right side of the
u
( ))
( ))
interface.arethenu(x0(t))=0andkwater x ( x 0 t = kice x ( x +0 t .Asaresult,atsteadystate,the

temperatureprofileintheiceregionis:
J
forx x 0 , L :utr ( x ) = u ( x ) = kice ( L x ) + T c .Tcisthetemperatureatx=Landsymbolizes

thecoolerstemperature.
IfaheatfluxJ=1.2*104W/cm2 calculatedtoimposeT=0Cinx=100misused,theinterface
willbeat x = x = 100m:themicrochannelisstill100mdeepatsteadystate.Nonethe
less,thetemperaturegradientintheliquidregionishighersincetheheatconductivityis
lower.Thesteadystatesolutionforx x 0 , L iseasilyderived,andthecompletesteady
statesolutionis:
utr ( x ) =
utr ( x ) =
J
kwater
J
kice
( x0 x )ifx 0, x 0
( L x ) + T c ifx x0, L
Essentially,thetemperaturegradientishigherintheliquidregion.Itleadstohighertemper
atures than in the case where only ice is considered. For instance it leads to a steady state
temperatureinx=0ofvalueT = utr (0) = 2.8 C = 276K
TransientState.Theinfluenceofphasetransitiononthetransientstateisnowaddressed.The
temperatureandheatfluxsolutionoftheheatconductionproblemareplotted(seeFigure1)
togivemoreinsightonthephysicsthattakeplace.Itcanbeseenthatthetemperaturereaches
themeltingpointafter0.8s.Uptothatpointasinglephasesolutiontotheheatconduction
System Integration of a Novel Cell Interrogation Platform
problemisaccuratesincenophasetransitionoccurs.Figure1illustratesthatthetemperature
andheatfluxprofilewhenmeltingbeginsarealreadyclosetosteadystate.Inparticular,heat
fluxisalmostconstant,whichisdetrimentaltomelting.Anapproximatecalculationofthe
meltingtimecanbeperformedbasedonthosecurves.Sincethemeltingregionissmall,the
problemcanbelinearizedforafirstorderapproximation.Theheatusedformeltingthe100m
regionthenresultsfromthedifferencebetweentheheatfluxenteringtheregionandtheheat
fluxleavingtheregion.TheheatfluxJenteringatx=0isconstantandhasbeencalculated
previously.Theheatfluxexitingtheregionatx=100mandt=0.8sisinferredfromthesingle
phaseheatconductionproblemsolution.Theirdifferenceequals280W/m2.WithHthewater
enthalpyoffusionperunitvolume,H=3.32*108J.m3,andlthelengthofthemeltingzone,l=
100m,thefirstorderapproximationforthemeltingtimeis:
Hl
J J ( x = 100m, t = 0.8s )
=
melting
= 119s 2min
Figure 1. Evolution of temperature and heat flux profiles in the first 5s after heating starts.
Unfortunately, melting
is much longer than the characteristic transient times calculated
previously.Becausemeanwhilethetemperatureprofilewillevolvetowardshomongeneiza
tionoftheheatflux,itentailsthattherealmeltingtimewillprobablybelonger.Atthispoint,
forbetterunderstandingandassessmentoftheproblem,morequalitativeinsightisrequired.
When melting starts, it creates an interface at the melting temperature that gradually pro
gressestowardsitssteadystatelocation.Thespeedofthisprogressionisdeterminedbythe
couplingofthelatentheatoffusionandthederivativeoftheheatfluxacrosstheinterface.
Becausethelatentheatismuchhigherthantheheatcapacity,phasetransitionisamuchslower
processthantheincreaseoftemperature.Italsoimposesafixedtemperature.Asaresult,phase
transitionthwartstheheatfluxhomogenizationprocessofheatconductionacrossthemelting
interface.Thecharacteristictransienttimeforaniceregionof1mmandawaterregionof100m
donotdependontheheatfluxandarerespectively ice = 0.4sand ice = 0.03s.Whencomparing
to melting
, it seems a reasonable assumption to consider that steady state is reached in a
negligible time with regard to melting, both in the liquid region and the ice region. If the
interfaceisrepresentedbyasmallinterval,theheatfluxdifferenceacrosssuchintervalcanbe
assessed.Asaconsequence,itturnsoutthattheoverallmeltingtimecanbecalculatedrather
301
302
preciselyusingthefollowingmethod.Theoverallmeltingregioncanbediscretizedinsmall
intervalsthateachcorrespondtoavalued(t).Foreachd(t),thesteadystateprofilesinthe
water and in the ice regions are estimated. The temperature in the liquid region will be
described as uwater ( x, t ) =
J
kwater
(d (t ) x ) while the temperature in the ice region will be
L x
d (t ) T c .ConsequentlytheinwardheatfluxisJwhiletheoutwardheatfluxis
kice T c
H
L d (t ) .Itentailsthatthemeltingtimeforeachintervalis = J J out (d (t )) ,where
uice ( x, t ) = T c L
J out (d (t )) =
is the size on an interval. Since the interface progression slows down, the interval can be
discretizedmorefinelynearthesteadystateposition.Here, = 10monthefirst80m,then
= 1mfromx=80mtox=99mandfinally = 100nmfromx=99mtox=100m.The
overalltimecalculatedthroughthismethodgivesanestimateofthemeltingdurationofa
100mchannelwithaprecisionof100nm.Itgives
melting
= = 181s 3min
ThemovementofthesolidliquidinterfaceisplottedinFigure2.Whenmelting
isaddedtothe
transienttime,namely2s,itbecomesthetimethata100mchannelneedstoreachstability
withaprecisionofahundrednanometers.Thetimestomeltsmallerregionswiththesame
heat flux are also noteworthy, even thought the steady state melted region will still be a
hundredmicrometerslong.Forinstance50maremeltedin18s,while10maremeltedin3s.
Besides, 90m and 99m are reached respectively in 56s and 106s. Interestingly, the first
approximationmelting
=119swasnotthatfarfetchedincomparison.Moreover,itmeansthat
twothirdofthetimearespentmeltingthelast10m.Varyingtheheatfluxcouldtherefore
shortenmeltingtimestremendously.Forinstanceahighheatfluxcanfirstbeusedtoreach
thedesiredsizeinashorttime;itcanthenbedecreasedtothevaluecorrespondingtothe
steadystaterelatedtothatsize.
Figure 2. Evolution of the location of the water/ice interface with time.
Finallythecaseoffreezingisalsonoteworthy.Microchannelfreezingindeedhappensatthe
verybeginningofthecoolingprocess:thetemperaturereaches0Cwithin0.1s.Asaresult,the
temperatureprofileintheicestillimposesahighheatfluxthroughFourierslawwhilethe
heatfluxatx=0isnowzero.Sincephasetransitionblocksthetemperatureevolutioninthe
icebyfixingthetemperatureatthefreezingpoint,freezingoccursataheatfluxvaluecloseto
the highest one in the device: J. The freezing time can therefore be evaluated as ice = 0.4s.
Steadystatewillbereachedtwosecondslateraccordingtothelastsection,henceafterabout
2.5s.Thetimesrequiredtomeltamicrochannelandfreezeamicrochannelhaveverydifferent
values;thisisduetotheirtimingwithinthetransientstate.Meltinghappensattheend,when
thederivativeoftheheatfluxisalreadyclosetozero:littleheatisavailableforphasetransition.
Onthecontraryfreezingoccursattheverybeginningofthetransientphaseandthusbenefits
fromhighheatfluxderivatives:freezingtimesareshort.Thisobservationbolstersthestrategy
forfastmeltingdescribedhereinbeforethatconsistsinusingahighheatfluxatthebeginning,
andtodecreaseitsvaluewhenthedesiredmicrochannelsizeisobtained.Indeed,meltingwill
startearlierinthetransientphaseprovidinghigherheatfluxgradientsforphasetransition.
Thosehigherheatfluxgradientsconcentrateacrosstheinterfacewhenheatfluxhomogenizes
intheliquidandsolidzones.Besides,byreducingheatfluxabruptly,theslowconvergence
oftheinterfacetoitssteadystatepositionisskipped.Therefore,sincevaryingtheheatflux
offersschemesforfastmelting,thisanalysistendstoprovethatfastreconfigurationcanbe
achieved.Additionally,itclearlyprovesthatmeltedmicrochannelswillbestable.Finally,the
questionofflowshasnotyetbeenaddressed.Afluidenteringachannelwillthermallyinteract
withitsenvironment.Itwillprobablyhaveahighertemperatureandcouldcausethechannel
shapestovary.Nevertheless,sincethetransienttimeforsmallvolumesisveryshortthisissue
canprobablybeeasilyaddressed.Itshouldnotbeamajorconcern.
Theoreticalmodelsandsimulationshaveproventhefeasibilityofmicrochannelreconfigura
tion by local melting. Preliminary experiments were also performed that corroborated the
analysisanddemonstratedliquidtransportinsuchchannels.Theyinvolvedpatternedmetal
electrodesthatwouldlocallythawicebyresistiveheating.Furtherworkhascharacterized
determinedtheoutputpowerofthephenomenonusedinOERM,optoelectronicheating[10].
Buildingonthosefindings,wedealtwiththecompleteplatformforoptoelectronicreconfig
urablemicrochannels,theOERMplatform.OERMconsistinlocallymeltingmicrochannelsin
afrozenworkingmediawithlightpatterns.Sinceitisareversibleprocess,microchannelscan
reconfigure when the actuating patterns change. Figure 3 illustrates this principle. The
previous paragraphs have reported the power needed for such task: direct melting by
illuminationwouldrequirehighpowerlightsources.Therearegreatadvantagestolowpower
light for parallel manipulation, flexibility, and device integration. OERM thus rely on a
transduction mechanism where lowpower light is converted into highpower heat; it is
achieved via optoelectronics and Joule effect. Therefore, such technology is built on the
synergeticharnessingofseveralphysicalphenomena:lighttransmission,transductionoflight
intoelectricalcurrents,conversionofelectricalcurrentsintoheat,melting,andmicroflows:
hence the complexity of the OERM platform. At its core lies a photoconductive material.
Because of its high light absorption and very low thermal conductivity compared to other
photoconductivematerialssuchascrystallinesilicon,hydrogenatedamorphoussiliconhas
provedthebestcandidate.
303
304
Figure 3. OERM principle: microchannel reconfiguration controlled by light patterns. 1) Working media is frozen. 2) A
first light pattern is displayed, which creates a corresponding microchannel; through external pumping green particles
(channel end, left-hand side of image) and purple particles (channel first branch, right-hand side of image) flow
through. 3) A second light pattern is displayed; the microchannel reconfigures accordingly, and the purple particles
(end of new channel) are directed towards another outlet.
OERMisanewmicrofluidicplatformthatcanprovideincomparableflexibilityintheplanar
manipulationofbiomoleculesinmicrochannels.Thankstothisnewtechnologyanymicro
fluidicnetworkcanbemanufacturedandreconfiguredwithinseconds.Itwillbeusedonthe
plasmonicsensorplatformforcell,drugandsensingprobedelivery(cf.Figure5).
Aschematicoverviewoftheoptoelectronicreconfigurablemicrochannelplatformisshown
inFigure4.Anoptoelectronicchipisfabricatedbysuccessivelydepositingonaglasssubstrate
atransparentconductivelayermadeofindiumtinoxide(ITO),aphotoconductivelayermade
ofhydrogenatedamorphoussilicon(aSi:H),andahighlyreflectiveconductivelayermadeof
titanium,platinumandsilverorgold.Aliquidlayerislaidupontheoptoelectronicchipand
isconfinedbetweenthechipandacoverslip.Thesystemisplacedonatransparentcoolerto
freeze the liquid. A voltage bias is then applied between the two conductive layers of the
optoelectronicchipresultinginaconstantelectricfieldacrossthephotoconductivelayer.A
light pattern is projected on the optoelectronic chip through the transparent cooler. It is
absorbedbythephotoconductivelayer(aSi:H).InthedarkpartsofthelightpatternaSi:H
behavesasaninsulator,hencenoelectricalcurrentflowsacrossit.Onthecontraryinthebright
partsofthelightpatternlocalcurrentsflowacrossaSi:H.ThosecurrentsthuscreatelocalJoule
heating. This process is called optoelectronic heating and is schematically represented (cf.
Figure4).SettingthelocalJouleheatingtotherightpowertriggerslocalmeltinginthefrozen
liquid, thus forming microchannels corresponding to the light pattern. When light stops
shiningonamicrochannel,localJouleheatingstopsandthefluidinthemicrochannelfreezes
thusclosingthemicrochannel.Ifthelightpatternhasreconfigured,othermicrochannelswill
openinotherlocationsonthechip.Flowscanbeproducedinthemicrochannelnetworkusing
anexternalpumpingsystemsootherphasesorparticlescanbemanipulatedinthenetwork.
Themeltedliquidiskeptasthemainphasetoallowthemeltedmicrochannelstofreezeback
whenlightpatternsdisappearorreconfigure.
Thekeynovelfeatureofthistechnologyisthatthelightpatterncanbechangedatwill,hence
makingthemicrofluidicnetworkreconfigureinstantaneously.Thispotentiallyallowsforany
dynamic twodimensional design of the microchannel network. The projected pattern is
controlledbyacomputer,whichentailsthatnotonlycanthelightpatternbepreprogrammed,
butalsoitcanevolveonitsownthankstoafeedbackcontrolloop.Thisisarealbreakthrough
formicrofluidicsandoffersawholenewrangeofpossibleapplications.
Last year a first version of the optoelectronic chip was manufactured that led to the first
demonstrationsoflocalmeltingcontrolledbylowpowerlightimages.Thesetupconsisting
oftheelectricalcontrols,thecoolingsystemandtheopticalprojectionsystemwasassembled
andoptimized.Itpavedthewaytothisyearsaccomplishments.
Figure 4. Working principle of OERM
Figure 5. OERM integrated with plamsonic sensor platform
305
306
3.Cellinterrogationplatform
OERM is being integrated in a biosensing platform. The platforms objective is to achieve
effectivedrugscreeningatthesinglecelllevel.Theplatformhasmainlytwopartsasshown
inFigure5:oneisreconfigurabledeliveryandtheotheriscellinterrogation.Inthereconfig
urabledeliveryarea,cells,drugsandmoleculesarepreparedandsuppliedtothecellinter
rogation area by the reconfigurable microchannel. In the cell interrogation area, cells area
attachedonthesubstrateandplasmonicsensorsareplacedaroundcellstodetectbiomarkers
fromcellsunderdrugstimulation.
To realize the platform, two different temperature conditions have to be realized in the
platform. The reconfigurable delivery needs low temperature condition to keep working
materialsfrozen.Ontheotherhand,ahightemperaturecondition(e.g.37C)hastobekept
in the cell interrogation part to keep cells alive. It is also important to supply drugs from
reconfigurable microchannel. The flow from the reconfigurable microchannel has to be
controlledtosupplydrugstocellsefficiently.Thissectionpresentsmainlythesetwoissues:
temperaturecontrolandflowcontrolforthecellinterrogationplatform.
3.1.Temperaturecontrol
Inthecellinterrogationplatform,water(orculturemedia)isusedasaworkingmediaofthe
reconfigurablemicrochannelforcelltransportation.Itisimportanttorealizethetemperature
onthereconfigurableareabelow0Ctokeepice.Ontheotherhand,temperaturehastobe
keptat37Ctoculturecellsinthecellinterrogationarea.Thermoelectricdevicesareusedto
realizethesetwotemperatureconditionsinthesmallplatform.
Atfirst,twothermoelectricdevicesareawereusedtocheckwhetherthesedevicescanmaintain
desiredtemperatures.Thermoelectricdevicesarewidelyusedtocontroltemperaturebelow
0Cinasmallareasuchasafewsquaremillimeters[11,12].Thermoelectricdeviceshavealso
beenintegratedinmicrofluidicdevicestocontroltemperatureformicrovalvesoperation[13],
makingiceinamicrochannel[14],andDNAanalysis[15].Figure6showstheexperimental
setupofthethermoelectricdevices.Thethermoelectriccoolerhasaheatsinkandafanonthe
warmsidetocooldownthewarmsideandachievelowtemperatureonthecoldside.Aglass
slidewasplacedonthethermoelectricdevicestocheckthetemperatureonthesubstrate.The
temperature on the glass substrate was measured by an infrared thermometer. 8 points as
showninFigure6weremeasuredinthisexperiment.Intheexperiment,12Vand2.3Vwere
appliedtothethermoelectriccoolerandheaterrespectively.Table1showstheresults.Onthe
cooler,thetemperaturewasbelow0Candfrostsgrewonthecoolerandtheglasssubstrate.
Ontheheater,thetemperatureshowedaround37C.Theresultsindicatethatthethermo
electricdevicescanbeusedtocontroltemperatureonthecellinterrogationplatform.
Figure 6. Experimental setup for temperature measurement on the thermoelectric devices
Measurement point
Temperature [C]
12
-6
-4
30
38
36
Table 1. Temperature measurement on the thermoelectric devices
Thetemperaturevariationhastobesmalltoformuniformmicrochannelsinthereconfigura
blearea.Tocheckthetemperaturedistributionsontheplatform,thesimulationwasconduct
edusingafiniteelementanalysismethod.Atfirst,theplatformwasdirectlyputonthetwo
thermoelectriccoolerstocheckthetemperatureuniformityontheplatforminthissimplesetting.
Figure7(a)showstheconstructedmodel.Twothermoelectriccoolersareplacedwithsome
distancetoprojectlightpatternsontotheplatforminbetweenthecoolers.Theplatformismade
ofglass,andthethermoelectriccoolersandroomtemperaturearesetto5Cand20C
respectively.Thethermalconductivityoftheglasssubstratewassetto0.74W/(mK)inthe
simulation.ThesimulationresultisshowninFigure7(b).Inthiscondition,thetemperature
variationontheglasssubstrateislargeandthetemperaturecannotmaintainbelow0Cinthe
reconfigurablearea.Inthisexperimentalsetup,icecannotbeformedonthereconfigurablearea.
307
308
Figure 7. Simulation results when the platform is directly placed on the two thermoelectric cooler (a) A schematic of
constructed model for the simulation (b) Simulation result.
Torealizelowtemperature(below0C)andsmallvariationofthetemperatureontheplatform,
analuminiumplatewasplacedinbetweenthethermoelectriccoolersandtheplatform.Figure
8(a)showstheconstructedmodel.Thealuminiumhashighthermalconductivity(225W/(m
K)).Thealuminiumplatecanbecooledbythethermoelectriccoolerspreciselyandfast.The
aluminiumplatehasa7mm7mmsquareholetoprojectlightimagesfromunderneathonto
thereconfigurablearea.Theplatformcanbecooledstronglybecausethebottomfaceofthe
platformcanbeattachedtothealuminiumplateoutsideofthereconfigurablearea.Asshown
inFigure8(b)and(c),thetemperatureinthereconfigurableareawasmaintainedbelow0C
andthetemperaturevariationwaslessthan0.7C.Thisresultindicatesthatthealuminium
plateiseffectiveforcooling.
Forpracticalusesofthiscoolingsystem,condensationofthevapourisoneofthesignificant
problems.LightpatternsfromtheDMDdevicecomefromunderneathofthereconfigurable
area. If there is condensation water on the platform, the light pattern is refracted by the
condensationandthelightpatternischanged.Topreventcondensationontheplatform,an
acrylicbarisplacedonthepathofthelightpatternfromtheDMDdevicetothereconfigurable
area.Acrylicmaterialshavelowthermalconductivity(0.19W/(mK))anditisconsideredthat
thebottomfaceoftheacrylicbarcanbemaintainedclosetotheroomtemperatureevenifthe
topfaceisunder0Cprovidedthattheacrylicbarislongenough.Inthesimulation,theacrylic
barisplacedinandundertheholeofaluminiumplateasshowninFigure9(a).Theheightof
theacrylicbarwasseton35mm.
Figure9(b)and(c)showsthesimulationresults.Thetemperatureinthereconfigurablearea
iskeptunder0Candthetemperaturevariationdecreasedtolessthan0.3C.Ontheother
hand,thebottomfaceoftheacrylicbarismorethan18C.Theresultsshowthattheuniform
lowtemperatureinthereconfigureareacanberealizedandcondensationundertherecon
figurableareacanbepreventedbyputtingtheacrylicbarunderthereconfigurablearea.
Figure 8. Simulation results when an aluminium plate was paced in between the cooler and platform (a) A schematic
of constructed model (b) Simulation result (c) Temperature variation in the reconfigurable area (temperature on the
red dash line in (b))
Figure 9. Simulation results when an acrylic bar was paced under the reconfigurable area (a) A schematic of con
structed model (b) Simulation result (c) Temperature variation in the reconfigurable area (temperature on the red
dash line in (b))
309
310
Athermoelectricheaterisaddedinthemodeltorealizecellinterrogationareaintheplatform.
Fig.10(a)showstheconstructedmodel.thethermoelectricheaterandcoolersareseton37C
and5Crespectively.Thedistancebetweenthecoolersandheaterwereseton10mm.
Figure 10. Simulation results when the thermoelectric coolers and heater are integrated for the platform. (a) A sche
matic of constructed model (b) Simulation result (c) Temperature variation in the reconfigurable area (temperature on
the red line in (b))
Thesimulationresultsshowsthatthereconfigurableareaiskeptunder0Ceveniftheheater
isadded.
3.2.Flowcontrol
TorealizethecellinterrogationplatformasshowninFigure5,cells,drugsandmoleculesfor
cellinterrogationhavetobesuppliedfromreconfigurabledeliveryarea.Tosupplycellsand
drugstothecellinterrogationareaefficiently,thespreadangleoftheflowfromthereconfig
urableareahastobesmall.Tochecktherelationshipinbetweentheflowrateandthespread
angle,apolydimetylsiloxane(PDMS)microchannelwasused.Thecrosssectionalshapeofthe
microchannelisabout200m100m.Intheexperiment,flowratewaschangedfrom700l/
minto1000l/minbyasyringepump.Tocheckthespreadangle,bluedyedwaterflowedout
fromthemicrochannelandtheoutletofthemicrochannelwasobservedunderamicroscope.
Figure 11 shows the experimental results. Bluedyed water flowed out from the PDMS
microchanneltotheopenspace.Thespreadangledecreasedwhentheflowrateincreased,
whichreached5whentheflowratewasseton1000l/min.Itisconsideredthatthespread
angleissmallenoughtosupplycellsanddrugstothetargetpositionintheplatform.
Tocheckthespreadofcellsbydischargingfromthemicrochannel,RAW264.7macrophage
cellswereused.Cellanalysishasbeenachievedinmicrofluidicdevicespreviously[16].The
glasssubstratewascoatedbypolyLlysinetoattachcellsonthesubstratebeforeexperiment
[18].Polylysineiscommonlyusedtoimprovecelladhesiononsubstratesandcanbeusedto
makearraysofcellsandhydrogelsforbiologicalapplication[18,19].Intheexperiment,the
cellsweredischargedfromthemicrochannelin5minutesbytheflowrateof1000l/min.
Figure12showstheexperimentalresult.Thecellsweredischargedfromthemicrochanneland
spreadonthesubstrate.Thespreadangleofcellswaslargecomparedtotheflowexperiment
as shown in Figure 3.7. When the syringe pump was stopped, the flow was not suddenly
stoppedbutgraduallysloweddown.Duringtheslowdownoftheflowspeed,thecellswere
spreadthroughoutthesubstrate.Itisexpectedthatthecellsnotattachedtothesubstratecan
beflushoutbyasolutionwithoutcellswhenthepatterningofthesubstrateisconductedto
attachcellspartiallybeforeexperiment.
Figure 11. Observation results of spread angle when the flow rate was changed from 700 to 1000 l/min.
311
312
Figure 12. Discharge of macrophages from the microchannel
4.Conclusion
ThischaptershowstheoreticalanalysisaboutphasetransitionintheOERMdeviceandan
applicationforabiosensingplatform.TheOERMplatformiscurrentlybeingintegratedwithin
a cell interrogation platform. Temperature and flow control are conducted to realize the
platform. The platform will realize sensing protocols automatically for labonthechip
analysisappliedtodiseasetherapies.
Acknowledgements
ThisprojectissupportedbyNSFnanomanufacturecenterSINAM(DMI0327077)
Authordetails
MasaruTakeuchi1,GauvainHaulot2andChihMingHo3
1GraduateSchoolofEngineering,NagoyaUniversity,Nagoya,Japan
2HenrySamueliSchoolofEngineeringandAppliedScience,UniversityofCalifornia,Los
Angeles,USA
3DepartmentofMechanical&AerospaceEngineering,DepartmentofBioengineering,Uni
versityofCalifornia,LosAngeles,USA
References
[1] LiuJ,TaiY.C,PongK,HoC.M.MicromachinedChannel/PressureSensorSystems
for Micro Flow Studies. June 1993, International Conference on Solid State Sensors
andActuators.1993;995997.
[2] HoC.M,TaiY.C.MicroElectroMechanicalSystemsandFluidFlows.AnnualRe
viewofFluidMechanics1998;30579612.
[3] Lillehoj P. B, Wei F, Ho C. M. A selfpumping labonachip for rapid detection of
botulinumtoxin.Labonachip2010;10(17)226570.
[4] WeiF,LillehojP.B,HoC.M.DNADiagnostics:NanotechnologyEnhancedElectro
chemicalDetectionofNucleicAcids.PediatricResearch2010;67(5)458468.
[5] HaulotG,BenahmedA.J,HoC.M.Optoelectronicreconfigurablemicrochannels.In:
ProdeedingsofIEEE24thInternationalConferenceonMicroElectroMechanicalSys
tems(MEMS).2327January2011.
[6] Bonnerot R, Jamet P. Numerical computation of the free boundary for the twodi
mensional Stefan problem by spacetime finite elements. Journal of Computational
Physics1977;25(2)163181.
[7] CrowleyA.B.Numericalsolutionofstefanproblems,InternationalJournalofHeat
andMassTransfer1978;21(2)215219.
[8] CiavaldiniJ.F.AnalyseNumeriqueDunProblemedeStefanADeuxPhasesParUne
MethodeDelementsFinis.SIAMJournalonNumericalAnalysis1975;12(3)464487.
[9] ChenS,MerrimanB,OsherS,SmerekaP.ASimpleLevelSetMethodforSolvingSte
fanProblems,JournalofComputationalPhysics1997;135(1)829.
[10] Haulot G, Ho C. M. Optoelectronic Heating for Fabricating Microfluidic Circuitry.
AdvancesinOptoElectronics2011;2011(237026).
[11] WalleB.L,GauthierM,ChailletN.PrincipleofaSubmergedFreezeGripperforMi
croassembly.IEEETransactionsonRobotics2008;24(4)897902.
[12] LiuJ,ZhouY.X,YuT.H.Freezetweezertomanipulatemini/microobjects.Journal
ofMicromechanicsandMicroengineering2004;14269276.
[13] ZimmermannM,BentleyS,SchmidH,HunzikerP,DelamarcheE.Continuousflow
inopenmicrofluidicsusingcontrolledevaporation.LabonaChip2005;513551359.
[14] Li Y, Wang F, Wanga H. Cell death along single microfluidic channel after freeze
thawtreatments.Biomicrofluidics2010;4014111.
[15] KhandurinaJ,McKnightT.E,JacobsonS.C,WatersL.C,FooteR.S,RamseyJ.M.
IntegratedSystemforRapidPCRBasedDNAAnalysisinMicrofluidicDevices.Ana
lyticalChemistry2000;7229953000.
313
314
[16] AmatoreC,ArbaultS,ChenY,CrozatierC,TapsobaI.Electrochemicaldetectionina
microfluidicdeviceofoxidativestressgeneratedbymacrophagecells.LabonaChip
2007;7233238.
[17] NilforoushanD,GramounA,GlogauerM,ManolsonM.F.Nitricoxideenhancesos
teoclastogenesispossiblybymediatingcellfusion.NitricOxide2009;212736.
[18] Chang W. C, Sretavan D. W. Novel HighResolution Micropatterning for Neuron
CultureUsingPolylysineAdsorptiononaCellRepellant,PlasmaPolymerizedBack
ground.Langmuir2008;241304813057.
[19] ZhangH,ShepherdbJ.N.H,NuzzoR.G.Microfluidiccontactprinting:aversatile
printing platform for patterning biomolecules on hydrogel substrates. Soft Matter
2010;622382245.
Chapter 16
Transferrin-Toxin Conjugates for Cancer

Daniel T. Kamei
Abstract
Forseveralyears,researchershavetargetedreceptorsoverexpressedoncancercellstoimprove
theselectivityoftoxicanticanceragents.Inadditiontoconjugatingthetoxintoanantibody
forsuchareceptor,thenaturalligandfortheoverexpressedreceptorhasalsobeenusedasa
proteinbaseddrugdeliveryvehicle.Thetherapeuticefficacyofsuchliganddrugmolecular
conjugates,however,maybelimited,sincetheynaturallyfollowtheintracellulartrafficking
pathwaysoftheendogenousligands,whichhavecertainlynotbeenoptimizedbynaturefor
drugdeliveryefficacy.Accordingly,noveldesigncriteriamaybeidentifiedfortheseligands
throughanunderstandingoftheirintracellulartraffickingpathways.Thisshortreviewbriefly
describes the transferrin ligand/transferrin receptor system, where intracellular trafficking
considerations have led to improvements in the therapeutic efficacy of transferrintoxin
molecularconjugates.
1.Introduction
CanceristhesecondleadingcauseofdeathintheUnitedStates[1,2].Unfortunately,current
treatmentsinvolveinvasivesurgeryfollowedbynonspecificradiationandchemotherapythat
harm both healthy and cancer cells. Accordingly, much research has been dedicated to
improving the tumor selectivity of chemotherapy treatments. Since the early 1980s, many
receptorswerefoundtohaveincreasedexpressionincancercellscomparedtotheirnormal
counterparts.Theseincludetransferrinreceptor,interleukin13receptor,andgrowthfactor
receptors[37].Thesereceptorshavebeeninvestigatedforseveralyearsascancerselective
targetsfortherapeuticpurposes.
Anunderstandingofthetraffickingpathwayofthenaturalligandcanleadtonoveldesign
criteriaforengineeringtheligandtobeamoreeffectivedrugcarrier[812].Thisshortreview
focusesontransferrinligandtoxinmolecularconjugateswithparticularemphasisgiventothe
2013 Kamei; licensee InTech. This is an open access article distributed under the terms of the Creative
316
intracellular trafficking properties of the ligand. Moreover, although transferrin has been
conjugatedtoliposomes[13]andnanoparticles[14]toenhancetheirtargetingtocancercells,
theyarenotincludedinthisreview,sincetheintracellulartraffickingpropertiesofthesedrug
deliveryvehiclescanvaryfromthatoftheligandaloneduetothelargedifferencesinsize.
Humanserumtransferrin(Tf)hasbeeninvestigatedforseveralyearsasatargetingagentdue
totheoverexpressionofitsreceptoroncancercells.Tfhasamolecularweightofapproximately
80 kDa and is responsible for transporting free iron from the circulation to cells. Each Tf
moleculehasthecapabilitytobindtotwoferric(Fe3+)ions,oneintheNterminallobe(Nlobe)
andtheotherintheCterminallobe(Clobe).Eachlobebindstoaferricionwithanequilibrium
dissociationconstant(KD)ofapproximately1022M[15,16].ThisironboundTf,orholoTf,
thenbindstoitscellsurfacereceptor(TfR).Afterbinding,TheTf/TfRcomplexisinternalized,
andholoTfdeliversitsirontothecell,promotingcellulargrowthandproliferation[17].Since
cancercellsrequiremoreirontosustaintheirrapidproliferation,theyhavebeenfoundto
overexpressTfR,andthishighexpressionlevelofTfRhasbeenexploitedtoachieveselective
targetingofanticanceragentstocancercells.
2.Intracellulartrafficking
TheintracellulartraffickingpathwayofTfanditsreceptorhasbeenstudiedforseveralyears
andhasbeenreviewedinmanyjournals[1820].AfterholoTfbindstoTfRonthesurfaceof
cellswithnanomolaraffinity(KD~109M),theTf/TfRcomplexisinternalizedaspartofan
endocyticvesicle.TheendosomethenmaturesandacidifiestoapHbetween5and6[21],
causingirontobereleasedfromTf.Onceironisreleased,itisreducedtotheferrous(Fe2+)
formduetothepresenceofoxidoreductasesintheendosome.Adivalentmetaltransporter
thenshuttlesFe2+intothecytosol.TheironfreeTf/TfRcomplexrecyclesbacktothecellsurface,
andsinceironfreeTf(apoTf)hasalowbindingaffinityforTfRatthenearneutralpHofthe
cellsurface,apoTfquicklydissociatesfromthecellsurfacereceptor.Thisentirecycleofthe
Tf/TfRtraffickingpathwaylastsonlyabout5minutes[22].
WhiletherapidrecyclingofTfcontributestotheefficienttransportofiron,italsolimitsthe
ability of the ligand to deliver its payload. Accordingly, drug delivery efficacy may be
improvedthroughabetterunderstandingofthekineticsinvolvedintheintracellulartraffick
ingpathwayofTf[23].Throughinvitroexperimentsandmathematicalmodeling,Murphy
andcoworkerspreviouslydemonstratedthatmonoclonalantitransferrinreceptorantibodies
wouldbemoreeffectivedrugdeliveryvehiclesiftheyassociatedwithcellsforagreaterperiod
oftime[2426].Inotherwords,thisincreasedcellularassociationwouldleadtoanincreasein
theexposureofcancercellstotheconjugateddrug.InsteadofinvestigatingantibodiesforTfR,
theKameilaboratorystudiedtheTfliganditself.Specifically,byderivingandanalyzinga
mathematicalmodelforTf/TfRtrafficking,Kameiandcoworkersidentifiedanoveldesign
criterionforengineeringTftoenhanceitsdrugdeliveryefficacy[810],asdiscussedbelow.
3.Transferrindiphtheriatoxinconjugates
Diphtheria toxin (DT) is a protein toxin secreted by Corynebacterium diphtheria. DT acts by
inhibitingproteinsynthesisthroughinactivationofelongationfactor2(EF2)[27].Tfconju
gatedtoamutantofDT,knownasCRM107,hasbeeneffectiveagainstmalignantgliomas[28].
ThemutationinCRM107significantlyinhibitsthebindingofthetoxintoitsnativereceptor,
aheparinbindingepidermalgrowthfactorlikegrowthfactorprecursor,therebyreducingthe
nonspecifictoxicityofTfCRM107[29].YouleandOldfieldsgroupperformedinvivostudies
usingTfCRM107onsolidhumangliomasintheflanksofnudemice,andobservedincreased
cytotoxicityexertedbytheconjugate.ThesuccessofTfCRM107eventuallyledtophaseIII
clinical trials, which were unfortunately canceled in late 2006 following the results of a
conditionalpoweranalysissuggestingthatitsefficacywouldnotsignificantlyimproveupon
thecurrentstandardofcaretreatments.
Asmentionedabove,Tfrecyclesveryrapidly,andthisshortdurationinsidethecellcanlimit
the ability of Tf to deliver its cytotoxic payload. Therefore, to identify new approaches to
improving the efficacy of TfCRM107, Kamei and coworkers used mass action kinetics to
deriveamathematicalmodeloftheTf/TfRtraffickingpathway.Analysisofthemodelhelped
determinethat,byreducingorinhibitingtheironreleaserateofTfwithintheendosome,its
drugdeliveryefficacywouldbesignificantlyimproved[8].Inthisscenario,ironwouldbe
retainedbyTfuponrecyclingtothecellsurfaceandtheconjugatewouldbereinternalizedto
participateinanothercycleoftraffickingduetothepreservedhighaffinityofholoTfforTfR,
increasingitscellularassociation.ThedrugdeliveryefficacyofTfcanthereforebeimproved,
sinceasingleTfdrugconjugatewouldundergomultipletraffickingcycles,therebyincreasing
theprobabilityofdeliveringthedrug.KameiandcoworkersengineeredtwoTfvariantsthat
satisfied the molecular level design criterion using sitedirected mutagenesis [9]. These Tf
mutantswereconjugatedtoDTandwereshowntobemoreeffectivethanwildtypeTfin
deliveringDTinvitrotoU87andU251humangliomacelllines[10].Furthermore,Kameiand
coworkersperformedinvivoexperiments,anddemonstratedthatbothmutantTfDTconju
gatesweremoreeffectivethantheirwildtypecounterpartinshrinkinggliomatumorsonthe
flanks ofmice[10].Studies arecurrently beingperformedwith these mutantTfmolecules
conjugatedtoCRM107.
4.Concludingremarks
Ligandtoxin molecular conjugates for cancer have been studied for several years. Though
theseconjugateshavedemonstratedsomesuccess,notmanyhaveobtainedFDAapprovalfor
thetreatmentofcancer.ThegenerallackofFDAapprovedligandtoxinconjugatesmaybe
attributedtothephysiologicalbehaviorofthetargetingligand.Althoughligandscaneffec
tivelytargetcancercellsviatheircellsurfacereceptors,theywillfollowtheligandsphysio
logicalpathwayonceboundtoreceptors.Forinstance,therapidrecyclingrateofTfwhich
aidsinirondeliverycanrestricttheabilitytodeliveraconjugatedtoxin[8].Throughquanti
317
318
tativeexperimentsandmodeling,amorethoroughunderstandingoftheintracellulartraf
ficking properties of various ligands can be achieved. The Kamei laboratory has used this
approachtofindaninnovativeapproachtomanipulatingtheTftraffickingpathwayforthe
advancementofTfbasedcancertherapeutics[8,9].Thisexampledemonstratesthatasystems
analysis of intracellular trafficking properties can result in the engineering of effective
targetingagentstocancers.
Authordetails
DanielT.Kamei
DepartmentofBioengineering,UniversityofCalifornia,LosAngeles,USA
References
[1] MininoA.M,XuJ,KochanekK.D,TejadaVeraB.DeathintheUnitedStates,2007,
NCHSdatabrief;2009.
[2] JemalA,SiegelR,XuJ,WardE.Cancerstatistics,2010,CA:acancerjournalforclini
cians2010;60277300.
[3] Mendelsohn J. Targeting the epidermal growth factor receptor for cancer therapy,
JournalofClinicalOncology2002;201S13S.
[4] Kobrin M. S, Yamanaka Y, Friess H, Lopez M. E, Korc M. Aberrant expression of
typeIfibroblastgrowthfactorreceptorinhumanpancreaticadenocarcinomas.Can
cerresearch1993;5347414744.
[5] LowP.S,KularatneS.A.Folatetargetedtherapeuticandimagingagentsforcancer.
Currentopinioninchemicalbiology2009;13256262.
[6] MurataT,ObiriN.I,DebinskiW,PuriR.K.StructureofIL13receptor:analysisof
subunit composition in cancer and immune cells. Biochemical and biophysical re
searchcommunications1997;2389094.
[7] RechtL,TorresC.O,SmithT.W,RasoV,GriffinT.W.Transferrinreceptorinnor
malandneoplasticbraintissue:implicationsforbraintumorimmunotherapy.Jour
nalofneurosurgery1990;72941945.
[8] LaoB.J,TsaiW.L,MashayekhiF,PhamE.A,MasonA.B,KameiD.T.Inhibitionof
transferrinironreleaseincreasesinvitrodrugcarrierefficacy.JournalofControlled
Release2007;117403412.
[9] YoonD.J,ChuD.S,NgC.W,PhamE.A,MasonA.B,HudsonD.M,SmithV.C,
MacGillivrayR.T,KameiD.T.Geneticallyengineeringtransferrintoimproveitsin
vitroabilitytodelivercytotoxins.JournalofControlledRelease2009;13378184.
[10] YoonD.J,KwanB.H,ChaoF.C,NicolaidesT.P,PhillipsJ.J,LamG.Y,MasonA.B,
WeissW.A,KameiD.T.Intratumoraltherapyofglioblastomamultiformeusingge
neticallyengineeredtransferrinfordrugdelivery.Cancerresearch2010;7045204527.
[11] LuY,LowP.S.Folatetargetingofhaptenstocancercellsurfacesmediatesimmuno
therapyofsyngeneicmurinetumors.CancerImmunolImmunother2002;51153162.
[12] Lu Y, Sega E, Low P. S. Folate receptortargeted immunotherapy: induction of hu
moral and cellular immunity against haptendecorated cancer cells. International
journalofcancer2005;116710719.
[13] Shmeeda H, Mak L, Tzemach D, Astrahan P, Tarshish M, Gabizon A. Intracellular
uptake and intracavitary targeting of folateconjugated liposomes in a mouse lym
phoma model with upregulated folate receptors. Molecular cancer therapeutics
2006;5818824.
[14] Davis M. E, Zuckerman J. E, Choi C. H, Seligson D, Tolcher A, Alabi C. A, Yen Y,
Heidel J. D, Ribas A. Evidence of RNAi in humans from systemically administered
siRNAviatargetednanoparticles.Nature2010;46410671070.
[15] AisenP,LeibmanA,ZweierJ.Stoichiometricandsitecharacteristicsofthebinding
ofirontohumantransferrin.TheJournalofbiologicalchemistry1978;25319301937.
[16] BakerE.N,BakerH.M,KiddR.D.Lactoferrinandtransferrin:functionalvariations
onacommonstructuralframework.Biochemistryandcellbiology=Biochimieetbi
ologiecellulaire2002;802734.
[17] CazzolaM,BergamaschiG,DezzaL,ArosioP.Manipulationsofcellularironmetab
olism for modulating normal and malignant cell proliferation: achievements and
prospects.Blood1990;7519031919.
[18] Li H, Sun H, Qian Z. M. The role of the transferrintransferrinreceptor system in
drugdeliveryandtargeting.Trendsinpharmacologicalsciences2002;23206209.
[19] LiH,QianZ.M.Transferrin/transferrinreceptormediateddrugdelivery.Medicinal
researchreviews2002;22225250.
[20] PonkaP,LokC.N.Thetransferrinreceptor:roleinhealthanddisease.Theinterna
tionaljournalofbiochemistry&cellbiology1999;3111111137.
[21] CiechanoverA,SchwartzA.L,DautryVarsatA,LodishH.F.Kineticsofinternaliza
tion and recycling of transferrin and the transferrin receptor in a human hepatoma
cell line. Effect of lysosomotropic agents. The Journal of biological chemistry 1983;
25896819689.
319
320
[22] Klausner R. D, Van Renswoude J, Ashwell G, Kempf C, Schechter A. N, Dean A,

BridgesK.R.ReceptormediatedendocytosisoftransferrininK562cells.TheJournal
ofbiologicalchemistry1983;25847154724.
[23] LaoB.J,KameiD.T.Improvingtherapeuticpropertiesofproteindrugsthroughal
terationofintracellulartraffickingpathways.Biotechnologyprogress2008;2427.
[24] Yazdi P. T, Murphy R. M. Quantitative analysis of protein synthesis inhibition by
transferrintoxinconjugates.Cancerresearch1994;5463876394.
[25] WenningL.A,YazdiP.T,MurphyR.M.Quantitativeanalysisofproteinsynthesis
inhibitionandrecoveryinCRM107immunotoxintreatedHeLacells.Biotechnology
andbioengineering1998;57484496.
[26] YazdiP.T,WenningL.A,MurphyR.M.Influenceofcellulartraffickingonprotein
synthesisinhibitionofimmunotoxinsdirectedagainstthetransferrinreceptor.Can
cerresearch1995;5537633771.
[27] GreenfieldL,JohnsonV.G,YouleR.J.Mutationsindiphtheriatoxinseparatebind
ingfromentryandamplifyimmunotoxinselectivity.Science1987;238536539.
[28] Weaver M, Laske D. W. Transferrin receptor ligandtargeted toxin conjugate (Tf
CRM107)fortherapyofmalignantgliomas.Journalofneurooncology2003;65313.
[29] NaglichJ.G,MetherallJ.E,RussellD.W,EidelsL.Expressioncloningofadiphthe
riatoxinreceptor:identitywithaheparinbindingEGFlikegrowthfactorprecursor.
Cell1992;6910511061.
Chapter 17
Tissue Engineering and

Regenerative Medicine for Bone Regeneration
Hideharu Hibi and Minoru Ueda
1.Introduction
The terms tissue engineering and regenerative medicine originally described the concepts for
producingtissueinvitroandinvivo,respectively.However,thegoalofbothapproachesisto
promotehealingandtissueregenerationandfunctionmorepredictably,morequickly,and
lessinvasivelythanallowedbyprevioustechniques.Regardlessoftheseconcepts,successful
outcomesdependonthestrategicemploymentofelementsfortissueregenerationandmay
leadtothedevelopmentofalternativetechniquestowholeorganandtissuetransplantation
fordiseased,failed,ormalfunctioningorgans.Ourresearchprojects,includingthoseintro
ducedinchapter8,rangeoverautologousstemcells,intelligentscaffolds,andvariouscell
signalingpathwayswithregardtotissueengineeringandregenerativemedicine,frombasic
principlestoclinicalapplications.Thischapterpresentsalandscapeoftissueengineeringand
regenerativemedicineforboneregeneration.
2.Principlesoftissueregeneration
Tissueregenerationrequires3keyelements:1)cellsthatareharvestedanddissociatedfrom
thedonortissue,2)scaffoldsubstratesasbiomaterialsinwhichcellsareattachedandcultured
resultingintheimplantationatthedesiredsiteofthefunctioningtissue,and3)cellsignals
thatpromoteorpreventcelladhesion,proliferation,migration,anddifferentiationbyupre
gulatingordownregulatingthesynthesisofproteins,growthfactors,andreceptors(Figure
1). Combined with these 3 key elements at different rates, timings, and so on, the clinical
strategyoftissueregenerationmainlycomprisescelltransplantation,implantationofbioarti
ficialtissueconstructs,andchemicalinductionofregenerationfromtissueatthesiteofinjury.
Accordingtothenatureofthetargettissueandtheextentofthedamagetoberepaired,the
treatment strategy needs to be modified. For bone tissue regeneration, mesenchymal cell
2013 Hibi and Ueda; licensee InTech. This is an open access article distributed under the terms of the
322
transplantationandosteogenicchemicalinductionarepreferredforsmalltissuedeficiencies,
whilebioartificialmineralconstructimplantationisconsideredsuitableforlargedeficiencies.
Figure 1. Elements required for tissue regeneration
3.Researchprojects
Thefollowingresearchprojectsareintroducedinthepresentchapter:1)clinicalresearchon
tissueengineeredosteogenicmaterial(TEOM),2)regulationofcellsignalingpathways,3)cell
transplantationforimprovingbonequality,and4)TEOMforpromotingboneformation.
3.1.ClinicalresearchonTEOM
TEOM,ie,injectabletissueengineeredbone,isagelcomprisinginvitroexpandedautologous
bonemarrowstromalcells(BMSCs)andplateletrichplasma(PRP).Alphagranulesinthe
plateletsreleasevariousgrowthfactorsandtheplasmaformsafibrinnetwork,whichserves
ascellscaffold.TranslationalresearchesonTEOMhaveproducedpositivedataandTEOMis
consideredapromisingmaterialforboneformation[14].Yamadaetal.reportedaclinical
study on injectable tissueengineered bone for maxillary sinus floor augmentation and
simultaneousplacementofdentalimplants(Figure2).Theirstudyincluded16sinusaugmen
tationsin12patientswhosealveolarcrestboneheightwas2to10mm.All41dentalimplants
installedwereclinicallystableatthetimeofsecondstagesurgery.Radiographyat2yearsafter
implantinstallationshowedincreasedmineralizedtissue(mean,8.81.6mm)comparedwith
thepreoperativeboneheight,andneitheradverseeffectsnorremarkableboneabsorptionwere
seen in the 2 to 6year followup periods. These results support the feasibility of using
injectabletissueengineeredboneforsuccessfulboneformationanddentalimplants.
Tissue Engineering and Regenerative Medicine for Bone Regeneration
Figure 2. Treatment procedures with tissue-engineered bone A: Preoperative view. B: Magnified view of dental im
plants placed into the area of maxillary sinus floor augmentation and dental implant thread exposure. C: Tissue-engi
neered bone injection. D: Magnified view during second-stage surgery at 6 months after implant placement. The
exposed thread is surrounded by newly formed bone (white arrow). The green arrow indicates the line of bone biop
sy. E: Provisional prosthesis. F: Micrograph of the regenerated bone area (4); right-upper panel (20). G: Orthopanto
mogram taken immediately after first-stage surgery. Line of substratum of the maxillary sinus mucosa (arrow). H:
Orthopantomogram taken at 6 months after first-stage surgery. Line of maxillary sinus mucosa after sinus augmenta
tion (arrow). The left lower panel is a computed tomography (CT) image indicating radiodensity. (From [1]. Reprinted
with permission).
323
324
3.2.Regulationofcellsignalingpathways
Improvements in cell culture efficiency may contribute to a shorter treatment time and
decreaseitscost.Katagirietal.examinedthecultivationprocessforhumanmesenchymalstem
cells(MSCs)byregulatingtheWntsignalingpathway[5].Wntsignalingwasactivatedand
inhibited with LiCl and secreted frizzledrelated protein3 (sFRP3), respectively. Human
MSCswereexaminedforproliferation(cellcountingandBrdUassays),osteogenicdifferen
tiation(alizarinredstaining),andosteogenicgeneexpressionondays7and14afterinduction
(reversetranscriptionpolymerasechainreaction[RTPCR]andquantitativerealtimeRTPCR
analyses). Cell counting and BrdU assays showed higher proliferation rate of LiCltreated
MSCs than of untreated MSCs. Alizarin red staining showed that sFRP3treated MSCs
mineralizedearlier(onday7)thanuntreatedMSCs(Figure3).BothRTPCRanalyses(ondays
7 and 14) demonstrated that sFRP3treated MSCs expressed higher levels of the alkaline
phosphatase gene than untreated MSCs. These results suggest that regulation of the Wnt
signalingpathwaycontributestocellproliferationandosteogenicdifferentiationofMSCs.This
studyimpliesthateffectiveandefficientboneregenerationtechnologiescanbedevelopedby
regulatingtheWntsignalingpathway.
Figure 3. Alizarin red staining of sFRP-3-treated human mesenchymal stem cells (MSCs) during osteogenic differentia
tion A: During osteogenic induction of MSCs, more mineralization occurred in MSCs treated with secreted frizzledrelated protein-3 (sFRP-3) than in untreated MSCs on days 7 and 14. B: Phase-contrast macroscopic views (40). (From
[5]. Reprinted with permission).
3.3.Celltransplantationforimprovingbonequality
Low bone quality is a risk factor for successful osseointegration, a prerequisite for dental
implants, and typically occurs in osteoporosis. Osteoporosis is a systemic skeletal disease
leadingtofragileboneswithdecreasedmicrostructuresduetothepostmenopausaldecrease
inestrogensecretion.Okamotoetal.investigatedwhetherBMSCscanpromotebonehealing
aroundtitaniumimplantsinratosteoporosismodels[6].SpragueDawleyratsweredivided
into3groups:thefirstgroupinwhichtheovarieswereremoved(OVXgroup),thesecond
groupinwhichashamsurgerywasperformed(SHAMgroup),andthethirdgroupinwhich
BMSCs were transplanted to an OVX group (OVXBMSCs group) (Figure 4). In the OVX
BMSCsgroup,1105BMSCsweretransplantedintothefemurwithimplant.Eachvalueof
thebonetoimplantcontactandtheboneareaofeachcorticalboneandcancellousbonewere
obtained histomorphometrically. Bone density was measured over 500 m distally and
proximally from the implants. Each ratio of bonetoimplant contact, bone area, and bone
densityintheOVXBMSCsgroupwassignificantlyhigherthanthoseoftheOVXgroupas
comparedtothecancellousbone.BMSCtransplantationtherapyimprovedlocalbonehealing
inthecancellousbonesurroundingimplantsandalsosignificantlyimprovedbonebinding
withimplants.
Figure 4. Experimental design (left) and histological evaluation of the distal femur at 28 and 56 days after implanta
tion (right) A-F: Micrographs of the distal femur at 28 days after implantation. A and D: SHAM; B and E: OVX (ovaries
had been removed); C and F: OVX-BMSCs (ovaries had been removed and bone marrow stromal cells had been trans
planted), (toluidine blue stain, A-C 1.25, bar = 1.1 mm; D-F 10, bar = 200 m). In the cortical bone area, most of the
implant surface was in direct contact with newly generated bones, and the thread was widely filled with bone tissues
in all groups. No remarkable difference was observed among the groups. Slightly more osteogenesis was observed in
the OVX-BMSCs group than in the OVX group. The osteogenetic area was remarkably smaller in the OVX group than
in the SHAM group outside the thread. In addition, osteogenesis was observed at a higher rate in the OVX-BMSCs
group than in the OVX group outside the thread. G-L: Micrographs of the distal femur at 56 days after implantation. G
and J: SHAM; H and K: OVX; I and L: OVX-BMSCs, (toluidine blue stain, G-I 1.25, bar = 1.1 mm; J-L 10, bar = 200 m).
In the cortical bone area, the area on the implant surface that was in contact with newly generated bones was greater
than that observed at 28 days, and the amount of bone tissue inside the thread had increased in all groups. In the
cancellous bone area, the amount of bone that was in contact with the implant surface had increased in all groups.
There was no remarkable difference between the SHAM group and the OVX-BMSCs group. In the OVX-BMSCs group,
trabecular structures were observed, presenting as web-like pattern. (From [6]. Reprinted with permission).
325
326
Figure 5. Two-step transport distraction osteogenesis (DO) for the posterior body and ramus of the mandible A: Man
dible with recurrent ameloblastoma. Computed tomography (CT) revealed a multilocular radiolucent area in the ra
mus and posterior body of the mandible. B: Custom-made distraction device with an artificial condyle on the patients
stereolithographic model. Locking plates and screws enabled supraperiosteal fixation of the transport segment. C: Su
praperiosteal fixation of the distraction device. A short buccal periosteal flap of the transport segment was reposi
tioned and sutured to cover the vertical osteotomy line. The transport segment was fixed with locking plates and
screws supraperiosteally. D: Regenerated posterior mandibular body. Eight months after distraction, CT showed that
the newly formed 50-mm-long bone in the distraction gap (arrowheads) progressed to continuous buccal and lingual
cortical surfaces. E: Secondary vertical DO for reconstructing the ascending ramus. A short buccal periosteal flap of the
secondary transport segment was repositioned and sutured to cover the horizontal osteotomy line. The transport seg
ment was fixed with another distraction device positioned supraperiosteally. F: Another transport segment. Immedi
ately after surgery, CT showed fixation of the secondary transport segment with the distraction device. G:
Regenerated ascending ramus. Nine months after the secondary distraction, CT showed that newly formed bone in
the distraction gap progressed to continuous buccal and lingual cortical surfaces of the ramus. H: Device removal and
mandibular reconstruction. CT showed that the regenerated ramus was connected to the preserved condylar segment
with a lag screw and plate. I: Regenerated mandibular body containing implants. CT showed 3 screw-type implants
installed in the right second premolar and molar regions with a buccal veneer bone graft fixed with 2 screws. J: Recon
structed mandible and osseointegrated implants for occlusal function. CT showed that the mandibular angle was ad
justed with osteotomy and screw fixation in the angle region. K: Provisional implant-supported prosthesis on the
reconstructed mandible. The ridge supporting the implants was surrounded by the attached mucosa, the buccal part
of which originated from the palatal mucosa. (From [8]. Reprinted with permission).
3.4.TEOMforpromotingboneformation
Distractionosteogenesis(DO)isconsideredinvivotissueengineeringandisatechniquethat
provides autologous and predictable bone formation. Hibi & Ueda developed an internal
devicewithwhichtheywereabletoreconstructamandibularsegmentalbonydefectofmore
than10cm[7,8].(Figure5).
DOisausefultechniqueforreconstructingbonydefectswithoutperforminggraftingproce
dures; however, it requires a long treatment time that includes latent, lengthening, and
consolidation periods. To shorten these periods, attempts of applying tissue engineering
technologies for DO have been made [9, 10]. Kinoshita et al. investigated whether locally
injected TEOM can promote bone regeneration in a rabbit highrate DO model (Figure 6).
Bilateralosteotomieswereperformedinthemaxilla;distractiondeviceswereactivatedata
rateof2.0mmoncedailyfor4daysaftera5daylatencyperiod.Twelverabbitsweredivided
into2groups.Attheendofdistraction,theexperimentalgroupofrabbitsreceivedaninjection
ofTEOMintothedistractedtissueononeside,whereassalinesolutionwasinjectedintothe
distractedtissueonthecontralateralside(asinternalcontrol).Anadditionalcontrolgroup
receivedaninjectionofPRPorsalinesolutionintothedistractedtissueinthesamewayasthe
experimentalgroup.Thedistractionregenerateswereassessedbyradiologicalandhistomor
phometricanalyses.TheradiodensityofthedistractiongapinjectedwithTEOMwassignifi
cantlyhigherthanthatinjectedwithPRPorsalinesolutionat2,3,and4weekspostdistraction.
Thehistomorphometricanalysisalsoshowedthatbothnewbonezoneandbonycontentin
thedistractiongapinjectedwithTEOMweresignificantlyincreasedwhencomparedwithPRP
orsalinesolution.TheseresultsdemonstratedthatthedistractiongapinjectedwithTEOM
showed significant new bone formation. Therefore, injections of TEOM may be able to
compensateforinsufficientdistractiongaps.
Figure 6. Experimental design (left) and radiographic view of the distracted maxilla (right) A: Two weeks after tissueengineered osteogenic material (TEOM) injection. B: Four weeks after TEOM injection. C: Two weeks after injection of
saline solution. D: Four weeks after saline solution injection. E: Two weeks after platelet-rich plasma (PRP) injection. F:
Four weeks after PRP injection. G: Two weeks after injection of saline solution. H: Four weeks after injection of saline
solution. (From [9]. Reprinted with permission).
327
328
4.Conclusionandfutureperspective
The research projects presented in this chapter and chapter 8 cover autologous stem cells,
intelligentscaffolds,andvariouscellsignalingpathwayswithregardtotissueengineering
and regenerative medicine, from basic principles to clinical applications. Future research
shouldfurtherclarifytheimportanceoffunctionsandmutualrelationshipsofthistriadon
various levels. In addition to the elements of this triad, efficient tissue regeneration and
maintenanceoftheregeneratedtissuerequiresufficientbloodsupplyandmechanicalstress,
respectively(Figure7).Theseadditionalelementshavebeenalsoapproachedinourongoing
researchprojects;allcombinedwillproducedramaticadvancesinourabilitytoregenerate
bony tissues, multilaterally improve the quality of life for all, and provide brighter future
perspectivesinthefieldoftissueengineeringandregenerativemedicine.
Figure 7. Elements required for efficient tissue regeneration and maintenance of the regenerated tissue
Authordetails
HideharuHibiandMinoruUeda
DepartmentofOral&MaxillofacialSurgery,GraduateSchoolofMedicine,NagoyaUniver
sity,Nagoya,Japan
References
[1] YamadaY,NakamuraS,ItoK,KohgoT,HibiH,NagasakaT,UedaM.Injectabletis
sueengineeredboneusingautogenousbonemarrowderivedstromalcellsformaxil
larysinusaugmentation:clinicalapplicationreportfroma26yearfollowup.Tissue
EngineeringPartA2008;14(10)16991707.
[2] UedaM,YamadaY,KagamiH,HibiH.Injectableboneappliedforridgeaugmenta
tion and dental implant placement: human progress study. Implant Dentistry
2008;17(1)8290.
[3] Ueda M, Yamada Y, Hibi H. Osteoperiosteal tissueengineered injectable bone. In:
TheOsteoperiostealFlap:Asimplifiedapproachtoalveolarbonereconstruction.Jen
senO.T.(Ed.),Illinois,USA:QuintessencePublishing;2010.
[4] ItoK,YamadaY,NaikiT,UsamiK,MizunoH,OkadaK,NaritaY,AokiM,KondoT,
MizunoD,MaseJ,NishiguchiH,KagamiH,KinoshitaK,HibiH,NagasakaT,Ueda
M. Bone. In: Applied Tissue Engineering, Ueda M., (Ed.), pp.2145, Rijeka: InTech;
2011.
[5] KatagiriW,YamadaY,NakamuraS,ItoK,HaraK,HibiH,UedaM.Regulationof
the Wnt signaling pathways during cell culture of human mesenchymal stem cells
forefficientboneregeneration.OralScienceInternational2010;7(2)3746.
[6] OkamotoY,TateishiH,KinoshitaK,TsuchiyaS,HibiH,UedaM.Anexperimental
study of bone healing around the titanium screw implants in ovariectomized rats:
Enhancementofbonehealingbybonemarrowstromalcellstransplantation.Implant
Dentistry2011;20(3)236245.
[7] Hibi H, Ueda M. Supraperiosteal transport distraction osteogenesis. In: The Osteo
periosteal Flap: A simplified approach to alveolar bone reconstruction. Jensen O.T.,
(Ed.),Illinois,USA:QuintessencePublishing;2010.
[8] Hibi H, Ueda M. Supraperiosteal transport distraction osteogenesis for reconstruct
ing a segmental defect of the mandible. Journal of Oral and Maxillofacial Surgery
2011;69(3)742746.
[9] KinoshitaK,HibiH,YamadaY,UedaM.Promotednewboneformationinmaxillary
distraction osteogenesis using a tissueengineered osteogenic material. Journal of
CraniofacialSurgery2008;19(1)8087.
[10] JensenO.T,LasterZ,HibiH,YamadaY,UedaM.Alveolardistractionosteogenesis
andtissueengineering.In:TissueEngineering:Applicationsinoralandmaxillofacial
surgery and periodontics, 2nd edition, Lynch S. E, Marx R. E, Nevins M, Wisner
LynchL.A.(Eds.),Illinois,USA:QuintessencePublishing;2010.
329
Chapter 18
MEMS Sensors and Their Applications
Mitsuhiro Shikida
1.Introduction
Microelectromechanicalsystems(MEMS)technologyhasopenedthedoorwaytominiatur
izingandintegratingvariousmechanicalandelectricalcomponentsonthesamechip.The
principal fabrication technologies for producing MEMS devices, such as anisotropic wet
etchingforsinglecrystalsilicon,sacrificialetchingforproducingmechanicalelements,and
anodicbondingforthepackaging,wereinventedinthe1960s.Thesetechnologieshavebeen
usedforproducingvariousminiaturizedsensorsfromthe1970sonwards.Silicondiaphragm
structures produced by anisotropic wet etching are used as sensing elements for pressure
sensors.Anionsensitivefieldeffectivetransistorwasdevelopedasachemicalsensor.MEMS
sensorsstartedtobeintegratedonsiliconwafersinthemiddleofthe1970s.Apressuresensor
anditselectricalcircuitsareintegratedonthesamesiliconwaferinordertoreducetheparasitic
capacitance and footprint. During this time, a miniaturized gas chromatograph for use in
aerospace was fabricated on a 2inch silicon wafer; it integrated a column, valve seat, and
sensing heater. Moreover, a massflow controller driven by piezoelectric actuators for
preciselyregulatinggasflowinapipewasdevelopedinthelate1980s.Sincethe1990s,MEMS
technologies have spread to many different fields, and these days there are miniaturized
sensors,forexample,foracceleration,gyroscopic,flow,infrared,andtactilesensors.
1.1.AttractivefeaturesofMEMSsensors
MEMStechnologiescanproduceminiaturizedsensorsafewmminsizethatofferthefollowing
veryusefulfeaturesforindustry.
1.
Lowercosts:MEMSsensorsarenormallyproducedinbatcheslikesemiconductorchips;
asaresult,theyofferlowerfabricationcosts.
2.
Highspaceandtimeresolution:MEMSsensorsaretypicallymeasurelessthanafewmm.
Thismeanstheycanhaveverysmallfootprints,whichinturnraisesthespatialresolution.
Thevolumereductionaffordedbythemcanbeusedtoincreasethetimeresolutionof
2013 Shikida; licensee InTech. This is an open access article distributed under the terms of the Creative
332
thermal sensor applications. This is because the heat capacity decreases as the sensor
volumedecreases.
3.
Small parasitic capacitance and dead space: MEMS sensors can be integrated with
electrical circuits on the same silicon wafer, and thus, one can reduce the parasitic
capacitanceinthesensingcircuit.MEMStechnologiescanintegratedifferentcomponents,
forexample,sensorsandactuators,onthesamewafertobuildupthesystem,andthus,
theymakeitpossibletoreducethedeadspaceinthesystem.
4.
2Danalysis:MEMSsensorscanbeintegratedonsiliconwaferintheformofanarray,and
thus,onecananalyzethedistributionofthephysicalquantities,forexample,pressure
andforce,actingonaplanewithhighspatialresolution.
1.2.DifficultiesindevelopingMEMSsensors
Despite the abovementioned advantages, MEMS sensors face a number of difficulties in
regardtotheircontinueddevelopment.
1.
Large capital investment: The MEMS technologies, especially in those used in the
fabricationprocess,arebasedonsemiconductortechnologiesthathavebeendeveloped
forproducing3Dmicrostructuresfromsiliconwafers.Alargeamountofcapitalinvest
mentisgenerallyrequiredinthefabricationofMEMSsensors.
2.
Variationsinthefabricationprocess:MEMSsensorstypicallyhavecantilever,diaphragm,
orcombstructures.Theyaresuspendedandactuatedtodetectphysicalquantities.Many
technologiesarerequiredforproducingsuchstructures.
a.
Sacrificialetchingisusedtomakemovableelementsonasiliconwafer.Partsofthe
deposited layers are selectively etched away. This etching is normally done by
isotropic wet etching, and it is especially used to produce combdrive actuators.
Stickingbetweenthereleasedmovableelementandthesubstrateunderneathitis
sometimesaproblem,butitcanbesolvedbyusingaspecializeddryingmethodand
thinfilmdeposition.
b.
Anisotropicwetetchingisusedtoproducediaphragmandcantileverstructures.The
microstructures are formed by orientationdependent etching of single crystal
silicon.
c.
Dryetchingisnormallyusedtomakecombdriveactuators.Toproducethedesigned
structureprecisely,onehastoovercomeloadingandnotcheffects.
d.
VariationinpackagingisanissuesinceeachMEMSsensorrequiresanordermade
packagingprocess.Forinstance,tactileforceandflowvelocitysensorshavetobe
exposedtotheirenvironmentinordertodetectphysicalandchemicalquantitiesof
interest.Ontheotherhand,accelerationandgyrosensorsarecompletelyencapsu
latedinashellstructure.Thepressureinsidetheshelliskeptatavacuumlevelin
ordertoincreasetheQfactorattheresonantfrequency.
e.
3.
Compatibilityofspecializedfabricationprocesses,suchthosedescribedin(a)to(c),
with semiconductor electrical circuit fabrication processes has to be carefully
consideredwhenMEMSsensorsaretobeintegratedwithelectricalcircuits.
Multidisciplinary knowhow: MEMS sensors development requires multiphysics

knowledge.Forexample,developersofaccelerationsensorsneedathoroughknowledge
ofthefollowingareas:
a.
Mechanics for designing an adequate deformation vs. force relation and ensuring
reliability,
b.
Vibrationanalysisfordesigningtheresonantfrequencyofthedevice,
c.
Electricalcircuitanalysisfordesigningthesensingcircuitsandelectricalproperties
(capacitanceandresistance)ofthesensingstructures,and
d.
Fabricationtechnologyforproducingthesensors.
The multiphysics approach is taken because the above points are related to each
other.
4.
Scalingeffects:Thesizesofstructuralelementssuchascantileversanddiaphragmsare
ontheorderof10micrometers.Suchalevelofminiaturizationmeansthatforcesthatare
proportionaltotheareaofthestructuralelementdominatethemotionofthesensor.Other
forces, such as magnetic and inertial forces which are proportional to the volume of
elements,aregreatlyreducedasaresultofminiaturization.Theeffectsoftheelectrostatic
forceandsurfacetensionhavetobecarefullyconsideredwhendesigningMEMSsensors.
1.3.ApplicationsofMEMSsensors
MEMStechnologieshaveawidevarietyofapplications,andtherearemanytypes(examples
areshowninTable1).Inthefollowingsections,tactilesensorsforhumaninterfacesandflow
sensorsforairconditioningsystemsandmedicalapplicationswillbefocused.
2.Tactilesensorsforhumaninterfaces
The single crystal silicon is often used as a MEMS material because it is a highquality
semiconductorwithexcellentmechanicalproperties[1].Itissaidthattheyieldstrengthof
siliconiscomparabletothatofstainlesssteel.Siliconsensorsforgyroscopesandformeasuring
pressure,acceleration,flow,andinclinationhavealreadybeencommercializedandarebeing
usedintheautomotiveandinformationindustries[25].Siliconhasalsobeenusedtomake
miniaturizedtactilesensorsasinterfacedevicesinroboticautomationsystemssinceinthe
middleofthe1980s.VarioustactilesensorsbasedonsiliconMEMStechnologieshavebeen
developed[69].Inparticular,Takaoetal.integratedatactilesensorwithaCMOSdeviceon
thesamechip[10].Moreover,Shikidaetal.proposedtointegrateanactuatormechanismonto
atactilesensorasmeansofmeasuringthehardnessofobjects[1113].
333
334
Table 1. MEMS sensor applications
MEMSsensorsarebeingusedtostudyhumanactivitybyembeddingtheminclothesand
everyday consumer products. Silicon tactile sensors have high space and time resolutions.
However,itisdifficulttomakethemwearablebecausesiliconissobrittle.Toovercomethis
problem, MEMS researchers have started using resin materials to fabricate flexible sensor
structures[1417].Notably,Konishietal.andN.Chenetal.devisedknittedstructuresfor
electricalwiring[18,19].Thesestructureshaveadvantageswherebytheycanfitonarbitrary
bendablesurfaces.Shikidaetal.proposedanartificialhollowfiberasaMEMSmaterialto
implementatactilesensingfunctiononafabricanddevelopedatactilesensorbyweaving
together fibers [2023]. They studied the units ability to detect force, crosstalk, etc., by
applyingconcentratedloadstoit.
Thefollowingsectionsfocusonsiliconbasedandfabrictactilesensors.
2.1.Sibasedtactilesensors
AschematicviewofamagneticallydriventactilesensorisshowninFigure1.Itconsistsofa
diaphragmwithamesaatthecenter,piezoresistivestrainsensorsfordetectingthedisplace
ment of the diaphragm at the periphery, a permanent magnet on the backside of the dia
phragm,andaflatcoil.Themagneticactuationsystemisformedbypairingthepermanent
magnetandtheflatcoil,andthesensingdiaphragmisdrivenupanddownbythemagnetic
forcegeneratedbythisactuationsystem.Thesensorisusedintwodifferentmodes,asshown
inFigure2.
Figure 1. Magnetically driven Si-based tactile sensor. Republished with permission of IOP Published Ltd from Ref. [13].
Figure 2. Detection principle of Si-based active tactile sensor. Republished with permission of IOP Published Ltd from
Ref. [13].
335
336
Quasistaticmode
Inthismodeofoperation,thetactilesensordetectstwophysicalquantities,thecontactforce
andtheelasticityofthecontactedobject.Thevaluesaremeasuredwhentheforcesworking
ontheobjectandtheactuatingmagneticforceareinequilibrium.Thedetectionprincipleisas
follows.Themesastructureonthediaphragmmovesonlyonedirectioninthismode.When
themesasonthediaphragmofthetactilesensorarrayscomeintocontactwiththeobject,some
ofthemesastructuresarepushedinwardbythebumpysurfaceoftheobject(Figure2(a)).The
contact force is detected by piezoresistive strain sensors measuring the deflection of the
diaphragm.Theforcedistributionanda2Dsurfaceimageoftheobjectcanbegeneratedby
usinganarrayofsensors.Afterthecontactforcedetection,thecontactingmesasaredriven
againsttheobjectbythemagneticactuationsystemtomeasurethehardnessoftheobject.As
aresult,thecontactingregionsaredeformedinawaythatdependsontheelasticityofthe
object.Thus,onecancalculatetheelasticityoftheobjectbyusingtherelationshipbetweenthe
displacementofthediaphragmandtheappliedforce,whichismeasuredquasistatically.
Vibrationmode
Inthismode,themesastructureisalternatelydrivenupanddownattheresonancefrequency
ofthediaphragm.Thesensorcanestimatetheelasticanddampingcoefficientsoftheobjectin
contact.Eachsensingdiaphragmisvibratedattheresonantfrequencybymagneticforce,as
shown in Figure 2(b). Sensor outputs, such as the amplitude of the diaphragm, resonant
frequency, and phase shift, change in accordance with the mechanical properties of the
contactedobject.Thelumpedparametercircuitmodelofthissensorsystemandtheformulas
fortheresonantfrequencyandQfactorareshowninFigure2(b3).Theresonantfrequency
and Qfactor depend on the elastic coefficient and equivalent mass of the object (see the
formulas).Theequivalentmass,ms,dependsonthevibrationamplitude,whichiscontrolled
with the measurements. The detection procedure depends on the magnitude of difference
betweenthedevicemassmd,andobjectmassms,asfollows:
1.
ms<<md
In this case, the vibrating mass of the sample is much less than that of the mesa with the
diaphragm.Themassofthemesaanddiaphragmdominatesthetotalmassinvibration;thus,
onecanestimateelasticanddampingcoefficientsofthecontactedobjectwiththefollowing
procedure.
a.
ElasticcoefficientKs
Theresonantfrequencydependsonlyontheelasticcoefficientoftheobject,anditincreases
astheelasticcoefficientKsincreases,asshowninFormula1.Thus,onecancalculateKsofthe
contactedobjectsimplybymeasuringthechangeintheresonantfrequency.
b.
DampingcoefficientCs
The Qfactor of the vibrating diaphragm structure depends on the elastic and damping
coefficients(KsandCs),asshowninFormula2.Thus,onecanestimateCsofthecontactedobject
fromtheQfactorandKs.
2.
msisnotnegligiblecomparedtomd
Inthiscase,thesamplesequivalentmassisnotnegligiblecomparedtothatofthemesawith
diaphragm.Thus,theresonancecharacteristics,suchastheresonantfrequencyandtheQ
factor,dependontheequivalentmassandtheelasticcoefficientofthecontactedobject.The
followingprocedureisusedtoestimatephysicalquantities.
a.
ElasticcoefficientKs
BothoperationmodesarerequiredtoestimateKsandtheequivalentmassmsoftheobject.
First,thesensorpreciselydetectstheelasticcoefficientoftheobjectusingquasistaticmode.
Then,theequivalentmassoftheobjectisestimatedfromtheelasticcoefficientoftheobject
andresonantfrequencyobtainedinvibrationmode.
b.
DampingcoefficientCs
CsofthecontactedobjectisestimatedfromKs,ms,andtheQfactor.
Thefabricationprocessandthesensorcharacteristicsdrivenbythetwomeansofactuation
aredescribedinthereferences[1113].
2.2.Fabrictactilesensors
Figure 3(a) shows the applications of fabric tactile sensors. These sensors can be used for
monitoring the force distribution at human interfaces. A schematic view and operation
principleareshowninFigures3(b)(d).Anartificialhollowfiberismadefromasingleelastic
hollowfibercoveredwithmetalandinsulationlayers.Thesensorisfabricatedbyweavingthe
artificialhollowfibersinareticularpattern.Theappliednormalloadisdetectedbymeasuring
thechangeincapacitancebetweenthewarpandweftfibersattheirintersectionpoint.When
a normal load is applied to an intersection point of the fabric tactile sensor, the resulting
deformationincreasesthecontactareabetweenthewarpandweftfibers.Thus,thecapacitance
betweenthemincreasesastheappliedloadincreases.Thefabrictactilesensoristhereforeable
todetectthenormalloadbymeasuringthiscapacitancechange.Thisfabrictactilesensorhas
thefollowingadvantages.
1.
Tactilesensingatarbitrarysurface
Thefabricsensorcanfitonanysurface,anditcanperformtactilesensingbyweavingthefibers
asiftheywerecloth.
2.
Wearablesensors
Wearabletactilesensorscanbemadebydirectlyweavingartificialhollowfibersintoclothes,
gloves,andshoes,orbypatchingthemtotheseitems.
3.
Detecting2Dcontactforcedistributions
Thesensorcandetectthe2Dnormalloaddistributionbysequentiallymeasuringthecapaci
tancechangesatallintersectionpoints.Theresolutionofthe2Dnormalforcedistributionis
337
338
Figure 3. Fabric tactile sensor and its applications. Republished with permission of IOP Published Ltd from Ref. [21].
determinedbythegridspacingbetweenthewarpandweftfibers,anditiseasilyadjustedfor
anypurposebycombiningtheartificialhollowfiberswithcottonyarn.
4.
Tactilesensingoveralargesurfacearea
SiliconbasedMEMSsensorshavelimitedareacoverage,becausetheyuseasmallsiliconwafer
asasubstrate.However,fabrictactilesensorsconsistoffibers;therefore,theycanbemadeas
bigasneededfortheapplication.
A large rectangular fabric tactile sensor was fabricated by weaving together hollow fibers,
brasswires,andconventionalcottonyarn(0.3mmindiameter).Thefabricsensorduringthe
weavingprocessisshownintherightofFig.4(a).Figure4(b)showsoneoftheproducedfabric
tactilesensors.Thesensormeasures87.5mmx52.0mm,anditssensingpointsarearranged
ina4x4array.Thepitchofdetectionpointsis7.0mmhorizontallyandvertically.
Figure 4. Fabrication of fabric tactile sensor. Republished with permission of The Institute of Engineering and Technol
ogy from Ref. [23].
Figure 5. Applied force and shape detection using a fabric tactile sensor. Republished with permission of The Institute
of Engineering and Technology from Ref. [23].
Afabrictactilesensormeasuring180.0mmx56.0mmwasevaluatedastoitsshapedetection
capability.Theunithada4x10arrayofsensingpoints(totalnumber:40).Thenumberofthe
warpwiresandweftfibersvariedfromAtoDand1to10,respectively.Thus,theintersection
pointswereeachlabeledwithaletterandnumbercombination(seeFig.5).Thepitchofthe
detectionpointswas18.0mmhorizontallyand14.0mmvertically.Figure5showshowthe
appliedpressureandtheobjectsshapearedetected.Asquareplatemeasuring20mmx20
mmwasplacedonthesensor(theareafromB3toB4andfromC3toC4),andaforceof2925
mNwasappliedtoit(Figure5(a)).Twodifferentmetalweightswereplacedonthesensor
(Figure5(b)).Aheavyweightof269gandalightweightof187gweredirectlyplacedonthe
areafromB3toB4andfromC3toC4andtheareafromB7toB8andfromC7toC8.Both
weightswerethesamesize,i.e.,24.2mmx24.2mm.Thefabricsensorsuccessfullydetected
thetwodimensionalshapeofthecontactedobjects.
3.Flowsensorsandtheirapplications
MEMSflowsensorsfabricatedonSiwafershavealonghistory,andvarioustypeshavebeen
developed.C.Liuetal.developedathermaltypeofflowsensortodetectshearstress[24].J.
Zheetal.usedamicromachinedcantileverstructureformeasuringshearstressfromtheflow
[25].SiMEMSflowsensorshaveexcellentspaceandtimeresolutions,andtheyareusedin
semiconductorequipmentandinautomobilesforcontrollinggasflowsprecisely.Addition
ally, Unnikrishan et al. developed a MEMSontube assembly to simplify the packaging
process[26].TheyintegratedSiMEMSdevicesdirectlyonaglasstube,whichiscompatible
339
340
withaSwagelokconnector,byfusionbonding.TheseandotherSibasedflowsensorsare
summarizedinthereferences[27,28].Thesensorsareassembledontoarigidflatboardinthe
packaging process, to compensate for the brittleness of the Si material. This means these
sensorsaredifficulttoputoncurvedsurfaces.
ImprovedflexibilityisthekeytomakingMEMSsensorssuitableforalargerrangeofappli
cations, and this can be accomplished by using resin materials as a substrate. Zhu et al.
fabricatedaflowsenordirectlyonaflexibleprintedcircuitboardwithelectricalcircuits,and
mounteditontothecurvedsurfaceofawingtocontrolanaircraft[29].Maetal.,produceda
flexibleflowsensormadeofpolyimidefilmfordetectingthedynamicwaveflowsinawater
channel [30]. For medical applications, Li et al., used Kapton film as a substrate, and they
integratedpressure,temperature,glucose,andoxygensensorsonitforbloodanalysis[31].
Naitoetal.fabricatedaflowsensoronaflexible7.5mthickpolyimidefilmandproduceda
miniaturizedonwallintubeflowsensor[32].Theyusedittofabricateacatheterflowsensor
formeasuringaspiratedandinspiredaircharacteristics[33,34].Recently,flowsensorshave
beenusedtocontrolairsuppliespreciselyinlargescaleairconditioningnetworksystemsfor
thepurposeofreducingwastedenergy.
3.1.Thermalflowsensors
Threeprinciples,thermalanemometry,calorimetricflowsensing,andtimeofflightsensing,
areusedinthermalflowsensing.Thermalanemometry,i.e.,hotwireanemometry,detectsthe
flowratebyitscoolingeffectonaheatedelement.Acalorimetricflowsensor,ontheother
hand,usesaheaterandtwosensingelements;theheaterisplacedbetweenthetwosensing
elements,andtheflowisdetectedfromthedifferenceinheatingeffectsatthetwosensing
elements.Thetimeofflightmethodusesthetravellingtimeofathermalpulsefromtheheater
tothesensingelementfordetectingflow.Themethodsareexplainedindetailinthereferences
[27,28].Hotwireanemometryflowsensorsarewidelyusedbecausetheycanmeasurehigh
flowratesandhavethesimpleststructure.
3.1.1.Hotwireanemometry
Hotwireanemometryusesthecoolingeffectofforcedconvectiontodetectchangesinflow.
Whenaheatedwireisplacedinaflow,theequilibriumtemperatureatthewiredependson
theamountofconvectioninthefluid.Ontheotherhand,theelectricalresistanceofthewire
generallydependsonitstemperature,anditcanbeexpressedas,
R2 R1 1 T2 T1
(1)
whereR1andR2,aretheelectricalresistancesofthewireattemperatureT1andT2,andisthe
temperaturecoefficientofresistance(TCR).Flowchangesaredetectedfromchangesinthe
electricalresistanceofthewire.Therefore,inpractice,thismethodrequiresalargeandstable
TCRvalue.
Asstatedabove,thehotwiremethoddetectsaflowbymeasuringtheheatitdissipatestothe
fluidbyforcedconvection.TherelationshipbetweenthedissipatedpowerQandflowvelocity
UinalowReynoldsnumberflowisgivenbyKingslaw[35]:
Q A BU n Th T0
(2)
whereA,B,andnareconstantsdependingonthegeometryofthewireelement.ThandT0are
thetemperaturesofthewireandinthefluid,respectively.Ifthewirewereinfinitelylong,n
wouldbe0.5.However,ndiffersform0.5inactualwiresoffinitelength.
Twodifferentoperationsarecommonlyusedforheatingthewireelectrically.Oneisconstant
voltage,orcurrentmode.Theflowrateiscalculatedfromthechangeinelectricalresistance,
and this method has the advantage of needing only relatively simple electrical circuits.
However,ithasadrawbackinthatitisdifficulttoshortentheresponsetime.Toimprovethe
responseofthesensingwire,anothermode,i.e.,constanttemperature(CT),isfrequentlyused
inflowratemeasurements.Inthismode,thetemperatureofthewireiskeptconstantbyusing
afeedbackcircuittoshortentheresponsetime,andtheflowrateiscalculatedfromthefeedback
voltage.CTmodeisnormallyusedinpractice.Theelectricalenergysuppliedtothehotwire
inCTmodeisequaltothepowerdissipatedfromthewiretothefluid.Asaresult,equation
(2)ismodifiedasfollows:
Vh2
A BU n Th T0
Rh
(3)
whereVhandRharethevoltagedifferenceandelectricalresistanceatthewire,respectively.
Here,thesquareofthevoltageatthewireisproportionaltothenthpoweroftheflowvelocity.
3.1.2.Onwallintubeflexiblethermalflowsensors
Themostcommonhotwireanemometersareplacedinthecenterofflowchannelstructures.
Thus,thedevicesdetecttheflowratethroughthecenterofatube.Thismeasurementmethod
hasthefollowingdrawbacks.
1.
Requirementoflargeinletlength
Theflowvelocityatthecenteroftubedependsontheflowconditions.Forexample,theflow
distributioninthetubedependsonthedistancefromtheentrance.Thus,thesensorhastobe
placedinthehydraulicallyfullydevelopedflowregiontoobtainaconstantvalue.Thismeans
thatitneedstobeacertaindistancefromthetubeentrance.
2.
Requirementofpositionaccuracy
Thesensoroutputdependsontheradialpositioninthetube,becauseoftheflowratedistri
bution.Thisistrueeveninafullydevelopedflow.
341
342
3.
Difficultyofmeasuringflowratesincurvedtubes
Theflowvelocitydistributionchangesdependingonthecurvatureradiusandangleofthe
tubeandonthedistancefromtheendofthecurvedregion.Thus,thesensorcannotoutputa
constantvalueinacurvedtubebecausethevelocitydistributionisasymmetricintheradial
direction.
Toovercometheseproblems,awallmountedintubethermalflexiblesensorabletomeasure
the flow rate under both hydraulically developing and fully developed conditions was
developed[36,37].Thesensorconsistsofaringshapedheaterelementonaflexiblepolyimide
film(Figure6),anditisfabricatedonapolyimidefilmsubstrateandmountedontheinner
wallsurfaceofthetubetoformaringshapedsensorstructure.
Figure 6. On-wall in-tube thermal flow sensor. Republished with permission of IOP Published Ltd from Ref. [36].
Theadvantagesofthesensorareasfollows.
1.
Itcanmeasureflowsinhydraulicallydevelopingregions
Theflowdevelopsinthedirectionfromtheinnerwallsurfacetothecenterofthetube.The
flowrateneartheinnerwallsurfacebecomessteadyfirst.Thus,thissensorcanmeasurethe
floweveniftheflowisnotfullydevelopedinallregionsinthetube,becauseitmeasuresthe
flow rate only near the inner wall. This means that the sensor can measure even in the
hydraulicallydevelopingregionofthetube,whichshortensthedistancefromtheentrance
requiredbythesensor.
2.
Itcanmeasureaxiallyasymmetricflows.
Theoutputsignalfromthesensorisinsensitivetothevelocitydistributioninthetube,because
ringshaped sensing structure averages out the flow distribution. This means the sensor
outputsasteadyvalueevenifthevelocitydistributionisnotaxialsymmetric.
Theonwallintubeflowsensorisdescribedinthereferences[36,37].Thefollowingsections
describetwoofitsapplications.
3.2.Flowsensorsforairconditioningnetworks
Airconditioningsystemsarelocatedoutsideofbuildingsandtheydelivertreatedairtorooms
through a network of ducts. These ducts usually have complicated configurations and a
numberofbendsbecausetheyhavetogoinsidealimitedamountemptyspacebetweenthe
beamsofthebuildingsceiling.Toreducewastedenergy,theairsuppliedtoeachroomhas
tobecontrolledatanidealsensingpointneartheoutletport.Thismeanstheflowratemust
tobepreciselymeasureddownstreamofthebentduct.
Tomeasuretheflowratedownstreamofabentduct,thefollowingpointshavetobeconsidered
inthesensorsdevelopment.
1.
Theflowvelocitydistributioninthetubeisexpressedasaquadraticfunction.Thismeans
thattheflowvelocityreachesamaximumattheductscenterandaminimumattheinside
surfaceoftheduct.Thus,toreducetheairflowresistanceofthesensoritself,thesensor
hastobeputontheinsidesurfaceoftheduct,anditsthicknesshastobeminimizedas
muchaspossible.
2.
Theflowsensorstructurehastobeflexibleinordertofitontheroundedinsidesurface
oftheduct.
3.
Theairflowdownstreamofabentductiscomplicatedbecauseofthesecondaryflow
causedbythebend.Thismeansthatitisdifficulttomeasuretheflowratebyusinga
singlepointmeasurement.Toovercomethisproblem,anumberofsensorshavetobe
attachedtotheinsidesurface,andtheiroutputsshouldbeaveragedtoreducetheeffect
ofthesecondaryflow.
Apatchtypeflexibleflowsensorwasdevelopedasawaytocontroltheairsupplyprecisely
inlargescaleairconditioningnetworksystems.Thesensorisbasedonpolyimidefilm,andit
was fabricated by photolithography and thinfilm deposition. First, the photoresist was
patternedbyphotolithography,andAu/Crfilmwasdepositedbysputtering.Themetalwas
then patterned by selectively removing the photoresist (liftoff process). The Au and Cr
thicknesses were 250 nm and 10 nm, respectively. Figure 7(a) shows flexible film sensors
fabricatedona3inchwafer.Eachsensormeasures10mmx10mm.Aflexibleprintedcircuit
wasusedfortheelectricalconnection.Thesensorwasplacedontheprintedcircuit,anditwas
bondedtotheboardbyadhesive.Theelectricalconnectionwasmanuallymadewithsilver
paste.Theelectricalcontactareawascoveredwithadhesivetoincreaseitsmechanicalstrength.
Toformacavityforthermalisolation,thesensorwasplacedonasiliconerubbersheetwitha
5.0mmx5.0mmholeinit,andthesensorwasfixedtothesheetwithadhesivepolyimidefilm.
Therubbersheetwas0.5mmthick.Figure7(b)showsanassembledflexibleflowsensor.
Inanexperimentalevaluation,fourflexibleflowsensorswereattachedatanangleof90to
theinsidesurfaceofan8inchduct,andtheoutputsfromthemwereaveraged.Therelationship
betweentheoutputandflowrateobeyedKingsequationforflowsoffrom0to3000m3h1.
Theaveragedsensoroutputsdependedonthedistanceofthesensorfromthebendoftheduct,
andtheirvalueswereslightlyhigherthanthoseobtainedinastraightduct.Aconversionfactor,
343
344
Figure 7. Patch-type flexible flow sensor for large scale air-conditioning network systems.
whichenablesustocalculatetheflowratevaluesfromtheobtainedsensoroutputsinbent
ducts,wasderived[38].
3.3.Flowsensorsformedicalapplications
Thenumbersofcardiacdiseaseandcerebralstrokepatientsaregraduallydecreasing,thanks
to developments in medical devices and improved health guidance. However, chronic
obstructivepulmonarydisease(COPD)isstillontheincrease.Spirometryisnormallyusedto
evaluatetheprogressofCOPD.Itmeasurestheflowrateinthehumanmouth.Therespiratory
systemconsistsofnumerouslevelsofbronchijustlikeatree,andthelungalveoliarelocated
attheendofdivergingbronchi.InthecaseofCOPD,thealveolistructuresgraduallycollapse
asthepatientagesandbecauseofabsorbedcigarettesmokeorotherairpollutingsubstances.
Lesionsdevelopattheendsofthedivergingbronchi.Thecurrentmethodofmeasurement,
whichevaluateslesionsinthemouth,cannotbeusedtoevaluatesuchsmalllesionsinthe
bronchi.
Toovercometheaboveproblem,Shikidaetal.[34,39]devisedacathetertypeflowsensorthat
canmeasureaspiratedandinspiredairflowcharacteristicstransbronchially.Theflowsensor
(Figure 8) can be inserted into a small bronchus for measuring aspirated and inspired air
characteristics.Anonwallintubethermalflowsensorismountedontheinsideofthetube,
anditisusedinthebronchoscope.Theexternaldiameterofthetubeisonlyafewmm,and
therefore,itcanreachintothesmallbronchus.Twoheaterswereformedonthefilminorder
todetecttheflowdirection.
Figure 8. Concept of catheter flow sensor for trans-bronchial measurement. Republished with permission of IOP Pub
lished Ltd from Ref. [34].
PolymerMEMStechnologiesandheatshrinkabletubesareusedtoproducethesensor.The
fabricationprocessconsistsoftwosteps.
1.
Flexiblefilmsensorfabrication
Polyimidefilmisusedasasubstrate.Thethicknessofthefilmaffectsthethermalresponse
performance(ithastoberelativelythick;7.5m).Themetalfilmheaterstructuresareformed
onthefilmbyusingphotolithographyandsputtering,andtheyarepatternedwithaliftoff
process.Thetypicalsizeofthesensoris2.8mmx5.5mm.Twoheatersareformedonthefilm
inordertodetecttheflowdirection.Eachfilmsensorismechanicallycutbeforebeingmounted
insidethetube.
2.
Mountingfilmsensorinsideatube
Thesensorismountedontheinsidesurfaceofthetubeasfollows.
a.
ThesensorisinsertedintoaheatshrinkabletubemadeofTeflon.
b.
Theheatshrinkabletubeisbakedat110C.TheTeflontubeshrinkingtoalmosthalfits
originalsizeasaresultofheating,andthefilmsensorisautomaticallymountedonthe
innerwallsurfaceduringtheshrinkingprocessandbecomesfixedonthetubesurface.
c.
Acavitystructureisformedundertheheatingelementtoimprovethermalisolation.To
producetheactivestructure,aslitisformedonthetube,anditiscoveredwithaone
modeTeflonheatshrinkabletubetosealit.Theouterdiameteroftheinnertubeisonly
alittlebitlargerthantheinnerdiameteroftheouterone,aftertheheatshrinkingprocess.
Thus,thesetwoTeflontubesaretightlyfixedtooneother.
Figures9(a)and9(b)showaschematicdiagramandaphotographofthefabricatedcatheter
type flow sensor. The inner and outer diameters of the tube are 1.0mm and 1.8 mm. The
345
346
temperaturecoefficientofresistanceofthesensoris0.0025K1.Thepackagemethodoftheon
wallintubefilmmountingissuitableforminiaturizationbecausethesensorstructureitself
doesnotdisturbtheflowstream.
Figure 9. Fabricated catheter type flow sensor. Republished with permission of IOP Published Ltd from Ref. [39].
Experiments were done under the regulations set out in the Nagoya University Animal
Experiments Guidelines, and they were approved by the animal ethics committee. The
experimentsweredoneunderanesthesia.Thus,thefollowingdataareaspiratedandinspired
aircharacteristicsunderanesthesia.Infiltrationofbodilyfluidsintothetubewasprevented
byusingTeflonasthetubematerial.Thecatheterflowsensorisintendedtobeincorporated
intoabronchoscopeandtobeinsertedinthesmallairwaysfromthemouth.Thus,thesensor
wasfirsttestedonrats.Anopticalfiberscopewithanouterdiameterof0.8mm,insteadofa
bronchoscope,wasusedinthetests.Insertingtheflowsensorintotheairwayfromthemouth
oftheratwiththefiberscopeinvolvedthreesteps.
1.
Onlythefiberscopewasinserted,fromthemouthtothetargetedlocation,byobserving
theinsideoftheairway.Then,thesensorwasinsertedtothetargetedlocationwitha
Teflontubeguide.
2.
Thefiberscopewaswithdrawnwhentheflowsensorreachedthetargetlocation.
3.
TheTeflontubeguidewascarefullyextractedandonlytheflowsensorremainedatthe
location.
ThebreathingwaveformoftheratisshowninFigure10.Aperiodof820msforinspiration
and aspiration was obtained. This means that the respiration frequency was 1.1 Hz. The
ventilatedairvolumewascalculatedfromthisbreathingwaveform,andavaluerangingfrom
1.011.09ccwasobtained.Theknownrespirationfrequencyandventilatedairvaluesofrats
range from 1.11.9 for the former and from 0.601.25 for the latter. The measured values
coincidedwiththephysiologicalvaluesintheliterature.
Theairwasinspiredforashorttimeperiod,suddenlybecomingaspiratedforatimeperiod.
Inaspiratedmode,alargeamountofairwasaspiratedatthebeginning,andtheaspiratedair
gradually decreased afterwards. Inspiration and aspiration were done by moving the dia
phragm.Theairwasinspiredbyexpandingthethoracis.Thiswasdonebycontractingthe
diaphragm.Theairwassimplyaspiratedbytherestorativeforceofthethoracis.Thus,theair
wasinspiredinashorttime,andalargeamountofairwasaspiratedatthebeginningofthe
aspirationmode,andtheamountgraduallydecreasedafterthat.Thesensorsignalquantita
tivelycorrespondedtothisnaturalrespiratorymechanism.Fromtheseresults,thecatheter
flowsensorwillbeusefulforevaluatingtheflowcharacteristicsinthesmallbronchusregion
inthefuture.
Figure 10. Breathing waveform of rat measured with intubated catheter flow sensor. Republished with permission of
IOP Published Ltd from Ref. [39].
Figure 11. Stent flow sensor for evaluation of nasal respiration. Republished with permission of IOP Published Ltd
from Ref. [40].
347
348
Ahumanbeingbreathesthroughthemouthandthenose.Bacteriaandvirusesaretrappedin
thenosecavity,andonlycleaninspiredairissuppliedtothehung.Theinspiredairisalso
humidifiedinthecavityinorderforlungstoadsorboxygeneffectively.Theinspiredairis
warmedinsidethebodyisthenexpiredoutside;thisisamethodofheatexchange.Theflow
characteristicsofthenosearedeeplyrelatedtovarioushealthconcerns.Thus,astenttypeof
thermalflowsensorwasalsodevelopedformeasuringnasalrespiration.TheMEMSstentflow
sensor is shown in Figure 11. It is a thermal flow sensor fabricated on polymer film and
monolithicallyintegratedonthestentstructure[40].
4.Conclusion
ThischapterdescribedthefabricationandapplicationsofmicrosensorsproducedbyMEMS
technologies. The attractive features and problems of MEMS sensors were discussed. In
particular,MEMSsensorshavelowcosts,highspaceandtimeresolutions,andreduceddead
space.TheapplicationsofMEMSsensorsincludetactileSisolidandfabricsensorsforhuman
interfacesandflowsensorsforairconditioningsystemsandmedicine.
Acknowledgements
TheresearchwassupportedbytheCentreofExcellence(COE)forEducationandResearchon
MicroNanoMechatronicsandaGrantinAidforScientificResearch(B)No.23310091from
theMinistryofEducation,Culture,Sports,ScienceandTechnology(MEXT),Japan.
Authordetails
MitsuhiroShikida
References
[1] Petersen K. E. Silicon as a mechanical material. Proceedings of the IEEE 1982;70(5)
420457.
[2] Muller R. S, Howe R. T, Senturia S. D, Smith R. L, White R. M. Microsensor. USA:
IEEEPRESS;1990.
[3] YamasakiH.,editor.HandbookofsensorsandactuatorsIntelligentsensors(vol.3).
TheNetherlands:ElsevierScienceB.V.;1996.3.
[4] RisticL.Sensortechnologyanddevices.USA:ArtechHouse,Inc.;1994.
[5] MalufN.Anintroductiontomicroelectromechanicalsystemsengineering.USA:Ar
techHouse,Inc.;2000.
[6] ChuZ,SarroP.M,MiddlehoekS.Siliconthreeaxialtactilesensor,In:TechnicalDi
gestofThe8thInternationalConferenceonSolidStateSensorsandActuators,June
1995,Stockholm,Sweden;1995.
[7] Kane B. J, Cutkosky M. R, Kovacs G. T. A. A traction stress sensor array for use in
highresolution robotic tactile imaging. Journal of microelectromechanical systems
2000;9(4)425434.
[8] Kasten K, Amelung J, Mokwa W. CMOScompatible capacitive high temperature
pressuresensors.SensorsandActuatorsA:Physical2000;85(13)147152.
[9] KimK,LeeK.R,KimY.K,LeeD.S,ChoN.K,KimW.H,ParkK.B,ParkH.D,Park
Y.K,KimJ.H,PakJ.J.3axesflexibletactilesensorfabricatedbySimicromachining
andpackagingtechnology.In:ProceedingsofIEEEMEMS06.January2006,Istanbul,
Turkey;2006.
[10] TakaoH,YawataM,SawadaK,IshidaM.ArobustandsensitivesiliconMEMStac
tileimagerwithscratchresistantsurfaceandoverrangeprotection.TechnicalDigest
ofThe14thInternationalConferenceonSolidStateSensors,ActuatorsandMicrosys
tems.June2007,Lyon,France;2007.
[11] ShikidaM,ShimizuT,SatoK,ItoigawaK.Activetactilesensorfordetectingcontact
force and hardness of an object. Sensors and Actuators A: Physical 2003;103(12)
213218.
[12] HasegawaY,ShimizuT,MiyajiT,ShikidaM,SasakiH,SatoK,ItoigawaK.Amicro
machined active tactile sensor for hardness detection. Sensors and Actuators A :
Physical2004;114(23)141146.
[13] HasegawaY,ShikidaM,SasakiH,SatoK,ItoigawaK.Anactivetactilesensorforde
tecting mechanical characteristics of contacted objects. Journal of micromechanics
andmicroengineering2006;16(8)16251632.
[14] EngelJ,ChenJ,LiuC.Developmentofpolyimideflexibletactilesensorskin.Journal
ofmicromechanicsandmicroengineering2003;13(3)359366.
[15] Kim SH, Engel J, Liu C, Jones D. L. Texture classification using a polymerbased
MEMS tactile sensor. Journal of micromechanics and microengineering 2005;15(5)
912920.
349
350
[16] LeeHK,ChangSI,YoonE.Aflexiblepolymertactilesensor:Fabricationandmodu
larexpandabilityforlargeareadeployment.Journalofmicroelectromechanicalsys
tems2006;15(6)16811686.
[17] HwangES,SeoJH,KimYJ.Apolymerbasedflexibletactilesensorforbothnormal
andshearloaddetectionsanditsapplicationforrobotics.Journalofmicroelectrome
chanicalsystems2007;16(3)556563.
[18] KonishiS,MaedaH,EzakiT,KawatoM,AsajimaS,MakikawaM.Aselectivestimu
lation of nerve bundles using a flexible multielectrode. In: Technical Digest of The
10thInternationalConferenceonSolidStateSensorsandActuators.June1999,Sen
dai,Japan;1999.
[19] Chen N, Engel J, Pandya S, Liu C. Flexible skin with twoaxis bending capability
made using weaving bylithography fabrication method. In: Proceedings of IEEE
MEMS06,January,2006,Istanbul,Turkey;2006.
[20] HasegawaY,ShikidaM,OguraD,SatoK.Glovetypeofwearabletactilesensorpro
ducedbyartificialhollowfiber.In:TechnicalDigestofThe14thInternationalConfer
ence on SolidState Sensors, Actuators and Microsystems. June 2007, Lyon, France;
2007.
[21] HasegawaY,ShikidaM,OguraD,SuzukiY,SatoK.Fabricationofawearablefabric
tactilesensorproducedbyartificialhollowfiber.Journalofmicromechanicsandmi
croengineering2008;18(8)085014.
[22] Kita G, Suzuki Y, Shikida M, Sato K. Fabric tactile sensor composed of ballshaped
umbonalfiberfordetectingnormalandlateralforce.Micro&NanoLetters2010;5(4)
211214.
[23] KitaG,ShikidaM,SuzukiY,TsujiY,SatoK.Largesizedfabrictactilesensorsforde
tectingcontactedobjects.Micro&NanoLetters2010;5(6)389392.
[24] LiuC,HuangJB,ZhuZ,JiandF,TungS,TaiYC,HoCM.Amicromachinedflow
shearstress sensor based on thermal transfer principle. Journal of Microelectrome
chanicalsystems1999;8(1)pp.9099.
[25] ZheJ,MidiV,FarmerJrK.R.Amicrofabricatedwallshearstresssensorwithcapaci
tativesensing.JournalofMicroelectromechanicalsystems2005;14(1)167175.
[26] Unnikrishnan S, Jansen H. V, Berenschot J. W, Mogulkoc B, Elwenspoek M. C.
MEMSwithinaSwagelok:anewplatformformicrofluidicdevices.LabonaChip
2009;9(13)19661969.
[27] GianchandaniY,TabataO,ZappeH.ComprehensiveMEMS,ElsevierScienceB.V.
2008.
[28] ElwenspoekM,WiegerinkR.Mechanicalmicrosensors.Germany,Springer;2001.
[29] Zhu R, Liu P, Liu X. D, Zhang F. X, Zhou Z. Y. A lowcost flexible hotfilm sensor
systemforflowsensinganditsapplicationtoaircraft.TechnicalDigestIEEEMicro
ElectroMechanicalSystemsConference.January2009,Sorrento,Italy;2009.
[30] MaB,RenJ,DengJ,YuanW.FlexiblethermalsensorarrayonPIfilmsubstratefor
underwater applications. In: Technical Digest IEEE Micro Electro Mechanical Sys
temsConference.January2010,HongKong;2010.
[31] LiC,WuPM,JungW,AhnC.H,ShutterL.A,NarayanR.K.Anovellabonatube
formultimodalmonitoringofpatientswithtraumaticbraininjury.In:Technicaldi
gestofThe15thInternationalConferenceonSolidStateSensorsandActuators.June
2009,Denver,USA;2009.
[32] NaitoJ,ShikidaM,HirotaM,TanZ,SatoK.Miniaturizationofonwallintubeflexi
blethermalflowsensorusingheatshrinkabletube.In:TechnicalDigestIEEEMicro
ElectroMechanicalSystemsConference.January2008,Tucson,USA;2008.
[33] Yokota T, Naito J, Shikida M, Kawabe T, Hayashi Y, Sato K. Catheter type of flow
sensor for transbronchial measurement for lung functions. Technical digest of The
15thInternationalConferenceonSolidStateSensorsandActuators.June,2009,Den
ver,USA;2009.
[34] Shikida M, Naito J, Yokota T, Kawabe T, Hayashi Y, Sato K. A Cathetertype Flow
Sensor for Measurement of Aspirated and Inspiredair Characteristics in Bronchial
Region.JournalofMicromechnicsandMicroengineering2009;19(10)105027
[35] KingL.V.OntheConvectionofHeatfromSmallCylindersinaStreamofFluid:De
terminationoftheConvectionConstantsofSmallPlatinumWireswithApplications
to HotWire Anemometry. Philosophical Transactions of the Royal Society of Lon
don. Series A, Containing Papers of a Mathematical or Physical Character
1914;90(622)563570.
[36] TanZ,ShikidaM,HirotaM,SatoK,IwasakiT,IriyeY.Experimentalandtheoretical
study of an onwall intube flexible thermal sensor. Journal of Micromechnics and
Microengineering2007;17(4)679686.
[37] TanZ,ShikidaM,HirotaM,XingY,SatoK,IwasakiT,IriyeY.Characteristicsofon
wallintubethermalflowsensorunderradiallyasymmetricflowcondition.Sensors
andActuatorsA2007;138(1)8796.
[38] Yoshikawa K, Iwai S, Shikida M, Sato K. Attachedtype flexible flow sensor for air
conditioningnetwork.TechnicaldigestofThe16thInternationalConferenceonSol
idStateSensorsandActuators.June2011,Beijing,China;2011.
[39] ShikidaM,YokotaT,KawabeT,FunakiT,MatsushimaM,IwaiS,MatsunagaN,Sa
toK.Characteristicsofanoptomizedcathetertypethermalflowsensorformeasur
ing reciprocating airflows in bronchial pathways. Journal of Micromechnics and
Microengineering2010;20(12)125030.
351
352
[40] ShikidaM,YokotaT,NaitoJ,SatoK.Fabricationofastenttypethermalflowsensor
for measuring nasal respiration. Journal of Micromechnics and Microengineering
2010;20(5)055029.
Chapter 19
Single Cell Nanosurgery System
Toshio Fukuda, Masahiro Nakajima,

Yajing Shen and Masaru Kojima
1.Introduction
Amodelorganismisoneofthespeciesthatisextensivelystudiedtounderstandparticular
biologicalphenomena,withtheexpectationthatdiscoveriesmadeintheorganismmodelwill
provideinsightintotheworkingsofotherorganisms[1].Inparticular,modelorganismsare
widely used to explore potential causes and treatments for human disease when human
experimentationwouldbeunfeasibleorunethical[2].Itshowssomeofthemodelorganisms
thathavebeenusedinabiomedicalresearch.
Thefirstandforemostconsiderationintheselectionofanymodelorganismsbeforeconducting
anybiorelatedresearchishowrelevancetheselectedmodelorganismstohuman.Ifthefirst
considerationisjustified,thenthesecondconsiderationwhichneedstobeaddressedishow
practicalandeasytheselectedmodelorganismsforexperimentalendeavors.
Yeast is one of the simplest eukaryotic organisms (organism whose cells contain a clear
defining membranebound structure of nucleus) but many essential cellular processes are
conservedbetweenyeastandhumans.Therearegenesinyeastandmammalsthatencode
very similar proteins [3]. Comparison of the yeast and human genomes, reported in 1997,
revealed that 30% of known genes involved in human disease have yeast orthologs (i.e.
functionalhomologs)[4].Furthermore,hundredsofyeastgenesexhibitalinktohumandisease
genesasreportedby[5].
Yeastisagoodexperimentaltoolformolecularandcellularbiologystudies.Yeastgrowthand
divisioncanbecontrolledefficientlyandeffectivelybyadjustingenvironmentalconditions.
Furthermore,yeastcellsdivideinasimilarmannertohumancells.
Becauseoftheseadvantageousfeatures,yeasthasbecomethemodelorganismofchoicefor
medicinerelatedresearch.Forexample,studieswithyeasthavecontributedgreatlytoour
knowledgeoftheregulationofeukaryoticcelldivision,includingthecancerrelateddistur
2013 Fukuda et al.; licensee InTech. This is an open access article distributed under the terms of the
354
bances[6].Uptonow,yeasthasmaintaineditsroleasausefulmodelsysteminfundamental
studiesofdiseaseprocesses.
2.Singlecellnanosurgerysystembasedonnanoroboticmanipulation
system
2.1.Singlecellnanosurgerysystem
WehavebeenproposedaNanolaboratorybasedonnanoroboticmanipulationsystemfrom
around2000[7].Itisoneofthesystemstorealizevariousnanoscalefabricationandassembly
to develop novel nanodevices to integrate borderless technologies based on nanorobotic
manipulationsystem.Itisreadilyappliedtothescientificexplorationofmacroscopicphe
nomenaandtheconstructionofprototypenanodevices.Itwouldbeoneofthemostsignificant
enablingtechnologiestorealizethemanipulationandfabricationtechnologywithindividual
atomsandmoleculesfortheassemblyofdevices.Recently,theinvestigationofNanoelectro
mechanicalSystems(NEMS)hasattractedmuchattention[811].Itisexpectedtorealizehigh
integrated, miniaturized, and multifunctional devices for various applications. One of the
effectivewaysisthedirectusageofthebottomupfabricatednanostructures.
Nanolaboratory can be applied for the single cell analysis and manipulations. As show in
Figure1,theintegrationisimportantforthesinglecellnanosurgerysystembetweenmicro
andnanoroboticmanipulatorsundervariousmicroscopes.Theapplicationsunderdry,semi
wet,andwetconditionscanbedoneunderTEM/SEM,ESEMandopticalmicroscope(OM).
The nanomanipulation system inside TEM/SEM is a fundamental technology for property
characterizationofnanomaterials,structuresandmechanisms,withthefabricationofnano
buildingblocks,andfortheassemblyofnanodevices.Thenanomanipulationsysteminside
ESEM provides single cell manipulation and analysis under nanoscale high resolution
imagesfortheapplicationofnanodevicesornanotoolsassembledunderdrycondition.OM
micromanipulationsystemisusedunderwater,hencethebiologicalcellscanbeculturedwith
medium.
2.2.Nanoroboticmanipulations
Nanorobotic manipulation; nanomanipulation, has been received much more attention,
becauseitisaneffectivestrategyforthepropertycharacterizationsofindividualnanoscale
materialsandtheconstructionofnanoscaledevices[12].Theymightfinallybethecoremost
partofnanotechnology.Oneoftheattractivefutureapplicationsofnanomanipulationisto
realizetheultimategoalofnanotechnology,ornanomanipulation,isconsidered.
Tomanipulatenanoscaleobjects,itisneededtoobservethemwitharesolutionhigherthan
nanoscale. Hence, the manipulators and observation systems, microscopes in general, are
necessaryfornanomanipulations.Figure1showsthestrategiesofnanomanipulationswith
variouskindsofmicroscopes.Thenanomanipulationundervariousmicroscopesfor2D/3D
nanomanipulations.Opticalmicroscope(OM)isoneofthemosthistoricalandbasicmicro
Figure 1. Single cell nanosurgery system based on micro/nanomanipulators under various microscopes (wet/semiwet/dry conditions).
scope.However,itsresolutionislimitedto~100nmbecauseofthediffractionlimitofoptical
wavelength(~400~800nm)explainedbythewellknownAbbeslaw[13].
The scanning tunnelling microscopes (STMs) or atomic force microscopes (AFMs), have
functionsofbothobservationandmanipulationinnanoscale.Theirhighresolutionmakes
themcapableofatomicmanipulation.In1990,EiglerandSchweizedemonstratedthatthefirst
atompracticenanomanipulationwithscanningtunnellingmicroscope(STM)[14].Avouriset
al.appliedanAFMtobendandtranslatecarbonnanotubes(CNTs)onasubstrate[15].They
combinedthetechniqueswithaninverseprocess,namelystraightening,bypushingalonga
benttube,andrealizedthetranslationofatubetoanotherplace.NingXiet.alatMichigan
StateUniversity,developedAFMbasednanomanipulationsystemwithinteractiveoperation
system[16,17].ThesystemrealizedarealtimevisualfeedbackduringAFMbasednanoma
nipulation.
Normally,SPMsystemshavelimitfortheobservationin2Dplanewithquitesmoothsurface.
Moreover,theobservationareaislimitedandlongtimeisneededtogetoneimage(morethan
mins).Thislimitationrisesupas3Dnanomanipulationofnanostructures.Ontheotherhand,
theelectronmicroscopes(EMs)provideatomicscaleresolutionwiththeelectronbeamwhich
wave length is less than ~0.1 . EMs are divided mainly two types as scanning electron
355
356
microscopes(SEMs)andtransmissionelectronmicroscopes(TEMs).Forexample,M.F.Yuet.
alpresentedthetensilestrengthofindividualCNTsinsideaSEM[18].Howevertheresolution
ofSEM,generally~1nmresolution,isapproximatelyoneorderinmagnitudelowerthanthat
ofaTEM.HighresolutionandtransmissionimageofTEMsareusefulformeasurementand
evaluationofnanoscaleobjects.Kizukaet.alproposedthemanipulationholderinsidehigh
resolution transmission electron microscope (HRTEM). The manipulator was specially
designedwithatomiclevelpositioningresolution[19].
However, the specimen chamber and observation area of TEM are too narrow to contain
manipulatorswithcomplexfunctions.Hence,specialsamplepreparationtechniquesarealso
needed.WeproposedahybridnanoroboticmanipulationsystemwhichisintegratedTEMand
SEMnanoroboticmanipulatorsascoresystemfortheNanolaboratory[20,21].Thestrategyis
namedashybridnanomanipulationsoastodifferentiateitfromthosewithonlyanexchange
ablespecimenholder.Themostimportantfeatureofthemanipulatoristhatitcontainsseveral
passiveDOFs,whichmakesitpossibletoperformrelativelycomplexmanipulationswhereas
tokeepcompactvolumetobeinstalledinsidethenarrowvacuumchamberofaTEM[22].
Recentlysinglecellsanalysishasbeenmuchmoreattentionsbecauseoftheprogressofthe
micro/nano scale techniques on the local environmental measurements and controls [23].
UnderconventionalSEMsandTEMs,thesamplechambersoftheseelectronmicroscopesare
setunderthehighvacuum(HV)toreducethedisturbanceofelectronbeamforobservation.
Toobservewatercontainingsamples,forexamplebiocells,theappropriatedryinganddying
treatments are needed before observations. Hence, direct observations of watercontaining
samplesarenormallyquitedifficultthroughtheseelectronmicroscopes.
Ontheotherhand,theenvironmentalSEM(ESEM)canberealizedthedirectobservationof
watercontainingsampleswithnanometerhighresolutionbyspeciallybuiltsecondlyelectron
detector[24].Theevaporationofwateriscontrolledbythesampletemperature(~0~40C)
andsamplechamberpressure(102600Pa).TheuniquecharacteristicoftheESEMisthe
directobservationofthehydroscopicsampleswithnondryingtreatment.Hence,thenano
manipulationinsidetheESEMisconsideredtobeaneffectivetoolforawatercontaining
samplewithnanometerresolution[2528].
3.SinglecellanalysisbasedonanESEMnanoroboticmanipulation
system
Singlecellsanalysisneedstobeinvestigatedthroughthemicro/nanoscaletechniquesbased
onthelocalenvironmentalmeasurementsandcontrols.WedevelopedtheEnvironmental
SEM(ESEM)nanoroboticmanipulationsystemtomanipulateandcontrolthelocalenviron
mentsforbiologicalsamplesinnanoscale(Figure2).Itrealizedthatdirectobservationand
manipulation of watercontaining biological samples under nanometer high resolution
imaging.
Figure 2. Single Cell Nano-surgery System with various nano-tools
In this chapter, the novel local stiffness evaluation, local cutting, and local extraction of
biological organism are presented by micronanoprobes based on the ESEM nanorobotic
manipulationsystemforfuturecelldiagnosisandsurgerysystem.
3.1.Adhesionforcemeasurementofsinglecellusingnanoputter
Cellactivities,suchasembryogenesis,mitosis,morphogenesis,cellorientation,cellmotility,
andsurvivaldependonattachmenttoneighboringcellsandtheextracellularmatrix.Cellular
attachmenttoextracellularmatricesinfluencescellmorphology,cellfunction,andsignaling
mechanismsthatdirectcellularproliferationanddifferentiation[29].Cellsurfaceinteraction
isimportantinthedevelopmentofanymaterialordeviceforbiomedicalapplications,since
theperformanceofamedicaldeviceinthebodymustbecompatiblewiththesurrounding
tissue[30,31].Understandingofthecelladhesionprocesswouldbenefitthedevelopmentof
suitablebiomaterialsordeviceforbothtissueengineeringandmedicalfields.
Celladhesionprocessesareinfluencedbynumerousparameters,suchasthenatureofthe
biomaterial and its surface characteristics (roughness, topography, chemical composition,
surfacewettability,surfacechargeandsurfacetreatments)havebeeninvestigated.However,
theeffectofambienthumidityoncelladhesionhashadlessattention,especiallyatthesingle
celllevel.Understandingtheadhesionforceatvarioushumidityconditionscouldhelpusto
better understand the processes of celldirected integration during water evaporation. The
understandingisalsousefulincontrollingyeastinfectionsatwetenvironment.Moreover,cell
adhesionisinfluencedbythesurfaceenergyofsubstratestrongly,butthemechanismisstill
notclear[32].Thestudyofcelladhesiononsubstrateswithdifferentsurfaceenergycouldhelp
ustounderstandtheadhesionmechanismbetter.
357
358
We presented a yeast cell adhesion force measurement performed using the nanorobotic
manipulationsysteminsidetheESEM[33,34,35].Figure3(A)showsatypicalforcedisplace
mentcurveduringthesinglecelladhesionforcemeasurement.Figure3(B)showstheinitial
positionofthemicroputterandthesinglecell.Themicroputterwasdrivenbythenanorobotic
manipulationsystem.First,itwasmovedtowardsthecelluntilitcontactedthecell(Figure3
(C)).Then,acontinuousmovementwasappliedtothemicroputterbythenanomanipulator.
Themicroputterbeamdeflectedowingtotheincreasingpushingforce.Figure3(D)shows
thedeflectionofthemicroputterduringtheadhesionforcemeasurement.Finally,thecellwas
detached from its initial position under a certain force (Figure 3 (E)). The maximum force
duringthismanipulationprocedurewasdefinedastheadhesionforce.
Singleyeastcelladhesionforcemeasurementwasperformedatthreehumidityconditions,i.e.
100%,70%and40%.Themeanadhesionforceandthedeviationarewith95%confidenceat
eachhumidityconditions.Itdemonstratesthattheyeastcelladhesionforcesrangefrom10to
25Natvarioushumidityconditions.Theadhesionforceswere11.05.1N,17.44.7N
and23.56.1Nat100%,70%and40%relativehumidityconditionsrespectively.Itshowed
clearlythatthecelladhesionwasaffectedbytheambienthumidity.Thecelladhesionforceis
largeratlowhumiditythanathighhumidity.Forexample,thecelladhesionforcewas23.5
Natahumidityof40%,whichwas1.14timeslargerthantheforce11.0Nathumidity100%.
3.2.Singlecellcuttingusingnanoknife
Cellcuttingisanimportantstepincellanalysisprocesses.Forinstance,itwaswidelyusedto
preparecellspecimenslicesfortheobservationofaninnerstructure[36].Differenttogroup
cellsanalysis,researchonindividualcellscouldgiveaccuratedataratherthanaverageresults.
Singlecellanalysiscanhelpustounderstandthebiologicalprocessesmoreaccurately.Insitu
singlecellcuttingtechniquecouldpotentiallybenefitcellanalysis,suchassinglecelloperation
anddiseasetreatment.
Recently,ananoknifefabricatedfromacarbonnanotube(CNT)hasbeendevelopedforthe
purpose of cell cutting [37]. The nano knife was designed by welding a CNT across two
tungsten needles inside a scanning electron microscopy (SEM). This device can reduce the
anglebywhichthesampleisbentduringcutting,duetothesmalldiameteroftheCNT.Itcan
beseenclearlythatthenanoknifecanleaveamarkontheeponresinsurface,whichmeansit
cancutverythinslicesofcells.However,thebondingforcebetweentheCNTandtungsten
probesbyusingelectronbeaminduceddeposition(EBID)methodisquitesmall.Thereare
stillcertaintypesofhardspecimenssuchasbone,plants,andthickwalledspores.TheCNT
basednanoknifemaynotbeabletodealwithsuchsamples,sincealargercuttingforceis
required,especiallywhenthesamplesizeislarge.
Wepresentedananoknifewithabufferingbeamwasdesignedforaninsitusinglecellcutting
purpose[38].Aschematicdrawingofthesinglecellcuttingusingananoknifeisshownin
Figure 4. The nano knife was immobilized to the nanomanipulator by using the electrical
conductivetapeinsidetheESEMchamber.Underthedrivingofthenanoroboticmanipulation
system,thenanoknifecanmovetowardsandcutthesinglecellfinally.Thecuttingforcecan
becalculatedbasedonthedeformationofthenanokinfesbeam.
Figure 3. Single cell adhesion force measurement using micro putter inside ESEM.
359
360
Theinsitusinglecellcuttingexperimentwasperformedusingthesethreenanoknives.Figure
5(A)showstheinitialpositionofthenanoknifeandasinglecell.Figure5(B)showsthetouching
betweenthenanoknifetipandthesinglecell.Thedeformationofthenanoknifebeamand
thesinglecellduringthecuttingisshowninFigure5(C).Thedeformationofthebeamcanbe
measuredfromtheESEMimagedirectlyusingimageanalysissoftware.Therefore,thecutting
forcecanbecalculatedbasedonHookeslaw.Theseparatedsinglecellaftercuttingisshown
inFigure5(D).ThesamplesliceanglecanbemeasuredfromtheESEMimagedirectlyaswell.
Figure5(E)showsthesinglecellcuttingimageusingthe25oknife.Figure5(F)showsthesample
sliceangleaftercutting.Theimagesofsinglecellcuttingandsamplesliceangleaftercutting
using45oknifeareshowninFigure5(G)andFigure5(H).
Figure 4. Schematic drawing of the single cell cutting experiment using nano knife inside ESEM.
Figure 5. Single cell cutting using nano knife inside ESEM.
4.Conclusion
Thischapterpresentsthesinglecellnanosurgerysystembasedonnanomanipulationtechni
ques.Themicronanotoolshavebeenproposedtoinvestigatesinglecellanalysistomanipulate
andcontrolthelocalenvironmentinmicronanoscale.TheESEMnanomanipulationsystem
wasconstructedtorealizethelocalstiffnessevaluation,localcutting,andlocalextractionof
biologicalorganisminnanometerscale.Theadhesionforcemeasurementwaspresentedby
microputterforsinglecells.Thesinglecellcuttingwasalsodescribedusingnanoknife.As
future direction, the multiple micronanotools are used continuously depending on the
purposesbyexchangingmachinerysystem(NTExS:NanotoolExchangerSystem)[39].Weare
investigatingonthenanoinjectionapplicationsfortheCaenorhabditiselegans(C.elegans)as
oneofthemodelorganisms[40,41]
Acknowledgements
TheauthorsaregratefultoProf.T.InadaN.UozumiforprovidingwithW303yeastcells.This
workwaspartiallysupportedbyaGrantinAidforScientificResearchfromtheMinistryof
Education,Culture,Sports,ScienceandTechnologyofJapan.
Authordetails
ToshioFukuda,MasahiroNakajima,YajingShenandMasaruKojima
References
[1] Fields S, Johnston M. Cell biology: Whither model organism research? Science
2005;30718851886.
[2] FoxM.A.Thecaseforanimalexperimentation:Anevolutionaryandethicalperspec
tive:UniversityofCaliforniaPress;1986.
[3] BotsteinD,ChervitzS.A,CherryJ.M.Yeastasamodelorganism.Science1997;277
12591260.
[4] Foury F. Human genetic diseases ? a crosstalk between man and yeast. Gene
1997;195110.
[5] Mager W. H, Winderickx J. Yeast as a model for medical and medicinal research.
TRENDSinPharmacologicalSciences2005;26265273.
361
362
[6] HartwellL.H.Yeastandcancer,BioscienceReports2002;22373394.
[7] DongL.X,AraiF,FukudaT.DestructiveConstructionsofNanostructureswithCar
bon Nanotubes Through Nanorobotic Manipulation. IEEE Transaction on Mecha
tronics2004;9350357.
[8] SiegelR.W,HuE,RocoM.C.NanostructureScienceandTechnology.KluwerAca
demicPublishers;1999.
[9] CraigheadH.G.NanoelectromechanicalSystems.Science2000;29015321535.
[10] StaplesM,DanielK,SimaM.J,LangerR.ApplicationsofMicroandNanoElectro
mechanicalDevicestoDrugDelivery.PharnaceuticalResearch2006;23847863.
[11] LearyS.P,LiuC.Y,ApuzzoM.L.J.TowardtheEmergenceofNanoneurosurgery:
PartIIINanomedicine:TargetedNanotherapy,Nanosurgery,andProgressToward
theRealizationofNanoneurosurgery.Neurosurgery2006;5810091026.
[12] DuE,CuiH,ZhuZ.ReviewofNanomanipulatorsforNanomanufacturing.Interna
tionalJournalofNanomanufacturing2006;183104.
[13] HellS.W.FarFieldOpticalNanoscopy.Science2007;31611531158.
[14] EiglerD.M,SchweizerE.K.PositioningSingleAtomswithaScanningElectronMi
croscope.Nature1990;344524526.
[15] Hertel T, Martel R, Avouris P. Manipulation of Individual Carbon Nanotubes and
theirInteractionwithSurfaces.TheJournalofPhysicalChemistryB1998;102910915.
[16] LiG,XiN,ChenH,PoneroyC,ProkosM.VideolizedAtomicForceMicroscopyfor
Interactive Nanomanipulation and Nanoassembly. IEEE Transaction on Nanotech
2005;4605615.
[17] LiG,XiN,YuM,FungWK.DevelopmentofAugmentedRealitySystemforAFM
basedNanomanipualiton.IEEETransactiononMechatronics2004;9358365.
[18] YuM.F,LourieO,DyerM.J,MoloniK,KelleyT.F,RuoffR.S.StrengthandBreak
ing Mechanism of Multiwalled Carbon Nanotubes under Tensile Load. Science
2000;287637640.
[19] Kizuka T, Yamada K, Deguchi S, Naruse M, Tanaka N. Crosssectional Time Re
solved HighResolution Transmission Electron Microscopy of AtomicScale Contact
and NoncontactType Scannings on Gold Surfaces. Physical Review B 1997;55
73987401.
[20] Nakajima M, Arai F, Fukuda T. In situ Fabrication and Electric Actuation of Tele
scoping Nanotube inside TEM through Hybrid Nanorobotic Manipulation System.
In:Proceedingsofthe2006IEEE/RSJInternationalConferenceonIntelligentRobotics
andSystems(IROS2006);2006.,19151920.
[21] NakajimaM,AraiF,FukudaT.InsituMeasurementofYoungsModulusofCarbon
Nanotube inside TEM through Hybrid Nanorobotic Manipulation System. IEEE
TranssactiononNanotechnology2006;5243248.
[22] LearyS.P,LiuC.Y,ApuzzoM.L.J.TowardtheEmergenceofNanoneurosurgery.
Neurosurgery2006;5810091026.
[23] Donald A. M. The use of environmental scanning electron microscopy for imaging
wetandinsulatingmaterials.NatureMaterials2003;2511516.
[24] Nakajima M, Arai F, Fukuda T. Nanofixation with Low Melting Metal Based on
NanoroboticManipulation.Proceedingsofthe6thIEEEInternationalConferenceon
Nanotechnology(IEEENano2006).,925928.
[25] AhmadM.R,NakajimaM,KojimaS,HommaM,FukudaT.Theeffectsofcellsizes,
environmental conditions and growth phases on the strength of individual w303
yeastcellsinsideesem.IEEETransactionsonNanobioscience2008;7185193.
[26] AhmadM.R,NakajimaM,KojimaS,HommaM,FukudaT.InsituSingleCellMe
chanics Characterization of Yeast Cells using Nanoneedles inside Environmental
SEM.IEEETransactionsonNanotechnology2008;7(5)607616.
[27] AhmadM.R,NakajimaM,KojimaS,HommaM,FukudaT.NanoindentationMeth
odstoMeasureViscoelasticPropertiesofSingleCellsusingSharp,FlatandBuckling
TipsinsideESEM.IEEETransactionsonNanobioscience2010;9(1)1223.
[28] Ahmad M. R, Nakajima M, Kojima S, Homma M, Fukuda T. Buckling Nanoneedle
forCharacterizingSingleCellsMechanicsinsideEnvironmentalSEM.IEEETransac
tionsonNanotechnology2011;10(2)226236.
[29] Gumbiner B. M. Cell adhesion: the molecular basis of tissue architecture and mor
phogenesis.Cell1996.,84345.
[30] LelievreS,WeaverV.M,BissellM.J.Extracellularmatrixsignalingfromthecellular
membraneskeletontothenuclearskeleton:amodelofgeneregulation.RecentProg
ressinHormoneResearch(1996).,51417.
[31] HongGellerE.Aroleforcelladhesioninberylliummediatedlungdisease.Journal
ofOccupationalandEnvironmentalHygiene2009;6727731.
[32] MarcotteL,TabrizianM.Sensingsurfaces:Challengesinstudyingthecelladhesion
processandthecelladhesionforcesonbiomaterials.Irbm2008;297788.
[33] Shen, Y, Nakajima, M, Kojima, S, Homma, M, Kojima, M, & Fukuda, T., Single cell
adhesionforcemeasurementforcellviabilityidentificationbyusingAFMcantilever
basedmicroputter.,Measurementscienceandtechnology,(2011),22(11)115802
[34] Shen,Y,Ridzuan,M,Nakajima,M,Kojima,S,Homma,M,&Fukuda,T.,Evaluation
ofthesingleyeastcellsadhesiontoITOsubstrateswithvarioussurfaceenergiesvia
363
364
ESEM nanorobotic manipulation system. IEEE Transactions on Nanobioscience,

(2011).,10(4),217224.
[35] Shen, Y, Nakajima, M, Ridzuan, M, Kojima, S, Homma, M, & Fukuda, T. Effect of
ambienthumidityonthestrengthoftheadhesionforceofsingleyeastcellinsideen
vironmentalSEM.Ultramicroscopy(2011).,111(8),11761183.
[36] Mobius,W.Cryopreparationofbiologicalspecimensforimmunoelectronmicrosco
py.AnnalsofAnatomyAnatomischerAnzeiger(2009).,191231.
[37] Singh, G, Rice, P, Mahajan, R, & Mcintosh, J. Fabrication and characterization of a
carbonnanotubebasednanoknife.Nanotechnology(2009).20(9),095701.
[38] Shen, Y, Nakajima, M, Yang, Z, Kojima, S, Homma, M, & Fukuda, T. Design and
characterization of nanoknife with buffering beam for in situ single cell cutting.
Nanotechnology(2011),22(30),305701.
[39] Nakajima,M,Kawamoto,T,HiranoT,Kojima,M.&Fukuda,T.NanotoolExchanger
SystembasedonESEMNanoroboticManipulationSystem.Proceedingsofthe2012
IEEE International Conference on Robotics and Automation (ICRA 2012), (2012),
27732778.
[40] Nakajima,M,Hirano,T,Kojima,M,Hisamoto,N,Homma,M,&Fukuda,T.Direct
Nanoinjection Method by Nanoprobe Insertion based on ESEM Nanorobotic Ma
nipulation under Hybrid Microscope. Proceedings of the 2011 IEEE International
ConferenceonRoboticsandAutomation(ICRA2011),(2011),41394144.
[41] Nakajima,M,Hirano,T,Kojima,M,Hisamoto,N,Nakanishi,N,Tajima,H,Homma,
M, & Fukuda T. Local Nanoinjection of Fluorescent Nanobeads inside C. elegans
basedonNanomanipulation.Proceedingsofthe2012IEEE/RSJInternationalConfer
enceonIntelligentRoboticsandSystems(IROS2012)(2012),32413246.

Paper

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Paper

Uploaded by

Copyright:

Available Formats

MICRO-NANO

Micro-Nano Mechatronics New Trends in Material, Measurement, Control,

Research and Technology on Micro-Nano Mechatronics 1

Neural Interfaces: Bilateral Communication

High Knudsen Number Flow

Precision Micro Machining

Micro-Nano Robotics and

Tissue Engineering and Regenerative Medicine 123

Electronic Structure Calculations for Nano Materials 167

Tissue Damage and Repair Caused by Immune System

Tribology for Biological and Medical Applications 221

Micro-Nano Materials Characterization and Inspection 241

Aerospace Application 271

Transferrin-Toxin Conjugates for Cancer 315

Tissue Engineering and

MEMS Sensors and Their Applications 331

Single Cell Nanosurgery System 353

Research fields and two-dimensional matrix structure

Mechatronic devices for regeneration &

Mechatronics for cell

Nagoya, December 2012

Research and Technology on Micro-Nano Mechatronics

Toshio Fukuda, Masahiro Nakajima and

Figure 1. Micro-Nano mechatronics

Figure 2. Micro-Nano mechatronics for social and industrial demands

Research and Technology on Micro-Nano Mechatronics

High Integration Level

Figure 4. Green and life innovations based on micro-nano mechatronics

Micro/nano devices for Discovery of oil resource, Management of water

Monitor, control and management of Environment, Distributed sensing &

Energy saving, harvesting and alternatives , Energy grid and

Food and Agriculture

Safety, testing and tracing, Efficient harvesting, nutritious products,

Table 1. Challenges for green innovation by micro-nano mechatronics

Medicine for life

Inspection and diagnosis , Regenerative medicine, Gene therapy and life

BiologyAnalysis and Synthesis

Sensing , manipulation and automation, New species, DNA diagnosis &

Table 2. Challenges for life innovation by micro-nano mechatronics

Research and Technology on Micro-Nano Mechatronics

Figure 5. Blood vessel simulator and surgical operation system

In medical applications, surgical supporting or simulating technologies are important We

Research and Technology on Micro-Nano Mechatronics

Figure 6. Milestone of Micro-Nano mechatronics research fields

Research and Technology on Micro-Nano Mechatronics

[7] Yoshida K, Hagihara Y, Tanaka S, Esashi M. Normallyclosed Electrostatic Micro

Research and Technology on Micro-Nano Mechatronics

Neural Interfaces: Bilateral Communication Between

Goro Obinata, Hitoshi Hirata, Chikara Nagai,

Figure 4. The electrodes placement

: Marker for measuring positions of body

Figure 5. Proposed experimental setup for reconstructing walking motion.

passivei i , i kiJ1 exp kiJ2

Figure 7. Structure of CPGs.

Figure 8. Walking simulation results.

Figure 9. CPGs outputs as commands for joint torques in steady state.

Joint Torque [Nm]

Figure 10. Generated joint torques.

Joint Angle [degree]

Figure 11. Joint angles at shoulders and hips.

High Knudsen Number Flow

High Knudsen Number Flow Optical Diagnostic Techniques

Figure 1. Principle of LIF

Figure 2. Supersonic free jet visualized by I2-LIF [1]