Clean Chemistry

Techniques for the Modern Laboratory
by Dr. Robert Richter
A practical guide to . . .
Clean Sample Preparation

for Trace Metals Analysis
LEAN
HEMISTRY
Techniques for the Modern Laboratory
by Dr. Robert Richter
A practical guide to . . .
Clean Sample Preparation

for Trace Metal Analysis
2003 Milestone Inc.
All rights reserved. No part of this publication may be reproduced,

stored in a retrieval system, or transmitted, in any form or by any
means, electronic, mechanical, photocopying, recording, or otherwise,
without the prior written consent of the publisher.
Printed in the U.S.A.

Edition 1
Teflon, Dacron, and Tyvek are registered trademarks

of E.I. DuPont Corporation.
Milestone Press
25 Controls Dr..
Shelton, CT 06484
CONTENTS
Section 1:
Controlling Contamination in Your Laboratory
Chapter 1:
The Analytical Blank
13
Chapter 2:
Laboratory Environment
15
Chapter 3:
Materials for Trace Analysis
21
Chapter 4:
Trace Analysis Reagents
33
Chapter 5:
The Analyst: A Source of Contamination
37
Section 2:
Advanced Sample Preparation Techniques
Chapter 6:
Closed-Vessel Microwave Digestion
43
Chapter 7:
Microwave Evaporation
57
Chapter 8:
Improving Method Detection Limits
65
Section 3:
Preparing Your Laboratory for Trace Analysis
Chapter 9:
Laboratory Housekeeping Techniques
77
Chapter 10:
Equipment & Supplies for Trace Metal Analysis
81
Appendix A:
References
91
5
Clean Chemistry: Techniques for the Modern Laboratory
Introduction
ver the past several decades, the need for trace and ultratrace elemental analysis has increased. Current drinking
water regulations require determinations in the low parts
per billion (ppb) range, while semiconductor applications require

parts per trillion (ppt) determinations. There is a growing awareness that improved sample handling and preparation techniques
are required to meet these new standards, and the instruments
and experimental techniques used to achieve them.
This book focuses on helping the analytical chemist understand and adapt to the latest
trends in trace metal analysis.
The book is divided into three
parts. The first part of the book
focuses on the importance of
the analytical blank, along with
techniques for its reduction and
Introduction
control. The second part deals

with the preparation of the homogeneous sample solutions for
trace metal analysis. Part three
talks about preparing your lab for trace metal analysiswhat
you need and where to get it.
10
SECTION ONE
Controlling
Contamination in
Your Laboratory
12
race elemental methods require the analyst to carry a special sample, known as an analytical blank, through all the
steps of the analysis. This analytical blank is a measure of
all external sources of elemental contamination, and is used to

make a correction to the measured sample concentration. The
variability of the analytical blank, rather than its absolute value,
is what determines the accuracy of trace metals analysis.
The overall uncertainty for an analytical measurement
is calculated using the following formula1:
The variability of the analytical blank has little or no effect on

the accuracy of the result when
the analyte concentration of the
sample is several orders of magni-
Chapter 1
The Analytical
Blank
tude higher than the blank result.

For example, given a sample with
a measured lead concentration
of 500 25 ng and a blank concentration of 10 5 ng, the total
uncertainty would be:

The reported analytical result would be 490 25 ng of lead.
In this example, the blank result has no effect on the accuracy of
the result.
13
On the other hand, as the analyte concentration approaches the blank level, the accuracy of the result can
be compromised by the uncertainty of the analytical
blank measurement. For example, given a sample with
a measured lead concentration of 50 2 ng and a blank
concentration of 10 5 ng, the total uncertainty would be:

The reported analytical result would be 50 5 ng of
lead. In this example, all of the uncertainty in the analysis
is due to the analytical blank.
In order to improve the precision and accuracy of
trace metal determinations, steps must be taken to control
and reduce the analytical blank. The contributions to the
analytical blank come from four major sources:
Laboratory environment
Apparatus and containers used for analysis
Reagents
Analyst
The rest of Section One discusses each of these
sources in detail, and offers some strategies for controlling
these contamination sources in your lab.
14
irborne contamination is a major contributor to

the analytical blank. Dust particles are created in
the laboratory from the slow weathering of solid
objects, or from chemical processes upon materials. The

main components of these dust particles are Ca, Si, Al,
Fe, Na, Mg, K, Tl, Cu, and Mn.2-4 If left isolated and undisturbed, dust particles will have little effect on trace metal
analysis. Unfortunately, this does not usually happen, and
the dust particles become airborne. Airborne particulate
concentrations can go from 0.2
x 106 in the morning to 1.5 x 106
by noon.5 Once airborne, the
heavier particles quickly settle
out, while the lighter ones are
suspended and carried by the
air currents throughout the
laboratory. Airborne contami-
Chapter 2
Laboratory
Environment
nates can also be introduced

to the laboratory from external
sources, through the ventilation system. In Dortmund, Germany, analysts discovered
that their high iron blanks were caused because nearby
industrial plants dispersed 20 tons of ferric oxide, which
contaminated their ventilation system.6
Particulate contamination is minimized by filtering
the laboratory air with a high efficiency particulate air
filter (HEPA). This filter was developed for the Manhattan Project to remove fissionable particles from the air.7
HEPA filters have an efficiency of 99.97% for particles 0.3
15
m and larger.
They effec-
tively remove bacteria, pollen, fly ash, and dust. Some

common contaminates, such
as tobacco smoke, are not
removed because their particles are smaller than 0.3 m.2
Particulate contamination in
a work area is also minimized
by using a laminar air flow
HEPA air filter
system to provide directional

air flow. Directional air flow
from one end of the work area to the other essentially

creates an air curtain, preventing airborne contamination
from entering, or settling out.
The best approach to controlling airborne contamination for trace metal analysis is to construct a clean laboratory, or clean room. This approach is expensive, and is
used in the production of computer, pharmaceutical, and
aerospace products, and in the production of standard
reference materials.
Clean rooms minimize metal-bear-
ing dust by filtering the air that enters the room through
HEPA filters, using specialized construction materials,
and operating under positive pressure so that the flow
of air is always out of the room. Clean rooms also utilize
antechambers to isolate them from the general laboratory
and provide an area for analysts to change into specialized
clean room attire. In a properly maintained and operated clean room, sub-parts-per-trillion measurements can be
16
Ch. 2: Laboratory Equipment
made.
More information
on clean rooms and clean

room design can be found
in references 811.
A more practical approach for trace metal
laboratories is the use
of laminar flow hoods.
These units provide areas
of HEPA filtered air for
sample handling, analysis,
and reagent storage. The
Laminar flow hood
standard units are good
for drying containers and apparatus after cleaning, protecting samples in autosampler trays during analysis, and
preparation of calibration and quality control standards. A
standard unit should not be used to handle or store toxic
or hazardous materials, because all the fumes that are
generated are blown out at the analyst. Exhausted laminar
hoods are available for the handling of hazardous and
corrosive substances. These units are constructed from
nonmetallic components and the HEPA filtered air flows
vertically and is drawn though a perforated work area into
the laboratory ventilation system.
Clean conditions can also be obtained with a
filtered air enclosure. These enclosures consist of a plexiglass box that is purged with filtered nitrogen or air to
remove air particles. The exhausted air is vented to a con17
ventional hood. These units are an inexpensive alternative

to laminar flow hoods, and are good for drying and storage
of containers and apparatus after cleaning, and storage of
samples and reagents.
Airborne contamination can also be controlled by
minimizing the generation, transportation, and deposition of atmospheric particles. The major source for trace
metal particulates in the laboratory is the degradation of
metals, paints, cements, plastics, and other construction
materials. Unnecessary shelving, partitions, and furniture
should be removed from the laboratory, because dust and
debris can accumulate on them. Metal furniture should be
replaced with wood. Stainless steel door handles, hinges,
and plumbing should be replaced with plastic equivalents or coated with pigment-free epoxy paint. Bench
tops should be coated with epoxy paint, and for added
protection, covered with Teflon, or polyethylene sheeting
(contact paper). Ceiling panels should be replaced with
ones that have a plastic laminate on each side to prevent
particle formation. Low fiber emitting and dissolvable content tissues should be used for wiping operations. Floors,
benches, and apparatus should be wiped down with D.I.
water regularly.
Bottles, containers, samples, reagents,
and equipment should be kept isolated from laboratory air

using laminar flow hoods, plastic snap-top Tupperware
boxes, or polyethylene bags.3,4,8-13
When an electrostatic charge is present, charged
atmospheric particles are attracted to oppositely charged
18
Sili
c
Tefl on R
u
Sili on bber
con
Vin
y
Po l
lyu
Po ret
h
ly
My styre ane
l
ne
Ra ar
yon
Go
l
Bra d, Pla
t
s
Ha s, Silvinum
rd
Wo Ruber
o
Ste d ber
el
Pap
Alu er
m
Silk inu
m
Wo
ol
Ny
lo
Hu n
m
Mi an H
ca
air
G la
s
Hu s
m
Air an S
k in
Ch. 2: Laboratory Equipment
Most Negative (-)
Most Positive (+)
Figure 1. Materials tendency to generate a static charge.
surfaces. Electrostatic charges are usually generated by

friction between, and/or separation of, two dissimilar
materials, at least one of which is a nonconductor or a
poor conductor of electricity. The accumulated charge
(static) resides on the surface of, rather than within, the
charged nonconductive object. The highest static charges
accumulate under low humidity, on insulating surfaces
that have a low moisture content.4 Teflon has a tendency
to generate a negative charge, while the components of
airborne contamination become positively charged, especially at low humidity (Figure 1). The opening of a Teflon
bottle produces a negative charge on the neck of the
bottle, so any positively charged airborne contamination
can immediately become attached to the cap or neck of
the bottle.12 Electrostatic charges on airborne contamination can be minimized by keeping humidity levels above
50%. The surface charges can be removed by wiping with
a lint-free cloth, lightly wetted with high-purity ethanol or
water. Commercially manufactured static eliminators are
also available.
19
20
efore a sample is analyzed for trace metals, it has been

collected, stored, processed, and prepared. During this
sequence of events, the sample comes in contact with
many different laboratory tools, containers, and apparatus,

which can deposit trace metal contamination into the sample.
Standard laboratory mortars are made of alumina or glass, and
will therefore contaminate the sample with Al, Si, and Fe. Mills
and blenders are made of stainless steel, and have tungsten
carbide blades, which will contaminate the sample with W, Ni,
Cr, Co, and Fe. Sieves are usually
made of stainless steel with copper wire mesh, which will contribute Cu, Fe, Co, Cr, and Ni. The
ashless filter paper often used to
filter samples contains trace metal contamination on the order of
1ppm, and the filtration assem-
Chapter 3
Materials for
Trace Analysis
blies can also impart contamination to the sample.
For trace
metal analysis it is best to avoid

these practices when possible. Solid samples should be homogenized by digesting several samples and combing the digestates,
or by digesting a large sample and aliquoting. Aqueous samples
should be centrifuged in plastic tubes (see below) to remove
particulate material.
Borosilicate Glass
The most common material used for laboratory containers and apparatus is borosilicate glass. It is resistant to most
21
acids, but should not be used with HF or boiling H3PO4.

Alkaline solutions should not be heated or stored in borosilicate glass, because they will gradually solubilize the
glass according to the following equation4:
2X NaOH + (SiO2)x
]XNa SiO
2
+ XH20
Borosilicate glass is not a good material for trace

analysis, because it contains high levels of trace metals
(Table 1), and has the potential to absorb analytes from
the sample according to the following equation4,14:
GlassSiOH + M+
] GlassSiOM + H
Quartz
An alternative to borosilicate glass is quartz.
Like
borosilicate glass, it is resistant to most acids, but should

not be used with HF, boiling H3PO4, or alkaline solutions.
Quartz is composed almost entirely of SiO2, and its trace
metal concentration depends on the type of quartz and the
method of production. Naturally occurring quartz laboratory components are made by electric (Type I) or flame
melting (Type II). Type II quartz has a lower trace metal
concentration because some of the metals are volatilized
in the flame. Synthetic quartz laboratory components are
made by the vapor phase hydrolysis (Type III) or oxidation and electrical fusion (Type IV) of SiCl4. Both of these
methods produce quartz with low trace metal contamination.3,4,15 The typical trace metal impurity levels for the
various types of quartz are shown in Table 1.
22
23
b
600
1,000
Major
2.9
3,000
3,000
Borosilicate Glass
Major
Major
1,000
Adapted from references 3 and 15.

ND = Not Detected.
Element
Al
B
Ca
Cr
Cu
Fe
K
Li
Mg
Mn
Na
Sb
Quartz (Type I)
74
4
16
0.1
1
7
6
7
4
1
9
0.3
Quartz (Type II)

68
0.3
0.4
NDb
1
1.5
<1
1
ND
0.2
5
0.1
Quartz (Type III)

< 0.25
0.1
< 0.1
0.03
<1
< 0.2
0.1
ND
ND
< 0.02
< 0.1
0.1
Table 1: Trace element concentrations (g/g) in borosilicate glass and various types of quartz.a
Ch. 3: Materials for Trace Analysis
Synthetic Polymers
The low levels of trace metal contamination make
quartz an ideal material for trace metal analysis, but the
cost and availability of common laboratory containers and
apparatus in quartz limit its use in trace metal analysis.
Synthetic polymeric materials are now being employed
as materials for containers and apparatus for trace metal
analysis. The trace metal impurities of these materials
are comparable to those of quartz, but will vary based on
the manufacturing environment, type of molding, molding
components, and polymerization process. A dirty manufacturing environment will lead to the incorporation of
airborne particles. Molded components can contain high
levels of the metals (Ni, Al, Mn, Cu, Fe) used to make the
mold.12,16 The most common materials used for trace metal
analysis are polyethylene, polypropylene, and fluorinated
polymers.
Polyethylene
There are two types of polyethylene used in trace
metal analysis, conventional (low density) and linear (high
density). Low density polyethylene (LDPE) is produced
by high pressure polymerization of ethylene. High density
polyethylene is produced at low pressures, catalyzed by
transition metal oxides ([Al]R3, TiCl4, ZrCl3, VCl3, CrCl3).
Polyethylene is resistant toward concentrated HCl and HF,
but is oxidized by dilute HNO3 and aqua regia. Prolonged
storage of dilute solutions of HNO3 causes the material to
turn brown or yellow. The maximum temperature for LDPE
24
Table 2:
Trace element concentrations (g/g) in some polymersa
Element
Al
Ca
K
Na
Sb
Ti
Mn
Zn
a
LDPE
0.5
> 5,000
1.3
0.005
HDPE
30
800
> 600
1.5
0.2
5
0.01
520
PP
55
4.8
0.6
60
0.2
PFA
FEP
PTFE
0.1
0.2
0.4
0.23
0.16
0.6
Adapted from reference 17.
is 80C and 110C for HDPE. The use of LDPE is preferable

to HDPE because it has less trace metal contamination.
(Table 2).
Polypropylene
Polypropylene is produced catalytically (Al, Ti) from
propylene, and, like HDPE, has elevated levels of some
trace metal contaminates (Table 2). Polypropylene is less
resistant to concentrated HCL and becomes yellow or
brown with prolonged exposure. Its resistance to dilute
HNO3 and aqua regia is similar to that of polypropylene.
Polypropylene is harder and more rigid than polyethylene
and is stable up to 135C. Polypropylene is well suited for
open vessel digestion containers, and applications that
require sterilization.
25
Fluorinated Polymers (Teflon)

Fluorinated polymers (Teflon) are more expensive
than polyethylene or polypropylene, but have lower
trace metal impurities (Table 2), and greater chemical
resistance. Fluorinated polymers are attacked only by
molten alkali metals and fluorinated organic compounds
at elevated temperatures. The greater resistance is due
to the high-energy C-F bonds and the protection of the
carbon backbone by the fluorine atoms. The three most
common fluorinated polymers used in trace metal analysis
are poly(tetrafluroethylene)(PTFE), perfluoroalkoxy-flurocarbon (PFA), fluorinated ethylene propylene (FEP).
Cleaning Methods
Picking a material with low trace metal impurities
doesnt guarantee low blank levels. Trace metal impurities can be leached from the container and apparatus, by
the sample and reagents used for trace metal analysis.
Comprehensive cleaning procedures must be adopted to
ensure the lowest blank levels. There have been a variety
of cleaning methods reported in the analytical literature,
and the method chosen will vary with the chemical behavior of the element of interest. A good general procedure for
cleaning all types of containers is sequential leaching with
hydrochloric and nitric acids (Tables 3 and 4). For polyethylene, a 48-hour soaking with 10% nitric acid is effective
for initial and routine cleaning.17, 18
26
Table 3:
Impurities (ng/cm2 of surface) leached from plastic containers in
one week with 1:1 HNO3/Water. HDPE and LDPE were leached at
room temperature while FEP was heated to 80C.a
Element
Pb
Tl
Ba
Te
Sn
Cd
Ag
Sr
Se
Zn
Cu
Ni
Fe
Cr
Ca
K
Mg
Al
Na
Total
a
b
LDPE
0.7
1
2
< 0.5
< 0.8
0.2
NDb
0.2
3
2
2
0.5
3
0.8
10
2
0.7
1
8
38
HDPE
2
<1
< 0.2
0.2
1
0.2
0.2
1
0.4
8
0.4
1.6
3
0.2
0.6
2
0.6
1
10
50

ND = Not Detected.
27
FEP
2
<1
4
0.6
1
0.4
<8
0.2
0.2
4
2
2
20
0.8
80
2
8
6
6
148
Table 4:
Impurities (ng/cm2 of surface) leached from plastic containers in
one week with 1:1 HCl/Water. HDPE and LDPE were leached at
room temperature while FEP was heated to 80C.a
Element
Pb
Tl
Ba
Te
Sn
Cd
Ag
Sr
Se
Zn
Cu
Ni
Fe
Cr
Ca
K
Mg
Al
Na
Total
a
b
LDPE
18
3
0.3
0.7
< 0.8
0.2
ND
0.2
< 0.3
1.0
0.7
0.3
1.0
0.3
0.8
0.7
0.7
10
42
38
HDPE
0.6
< 0.6
1
NDb
<1
0.2
ND
0.2
0.4
9
1
0.8
1
0.8
60
1
0.4
4
6
50

ND = Not Detected.
28
FEP
2
<1
2
2
1
0.6
<6
<1
0.8
4
6
0.8
16
4
2
1.6
1.0
4
2
148
Steam cleaning with nitric or hydrochloric acid is also

a very effective cleaning method for containers and apparatus (Table 5).19,20 In this method the container is placed
over a PTFE-coated glass rod. Acid in a lower reservoir
is heated, and purified acid vapor travels up through the
glass rod and condenses on the container, removing surface contamination (Figure 2). This method of cleaning is a
popular alternative to the traditional soaking methods for
the following reasons:
1. The trace metal contamination found in the reagent grade acid remains in the lower reservoir and
does not come in contact with the component to be
cleaned.
2. The clean component does not remain in contact
with the cleaning acid after the surface contamination is removed.

29
Figure 2. Acid
steam cleaning
system.
3. The critical surfaces of the clean component are

dry when the cleaning process is complete.
This
eliminates the need for rinsing and air drying.

4. The cleaning process takes place in a sealed container, which minimizes airborne contamination and
provides a clean environment for the components to
be stored until they are needed.
A Closer Look
A machine that performs fully automated acid
steam cleaning, as described in the text above.
Purified acid vapor condenses on the containers
placed in the machine, and the sealed environment
protects them until they are needed. (See page 82
for more information.)
30
85
55
56
144 39
85
72
261
57
995 80
Ni
Co
Cu
Cr
Cd
Tl
Pb
Zn
117 12
474
Fe
876
57
261
72
56
55
474
121
121
Element
Al
Mg
Na
TFM Teflon Vessel

Acid Leachingb
Steam Cleaningb
287 46
258 24
289 22
232 15
1,005 124
57
261
72
85
109 9
56
55
474
Error expressed as one standard deviation (n = 3).
1,640 1,000
57
261
72
170 15
176 57
56
55
474
Quartz Vessel
Acid Leaching
Steam Cleaningb
398 28
327 18
441 56
347 26
1190 350
608 67
b
Table 5: Comparison of high-temperature acid leaching cleaning vs. acid steam cleaning. Trace metal
contamination (pg/g) in 5% HNO3 blanks prepared after cleaning are listed below. The acid leaching was
performed at 180C with mixture of HCL and HNO3. The steam cleaning performed with HNO3 only.a
32
he instrumentation used for trace analysis (ICP/OES or

ICP/MS) requires homogeneous solutions for calibration
and analysis. Calibration solutions are prepared by dilu-
tion with water, and samples are treated with mineral acids. The
purity of the reagents is important because the amount of reagent used is usually several orders of magnitude larger than the
original sample size. Trace metal contamination in the reagents
must be low enough to measure accurately the analyte concentration in the sample. For example, to measure 10 ng of lead
in a sample that was prepared
with 25 mL of reagents, the lead
contamination in the reagents
must be less than 0.2 ng.
No single purification method
is capable of removing all impurities from reagents.
Sub-boiling
distillation has been shown to
Chapter 4
Trace Analysis
Reagents
be the method of choice for acid

purification.
This method uses
infrared heaters to vaporize the

surface liquid. The vaporized liquid is collected on an inclined
water-cooled condenser and drips into the collection container
(Figure 3). Vaporization without boiling is the key element of this
purification process because it prevents aerosolized particles
from depositing on the surface of the condenser and being carried over to the purified acid. High-purity acids produced in this
way have essentially the same concentration as the acids used to
produce them.12,21-23 Table 6 shows the quality of acids that can
be obtained from sub-boiling distillation.
33
Figure 3. Sub-boiling acid distillation system.
A Closer Look
A fully automated sub-boiling distillation
system allows chemists to produce their
own high-purity acids, and even re-purify
contaminated acids. (See page 81 for more
information.)
A good water supply is also essential for trace metal

analysis. The most common methods for achieving this
are sub-boiling distillation (previously discussed), reverse
osmosis, and ion-exchange. Reverse osmosis separates
dissolved material from the water by forcing contami34
Ch. 4: Trace Analysis Reagents
Table 6:
Trace metal contamination (pg/g) in nitric acid produced by
sub-boiling distillation in a quartz still.
Element
Mg
Al
Ca
Ti
V
Cr
Mn
Fe
Co
Ni
Cu
Zn
Se
Sr
Ag
Cd
Sn
Ba
Tl
Pb
B
Source 1a
42
147
157
8
11
5
2
210
1
23
21
49
1
1
2
2
9
4
<1
3
ND
Source 2b
90
700
110
ND
ND
60
ND
350
ND
80
50
60
60
20
6
20
20
20
60
30
ND
Source 3c
400
900
400
800
<3
100
7
800
< 10
30
200
80
ND
ND
ND
< 30
ND
20
ND
40
200
Reference 24 Microwave evaporation with ICP-MS analysis.

Reference 22 Hot plate evaporation in Class 100 hood with
ID-SSMS analysis.
c
Reference 23 Inverted Pyrex with filtered air evaporation with
SSMS analysis.
a
35
nated water through a membrane against osmotic pressure. The membrane preferentially allows water to pass,
rejecting 90-99% of dissolved ions and particulates. The
reverse osmosis system is generally used as a pre-treatment technique for water before it is further purified by
either sub-boiled distillation (previously described) or
ion-exchange. In the ion-exchange method the contaminated water passes through a column of resin. The resin is
composed styrene-divinylbenzene copolymers engineered
to have an affinity for either cations or anions. The resins
exchange hydrogen and hydroxyl ions for the charged
metal contaminates. This results in an exchange of the
trace metal contamination for clean water.
Sub-boiling
distillation and ion exchange both produce water with low

levels of trace metal contamination (Table 7).
Table 7:
Trace metal contamination (pg/g) in high-purity water.
Element
Mg
Al
Ca
Ti
V
Cr
Mn
a
Source 1a
42
147
157
8
11
5
2
Source 2b
90
700
110
ND
ND
60
ND

36
early all trace metals analysis procedures require some

intervention by the analyst. Serious sample contamination can occur as a result of careless manipulation of the
sample. Touching or handling of equipment with bare hands

can cause serious contamination. Dried skin contains 6 g/g of
Zn and 0.7 g/g of copper, and sweat contains Na, K, Pb, Ca, and
Mg. Hand lotions and creams contain Al, Zn, Ti, and Mg oxides,
and other trace metal contaminates.
Traces of iron, copper,
gold, silver, platinum, and chromium can be deposited from

watches, rings, and bracelets.
Human hair contains on average
100 g/g of zinc, 20 g/g of copper
and iron and 10 g/g of lead. The
use of cosmetics by the analyst
can be a significant source of contamination, because of the metal
oxides and other materials which
are added to such products for
Chapter 5
The Analyst:
A Source of
Contamination
color and texture (Table 8). Some

hair dyes and shampoos contain selenium and lead. Analysts should avoid the use of these
products when carrying out analyses.
In order to control analyst contamination the analyst must
first be isolated from the samples.
The most fundamental
precaution is to wear gloves when performing procedures that

require manipulation of the sample. The gloves must be impervious to skin oils and perspiration and must be powder-free. Clear
polyvinyl chloride or polyethylene gloves are the best for routine
handling of the samples. When working with concentrated acids,
37
Table 8:
Trace metal contamination in cosmetics.
Cosmetic
Lipstick
Eye Shadow
Blush
Mascara
Foundation
Face Powder
Trace Metals Present

Bi, Zn, Fe, Mg, Ti, Mn
Bi, Si, Fe, Mn, Ti, Al, Cr, Mg
Si, Fe, Mg, Ti, Ca
Na, Fe, Mg, Ti, Cr, Al
Ti, Al, Zn, Fe
Ti, Si, Bi, Fe, Zn, Mg, Ca
nitrile gloves are a good compromise between chemical

resistance and cleanliness, and are often worn in conjunction with long-cuff vinyl gloves. One should remember
that gloves are only as clean as the last thing touched, and
must be changed on a routine basis to avoid the gloves
becoming a contamination source. The best results are
achieved when gloves are worn in conjunction with a head
cover, shoe covers, and a laboratory coat.
Protective garments (lab coats, head and shoe covers)
worn during trace metal analysis should not be cotton or
linen. These materials should be avoided because they
produce a considerable amount of lint, which can contaminate the sample. The best materials for laboratory
garments are nylon, Dacron polyester, and Tyvek. Garments made from these materials do not shed fibers, are
lightweight, and are resistant to acids and other reagents.
Garments used in trace metal analysis should also have
all cut edges enclosed, and should be made without com38
Ch. 5: The AnalystA Source of Contamination
ponents that are susceptible to

corrosion, such as metal buttons
and zippers.
The analyst can also be responsible for cross-contamination from other samples or work
being performed in the laboratory. If the analyst is not aware
of the history of the laboratory
containers and apparatus used
for the analysis, contamination
can occur as a result of using a
vessel or container that was exposed to high concentrations of
the analytes of interest. The use
of paper towels in another part
of the laboratory can introduce
large quantities of airborne contamination into the atmosphere.
Protective garbincluding
gloves, head cover, shoe covers,
and lab coatis essential for
best results.
Careless handling of the reagents

can lead to contamination of the reagents, the sample,
and the calibration standards. Contamination from these
sources is very unpredictable. As an analyst, one must
be constantly aware of ones actions, and think about how
those actions will affect the blank. Most importantly, the
analyst must avoid those actions that tend to increase the
blank, or whose effects on the blank are unknown.
39
40
SECTION TWO
Advanced
Sample Preparation
Techniques
42
he reliability of most methods of analysis depends on

quantitative conversion of solids to homogeneous solutions. Conventional wet-sample preparation methods for
the decomposition of solid samples are usually carried out in

vessels containing the sample and a large volume of decomposition reagent(s), typically 15 to 100 mL. This mixture is heated
for several hours using hot plates, heating mantles, or ovens.
Heating is terminated when the analyst decides that the decomposition of the sample is sufficiently complete.
This type of
open-vessel digestion has many

drawbacks, which include the use
of large volumes (and multiple
additions) of reagents, potential
for contamination of the sample
by materials and laboratory environment, and the exposure of
the analyst and the laboratory to
corrosive fumes.
The
high-pressure
Chapter 6
Closed-Vessel
Microwave
Digestion
closed-
vessel wet-ashing technique, originally described by Carius, is

a more efficient way to decompose samples for analysis. The
increased pressures allow temperatures beyond the atmospheric
boiling point of the reagent to be reached. While Cariuss method
improves the efficiency of decomposition, it also has several
drawbacks. Analytes can be lost during the opening of the tube
when the contents, under elevated pressure, are released suddenly. The analyst is frequently exposed to corrosive reagents, as
well as flying pieces of glass, during the opening of the tube.
43
A Closer Look
A closed-vessel microwave sample
digestion machine is the most
efficient means of converting solids
to homogeneous solutions. (See pp.
81-82 for more information.)
Steel-jacketed Teflon lined bombs are now available

to perform similar high-pressure and temperature reactions in thermal ovens. While the higher pressures and
temperatures inside the closed vessels increase reaction
rates, digestions may still require several hours due to the
inefficient, outside-in heating mechanism. In addition,
the high pressures involved with both of these conventional closed vessel methods tend to increase the safety
risk of applying these techniques. Closed-vessel microwave decomposition, on the other hand, uses significantly
different technology and fundamentally unique principles
to accomplish sample decomposition.
Heating by microwave energy is a cold in situ
process, producing heat only when there is absorption or
coupling of the microwave energy to the solution or microwave-absorbing objects. The two primary mechanisms
for the absorption of microwave energy by a solution are
dipole rotation and ionic conductance. In the dipole rotation mechanism, molecular dipoles align with the applied
44
Ch. 6: Closed-Vessel Microwave Digestion
electric field. Oscillation of the electric field results in

forced molecular movement of the dipole molecules with
the resulting friction heating the solution. At 2.45 GHz, the
frequency of most laboratory microwave ovens, the dipoles align, then randomize 5 billion times a second. In the
ionic conduction mechanism, the ionic species present in
solution migrate in one direction or the other according to
the polarity of the electromagnetic field. The accelerated
ions meet resistance to their flow, and heating is a natural
consequence.1
These two unique heating mechanisms
result in rapid heating of solutions, in comparison with

conduction and convection. The heating is so fast that, in
open vessels, vaporization alone can dissipate the excess
energy. This results in solutions being able to sustain superheating above their normal boiling points by as much
as 50C for water to 260C for acetonitrile.2,3
To use closed-vessel microwave decomposition
effectively, one must understand the unique temperature
and pressure relationships involved.
Gas pressures
inside microwave-closed vessels are not what would be

predicted from the temperature of the liquid phase. The
pressure inside a microwave vessel is dependent upon the
volume of the vessel, and the temperature and composition of the gas phase. For example, when water is placed
in a high-pressure steel-jacketed Teflon bomb and heated
in a convection oven, an equilibrium vapor pressure is
established.
This vapor pressure is dependent upon
the rate of evaporation and condensation of the water

vapor.
When the temperature is increased, there is a

45
corresponding increase in the evaporation rate and a decrease in the condensation rate, because the vessel walls
heat both the solution and gas phase. The decrease in
condensation rate leaves more water in the vapor phase,
increasing the internal pressure. In contrast, when water
is heated to the same temperature in a microwave closedvessel, the internal pressure is significantly less than its
steel-jacketed counterpart. This phenomenon is a direct
result of the microwave heating mechanism used, and the
materials of which the microwave vessel is composed.
The microwave-closed vessels liner and outer casing are
microwave transparent and have no insulating capacity. Thus, they remain relatively cool during the heating
process. The cooler the vessel walls, the more efficient
they will be at removing
water molecules from
the vapor phase.
The
increased condensation
rate results in lower
internal
pressures
at
higher
temperatures.
This microwave reflux

action is illustrated in
Figure 1.
A more complex
example of this pheFigure 1. Reflux conditions inside a microwave closed vessel.
46
nomenon is the closed-vessel microwave heating of nitric

acid. Nitric acid is a polar and partially ionized mixture
which heats rapidly in a microwave field. When heated,
nitric acid partially decomposes into NOx gas. The gas
phase inside a closed vessel becomes a mixture of NOx
gas, nitric acid and water vapor. The pressure of nitric
acid during closed-vessel microwave heating is still lower
than predicted, even after taking into account the partial
pressure of the NOx gas (Figure 2). The decrease in pressure results from the previously described reflux action
and from the loss of the ionic conductance mechanism
of microwave heating in the gas phase. The loss of the
70
60
Thermal Equilibrium
Microwave Vessel
Pressure (atm)
50
40
30
20
10
0
121
130
133
150
165
170
190
192
Temperature (Celcius)
Figure 2. Internal pressure of nitric acid at different temperatures.

47
210
219
ionic conductance mechanism in the gas phase means

the NOx gas does not convert microwave energy into heat
efficiently, keeping the pressure increase associated with
the heating of the gas to a minimum.
This unique temperature and pressure relationship,
found only in closed-vessel microwave heating, becomes
more complex and unpredictable as additional reagents
and samples are added to the solution. Also, the condensation rate varies with the microwave vessel materials and
geometry, the liquid volume, the duration of heating, and
the systems ability to dissipate the excess energy. At the
present time there is no conventional method of predicting
the decrease in internal pressure associated with closedvessel microwave heating. This phenomenon is one of
the reasons that pressure control is not applicable for
standardizing microwave sample preparation methods.
The use of closed vessels in microwave decomposition allows the reagents to be heated above their atmospheric boiling points. The higher temperatures achieved
in the closed system give microwave decomposition a
kinetic advantage over hot plate digestion, as described by
the Arrhenius Equation:
dlnk
dT
Ea
RT 2
Integration of this equation gives:
ln
k2
Ea
1
1
=
k1
2.303R T1 T2
48
In this expression k1 and k2 are rate constants for the

reaction of interest at T1 and T2 respectively, Ea is the
activation energy, and R is the ideal gas constant. These
equations show that the reaction rate increases exponentially with increasing temperature. This translates into
approximately a 100-fold decrease in the time required to
carry out a digestion at 175 C when compared to 95 C digestion.4,6,7 In addition, because the mineral acid converts
the microwave energy into heat almost instantaneously,
rapid heating of the sample is achieved, further decreasing
the reaction times.
The effectiveness and safety of the microwave
digestion procedure depends on the choice of digestion
reagents. The most common reagents used for microwave
digestion are nitric, hydrochloric, and hydrofluoric acids.
Nitric acid is a strong oxidizing agent and liberates trace
elements as highly soluble nitrate salts. The oxidizing
properties of nitric acid are retained only in concentrated
form and are lost when the acid is diluted below 2M. Nitric
acid is used for the dissolution of metals and organic/biological materials. Some metals form insoluble oxide films
(Al, Cr, Ti, Nb, Ta) and require the addition of a complex
forming species in order to obtain complete dissolution.
The nitrate ion forms weak complexes with tin, tungsten,
and antimony, and these elements can become hydrolyzed
and precipitate from solution.
Hydrochloric and hydrofluoric acids are complexing acids. Many metal carbonates, peroxides and alkai hy49
droxides are dissolved by hydrochloric acid. Hydrofluoric

acid is the only acid which will readily dissolve silica-based
materials, and creates strong complexes with elements
that form stable refractory oxides. Alkaline earths, lanthanide and actinide elements form insoluble or sparingly
soluble complexes with hydrofluoric acid. These acids are
often used in conjunction with nitric acid to dissolve metal
alloys and noble metals.
Microwave Decomposition and

Clean Chemistry
Closed-vessel microwave decomposition has several
unique characteristics that have led analysts to reduce
the sources of error and contributions to the analytical
blank that were previously obscured by lengthy sample
preparation procedures.
As described in Chapter 2, the laboratory air that
comes in contact with the sample can deposit airborne
contaminants into the sample. Open-vessel methods cannot prevent this type of contamination because the samples are continuously exposed to laboratory air currents.
The use of closed vessels for microwave decomposition
isolates the sample from laboratory air currents during the
decomposition process, reducing airborne contamination.
By preparing the samples for closed-vessel microwave
dissolution under clean air conditions, we can isolate the
sample, so that it will not come in contact with laboratory
air. Thus, the possibility for contamination by airborne
50
particles in the laboratory air is eliminated. Similarly, by

performing post sample processing, such as rinsing and
dilution in clean environments, airborne contamination
can be further reduced.
Reagent contamination contributes to the concentration of the analytes present in the sample. The amount
contributed is a function of the total quantity of reagent
used. For example, using 50 mL of a reagent that contains
a contaminant at the 100 ppb level will contribute 5 g of
that species as contamination. Unlike open-vessel methods that require large quantities of reagent due to evaporation losses, closed-vessel microwave methods continually
reflux the reagents and do not allow the reagent vapors to
escape. The reflux conditions inside a microwave-transparent vessel allow much smaller volumes of the reagent
to be used, thus reducing the contamination from the
decomposition reagents.
The materials used for microwave vessels (fluorinated polymers and quartz) are chemically inert to most
dissolution reagents, and provide a non-contaminating environment for sample preparation. TFM is preferred over
PFA for the construction of microwave vessels because of
its superior chemical and thermal resistance and lower
blank levels (Tables 1 and 2).
The skill of the analyst is perhaps the most difficult
factor to evaluate. For open-vessel methods, the analyst
must monitor the progress of the dissolution in several
51
Table 1:
Trace element concentrations (g/L) of acid blanks prepared
from identical reagents in PFA and TFM vesselsa
Element
Al
B
Ba
Bi
Cd
Co
Cr
Mo
Pb
Sb
W
Zn
Zr
PFA
2.7
7.5
0.9
0.5
4.7
0.8
0.8
0.6
6.7
0.6
5.4
0.6
1.0
TFM
ND
1.8
DL
DL
0.1
DL
0.1
DL
0.03
DL
DL
0.2
DL

ND = Not determined.
DL = Concentration below instrument detection limit.
a
vessels simultaneously. Compounding the problem, the

temperature of each vessel varies with its position on
the hot plate. The analyst must then subjectively decide
whether dissolution is sufficiently complete, and remove
the sample from the heat source. In microwave methods,
the efficiency of the heating, coupled with the direct monitoring of the conditions inside the vessel, removes the
subjective judgment of the analyst from the digestion pro52
Table 2:
Analytical blank values for TFM microwave vessels after fifty
digestions of environmental samplessa
Element
As
B
Ba
Be
Cd
Ce
Co
Cu
Ga
a
g/L
0.24
0.40
0.04
0.03
0.01
0.19
0.02
0.03
0.02
Element
Li
Mo
Pb
Sb
Sc
Se
Sr
Ta
Zn
g/L
0.09
0.26
0.15
0.01
0.04
0.01
0.05
0.01
0.31
cess. The automation and standardization of the sample

preparation method through the use of closed-vessel microwave decomposition serves to reduce the impact of the
analysts judgment upon much of the sample preparation
process.
The combined effect of closed-vessel microwave
decomposition and clean chemistry techniques is shown
in Tables 3 and 4, and Figure 3. The blank results show
significant reduction in the contributions of the outside
environment, reagents, and materials to the analytical
blank level and overall measurement uncertainty. From
Table 4, we can see that this combination of techniques
allows trace metal analysis to be completed with increased
53
accuracy and precision. Figure 3 (on page 56) shows

that closed-vessel microwave decomposition, coupled
with clean chemistry techniques, is accurate and precise
enough to prepare instrument calibration standards.
Table 3:
Comparison of analytical blank results obtained from hot
plate and microwave digestion of a certified reference soil.a
Concentration expressed as g/g.
Analyte
As
Cd
Cr
Cu
Pb
Hg
Ni
Se
Tl
V
Zn
a
b
Hot Plateb
0.204 0.106
0.318 0.122
3.35 2.85
0.060 0.020
0.171 0.076
0.037 0.004
0.375 0.069
0.548 0.264
0.028 0.020
2.35 0.45
2.92 1.43
Microwaveb
0.074 0.013
0.029 0.019
0.104 0.059
0.030 0.017
0.040 0.019
0.017 0.007
0.060 0.063
0.172 0.022
0.028 0.020
1.20 0.51
1.66 0.93
From reference 10.

Error expressed as 95% confidence interval (n=4).
54
Table 4:
Comparison of analytical blank results obtained from hot
plate and microwave digestion of a certified reference soil.a
Concentration expressed as g/g.
Analyte
As
Cd
Cr
Cu
Pb
Hg
Ni
Se
Tl
V
Zn
Hot Plateb
12.3 2.27
0.31 0.09
68.8 7.5
23.8 2.7
10.8 1.3
0.97 0.14
63.4 3.9
1.61 0.34
0.29 0.05
65.6 8.8
113 13.5
Microwaveb
17.6 0.9
0.41 0.06
123 3
33.5 1.2
17.5 1.1
1.42 0.10
83.2 3.0
1.54 0.33
0.63 0.02
119 6
102 6.1
Certified Valuec
17.7 0.8
0.38 0.01
130 4
34.6 0.7
18.9 0.5
1.40 0.08
88 5
1.57 0.08
0.74 0.05
112 5
106 3
From reference 10.

Error expressed as 95% confidence interval (n=4).
c
95% confidence interval as reported on certificate.
a
55
Figure 3. ICP-MS calibration using microwave digested certified reference

materials. The three calibration points represent separate digestions. Adapted
from reference 10.
56
olutions submitted for trace elemental analysis often need

to be evaporated prior to analysis. The most common reason for evaporation is to concentrate the sample, because
initial analyte levels are below the instrument detection limits.

The other reason for evaporation is that the sample solution
contains matrix elements that will present problems for analysis.
Traces of hydrofluoric acid will etch the glass components of
ICP and ICP-MS systems, releasing trace element contamination. Chlorine and fluorine form polyatomic ions that interfere
with the ICP-MS analysis of many
common elements. For example,
40
Ar35Cl+ interferes with 75As, and
35
Cl16O+ interferes with 51V.

Obtaining
solutions
for
trace elemental analysis is often

complicated due to the formation of volatile species during
Chapter 7
Microwave
Evaporation
the evaporation process. As the

number of solvent molecules decreases, the ions begin to recombine, and at dryness the residue
will consist of a mixture of recombined salts. These salts will
have an associated vapor pressure and boiling point that will
vary with the oxidization state and counter anion (Table 5). The
use of traditional heating methods to perform evaporations can
lead to the loss of volatile analytes, because as the evaporation
proceeds, the temperature of the evaporation vessel approaches
the temperature of the heat source. At dryness these temperatures can be in excess of 150C.
57
Table 5:
Potentially volatile salts from solution.
Element
Arsenic
Antimony
Selenium
Tin
Vanadium
Chromium
Volatile Salts
AsCl3
AsF3
SbF5
SbCl5
SeCl4
SeF4
SnCl4
VCl4
VF5
CrF5
Boiling Point (C)

130.2
-63
150
79
170196a
107.8
115
152
111.2
117
SeCl4 sublimes.
Data taken from the following references:
a
Dean, J.A., ed., Langes Handbook of Chemistry, 12th ed., New York:
McGraw-Hill, 1979.
David, R. Linde, ed., CRC Handbook of Chemistry and Physics, 71st
ed., Cleveland: CRC Press, 1990.
In contrast to traditional evaporation methods, the

unique heating mechanism exclusive to microwave-assisted heating allows for the retention of volatile analytes
during evaporation. As the evaporation proceeds and the
solvent is removed from the system, the matrix volume will
be reduced. As the mass of the sample solution decreases,
the amount of microwave energy absorbed decreases
according to the following equation:
58
Ch. 7: Microwave Evaporation
A Closer Look
Microwave-assisted heating methods allow for the
retention of volatile analytes
during evaporation. (See pp.
81-82 for more information.)
KCpTm
t
In this equation P is the apparent absorbed power in
watts, K is the conversion factor for calorie/s to watts,
Cp is the specific heat, T is change in temperature, m
is the total mass of sample in the microwave, and t is irradiation time. This unique relationship between sample
mass and energy absorption, along with the microwave
vessels being microwave transparent, leads to a decrease
in temperature as the sample approaches dryness. Lower
temperatures at dryness decrease the potential for loss of
volatile species, resulting in more complete recoveries for
volatile analytes11,12 (Figures 4 and 5).
Pabs=
Another factor that plays a significant role in

evaporation losses is the oxidization state of the analyte
of interest. It has been reported that even mild heating
of an HF solution results in a 20% loss of Se(IV) and a
45% loss of As(III), and a losses of 65%-100% upon dryness. Losses of antimony during evaporation are due to
59
Figure 4. Evaporation recoveries of select elements from a 9:3 solution of

HNO3/HCl. Initial solution concentration was 500 ppb.
the formation of poorly soluble compounds, and mercury

is due to its being present in its elemental form. These
losses can be prevented by converting these elements to
higher oxidation states i.e. Se(VI), As(V), Sb(V), Hg(II). The
use of traditional decomposition methods can not ensure
a uniform oxidation state, due to matrix interferences
and reaction rate limitations.14 This can be overcome by
coupling closed-vessel microwave decomposition with
microwave evaporation. The elevated temperatures and
pressures decompose the matrix interferences and form
stable complex ions.
60
Figure 5. Percent recovery of 2.5 ng spikes from 10ml of HCl. Uncertainties

are expressed as 95% confidence intervals with n4. Adapted from references
4 and 13.
For example:
Sb2O2(s) + HCL + HNO3
AsF3(aq) + HF + HNO3
SbCl6-(aq)
AsF6-(aq)
The formation of these stable complex anions leads to

complete retention of these traditionally volatile elements.
(Figures 6 and 7).
61
Concentration (g/g)
Element
Figure 6. Concentration of analytes in SRM 1566A (Oyster Tissue) following

Microwave-Assisted Evaporation of the digestate compared with the certified
total concentrations. Uncertainties are expressed as 95% confidence intervals
with n3. Adapted from references 4 and 11.
62
Concentration (g/g)
Element
Figure 7. Concentration of analytes in SRM 2710 (Montana Soil) following

Microwave-Assisted Evaporation of the digestate compared with the certified
total concentrations (* = noncertified concentration). Uncertainties are
expressed as 95% confidence intervals with n3. Adapted from references 4
and 11.
63
64
nalytical chemists are being required to measure lower

and lower levels of trace metals in samples. This frequently requires the analyst to make measurements near
the method detection limit, which usually results in decreased

accuracy and precision. The advantages of using closed-vessel
microwave decomposition in combination with clean chemistry
techniques for lowering method detection limits have been discussed in Chapter 6 of this book, but sometimes these methods
are not enough, and another approach is required to achieve the
desired levels.
Closed-vessel microwave
digestion techniques require a
minimum volume of 10 mL to
achieve accurate temperature
monitoring of the reaction conditions. Modern spectroscopic
techniques
require
samples
submitted for analysis to have
Chapter 8
Lowering Your
Method Detection
Limits
acid concentration of 10% (v/v)

or less. This requires samples
that were digested with 10 mL of acid to be diluted by a factor
of 100 or more. This problem can be overcome by increasing
the sample size. This works well for samples that do not contain a large amount of organic material. For samples with high
organic content this is usually not an option, because the secondary gases (CO2 and NOx) produced during the digestion can
cause the vessel to vent when larger sample sizes (greater than
0.5 grams) are used. Recent advances in microwave chemistry
have led to the development of vessel-inside-vessel technol65
ogy as a means to lower method detection limits for highly

organic samples.
Vessel-inside-vessel technology was developed
through by the collective efforts of Milestone Inc, srl, and
GmbH in the late 1990s. Vessel-inside-vessel technology
uses a smaller secondary vessel inside the primary microwave vessel. The secondary vessel contains the sample
and digestion reagents, and the primary vessel contains
the 10 mL of solution required to achieve accurate temperature monitoring (Figure 8). This configuration reduces the
amount of acid required for digestion to near stoimetric
quantities, which reduces the dilution factor and increases
the detection limit.
The use of vessel-inside-vessel technology is also
used for the processing of larger organic sample sizes.
This is accomplished by controlling the reaction kinetics
and lowering the pressure inside the microwave vessel.
Controlling reaction kinetics is especially important when
trying to digest large quantities (0.5 to 1.0 g) of organic material, because the potential for auto-catalytic decomposition increases. When the sample size is small (0.25 g), the
heat released by the oxidization of the organic material
does not cause a significant change in the temperature of
the reaction mixture. As the sample size is increased, the
heat released from the oxidation can cause the reaction
mixture to heat faster that the programmed rate. The rise
in temperature promotes further decomposition, which
results in the microwave vessels venting (sometimes at
66
Ch. 8: Lowering Your Method Detection Limits
Figure 8 (a and b). Photo and schematic of vessel-inside-vessel technology.
67
pressures lower than its rating) due to the sudden increase

in pressure resulting from the self-sustaining auto-catalytic
decomposition (runaway reaction) of the sample (Figure
9). The use of vessel-inside-vessel technology helps to
control these self-sustaining auto-catalytic reactions by
providing a heat sink for the energy liberated during oxidization. This is accomplished by placing water in the outer
microwave vessel. The water draws the heat away from
the reaction mixture, slowing down the reaction kinetics
and preventing a runaway reaction. (Fig 10).
The amount of sample that can be safely digested
is limited by the amount of pressure generated during the
decomposition process. Current microwave vessel tech-
Temperature (C)
nology limits the internal pressure to 100 atm (1450 psi).
Figure 9. Example of runaway reaction during a closed-vessel microwave

digestion.
68
Temperature (C)
Temperature (C)
Figure 10a. Digestion of 5 grams of fresh liver using conventional microwave

decomposition. Note the runaway reaction.
Figure 10b. Digestion of 5 grams of fresh liver using vessel-inside-vessel

technology. There is no runaway reaction.
69
For most organic samples, this limits the sample size to

0.5 to 0.7 grams. In order to digest organic samples larger
than 0.7 grams, secondary reaction chemistry must be employed to lower the pressure during microwave digestion.
This is accomplished with vessel-inside-vessel technology
by adding H2O2 to the outer microwave vessel to convert
NOx and CO2 into HNO3 and HCO3 respectively.
Primary decomposition reaction
(CH2)x + 2xHNO3
xCO2 + 2xNO + 2xH2O
Secondary reactions
2H2O2
2H2O + O2
2NO + O2
2NO2
2NO2 + H2O
HNO2
CO2 +H2O
HNO3 + HNO2
HNO3 + 2NO + H2O
H2CO3
The quantitative effect of this technique is shown in

Figure 11. As you can see, this technique effectively doubles the amount of organic sample that can be digested
using closed-vessel microwave technology. There is also
no transfer of analytes from the inner vessel to the outer
vessel. (Tables 6-8.)
70
Red line = Temperature
Blue line = Pressure
Figure 11. Quantitative effect of vessel-inside-vessel technology. These

are overlay plots for the microwave digestion of polypropylene pellets.
Line A represents 0.35 grams digested using conventional microwave
digestion. Line B represents 0.35 grams digested with vessel-inside-vessel
technology. Line C represents 0.60 grams digested with vessel-inside-vessel
technology.
71
Table 6:
Microwave digestion of cell culture media, using vessel-insidevessel technology followed by ICP-AES analysis. Results of six
replicate samples.
Element
Ca
Zn
Fe
Mo
Mg
K
P
Na
Average (g/g)
454
2.22
9.44
7.03
428
19,552
7,527
79,461
% RSD
5.04
8.24
2.52
5.49
4.13
5.66
3.69
2.88
Table 7:
Iron spike recoveries for microwave digestion of cell culture
media, using vessel-inside-vessel technology followed by
ICP-AES analysis.
Spike Amount
10 g
25 g
50 g
% Recovery
95.2
103
98.8
The average spike recovery is 99.2% with an RSD of 4.3%.
72
Table 8:
Microwave digestion of SRM 1577A (Bovine Liver), using vesselinside-vessel technology followed by ICP-MS analysis.
Element
Cd
Co
Mo
Pb
V
a
Measureda (ng/g)
427 50
208 1.9
3,540 46
135 14
104 7
Certifieda (ng/g)
440 60
210 50
3,500 500
135 15
98.7 1.6
Error expressed as 95% confidence interval.
73
74
SECTION THREE
Preparing
Your Laboratory for
Trace Analysis
76
he first two parts of this book provide the theory and

background for low level trace metal analysis. This chapter
outlines simple housekeeping and laboratory techniques
that can have an immediate effect on your trace metals determinations.
Laboratory Environment
As previously discussed, the laboratory environment
can significantly affect low level trace metal analysis. If clean air
facilities are not available for any
reason, then the sample preparation and analysis areas must be
isolated from the main laboratory. Rooms with access to the
outside via doors or windows,
with high pedestrian traffic, or
with
exposed
metal
surfaces
should be avoided. Any exposed
Chapter 9
Laboratory
Housekeeping
Techniques
metal surfaces should be painted

with epoxy paint to prevent metal
contamination. The work surfaces should be sealed with a polyurthane or polyacrylic finish (clear heavy-duty contact paper can
also be used) to trap potential contamination. Airborne contamination can be controlled by covering air vents with additional
filtration media and using a portable HEPA filter system that can
be purchased from a home supply store. Floors, benches, and
apparatus should be wiped down with D.I. water regularly with
lint free towels to remove any dust accumulation.
77
Reagent Handling
Clean reagents are essential to the success of trace
metal analysis. All reagents and solutions used in trace
metal analysis must be handled with care, in order to
prevent contamination from outside sources. Reagents
should be dedicated to trace metal analysis, and not be
used for other laboratory procedures. Reagents bottles
should be stored in a clean environment to prevent the accumulation of dust, which can be transferred to the sample
during handling. For calibration solutions, a commercially
available desiccator cabinet (without the dessicant) works
well. For digestion acid bottles, large polyethylene bags
used in layers work well for 100-1,000 mL Teflon bottles.
No foreign objects should ever be introduced into the
original reagent container, and the original reagent bottle
should be opened as little as possible, and for the minimum amount of time. To minimize opening of the original
container, a sub-sample should be transferred into a clean
container for everyday use. Excess reagent should not be
returned to the original bottle under any circumstances.
Preparation Procedures
Preparation procedures for standards and samples
should be designed to minimize the potential for contamination. Large dilutions of standards or samples should be
prepared using a micropipette with disposable tips. This
approach minimizes the number of dilutions, and prevents
contamination from a glass pipette. When using this approach, the tips should be rinsed first with DI water to
78
Ch. 9: Laboratory Housekeeping Techniques
remove any dust, then with a 10% acid solution to remove

any trace metal contamination, and last with two volumes
of the final solution before use.
Traditional glass volumetric apparatus should also be
avoided, because they can contribute trace metal contamination to the solution. Plastic volumetric apparatus are
available, but disposable graduated polypropylene centrifuge tubes work best. The graduation marks are accurate
to 1% RSD or better (dilutions can be done by weight if
greater accuracy is needed). These tubes can be used for
preparation of calibration solutions as well as sample dilutions, and, because they are disposable, carry-over and
cross-contamination problems are eliminated.
Miscellaneous
As an analyst, one must be constantly aware of ones
actions, and think about how those actions will affect the
blank. Most importantly, the analyst must avoid those
actions that tend to increase the blank, or whose effects
on the blank are unknown. For example, if you are wearing
gloves to prevent contamination from your hands, but you
touch a dusty or metallic surface, your glove will pickup
trace metal contamination that can be transferred to the
sample. The best approach to trace metal analysis is to
have a routine and follow that routine for every analysis.
If you have a set routine, it will help when a contamination
problem occurs, because you can check each step until
the source of the contamination is found.
79
80
his chapter lists some of the laboratory equipment and

supplies you will need to perform trace metal analysis,
and provides contact information for many reputable
suppliers of such equipment.
Instrumentation
Acid Purification
Milestone has developed a self-contained sub-boiling
distillation system called duoPUR, which allows chemists
to make their own high-purity
acids.
The system consists of
two high-purity quartz distillation units.
Each unit contains
two infrared heating elements

that supply a maximum power
of 1,250 W, a water cooled
condenser,
bottle.
is
and
collection
Chapter 10
Equipment and
Supplies for Trace
Metal Analysis
The distillation process
microprocessor-controlled,
allowing the user to set distillation times and power level.

Distillation rates range from 10 to 200 mL per hour, depending
on the power setting
and the temperature
of the cooling water.
An added benefit of
the system is that you
can re-purify contaminated acids instead of
downgrading them.
81
Closed-Vessel Microwave Decomposition

Milestones Ethos Plus line
of microwave digestion systems
was designed with the trace
metals chemist in mind.
The
system is constructed from corrosion-resistant stainless steel.

The inner chassis is coated with
five layers of electrosprayed
PTFE for added corrosion resistance. The unique software
interface allows precise temperature ramping. You simply create a temperature profile,
and software modulates the power output to follow the
defined heating profile. A variety of digestion rotors and
accessories (evaporation, micro-inserts, etc.) are available
to accommodate all your digestion needs.
Ultra-trace Cleaning
Milestones
traceCLEAN system is a fully
automated
acid
steam
cleaning system for trace

metal analysis accessories. This self-contained
system houses an accessories rack and an acid
reservoir.
Once loaded,
the accessories are lowered into a sealed cham82
Ch. 10: Equipment and Supplies for Trace Metal Analysis
ber. Nitric acid is repeatedly evaporated and condensed

throughout the chamber, thoroughly cleaning the accessories. The main benefit of the system is that any trace
metal impurities that are present in the acid do not come
in contact with the cleaned accessories.
Clean Benches and Hoods
Several companies offer
clean benches and hoods for
trace metal sample preparation.
Labconco offers the
Purifier
Trace Metals Work
Station, designed specifically

for the demands of trace metals analysis. This work station
is made of non-metallic components and provides a class
100 working environment. This
enclosure is well-suited to
applications involving corrosive chemicals, such as acid
digestions, with its optional exhaust fan and PVC duct
work. Several other models are also available.
Terra Universal Inc. offers everything from free standing modular clean rooms to portable clean booths. All
units offer HEPA filtration, with some units providing class
1-10 working conditions. Terra also offers static control via
optional ionizers, and static dissipative materials neutralize any static imbalances that may exist in the work area.
83
High-Purity Water
There are three
major
ers
of
water
manufacturhigh-purity
equipment:
Millipore, Barnstead
International,
and
Labconoco. All three
manufacturers offer
Free-standing modular cleanroom
a variety of systems,
from water softeners and reverse osmosis to high end

polishers capable of producing Type 1 18.2 megohm-cm
water.
Reagents and Standards

Calibration and Quality Control Standards
There are three major suppliers of calibration standards for trace metals analysis. They are SPEX CertiPrep
Inc., Inorganic Ventures, and High Purity Standards. All
three companies offer single, multi-element, and custom
calibration standards for AA, ICP, and ICP-MS analysis.
SPEX offers proficiency testing samples, and Inorganic Ventures and High Purity Standards offer certified reference
materials for quality control procedures.
Resource Technologies Corporation (RTC) and the
National Instituted of Standards and Technology (NIST)
provide real world quality control material for trace
metals analysis. Samples range from soils and sediments
84
to glasses and metal alloys. Each sample comes with a

certificate of analysis identifying major and minor trace
element constitutes.
Reagents
If you are not going to make your own high-purity acid,
then you will have to buy it. For general and low trace metal analysis, high-purity acids can be purchased from J.T.
Baker, Fisher Scientific, and High Purity Standards. Each
company offers double-distilled nitric and hydrochloric
acids in high-purity Teflon bottles. For ultra-trace work,
TAMA Chemicals offers TAMAPURE AA-100, guaranteed
to contain trace metal impurities less than 100 pg/mL,
and TAMAPURE AA-10, guaranteed to contain trace metal
impurities less than 10 pg/mL. All bottles of high-purity
acid come with a certificate of analysis.
Lab Supplies
Gloves
There are many different types and brands of gloves
that are suitable for trace metals analysis. I prefer to use
N-Dex Nitrile gloves (available from Fisher Scientific) as
my first layer, and Oak powder-free vinyl gloves as my
top layer. The Oak gloves are available in standard or long
cuff. I use the long cuff because it offers better protection. The Oak gloves are available from Fisher Scientific or
Terra Universal.
85
Plasticware
As discussed in Section One, plastics are the materials
of choice for trace metals work. Nalgene makes a wide
variety of plastic bottles and accessories for trace metal
analysis. One hard-to-find item that they make is plastic
volumetric flasks. Sizes range from 50mL to 1,000mL, and
flasks are made from either polypropylene or polymethylpentene. Another item that I have found I cant live without is the free-standing graduated polypropylene tube.
These tubes are available from Fisher Scientific, and are
perfect for sample dilutions and storage.
Miscellaneous
A variety of clean chemistry accessories are available
from Terra Universal and Fisher Scientific. Items range
from tacky mats to lint free towels.
86
Suppliers Contact Information

Milestone Inc.
25 Controls Drive
Shelton, CT 06484
Telephone: 866-995-5100
Fax: 203-925-4241
Website: www.milestonesci.com
Labconco Corporation
8811 Prospect Avenue
Kansas City, Missouri 64132-2696
Phone: 816-333-8811
Fax: 816-363-0130
Website: www.labconco.com
Terra Universial Inc.
700 N. Harbor Blvd.
Anaheim, CA 92801
Phone: 714-526-0100
Fax: 714-992-2179
Website: www.terrauniversal.com
Millipore Corporation
80 Ashby Road
Bedford, MA 01730
Phone: 1-800-MILLIPORE
Website: www.millipore.com
87
Barnstead International
255 Kerper Blvd.
Dubuque, IA 52001
Phone: 1-800-446-6060
Fax: 536-589-0516
Website: www.barnsteadthermolyne.com
SPEX CertiPrep
203 Norcross Avenue
Metuchen, NJ 08840
Phone: 1-800-LAB-SPEX
Fax: 732-603-9647
Website: www.spexcsp.com
Inorganic Ventures
195 Lehigh Avenue
Suite 4
Lakewood, NJ 08701
Phone: 1-800-569-6799
Fax: 732-901-1903
High Purity Standards
4741 Franchise Street
Charleston SC 29423
Phone: 843-767-7900
Fax: 843-767-7906
Website: www.hps.net
88
Resource Technologies Corporation

2931 Soldier Springs Rd.
Laramie, WY 82070
Phone: 1-800-567-5690
Fax: 307-745-7936
Website: www.RT-Corp.com
National Institute of Standards and Technology
Standard Reference Materials Program
Building 202, Room 204
Gaithersburg, MD 20899
Phone: 301-975-6776
Fax: 301-948-3730
Website: www.nist.gov
Mallinckrodt Baker, Inc. (J.T. Baker)
222 Red School Lane
Phillipsburg NJ 08865 U.S.A.
Phone: 1-800-582-2537
Fax: 908-859-6905
Website: www.JTBaker.com
Fisher Scientific
Phone: 1-800-766-7000
Website: www.fishersci.com
89
Tama Chemicals
Distributed by
Moses Lake Industries
8249 Randolph Rd
Moses Lake, WA 98837
Phone: 509-762-5336
Fax: 509-762-5981
Nalge Nunc International
75 Panorama Creek Drive
Rochester, NY 14625
Phone: 1-800-625-4327
Fax: 585-586-8987
Website: www.nalgenunc.com
90
For Section One

1.
Harris, D.C. Quantitative Chemical Analysis, 2nd Ed., W.H.
Freeman and Company, New York, 1987. p. 38.

2.
J. Ruzicka and J. Stary, Substoichiometry in Radiochemical
Analysis, Pergamon Press, New York, 1968 p. 54-58.
3.
Murphy, T.J. The Role of the Analytical Blank in Accurate
Trace Analysis, Proceedings from the Seventh Materials Research
Symposium; U.S. Government Printing Office, Washington D.C.,
1976. p. 509-539.
4.
Howard, A.G. and Statham,
P.J. Inorganic Trace Analysis: Philosophy and Practice, Wiley, New
York, 1993 Murphy, T.J. The Role
of the Analytical Blank in Accurate Trace Analysis, Proceedings
from the Seventh Materials Re-
Appendix A
References
search Symposium; U.S. Government Printing Office, Washington

D.C., 1976. p. 509-539.
5.
P.W. Morrison, ed. Contamination Control in Electronic
Manufacturing, Van Nostrand Reinhold, New York, 1973, p. 245.
6.
Specker, H.I. Z. Erzbergbau Metallhue 17, 132 (1964).
7.
Gilbert, H. and Palmer, J.H. High Efficiency Particulate Air
Filter Units, TID-7023 USAEC , Washington D.C., August 1961.
8.
Moody, J.R., Analytical Chemistry 52, (1982) p. 1358A1376A.
9.
Whyte, W. (Ed.) Cleanroom Design, Wiley, Chichester,
1991.
91
10.
Boutron, C.F., Fresenius Journal of Analytical Chem-
istry, 337 (1990) p. 482-491.

11.
Austin, P.R., Encyclopedia of Clean Rooms, Bio-Cleanrooms and Aseptic Areas, 3rd ed., Acorn Industries, 2000.
12.
Zief, M. and Mitchell, J.L. Contamination Control in
Analytical Chemistry, Wiley, New York, 1976.
13.
Patterson, C.C. and Settle, D.M. The Reduction of
Orders of Magnitude Errors in Lead Analysis of Biological
Materials and Natural Waters by Evaluating and Controlling the Extent and Sources of Industrial Lead Contamination Introduced During sample collecting, Handling and
Analysis, Proceedings from the Seventh Materials Research
Symposium; U.S. Government Printing Office, Washington
D.C., 1976. p. 321-351.
14.
Sulcek, Z. and P. Povondra. Methods of Decomposition in Inorganic Analysis, CRC Press, Boca Raton, FL, 1989.
15.
Hetherington, G., Stephenson, G.W., Witerburn, J.A.
Electronic Engineering, May 1969 p. 52
16.
Stevens, M.P. Polymer Chemistry: An Introduction,
Oxford, New York, 1990
17.
Moody, J.R. and Lindstrom, R.M. Analytical Chemistry 49, (1977) p. 2264-2267.
18.
Laxen, D.P.H and Harrison, R.M. Analytical Chemistry
53, (1981) p. 345-350.
19.
Tschopel, P.,Kotz, L., Schulz, W., Veber, G. and Toelg
G. Fresenius Z. Anal. Chem. 302, (1980) p. 1
20.
Richter, R.C. Spectroscopy, 16(6), (2001) p. 21-24.
21.
Kuehner, E.C., Alvarez, R., Paulsen, P.J., and Murphy,
T.J. Analytical Chemistry, 44, (1977) p. 2050-2056
92
Appendix A: References
22.
Moody, J.R and Beary E.S.Talanta, 29, (1982) p.
1003-1010.
23.
Dabeka, R.W, Mykytiuk, A., Berman, S.S., and Rus-
sell, D.S. Analytical Chemistry, 48, (1976) p. 1203-1207.

24.
Richter, R.C., Link, D., Kingston, H.M., Spectroscopy,
15(1), (2000) p. 38.
For Section Two

1.
Mingos, D. M. P.; Baghurst, D. R. Chem. Soc. Rev. 20,
(1991) p. 1-47.
2.
Hoopes, T.; Neas, E.; Majetich, G. Abstr. Pap. Am.
Chem. Soc. 201, (1991) p. 231.
3.
Baghurst, D. R.; Mingos, D. M. P. J. Chem. Soc., Chem
Commun. (1992) p. 674.
4.
Richter, R.C., Link, D.D., Kingston, H.M. Analytical
Chemistry (2001) 73(1) p30A
5.
Journal of Research of the National Bureau of Standards 30, (1943) p. 110.
6.
Kingston, H. M.; Walter, P. J. Spectroscopy 7 (1992)
p. 22-27.
7.
Kubrakova, I. V.; Formanovskii, A. A.; Kudinova, T.
F.; Kuzmin, N. M. J-Anal-Chem (1999) p. 460-465.
8.
T. Noltner, Spectroscopy 5(4) 1989.
9.
Visini and Rampazzo, Department of Environmental
Science, University of Venice, Private Communication
10.
Richter, R.C. and H.M. Kingston, Department of
Chemistry, Duquesne University Private Communication.
93
11.
Link, D.L. and Kingston, H.M. Analytical Chemistry
(2000) 72(13) p. 2908-2913.

12.
Link, D.L. and Kingston, H.M., Havirilla, G.L, Colletti,
L.P. Analytical Chemistry (2002) 74(5) p. 1165-1170.
13.
Han, Y., Kingston, H.M., Richter, R.C., Pirola, C.,
Analytical Chemistry (2001) 73(6) p. 1106-1111.
14.
Sulcek, Z. and Povondra, P., Methods of Decomposition in Inorganic Analysis, CRC Press, Boca Raton, FL, 1989.
94
Appendix A: References
95
About the Author

Dr. Robert C. Richter is a research professor at Chicago State University, where he develops and coordinates
student research projects involving microwave-enhanced
chemistry. He has taught analytical chemistry and sample
preparation, for various universities and the American
Chemical Society, since 1992. He has also served as a
consultant on a wide range of corporate and governmental
projectsdeveloping microwave-assisted methods for a
variety of applications, and conducting seminars on clean
chemistry techniques.
Dr. Richter is also Senior Applications Chemist for
Milestones clean chemistry, microwave digestion, and
microwave extraction product lines. In this role, he has
helped academic and commercial laboratories across
North America to adapt themselves for clean chemistry
procedures and standards.
Articles written by Dr. Richter have been published
in Analytical Chemistry, American Laboratory News, Spectroscopy, Chemosphere, and Fresenius Journal of Analytical
Chemistry, among other publications. He has contributed
to papers presented at the Pittsburgh Conference, and at
meetings of the American Chemical Society.
96

Clean Chemistry

Uploaded by

Document Information

Original Title

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

Clean Chemistry

Uploaded by

Copyright:

Available Formats

Techniques for the Modern Laboratory

by Dr. Robert Richter

Clean Sample Preparation

by Dr. Robert Richter

Clean Sample Preparation

2003 Milestone Inc.

All rights reserved. No part of this publication may be reproduced,

Printed in the U.S.A.

Teflon, Dacron, and Tyvek are registered trademarks

Clean Chemistry: Techniques for the Modern Laboratory

Clean Chemistry: Techniques for the Modern Laboratory

per billion (ppb) range, while semiconductor applications require

control. The second part deals

Clean Chemistry: Techniques for the Modern Laboratory

Clean Chemistry: Techniques for the Modern Laboratory

all external sources of elemental contamination, and is used to

The variability of the analytical blank has little or no effect on

tude higher than the blank result.

Clean Chemistry: Techniques for the Modern Laboratory

irborne contamination is a major contributor to

objects, or from chemical processes upon materials. The

nates can also be introduced

Clean Chemistry: Techniques for the Modern Laboratory

tively remove bacteria, pollen, fly ash, and dust. Some

system to provide directional

from one end of the work area to the other essentially

Clean rooms minimize metal-bear-

Ch. 2: Laboratory Equipment

on clean rooms and clean

Laminar flow hood

standard units are good

Clean Chemistry: Techniques for the Modern Laboratory

ventional hood. These units are an inexpensive alternative

Bottles, containers, samples, reagents,

and equipment should be kept isolated from laboratory air

Ch. 2: Laboratory Equipment

Most Negative (-)

Most Positive (+)

Figure 1. Materials tendency to generate a static charge.

surfaces. Electrostatic charges are usually generated by

Clean Chemistry: Techniques for the Modern Laboratory

efore a sample is analyzed for trace metals, it has been

many different laboratory tools, containers, and apparatus,

blies can also impart contamination to the sample.

metal analysis it is best to avoid

Clean Chemistry: Techniques for the Modern Laboratory

acids, but should not be used with HF or boiling H3PO4.

Borosilicate glass is not a good material for trace

borosilicate glass, it is resistant to most acids, but should

Adapted from references 3 and 15.

Quartz (Type II)

Quartz (Type III)

Ch. 3: Materials for Trace Analysis

Clean Chemistry: Techniques for the Modern Laboratory

Ch. 3: Materials for Trace Analysis

Adapted from reference 17.

is 80C and 110C for HDPE. The use of LDPE is preferable

Clean Chemistry: Techniques for the Modern Laboratory

Fluorinated Polymers (Teflon)

Ch. 3: Materials for Trace Analysis

Adapted from reference 17.

Clean Chemistry: Techniques for the Modern Laboratory

Adapted from reference 17.