Absorption property of light weight soil with air foam under wet sand

condition
Y. Kikuchi
Port & Airport Research Institute, Yokosuka, Japan

T. Nagatome
Toa Corporation, Yokohama, Japan

ABSTRACT: Recently, light weight soil made of dredged slurry with cement and air foam has been used for
waterfront constructions. However, the characteristics of this material can be influenced by seawater when it
is used in coastal areas, and can be weathered if it absorbs water. These phenomena are affected by the ab-
sorption properties of the material, so this study set out to evaluate the absorption properties of light weight
soil with air foam under the wet sand condition using an X-ray CT scanner. In this study, a series of absorp-
tion tests was conducted for specimens under both the water and wet sand conditions, and the absorption
mechanism was investigated by CT scanning the specimens during absorption tests. Furthermore, the light
weight soil with air foam used for backfilling a quay wall was sampled at Kobe Port Island in 2005; this site
was the first place where this material was applied, in 1995. The absorption properties of these samples sur-
rounding wet sand were investigated by X-ray CT. The results revealed that the absorption properties of light
weight soil with air foam under the wet sand condition were different from those under the water condition.

1 INTRODUCTION used for engineering purposes as a nondestructive

Light weight soil was developed for reducing self-
weight and forces acting on revetments or quay walls
induced by embankment and reclamation on soft
ground. Adding cement and light weight materials
such as air foam to soil is one means of making sta-
bilized light geo-materials. In this paper this material
is called super geo material (SGM). Recently, the
method has been widely used to counter the increase
of solid waste such as dredged slurry from construc-
tion activities in Japan (CDIT 1999, Tsuchida et al.
1996). Port construction engineers need to find an
effective way of using dredged clay, because it is
produced in large quantities by the dredging of navi-
gation channels. One typical application is as a back-
fill material for the quay walls of seaports for reduc-
ing both earth pressure and its settlement with a soil
improvement technique on soft ground.
SGM is intended to be used for waterfront con-
structions. However, the characteristics of the mate-
rial can be influenced by seawater when it is used in
coastal areas, and there is a risk that some air in the
SGM may be replaced by water over time. The most
important issue is to predict of the change of unit
weight with time. These phenomena are affected by
the permeability and absorption properties of the ma-
terial.
The X-ray computed tomography (CT) scan-
ner,which is used in medical diagnosis, has also been
testing method (Otani and Obara 2003, Desrues et al.
2006, Kikuchi 2006). In this study, a micro-focus X-
ray CT scanner installed at the Port and Airport Re-
search Institute, Japan in 2004 (Kikuchi et al., 2006)
was used to investigate the absorption mechanism of
SGM.
The purpose of this study was to evaluate the ab-
sorption properties of SGM under the wet sand con-
dition using the X-ray CT scanner. A series of ab-
sorption tests was performed with specimens under
both the water and wet sand conditions, and the ab-
sorption mechanism was investigated by CT scan-
ning the specimens during absorption tests. Then,
SGM used for backfilling a quay wall was sampled
at Kobe Port Island in 2005. This site was the first
place where SGM was applied, in 1995 (Wako et al.,
1998). The absorption properties of these samples
surrounding wet sand were investigated using the X-
ray CT scanner.
2 MICRO FOCUS X-RAY CT SCANNER AT
PARI
2.1 Micro focus X-ray CT scanner
A micro focus X-ray CT scanner system was in-
stalled at the Port and Airport Research Institute in
2004. The basic specifications and configuration are
shown in Table 1 and Figure 1.
 In this system, the voltage and current can be 2.2 X-ray CT data
changed manually and the distance between the X- X-ray computed tomography (CT) is widely used in
ray tube and specimen or X-ray television device can medical applications, but it is also an effective non-
be changed. As a result, the micro focus X-ray CT destructive inspection method that can visualize the
scanner is very flexible and can be used for scanning internal structure of specimens. CT image data are
with various magnifications. Furthermore, its 3D evaluated quantitatively using the so-called “gray
cone beam imaging function can obtain consecutive level,” which is expressed as a numerical number
images of specified cross sections. and is proportional to the density of the material.
Figure 2 shows an overview of the system. The Figure 3 shows the relationship between the aver-
shield box is designed such that triaxial compression age gray level and average density of the tests re-
tests can be performed in the shield box. sults. These results show a linear relationship, which
Table 1. Specifications of micro focus X-ray CT scanner
means that the gray level can be used as a parameter
X-ray unit Type Non-enclosure
for evaluating density distribution. In other words,
X-ray tube voltage (max.) 225 kV
the gray level method enables the structural features
X-ray tube current (max.) 1000 μA of a material to be evaluated quantitatively based on
Maximum output 135 W this relationship with CT data. Based on the results,
Focus size (min.) 4 μm a quantitative discussion of density changes in soil is
X-ray irradiation angle Corn ･ 60 deg. also considered possible.
Test samples Maximum mounting dimensions φ250×800 mm
Maximum dimensions φ200 mm
Maximum weight 588.6 N 60000
X-ray shield box Dimensions 2150×1310×2400 mm rubber
Weight Approx. 45100 N 55000
SGM
50000
cement treated clay
Gray level

X-ray shield box 45000

X-ray television device Sub CRT

(Fluoroscopic 40000
(Image intensifier, X-ray image)
CCD camera)
35000
Rotation stage Image processing air
X-ray tube computer
30000
High-voltage Monitor input 0.0 0.5 1.0 1.5 2.0
transformer board
3
Density (g/cm )
Specimen
Main CRT
Figure 3. Relationship between gray level and wet density
Image display/
Rotary Stage controller collect/reconstruct
Inter lock
vacuum pump
3 ABSORPTION TEST
Control unit
3.1 Specimens
X-ray source
Dredged clay, cement and air foam were used to
Figure 1. Configuration of micro focus X-ray CT scanner make the SGM. The dredged clay was Tokyo Bay
clay, the physical properties of which are summa-
rized in Table 2. The water content of the clay was
X-ray shield box (2,150x1,310x2,400mm)
controlled to 3.46 wL by adding seawater. Cement
was blast furnace slag cement B, the particle density
X-ray tube
of which was 3.04 g/cm3. Foaming agent was animal
protein series.
Sample SGM was made by mixing these materials under
the designated mixing conditions as shown in Table
3. The target density of SGM was 1.05 g/cm3, and
the target unconfined compression strength at 28
days of curing was 400 kN/m2.

Table 2. Physical properties of Tokyo Bay clay

Particle density Liquid limit Plastic limit Grain size distribution (%)
3
(g/cm ) (%) (%) Gravel Sand Silt Clay
2.663 103.8 33.6 0.0 4.4 36.2 59.4
Figure 2. Overview of micro focus X-ray CT scanner
Table 3. Mixing condition of SGM
Dry mass Water Cement Air foam Total
3
100
Mass (kg/m ) 211.5 760.4 70.0 8.1 1050
3
78.4 741.8 23.0 156.8 1000

Percent finer by weight (%)

Volume (l/m )
80

Plastic molds of 50 mm in diameter and 120 mm 60

in height were filled with the mixture. The top sur-
face of the mold was sealed with a plastic cover, 40

then the molds were placed in a closed box in which

20
the temperature and humidity were maintained at
20oC and 95%. The specimens were cured for 28 0
days, and then the top and bottom surfaces were 0.01 0.1 1 10 100
trimmed to 50 mm in diameter and 100 mm in Grain size (mm)
height.
Figure 5. Grain size distribution curve

3.2 Test procedure

Figure 4 shows outline images and photographs of
the absorption test.
In the absorption test under the water condition,
the specimen was soaked in water as shown in Fig-
ure 4(a). Therefore, the water could be absorbed by
the specimen not only from the side surface, but also
from the bottom surface and top surface.
In the absorption test under the wet sand condi-
tion, the specimen was soaked in wet sand as shown
in Figure 4(b). The sand used in this test was Sagami
river sand and the soil particle density was 2.74
g/cm3. The grain size distribution curve of this sand
was as shown in Figure 5. In this test, the water level
of wet sand was higher than the top surface of the
specimen. The natural water content of wet sand be-
low water level was 24%, and the degree of satura-
tion of the wet sand was about 85%.
In each condition, the water temperature was con-
trolled to 20oC. In these tests, the weight of each
specimen was measured over an elapsed period, and
the specimen was scanned by the X-ray CT scanner.
These tests were continued for 196 days.
5cm
3.3 Absorption test results
Figure 6 shows the relationship between the wet
density and elapsed time for each specimen. It was
found that the density of the specimen under the wa-
ter condition increased with elapsed days. In con-
trast, the density of the specimen under the wet sand
condition hardly increased compared with that under
the water condition.
Figure 7 shows horizontal cross-sectional images
of each specimen during the absorption test. These
images were chosen at the middle height of each
specimen. In the X-ray CT images, higher densities
are shown lighter and lower densities are shown
darker. The images of the specimen under the water
condition showed extended lighter parts from the
water contact surface with elapsed time.
Figure 8 shows high magnification CT images
around lighter parts of another SGM specimen dur-
ing the absorption test under the water condition. No
air voids were observed in the outer whiter ring area.
This observation means that there is a distinct
boundary between the absorbed zone and non ab-
sorbed zone and that the front of the absorbed zone
can be defined.
As shown in Figure 7, the absorption front ex-
tended gradually to the center of the specimen under

Specimen
Specimen the water condition. Meanwhile, the images of the
specimen under the wet sand condition showed that
10cm

lighter parts were only the water contact surface.

Figure 9 shows the vertical cross section of each
5cm

Water
specimen during the absorption tests. The images of
(a) Water condition the specimen under the water condition were ex-
tended as lighter parts from the water contact surface
to the center. This tendency was similar to that of the
absorption process in the horizontal cross-sectional
10cm 5cm

Specimen images. On the other hand, the images of the speci-

men under the wet sand condition showed that the
lighter parts were only the water contact surface.
5cm

Wet sand

(b) Wet sand condition

Figure 4. Absorption test

 1.3 50mm

water
wet sand
Wet density (g/cm )
3

1.2

100mm
1.1

1.0 Initial 56 days 196 days

0 50 100 150 200 (a) Water condition
Elapsed time (days)

Figure 6. Relationship between wet density and elapsed time

Initial 56 days 196 days

10mm
(b) Wet sand condition
Initial 56 days 196 days
Figure 9. Vertical cross-sectional images
(a) Water condition
Figure 10 shows the relationship between the ab-
sorbed distance from the surface and elapsed time.
Here, the absorbed parts were where air foam was
replaced by water. The absorbed distance was calcu-
lated using the horizontal cross-sectional X-ray CT
data. The absorption velocity of the specimen under
the water condition was about 0.1 mm/day, and un-
Initial 56 days 196 days der the wet sand condition was less than 0.01
(b) Wet sand condition mm/day, indicating a significant difference in ab-
sorption properties of SGM under the two condi-
Figure 7. Horizontal cross-sectional images tions. The reasons for these results are now being in-
vestigated.

10mm 2mm 25
Absorbed distance from surface (mm)

20

15

10 water
wet sand

5
(a) Whole (b) High magnification
0
Figure 8. Whole and high magnification images of SGM under 0 50 100 150 200
absorption test
Elapsed time (days)

Figure 10. Relationship between absorbed distance from sur-

face and elapsed time
4 FIELD INVESTIGATION
4.1 Sampling of SGM at Kobe Port Island
The SGM for backfilling a quay wall was sampled at
Kobe Port Island in 2005. This site was destroyed by
the Kobe earthquake in 1995, and SGM was then
used for the restoration construction work.
Figure 11 shows the cross section of a recon-
structed quay wall at Kobe Port Island at that time.
The top surface of SGM was covered by wet sand to
a thickness of 65 cm. The soil particle density was
2.70 g/cm3, and the natural water content of the wet
sand was about 10%. The top of the wet sand was
covered by 22 cm thick pavement. The top surface of
SGM was D.L. +3.4 m and the bottom surface of
SGM height was D.L. -1.0 m. The high water level
of the front quay was D.L. +1.7 m and the low water
level was D.L. +0.0 m.
The dredged clay was Kobe Bay clay, the physical
properties of which are summarized in Table 4. The
water content of the clay was controlled to 2.79 wL
by adding seawater. The cement was blast furnace
slag cement B, and the foaming agent was surfactant.
SGM was made by mixing these materials in the
designated mixing condition as shown in Table 5.
Sampling point (D.L.+3.40)
Caisson
Backfill stone
Rubble
Embankment mound
Replaced sand
Figure 11. Cross section of reconstructed quay wall at Kobe
Port Island
Table 4. Physical properties of Kobe Bay clay
Particle density Liquid limit Plastic limit
3
(g/cm ) (%) (%) Gravel Sand
2.709 97 41 3 14
Table 5. Mixing condition of SGM
Controlled clay Cement Air foam
3 3 3 3
(kg/m ) (kg/m ) (l/m )
849 140 279
This study used the X-ray CT scanner to confirm
the stability of the internal structure of the field
SGM that had been cast 10 years earlier. The top
surface of the SGM was sampled as a 300 mm cubic
block at the site. Then, this block was cut into a col-
umn of 50 mm in diameter and 100 mm in height in
the laboratory. The specimens were taken at 0-100
mm and 120-220 mm from the top surface of the
SGM. The density of the 0-100 mm specimen was
1.12 g/cm3 higher than the design density of 1.00
g/cm3, and the density of the 120-220 mm specimen
was 1.02 g/cm3.
4.2 Investigation of the inside of the field SGM
Figure 12 shows horizontal cross-sectional images at
several depths. The image of 25 mm depth from the
top surface of the SGM shows almost no black parts
such as air foam, but the black parts gradually in-
crease with greater depth.
Figure 13 shows the vertical cross section of each
specimen. The image of the 0-100 mm specimen
shows lighter parts extending from the top to the
middle height. These lighter parts are considered to
be absorption zones. The bottom of this specimen
showed some black parts such as air foam. The im-
age of the 120-220 mm specimen revealed that it
was almost entirely air foam.
Crane foundation
Light weight soil (SGM)
Reclaimed sand
improved by SCP
Reclaimed sand
10mm
(a) 25 mm depth (b) 50 mm depth (c) 75 mm depth
Soft marine clay
(d) 145 mm depth (e) 170 mm depth (f) 195 mm depth
Figure 12. Horizontal cross-sectional images
50mm
Grain size distribution (%)
Silt Clay
54 28
100mm
Target density
(g/cm )
1.0
(a) 0-100 mm depth (b) 120-220 mm depth
Figure 13. Vertical cross-sectional images
 Figure 14 shows the density distribution of the
SGM specimens estimated from CT data. The den-
sity from the top surface to 30 mm depth was con-
stant at about 1.20 g/cm3. The density from 30 mm
to 70 mm depth decreased with increasing depth, and
the density from 70 mm depth was about 1.03 g/cm3.
The area of complete absorption was to 30 mm deep
from the SGM surface after 10 years, so the absorp-
tion velocity under this condition was about 0.01
mm/day.
0.9
0
40
Depth from the SGM surface (mm)
80
120
160
200
240
Figure 14. Density distribution
Next, another specimen from this site was cut
from a 30 cm cubic block sample, and was soaked in
water. The specimen size was 50 mm in diameter
and 100 mm in height, and the depth from the top
surface of the SGM was 120-220 mm. The condi-
tions of this test were the same as those shown in
Figure 4(a). Figure 15 shows horizontal cross-
sectional images of the specimen during the absorp-
tion test. These images were chosen at the middle
height of the specimen. The images reveal extension
of lighter parts from the water contact surface after
93 days; these lighter parts were absorption zones.
The absorption velocity under the water condition
was about 0.1 mm/day. This result was almost the
same tendency as the laboratory test results. It is
considered that the absorption velocity in the field
was very slow, because the field SGM was sur-
rounded by wet sand.
10mm
(a) Initial
Figure 15. Horizontal cross-sectional images
5 CONCLUSION
This paper reported the absorption properties of
SGM under the wet sand condition using an X-ray
CT scanner.
A series of absorption tests was conducted for
specimens under both the water and wet sand condi-
tions. The absorption properties of SGM under the
wet sand condition were different from those under
the water condition. The absorption velocity of SGM
3
Wet density (g/cm )
under the wet sand condition was more than 10
times slower than that under the water condition.
1.0 1.1 1.2 1.3
SGM used for backfilling a quay wall was sam-
pled at Kobe Port Island in 2005 and investigated.
The absorption part was several centimeters from the
top surface of the SGM, and the absorption velocity
was about 0.01 mm/day. It is considered that the ab-
sorption velocity of the field SGM was low because
it was covered by wet sand. In this way, the absorp-
tion properties of SGM depend on the surrounding
conditions.
REFERENCES
Coastal Development Institute of Technology (CDIT): Techni-
cal manual on Light-Weight soil method for constructions
of port and harbor facilities, 1999. (in Japanese)
Desurues, J., Viggiani, G. and Besuelle, P.: International
Workshop Advances in X-ray Tomography for Geomateri-
als, ISTE, 2006.
Kikuchi, Y.: Engineering properties investigation of man-made
composite geo-materials with micro X-ray CT, Interna-
tional Workshop Advances in X-ray Tomography for Geo-
materials, ISTE, pp53-78, 2006.
Kikuchi, Y., Mizutani, T., Nagatome, T. and Hata, T.: Applica-
bility of Micro-focus X-ray CT Scanner for Geomaterials,
Technical Note of the Port and Airport Research Institute
(1125), 21p., 2006. (in Japanese)
Otani, J. and Obara, Y.: International Workshop on X-ray CT
for Geomaterials –GeoX2003 –, Balkema, 2003.
Tsuchida, T., Takeuchi, D., Okumura, T. and Kishida, T.:
Development of light-weight fill from dredgings, Proc. of
Environmental Geotechnics, Balkema, pp.415-420, 1996.
Wako, T., Tsuchida, T., Matsunaga, Y., Hamamoto, K., Ki-
shida, T. and Fukasawa, T.: Use of artificial light weight
material (treated soil with air foam) for port facilities, Jour-
nal of Geotechnical Engineering, JSCE, No.602/IV-40,
pp35-52, 1998. (in Japanese)
(b) 93 days