Appendix A

Japanese laboratory mix

test procedure

1 INTRODUCTION

The shear strength of stabilized soil is considered the most important geotechnical
characteristic which leads to the improved stiffness, homogeneity and long term sta-
bility of stabilized soil. In general, the shear strength of stabilized soil is influenced by
many factors, including characteristics of soil (water content, organic matter content,
etc.), non-uniformity of soil (due to complex natural soil structure), type and amount
of binder, curing period and temperature, and the degree of mixedness (Babasaki et al.,
1996). Hence, it is difficult to predict the strength of field stabilized soil precisely solely
by soil investigations prior to mixing. In order to determine the mix design for actual
production it is very important to perform a laboratory mix test which examines the
unconfined compressive strength of stabilized soils prepared in the laboratory, qul by
changing the type and amount of binder, curing time, and water-cement ratio. This
mix design process also contributes to quality control at the construction site. It is
important to recognize that the strength of laboratory mixed stabilized soil, qul is not
always same as the strength of field mixed stabilized soil, quf . This knowledge may pre-
vent troubles encountered at the construction site. The strength of laboratory mixed
stabilized soil is influenced by the procedure of making and curing stabilized soil.
According to the recent questionnaire survey regarding protocols for laboratory mix
test procedures, molding methods and curing conditions exhibit notable international
differences (Kitazume et al., 2009).
In this Appendix, a procedure of making and curing stabilized soil specimen is
introduced which is frequently applied in Japan to obtain the mixing condition to
assure the target strength, and to develop new binder. This procedure conforms to the
Japanese Geotechnical Society Standard (Japanese Geotechnical Society, 2009).

2 TESTING EQUIPMENT

2.1 Equipment for making specimen

2.1.1 Mold
The standard mold size is 50 mm in diameter and 100 mm in height. However, depend-
ing on the soil characteristic, the specimen diameter may be varied. In the case of clayey
396 Appendix A: Japanese laboratory mix test procedure

Figure A.1 Standard – sized lightweight mold.

or sandy soil without gravels and when the amount of soil is limited, the diameter less
than 50 mm has been used. Conversely, if the soil contains a large amount of gravels
or decayed plants, a diameter larger than 50 mm can be accepted. In the both cases,
the height of specimen is set to be 2.0 to 2.5 times the diameter.
The material for the mold is usually either cast-iron, plastic, or tin. The latter two
types of mold are referred to as lightweight molds and are popular choices today. The
merits of lightweight molds are that they are easy to tap against the surface of a table
or floor to remove air bubbles and easy to remove the specimen out of the mold. Also,
the specimen can be cured in the mold without the risk of the mold rusting. Figure A.1
shows photos of a standard-sized lightweight mold with 100 mm in height and 50 mm
in diameter. Splittable cast-iron molds are also available in various sizes based on JIS
A 1132 (Japanese Industrial Standard, 2006).

2.1.2 Mixer
A mixer should be capable of mixing soil and binder uniformly. An electric mixer
consisting of three basic parts: motor, stirring blades, and mixing bowl is specified
in the JGS standard, because the electric mixer is suitable for most types of soil:
clayey, organic, and sandy soils in most cases. Figure A.2 shows an example of electric
mixer which has been often used in Japan. The capacity of bowl ranges from 5,000 to
30,000 cm3 . Different types of mixing paddles are available as shown in Figure A.3,
but for most of the case hook type is preferred for uniform mixing. In this particular
soil mixer, the paddle revolves at 120 to 300 rpm with planetary motions of 30 to
125 rpm. The stand of the mixer enables the raising and lowering of the bowl during
mixing.

2.1.3 Binder mixing tool

When binder in slurry form is used, use a mixing bowl (typically a metal bowl) and
rubber spatula or spoon to mix the binder and water.
 Appendix A: Japanese laboratory mix test procedure 397

Figure A.2 Electric mixer.

Figure A.3 Examples of mixing paddle.

2.2 Soil and binder

2.2.1 Soil
For laboratory mix test for actual construction purpose, it is a basic principle to collect
soil samples from all soil layers to be stabilized. In order to collect soil samples from
deeper layers, a thin-walled sampler is typically used. Sampled soil should be stored
at its natural water content. The soil samples are classified based on their observation
records and soil testing results. Natural water content, consistency limits, organic
matter content, pH and grain size distribution are good indices for the classification
(see Chapter 2). The soil samples are separated into the identified layers. However,
the soil sample in a thick layer is sometimes further divided into sub-layers to take
variation in water content into consideration. Conversely, in the case where a layer
is thin and its soil characteristics are similar to those of its neighboring layer, these
layers are combined to reduce testing complexity. Each grouped soil sample is sieved
through a 9.5-mm sieve. In the case the diameter of mold used is less than 50 mm, the
soil sample is sieved through an appropriate size sieve so that the maximum grain size
of the sieved sample should be less than 1/5th of the inner diameter of mold. While
sieving, large obstacles such as shells and plants should be removed. If it is clearly found
398 Appendix A: Japanese laboratory mix test procedure

that the grain size is less than 1/5th of the inner diameter of the mold and the sample
does not contain any obstacles, this procedure can be skipped. Then, each grouped soil
sample is stirred by a mixer and its water content is measured. If it is considered that
the water content of the soil sample has been changed during the process of sampling,
transportation, and storage, the water content of the soil sample should be adjusted
to its natural water content.
The required amount of soil sample is about 500 g for a standard-sized specimen.
The total number of specimens to be tested is determined by the variations in binder
types, binder factor (or binder content), curing period (curing time), and other con-
struction control values (such as the influence of water/binder ratio), or a combination
thereof. Three or more specimens should be prepared for each mixing conditions and
curing period. It is desirable to have an extra amount of soil samples for the case of
follow-up tests or repeated tests (due to procedural errors).
Note: The sampling strategy mentioned above is applicable for mechanical mixing
with vertical rotary shafts and blades. For the shallow mixing techniques or chainsaw
type deep mixing system which involve the vertical movement of soil-binder mixture
in the actual production, soil samples may be prepared to simulate the in-situ mixing
condition such by combining the soils taken from different layers according to the
weighted average.
2.2.2 Binder
The quality of binder should be stringently assured. In general, it is desirable to use fresh
binder for the test. However, if aged binder is unavoidably used, it should be inspected
thoroughly for any quality degradation. For instance, degraded cement becomes grainy.
The binder form in the mixing test is roughly divided between the slurry form or
powder or granular form. Chemical additives are sometimes used together with the
binder, which provide a specific effect, such as accelerating or decelerating the rate
of hardening. For instance, retarding chemical additives may be used for the ease of
overlapping process of stabilized soil columns.
The required amount of binder is determined by binder factor (or binder content)
and number of specimens. Similar to the required amount of soil sample, it is desirable
to have an extra amount of binder. Tap water is generally used to make binder slurry.
However, seawater may be used for marine construction.

3 MAKING AND CURING OF SPECIMENS

3.1 Mixing materials

An optimal duration to mix soil and binder varies due to many factors such as the
type and amount of soil, the type and amount of binder and the consistency of soil-
binder mixture. The JGS standard specifies that the binder should be mixed with
the soil thoroughly to achieve uniform mixture and notes that about 10 minutes is the
ordinary practice. However 10 minutes is accepted as de facto standard. When the
mixing duration is too long, it becomes difficult to remove air bubbles from stabilized
soil in a mold since the stabilized soil may begin to harden.
Notes: It is desirable to suspend the mixing after about 5 minutes, to detach the
mixing bowl from the mixer, and to pour the stabilized soil in the mixing bowl and
 Appendix A: Japanese laboratory mix test procedure 399

Figure A.4 Tapping technique in molding procedure.

that adhered to the stirring blades to another container using a rubber spatula, to mix
it briefly by hand, then to return it to the mixing bowl, and to restart to mix it by
the mixer for another 5 minutes. Another option is to suspend the mixing every two
minutes and to mix the soil in the mixing bowl by hand. These procedures can provide
uniform mixing of the soil including the soil stuck on the mixing bowl and blades.
In the case of slurry form binder is used, splashing of the slurry may occur when
starting the mixer right after pouring the binder slurry to the soil in the mixing bowl.
It is desirable to mix the soil and the slurry by hand briefly before starting the mixer.

3.2 Making specimen

A thin layer of grease may be applied on an inner surface of mold to allow easy removal
of the specimen out of the mold after curing. Then the stabilized soil is filled in a mold
in three separate layers. After filling each stabilized soil layer, air bubbles should be
removed. Typical methods for removing air bubbles are (1) lightly tapping the mold
against a table or a concrete floor (Figure A.4), (2) hitting the mold with a mallet, and
(3) subjecting the mold to vibration. The air removing procedure is terminated once
air bubbles are no longer found on the soil surface.
In general, it is hard to remove air bubbles from stabilized soil with low consis-
tency. Also, some stabilized soils decrease in volume over time, resulting in insufficient
specimen height. To assure the proper specimen height, a sheet of hard polymer film
of 10 to 15 mm taller than the mold height is placed around the inner perimeter of the
mold so that stabilized soil can be filled above the top edge of the mold and be sealed
by sealant as shown in Figure A.5. The hard polymer film also functions to protect a
specimen when it is removed from the mold.
The water content of stabilized soil is measured for each mixing bowl. By com-
paring the water contents before and after mixing, any mistakes in material amounts
can be spotted in the early experimental stage.
Some stabilized soils become hard quickly to cause removing air bubbles difficult.
In such a case, the stabilized soil mixture should be filled in molds as quickly as possible
by increasing the number of personnel and/or dividing the making into several times
by reducing the quantity of one batch.
400 Appendix A: Japanese laboratory mix test procedure

Figure A.5 Sealing by plastic film.

The sandy soil and binder sometimes separate easily during mixing and filling into
molds. Especially it happens in the case of slurry form binder. This causes a strength
decrease of laboratory mix stabilized soil, which is thought one of the reasons for the
high strength ratio of the field strength quf to the laboratory strength qul (Sasaki et al.,
1996; Ishibashi et al., 1997). In order to prevent the separation, mix the stabilized soil
by hand in a mixing bowl and scoop and fill it in molds quickly.
In the case of uniform sandy soil with less fines content being mixed with slurry
form binder, excessive tapping of the mold for air removal may cause the density
and strength decrease. In the case of loam or clayey soil with sand being mixed with
powder form binder, the mixture can form lumps during mixing by an electric mixer.
If it happens, the lumps should be broken and filled in molds.

3.3 Curing
The specimen in the mold is covered by sealant to prevent the change of water content
and cured at 20 ± 3◦ C over a prescribed curing period. The curing period may be
selected from 1, 3, 7, 14, 28, and 91 days, etc. depending on the purpose of test and
type of binder, while it is common and desirable to include 7 and 28 days.
The following are desirable curing procedures: (i) Sealed mold and/or specimen
should be placed in a temperature and humidity-controlled container (Figure A.6(a)),
(ii) Sealed mold and/or specimen should be placed in a humid box (relative humidity
above 95%) and the box should be placed in a temperature controlled room (Figure
A.6(b)). The best care should be paid to prevent tears in the sealant material to assure
tight sealing. The reason for not curing the specimen directly under water is that the
actual stabilized soil is mostly cured underground with negligible migration of water.

3.4 Specimen removal

Once the strength of a stabilized soil specimen is found to reach sufficient level, the
specimen may be taken out of the mold during the curing process. Figure A.7 shows an
 Appendix A: Japanese laboratory mix test procedure 401

Figure A.6 Examples of curing container and box.

Figure A.7 Removal of specimen by splitting along pre-processed slits of lightweight plastic mold.

Figure A.8 Unconfined compression test on stabilized soil.

402 Appendix A: Japanese laboratory mix test procedure

Table A.1 Example format for test report.

Specimen Binder Type: Additive∗

condition Binder factor (%) Additive amount∗∗ (%)
Binder content (kg/m3 ) Number of specimens
Slurry mixing water type∗ Mold dimension (volume) φ cm × cm
( cm3 )
Water/binder ratio∗ (%)
Curing period (days)

Raw soil Container No.

water ma (g)
content mb (g)
mc (g)
w (%)
Average w (%)

Required amount of Soil, mT (g) Binder, mH (g)

materials per Water∗ , mW (g) Additive∗ , mA (g)
one batch

Stabilized Container No.

Soil ma (g)
water mb (g)
content mc (g)
w (%)
Average ws (%)

For saturated soil, determine the required material amounts by the equations below.
➀ Soil mass, mT (g) :
w ρs w
Wet density, ρT (g/cm3 ) = {1 + /100}/{1/ + /100}
=
mT (g) = × × ×
# of specimens volume of mold ρt extra∗∗∗
➁ Binder mass, mH (g):
mT w
Soil dry mass, mD (g) = /(1 + /100) =
mH (g) = × /100 =
mD binder factor
Binder content (kg/m3 ) = {10ρT /(1 + w/100)} × Binder factor
➂ Mass of slurry mixing water, mW (g)
mW (g) = × /100 =
mH water to binder ratio (%)
➃ Additive mass, mA (g)
mA (g) = × /100 =
mH Additive amount (%)
notes:
* only if used
** ratio w.r.t. binder mass
*** normally 1.1∼1.2
Table A.2 Unconfined compressive strength of various stabilized soils.

Soil Binder
Unconfined
Grain Size Composition Binder/Soil Compressive Strength
Water Ratio qu (kN/m2 )
Sample Content Sand Silt Clay Liquid Plastic Organic Powder/Slurry (amount of
Location Soil Type (%) (%) (%) (%) Limit (%) Limit (%) Content (%) Type (W/C) binder) 7 days 28 days

Yokohama Marine Soil 97.9 6.4 37.5 56.1 95.4 32.3 3.6 NP C slurry (60%) 13.5 (100) 2,140 2,870
Bay BF 13.5 (100) 1,180 1,990
NP 27.0 (200) 4,050 5,490
BF 27.0 (200) 3,690 5,960
Osaka Bay Marine Soil 93.9 3.5 30.8 65.7 79.3 30.2 2.7 NP C slurry (60%) 13.1 (100) 950 1,400
BF 13.1 (100) 980 1,470
NP 26.2 (200) 1,490 2,750
BF 26.2 (200) 3,150 4,890
Imari Bay Marine Soil 83.3 2.2 44.5 53.3 70.4 24.2 4.3 NP C slurry (60%) 12.0 (100) 540 830
BF 12.0 (100) 490 830
NP 24.0 (200) 1,130 2,060
BF 24.0 (200) 2,190 4,250
Tokyo Land Soil 54.0 5.0 53.0 42.0 44.7 23.9 3.8 NP C slurry (80%) 4.6 (50) 530 730
Prefecture BF 4.6 (50) 160 350
NP 6.8 (75) 1,260 1,760
BF 6.8 (75) 580 1,090
NP CB slurry (200%) 22.8 (250) 700 1,510
BF 22.8 (250) 1,110 2,410
Funabashi, Land Soil 14.2 95.6 3.1 1.3 – – – NP CB slurry (80%) 15.3 (300) 460 910
Chiba BF 15.3 (300) 560 1,800
Slag 15.3 (300) 1,110 2,860

(Continued)
Table A.2 Continued.

Soil Binder
Unconfined
Grain Size Composition Binder/Soil Compressive Strength
Water Ratio qu (kN/m2 )
Sample Content Sand Silt Clay Liquid Plastic Organic Powder/Slurry (amount of
Location Soil Type (%) (%) (%) (%) Limit (%) Limit (%) Content (%) Type (W/C) binder) 7 days 28 days

Fujishiro, Land Soil 236 – – – 251 92.7 25.2 NP C slurry (80%) 72.5 (250) 130 190
Ibaragi BF 72.5 (250) 140 160
For 72.5 (250) 490 780
Organic
Soil
Nangoku, Land Soil 295 – – – 272 69.1 17.6 NP C slurry (80%) 85.0 (250) 140 250
Kouchi BF 85.0 (250) 98 200
For 85.0 (250) 590 1,570
Organic
Soil
Haneda Reclaimed 160 1.0 33.0 66.0 99.1 39.7 4.8 Quicklime Powder 10 (–) 540 740
Land Soil 20 (–) 640 1,370
Yokohama Reclaimed 102.5 9.9 44.6 45.5 78.8 39.1 2.95 Quicklime Powder 10 (–) 1,670 2,740
Land Soil
20 (–) 2,350 3,720
Naruo, Marine Soil 90.2 2.0 26.1 71.9 83.0 31.4 – Quicklime Powder 10 (–) 250 690
Hyogo

Notes:
1) NP: ordinary Portland cement; BF: blast furnace cement type B
2) C slurry: cement slurry; CB slurry: cement-bentonite slurry
3) W/C: water/cement ratio
4) Binder/Soil ratio (%): Ratio of binder mass to dry soil mass;Amount of binder: binder mass (kg) per m3 of test soil
5) The unconfined compressive strengths of stabilized soil with quicklime is obtained from the figures (Terashi et al., 1997).
6) Organic contents of soil are determined according to JGS T 231 “Testing Procedure for organic content of soil’’ (chromic acid oxidation method)
 Appendix A: Japanese laboratory mix test procedure 405

example of removal of specimen by splitting along pre-processed slits of a lightweight

plastic mold. The exposed end of the specimen must be trimmed properly before remov-
ing out of the mold. The removed specimen should be put in a polyethylene bag or
wrapped with a sheet of high polymer film (such as food storage-type plastic wrap) and
placed back in the curing container to complete the curing process. To avoid possible
deformation due to excess load, the wrapped specimens should not be stacked.

4 REPORT

In the report, it is desirable to report both the binder factor as well as the binder
content, as they are most commonly used. There are other expressions on the binder
amount such as (1) the ratio of the dry weight of binder to the wet weight of soil and
(2) the ratio of volume of binder-slurry to the volume of soil.
The type and amount of chemical additives should be reported if used. Also, it
is desirable to report any data on the amounts of all materials such as soil sample
and binder measured during the preparation procedure. Table A.1 shows an example
format for the specimen preparation report (Japanese Geotechnical Society, 2009).

5 USE OF SPECIMENS

The stabilized soil specimens are mostly used for the unconfined compression test.
However they can also be used for the triaxial test, simple tensile strength test, splitting
tensile strength test, cyclic triaxial test, and fatigue strength test.

REFERENCES

Babasaki, R., Terashi, M., Suzuki, K., Maekawa, J., Kawamura, M. & Fukazawa, E. (1996)
Factors influencing the Strength of improved soils. Proc. of the Symposium on Cement Treated
Soils. pp. 20–41 (in Japanese).
Ishibashi, M., Yamada, K. & Saitoh, S. (1997) Fundamental study on laboratory mixing test for
sandy ground improvement by deep mixing method. Proc. of the 32nd Annual Conference of
the Japanese Geotechnical Society. pp. 2399–2400 (in Japanese).
Japanese Geotechnical Society (2009) Practice for making and curing stabilized soil specimens
without compaction. JGS 0821-2009. Japanese Geotechnical Society. Vol. 1. pp. 426–434
(in Japanese).
Japanese Industrial Standard (2006) Method of making and curing concrete specimens, JIS A
1132: 2006 (in Japanese).
Kitazume, M., Nishimura, S., Terashi, M. & Ohishi, K. (2009) International collaborative study
Task 1: Investigation into practice of laboratory mix tests as means of QC/QA for deep mixing
method. Proc. of the International Symposium on Deep Mixing and Admixture Stabilization.
pp. 107–126.
Terashi, M., Okumura, T. & Mitsumoto, T. (1977) Fundamental properties of lime-treated soils.
Report of the Port and Harbour Research Institute. Vol. 16. No. 1. pp. 3–28 (in Japanese).