Environ Earth Sci (2016) 75:1262

DOI 10.1007/s12665-016-6073-8

ORIGINAL ARTICLE

Investigating the effect of sample disturbance, compaction

and stabilization on the collapse index of soils
R. Noorzad1 • H. Pakniat1

Received: 29 December 2015 / Accepted: 8 September 2016 / Published online: 16 September 2016
 Springer-Verlag Berlin Heidelberg 2016

Abstract Collapsible soils are problematic soils, which Introduction

support heavy loads at their natural water content, but as
their water content increases, they undergo a considerable Loess usually consists of silt-sized primary quartz particles
reduction in volume. This research presents results of sin- formed as a result of high-energy earth-surface processes.
gle oedometer tests for determining the collapse index of These processes have resulted in forming an almost con-
undisturbed and remolded soil samples. The alteration in tinuous deposit layer from the northern China plain to the
this index was investigated by applying disturbance to southeastern England with other comparable deposits
samples. Samples were remolded by different dry unit found in North America, South America and New Zealand,
weights and water contents. Additionally, some samples equating to approximately 10 % of the world’s landmass
were stabilized with cement kiln dust, lime, ammonium (Jefferson et al. 2008).
sulfate and potassium chloride to investigate their effect on These soils exhibit low dry density and moisture content
the collapse index. The results indicated a greater collapse in all testing schemes due to weak cementation bonds or
index for undisturbed samples compared to that of remol- loess structure, especially at unsaturated state in arid or
ded ones. It was also found that increasing water content semiarid areas. Moreover, large deformation, rapid settle-
and dry unit weight contribute to collapsibility reduction ment and drastic decrease in void ratio of a metastable soil
and stabilization, making it possible to minimize the col- structure occur during wetting and loading phases
lapse index to the degree of non-collapsible soil. Reduction (Al-Obaidi et al. 2013).
procedure is affected by amount of additives, dry unit The physical state of subgrade soils includes relative
weight and water content of samples. In addition, curing compaction and water content, which are mainly reflecting
time of samples was another effective factor for cement the in situ changes. For fine-grained soils, the permanent
kiln dust and lime. deformation is affected by many conditions, such as
deviator stress, load frequency, the number of load cycles,
Keywords Collapsible soil  Soil stabilization  Cement stress history, confining pressure and physical states of
kiln dust  Lime  Ammonium sulfate  Potassium chloride soils (Xiao et al. 2014).
Evidently construction projects require soils with suffi-
ciently good engineering properties. However, it is difficult
to find proper sites of construction or appropriate materials
for different construction projects within a given economic
range. Unsuitable materials with poor engineering proper-
ties are often encountered in construction projects, so it is
necessary to find a way to improve their properties in order
& R. Noorzad to make them acceptable for construction applications
rn0864@gmail.com
(Harichane et al. 2012).
1
Faculty of Civil Engineering, Babol University of Traditional techniques in geotechnical engineering are
Technology, Babol, Mazandaran, Iran often associated with some problems in terms of high costs

123
1262 Page 2 of 9 Environ Earth Sci (2016) 75:1262

and/or environmental issues. For instance, using granular In addition to modifying the characteristics of existing
bases for road construction fails its feasibility when the soil, mitigation of environmental problems due to waste
borrow site is far away from construction site. Another disposal was the main objective of these studies.
example is construction of foundations in soils with poor Usage of lime-stabilized loess in construction of irri-
bearing capacities, where costs of a deep foundation gational water reservoirs is an effective and sustainable
solution can be incompatible with the overall expected method offering good strength while contributing to per-
costs for low-budget building projects. In these cases, meability reduction (George 2001; Harichane et al. 2012;
alternative solutions such as addition of cementitious Jefferson et al. 2005; Soliman and Hanna 2010; Sujit
agents can be used to improve local soil properties (Con- Kumar and Monowar 2012).
sololi et al. 2010). Enhancement of geotechnical properties of a collapsible
Soil stabilization technique has been introduced many soil and chemical stabilization process using lime takes
years ago with the main purpose of making soils capable of place through two set of chemical reactions: short-term and
meeting specific requirements of engineering projects long-term reactions. Short-term reactions include cation
(Harichane et al. 2012). exchange, flocculation and agglomeration; these processes
Scientific techniques of soil stabilization have been are primarily responsible for modification of soil’s engi-
introduced in recent years, and the use of cementitious neering properties such as workability and plasticity
materials like Portland cement, hydraulic lime, bitumen, reduction. Long-term reactions, called pozzolanic reac-
asphalt and certain resins as stabilizer is quite common. tions, lead to create new calcium hydrates, which con-
The potential for using industrial by-products for stabi- tribute to flocculation by bonding adjacent soil particles
lization of clayey soils is promising and has been investi- together and strengthen the soil as the curing occurs.
gated. There are so many by-products that can be used for Pozzolanic reactions are time and temperature dependent,
this purpose, such as cement kiln dust, fly ashes and bottom resulting in a gradual gain in strength over time (Aldaood
ashes from coal burning, rich husk ash or ladle furnace slag et al. 2014).
(Hossain and Mol 2011; Manso et al. 2013). Investigating many by-products (alone or in combina-
Engineering properties of collapsible soils are not suf- tion with cement and lime), has shown that cement kiln
ficient for geotechnical applications; therefore, it is a dust (CKD) provides significant and durable benefits when
common practice to incorporate additives, ranging from is used for stabilization of soils. Addition of CKD increases
by-products to commercially manufactured stabilizers, to unconfined compression strength while reducing swell
improve their engineering properties. Effectiveness of an factor in stabilized soil (Kolias et al. 2005; Onsy Mohamed
additive depends on soil type as well as the field condi- and El Gamal 2012; Parsons and Kneebone 2004; Sarios-
tions. Compaction control and performance specifications seiri and Muhunthan 2008).
are needed to ensure the feasibility of construction (Evs- The addition of cement to soil produces primary and
tatiev 1988). It is possible that sampling and preparation of secondary cementitious compounds in the soil–cement
laboratory specimens may induce disturbances and exper- matrix, which improves soil properties. The primary
imental errors that lead to either an underestimate or an cementation compounds are formed by a hydration reaction
overestimate of field collapse settlements (Houston and El- and comprised of hydrated calcium silicates, calcium alu-
Ehwany 1991). Careful techniques will minimize sample minates and hydrated lime. A secondary pozzolanic reac-
disturbance and maintain basic structural integrity as fine tion between the hydrated lime and the silica and alumina
as particle to particle orientation in clay samples. Barden from the clay minerals leads to form additional calcium
and Sides investigated the engineering behavior and silicate hydrates and calcium aluminate hydrates. This
structure of selected compacted clay samples. Samples soil–cement reaction is responsible for the improvement in
compacted both wet and dry of optimum water content the strength and compressibility of treated soil (Pakbaz and
revealed no marked difference in structure but showed Alipour 2012). Successful attempts have been made in the
turbostratic structure at high scanning electron microscopy Soviet Union and Bulgaria to reduce infiltration leakage of
magnification (Bennett and Hulbert 1986). irrigation canals into loess soils by treatment with salt
Stabilization methods improve binding properties of solutions (Soliman and Hanna 2010). Also, obtained results
loess through formation of new structural bonds, as a result from reconstituted collapsible soils at the laboratory have
of which cohesion increases. This is realized by breaking shown that salts can be effectively used to reduce the
the natural skeleton in some methods, while in other, the collapse potential of soil (Abbeche et al. 2010).
skeleton is fixed and even strengthened by injection. In In the absence of mechanical disturbance of the sedi-
road construction, stabilization is commonly implemented ment, stability may be recovered through the addition of
by using cement, lime and some kind of waste materials salt to the pore water. Sides and Barden concluded that as
(Evstatiev 1988). the clay particle size is reduced, chemical additives become

123
Environ Earth Sci (2016) 75:1262 Page 3 of 9 1262

more effective in producing dispersed or flocculate struc- Test materials

tures (Bennett and Hulbert 1986).
Most of these researches have been conducted for Soil
studying the influence of stabilizers on unconfined com-
pression strength, Atterberg limits variations, maximum As mentioned before, loessial soil used in this research was
dry unit weight and optimum water content of a mixture. obtained from the vicinity of Golestan | earth dam. Figure 1
However, limited researches have focused on either the shows the location where the soil was obtained.
effects of lime and CKD on the collapse index of soils, or Soil samples were obtained from depth of about 2 m
comparison between results from pozzolan and chemical within a test pit. Undisturbed soil samples were taken from
stabilizations. thin-wall tube samplers, which were made of 110-mm
This paper presents an investigation on the effects of inside-diameter stainless steel tubing. The area ratio of the
sample disturbance, dry unit weight, water content, and tubes was about 7 %.
stabilization with CKD, lime, ammonium sulfate and The principle of the operation is to push the sampler
potassium chloride on the collapse index of soil. without rotation into the soil at a controlled penetration rate
and pressure. To inhibit drying of sample after the samples
were cut and trimmed to the desired size, they were cov-
Materials and methods ered with two layers of cloth and sealing wax (Earth
Manual 1998).
To investigate the effect of sample disturbance, dry unit A number of laboratory tests including specific gravity,
weight, water content, and stabilization with CKD, lime, particle size analysis, Atterberg limits and standard proctor
ammonium sulfate and potassium chloride on the col- test were conducted to determine physical properties of the
lapse index of samples, 140 single oedometer tests were soil. Table 1 presents a summary of physical properties of
conducted on undisturbed and remolded samples. For the soil.
this purpose, first, seven undisturbed samples were taken
using thin-wall tube sampler to determine the collapse
Lime
degree of the soil in the vicinity of Golestan | earth dam,
located in the northeastern Iran. After the tests showed a
Lime treatment is one of the most interesting techniques
moderately severe degree of collapsibility, 18 single
proposed to reduce the impact of wetting and drying cycles
oedometer tests were conducted on remolded samples
on soils. Indeed, it induces time-dependent physicochemi-
prepared without any stabilizers at 5 different dry unit
cal processes (cation exchange, pozzolanic reactions, etc.)
weights with both natural and optimum water contents to
that result in the improvement of soil behavior (Cuisinier
investigate the influence of sample disturbance, com-
et al. 2014). The main chemical compositions of lime are
paction and water content on the collapse index. Then,
indicated in Table 2.
more than 100 specimens were remolded at different dry
unit weights, different CKD and lime contents and dif-
CKD
ferent concentrations of ammonium sulfate and potas-
sium chloride; also some of them were cured with
Cement industries generate millions of tons of CKD, a by-
different curing times.
product of cement manufacturing, as a measure to control
A summary of test parameters is given below:
product quality. CKD is a fine powdery material that is
• Five dry unit weights (81, 85, 88, 90 and 95 % of collected from cement kiln exhaust gases and consists of
maximum dry unit weight), entrained particles of clinker, unreacted and partially cal-
• Two water contents (7 and 14 %). cined raw materials, and fuel ash enriched with alkali
• Four additives (CKD, lime, ammonium sulfate and sulfates, halides, and other volatiles (Sreekrishnavilasam
potassium chloride), et al. 2007).
• Three concentrations of ammonium sulfate and potas- The beneficial properties of CKD and its cost-effec-
sium chloride (0.5, 1.25 and 2 mol/lit), tiveness compared to other types of stabilizers have led to
• Seven percentages of lime and CKD (0.5, 1, 3, 5, 7.5, its use as a popular stabilization agent during the last
10 and 15 % by dry weight of the soil), decade (Sariosseiri and Muhunthan 2008). The main
• Four curing times (1 h, 7, 14 and 21 days). chemical compositions of CKD are presented in Table 3.

123
1262 Page 4 of 9 Environ Earth Sci (2016) 75:1262

Fig. 1 Location of Golestan | earth dam where the soil was obtained

Table 2 Chemical compositions of lime

Table 1 Physical properties of the soil used in experiments Chemical analysis Percentage (%)
Property value
Silicon dioxide, SiO2 1.20
USCS classification CL Aluminum oxide, Al2O3 0.69
Passing#200 sieve 99 Iron oxide, Fe2O3 0.12
Particle size analysis Calcium oxide, CaO 71.10
Specific gravity 2.69 Magnesium oxide, MgO 0.53
Liquid limit (%) 25 Sodium oxide, Na2O [0.10
Plastic limit (%) 17 Potassium oxide, K2O [0.10
Plasticity index (%) 8 Sulfur trioxide, SO3 0.10
Maximum dry unit weight (kN/m3) 18.4 Specific gravity 2.30
Optimum water content (%) 14.25 Loss on ignition 25.86

123
Environ Earth Sci (2016) 75:1262 Page 5 of 9 1262

Table 3 Chemical compositions of cement kiln dust (CKD)

Chemical analysis Percentage (%)

Silicon dioxide, SiO2 17.62

Aluminum oxide, Al2O3 4.90
Iron oxide, Fe2O3 2.58
Calcium oxide, CaO 62.09
Magnesium oxide, MgO 1.93
Sodium oxide, Na2O 0.56
Potassium oxide, K2O 3.76
Sulfur trioxide, SO3 5.79
Moisture content 0.07
Loss on ignition 4.97
Available lime index, CaO 33.70

Fig. 2 Typical compression curve in the collapse potential test

Ammonium Sulfate (ASTM 2004)

Ammonium sulfate, (NH4)2SO4, is an inorganic salt with a

number of commercial applications. It is primarily used as
a fertilizer for alkaline soils. In the soil, the ammonium ion Table 4 Classification of collapse index (Ie) (ASTM 2004)
is released and forms a small amount of acid, lowering the
Degree of collapse Collapse index Ie, (%)
PH balance of the soil, while contributing to the release of
essential nitrogen for plant growth. The produced ammo- None 0
nium sulfate is almost entirely consumed as fertilizer, Slight 0.1–2.0
while very little amount is used in the industrial aspects. In Moderate 2.1–6.0
Africa and Asia, ammonium sulfate is used, especially to Moderately severe 6.1–10.0
fertilize rice and tea. In Europe, USA and Brazil, it is often Severe [10
a component of blended and complex fertilizers. (NH4)2-
SO4 is used to produce persulfates and fire-extinguishing
powders while it has a number of applications in photog- determine the magnitude of potential collapse that may
raphy, textile, and glass industries also serves as a nutrient occur for a given vertical (axial) stress and to establish an
for yeast and bacterial cultures. index for rating the collapse potential. The collapse index
(Ie) can be determined as follows:
Potassium Chloride
df  di
Ie ¼  100 ð1Þ
h0
Potassium minerals are mined from salt deposits formed by
crystallization and/or evaporation of marine regions during where df : Dial reading at 200 kPa stress level after wetting,
various stages of the earth’s evolution. Potassium chloride mm; di : Dial reading at 200 kPa stress level before wetting,
(KCl) is widely used as a fertilizer to increase crop outturn. mm; h0 : Initial specimen height, mm (ASTM 2004).
If moisture is introduced, even without additional load,
Test method the structure collapses and significant deformations occur.
This behavior is illustrated in Fig. 2, representing a typical
Collapse potential is traditionally measured using double curve of a single oedometer test for such soil. Ie is classi-
and single oedometer tests. The amount of collapse strain fied in Table 4 (ASTM 2004).
produced when the test specimen is inundated under a Undisturbed samples were taken in the field and then
given pressure indicates a sample’s susceptibility to col- carefully trimmed in the laboratory to fit into the sample
lapse (Jefferson et al. 2005). ring (Rendell 1988). Disturbed soil samples were com-
Collapse can be defined as a decrease in height of a pacted, by hand, in the oedometer ring using distilled water
confined soil following wetting at a constant applied ver- to achieve desired density and water content (Houston and
tical stress. Single oedometer test method is used to El-Ehwany 1991).

123
1262 Page 6 of 9 Environ Earth Sci (2016) 75:1262

Results and discussion particles are pressed together, reducing pore space between
them. These pores efficiently affect the movement of water
Sample disturbance through the soil. Therefore, as pore space is decreased
within a soil, volume change potential of the soil also
The results obtained from undisturbed samples showed that decreased.
the average of natural water content was equal to 7 % and
the dry unit weight of samples was about 15 kN/m3. Single Stabilization
oedometer tests were conducted on both undisturbed and
remolded samples at natural water content and dry unit When the remolded samples were stabilized with ammonium
weight. The collapse indices of undisturbed and remolded sulfate and potassium chloride, there was a significant
samples were found to be 7.83 and 5.71, respectively. It has decrease in collapse index. The results show the collapse
been observed that although both undisturbed and remol- index of 2.69 and 3.04 for samples with relative compaction
ded samples underwent a dramatic reduction in volume equals to 85 % which stabilized with 0.5 mol/lit concentra-
when they are saturated under load, sample disturbance tion of potassium chloride and ammonium sulfate solution. It
could decrease the collapse index by about 20 %. This is can therefore be concluded that the magnitude of collapse
due to the difference in microscopic structures between index of the stabilized samples is approximately 50 % less
undisturbed and remolded samples. than that of remolded samples with no stabilization. At the
same concentration, increasing the dry unit weight of sam-
Compaction ples to 90 % of maximum dry unit weight was associated
with a reduction in collapse index down to 1.3 for samples
In order to investigate the effect of compaction on collapse stabilized with potassium chloride and 0.89 for samples
index, samples were remolded at their optimum water stabilized with ammonium sulfate. For samples with a dry
content, because their natural water content was only about unit weight equal to 95 % of the maximum dry unit weight,
7 %, and it is difficult to remold uniform samples at such a this index was reduced to 0.39 for samples stabilized with
low water content. Figure 3 shows the results of single potassium chloride and 0.17 for samples stabilized with
oedometer tests on remolded samples without additives, ammonium sulfate. This procedure was also observed for
which are compacted at different dry unit weights (in the samples that stabilized with 1.25 and 2 mol/lit concentration
following figures, Rc is defined as: Relative compaction of solution. Figure 4 shows the results of single oedometer
¼ cd=c ). Results show the collapse index of samples tests on samples stabilized with ammonium sulfate and
dmax
remolded with their natural dry unit weight (81 % of potassium chloride for 1.25 mol/lit concentration.
maximum dry unit weight) is 5.26 while it is 3.91 for Also, Fig. 5 shows that the magnitude of collapse index
samples remolded with 88 % of maximum dry unit weight decreases with an increase in solution concentration. Sta-
and 0.74 for samples remolded with 95 % of maximum dry bilized samples compacted to 85 % of maximum dry unit
unit weight. As a result, the increase in dry unit weight can weight with a low concentration of 0.5 mol/lit had a col-
reduce the collapse index by more than 80 % and change lapse index of 2.69, while it was reduced to 2.32 for the
the degree of collapse from moderate to slight. Soil com- concentration of 2 mol/lit. Although the magnitude of
paction (increasing the dry unit weight) occurs when soil collapse index decreased, it should be noted that in com-
parison with the effect of dry unit weight of samples, the
concentration of solution did not make significant change.
Applied Vertical Stress (kPa) Comparison of the collapse indices of different tested
1 10 100 1000
samples, as shown in Fig. 5, reveals that the magnitude of
0
1
collapse index decreases with an increase in solution con-
2 centration. Stabilized samples compacted to 85 % of
3 maximum dry unit weight with a low concentration of
Strain (%)

4 0.5 mol/lit had a collapse index of 2.69, while it was

5 Rc=81% reduced to 2.32 for the concentration of 2 mol/lit. Although
6 the magnitude of collapse index decreased, it should be
Rc=85%
7 noted that in comparison with the effect of dry unit weight
Rc=90%
8 of samples, the concentration of solution did not make
Rc=95%
9
significant change.
10
Also from this figure, it can be seen that samples sta-
Fig. 3 Results of single oedometer tests on remolded samples bilized with 0.5 mol/lit concentration of ammonium sulfate
without additives and compacted at different dry unit weights and compacted to 95 % of maximum dry unit weight have

123
Environ Earth Sci (2016) 75:1262 Page 7 of 9 1262

Fig. 4 Results of single (a) (b)

Applied Vertical Stress (kPa) Applied Vertical Stress (kPa)
oedometer tests on stabilized
1 10 100 1000 1 10 100 1000
samples with a ammonium
0 0
sulfate and b potassium chloride
for 1.25 mol/lit concentration 1 1
2 2

Strain (%)

Strain (%)
3 3
4 Rc=85% 4 Rc=85%
5 Rc=90% 5 Rc=90%
Rc=95% 6 Rc=95%
6
7 7

Fig. 5 Effects of concentration 5

of solution on the collapse index 4.5 Potassium Chloride , Rc=85%
of stabilized samples with 4
Collapse Index (%)

potassium chloride and Potassium Chloride , Rc=90%

3.5
ammonium sulfate
3 Potassium Chloride , Rc=95%
2.5
Ammonium Sulfate, Rc=85%
2
1.5 Ammonium Sulfate , Rc=90%
1 Ammonium Sulfate , Rc=95%
0.5
0
0 0.5 1 1.5 2 2.5
Concentration (mol/lit)

Fig. 6 Effect of compaction (a) (b)

and percentage of additive on 5 5
Collapse Index (%)
the collapse index of stabilized
Collapse Index (%)

4 Rc=85% 4 Rc=85%
samples with a lime and b CKD
which cured for an hour 3 Rc=90% 3 Rc=90%

2 2

1 1

0 0
0 2 4 6 0 2 4 6
Percentage (%) Percentage (%)

a collapse index of 0.17, while the samples stabilized with stabilized with lime and CKD equal to 5 % by dry weight
potassium chloride (with the same dry unit weight and of soil and compacted to 85 % of maximum dry unit
concentration) have a higher collapse index of 0.39. This weight was 0.13 and 0.23 for CKD and lime, respectively,
procedure was repeated with solutions with concentrations which is significant as it reduces the degree of collapse to
of 1.25 and 2 mol/lit and for compaction to 90 and 95 % of slight; on the other hand, samples stabilized with 5 % lime
the maximum dry unit weight; ammonium sulfate has more or CKD by dry weight of soil and compacted to 90 % of
contributions to the reduction in collapse index of the soils. maximum dry unit weight have the collapse index magni-
Figure 6 shows that when the compacted samples are tude of less than 0.1, which relates to non-collapsible soils.
stabilized with CKD and lime, there would be a significant This procedure was also observed in samples, which were
decrease in collapse index. Samples stabilized with 0.5 % cured for 7, 14 and 21 days. Figure 7 shows the effect of
CKD by dry weight of soil and compacted to 85 % of the compaction and percentage of additive on the collapse
maximum dry unit weight while not cured had a collapse index of stabilized samples with lime and CKD, which
index of 2.69, while the collapse index of samples at the were cured for 21 days. It should be noted that in stabilized
same condition but stabilized with lime was 2.78. The samples with lime or CKD, the difference between results
collapse index decreases with increasing stabilizer content. from remolded samples with different dry unit weights is
The results show that the collapse index of samples negligible.

123
1262 Page 8 of 9 Environ Earth Sci (2016) 75:1262

Fig. 7 Effect of compaction (a) (b)

and percentage of additive on 2.5 1.2

Collapse Index (%)

the collapse index of stabilized

Collapse Index (%)

2 Rc=85% 1 Rc=85%
samples with a lime and b CKD
which cured for 21 days Rc=90% 0.8
1.5 Rc=90%
0.6
1
0.4
0.5 0.2
0 0
0 2 4 6 0 2 4 6
Percentage (%) Percentage (%)

Fig. 8 Effects of curing time 3

and lime percentages on the
collapse index of stabilized 2.5 An hour
samples which compacted to
Colllapse Index (%)

7 days
85 % of maximum dry unit 2
weight 14 days
21 days
1.5

0.5

0
0 1 2 3 4 5 6
Percentages (%)

Figure 8 shows the effect of curing time on the collapse 1

index of samples. It can be seen that in samples with 0.5 % 0.9
Lime
Collapse Index (%)

lime (by dry weight of soil), increasing curing time to 0.8

0.7 CKD
21 days will be associated with a change in collapse index
0.6
from 2.78 (not cured samples) to 2.07. This procedure has
0.5
been also considered for CKD stabilized samples; in these 0.4
samples with 0.5 % CKD (by dry weight of soil), the 0.3
collapse index without curing time was 2.89, while it was 0.2
reduced to 1.14 after 21 days of curing time. Even though 0.1
this reduction ratio is less than 50 %, the degree of collapse 0
may be considered, which has changed from moderate to 0 5 10 15 20 25
Curing Time (days)
slight. This result also has been seen in samples stabilized
with CKD, and samples compacted at 90 % of maximum Fig. 9 Effects of lime and CKD on the collapse index of samples
dry unit weight either. stabilized with 3 % additives and compacted to 85 % of maximum
Figure 9 shows the comparison between samples stabi- dry unit weight
lized with CKD and lime. The results show that under the
same percentage of additives, dry unit weight of samples can be reconstituted with various densities, water contents,
and curing time, samples stabilized with CKD have less stabilizer percentages and curing times at the laboratory.
collapse index than samples stabilized with lime. According to the results of single oedometer tests car-
ried out on undisturbed and remolded samples of col-
lapsible soils stabilized with different concentrations of
Conclusions ammonium sulfate or potassium chloride solutions, dif-
ferent percentages of CKD or lime and treated for different
The collapse of soils is a very complex phenomenon periods of time, the following conclusions may be drawn:
affected by various physical and environmental parameters.
It is mainly caused by soil grains rearrangement after 1. Sample disturbance could reduce the collapse index of
moisture intrusion or flooding. However, collapsible soil soils.

123
Environ Earth Sci (2016) 75:1262 Page 9 of 9 1262

2. The collapse of soil decreased with an increase in dry George KP (2001) Soil stabilization field trial. Department of Civil
unit weight of samples. Engineering University of Mississippi: 1–117
Harichane K, Ghrici M, Kenai S (2012) Effect of the combination of
3. Results showed that CKD, lime, ammonium sulfate lime and natural pozzolana on the compaction and strength of
and potassium chloride reduce the collapse index. soft clayey soils: a preliminary study. J Environ Earth Sci
4. Magnitude of collapse index decreased with an 66:2197–2205. doi:10.1007/s12665-011-1441-x
increase in the concentration of ammonium sulfate Hossain KMA, Mol L (2011) Some engineering properties of
stabilized clayey soils incorporating natural pozzolans and
and potassium chloride. industrial wastes. J Constr Build Mater 25:3495–3501. doi:10.
5. At the same concentration and dry unit weight, ammo- 1016/j.conbuildmat.2011.03.042
nium sulfate had a greater effect on reduction in the Houston SL, El-Ehwany M (1991) Sample disturbance of cemented
collapse index of soils, compared to potassium chloride. collapsible soils. J Geotech Eng 117:731–752. doi:10.1061/
(ASCE)0733-9410(1991)117:5(731),731-752
6. In stabilized samples with lime and CKD, an increase Jefferson I, Rogers C, Evestatiev D, Rogers CDF (2005) Treatment of
in stabilizer content was associated with a decrease in metastable loess soils: Lessons from Eastern Europe. Ground
collapse index. improvement-case histories. Oxford, Elsevier: 723–762. doi: 10.
7. The collapse index of stabilized samples with lime and 1016/S1571-9960(05)80028-X
Jefferson I, Evstatiev D, Karastanev D (2008) The treatment of
CKD can decrease with an increase in curing time. It collapsible loess soils using cement materials. Proc Geosustain
should be noted that the role of stabilizer content is Geohazard Mitig 178:662–669. doi:10.1061/40971(310)82
more significant than curing time. Kolias S, Kasselouri-Rigopoulou V, Karahalios A (2005) Stabilisation
8. Comparing samples stabilized with CKD and lime of clayey soils with high calcium fly ash and cement. J Cem
Concr Compos 27:301–313. doi:10.1016/j.cemconcomp.2004.
showed that at the same percentage of additives, dry 02.019
unit weight of samples and curing time, samples Manso JM, Ortega-López V, Polanco JA, Setién J (2013) The use of
stabilized with CKD have lower collapse index than ladle furnace slag in soil stabilization. J Construct Build Mater
samples stabilized with lime. 40:126–134. doi:10.1016/j.conbuildmat.2012.09.079
Onsy Mohamed AM, El Gamal MM (2012) Treatment of collapsible
soils using sulfur cement. Int J Geotech Eng 6(1):65–77. doi:10.
3328/IJGE.2012.06.01.65-77
Pakbaz MS, Alipour R (2012) Influence of cement addition on the
geotechnical properties of an Iranian clay. J Appl Clay Sci
References 67:1–4. doi:10.1016/j.clay.2012.07.006
Parsons RL, Kneebone ER (2004) Use of cement kiln dust for the
Abbeche K, Bahloul O, Ayadat T, Bahloul A (2010) Treatment of stabilization of soils. Proc Geotech Eng Transp Proj
collapsible soils by salt using double consolidation method. Proc 126:1124–1131. doi:10.1061/40744(154)101
Exp Appl Model Unsatur Soils 202:69–78. doi:10.1061/ Rendell HM (1988) Comparison between naturally consolidated and
41103(376)10 laboratory consolidated loess. J Eng Geol 25:229–233. doi:10.
Aldaood A, Bouasker M, Al-Mukhtar M (2014) Soil–water charac- 1016/0013-7952(88)90028-2
teristic curve of lime treated gypseous soil. J Appl Clay Sci Sariosseiri F, Muhunthan B (2008) Geotechnical properties of palouse
102:128–138. doi:10.1016/j.clay.2014.09.024 loess modified with cement kiln dust and portland cement. Proc
Al-Obaidi QAJ, Saad Farhan I, Schanz T (2013) Evaluation of Charact Monitor Model Geosyst 179:92–99. doi:10.1061/40972
collapse potential investigated from different collapsible soils. (311)12
Multiphys Test Soils Shales, Springer Ser Geomech Geoeng: Soliman SH, Hanna A (2010) Performance of reinforced collapsible
117–122. doi: 10.1007/978-3-642-32492-5_12 soil. J Adv Anal Model Des 199:356–374. doi:10.1061/41095
American Society for Testing and Materials, ASTM (2004) Standard (365)32
test method for measurement of collapse potential of soils. Sreekrishnavilasam A, Rahardja S, Kmetz R, Santagata M (2007) Soil
Annual Book of ASTM Standards 04–08 treatment using fresh and landfilled cement kiln dust. J Construct
Bennett RH, Hulbert MH (1986) Clay microstructure 1st edn. Build Maters 21:318–327. doi:10.1016/j.conbuildmat.2005.08.
D.Reidel Publishing, Boston 015
Consololi NC, Bassani MAA, Festugato L (2010) Effect of fiber- Sujit Kumar D, Monowar H (2012) Lime stabilization of soils:
reinforcement on the strength of cemented soils. J Geotext reappraisal. J Mater Civ Eng 6:707–714. doi:10.1061/(ASCE)
Geomembr 28:344–351. doi:10.1016/j.geotexmem.2010.01.005 MT.1943-5533.0000431
Cuisinier O, Stoltz G, Masrouri F (2014) Long term behavior of lime- Xiao J, Hsein Juang C, Changjie X, Xiongwei L, Wang L (2014)
treated clayey soil exposed to successive drying and wetting. In: Strength and deformation characteristics of compacted silt from
Proceeding of geo-congress 2014 technical papers 234: the lower reaches of the Yellow River of China under monotonic
4146–4155. doi: 10.1061/9780784413272.403 and repeated loading. J Eng Geol 178:49–57. doi:10.1016/j.
Earth manual (1998) U.S. Department of the Interior Bureau of enggeo.2014.06.008
Reclamation, 3rd ed
Evstatiev D (1988) Loess improvement methods. J Eng Geol
25:341–366. doi:10.1016/0013-7952(88)90036-1

123