Beruflich Dokumente
Kultur Dokumente
h i g h l i g h t s
The mechanical properties of soft clay stabilized with cement-based stabilizer were investigated.
The predicted models for compressive strength and secant modulus were analyzed.
The effect of the supplementary cementing materials was conrmed by SEM imaging.
a r t i c l e
i n f o
Article history:
Received 10 June 2014
Received in revised form 18 August 2014
Accepted 27 August 2014
Available online 18 September 2014
Keywords:
Clay stabilizer
Development and analysis
Unconned compressive strength
Secant modulus
Pozzolanic effect
a b s t r a c t
Sodium silicate and promoters have been widely used as supplementary cementing materials substituting ordinary Portland cement to improve the properties of stabilized soils. In this paper, the developments of mechanical properties of specimens stabilized with cement-based stabilizer, which consists
of cement, sodium silicate and composite promoter, is investigated by unconned compressive strength
(UCS) test and scanning electron microscope (SEM). The failure strain and secant modulus are also
obtained in UCS tests. The test results indicate that the supplementary cementing materials perform
effectively pozzolanic reactions and improve the mechanical properties of cement stabilized clay. The
contribution of pozzolanic effect is regarded akin as an addition of cement and the total cement content
is the summation of cement content and equivalent cement content of the supplementary cementing
materials. According to this premise, the claywater/cement ratio hypothesis for stabilized clay is proposed for analyzing and assessing the development of the mechanical properties. The phenomenological
models for predicting the compressive strength and secant modulus of specimens stabilized with the
selected cement-based stabilizer are developed and veried. The comparison of predicted results and laboratory results indicates that the deviation is mostly within 10%. The microstructural analysis observes
the changes of cementitious products in stabilized clays and conrms the pozzolanic effect of the supplementary cementing materials. On account of the efciently pozzolanic effect, the addition of sodium silicate (<2%) and composite promoter (<4%) can be equivalent to several times cement content. Hence, the
economic and environmental mix design is developed with the addition of the supplementary cementing
materials.
2014 Elsevier Ltd. All rights reserved.
1. Introduction
Clayey soils occur in most regions of South and East China [1,2].
Soft clays are often encountered on many civil engineering project
sites, which lack sufcient strength to support the loading either
during construction or throughout the service life. To improve
the strength and stiffness of those less competent soils, soil stabilization with cementitious materials has been widely practiced
[35].
Corresponding author. Tel.: +86 13917109022.
E-mail address: hntchen@sjtu.edu.cn (C. Bing).
http://dx.doi.org/10.1016/j.conbuildmat.2014.08.087
0950-0618/ 2014 Elsevier Ltd. All rights reserved.
355
Table 1
Chemical composition of clay and OPC.
Oxide
OPC
57.02
3.63
16.42
6.79
3.68
0.81
3.59
0.05
6.43
21.60
64.44
4.13
4.57
1.06
0.11
0.56
1.74
0.76
The binding agents consist of Portland cement and the supplementary cementing materials. ASTM Type I ordinary Portland cement (hereinafter referred to as
cement, OPC) was used for all stabilized clay mixtures in this study. The chemical
composition of OPC is also shown in Table 1. The supplementary cementing materials are sodium silicate and composite promoter. Sodium silicate (SS), a syrupy
liquid, consists of SiO2 (29.48%), Na2O (9.52%) and the silica modulus (molar ratio
of SiO2 and Na2O) is 3.2. The density and pH are 1.43 g/cm3 and 11.98, respectively.
Composite promoter (CN) consists of sodium hydroxide and calcium chloride at the
mass ratio of 1:1. And sodium hydroxide and calcium chloride used in this study are
both chemically pure.
2.2. Casting, curing, and testing of specimens
In order to investigate the mechanical properties of OPC and CSCN stabilized
clays, different contents of clay stabilizers were admixed with specic amounts of
oven-dry clay. For preparing the samples, the water was added through a two-step
process. Firstly, the predetermined quantity of water was initially mixed with the
oven-dry clay, and the mass ratio of water and clay was 0.7. Then the rest of the
water was blended with the binders at the mass ratio of 0.5, and the binders consisting of the desired amount of promoter, OPC and sodium silicate were added
by the weight of dry clay. The soluble promoter was added into the mixtures in
the form of solution, and the sequential mixing with CaCl2 solution followed by
NaOH solution was selected. Initial mixing was carried out in a laboratory mixer
for at least 10 min and the mix was subsequently transferred to a cubic mold
(70.7 mm in length). For squeezing the air and achieve a homogeneous mixture,
the mold was put onto vibrating table to vibrate for at least 2 min. Then, the mold
was sealed and stored in the curing room (20 2 C, 98 2% RH) for 7 days. Afterwards, the stabilized samples were demolded and put into airtight vinyl bags in
the curing room until the testing ages.
The unconned compressive strength (UCS) test was performed on the samples
after 7, 28, 60 and 90 days of curing. The rate of vertical displacement was xed at
0.5 mm/min. At least three replicates of each sample set were prepared and tested
under the same conditions to assure reproducibility. In most cases, the results
under the same testing condition were reproducible with low mean standard deviations, SD (SD/
x < 10%, where
x = the mean strength value). The failure strain (ef) of
the stabilized specimens was determined in UCS tests. The secant modulus (Es) is
dened as the ratio of one half of the compressive strength to the axial strain corresponding to this stress.
To conrm the pozzolanic reactions of the supplementary cementing materials
and examine the micromorphological change of the stabilized specimens, SEM
imaging was performed on selected samples. SEM samples were prepared by following a procedure suggested in previous study [35]. A 1 1 1 cm cubic specimen was trimmed off and then air dried in a desiccator at ambient temperature.
The dried specimen was broken into two parts and the debris on the surface was
removed with an adhesive tape. The specimen was mounted on an alumina stud
with conductive tapes, and then sputter coated with goldpalladium alloy.
356
Table 2
Mechanical properties of OPC stabilized clay at different curing ages.
OPC (%)
wc (%)
wc/C
10
12
15
20
25
30
40
50
60
70
80
75.0
76.0
77.5
80.0
82.5
85.0
90.0
95.0
100.0
105.0
110.0
7.50
6.33
5.17
4.00
3.30
2.83
2.25
1.90
1.67
1.50
1.38
7 days
28 days
60 days
7 days
28 days
60 days
7 days
28 days
60 days
34.4
51.4
120.0
198.1
286.6
385.7
706.6
1473.9
1556.8
2343.7
2663.4
79.7
99.8
232.7
432.9
565.3
683.9
1147.4
2376.3
3277.7
3687.3
4806.7
106.8
142.7
335.9
538.6
736.8
845.2
1577.0
2585.6
4253.7
5043.9
6242.2
3.86
1.94
2.64
1.58
1.51
1.40
1.38
1.63
1.77
1.44
1.61
2.85
1.66
1.91
1.09
1.38
1.28
1.54
1.82
1.72
1.95
2.14
1.16
1.46
1.63
1.68
1.45
1.34
1.46
1.76
1.89
2.09
2.22
1.4
6.1
16.1
47.3
46.4
52.3
67.8
137.5
171.6
196.3
228.3
4.4
16.0
30.6
57.8
78.9
71.9
113.8
270.0
277.4
309.4
377.5
32.3
33.1
40.8
69.7
82.5
84.6
154.2
325.8
306.6
349.8
421.6
Table 3
Mechanical properties of CSCN stabilized clay at different curing ages.
Mix No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
OPC (%)
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
SS (%)
1
1
1
1
2
2
2
2
3
3
3
3
4
4
4
4
CN (%)
1
2
4
6
1
2
4
6
1
2
4
6
1
2
4
6
wc (%)
76.0
76.5
77.5
78.5
76.5
77.0
78.0
79.0
77.0
77.5
78.5
79.5
77.5
78.0
79.0
80.0
7 days
28 days
60 days
7 days
28 days
60 days
7 days
28 days
60 days
192.6
213.8
312.7
299.1
277.3
312.3
489.2
347.5
217.3
359.3
325.7
345.1
5.3
25.6
162.7
138.8
348.7
436.6
566.3
627.9
426.9
546.7
739.4
717.3
301.7
456.4
441.5
474.5
40.7
105.3
301.4
266.9
497.6
548.0
756.5
697.8
529.9
627.8
917.6
817.7
386.5
535.0
568.6
625.8
2.00
1.99
1.07
0.64
1.32
1.17
1.05
0.76
0.99
1.74
1.53
2.43
3.76
3.44
2.37
1.66
1.28
1.15
0.92
0.83
1.38
1.46
0.91
0.76
1.97
1.12
2.02
1.27
2.71
2.73
1.30
1.21
1.13
1.34
1.41
1.73
1.45
1.12
0.85
0.93
1.57
1.44
1.81
1.63
24.2
30.3
55.23
137.8
50.3
58.5
86.1
173.3
38.6
63.2
65.4
48.9
0.1
0.5
31.0
18.3
64.4
78.4
99.6
88.5
75.0
91.3
106.1
113.3
26.9
97.1
40.0
93.9
0.8
12.3
83.8
71.3
81.4
92.2
112.9
53.8
88.1
100.6
121.5
98.4
63.4
79.5
56.4
117.8
For the reason of the very high solubility and diffusion rates of CaCl2
and NaOH in water in contrast to the solubility and diffusion rates
of Ca(OH)2, CaCl2 and NaOH solutions are sequentially permeated in
microporosity or very small pores of stabilized clays. The formation
of Ca(OH)2 in microporosity or very small pores developed much
stronger soilCa(OH)2 pozzolanic reactions than those caused by
directly adding Ca(OH)2 [36]. The possible mechanisms may be that
render CaCl2 and NaOH solutions are sequentially permeated stabilized clays. Hence, the pozzolanic effect of the supplementary
cementing materials can be regarded akin as an addition of cement
[37]. By considering that the content of sodium silicate and composite promoter can be equivalent to cement content, the equivalent
cement content (Ce) is calculated by the following equation:
C e k1 C1 k2 C2
where k1 and k2 are pozzolanic factors, C1 and C2 are sodium silicate and composite promoter content, respectively.
The total cement content (C) for the stabilized specimens is thus
the summation of input of cement (Ci) and equivalent cement content (Ce). The claywater/cement ratio hypothesis for the stabilized
clays developed as follows For given set of cement stabilized clay
specimens, the strength development depends only on the clay
water/cement ratio, wc/C [38]. The Ce is mainly dependent upon
the pozzolanic effect governing by the content of supplementary
cementing materials and alters with curing age.
To predict the mechanical properties of CSCN stabilized clay
specimens, the phenomenological models of OPC stabilized clay
and values of the pozzolanic factors are needed. The phenomenological models are the mathematical formulas which can
357
4000
800
15% OPC wc=5.17
3500
600
3000
2500
8% SCSCN
10% SCSCN
12% SCSCN
14% SCSCN
16% SCSCN
700
2000
1500
500
400
300
1000
200
500
100
0
0
7 days
28 days
60 days
6000
5000
4000
3000
2000
1000
0
1
wc /C
Fig. 3. Fitted curves of strength development in the OPC stabilized clay using the
claywater/cement ratio hypothesis.
qu
wc =CB
Table 4
Mechanical properties of SCSCN stabilized clay at different curing ages.
Mix No.
1
2
3
4
5
OPC (%)
5
6.25
7.5
8.75
10
SS (%)
1
1.25
1.5
1.75
2
CN (%)
2
2.5
3
3.5
4
wc (%)
74.0
75.0
76.0
77.0
78.0
7 days
28 days
60 days
7 days
28 days
60 days
28 days
60 days
86.3
124.4
203.1
328.6
489.2
164.5
247.7
416.9
583.4
739.4
221.4
327.6
532.7
716.2
917.6
2.29
1.43
1.26
1.30
1.05
1.84
1.55
1.46
1.01
0.91
1.50
1.17
1.11
1.37
0.85
10.5
36.4
49.0
62.7
86.1
36.9
43.5
68.1
94.0
106.1
41.5
61.8
78.6
120.0
121.5
358
Table 5
The C, Ce and pozzolanic factors for the compressive strength of CSCN stabilized clay at different curing ages.
Mix no.
Ci
C1
C2
1
2
3
5
6
7
10
10
10
10
10
10
1
1
1
2
2
2
1
2
4
1
2
4
7 days
28 days
Ce
19.26
20.21
23.84
22.43
23.67
28.70
9.26
10.21
13.84
12.43
13.67
18.70
Factors
Ce
k1 = 5.13
k2 = 2.19
19.91
21.90
24.57
21.71
24.07
27.45
9.91
11.90
14.57
11.71
14.07
17.45
5958
qu7days
qu28days
qu60days
R2 0:997
2:5
wc =C
10601
wc =C2:5
13732
wc =C2:5
4a
R 0:998
4b
R2 0:996
4c
wc
C 0 1
B
60 days
A
qu1
Ce
Factors
k1 = 4.89
k2 = 2.35
20.16
21.09
24.31
20.81
22.41
26.43
10.16
11.09
14.31
10.81
12.41
16.43
k1 = 4.31
k2 = 2.39
sodium silicate. The C of CSCN stabilized clay specimens at different curing ages can be determined as the following equations:
6a
6b
6c
That is to say, to achieve the equivalent strength of specimens stabilized with 20% OPC, only about 8.5% OPC, 1.7% sodium silicate and
3.4% composite promoter are needed when the mass ratio of
sodium silicate and composite promoter is 1:2.
Based on the above analysis, the theoretical prediction models
for the compressive strength of CSCN stabilized clay specimens
are drawn and its mathematical expressions are expressed by combination of Eqs. (4) and (6):
5958
q0u7days
wc
C i 5:13C12:19C2
7a
2:5
10601
q0u28days
2:5
7b
2:5
7c
wc
C i 4:89C12:35C2
13732
q0u60days
wc
C i 4:31C12:39C2
1100
600
1000
actual
predicted = actual
10% from predicted
actual
predicted = actual
10% from predicted
900
500
Factors
400
300
200
800
700
600
500
400
300
200
100
100
0
0
100
200
300
400
500
600
100
200
300
400
500
600
700
800
359
3.0
1200
1100
actual
predicted = actual
10% from predicted
1000
900
7 days
28 days
60 days
2.5
800
2.0
700
600
1.5
500
400
1.0
300
200
0.5
100
0
0
100 200 300 400 500 600 700 800 900 1000 1100 1200
0.0
2.4
2.8
3.2
3.6
4.0
4.4
4.8
wc/C'
Es
wc =CN
where Es is the secant modulus of OPC stabilized clay at a given curing age, wc is water content in the mixture expressed in percentage,
C is the total cement content, A and B are empirical constants.
Fig. 9 shows the phenomenological models of secant modulus
development in the OPC stabilized clay by tting the experimental
results on the basis of the generalized two-dimensional Eswc/C
plot. According to the tted curves, the general relationship
between wc/C and Es at different curing ages can be advanced as:
480
4.0
420
3.5
3.0
7 days
28 days
60 days
2.5
2.0
1.5
7 days
28 days
60 days
360
300
240
180
120
60
1.0
0
0.5
wc/C
Fig. 7. Failure strain of OPC stabilized specimens at different curing ages.
wc/C
Fig. 9. Fitted curves of secant modulus development in the OPC stabilized clay
using the claywater/cement ratio hypothesis.
360
Table 6
The C, Ce and pozzolanic factors for the secant modulus of CSCN stabilized clay at different curing ages.
Mix no.
Ci
C1
C2
1
2
3
5
6
7
10
10
10
10
10
10
1
1
1
2
2
2
1
2
4
1
2
4
417
Es7days
Es28days
Es60days
7 days
28 days
Ce
16.43
18.68
26.14
24.53
26.79
33.40
6.43
8.68
16.14
14.53
16.79
13.40
Factors
Ce
k1 = 5.70
k2 = 2.63
21.22
23.74
27.35
23.18
25.84
28.48
11.22
13.74
17.35
13.18
15.84
18.48
R2 0:980
wc =C1:86
691
wc =C1:86
790
wc =C1:86
R2 0:964
Ce
Factors
k1 = 5.33
k2 = 2.74
22.39
24.10
27.23
23.52
25.42
28.51
12.39
14.10
17.23
13.52
15.42
18.51
k1 = 5.48
k2 = 2.69
9b
E0s7days
417
wc
C i 5:70C12:63C2
R2 0:955
9c
wc
C 0 1
N
12a
1:86
691
E0s28days
1:86
12b
1:86
12c
wc
C i 5:33C12:74C2
790
E0s60days
wc
C i 5:48C12:69C2
10
M
Es1
140
11a
11b
100
60
40
20
20
40
60
80
100
120
140
180
actual
predicted = actual
10% from predicted
80
80
11c
The pozzolanic factors vary with curing age. The 60-day pozzolanic
factors are higher than 28-day pozzolanic factors, which is different
from the tendency in the pozzolanic factors for compressive
strength. By combination of Eqs. (9) and (11), several mathematical
90
actual
predicted = actual
10% from predicted
120
actual
predicted = actual
10% from predicted
160
70
60
50
40
30
20
140
120
100
80
60
10
0
Factors
9a
Similar with the phenomenological models for the strength development in OPC stabilized clay, the increase of parameter M with
curing age is signicant and parameter N keeps constant.
The total cement content (C) for secant modulus of CSCN stabilized clay can be calculated based on the following expression:
60 days
40
0
10
20
30
40
50
60
70
80
90
40
60
80
100
120
140
160
361
(Fig. 13(a)), there are some brous CSH gels appeared in 10%
OPC stabilized clay (Fig. 13(b)). Besides, more brous CSH gels
are formed as the OPC content increasing to 20% (Fig. 13(c)). However, as shown in Fig. 13(d)(f), the SCSCN stabilized clay does not
have many brous CSH gels. There exist much anomalously
agglomeration and thin laminated CSH gels in the specimen of
SCSCN stabilized clay, which is much higher than that of the OPC
stabilized clay. Although the microstructure of specimens stabilized with 12% and 16% SCSCN is still rough, the discrete particles
were bonded signicantly closely and the gaps among the particles
were more tightly lled. This implies that the supplementary
cementing materials can react with OPC and active clay minerals
and generates higher amount of hydration and pozzolanic products
which bond among the clay particles. The higher degree of bonding
and more compact microstructure are believed to result in higher
compressive strength and secant modulus.
Fig. 13. SEM photos of (a) untreated clay, (b) 10% OPC, (c) 20% OPC, (d) 8% SCSCN, (e) 12% SCSCN and (f) 16% SCSCN stabilized specimens for 28 days.
362
6. Conclusions
1. The addition of the supplementary cementing materials
(i.e. sodium silicate and composite promoter) can signicantly improve the compressive strength and secant
modulus of OPC stabilized clay. The optimal improvement
is found at the sodium silicate content of 2% and the
composite promoter content of 4%. Hence, the analysis for
compressive strength and secant modulus is based on the
test results of specimens stabilized with less than 2% of
sodium silicate and 4% of composite promoter.
2. Based on claywater/cement ratio and pozzolanic effect
hypothesis, wc/C is used as the prime parameter to achieve
the phenomenological models for predicting the mechanical properties of specimens stabilized CSCN and the supplementary cementing materials can be regarded akin as the
corresponding addition of cement. Therefore, the summation of OPC content and equivalent cement content of the
supplementary cementing materials is the total cement
content.
3. The interrelationship between mechanical properties (i.e.
compressive strength and secant modulus) and wc/C at
different curing ages is proposed. The models are useful
in predicting the compressive strength and secant modulus
wherein the content of OPC and the supplementary
cementing materials varies. And the deviation between
predicted values and laboratory results is mostly within
10%. Its notable that it is difcult to identify an accurate
relationship between failure strain and claywater/cement
ratio.
4. With the aid of SEM imaging, brous and thin laminated
CSH gels are formed in the specimens stabilized with
OPC and SCSCN, respectively. A more compact microstructure is formed in the specimens stabilized with SCSCN,
which conrmed that the supplementary cementing
materials perform effectively pozzolanic reactions. The
addition of the supplementary cementing materials is
recommended for the economic and environmental mix
design.
Acknowledgement
This research work was nancially supported by the National
Natural Science Foundation of China (Nos. 50708059 and
51379122) and National Key Technology R&D Program (No.
2014BAL03B03).
References
[1] Gao GR. Formation and development of the structure of collapsing loess in
China. Eng Geol 1988;25(2):23545.
[2] Gao GR. The distribution and geotechnical properties of loess soils, lateritic
soils and clayey soils in China. Eng Geol 1996;42(1):95104.
[3] Terashi M, Tanaka H, Mitsumoto T, Niidome Y, Honma S. Fundamental
properties of lime and cement treated soils (2nd report). Rep. Port Harbour
Res. Inst. 1980;19(1):3362 [in Japanese].
[4] Miller GA, Azad S. Inuence of soil type on stabilization with cement kiln dust.
Constr Build Mater 2000;14(2):8997.
[5] Degirmenci N, Okucu A, Turabi A. Application of phosphogypsum in soil
stabilization. Build Environ 2007;42(9):33938.
[6] Rogers CDF, Glendinning S, Roff TEJ. Lime modication of clay soils for
construction expediency. Geotech Eng 1997;125(4):2429.
[7] Kawasaki T, Niina A, Saitoh S, Suzuki Y, Honjo Y. Deep mixing method using
cement hardening agent. In: Proceedings of the 10th international conference
on soil mechanics and foundation engineering. Stockholm; 1981. p. 7214.
[8] Bruce DA, Bruce MEC, Dimillio AF. Dry mix methods: a brief overview of
international practice. In: Proceedings of international conference on dry mix
methods for deep soil stabilization. Rotterdam; 1999. p. 1525.