Consolidated Undrained Triaxial Compression Tests and Strength Criterion of Solidified Dredged Materials

Ph.D, Associate Professor, Institute of Geotechnical Engineering, School of Transportation, Southeast University, Nanjing 210096, China Ph.D, Jiangsu Water Resources Company Limited, +e Water Diversion Project from South to North in China, Nanjing 210029, China Ph.D, Lecturer, College of Transportation, Shandong University of Science and Technology, Qingdao 266590, China Ph.D Candidate, Institute of Geotechnical Engineering, School of Transportation, Southeast University, Nanjing 210096, China Ph.D, Jiangsu Water Resources Company Limited, +e Water Diversion Project from South to North in China, Nanjing 210029, China


Introduction
Large volumes of sediment are dredged annually to maintain the navigational depth of channels and harbors, to prevent rivers from flooding, and to restore the ecosystem of degenerative water bodies [1][2][3][4][5][6].How to deal with the vast amounts of dredged materials (DM) becomes an important issue in engineering practice.Many studies have been conducted regarding the use of DM as land reclamations backfill or engineering construction materials [2,5,[7][8][9].Solidification/stabilization is an ideal way that can improve the engineering properties of the abandoned DM [5] ( Kamon et al. 2005) [6,8,10], showing economical and environmental advantages and avoiding borrowing soils from elsewhere.e strength properties of structured clays are different from remolded clays [11][12][13][14][15].However, the existing strength criterions are mostly established on the basis of the properties of reconstituted soils [16][17][18][19] that cannot well reflect the influence of soil structures on the properties of structured soils [14].It has been well reported by researchers [15,20] that the strength envelope of the natural structured sedimentary clays can be divided into two parts in terms of the structure yield stress (i.e., preyield state and postyield state).And there also exists a transitional stress on the strength envelope of the cement-treated soil [21,22].e structure formation of natural soil is attributed to the soil structure development during depositional and postdepositional processes [23][24][25][26].Similar to natural structured soils, the observed high yield stresses are evidences for the structure of cement-treated soil [1,2,22,27,28].e structure of the SDM is formed by cement hydration and pozzolanic reaction, and a mechanically stable soil matrix was produced due to cementation bonds [1,6,21,28,29].Hence, the SDM can be considered as an artificially structured soil, and it can exhibit strongly stable structure and higher yield stress compared with the untreated DM.
e objectives of this study are (1) to investigate shear strength behavior of the SDM, (2) to establish strength criterion for the SDM, and (3) to verify the validity of the proposed strength criterion and discuss the relevant parameters.

Materials.
e dredged material (DM, as shown in Figure 1) used in this study was taken from the bottom of the Baima Lake located in Huaian, Jiangsu Province, China (sampling site as shown in Figure 2).e basic physical properties of the DM are summarized in Table 1.e liquid limit and plastic limit are 66.1% and 26.6%, respectively.According to the Unified Soil Classification System, the DM is classified as high plasticity clay.
Based on the conventional cement-treated method, in this study, phosphogypsum was used together with cement to stabilize the DM at high water content.It should be mentioned that phosphogypsum is a by-product of the production of phosphoric acid, only a small portion can be recycled, other most is deposited without any prior treatment.Deserted phosphogypsum occupies land areas and possibly causes serious environmental contamination [31][32][33].Type I ordinary Portland cement was used in this study, and phosphogypsum (as shown in Figure 3) used here was taken from a chemical company in Nanjing, China.

Sample Preparing and Test Method.
Mixing proportion design of the SDM is presented in Table 2, and current dredging method characterized with high water content of the DM [34] has been taken into account.In Table 2, the mixing proportion can be expressed as 2.5W+C100+P40 and so on.In the expression, 2.5W denotes the water content of DM slurry is 2.5 times the liquid limit, C100 denotes the cement content is 100 kg for 1 m 3 DM slurry (i.e., 100 kg/m 3 ), and P40 denotes the weight of phosphogypsum is 40% of the cement.All triaxial tests were performed at the curing period of 28 days, and the confining pressure is designed in the range of 100-1200 kPa, which are shown in Table 2.
e specimens were prepared by mixing the DM slurries with dry cement powder and phosphogypsum powder.
en, the uniform paste was placed into a cylindrical mold.e bottom of the mold is specially designed which can be dismantled freely so that the sample is easy to dismantle.Friction was minimized by using a polished inner wall and smeared with a thin film of grease.e cylindrical specimens were dismantled after 24 hours, wrapped with plastic bags, and then stored in a standard curing room with a constant ambient temperature of 20 ± 2 °C and a relative humidity above 95%.e dimension of all specimens (as shown in Figure 4) is 39.1 mm in diameter and 80 mm in height.
Consolidated undrained triaxial compression tests were conducted at the designed curing period of 28 days.Before installation in the triaxial cell, the specimens were placed in a vacuum chamber filled with distilled water for saturation at least 24 hours.After installation, undrained isotropic loading was used to check the pore pressure coefficient "B." e triaxial tests were conducted at an axial compression rate of 0.073 mm/min.Excess pore water pressure was measured during the shear phase.It should be mentioned that even adopted with a back pressure of 200 kPa after vacuum saturation, the measured pore pressure parameter "B" was only around 0.9, which is in agreement with the phenomenon reported by Kamruzzaman et al. [28].

Results and Discussion
e variation of deviator stress under different confining pressures is shown in Figure 5.It can be seen that there exists a peak deviator stress for all SDM samples, which is similar to the overconsolidated soils [28,[35][36][37][38][39]. Figure 5 illustrates that the peak deviator stress increases with the applied confining pressure and the failure strain also increases with the increase in confining pressure.It is interesting to note that the increment of the peak deviator stress is different under different stress levels.
e peak deviator stress increases very small when the confining pressure is less than 400 kPa, which is similar to the observations reported by other researchers [21,28].e change of the fabric in the consolidation process is insignificant at low confining pressures, and the peak strength is mainly governed by the soil structure resistance [15,21].However, the peak strength significantly increases when the confining pressure is more than 400 kPa, for the cementation bond begins to break and its artificial structure disappears progressively at high confining pressures, and the shear strength mainly depends on the applied stress level.Hence, the behavior of the SDM is similar to heavily over-consolidated natural structured clays, although its structure yield stress is related to bonding and not to stress history [1,22,28,40].e variation of excess pore water pressure under different confining stress is shown in Figure 6.It can be seen that there exists a peak value when the confining pressure is lower than 400 kPa; however, the peak value becomes insignificant when the confining pressure is greater than 400 kPa.Similar results were observed by other researchers on cement-treated or untreated clays [28,41,42].e change law of excess pore water pressure with the confining pressure is different with the deviator stress, the peak pore water pressure increases notably with an increase in confining pressure, both in the lower stress level stage and in the higher stress level stage.Similar results were reported by other researchers on cement-treated clays [28,41].
e relationship between undrained shear strength and applied normal stress is shown in Figure 7.It can be seen that the undrained shear strength envelope is approximately consisted of two straight lines, and there is a transitional 2 Advances in Civil Engineering stress at the intersection point of the two straight lines on the strength envelope.Such a bilinear strength envelope for natural sedimentary clays has been well reported by other researchers [15,20].
In Figure 7, it can be seen that the undrained shear strength increases slightly with the increase in applied normal stress at first (i.e., in the preyield state), which indicates that the shear strength of the SDM is mainly affected by the soil structure resistance when the applied stress is smaller than the yield stress.However, the shear strength increases significantly when the applied normal stress is greater than yield stress (i.e., in the postyield state).It is interesting to note that all the strength envelopes are nearly a straight line with the extension through the origin, which is consistent with the observation of other researchers [15,22].It indicates that the undrained shear strength mainly depends on the applied normal stress in the postyield stage, for the cementation bond is progressively destroyed and the soil structure resistance gradually vanishes when the stress level exceeds its yield stress [28,40].

Strength Criterion for the SDM
Based on the homogenization theory of heterogeneous materials [43] and considering the breakage mechanics of geomaterials [44], Shen [45] proposed the binary medium model for structured soil and suggested a basic equation for the mean stress tensor σ { } as Equation ( 1).In the model, the     Advances in Civil Engineering structured soil is conceptualized as binary medium consisting of bonding element and frictional element: where σ i   is the mean stress tensor of the bonding element and σ f   is the mean stress tensor of the frictional element, respectively, and b is the resistance share ratio.
In addition, according to the basic theory of soil mechanics, cohesion is independent of the normal stress, whereas internal friction is the proportional function of normal stress in the shear surface.Hence, the shear strength of the SDM can be expressed as where f(τ b ) is the shear strength provided by the bonding element and f(τ f ) is the shear strength provided by the frictional element.
Substituting Equation ( 1) into (2), the shear strength of solidified soil can be expressed as  4 Advances in Civil Engineering where τ b and τ f are the shear strengths provided by bonding element and frictional element, respectively, and β is the shear resistance share ratio.It has been well documented that atmospheric pressure can be used in the strength reference framework to investigate the strength of the solidified soil [14,46,47].e expression of shear strength provided by bonding element is assumed as follows: where q u is the unconfined compressive strength, σ bm � (σ 1b + σ 2b + σ 3b )/3 is the mean stress of bonding element, P a is the standard atmospheric pressure, and n is a undetermined constant.
where φ is the internal friction angle of frictional elements and σ f � (σ 1f + σ 3f )/3 is the average of the minor principal stress and the major principal stress.
According to the triaxial tests data, the evolution law of shear strength is significantly different between the preyield state and the postyield state.e parameter β used here is introduced from the expression suggested by Liu and Shen [14], as shown in Equation ( 6): where σ vy is the structure yield stress which can be obtained from oedometer test and also can be indirectly derived from unconfined compression test according to the empirical relationship suggested by some researchers [27,48,49], σ 3 is the minor principal stress, and m is an undetermined constant.

Parameters Discussion and Strength Criterion Verification
5.1.Parameters Discussion.e influence of parameters m and n is discussed as follows.First, the influence of m is discussed with the assumption that n is a constant of 0.2 referenced the value suggested by Liu and Shen [14].Other parameters are listed in Table 3. e influence of the parameter m to evolution laws of shear strength is shown in Figure 8.It can be seen that when m � 0.1, shear resistance is nearly one straight line and similar with the conventional Mohr-Coulomb criterion.In addition, it is interesting to note that when m is ≥3, not only all curves are very close but also the extensions of strength envelope are nearly through the origin in the postyield state.By comparison, the shape of curve is closer to the experimental curve when m � 3. Hence, m � 3 is suggested in this paper.
Subsequently, the value of n was discussed with the assumption that m is a constant of 3, and the other parameters are shown in Table 3.
e influence of the parameter n to evolution laws of shear strength is shown in Figure 9.It can be seen that the parameter n has more significant influence when the stress level is lower; however, when the stress level is higher, the parameter n almost has no effect on shear resistance and all curves nearly concentrate to one straight line, what is more, with the extension of the strength envelope through the origin in the postyield state.Figure 9 indicates that the cohesion of bonding element gradually vanishes, and the shear resistance is mainly provided by the frictional element after structure yield.Hence, the proposed strength criterion can well describe the evolution laws of shear strength for the SDM.

Strength Criterion Verification.
In order to further verify the validity of the proposed shear strength criterion,  6 Advances in Civil Engineering comparisons between the simulation results obtained from the proposed strength criterion and the experimental data from triaxial tests are plotted in Figure 10.e parameters are determined as follows: m � 3, P a � 101.4 kPa, σ vy is determined according to one-dimensional compression test [27], q u is determined according to unconfined compression test [50], φ is determined according to triaxial test in the postyield state, and n is an undetermined constant.e values of main parameters are summarized in Table 4. Figure 10 presents comparisons of shear strength between simulation results and experimental data.e curve is mainly composed of two segments, the former is a slightly upward curve and the latter is nearly one straight line.e suggested range of n is 0.1-0.3.And it can be seen that the extension of strength envelope is nearly through the origin when stress level is higher.e comparisons between predicted results and experimental data demonstrate that the proposed strength criterion is valid, and it can well describe the mechanical features and shear strength evolution laws.
Figure 11 presents the evolution laws of percentage of cohesion and internal friction with the increase in the normal stress for the mixing proportion of 2.5W+C100+ P40, and the others are same.It can be seen that the total shear resistance is all provided by the bonding element at first, and the percentage of cohesion decreases gradually with the increase in the stress level.However, the Advances in Civil Engineering percentage of the internal friction increases with the increase in the stress level.Finally, with the bonding elements broken completely, the total shear resistance is fully provided by the frictional elements.Figure 11 indicates that the internal friction increases gradually with the increase in the stress level because of the bonding elements progressively breaking down and turning into frictional elements in essence.

Conclusions
e main conclusions obtained in this study are summarized as follows: (1) e peak deviator stress increases small when the confining pressure is low, for the change of the fabric in the consolidation process is insignificant at low confining pressures.(2) e undrained shear strength envelope is approximately consisted of two straight lines.e strength increases slightly with the increase in normal stress in preyield state; however, it increases significantly in postyield state with the extension of the strength envelope nearly through the origin.
(3) In preyield state, the shear strength of the SDM is mainly affected by cementation bonding effect and governed by structure resistance; however, in postyield state, it is mainly attributed to internal friction and controlled by the applied normal stress.(4) A strength criterion considering strength evolution mechanism is proposed for the SDM.Comparisons of the predicted results and experimental data demonstrate that the proposed strength criterion can properly describe the mechanical features and strength evolution laws.

e
shear strength provided by frictional element follows the Mohr-Coulomb criterion.It is the proportional function of normal stress and can be expressed as

Figure 8 :Figure 9 :
Figure 8: Influence of parameter m to evolution laws of shear strength.

Figure 11 :
Figure 11: Percentage of cohesion and internal friction versus normal stress.

Table 1 :
Basic physical properties of the dredged material.

Table 2 :
Programme of triaxial test.

Table 3 :
Values of other parameters for "m" discussion.

Table 4 :
Values of parameters for strength criterion veri cation.