- Email: [email protected]
Combination of vacuum preloading and lime treatment for improvement of dredged fill
ENGEO-04493; No of Pages 10 Engineering Geology xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Engineering Geology journal homepage: www.elsevier.com/locate/enggeo
Combination of vacuum preloading and lime treatment for improvement of dredged fill Jun Wang a,c,d, Junfeng Ni a,c,d, Yuanqiang Cai b,c,d,⁎, Hongtao Fu a,c,d, Peng Wang a,c,d a
College of Architecture and Civil Engineering, Wenzhou University, Chashan University Town, Wenzhou 325035, China College of Civil Engineering and Architecture, Zhejiang University of Technology, Hangzhou 310014, China c The Key Laboratory of Engineering and Technology for Soft Soil Foundation and Tideland Reclamation of Zhejiang Province, China d Innovation Centre of Tideland Reclamation and Ecological Protection, Wenzhou University, Wenzhou, Zhejiang 325035, China b
a r t i c l e
i n f o
Article history: Received 3 May 2016 Received in revised form 31 January 2017 Accepted 12 February 2017 Available online xxxx Keywords: Vacuum preloading Lime treatment Flocculation Optimum lime concentration Hydrated lime
a b s t r a c t Numerous land reclamation projects that utilise dredged fill are currently being executed in the eastern regions of China to cater to the demands of economic development and address the land shortage problem. However, the dredged fill has high water content, low shear strength, and mainly consists of fine-grained particles. A long period is thus normally required for the treatment of dredge fill by vacuum preloading, and clogging always occurs around the prefabricated vertical drains (PVDs), resulting in insufficient consolidation of the soil. To address this issue, combination of vacuum preloading and lime treatment is proposed in this paper. In this method, a certain percentage of hydrated lime (Ca(OH)2) is added into the dredge fill slurry before vacuum preloading, to enhance the engineering properties of the fill, such as the shear strength and permeability. The added hydrated lime would induce cation exchange on the clay surface and flocculation of the fine soil particles. As a result, this method would significantly mitigate the risk of clogging around the PVDs with the enhanced soil permeability, thereby improving the consolidation efficiency. In this study, dredged slurry with a water content of approximately 187% was utilised. Lime modification optimum (LMO) was first determined using the vacuum preloading method, and a comparison test was then conducted to verify the effectiveness of the proposed method. It was found that, compared to the conventional vacuum preloading method, the proposed method significantly increased the vane shear strength of all the soil layers, especially the deep soil layers, and afforded a higher consolidation rate. © 2017 Published by Elsevier B.V.
1. Introduction The Oufei land reclamation project is currently being executed in eastern Wenzhou, China to cater to the demands of urban expansion and infrastructural development. The project utilises slurry dredged from the coastal seabed as the landfill material (Fig. 1). However, the engineering properties of such dredged fills are too poor to support the construction of structures. Even after years of self-weight consolidation, the fills still have a high water content, low bearing capacity, and high compressibility (Chu et al., 2000). Therefore, it is essential to stabilise the soft clayey soil before any construction work is conducted. One of the methods most widely used for the improvement of soft clay ground is vacuum preloading. The method, which was first proposed by Kjellman (1952), is convenient and economical, and is increasingly being used to improve the mechanical performance of dredged fills in numerous projects (Chai et al., 2010, 2006; Indraratna et al., 2011; Shang et al., 1998; Tang and Shang, 2000; Yan and Chu, 2005). Prefabricated vertical drains (PVDs) are used in the vacuum preloading ⁎ Corresponding author. College of Civil Engineering and Architecture, Zhejiang University of Technology, Hangzhou 310014, China. E-mail address: [email protected] (Y. Cai).
process to distribute the vacuum pressure and discharge the pore water (Indraratna et al., 2005; Yan and Chu, 2003). However, due to the dredged slurry mainly consists of fine-grained soils, these particles would be moved as the water drainage during the vacuuming and eventually aggregate around the PVDs (Wang et al., 2016b). This leads to the formation of an obvious soil column and clogging around the PVDs, ultimately resulting in insufficient consolidation and limited shear strength improvement (Wang et al., 2016a). There is thus the need for the enhancement of vacuum consolidation. This paper proposes a method for improving the effectiveness of vacuum consolidation. The proposed method involves the addition of a certain percentage of hydrated lime to the dredged slurry before vacuum preloading. When a hydrated lime is added to a soil, a modification reaction occurs immediately as a result of the substitution of calcium with the existing cations at negative charge sites on the clay mineral surface. Consequently, the interparticle repulsive forces between clay particles are decreased, which induced soil particles aggregate and flocculation (Le Runigo et al., 2009; Salehi and Sivakugan, 2009). The change of soil particles from the initial dispersed orientation state to flocculation increases the void ratio, thereby enhancing the permeability of the soil (Brandl, 1981; Rajasekaran and Narasimha Rao, 2002). These behaviours are beneficial to water flow and drainage of the soil.
http://dx.doi.org/10.1016/j.enggeo.2017.02.013 0013-7952/© 2017 Published by Elsevier B.V.
Please cite this article as: Wang, J., et al., Combination of vacuum preloading and lime treatment for improvement of dredged fill, Eng. Geol. (2017), http://dx.doi.org/10.1016/j.enggeo.2017.02.013
2
J. Wang et al. / Engineering Geology xxx (2017) xxx–xxx
and plastic limit (WP) were approximately 81% and 42%, respectively, while the specific gravity was determined to be 2.61. 3. Determination of LMO
Fig. 1. Layout of the tideland reclamation site in Lingkun (Longwan District), Wenzhou, China.
This approach can also reduce the occurrence of clogging around the PVDs by retarding the formation of soil columns, thereby improving the efficiency of vacuum consolidation. However, the permeability does not indefinitely increase with increasing lime content, but begins to slowly decrease beyond a threshold value. This decrease is due to the flow channels being blocked by the precipitation of cementitious compounds generated by the lime-soil reactions, resulting in reduced water flow (Le Runigo et al., 2009). Hence, it is imperative to determine the threshold for subsequent vacuum preloading. This threshold was also known as the lime modification optimum (LMO), which was first introduced by McCallister and Petry (1992) through leach tests conducted on lime-treated clay. They reported that the maximum permeability occurs at the LMO. Eades and Grim (1966) first developed the pH method for determining the LMO, and Rogers and Glendinning (1997) subsequently improved the method by considering the change in plasticity. Salehi and Sivakugan (2009) also determined the LMO for dredged mud based on the permeability and void ratio. However, these methods only consider the changes in the permeability or pH without directly assessing the efficiency of the vacuum preloading process. In the present study, a determination test was firstly conducted to identify the LMO by vacuum preloading with the purpose of quantifying the changes produced by the use of hydrated lime. Shear strength, pore water pressure, discharged water volume, and water content are regarded as the indicators for the determination of LMO by the vacuum preloading method. A comparison test was also performed to verify the effectiveness of the proposed method. Briefly, the objectives of this study were as follows: 1. To develop a method for determining the optimum lime content for enhancing the dredged slurry in Wenzhou. 2. To investigate the consolidation behaviour of natural and lime-treated dredged slurry during vacuum preloading. 3. To evaluate the efficiency and feasibility of the proposed method. 2. Properties of dredged slurry The clay slurry used for the tests was obtained from the site of the Oufei project in Wenzhou, China. The geological setting and depositional environment of the slurry are shown in Fig. 2. The sample slurry had just dredged from the seabed and had a water content of 187%. It mainly consisted of fine clay particles and had a pH of 6.8. The liquid limit (WL)
The void ratio and permeability of the soil are generally increased by the lime-clay reaction (Rajasekaran and Narasimha Rao, 2002). This means that the chemical reaction generates more pore channels, which enhance water discharge from the soil during subsequent vacuum consolidation. However, as implied earlier, the pore channels do not indefinitely increase with increasing lime content, but there exists an optimum lime content that maximises the number of pore channels. Below this optimum content, only flocculation occurs, while pozzolanic reactions are also triggered above it (McCallister and Petry, 1992). To determine this optimum lime concentration, the following laboratory test was developed and conducted. Slurry samples of equal mass (34 kg) were poured into five model buckets of height 40 cm and diameter 30 cm. Different amounts of lime representing 0%, 1%, 1.5%, 2%, and 2.5% of the dry soil by weight were respectively added to the slurry samples. Each mixture of lime and slurry was then thoroughly stirred by a slurry agitator until homogeneity was achieved, to ensure complete reaction between the lime and slurry (the first ‘mixture’ containing 0% lime was the natural slurry sample). PVDs with cap connections were subsequently installed horizontally in all the samples. A pore pressure sensor was placed at the bottom of each model bucket, which was filled with a sample to a height of 35 cm. A layer of geotextile and two layers of geomembranes were placed on top of the slurry in each bucket to seal the sample from the atmosphere. The cap, air–water separation flask, and vacuum pump were then successively connected by vacuum pipes. The complete test setup is shown in Fig. 3. During the vacuum consolidation, a minimum vacuum load of 80 kPa was applied to the surface of the soil and maintained until there was no further apparent increment in the measured volume of discharged water in the air-water separation flask. Once the steady state was attained, the vacuum pump was switched off, and the vane shear strength and water content of the consolidated soil were measured. The variations of the discharged water volume with time of all the samples are depicted in Fig. 4. It is obvious that, for all the samples, the discharged water volume increases sharply at the beginning of preloading. A stable value is attained for all the lime-treated slurry samples within 80 h, whereas the untreated sample continues to discharge water for an additional 88 h. This suggests enhanced drainage capability of the lime-treated slurries, especially that containing 2% lime, which stops draining within 48 h. Moreover, as the lime content increases from 1% to 2%, there is an increase in the volume of discharged water to the highest value among the treated samples. This can be explained by the permeability, which undoubtedly affects the draining during vacuum preloading; a higher hydraulic conductivity accelerates draining. When the lime is added to the slurry, cation exchange and flocculation immediately begin, with the permeability increased by the flocculation and resultant soil aggregation (Nalbantoglu and Tuncer, 2001; Rajasekaran and Narasimha Rao, 2002). The increased permeability in the present tests can be expressed in terms of not only the volume of discharged water, but also the rate of drainage. However, as noted, the permeability only increases up to a maximum value afforded by an optimal lime content, the LMO, and then begins to decrease due to the formation of cementitious compounds among the soil particles (Alhassan, 2008; Milburn and Parsons, 2004; Onitsuka et al., 2001). The cementitious compounds block the flow channels, reducing the water flow (Quang and Chai, 2015) and locking part of the water within the soil. This is evident from Fig. 4, which shows that 2.5% lime content produces the lowest volume of discharged water despite the accelerated drainage. It is very noteworthy that the natural slurry produces the highest volume of
Please cite this article as: Wang, J., et al., Combination of vacuum preloading and lime treatment for improvement of dredged fill, Eng. Geol. (2017), http://dx.doi.org/10.1016/j.enggeo.2017.02.013
J. Wang et al. / Engineering Geology xxx (2017) xxx–xxx
3
Fig. 2. Geological section of the Oufei project site.
discharged water. According to Mallela et al. (2004), lime-treated clays exhibit enhanced capability to hold a large amount of water. This is because some of the water is attracted by the modified soil particles and difficult to discharge despite the improved permeability. Fig. 5 shows the dissipation of the pore water pressure monitored at 30 cm below the surface of each sample during the vacuum consolidation. It is obvious that the dissipation of the pore water pressure in the lime-treated samples is almost completed within 80 h, whereas 150 h is required for the case of the natural slurry. This suggests that the addition of the lime accelerates the dissipation of the pore water pressure. Similar to the water drainage, the reduction of the pore water pressure increases as the lime content increases from 1% to 2%, after which it decreases for further increase of the lime content to 2.5%. The reason for this is also similar to that for the trends of the water drainage (Fig. 4).
This is predictable considering that drainage of water generally leads to dissipation of the pore water pressure according to the Terzaghi theory (Terzaghi and Peck, 1967). The lime treatment of slurry thus not only increases the reduction of the pore water pressure, but also accelerates the reduction. The water contents of all the samples were measured after the vacuum preloading. Fig. 6 shows the determined variation of the final water content with the lime content with respect to the depth in the samples. As can be seen, the final water content increases with increasing lime content from 0% to 2.5% (dry weight basis), without peaking. The cementitious compounds produced by the lime-slurry reactions are calcium silicate hydrate (CSH) and calcium aluminate hydrate (CAH), which are composed of the solid products of hydration and water (Le Runigo et al., 2009; Rajasekaran et al., 1997). The reactions
Please cite this article as: Wang, J., et al., Combination of vacuum preloading and lime treatment for improvement of dredged fill, Eng. Geol. (2017), http://dx.doi.org/10.1016/j.enggeo.2017.02.013
4
J. Wang et al. / Engineering Geology xxx (2017) xxx–xxx
Fig. 3. Schematic diagram of the test setup.
The above equations indicate that more water-containing agents are generated in the pore with increasing lime content. This is the reason why, as shown in Fig. 6, the water content increases with increasing lime content to 2.5%, despite the above-mentioned increase in the water discharge only with increasing lime content from 1% to 2%. Vane shear tests were also conducted at different depths of the samples after the vacuum preloading. The determined variation of the vane shear strength with the lime concentration with respect to the depth is shown in Fig. 7. As indicated, the vane shear strength gradually increases with increasing lime content from 0 to 2% (dry weight basis), at which it reaches the maximum value of 26 kPa. Thereafter, the vane shear strength decreases with further increase of the lime content to 2.5%, at which the value is approximately the same as that for 1.5% lime. There are two primary causes of the increase in the shear strength. First, flocculation is immediately induced by the addition of the lime to the slurry, resulting in changes in the soil fabric through rearrangement of the soil particles. The new fabric is relatively open, which is beneficial for the discharge of pore water. As the water discharged by PVDs, soil particles aggregated under the vacuum preloading, shear strength was enhanced. This is the main cause of the increase in shear strength. The other cause is the CSH and CAH generated during the long curing
period. As a type of gel, these compounds cement the flocculated soil particles, resulting in the formation of a stronger fabric with a higher shear strength (Dash and Hussain, 2012). In addition, the shear strength is affected by the arrangement of the particles within the soil, which indicates that a better fabric of the lime-treated soil essentially enhances the interparticle shear resistance. However, the vane strength does not increase indefinitely with the lime content, but begins to decrease when the amount of lime becomes excessive, as shown in Fig. 7. Considering that the initial increase in shear strength during the vacuum consolidation is mainly due to drainage of the slurry, the decrease with excessive addition of lime may be attributed to the resultant blockage of the water flow. As it mentioned in Fig. 4, the discharged water did not continuously increase with increasing lime content because of the pozzolanic reactions induced decrease in the permeability. Similarly, the decrease of shear strength is also due to the triggering of pozzolanic reactions between the lime and slurry. When sufficient lime is added to the slurry, the precipitation of the cementitious compounds blocks the flow channels and reduces the water flow, resulting in the retention of a relatively high water content (Milburn and Parsons, 2004; Onitsuka et al., 2001). As the water is trapped in the slurry, the efficiency of the vacuum consolidation is limited and the bonds among the soil particles are weakened, resulting in lower vane shear strength. The above comprehensive investigation of the effect of the amount of lime added to dredged slurry indicates that 2% lime content (dry weight basis) produces the best drainage and shear strength. This amount is thus considered the optimum lime concentration for vacuum preloading.
Fig. 4. Discharged water volume vs. time curve of the slurry with respect to the lime content.
Fig. 5. Variation of the pore water pressure at a depth of 30 cm in the slurry with respect to the lime content.
are as follows (Jawad et al., 2014; Mallela et al., 2004; Yong and Ouhadi, 2007): CaðOHÞ2 þ SiO2 →CaO−SiO2 −H2 O
ð1Þ
CaðOHÞ2 þ Al2 O3 →CaO−Al2 O3 −H2 O
ð2Þ
Please cite this article as: Wang, J., et al., Combination of vacuum preloading and lime treatment for improvement of dredged fill, Eng. Geol. (2017), http://dx.doi.org/10.1016/j.enggeo.2017.02.013
J. Wang et al. / Engineering Geology xxx (2017) xxx–xxx
5
4. Comparison test Because the model buckets used in the above tests were too small for the simulation of the scenario of a real land reclamation project, and the acquired data was limited, a comparison test was conducted to verify the effectiveness of the proposed method for improving the effectiveness of vacuum consolidation. For simplicity, only two slurry samples (without lime and with 2% lime) were used for the test. It was expected that the exact increment in the vane shear strength and the reduction in period of consolidation would be measured after the vacuum preloading. 4.1. Experimental apparatus
Fig. 6. Variation of the water content with the lime content of the slurry with respect to the depth.
The apparatus used for the comparison test mainly consisted of model boxes (height 1 m and width and length 2 m), vacuum pipes, a vacuum pump, and prefabricated vertical drains (PVDs). A special slurry agitator convenient for mixing a large amount of slurry with lime and the accompanying mixing method was developed for the test. In this method, lime was previously mixed with water. As shown in Fig. 8, the slurry agitator had two inlets in its upper part for the respective inflow of the limewater and slurry, and one outlet underneath for the outflow of the lime-slurry mixture. The homogeneous lime-treated slurry was attained by pumping lime and slurry into the agitator at the same time and thoroughly stirred. 4.2. Sample preparation and test procedure
Fig. 7. Variation of the vane shear strength with the lime content of the slurry with respect to the depth.
Two different types of slurries were actually used for the test. One was obtained from the site of the Oufei project. This slurry had an original water content of 187%, but had undergone a half-year of self-weight consolidation, which reduced the water content to 110%. It was used to simulate a practical project, wherein vacuum preloading is generally implemented after the slurry has undergone self-weight consolidation. However, considering the difficulty of mixing lime with hardened slurry in practice, the lime treatment should be performed at the time of pumping the slurry unto the site by injection of the lime into the slurry pipeline. Therefore, to simulate this practical engineering requirement, slurry with the original water content of 187% was obtained from the same site and thoroughly mixed with lime using the developed mixing method. Furthermore, to prevent the lime from causing long-term
Fig. 8. Schematic of the employed slurry agitator and illustration of the lime-slurry mixing method.
Please cite this article as: Wang, J., et al., Combination of vacuum preloading and lime treatment for improvement of dredged fill, Eng. Geol. (2017), http://dx.doi.org/10.1016/j.enggeo.2017.02.013
6
J. Wang et al. / Engineering Geology xxx (2017) xxx–xxx
Fig. 9. (a) Monitoring points of the vacuum pressure and pore water pressure during the vacuum preloading, and (b) plan view of the vacuum system.
alteration of the slurry and to take full advantage of the immediate limeslurry reactions such as flocculation (Jha and Sivapullaiah, 2015; Ural, 2015), vacuum preloading was immediately implemented on the lime-treated sample. The characteristics of the two slurries were the same as those used for the previous tests, except that the first slurry of the present test had a lower water content and void ratio, naturally acquired through its self-weight consolidation. However, considering that the ultimate strength is generally used as the main indicator to estimate the efficiency of vacuum preloading, the difference between the initial water contents of the two slurries of the present test only marginally affected the results. The samples were prepared as follows. The slurry with a water content of 187% was first poured into a model box and then mixed with limewater using the slurry agitator shown in Fig. 8. The lime-treated slurry was then pumped into another model box up to a height of 90 cm. For the comparison, the natural slurry with a water content of
110% was poured into another model box up to a height of 62 cm. The dry weights of the two slurry soils were the same. After the above preparation of the samples, the vacuum system was connected as described below. The vacuum pipes with needles were first inserted into the PVDs and connected to the vacuum pressure gauge. Two types of PVDs of different sizes (Fig. 9), namely mini PVD and PVD with cap (CPVD), were used to fix the vacuum pipes with needles and transfer the vacuum pressures in the PVDs and soil. All of the four employed CPVDs was connected to a vacuum pump and vertical drain through vacuum tubes, while the single mini PVD was only used to hold the vacuum pipe needles. This method reduced clogging in the vacuum pipes effectively and therefore facilitated measurement of the vacuum pressure in the PVDs and soil. The four CPVDs and one mini PVD, together with the pore water pressure sensors, were installed on an iron rack, to prevent the PVDs from being bent and to limit the movements of the PVDs and sensors during
Please cite this article as: Wang, J., et al., Combination of vacuum preloading and lime treatment for improvement of dredged fill, Eng. Geol. (2017), http://dx.doi.org/10.1016/j.enggeo.2017.02.013
J. Wang et al. / Engineering Geology xxx (2017) xxx–xxx
7
the vacuum preloading. Fig. 12. Pore water pressure vs. time curves of the slurries with respect to the depth 35 cmTaway from the PVDs. he iron rack was laid on the bottom of the model box and one layer of geotextile and two layers of geomembranes were used to seal the surface of the slurry. The same test setup was used for the two considered slurries. The distribution of the monitoring points of the vacuum pressure and pore water pressure during the vacuum preloading is shown in Fig. 9. The application of the vacuum pressure and consolidation of the slurry commenced when the vacuum pump was switched on.
4.3. Results and discussion
Fig. 10. Vacuum pressure vs. time curves with respect to the depth in the PVDs: (a) natural slurry, (b) slurry containing 2% lime.
Fig. 11. Vacuum pressure vs. time curves of the slurries with respect to the depth 30 cm away from the PVDs.
4.3.1. Vacuum pressure Fig. 10a and b respectively show the vacuum pressures in the natural and lime-treated slurries at different depths within the PVDs during the preloading. It can be clearly seen from the figures that a vacuum pressure of above 80 kPa was maintained at all depths in both slurries throughout the durations of the tests, with the pressure decreasing with increasing depth. Moreover, both slurries have almost the same vacuum pressure gradient in the vertical direction of approximately 10 kPa/m. This indicates that it is impossible to maintain a uniform vacuum pressure along the depth of the consolidated soil, as has been previously observed (Chai et al., 2005; Indraratna et al., 2005; Shang et al., 1998). There is, however, significant difference between the vacuum pressures in the two slurries outside the PVDs. As shown in Fig. 11, the vacuum pressure in the slurry containing 2% lime reaches as much as 75 kPa, which is much higher than that in the natural slurry. In addition, the lime-treated slurry requires a shorter time to attain a steady vacuum pressure. It is generally accepted that the addition of lime to soil increases the hydraulic conductivity with a more open the soil fabric (Nalbantoglu and Tuncer, 2001; Rajasekaran and Narasimha Rao, 2002). Moreover, Chai et al. (2010) found that the final vacuum pressure distribution was primarily determined by the hydraulic conductivity, with the other parameters having no considerable effect. The present results also show that the vacuum pressure outside the PVDs in the lime-treated slurry begins to increase as soon as the vacuuming begins, whereas 70 h is required for any change in vacuum pressure to occur in the natural slurry. Further, the addition of lime to soil has been observed to cause partial saturation (Zhang et al., 2015), which results in the immediate transfer of vacuum pressure within the soil. However, the presence of an unsaturated zone during preloading (Qiu et al., 2007) enables the loosened fabric to accelerate the drainage of water and facilitate the distribution of pressure in the soil. This accelerates the increase of the vacuum pressure in the soil and increases the maximum pressure attained, as shown in Fig. 11.
Please cite this article as: Wang, J., et al., Combination of vacuum preloading and lime treatment for improvement of dredged fill, Eng. Geol. (2017), http://dx.doi.org/10.1016/j.enggeo.2017.02.013
8
J. Wang et al. / Engineering Geology xxx (2017) xxx–xxx
Fig. 13. Plan view of the settlement and vane shear strength measurement points.
4.3.2. Pore water pressure Fig. 12 shows the variations of the pore water pressure at different depths in the two samples during the vacuum preloading. As can be seen, the pore water pressure in the lime-treated slurry reduces by nearly 80 kPa over 154 h and then becomes steady, whereas there is a continuous slow decrease over 478 h in the natural slurry. The limetreated slurry not only has a higher pore water pressure dissipation rate, but also a lower final value. The reasons for this are the same as those adduced in the discussion of the LMO determination tests. Moreover, the present observations agree with previous findings, that a higher vacuum pressure increases the drain rate and accelerates the dissipation of the pore water pressure (Indraratna, 2010). The present results further confirm the effectiveness of lime treatment.
4.3.3. Settlement and degree of consolidation The settlement data at five locations in each of the model boxes (PA, PB, PC, PD, and PE in Fig. 13) were recorded during the vacuum preloading. The obtained curves of the settlement in the two slurries are shown in Fig. 14. The settlement in the lime-treated slurry is clearly greater, and the rate of consolidation under vacuum preloading is higher compared to the natural slurry. In addition, the settlement in the lime-treated slurry is almost completed within 270 h, whereas that in the natural slurry requires about 400 h. This shows that lime treatment reduces the consolidation time by about a third part. In addition, as can be observed from Fig. 14, lime-treated soil reveals a more uniform settlement, which indicates that the clogging and soil columns may less than untreated one. However, considering that the initial volumes of the two slurries were different, the observed relative settlements cannot be accurately used to assess the effectiveness of lime treatment for soil improvement. The degree of consolidation (DOC), which is based on the settlement and pore water pressure, was used as a better measure. For this purpose, the method Asaoka was used to determine the ultimate settlements of both slurry samples (Asaoka, 1978). In this method, the settlement curve of a sample is obtained from the settlement data for a given period, and the ultimate settlement is determined as the intersection point of the settlement curve with the 45° line (Chu and Yan, 2005). Based on the ultimate settlements obtained in this study, the DOCs of the natural
and lime-treated slurries were determined to be 92% and 94%, respectively. Alternatively, the DOC at a given elevation can be estimated as (1 − Δuf/Pvac) × 100%, where Δuf is the pore water pressure at the completion of vacuum preloading, measured as the difference between the final pore water pressure and the suction line pressure (Chu et al., 2000); and Pvac is the corresponding vacuum pressure in the soil. The DOCs of the natural and lime-treated slurries estimated by this method using data obtained from Fig. 11 are 51% and 87% respectively. This shows that the DOC estimated from the pore water pressure is lower than that estimated from the settlement. The use of both methods is recommended (Chu and Yan, 2005). However, the DOCs determined from the pore water pressure in the present study are considered more accurate, because the settlement in the natural slurry under vacuum preloading was not uniform, which may cause a distinct DOC using settlement from different points. 4.3.4. Vane shear strength After the vacuum preloading, vane shear strength tests were conducted on the slurry samples at different points in the model box (A–E in Fig. 13). The results are presented in Fig. 15, which reveals a significantly higher overall vane shear strength of the lime-treated slurry, including at the deep layers, with the surface value being as high as 28.8 kPa. In addition, there is a lower rate of reduction of the vane shear strength of the treated sample with increasing depth. The improvement of the vane shear strength of the natural slurry was basically limited to the surface. These observations are due to the faster and higher vacuum pressure distribution to the deeper layers of the limetreated slurry. They confirm the effectiveness of lime treatment for comprehensively improving the shear strength of dredged slurry by vacuum preloading. 5. Conclusions The combined use of lime treatment and vacuum preloading was proposed for improving the consolidation of dredged slurry, and a series of laboratory tests were used to investigate the effectiveness of the method. Following is a summary of the findings and conclusions drawn from the study:
Please cite this article as: Wang, J., et al., Combination of vacuum preloading and lime treatment for improvement of dredged fill, Eng. Geol. (2017), http://dx.doi.org/10.1016/j.enggeo.2017.02.013
J. Wang et al. / Engineering Geology xxx (2017) xxx–xxx
9
2. Comparison tests using model boxes confirmed the effectiveness of the proposed method. The lime treatment was found to substantially improve the efficiency of the vacuum preloading, resulting in the achievement of a higher vacuum pressure within the soil and higher rate of consolidation. 3. The addition of the lime to the slurry reduced clogging around the PVDs and retarded the formation of soil columns during the vacuum preloading. 4. The most remarkable feature of the proposed method is that the material is prepared in slurry form through the rapid and convenient mixing of the pure slurry with lime. A specially developed apparatus was used for the present study. The foregoing discussion highlights the beneficial effect of lime on increasing the rate of vacuum consolidation and the values of vane shear strength of dredged fill. These laboratory tests prove the effectiveness and operational feasibility of the proposed method with lime treatment before the vacuum preloading. The advantages of this method may be used to develop an innovative ground improvement technique in land reclamation work using the dredged slurry as the fill material. Funding This work was supported by National Key Research and Development Program of China (Grant No. 2016YFC0800203), the Program of International Science and Technology Cooperation (Grant No. 2015DFA71550), the National Natural Science Foundation of China (Grant No. 51620105008, No. 51622810, No. 51408440 and No. 51478365), the Natural Science Foundation of Zhejiang Province Project (Grant No. LY17E080010), the Project of Science and Technology of Wenzhou (Grant No. S20150015). Acknowledgments The authors thank the Key Laboratory of Engineering and Technology for Soft Soil Foundation and Tideland Reclamation, and the Innovation Centre of Tideland Reclamation and Ecological Protection for the provision of the apparatus used for this study. Fig. 14. Surface settlement vs. time curves of the slurries: (a) natural slurry, (b) slurry containing 2% lime.
1. The optimum lime content for maximising the water discharge rate and vane shear strength of the dredged slurry utilised in the ongoing Wenzhou landfill project is 2% (dry mass basis).
Fig. 15. Variations of the shear strengths of the slurries with depth.
References Alhassan, M., 2008. Permeability of lateritic soil treated with lime and rice husk ash. Assumption Univ. J. Thailand 12, 115–120. Asaoka, A., 1978. Observational procedure of settlement prediction. Soils Found. 18, 87–101. Brandl, H., 1981. Alteration of soil parameters by stabilization with lime. Proceedings of the International Conference on Soil Mechanics and Foundation Engineering. 3, pp. 587–594. Chai, J.C., Carter, J.P., Hayashi, S., 2005. Ground deformation induced by vacuum consolidation. J. Geotech. Geoenviron. 131, 1552–1561. Chai, J.C., Carter, J.P., Hayashi, S., 2006. Vacuum consolidation and its combination with embankment loading. Can. Geotech. J. 43, 985–996. Chai, J., Hong, Z., Shen, S., 2010. Vacuum-drain consolidation induced pressure distribution and ground deformation. Geotext. Geomembr. 28, 525–535. Chu, J., Yan, S.W., 2005. Estimation of degree of consolidation for vacuum preloading projects. Int. J. Geomech. 5, 158–165. Chu, J., Yan, S.W., Yang, H., 2000. Soil improvement by the vacuum preloading method for an oil storage station. Geotechnique 50, 625–632. Dash, S.K., Hussain, M., 2012. Lime stabilization of soils: reappraisal. J. Mater. Civ. Eng. 24, 707–714. Eades, J.L., Grim, R.E., 1966. A quick test to determine lime requirements for lime stabilization: in behaviour characteristics of lime-soil mixtures. Highw. Res. Rec. 139, 61–72. Indraratna, B., 2010. Recent advances in the application of vertical drains and vacuum preloading in soft soil stabilisation. Aust. Geomech. J. 45, 1–44. Indraratna, B., Sathananthan, I., Rujikiatkamjorn, C., Balasubramaniam, A.S., 2005. Analytical and numerical modeling of soft soil stabilized by prefabricated vertical drains incorporating vacuum preloading. Int. J. Geomech. 5, 114–124. Indraratna, B., Rujikiatkamjorn, C., Ameratunga, J., Boyle, P., 2011. Performance and prediction of vacuum combined surcharge consolidation at port of Brisbane. J. Geotech. Geoenviron. 137, 1009–1018. Jawad, I.T., Taha, M.R., Majeed, Z.H., Khan, T.A., 2014. Soil stabilization using lime: advantages, disadvantages and proposing a potential alternative. Res. J. Appl. Sci. Eng. Technol. 8, 510–520.
Please cite this article as: Wang, J., et al., Combination of vacuum preloading and lime treatment for improvement of dredged fill, Eng. Geol. (2017), http://dx.doi.org/10.1016/j.enggeo.2017.02.013
10
J. Wang et al. / Engineering Geology xxx (2017) xxx–xxx
Jha, A.K., Sivapullaiah, P.V., 2015. Mechanism of improvement in the strength and volume change behavior of lime stabilized soil. Eng. Geol. 198, 53–64. Kjellman, W., 1952. Consolidation of clay soil by means of atmospheric pressure. Proceedings of a Conference on Soil Stabilization, pp. 258–263. Le Runigo, B., Cuisinier, O., Cui, Y.J., Ferber, V., Deneele, D., 2009. Impact of initial state on the fabric and permeability of a lime-treated silt under long-term leaching. Can. Geotech. J. 46, 1243–1257. Mallela, J., Von Quintus, P.H., Smith, K.L., Consultants, E., 2004. Consideration of Lime-stabilized Layers in Mechanistic-empirical Pavement Design. The National Lime Association, Arlington, Virginia, USA. McCallister, L.D., Petry, T.M., 1992. Leach tests on lime-treated clays. Geotech. Test. J. 15, 106–114. Milburn, J.P., Parsons, R., 2004. Performance of Soil Stabilization Agents. Kansas Department of Transportation, Topeka, KS. Nalbantoglu, Z., Tuncer, E.R., 2001. Compressibility and hydraulic conductivity of a chemically treated expansive clay. Can. Geotech. J. 38, 154–160. Onitsuka, K., Modmoltin, C., Kouno, M., 2001. Investigation on microstructure and strength of lime and cement stabilized Ariake clay. Rep. Fac. Sci. Eng. Saga Univ. 30, 49–63. Qiu, Q.C., Mo, H.H., Dong, Z.L., 2007. Vacuum pressure distribution and pore pressure variation in ground improved by vacuum preloading. Can. Geotech. J. 44, 1433–1445. Quang, N.D., Chai, J.C., 2015. Permeability of lime- and cement-treated clayey soils. Can. Geotech. J. 52, 1221–1227. Rajasekaran, G., Narasimha Rao, S., 2002. Permeability characteristics of lime treated marine clay. Ocean Eng. 29, 113–127. Rajasekaran, G., Murali, K., Srinivasaraghavan, R., 1997. Fabric and mineralogical studies on lime treated marine clays. Ocean Eng. 24, 227–234.
Rogers, C.D.F., Glendinning, S., 1997. Improvement of clay soils in situ using lime piles in the UK. Eng. Geol. 47, 243–257. Salehi, M., Sivakugan, N., 2009. Effects of lime-clay modification on the consolidation behavior of the dredged mud. J. Waterw. Port Coast. Ocean Eng. 135, 251–258. Shang, J.Q., Tang, M., Miao, Z., 1998. Vacuum preloading consolidation of reclaimed land: a case study. Can. Geotech. J. 35, 740–749. Tang, M., Shang, J.Q., 2000. Vacuum preloading consolidation of Yaoqiang airport runway. Geotechnique 50, 613–623. Terzaghi, K., Peck, R.B., 1967. Soil Mechanics in Engineering Practice. Ural, N., 2015. Effects of additives on the microstructure of clay. Road Mater. Pavement Des. 17, 104–119. Wang, J., Cai, Y., Ma, J., Chu, J., Fu, H., Wang, P., Jin, Y., 2016a. Improved vacuum preloading method for consolidation of dredged clay-slurry fill. J. Geotech. Geoenviron. 142. Wang, J., Ma, J., Liu, F., Mi, W., Cai, Y., Fu, H., Wang, P., 2016b. Experimental study on the improvement of marine clay slurry by electroosmosis-vacuum preloading. Geotext. Geomembr. 44, 615–622. Yan, S.W., Chu, J., 2003. Soil improvement for a road using the vacuum preloading method. Ground Improv. 7, 165–172. Yan, S.W., Chu, J., 2005. Soil improvement for a storage yard using the combined vacuum and fill preloading method. Can. Geotech. J. 42, 1094–1104. Yong, R.N., Ouhadi, V.R., 2007. Experimental study on instability of bases on natural and lime/cement-stabilized clayey soils. Appl. Clay Sci. 35, 238–249. Zhang, X., Mavroulidou, M., Gunn, M.J., 2015. Mechanical properties and behaviour of a partially saturated lime-treated, high plasticity clay. Eng. Geol. 193, 320–336.
Please cite this article as: Wang, J., et al., Combination of vacuum preloading and lime treatment for improvement of dredged fill, Eng. Geol. (2017), http://dx.doi.org/10.1016/j.enggeo.2017.02.013
Contents lists available at ScienceDirect
Engineering Geology journal homepage: www.elsevier.com/locate/enggeo
Combination of vacuum preloading and lime treatment for improvement of dredged fill Jun Wang a,c,d, Junfeng Ni a,c,d, Yuanqiang Cai b,c,d,⁎, Hongtao Fu a,c,d, Peng Wang a,c,d a
College of Architecture and Civil Engineering, Wenzhou University, Chashan University Town, Wenzhou 325035, China College of Civil Engineering and Architecture, Zhejiang University of Technology, Hangzhou 310014, China c The Key Laboratory of Engineering and Technology for Soft Soil Foundation and Tideland Reclamation of Zhejiang Province, China d Innovation Centre of Tideland Reclamation and Ecological Protection, Wenzhou University, Wenzhou, Zhejiang 325035, China b
a r t i c l e
i n f o
Article history: Received 3 May 2016 Received in revised form 31 January 2017 Accepted 12 February 2017 Available online xxxx Keywords: Vacuum preloading Lime treatment Flocculation Optimum lime concentration Hydrated lime
a b s t r a c t Numerous land reclamation projects that utilise dredged fill are currently being executed in the eastern regions of China to cater to the demands of economic development and address the land shortage problem. However, the dredged fill has high water content, low shear strength, and mainly consists of fine-grained particles. A long period is thus normally required for the treatment of dredge fill by vacuum preloading, and clogging always occurs around the prefabricated vertical drains (PVDs), resulting in insufficient consolidation of the soil. To address this issue, combination of vacuum preloading and lime treatment is proposed in this paper. In this method, a certain percentage of hydrated lime (Ca(OH)2) is added into the dredge fill slurry before vacuum preloading, to enhance the engineering properties of the fill, such as the shear strength and permeability. The added hydrated lime would induce cation exchange on the clay surface and flocculation of the fine soil particles. As a result, this method would significantly mitigate the risk of clogging around the PVDs with the enhanced soil permeability, thereby improving the consolidation efficiency. In this study, dredged slurry with a water content of approximately 187% was utilised. Lime modification optimum (LMO) was first determined using the vacuum preloading method, and a comparison test was then conducted to verify the effectiveness of the proposed method. It was found that, compared to the conventional vacuum preloading method, the proposed method significantly increased the vane shear strength of all the soil layers, especially the deep soil layers, and afforded a higher consolidation rate. © 2017 Published by Elsevier B.V.
1. Introduction The Oufei land reclamation project is currently being executed in eastern Wenzhou, China to cater to the demands of urban expansion and infrastructural development. The project utilises slurry dredged from the coastal seabed as the landfill material (Fig. 1). However, the engineering properties of such dredged fills are too poor to support the construction of structures. Even after years of self-weight consolidation, the fills still have a high water content, low bearing capacity, and high compressibility (Chu et al., 2000). Therefore, it is essential to stabilise the soft clayey soil before any construction work is conducted. One of the methods most widely used for the improvement of soft clay ground is vacuum preloading. The method, which was first proposed by Kjellman (1952), is convenient and economical, and is increasingly being used to improve the mechanical performance of dredged fills in numerous projects (Chai et al., 2010, 2006; Indraratna et al., 2011; Shang et al., 1998; Tang and Shang, 2000; Yan and Chu, 2005). Prefabricated vertical drains (PVDs) are used in the vacuum preloading ⁎ Corresponding author. College of Civil Engineering and Architecture, Zhejiang University of Technology, Hangzhou 310014, China. E-mail address: [email protected] (Y. Cai).
process to distribute the vacuum pressure and discharge the pore water (Indraratna et al., 2005; Yan and Chu, 2003). However, due to the dredged slurry mainly consists of fine-grained soils, these particles would be moved as the water drainage during the vacuuming and eventually aggregate around the PVDs (Wang et al., 2016b). This leads to the formation of an obvious soil column and clogging around the PVDs, ultimately resulting in insufficient consolidation and limited shear strength improvement (Wang et al., 2016a). There is thus the need for the enhancement of vacuum consolidation. This paper proposes a method for improving the effectiveness of vacuum consolidation. The proposed method involves the addition of a certain percentage of hydrated lime to the dredged slurry before vacuum preloading. When a hydrated lime is added to a soil, a modification reaction occurs immediately as a result of the substitution of calcium with the existing cations at negative charge sites on the clay mineral surface. Consequently, the interparticle repulsive forces between clay particles are decreased, which induced soil particles aggregate and flocculation (Le Runigo et al., 2009; Salehi and Sivakugan, 2009). The change of soil particles from the initial dispersed orientation state to flocculation increases the void ratio, thereby enhancing the permeability of the soil (Brandl, 1981; Rajasekaran and Narasimha Rao, 2002). These behaviours are beneficial to water flow and drainage of the soil.
http://dx.doi.org/10.1016/j.enggeo.2017.02.013 0013-7952/© 2017 Published by Elsevier B.V.
Please cite this article as: Wang, J., et al., Combination of vacuum preloading and lime treatment for improvement of dredged fill, Eng. Geol. (2017), http://dx.doi.org/10.1016/j.enggeo.2017.02.013
2
J. Wang et al. / Engineering Geology xxx (2017) xxx–xxx
and plastic limit (WP) were approximately 81% and 42%, respectively, while the specific gravity was determined to be 2.61. 3. Determination of LMO
Fig. 1. Layout of the tideland reclamation site in Lingkun (Longwan District), Wenzhou, China.
This approach can also reduce the occurrence of clogging around the PVDs by retarding the formation of soil columns, thereby improving the efficiency of vacuum consolidation. However, the permeability does not indefinitely increase with increasing lime content, but begins to slowly decrease beyond a threshold value. This decrease is due to the flow channels being blocked by the precipitation of cementitious compounds generated by the lime-soil reactions, resulting in reduced water flow (Le Runigo et al., 2009). Hence, it is imperative to determine the threshold for subsequent vacuum preloading. This threshold was also known as the lime modification optimum (LMO), which was first introduced by McCallister and Petry (1992) through leach tests conducted on lime-treated clay. They reported that the maximum permeability occurs at the LMO. Eades and Grim (1966) first developed the pH method for determining the LMO, and Rogers and Glendinning (1997) subsequently improved the method by considering the change in plasticity. Salehi and Sivakugan (2009) also determined the LMO for dredged mud based on the permeability and void ratio. However, these methods only consider the changes in the permeability or pH without directly assessing the efficiency of the vacuum preloading process. In the present study, a determination test was firstly conducted to identify the LMO by vacuum preloading with the purpose of quantifying the changes produced by the use of hydrated lime. Shear strength, pore water pressure, discharged water volume, and water content are regarded as the indicators for the determination of LMO by the vacuum preloading method. A comparison test was also performed to verify the effectiveness of the proposed method. Briefly, the objectives of this study were as follows: 1. To develop a method for determining the optimum lime content for enhancing the dredged slurry in Wenzhou. 2. To investigate the consolidation behaviour of natural and lime-treated dredged slurry during vacuum preloading. 3. To evaluate the efficiency and feasibility of the proposed method. 2. Properties of dredged slurry The clay slurry used for the tests was obtained from the site of the Oufei project in Wenzhou, China. The geological setting and depositional environment of the slurry are shown in Fig. 2. The sample slurry had just dredged from the seabed and had a water content of 187%. It mainly consisted of fine clay particles and had a pH of 6.8. The liquid limit (WL)
The void ratio and permeability of the soil are generally increased by the lime-clay reaction (Rajasekaran and Narasimha Rao, 2002). This means that the chemical reaction generates more pore channels, which enhance water discharge from the soil during subsequent vacuum consolidation. However, as implied earlier, the pore channels do not indefinitely increase with increasing lime content, but there exists an optimum lime content that maximises the number of pore channels. Below this optimum content, only flocculation occurs, while pozzolanic reactions are also triggered above it (McCallister and Petry, 1992). To determine this optimum lime concentration, the following laboratory test was developed and conducted. Slurry samples of equal mass (34 kg) were poured into five model buckets of height 40 cm and diameter 30 cm. Different amounts of lime representing 0%, 1%, 1.5%, 2%, and 2.5% of the dry soil by weight were respectively added to the slurry samples. Each mixture of lime and slurry was then thoroughly stirred by a slurry agitator until homogeneity was achieved, to ensure complete reaction between the lime and slurry (the first ‘mixture’ containing 0% lime was the natural slurry sample). PVDs with cap connections were subsequently installed horizontally in all the samples. A pore pressure sensor was placed at the bottom of each model bucket, which was filled with a sample to a height of 35 cm. A layer of geotextile and two layers of geomembranes were placed on top of the slurry in each bucket to seal the sample from the atmosphere. The cap, air–water separation flask, and vacuum pump were then successively connected by vacuum pipes. The complete test setup is shown in Fig. 3. During the vacuum consolidation, a minimum vacuum load of 80 kPa was applied to the surface of the soil and maintained until there was no further apparent increment in the measured volume of discharged water in the air-water separation flask. Once the steady state was attained, the vacuum pump was switched off, and the vane shear strength and water content of the consolidated soil were measured. The variations of the discharged water volume with time of all the samples are depicted in Fig. 4. It is obvious that, for all the samples, the discharged water volume increases sharply at the beginning of preloading. A stable value is attained for all the lime-treated slurry samples within 80 h, whereas the untreated sample continues to discharge water for an additional 88 h. This suggests enhanced drainage capability of the lime-treated slurries, especially that containing 2% lime, which stops draining within 48 h. Moreover, as the lime content increases from 1% to 2%, there is an increase in the volume of discharged water to the highest value among the treated samples. This can be explained by the permeability, which undoubtedly affects the draining during vacuum preloading; a higher hydraulic conductivity accelerates draining. When the lime is added to the slurry, cation exchange and flocculation immediately begin, with the permeability increased by the flocculation and resultant soil aggregation (Nalbantoglu and Tuncer, 2001; Rajasekaran and Narasimha Rao, 2002). The increased permeability in the present tests can be expressed in terms of not only the volume of discharged water, but also the rate of drainage. However, as noted, the permeability only increases up to a maximum value afforded by an optimal lime content, the LMO, and then begins to decrease due to the formation of cementitious compounds among the soil particles (Alhassan, 2008; Milburn and Parsons, 2004; Onitsuka et al., 2001). The cementitious compounds block the flow channels, reducing the water flow (Quang and Chai, 2015) and locking part of the water within the soil. This is evident from Fig. 4, which shows that 2.5% lime content produces the lowest volume of discharged water despite the accelerated drainage. It is very noteworthy that the natural slurry produces the highest volume of
Please cite this article as: Wang, J., et al., Combination of vacuum preloading and lime treatment for improvement of dredged fill, Eng. Geol. (2017), http://dx.doi.org/10.1016/j.enggeo.2017.02.013
J. Wang et al. / Engineering Geology xxx (2017) xxx–xxx
3
Fig. 2. Geological section of the Oufei project site.
discharged water. According to Mallela et al. (2004), lime-treated clays exhibit enhanced capability to hold a large amount of water. This is because some of the water is attracted by the modified soil particles and difficult to discharge despite the improved permeability. Fig. 5 shows the dissipation of the pore water pressure monitored at 30 cm below the surface of each sample during the vacuum consolidation. It is obvious that the dissipation of the pore water pressure in the lime-treated samples is almost completed within 80 h, whereas 150 h is required for the case of the natural slurry. This suggests that the addition of the lime accelerates the dissipation of the pore water pressure. Similar to the water drainage, the reduction of the pore water pressure increases as the lime content increases from 1% to 2%, after which it decreases for further increase of the lime content to 2.5%. The reason for this is also similar to that for the trends of the water drainage (Fig. 4).
This is predictable considering that drainage of water generally leads to dissipation of the pore water pressure according to the Terzaghi theory (Terzaghi and Peck, 1967). The lime treatment of slurry thus not only increases the reduction of the pore water pressure, but also accelerates the reduction. The water contents of all the samples were measured after the vacuum preloading. Fig. 6 shows the determined variation of the final water content with the lime content with respect to the depth in the samples. As can be seen, the final water content increases with increasing lime content from 0% to 2.5% (dry weight basis), without peaking. The cementitious compounds produced by the lime-slurry reactions are calcium silicate hydrate (CSH) and calcium aluminate hydrate (CAH), which are composed of the solid products of hydration and water (Le Runigo et al., 2009; Rajasekaran et al., 1997). The reactions
Please cite this article as: Wang, J., et al., Combination of vacuum preloading and lime treatment for improvement of dredged fill, Eng. Geol. (2017), http://dx.doi.org/10.1016/j.enggeo.2017.02.013
4
J. Wang et al. / Engineering Geology xxx (2017) xxx–xxx
Fig. 3. Schematic diagram of the test setup.
The above equations indicate that more water-containing agents are generated in the pore with increasing lime content. This is the reason why, as shown in Fig. 6, the water content increases with increasing lime content to 2.5%, despite the above-mentioned increase in the water discharge only with increasing lime content from 1% to 2%. Vane shear tests were also conducted at different depths of the samples after the vacuum preloading. The determined variation of the vane shear strength with the lime concentration with respect to the depth is shown in Fig. 7. As indicated, the vane shear strength gradually increases with increasing lime content from 0 to 2% (dry weight basis), at which it reaches the maximum value of 26 kPa. Thereafter, the vane shear strength decreases with further increase of the lime content to 2.5%, at which the value is approximately the same as that for 1.5% lime. There are two primary causes of the increase in the shear strength. First, flocculation is immediately induced by the addition of the lime to the slurry, resulting in changes in the soil fabric through rearrangement of the soil particles. The new fabric is relatively open, which is beneficial for the discharge of pore water. As the water discharged by PVDs, soil particles aggregated under the vacuum preloading, shear strength was enhanced. This is the main cause of the increase in shear strength. The other cause is the CSH and CAH generated during the long curing
period. As a type of gel, these compounds cement the flocculated soil particles, resulting in the formation of a stronger fabric with a higher shear strength (Dash and Hussain, 2012). In addition, the shear strength is affected by the arrangement of the particles within the soil, which indicates that a better fabric of the lime-treated soil essentially enhances the interparticle shear resistance. However, the vane strength does not increase indefinitely with the lime content, but begins to decrease when the amount of lime becomes excessive, as shown in Fig. 7. Considering that the initial increase in shear strength during the vacuum consolidation is mainly due to drainage of the slurry, the decrease with excessive addition of lime may be attributed to the resultant blockage of the water flow. As it mentioned in Fig. 4, the discharged water did not continuously increase with increasing lime content because of the pozzolanic reactions induced decrease in the permeability. Similarly, the decrease of shear strength is also due to the triggering of pozzolanic reactions between the lime and slurry. When sufficient lime is added to the slurry, the precipitation of the cementitious compounds blocks the flow channels and reduces the water flow, resulting in the retention of a relatively high water content (Milburn and Parsons, 2004; Onitsuka et al., 2001). As the water is trapped in the slurry, the efficiency of the vacuum consolidation is limited and the bonds among the soil particles are weakened, resulting in lower vane shear strength. The above comprehensive investigation of the effect of the amount of lime added to dredged slurry indicates that 2% lime content (dry weight basis) produces the best drainage and shear strength. This amount is thus considered the optimum lime concentration for vacuum preloading.
Fig. 4. Discharged water volume vs. time curve of the slurry with respect to the lime content.
Fig. 5. Variation of the pore water pressure at a depth of 30 cm in the slurry with respect to the lime content.
are as follows (Jawad et al., 2014; Mallela et al., 2004; Yong and Ouhadi, 2007): CaðOHÞ2 þ SiO2 →CaO−SiO2 −H2 O
ð1Þ
CaðOHÞ2 þ Al2 O3 →CaO−Al2 O3 −H2 O
ð2Þ
Please cite this article as: Wang, J., et al., Combination of vacuum preloading and lime treatment for improvement of dredged fill, Eng. Geol. (2017), http://dx.doi.org/10.1016/j.enggeo.2017.02.013
J. Wang et al. / Engineering Geology xxx (2017) xxx–xxx
5
4. Comparison test Because the model buckets used in the above tests were too small for the simulation of the scenario of a real land reclamation project, and the acquired data was limited, a comparison test was conducted to verify the effectiveness of the proposed method for improving the effectiveness of vacuum consolidation. For simplicity, only two slurry samples (without lime and with 2% lime) were used for the test. It was expected that the exact increment in the vane shear strength and the reduction in period of consolidation would be measured after the vacuum preloading. 4.1. Experimental apparatus
Fig. 6. Variation of the water content with the lime content of the slurry with respect to the depth.
The apparatus used for the comparison test mainly consisted of model boxes (height 1 m and width and length 2 m), vacuum pipes, a vacuum pump, and prefabricated vertical drains (PVDs). A special slurry agitator convenient for mixing a large amount of slurry with lime and the accompanying mixing method was developed for the test. In this method, lime was previously mixed with water. As shown in Fig. 8, the slurry agitator had two inlets in its upper part for the respective inflow of the limewater and slurry, and one outlet underneath for the outflow of the lime-slurry mixture. The homogeneous lime-treated slurry was attained by pumping lime and slurry into the agitator at the same time and thoroughly stirred. 4.2. Sample preparation and test procedure
Fig. 7. Variation of the vane shear strength with the lime content of the slurry with respect to the depth.
Two different types of slurries were actually used for the test. One was obtained from the site of the Oufei project. This slurry had an original water content of 187%, but had undergone a half-year of self-weight consolidation, which reduced the water content to 110%. It was used to simulate a practical project, wherein vacuum preloading is generally implemented after the slurry has undergone self-weight consolidation. However, considering the difficulty of mixing lime with hardened slurry in practice, the lime treatment should be performed at the time of pumping the slurry unto the site by injection of the lime into the slurry pipeline. Therefore, to simulate this practical engineering requirement, slurry with the original water content of 187% was obtained from the same site and thoroughly mixed with lime using the developed mixing method. Furthermore, to prevent the lime from causing long-term
Fig. 8. Schematic of the employed slurry agitator and illustration of the lime-slurry mixing method.
Please cite this article as: Wang, J., et al., Combination of vacuum preloading and lime treatment for improvement of dredged fill, Eng. Geol. (2017), http://dx.doi.org/10.1016/j.enggeo.2017.02.013
6
J. Wang et al. / Engineering Geology xxx (2017) xxx–xxx
Fig. 9. (a) Monitoring points of the vacuum pressure and pore water pressure during the vacuum preloading, and (b) plan view of the vacuum system.
alteration of the slurry and to take full advantage of the immediate limeslurry reactions such as flocculation (Jha and Sivapullaiah, 2015; Ural, 2015), vacuum preloading was immediately implemented on the lime-treated sample. The characteristics of the two slurries were the same as those used for the previous tests, except that the first slurry of the present test had a lower water content and void ratio, naturally acquired through its self-weight consolidation. However, considering that the ultimate strength is generally used as the main indicator to estimate the efficiency of vacuum preloading, the difference between the initial water contents of the two slurries of the present test only marginally affected the results. The samples were prepared as follows. The slurry with a water content of 187% was first poured into a model box and then mixed with limewater using the slurry agitator shown in Fig. 8. The lime-treated slurry was then pumped into another model box up to a height of 90 cm. For the comparison, the natural slurry with a water content of
110% was poured into another model box up to a height of 62 cm. The dry weights of the two slurry soils were the same. After the above preparation of the samples, the vacuum system was connected as described below. The vacuum pipes with needles were first inserted into the PVDs and connected to the vacuum pressure gauge. Two types of PVDs of different sizes (Fig. 9), namely mini PVD and PVD with cap (CPVD), were used to fix the vacuum pipes with needles and transfer the vacuum pressures in the PVDs and soil. All of the four employed CPVDs was connected to a vacuum pump and vertical drain through vacuum tubes, while the single mini PVD was only used to hold the vacuum pipe needles. This method reduced clogging in the vacuum pipes effectively and therefore facilitated measurement of the vacuum pressure in the PVDs and soil. The four CPVDs and one mini PVD, together with the pore water pressure sensors, were installed on an iron rack, to prevent the PVDs from being bent and to limit the movements of the PVDs and sensors during
Please cite this article as: Wang, J., et al., Combination of vacuum preloading and lime treatment for improvement of dredged fill, Eng. Geol. (2017), http://dx.doi.org/10.1016/j.enggeo.2017.02.013
J. Wang et al. / Engineering Geology xxx (2017) xxx–xxx
7
the vacuum preloading. Fig. 12. Pore water pressure vs. time curves of the slurries with respect to the depth 35 cmTaway from the PVDs. he iron rack was laid on the bottom of the model box and one layer of geotextile and two layers of geomembranes were used to seal the surface of the slurry. The same test setup was used for the two considered slurries. The distribution of the monitoring points of the vacuum pressure and pore water pressure during the vacuum preloading is shown in Fig. 9. The application of the vacuum pressure and consolidation of the slurry commenced when the vacuum pump was switched on.
4.3. Results and discussion
Fig. 10. Vacuum pressure vs. time curves with respect to the depth in the PVDs: (a) natural slurry, (b) slurry containing 2% lime.
Fig. 11. Vacuum pressure vs. time curves of the slurries with respect to the depth 30 cm away from the PVDs.
4.3.1. Vacuum pressure Fig. 10a and b respectively show the vacuum pressures in the natural and lime-treated slurries at different depths within the PVDs during the preloading. It can be clearly seen from the figures that a vacuum pressure of above 80 kPa was maintained at all depths in both slurries throughout the durations of the tests, with the pressure decreasing with increasing depth. Moreover, both slurries have almost the same vacuum pressure gradient in the vertical direction of approximately 10 kPa/m. This indicates that it is impossible to maintain a uniform vacuum pressure along the depth of the consolidated soil, as has been previously observed (Chai et al., 2005; Indraratna et al., 2005; Shang et al., 1998). There is, however, significant difference between the vacuum pressures in the two slurries outside the PVDs. As shown in Fig. 11, the vacuum pressure in the slurry containing 2% lime reaches as much as 75 kPa, which is much higher than that in the natural slurry. In addition, the lime-treated slurry requires a shorter time to attain a steady vacuum pressure. It is generally accepted that the addition of lime to soil increases the hydraulic conductivity with a more open the soil fabric (Nalbantoglu and Tuncer, 2001; Rajasekaran and Narasimha Rao, 2002). Moreover, Chai et al. (2010) found that the final vacuum pressure distribution was primarily determined by the hydraulic conductivity, with the other parameters having no considerable effect. The present results also show that the vacuum pressure outside the PVDs in the lime-treated slurry begins to increase as soon as the vacuuming begins, whereas 70 h is required for any change in vacuum pressure to occur in the natural slurry. Further, the addition of lime to soil has been observed to cause partial saturation (Zhang et al., 2015), which results in the immediate transfer of vacuum pressure within the soil. However, the presence of an unsaturated zone during preloading (Qiu et al., 2007) enables the loosened fabric to accelerate the drainage of water and facilitate the distribution of pressure in the soil. This accelerates the increase of the vacuum pressure in the soil and increases the maximum pressure attained, as shown in Fig. 11.
Please cite this article as: Wang, J., et al., Combination of vacuum preloading and lime treatment for improvement of dredged fill, Eng. Geol. (2017), http://dx.doi.org/10.1016/j.enggeo.2017.02.013
8
J. Wang et al. / Engineering Geology xxx (2017) xxx–xxx
Fig. 13. Plan view of the settlement and vane shear strength measurement points.
4.3.2. Pore water pressure Fig. 12 shows the variations of the pore water pressure at different depths in the two samples during the vacuum preloading. As can be seen, the pore water pressure in the lime-treated slurry reduces by nearly 80 kPa over 154 h and then becomes steady, whereas there is a continuous slow decrease over 478 h in the natural slurry. The limetreated slurry not only has a higher pore water pressure dissipation rate, but also a lower final value. The reasons for this are the same as those adduced in the discussion of the LMO determination tests. Moreover, the present observations agree with previous findings, that a higher vacuum pressure increases the drain rate and accelerates the dissipation of the pore water pressure (Indraratna, 2010). The present results further confirm the effectiveness of lime treatment.
4.3.3. Settlement and degree of consolidation The settlement data at five locations in each of the model boxes (PA, PB, PC, PD, and PE in Fig. 13) were recorded during the vacuum preloading. The obtained curves of the settlement in the two slurries are shown in Fig. 14. The settlement in the lime-treated slurry is clearly greater, and the rate of consolidation under vacuum preloading is higher compared to the natural slurry. In addition, the settlement in the lime-treated slurry is almost completed within 270 h, whereas that in the natural slurry requires about 400 h. This shows that lime treatment reduces the consolidation time by about a third part. In addition, as can be observed from Fig. 14, lime-treated soil reveals a more uniform settlement, which indicates that the clogging and soil columns may less than untreated one. However, considering that the initial volumes of the two slurries were different, the observed relative settlements cannot be accurately used to assess the effectiveness of lime treatment for soil improvement. The degree of consolidation (DOC), which is based on the settlement and pore water pressure, was used as a better measure. For this purpose, the method Asaoka was used to determine the ultimate settlements of both slurry samples (Asaoka, 1978). In this method, the settlement curve of a sample is obtained from the settlement data for a given period, and the ultimate settlement is determined as the intersection point of the settlement curve with the 45° line (Chu and Yan, 2005). Based on the ultimate settlements obtained in this study, the DOCs of the natural
and lime-treated slurries were determined to be 92% and 94%, respectively. Alternatively, the DOC at a given elevation can be estimated as (1 − Δuf/Pvac) × 100%, where Δuf is the pore water pressure at the completion of vacuum preloading, measured as the difference between the final pore water pressure and the suction line pressure (Chu et al., 2000); and Pvac is the corresponding vacuum pressure in the soil. The DOCs of the natural and lime-treated slurries estimated by this method using data obtained from Fig. 11 are 51% and 87% respectively. This shows that the DOC estimated from the pore water pressure is lower than that estimated from the settlement. The use of both methods is recommended (Chu and Yan, 2005). However, the DOCs determined from the pore water pressure in the present study are considered more accurate, because the settlement in the natural slurry under vacuum preloading was not uniform, which may cause a distinct DOC using settlement from different points. 4.3.4. Vane shear strength After the vacuum preloading, vane shear strength tests were conducted on the slurry samples at different points in the model box (A–E in Fig. 13). The results are presented in Fig. 15, which reveals a significantly higher overall vane shear strength of the lime-treated slurry, including at the deep layers, with the surface value being as high as 28.8 kPa. In addition, there is a lower rate of reduction of the vane shear strength of the treated sample with increasing depth. The improvement of the vane shear strength of the natural slurry was basically limited to the surface. These observations are due to the faster and higher vacuum pressure distribution to the deeper layers of the limetreated slurry. They confirm the effectiveness of lime treatment for comprehensively improving the shear strength of dredged slurry by vacuum preloading. 5. Conclusions The combined use of lime treatment and vacuum preloading was proposed for improving the consolidation of dredged slurry, and a series of laboratory tests were used to investigate the effectiveness of the method. Following is a summary of the findings and conclusions drawn from the study:
Please cite this article as: Wang, J., et al., Combination of vacuum preloading and lime treatment for improvement of dredged fill, Eng. Geol. (2017), http://dx.doi.org/10.1016/j.enggeo.2017.02.013
J. Wang et al. / Engineering Geology xxx (2017) xxx–xxx
9
2. Comparison tests using model boxes confirmed the effectiveness of the proposed method. The lime treatment was found to substantially improve the efficiency of the vacuum preloading, resulting in the achievement of a higher vacuum pressure within the soil and higher rate of consolidation. 3. The addition of the lime to the slurry reduced clogging around the PVDs and retarded the formation of soil columns during the vacuum preloading. 4. The most remarkable feature of the proposed method is that the material is prepared in slurry form through the rapid and convenient mixing of the pure slurry with lime. A specially developed apparatus was used for the present study. The foregoing discussion highlights the beneficial effect of lime on increasing the rate of vacuum consolidation and the values of vane shear strength of dredged fill. These laboratory tests prove the effectiveness and operational feasibility of the proposed method with lime treatment before the vacuum preloading. The advantages of this method may be used to develop an innovative ground improvement technique in land reclamation work using the dredged slurry as the fill material. Funding This work was supported by National Key Research and Development Program of China (Grant No. 2016YFC0800203), the Program of International Science and Technology Cooperation (Grant No. 2015DFA71550), the National Natural Science Foundation of China (Grant No. 51620105008, No. 51622810, No. 51408440 and No. 51478365), the Natural Science Foundation of Zhejiang Province Project (Grant No. LY17E080010), the Project of Science and Technology of Wenzhou (Grant No. S20150015). Acknowledgments The authors thank the Key Laboratory of Engineering and Technology for Soft Soil Foundation and Tideland Reclamation, and the Innovation Centre of Tideland Reclamation and Ecological Protection for the provision of the apparatus used for this study. Fig. 14. Surface settlement vs. time curves of the slurries: (a) natural slurry, (b) slurry containing 2% lime.
1. The optimum lime content for maximising the water discharge rate and vane shear strength of the dredged slurry utilised in the ongoing Wenzhou landfill project is 2% (dry mass basis).
Fig. 15. Variations of the shear strengths of the slurries with depth.
References Alhassan, M., 2008. Permeability of lateritic soil treated with lime and rice husk ash. Assumption Univ. J. Thailand 12, 115–120. Asaoka, A., 1978. Observational procedure of settlement prediction. Soils Found. 18, 87–101. Brandl, H., 1981. Alteration of soil parameters by stabilization with lime. Proceedings of the International Conference on Soil Mechanics and Foundation Engineering. 3, pp. 587–594. Chai, J.C., Carter, J.P., Hayashi, S., 2005. Ground deformation induced by vacuum consolidation. J. Geotech. Geoenviron. 131, 1552–1561. Chai, J.C., Carter, J.P., Hayashi, S., 2006. Vacuum consolidation and its combination with embankment loading. Can. Geotech. J. 43, 985–996. Chai, J., Hong, Z., Shen, S., 2010. Vacuum-drain consolidation induced pressure distribution and ground deformation. Geotext. Geomembr. 28, 525–535. Chu, J., Yan, S.W., 2005. Estimation of degree of consolidation for vacuum preloading projects. Int. J. Geomech. 5, 158–165. Chu, J., Yan, S.W., Yang, H., 2000. Soil improvement by the vacuum preloading method for an oil storage station. Geotechnique 50, 625–632. Dash, S.K., Hussain, M., 2012. Lime stabilization of soils: reappraisal. J. Mater. Civ. Eng. 24, 707–714. Eades, J.L., Grim, R.E., 1966. A quick test to determine lime requirements for lime stabilization: in behaviour characteristics of lime-soil mixtures. Highw. Res. Rec. 139, 61–72. Indraratna, B., 2010. Recent advances in the application of vertical drains and vacuum preloading in soft soil stabilisation. Aust. Geomech. J. 45, 1–44. Indraratna, B., Sathananthan, I., Rujikiatkamjorn, C., Balasubramaniam, A.S., 2005. Analytical and numerical modeling of soft soil stabilized by prefabricated vertical drains incorporating vacuum preloading. Int. J. Geomech. 5, 114–124. Indraratna, B., Rujikiatkamjorn, C., Ameratunga, J., Boyle, P., 2011. Performance and prediction of vacuum combined surcharge consolidation at port of Brisbane. J. Geotech. Geoenviron. 137, 1009–1018. Jawad, I.T., Taha, M.R., Majeed, Z.H., Khan, T.A., 2014. Soil stabilization using lime: advantages, disadvantages and proposing a potential alternative. Res. J. Appl. Sci. Eng. Technol. 8, 510–520.
Please cite this article as: Wang, J., et al., Combination of vacuum preloading and lime treatment for improvement of dredged fill, Eng. Geol. (2017), http://dx.doi.org/10.1016/j.enggeo.2017.02.013
10
J. Wang et al. / Engineering Geology xxx (2017) xxx–xxx
Jha, A.K., Sivapullaiah, P.V., 2015. Mechanism of improvement in the strength and volume change behavior of lime stabilized soil. Eng. Geol. 198, 53–64. Kjellman, W., 1952. Consolidation of clay soil by means of atmospheric pressure. Proceedings of a Conference on Soil Stabilization, pp. 258–263. Le Runigo, B., Cuisinier, O., Cui, Y.J., Ferber, V., Deneele, D., 2009. Impact of initial state on the fabric and permeability of a lime-treated silt under long-term leaching. Can. Geotech. J. 46, 1243–1257. Mallela, J., Von Quintus, P.H., Smith, K.L., Consultants, E., 2004. Consideration of Lime-stabilized Layers in Mechanistic-empirical Pavement Design. The National Lime Association, Arlington, Virginia, USA. McCallister, L.D., Petry, T.M., 1992. Leach tests on lime-treated clays. Geotech. Test. J. 15, 106–114. Milburn, J.P., Parsons, R., 2004. Performance of Soil Stabilization Agents. Kansas Department of Transportation, Topeka, KS. Nalbantoglu, Z., Tuncer, E.R., 2001. Compressibility and hydraulic conductivity of a chemically treated expansive clay. Can. Geotech. J. 38, 154–160. Onitsuka, K., Modmoltin, C., Kouno, M., 2001. Investigation on microstructure and strength of lime and cement stabilized Ariake clay. Rep. Fac. Sci. Eng. Saga Univ. 30, 49–63. Qiu, Q.C., Mo, H.H., Dong, Z.L., 2007. Vacuum pressure distribution and pore pressure variation in ground improved by vacuum preloading. Can. Geotech. J. 44, 1433–1445. Quang, N.D., Chai, J.C., 2015. Permeability of lime- and cement-treated clayey soils. Can. Geotech. J. 52, 1221–1227. Rajasekaran, G., Narasimha Rao, S., 2002. Permeability characteristics of lime treated marine clay. Ocean Eng. 29, 113–127. Rajasekaran, G., Murali, K., Srinivasaraghavan, R., 1997. Fabric and mineralogical studies on lime treated marine clays. Ocean Eng. 24, 227–234.
Rogers, C.D.F., Glendinning, S., 1997. Improvement of clay soils in situ using lime piles in the UK. Eng. Geol. 47, 243–257. Salehi, M., Sivakugan, N., 2009. Effects of lime-clay modification on the consolidation behavior of the dredged mud. J. Waterw. Port Coast. Ocean Eng. 135, 251–258. Shang, J.Q., Tang, M., Miao, Z., 1998. Vacuum preloading consolidation of reclaimed land: a case study. Can. Geotech. J. 35, 740–749. Tang, M., Shang, J.Q., 2000. Vacuum preloading consolidation of Yaoqiang airport runway. Geotechnique 50, 613–623. Terzaghi, K., Peck, R.B., 1967. Soil Mechanics in Engineering Practice. Ural, N., 2015. Effects of additives on the microstructure of clay. Road Mater. Pavement Des. 17, 104–119. Wang, J., Cai, Y., Ma, J., Chu, J., Fu, H., Wang, P., Jin, Y., 2016a. Improved vacuum preloading method for consolidation of dredged clay-slurry fill. J. Geotech. Geoenviron. 142. Wang, J., Ma, J., Liu, F., Mi, W., Cai, Y., Fu, H., Wang, P., 2016b. Experimental study on the improvement of marine clay slurry by electroosmosis-vacuum preloading. Geotext. Geomembr. 44, 615–622. Yan, S.W., Chu, J., 2003. Soil improvement for a road using the vacuum preloading method. Ground Improv. 7, 165–172. Yan, S.W., Chu, J., 2005. Soil improvement for a storage yard using the combined vacuum and fill preloading method. Can. Geotech. J. 42, 1094–1104. Yong, R.N., Ouhadi, V.R., 2007. Experimental study on instability of bases on natural and lime/cement-stabilized clayey soils. Appl. Clay Sci. 35, 238–249. Zhang, X., Mavroulidou, M., Gunn, M.J., 2015. Mechanical properties and behaviour of a partially saturated lime-treated, high plasticity clay. Eng. Geol. 193, 320–336.
Please cite this article as: Wang, J., et al., Combination of vacuum preloading and lime treatment for improvement of dredged fill, Eng. Geol. (2017), http://dx.doi.org/10.1016/j.enggeo.2017.02.013