Journal of Marine Science and Engineering

Article Analysis of the Effectiveness of the Step Vacuum Preloading Method: A Case Study on High Clay Content Dredger Fill in Tianjin, China

Jinfeng Li, Huie Chen , Xiaoqing Yuan * and Wenchong Shan

College of Construction Engineering, Jilin University, No.938 Ximinzhu Street, 130026 Changchun, China; jfl[email protected] (J.L.); [email protected] (H.C.); [email protected] (W.S.) * Correspondence: [email protected]



Received: 30 December 2019; Accepted: 8 January 2020; Published: 13 January 2020


Abstract: As a solution to avoid the blockage of the drainage pipe by traditional vacuum preloading, step vacuum preloading (SVP) has been progressively studied. However, the effectiveness of this technique has yet to be systematically analyzed. In this study, an indoor model test was conducted in which vacuum pressure was applied in five stages (10, 20, 40, 60, and 80 kPa) to dredger soil with high clay content at a reclamation site in Binhai New Area, Tianjin, China. The extent of the consolidation effect of the soil was determined, and the effectiveness of the step vacuum preloading method to address drainage pipe blockage was evaluated. The results indicate that soil settlement increases at each stage of vacuum pressure treatment and the degree of vertical consolidation at each stage exceeds 90%. At the end of the treatment stage with vacuum pressure of 80 kPa, the weakly bound water was discharged. Dissipation of pore water pressure occurred in all stages. On the basis of these results, it is shown that SVP can efficiently reinforce dredger fill. Moreover, after SVP, the grain size of the soil and void ratio are still uniformly distributed. Regardless of their location from the drainage pipe, soil exhibits permeability coefficients within the same order of magnitude. The consolidation effect of soil in each stage and the increased drainage rate in the initial stage of vacuum preloading with 80 kPa indicate that the test in the current study can decrease the horizontal displacement of fine particles and can avoid drainage pipe blockage.

Keywords: step vacuum preloading; dredger fill; permeability coefficient; consolidation effect; drain pipe blockage

1. Introduction In 1952, the Swedish scholar Walter Kjellman proposed vacuum preloading. This technique has been used in many countries for various large-scale projects after decades of research and experimentation. These projects include land reclamation projects and subgrade projects in which the soil mass needs to be strengthened. The technique is low cost, requires a short period to complete construction, requires no filling materials, involves no heavy machinery, and provides good reinforcement [1–7]. Regardless, the traditional vacuum preloading (TVP) method still has several disadvantages: On the one hand, under the action of vacuum pressure and water flow, fine soil particles gradually migrate to the vicinity of the prefabricated vertical drainage pipe, forming a low-permeability layer near the drainage pipe. This process reduces the vacuum pressure on the soil, causing drainage pipe blockage and a reduction in drainage capacity [8–10]. On the other hand, owing to the temporal and spatial effects of vacuum suction transfer in space, the process can easily lead to the uneven settlement of soil mass [11]. Numerous solutions to these problems have been proposed in various studies. Adding coarse sand, lime, or FeCl3 to soft soil can influence flocculation, significantly improve

J. Mar. Sci. Eng. 2020, 8, 38; doi:10.3390/jmse8010038 www.mdpi.com/journal/jmse J. Mar. Sci. Eng. 2020, 8, 38 2 of 16 the drainage capacity of the soil, shorten its consolidation time, and solidify heavy metals in soft soil. However, this method is too expensive and complex in high-scale projects and may cause secondary pollution, with even worse effects in coastal areas [12–14]. Compared with the traditional vacuum preloading technique, the combination of electroosmosis and vacuum preloading is more efficient and produces higher negative pore water pressures but entails considerably higher costs [15]. The low conductivity of the dredger fill in some cases, such as drainage to a low moisture content, renders this method unsatisfactory. Moreover, the metal electrode in the cathode is easily corroded by seawater, which also leads to the blockage of drainage pipes [16,17]. In 2006, Qing Wang (China) presented the step vacuum preloading (SVP) method. The layer-by-layer application of vacuum pressure alleviates the rapid mass migration of fine particles and fundamentally improves TVP. The advantages and disadvantages of SVP have been studied in recent years. Wang et al. [13] conducted a two-stage (at 40 and 80 kPa) vacuum preloading test [18] and a three-stage (at 20, 40, and 80 kPa) vacuum preloading test [19]. Compared with the direct vertical-vacuum preloading test, the step vacuum preloading method can improve the consolidation efficiency of the hydraulic fill and significantly reduce the blockage of the drainage pipe. Wu et al. [20] explored different approaches to vacuum loading and determined that small-vacuum loading followed by large-vacuum loading on soft soil can more easily allow the soil to settle, compared with direct large-vacuum loading; moreover, the final strength of the soil is greater. Yan [21] then explored a five-stage (10, 20, 40, 60, and 80 kPa) vacuum preloading test, and porosity testing of soil microstructure indicated a change in the soil sample from nondirectional to directional and ultimately to nondirectional. The change in directionality can indirectly reflect the drainage situation and the degree of reinforcement. The change in moisture content and drainage capacity of soil under SVP was examined by Liu et al. [22]. The results show that SVP can significantly reduce moisture content in soil, and drainage in SVP is similar to that in TVP, including the rapid growth area, slow growth area, and stable area. Using the indoor model experiment of SVP, Yuan [23] demonstrated that moisture content in soil decreased in each stage and that bearing capacity increased with a decrease in moisture content, which proved that the SVP method exerts a consolidation effect in each stage. Fang et al. [24] conducted an SVP test to apply vacuum pressure at the subsequent level on the basis of time rather than the completion of consolidation in each stage. Compared with TVP, the SVP method required longer to complete and exhibited a lower initial drainage rate. However, the accumulated drainage volume and final settlement were larger. Changes in the soil consolidation state, moisture content, drainage, and pore water pressure during step vacuum preloading have been reported. However, the correlation between these properties has yet to be effectively analyzed. The settlement, drainage rate, pore moisture content and type, pore water pressure, and so on are interrelated and influence one another during reinforcement. Thus, a comprehensive analysis of the change law of various properties should be conducted. In addition, blockage of the drainage pipe is essentially attributable to the low permeability coefficient of soil to a certain extent, impeding water outflow. The change in permeability of soil is one of the important factors affecting the consolidation. In the study of soil treatment by vacuum preloading, the permeability of soil during TVP was evaluated. Indraratna et al. [25] conducted finite element analysis and numerical simulations by using the ABAQUS (ABAQUS, Inc., Providence, Rhode Island, USA) software and presented the conversion formula of permeability and vacuum preloading pressure under axisymmetric and equivalent plane strain conditions. However, with respect to soil permeability, the consolidation of SVP on soil with high clay content is rarely reported. With the aforementioned problems considered, this study uses a five-stage vacuum preloading method to reinforce a coastal dredger fill with high clay content in Tianjin, China. The settlement, drainage rate, moisture content, and pore water pressure during consolidation were monitored in real-time. Moreover, the grain size distribution, void ratio, and permeability testing were conducted, and these characteristics were used to analyze the effectiveness of SVP in reinforcing the dredger fill with high clay content. J. Mar. Sci. Eng. 2020, 8, 38 3 of 16

J. Mar. Sci. Eng. 2019, 7, x FOR PEER REVIEW 3 of 16 2. Soil Properties 2. Soil Properties The dredger fill used in this test was from the southeast bank of the Nangang Industrial Zone, The dredger fill used in this test was from the southeast bank of the Nangang Industrial Zone, Binhai New Area, Tianjin, China (E 117◦37025.84”, N 38◦42016.03”), a large-scale land reclamation Binhai New Area, Tianjin, China (E 117°37′25.84″, N 38°42′16.03″), a large-scale land reclamation site site that started in 2016. Sampling was conducted in November 2017. The site was originally at the that started in 2016. Sampling was conducted in November 2017. The site was originally at the junction of a shallow sea and tidal flat and was artificially filled into land (Figure1). Currently, the junction of a shallow sea and tidal flat and was artificially filled into land (Figure 1). Currently, the site is relatively flat in general, with slight fluctuations in several portions. The dredger fill has not site is relatively flat in general, with slight fluctuations in several portions. The dredger fill has not been been vacuum-preloaded, but the surface layer has been dried and exhibits a certain strength. Two soil vacuum-preloaded, but the surface layer has been dried and exhibits a certain strength. Two soil samples were selected in this area for mixing; a spade was used to remove the hard surface layer from samples were selected in this area for mixing; a spade was used to remove the hard surface layer from the soil. The soil taken by digging contained moisture (30–80% and up to 120%) and was transferred the soil. The soil taken by digging contained moisture (30–80% and up to 120%) and was transferred into a bag for sealing. into a bag for sealing.

FigureFigure 1. Sampling sites of of the the soft soft soil soil used used in in this this study. study.

TheThe basicbasic physicalphysical propertiesproperties of the dredger fill fill sample are are listed listed in in Table Table 11.. Figure 22 showsshows the particleparticle distributiondistribution curve curve of of the the soil soil sample sample after after desalting desalting treatment, treatment, which which contained contained clay (52.06%) clay (52.06%) and andsilt (47.91%). silt (47.91%). The soil The was soil classified was classified as lean clay as leanin accordance clay in accordance with the Standard with the Practice Standard for Classification Practice for C = 24.17 Classificationof Soils for Engineering of Soils for Purposes Engineering (ASTM Purposes D2487-17). (ASTM The D2487-17). nonuniformity The coefficient nonuniformity u coeffi andcient = the curvature coefficient Cc 2.06 indicate that the soil is well-graded and that the particle size Cu = 24.17 and the curvature coefficient Cc = 2.06 indicate that the soil is well-graded and that the particledistribution size distributionis continuous. is continuous.The crystal Theminerals crystal in mineralsthe soil samples in the soil were samples characterized were characterized by X-ray bydiffraction X-ray di ffanalysisraction (Table analysis 2). (TableThe results2). The show results that show the primary that the minerals primary of minerals the soil ofmainly the soil consist mainly of consistquartz ofand quartz calcite, and the calcite,secondary the minerals secondary have minerals high illite-smectite have high illite-smectitelayer content, and layer the content, hydrophilicity and the hydrophilicityis high, indicating is high, that indicating the water could that the not water be easily could drained not be from easily the drained soil. from the soil.

Table 1. Basic physical properties of the soil sample. Table 1. Basic physical properties of the soil sample. Specific Gravity Plastic Limit (%) Liquid Limit (%) Soluble Salt Content (%) pH Specific Gravity Plastic Limit (%) Liquid Limit (%) Soluble Salt Content (%) pH 2.74 26 45 1.756 7.12 2.74 26 45 1.756 7.12

Table 2. Mineral composition of the soil sample. Table 2. Mineral composition of the soil sample. Secondary (Clay) Mineral (%) Primary Mineral (%) Secondary (Clay) MineralIllite– (%) Primary Mineral (%) Alkali Kaolinite Illite Chlorite Quartz Plagioclase Hornblende Calcite Muscovite Illite–Smectite Alkali Kaolinite Illite ChloriteSmectite layer Quartz Plagioclase Hornblende Calcite Muscovite Feldspar layer Feldspar 3.2 8.0 4.16 16.64 36.1 10.3 0.7 15.8 2.2 2.9 3.2 8.0 4.16 16.64 36.1 10.3 0.7 15.8 2.2 2.9

J. Mar. Sci. Eng. 2019, 7, x FOR PEER REVIEW 4 of 16 J. Mar. Sci. Eng. 2020, 8, 38 4 of 16 J. Mar. Sci. Eng. 2019, 7, x FOR PEER REVIEW 4 of 16

Figure 2. Grain size distribution of the dredger soil. Figure 2. GrainGrain size size distribution distribution of of the the dredger dredger soil. 3. Testing Methods and Procedure 3. Testing Methods and Procedure 3. Testing Methods and Procedure 3.1. Consolidation Testing 3.1. Consolidation Testing 3.1. Consolidation Testing As shown in Figure 3, the experimental device mainly included three parts: the main body of As shown in Figure3, the experimental device mainly included three parts: the main body of the the Assettler, shown the indrainage Figure 3,and the air experimental extraction systdeviceem, mainlyand the included observation three system. parts: theThe main settler body was of a settler, the drainage and air extraction system, and the observation system. The settler was a cylindrical thecylindrical settler, themetal drainage bucket andmeasuring air extraction 70 cm insyst diameterem, and and the 50 observation cm in height. system. The inner The settlersurface was of the a metal bucket measuring 70 cm in diameter and 50 cm in height. The inner surface of the settler was cylindricalsettler was metal wrapped bucket with measuring two layers 70 of cm plastic in diameter film to andisolate 50 cmand in seal height. the soil. The The inner drainage surface andof the air wrapped with two layers of plastic film to isolate and seal the soil. The drainage and air extraction settlerextraction was wrappedsystem included with two a layersdrain board,of plastic a drainfilm to pipe, isolate a drain and sealbarrel, the andsoil. aThe vacuum drainage pump. and Theair system included a drain board, a drain pipe, a drain barrel, and a vacuum pump. The observation extractionobservation system system included consisted a drain of a settlement board, a drain scale pipe,(observation a drain ofbarrel, the settlement), and a vacuum a vacuum pump. gauge The system consisted of a settlement scale (observation of the settlement), a vacuum gauge (real-time observation(real-time detection system consisted of vacuum of degreea settlement), and ascale pore (observation water pressure of thegauge. settlement), a vacuum gauge detection of vacuum degree), and a pore water pressure gauge. (real-time detection of vacuum degree), and a pore water pressure gauge.

Figure 3. Schematic of the experimental apparatus. Figure 3. Schematic of the experimental apparatus. In the actual consolidationFigure 3. process Schematic of of the the project,experimental the apparatus. dredger fill undergoes self-weight consolidationIn the actual while consolidation drying, that is,process soil-water of the separation project, andthe self-weightdredger fill sedimentationundergoes self-weight [23]. In thisconsolidation indoorIn the simulationactual while consolidation drying, test, the that soil is, samplesprocess soil-water withoutof the sepa desaltingprrationoject, andthe were self-weightdredger dried, crushed,fill sedimentation undergoes prepared, self-weight [23]. according In this consolidationtoindoor a moisture simulation contentwhile test, drying, of the 120%, soilthat fully samples is, soil-water stirred, without and sepa soaked desrationalting for and were 24 self-weight h. dried, The mud crushed, sedimentation was then prepared, poured [23]. according into In this the indoortestto a barrel. moisture simulation With content this test, method ofthe 120%, soilof samples fully refilling, stirred, without the and settlement des soakedalting for and were 24 change h.dried, The processmudcrushed, was of prepared,then the dredgerpoured according into fill can the toetestff aectively moisture barrel. be With studiedcontent this of[method26 120%,]. fullyof refilling, stirred, the and settlement soaked for and 24 h.change The mud process was of then the poured dredger into fill thecan testeffectively barrel. Withbe studied this method [26]. of refilling, the settlement and change process of the dredger fill can effectively be studied [26].

J. Mar. Sci. Eng. 2020, 8, 38 5 of 16

Under the influence of self-weight stress, soil precipitation and water were separated. In this stage, both drain valves were closed. The soil-water separation stage ended when the soil and water surfaces became stable. Drain valve 1 was then opened for the self-weight sedimentation stage. The water in the settler was discharged along the drainage pipe, and the drainage volume was determined. The self-weight sedimentation stage ended when the soil surface coincided with the water surface, the height of the mud surface did not change within 48 h, and the pore water pressure stabilized at a certain value. The soil-water separation and self-weight sedimentation stages are referred to as the self-weight consolidation stage. After the self-weight consolidation stage, SVP testing was started. In this stage, the plastic film on the inner surface of the settler separated the soil from the outside; thus, only the drainage pipe could discharge the water in the settler. Drain valve 1 was closed, drain valve 2 was opened, and the water discharged from the water collector was calculated. The vacuum pump was used to apply vacuum pressure to the whole system. Five levels (10, 20, 40, 60, and 80 kPa) of vacuum pressure were applied based on the testing time sequence. The consolidation stability standard of the dredger fill under the action of vacuum suction in each stage was a soil surface settlement of less than 1 mm within 48 h, meaning that the consolidation of soil in this stage of pressure was completed. Even if increasing consolidation time had no further effect on consolidation [27], the next level of vacuum pressure could be applied.

3.2. Monitoring Testing During testing, the settlement, drainage rate, and pore water pressure of the soil in the test cylinder were monitored. At the end of each stage of vacuum preloading, the moisture content and permeability of the soil samples were determined. At the end of SVP, the grain size distribution and void ratio at different positions of the soil were tested. The results of the sampling and test scheme are listed in Table3. The inclinometer was embedded at the edge of the settler to measure the horizontal displacement. The horizontal displacement of the upper surface at the edge was considered as the total maximum horizontal displacement. Owing to the high moisture content and poor consolidation effect of the soil in the vacuum phase of SVP at 10 kPa, sampling and permeability testing were not conducted in the vacuum phase of the technique at 10 kPa. The soil at the bottom of the settler was also neither sampled nor evaluated to prevent disturbance. The TST-55 model penetrameter (Nanjing Soil Instrument Factory Co., Ltd., Jiangsu Province, China) was used for penetration testing (Figure4). At the end of each stage, samples were taken with a 61.8 mm diameter and 40 mm high ring knife, in accordance with the results listed in the sampling table. The ring cutter with a penetration sample was put into the saturator for saturation and then placed in the permeameter to adjust the water head for testing.

Table 3. Soil sampling and testing schemes.

Test stage Post-20 kPa Post-40 kPa Post-60 kPa Post-80 kPa Time (d) 68.13 83.69 99.98 116.37 Test Moisture Moisture Moisture Moisture Permeability Permeability Permeability Permeability Types Content Content Content Content UC √ √ √ √ √ √ √ √ LC √ – √ – √ – √ √ UE √ √ √ √ √ √ √ √ LE √ – √ – √ – √ √ J. Mar. Sci. Eng. 2020, 8, 38 6 of 16

J. Mar. Sci. Eng. 2019, 7, x FOR PEER REVIEW 6 of 16

Figure 4. (a,b) TST-55 permeameter, (c) soil samples saturated in a saturator, and (d) soil samples after Figure 4. (a,b) TST-55 permeameter, (c) soil samples saturated in a saturator, and (d) soil samples after permeability testing. permeability testing.

4. Test Test Results and Discussion

4.1. Settlement Settlement Figure 55 showsshows thethe processprocess ofof settlementsettlement inin eacheach stagestage andand TableTable4 4shows shows the the specific specific data data of of settlement in each stage. When When mud mud was was poured into th thee settler, the initial height of the soil surface from the bottom bottom plate plate was was 25.67 25.67 cm. cm. The The total total water water volume volume did did not not change change during during soil-water soil-water separation, separation, and theand water the water surface surface settlement settlement was 1.39 was cm. 1.39 This cm. small This settlement small settlement resulted from resulted the flattening from the flatteningof the plastic of filmthe plasticby the filmmud by and the the mud adjustment and the adjustmentof the mud ofpo thesition, mud which position, rendered which the rendered mud fully the filled mud in fully the settler.filled in However, the settler. at However,the beginning at the of beginning this stage, of the this soil stage, surface the soil decreased surface decreasedrapidly and rapidly gradually and stabilizedgradually owing stabilized to the owing accumulation to the accumulation of soil partic of soilles and particles filling and between filling particles. between particles.The soil itself The exhibitedsoil itself uniform exhibited particle uniform size particle distribution, size distribution, resulting in good resulting gradation in good and gradationsatisfactory and filling satisfactory between soilfilling particles. between The soil soil particles. surface Thesank soil considerably, surface sank reaching considerably, 8.04 cm. reaching After soil-water 8.04 cm. Afterseparation, soil-water self- weightseparation, sedimentation self-weight occurred. sedimentation In this occurred.stage, the water In this surface stage, thedecreased water surfacerapidly decreasedand overlapped rapidly again and withoverlapped the soil againsurface with after the a soilcertain surface time, after owing a certain to the time,opening owing of the to thedrain opening valve. ofConsequently, the drain valve. the waterConsequently, surface was the no water longer surface higher was than no the longer soil surfac highere; thanthus, thethe soilchange surface; in water thus, surface the change was no in longer water recorded.surface was When no longer the soil recorded. surface and When water the soilsurface surface coincided, and water the surfacesettlement coincided, and time the showed settlement a linear and relationship,time showed and a linear soil relationship,drainage continued. and soil The drainage settlement continued. of the Thesoil settlementsurface was of 1.33 the soilcm during surface self- was weight sedimentation, and all occurred after the soil surface coincided with the water surface. This

J. Mar. Sci. Eng. 2020, 8, 38 7 of 16

1.33J. Mar. cm Sci. during Eng. 2019 self-weight, 7, x FOR PEER sedimentation, REVIEW and all occurred after the soil surface coincided with 7 of the 16 water surface. This occurrence indicated that the water above the soil surface was discharged before theoccurrence coincidence, indicated and thethat water the water between above the the soil soil pores surface was was discharged discharged after before the coincidence, the coincidence, resulting and inthe soil water settlement. between the soil pores was discharged after the coincidence, resulting in soil settlement.

Figure 5. Time-dependent settlement curves in didifferentfferent stages.stages. Table 4. Soil sampling and testing schemes. Table 4. Soil sampling and testing schemes. Water/Soil Observation Accumulated Settlement in Various StagesEnd Time End Time (d) WaterSurface/Soil Height Surface Accumulated Settlement in Various Stages ObservationPosition Position Settlement (cm) Each Stage (cm) (d) Height(cm) (cm) Settlement (cm) Each Stage (cm) Soil-waterSoil-water WaterWater surface surface 24.28 24.28 1.39 1.39 1.39 1.39 9.74 9.74 separationseparation Soil surfaceSoil surface 17.63 17.63 8.04 8.04 8.04 8.04 Self-weightSelf-weight WaterWater surface surface 16.3 16.3 9.37 9.37 7.98 7.98 36.56 36.56 sedimentationsedimentation Soil surfaceSoil surface 16.3 16.3 9.37 9.37 1.33 1.33 Soil surfaceSoil (center) surface 14.1 11.57 2.2 10 kPa 52.43 14.1 11.57 2.2 10 kPa 52.43 Soil surface(center) (edge) 13.9 11.77 2.4 Soil surface Soil surface (center) 13.9 13.6 11.77 12.07 2.4 0.5 20 kPa 68.13 (edge) Soil surface (edge) 12.9 12.77 1 Soil surface 13.6 12.07 0.5 Soil surface(center) (center) 12.88 12.79 0.72 40 kPa20 kPa 83.69 68.13 Soil surface Soil surface (edge) 12.9 12.1 12.77 13.57 1 0.8 (edge) Soil surface (center) 12.4 13.27 0.48 60 kPa 99.98 Soil surface Soil surface (edge) 12.88 11.4 12.79 14.27 0.72 0.7 40 kPa 83.69 (center) Soil surfaceSoil (center) surface 12.1 13.57 0.3 80 kPa 116.37 12.1 13.57 0.8 Soil surface(edge) (edge) 10.9 14.77 0.5 Soil surface 12.4 13.27 0.48 60 kPa 99.98 (center) During vacuum preloading, the smallerSoil surface settlement of the soil surface at the center and the larger 11.4 14.27 0.7 settlement of the soil surface at the edge (edge)were observed. The differential settlement increased with an Soil surface increase in vacuum pressure. Moreover, the total settlement12.1 at the center 13.57 was 4.20 0.3cm, the total 80 kPa 116.37 (center) settlement at the edge was 5.40 cm, andSoil the surface maximum differential settlement was 1.20 cm. Figure 5 10.9 14.77 0.5 presents the settlement curve of the three-stage(edge) (at 20, 40, and 80 kPa) vacuum preloading conducted by Yuan et al. [23]. The soil used for testing in the current study exhibited similar basic properties as thoseDuring reported vacuum by Yuan preloading, et al. [23], the which smaller was settlement attributed of to the the soil similar surface locations at the centerof both and types the of larger soil. settlementThe vacuum-preloaded of the soil surface soil atexhibited the edge a were consistent observed. trend The in di settlementfferential settlement under different increased loading with anmethods. increase The in two vacuum loading pressure. techniques Moreover, also effectivel the totaly settlementconsolidated at thethe centersoil. Compared was 4.20 cm,with the the total test settlementreported by at Yuan the edge et al. was [23], 5.40 the cm, test and in the the maximum current study differential required settlement a longer was settlement 1.20 cm. time Figure but5 presentsexhibited the a smoother settlement settlement curve of the curve. three-stage Moreover, (at 20,the 40,maximum and 80 kPa) differential vacuum settlement preloading was conducted smaller bybetween Yuan etthe al. center [23]. Theand soilthe edge used than for testing that obtained in the current by Yuan study et al. exhibited A maximum similar differential basic properties settlement as between the center and the edge obtained by Yuan et al. [23] was 1.45 cm and a maximum differential settlement to horizontal settlement distance ratio was 0.0483. Compared with this result, the ratio determined in the present study was smaller—that is, 0.04.

J. Mar. Sci. Eng. 2020, 8, 38 8 of 16 those reported by Yuan et al. [23], which was attributed to the similar locations of both types of soil. The vacuum-preloaded soil exhibited a consistent trend in settlement under different loading methods. The two loading techniques also effectively consolidated the soil. Compared with the test reported by Yuan et al. [23], the test in the current study required a longer settlement time but exhibited a smoother settlement curve. Moreover, the maximum differential settlement was smaller between the center and the edge than that obtained by Yuan et al. A maximum differential settlement between the center and the edge obtained by Yuan et al. [23] was 1.45 cm and a maximum differential settlement to horizontal settlement distance ratio was 0.0483. Compared with this result, the ratio determined in the present study was smaller—that is, 0.04. Moreover, under vacuum pressure, the displacement of the soil toward the central drainage pipe resulted in a higher soil surface at the center than the edge. Horizontal displacement occurred in each stage of vacuum pressure treatment. The higher the vacuum pressure, the larger the differential settlement and the greater the cumulative horizontal displacement. By applying vacuum pressure in stages, soil displacement was entirely distributed in sections, and the uneven settlement of soil was delayed. Mesri [28] performed traditional vacuum loading on various soft clay and muddy sediments; the horizontal displacement of the surface soil and the surface settlement ratio at the center ranged from 0.2 to 0.5, with an average of 0.36. In this test, the cumulative maximum horizontal displacement and the cumulative surface settlement ratio at the center were 0.09, 0.26, 0.22, 0.25, and 0.29, respectively. The average ratio was 0.22 and the maximum ratio was 0.29. The average and maximum ratios were smaller than the values obtained in the study by Mesri [28], confirming that the loading method of this test can reduce horizontal displacement.

4.2. Consolidation Degree In this study, the logarithmic curve method was used to estimate the final settlement in each stage and calculate the degree of consolidation [29]. The theoretical general solution of the degree of consolidation is as follows: βt U = 1 α e− , (1) − · where U is the degree of consolidation, e is the natural constant, and α and β are parameters under different drainage paths and different consolidation conditions. Whether vertical drainage, horizontal outward drainage, or horizontal central drainage can be used, only the values of α and β vary. The degree of consolidation is defined as S U = t , (2) S ∞ where St is the settlement at time t and S is the final settlement. (t1, S1), (t2, S2), and (t3, S3) are ∞ any three points satisfying t t = t t in the measured settlement curve. These three points are 3 − 2 2 − 1 substituted into Equation (1) and Equation (2), respectively, to derive the following equations:

 βt1  S1 = S (1 α e− )  ∞ − · βt  S2 = S (1 α e− 2 ) (3)  ∞ − ·  βt3 S3 = S (1 α e− ) ∞ − · Further, the following is obtained:

2 S S1S3 S3(S2 S1) S2(S3 S2) S = 2 − = − − − (4) ∞ 2S2 S1 S3 (S S ) (S S ) − − 2 − 1 − 3 − 2 The calculated final settlement in each stage is substituted into Equation (2), and the degree of consolidation in each stage is listed in Table5. The degree of consolidation in each stage exceeded 90%, which verifies the effectiveness of SVP. J. Mar. Sci. Eng. 2020, 8, 38 9 of 16

Table 5. Settlement and consolidation degree in different stages.

Self-Weight 10 kPa 20 kPa 40 kPa 60 kPa 80 kPa Consolidation Settlement at the center (cm) 1.33 2.20 0.50 0.72 0.48 0.30 Consolidation degree at the center 98.50% 95.65% 92.60% 98.23% 98.63% 96.77% Settlement at the edge (cm) 1.33 2.40 1.00 0.80 0.70 0.50 Consolidation degree at the edge 98.50% 99.45% 90.90% 93.46% 91.58% 98.04% J. Mar. Sci. Eng. 2019, 7, x FOR PEER REVIEW 9 of 16

4.3. Drainage Drainage Rate The drain valve was opened at the end of the soil– soil–waterwater separation stage—that is, at the beginning of the self-weightself-weight sedimentation stage; thus, the drainage rate was measuredmeasured fromfrom thisthis time.time. The drainage rates in different different stages are shown in Figure6 6.. InIn thethe initialinitial stagestage ofof self-weightself-weight sedimentationsedimentation and vacuum preloading, the drainagedrainage rate immediately reached its maximum and then decreased. The peak value of the drainagedrainage raterate duringduring self-weightself-weight sedimentation was considerably greater than that during vacuum vacuum preloading. preloading. The The reason reason is isthat that the the soil soil was was deposited deposited during during soil-water soil-water separation separation and theand distance the distance between between the soil the surface soil surface and the and water the surface water surfacewas 6.28 was cm, 6.28 allowing cm, allowing direct contact direct between contact thebetween water the above water the above soil surface the soil and surface the drainage and the pi drainagepe. After pipe. the drainage After the va drainagelve was valveopened, was the opened, water wasthe water discharged was discharged rapidly, resulting rapidly, from resulting the gravitatio from the gravitationalnal potential potentialenergy. With energy. a decrease With a decreasein distance in betweendistance the between soil surface the soil and surface the water and thesurface, water the surface, drainage the rate drainage decreased rate to decreased a relatively to astable relatively area andstable continued area and to continued decrease toslowly. decrease In the slowly. initial In stage the initialof vacuum stage pressure of vacuum treatment, pressure the treatment, water in the watersoil was in forced the soil to wasdischarge forced under to discharge an external under force, an with external a large force, flow with rate a and large a large flow drainage rate and rate. a large In eachdrainage stage rate. of vacuum In each stage pressure of vacuum treatment, pressure the treatment,bonding force the bondingbetween force the betweensoil and thethe soilwater and was the watergradually was balanced gradually with balanced vacuum with pressure vacuum over pressure time, thereby over time, reducing thereby the reducingdrainage and the drainagedrainage andrate [30].drainage At the rate end [30 of]. each At the stage, end ofthe each drainage stage, rate the drainageapproached rate 0. approached This finding 0. implied This finding that the implied bonding that forcethe bonding between force the soil between and water the soil was and nearly water balanced was nearly with balanced the vacuum with pressure the vacuum in the pressure entire settler in the andentire that settler soil drainage and that soilwas drainageimpeded. was At the impeded. initial stage At the of initial vacuum stage pressure of vacuum treatment pressure with treatment vacuum withpressure vacuum of 80 pressure kPa, the of peak 80 kPa, value the of peak the valuedraina ofge the rate drainage remained rate high, remained demonstrating high, demonstrating that there occurredthat there no occurred blockage no of blockage drainage of pipes. drainage pipes.

FigureFigure 6. Time-dependentTime-dependent curve of drainage rate in the different different stages of treatment.

4.4. Moisture Content At the end of each stage, the change in moisture content in soil at different positions (UE, UC, LE, and LC, these locations are shown in Figure 3) was evaluated (Figure 7). During consolidation, free water was discharged first, followed by bound water. To a certain extent, the total bound moisture content adsorbed by the soil mainly consisting of clay could be characterized by its liquid limit, and the maximum moisture content of bound water in clay was close to the liquid limit [31]. In addition, the plastic limit resulted from either cavitation or air entry, preventing the water phase from acting as a continuum within the soil thread [32]. During consolidation, the moisture content in soil spanned the liquid limit and the plastic limit, which could be regarded as the drainage of soil mass from free water to loosely bound water. As shown in Figure 7, moisture content decreased sharply and then gradually. The rate of change in moisture content was particularly evident on the upper layer. In the treatment

J. Mar. Sci. Eng. 2020, 8, 38 10 of 16

4.4. Moisture Content At the end of each stage, the change in moisture content in soil at different positions (UE, UC, LE, and LC, these locations are shown in Figure3) was evaluated (Figure7). During consolidation, free water was discharged first, followed by bound water. To a certain extent, the total bound moisture content adsorbed by the soil mainly consisting of clay could be characterized by its liquid limit, and the maximum moisture content of bound water in clay was close to the liquid limit [31]. In addition, the plastic limit resulted from either cavitation or air entry, preventing the water phase from acting as a continuum within the soil thread [32]. During consolidation, the moisture content in soil spanned the liquid limit and the plastic limit, which could be regarded as the drainage of soil mass from free J.water Mar. Sci. to Eng. loosely 2019, bound7, x FOR water.PEER REVIEW As shown in Figure7, moisture content decreased sharply and 10 then of 16 gradually. The rate of change in moisture content was particularly evident on the upper layer. In stagesthe treatment with a vacuum stages with pressure a vacuum of 10 pressure and 20 kPa, of 10 the and moisture 20 kPa, thecontent moisture approached content the approached liquid limit, the whichliquid could limit, be which regarded could as be the regarded transformation as the transformation of the drainage of form the drainage from free form water from to bound free water water. to Thebound moisture water. content The moisture on the content upper onlayer the was upper always layer lower was always than that lower on thanthe lower that on layer, the lowerindicating layer, thatindicating the transformation that the transformation of free water of free discharged water discharged from the fromupper the layer upper to layerloosely to looselybound boundwater occurredwater occurred before before that from that fromthe lower the lower layer. layer. Similarl Similarly,y, the theconversion conversion of offree free water water discharged discharged from from thethe center center to to loosely loosely bound bound water water occurred occurred before before that that from from the the edge. edge. Near Near the the end end of of the the treatment treatment stagestage with with a a vacuum vacuum pressure pressure of of 80 80 kPa, kPa, the the soil soil mo moistureisture content content in in the the settler settler approached approached the the plastic plastic limitlimit and and the the degree degree of of drainage drainage was was maximized maximized without without gas gas inflow. inflow. This This finding finding also also indicated indicated that that thethe loosely loosely bound bound water water was was basically basically discharged, discharged, co confirmingnfirming that that the the drainage drainage capacity capacity of of SVP SVP was was highlyhighly effective. effective.

FigureFigure 7. 7. MoistureMoisture content content in in different different locations locations at at the the end end of of each each stage. stage.

DuringDuring self-weight self-weight sedimentation, sedimentation, no no vacuum vacuum pressure pressure was was applied, applied, and and the the water water was was discharged byby water water seepage seepage itself. itself. After After the the soil soil surface surface coinci coincidedded with with the the water water surface, surface, the the free free water water on on the the upperupper layer layer no no longer longer exhibited exhibited the the trend trend of of transverse transverse flow flow because because of of the the absence absence of of an an upper upper pressure pressure effect,effect, and and the free water among particles tended to to flow flow vertically under the the effect effect of gravitational potentialpotential energy. energy. Thus, Thus, the the soil soil moisture moisture content content at at the the center center and and the the edge edge showed showed no no difference, difference, andand both both were were smaller smaller than than that that at atthe the bottom. bottom. For the For moisture the moisture content content of the of bottom the bottom layer, the layer, water the atwater the atcenter the center easily easily flowed flowed laterally laterally from from the thedrai drainagenage pipe pipe because because of ofthe the effect effect of of gravitational gravitational potentialpotential energy energy of of the the soil soil and and water water above above and and the the sh shortort seepage seepage path path close close to tothe the drainage drainage pipe pipe at theat the center. center. Thus, Thus, thethe moisture moisture content content at the at thecenter center was was slightly slightly less less than than the thethat that at the at theedges. edges. In all In stages of vacuum pressure treatment, the moisture content formed a constant gradient difference, ω ω ω ω and the moisture content gradient was consistently UC < UE < LC < LE . During vacuum pressure consolidation, water flowed downward and toward the center under the action of gravitational potential and vacuum pressure potential. The equipotential surface in the settler was funnel-shaped, which was low at the center and high at the edge. After the consolidation settlement, the soil moisture content in the settler was equal.

4.5. Pore Water Pressure Pore water pressure gauges were buried at the bottom of the center and at the edge of the settlement bucket to measure the pore water pressure in different stages in real-time. Under the influence of temperature and air pressure, the measured pore water pressure fluctuated slightly but did not affect the overall result. The dissipation of pore water pressure when vacuum pressure was

J. Mar. Sci. Eng. 2020, 8, 38 11 of 16 all stages of vacuum pressure treatment, the moisture content formed a constant gradient difference, and the moisture content gradient was consistently ωUC < ωUE < ωLC <ωLE. During vacuum pressure consolidation, water flowed downward and toward the center under the action of gravitational potential and vacuum pressure potential. The equipotential surface in the settler was funnel-shaped, which was low at the center and high at the edge. After the consolidation settlement, the soil moisture content in the settler was equal.

4.5. Pore Water Pressure Pore water pressure gauges were buried at the bottom of the center and at the edge of the settlement bucket to measure the pore water pressure in different stages in real-time. Under the

J.influence Mar. Sci. Eng. of temperature 2019, 7, x FOR PEER and air REVIEW pressure, the measured pore water pressure fluctuated slightly but 11 of did 16 not affect the overall result. The dissipation of pore water pressure when vacuum pressure was applied appliedduring testingduring intesting the current in the current study andstudy in and the studyin the bystudy Yuan by etYuan al. [ 23et ]al. is [23] illustrated is illustrated in Figure in Figure8. Both 8. Bothtests tests exhibited exhibited similar similar trends trends in pore in pore water water pressure pressure when when the the vacuum vacuum pressure pressure changed. changed. As As shown shown in inFigure Figure8, the8, the pore pore water water pressure pressure was was consistently consistently lower lower at at the the center center than than at at the the edge, edge, suggestingsuggesting thatthat when the the distance distance of of the the soil soil to to the the drainage drainage pipe pipe was was smaller, smaller, a better consolidation effect effect was observed. The The dissipation dissipation rate rate of of pore pore water water pressure pressure was lower in the current study than that in the study by Yuan etet al.al. [23[23].]. Similarly,Similarly, the the di differentialfferential dissipation dissipation rate rate of of pore pore water water pressure pressure between between the thecenter center and and the the edge edge after after the the test test were were also also lower. lower.

FigureFigure 8. Dissipation 8. Dissipation of pore of water pore pressure water pressure by step vac byuum step preloading vacuum preloading (SVP) with (SVP)different with loading different modes. loading modes. The change process of pore water pressure in this test is shown in Figure 9. In the initial stage of soil–waterThe change separation, process the of soil pore particles water flocculated pressure in and this sank, test is and shown water in moved Figure9 upward. In the initialrelatively. stage The of drainsoil–water valve separation,was not opened, the soil and particles the total flocculated water volume and sank,in the and test waterbarrel movedremained upward constant. relatively. However, The whendrain valvethe soil-water was not opened,separation and particles the total flocculated water volume and sank, in the the test pores barrel in remained the soil increased constant. relatively, However, andwhen the the effective soil-water seepage separation areas particleslargely varied flocculated when and water sank, seepage the pores occurred. in the soilTherefore, increased in the relatively, initial stageand the of soil–water effective seepage separation, areas the largely permeability varied when per un waterit area seepage was higher occurred. than the Therefore, pore water in thepressure, initial resultingstage of soil–water in an increase separation, in pore thewater permeability pressure at per the unit beginning area was of higherthe test. than This the effect pore was water reflected pressure, in theresulting pore water in an pressure increase in in the pore initial water stage pressure of soil–wa at theter beginning separation, of which the test. was This nearly eff ectdouble was that reflected of its valuein the before pore water the process. pressure The in clear the initialsoil surface stage then of soil–water appeared. separation, The soil and which water was continued nearly doubleto separate, that andof its the value pore before water thepressure process. decreased The clear slightly. soil surfaceDuring thenself-weight appeared. sedimentation, The soil and the water water continuedabove the soilto separate, surface was and discharged the pore water outward pressure and the decreased total wate slightly.r volume During in the self-weightsettler began sedimentation, to decrease before the thewater coincidence above the of soil soil surface surface was and discharged water surface. outward Thus, and the thegravity total above water the volume soil at in the the bottom settler beganof the settler was reduced and the drainage rate was larger relatively, resulting in an obvious decrease in pore water pressure. After this coincidence, the free water in the soil began to drain and the moisture content gradually decreased. However, the drainage rate was small, so the decrease in pore water pressure was small. During vacuum preloading, the pore water pressure generally exhibited a downward trend. The change in pore water pressure was not only related to the distance between drainage pipes but also to the type of water discharged from the soil [33]. As shown in Figure 8, the pore water pressure decreased steadily during the entire process when the center and the edge were in the self-weight consolidation stage and in the treatment stage with a vacuum pressure of 10 kPa. However, in the center of the final vacuum pressure stage, that is, when pressure levels of 20, 40, and 60 kPa were applied while at the edge of the final vacuum pressure stage, that is, when pressure levels of 40 and 60 kPa were applied, the pore water pressure increased slightly, which was similarly observed during the three-stage vacuum preloading test conducted by Yuan and Lei [9].

J. Mar. Sci. Eng. 2020, 8, 38 12 of 16 to decrease before the coincidence of soil surface and water surface. Thus, the gravity above the soil at the bottom of the settler was reduced and the drainage rate was larger relatively, resulting in an obvious decrease in pore water pressure. After this coincidence, the free water in the soil began to drain and the moisture content gradually decreased. However, the drainage rate was small, so the J.decrease Mar. Sci. Eng. in pore 2019, water7, x FOR pressure PEER REVIEW was small. 12 of 16

FigureFigure 9. 9. Time-dependentTime-dependent pore pore water water pressure pressure curves curves in in the the different different stages of treatment.

SectionDuring 4.4 vacuum of this preloading, paper indicates the pore that water the period pressure of generallytransition exhibited from free a water downward at the trend. center The to looselychange bound in pore water water occurred pressure earlier was not than only the relatedtransition to theat the distance edge. With between this observation, drainage pipes the butcurrent also studyto the concludes type of water that an discharged increase in from pore the water soil pr [33essure]. As was shown a distinct in Figure phen8,omenon the pore occurring water pressure during dischargedecreased of steadily loosely duringbound thewater. entire By processbackward when reasoning, the center when and the the center edge of were the insettler the self-weightwas in the treatmentconsolidation stage stage with and a vacuum in the treatment pressure stageof 20 with kPa, a transformation vacuum pressure into of loosely 10 kPa. bound However, water in thefor dischargecenter of theoccurred, final vacuum while the pressure edge was stage, in the that trea is,tment when stage pressure a with levels vacuum of 20, pressure 40, and of 60 40 kPa kPa. were In Figureapplied 7, whilein the atlater the stages edge ofof thevacuum final vacuumpressure pressure treatment stage, (40, 60, that and is, 80 when kPa), pressure the drainage levels rate of 40 reached and 60 0kPa in advance—that were applied, theis, when pore waterthe whole pressure soil mass increased in the slightly, settler was which in wasthe period similarly of loosely observed bound during water the dischargingthree-stage vacuumoutward, preloading the soil mass test in conducted the later stag by Yuane could and not Lei continue [9]. to drain outward. Combining FiguresSection 5 and 4.4 7, whenof this the paper vacuum indicates pressure that was the period40 and 60 of kPa, transition from the from time free when water the at drainage the center rate to reachedloosely bound0 until water the end occurred of each earlier stage, than the the soil transition continued at theto settle edge. by With 2.2 this and observation, 2.5 mm, respectively. the current Whenstudy concludesthe drainage that rate an increasereached in 0 porein the water treatmen pressuret stage was with a distinct a vacuum phenomenon pressure occurring of 80 kPa, during the settlementdischarge of looselythe soil boundwas stable, water. and By the backward settlement reasoning, curve of the when soil the no centerlonger ofchanged. the settler Therefore, was in inthe the treatment treatment stage stages with with a vacuum a vacuum pressure pressure of 20 of kPa, 40 and transformation 60 kPa, drainage into loosely stability bound was observed water for earlierdischarge in the occurred, settler, compared while the with edge the was settlement in the treatment stability of stage the asoil with mass. vacuum In this pressuretype of soil, of 40which kPa. wasIn Figure no longer7, in drained the later but stages still ofshow vacuums a small pressure amount treatment of settlement, (40, 60, the and soil 80pore kPa), was the correspondingly drainage rate reduced,reached 0which in advance—that caused the soil is, pore when water the wholepressure soil to massslightly in rise. the settler was in the period of loosely boundOn water the other discharging hand, with outward, the gradual the soil decrease mass in in the the later loosely stage bound could notwater continue in the tosoil, drain the outward.vacuum pressureCombining could Figures not further5 and7 drain, when the the water vacuum at a certai pressuren time. was Owing 40 and to pore 60 kPa, reduction, from the the time bonding when force the betweendrainage soil rate and reached water 0was until slightly the end higher of each than stage, the vacuum the soil pressure continued at this to settle time. byThe 2.2 loosely and 2.5 bound mm, waterrespectively. partly back When flowed the to drainage create a ratebalance reached between 0 in the the vacuum treatment suction stage and with water a vacuum absorption pressure capacity of of80 the kPa, soil; the this settlement process ofcaused the soil the was pore stable, water and pre thessure settlement in the soil curve to rise. of the Consequently, soil no longer part changed. of the looselyTherefore, bound in the water treatment flowed stages back withto balance a vacuum the two pressure forces, of prompt 40 anding 60 kPa,an increase drainage in stability pore water was pressureobserved in earlier soil from in the another settler, angle. compared At the with end theof the settlement treatment stability stage with of the a vacuum soil mass. pressure In this of type 80 kPa,of soil, the which loosely was bound no longer water drained was drained, but still and shows no return a small flow amount occurred, of settlement, resulting thein no soil increase pore was in porecorrespondingly water pressure. reduced, which caused the soil pore water pressure to slightly rise. In the current study, the decrease in pore water pressure was low in the self-weight consolidation stage and the treatment stage, with a vacuum pressure of 10 kPa, but was large in the treatment stage with a vacuum pressure of 20 and 80 kPa.

4.6. Grain Size Distribution and Void Ratio

J. Mar. Sci. Eng. 2020, 8, 38 13 of 16

On the other hand, with the gradual decrease in the loosely bound water in the soil, the vacuum pressure could not further drain the water at a certain time. Owing to pore reduction, the bonding force between soil and water was slightly higher than the vacuum pressure at this time. The loosely bound water partly back flowed to create a balance between the vacuum suction and water absorption capacity of the soil; this process caused the pore water pressure in the soil to rise. Consequently, part of the loosely bound water flowed back to balance the two forces, prompting an increase in pore water pressure in soil from another angle. At the end of the treatment stage with a vacuum pressure of 80 kPa, the loosely bound water was drained, and no return flow occurred, resulting in no increase in pore water pressure. In the current study, the decrease in pore water pressure was low in the self-weight consolidation stage and the treatment stage, with a vacuum pressure of 10 kPa, but was large in the treatment stage with a vacuum pressure of 20 and 80 kPa.

4.6. Grain Size Distribution and Void Ratio Determination of the grain size distribution and void ratio of the soil was conducted after SVP. The results are presented in Table6. Desalting was not conducted during the grain size distribution test to measure the actual grain size distribution of soil in various positions. Subsequently, the clay contents of the soil at the center and at the edge after the test were measured. A higher clay content and a lower silt content were found in the soil at the center than at the edge. Moreover, the soil at the center was more compact than that at the edge, as indicated by the void ratios. These differences were attributed to vacuum pressure. However, soil in various locations generally showed highly similar grain size distribution and void ratios, suggesting that after treatment with SVP, the soil mass was relatively even.

Table 6. Grain size distribution and void ratio in different positions of the soil at the end of SVP.

UC UE LC LE Grain Size Clay Silt Sand Clay Silt Sand Clay Silt Sand Clay Silt Sand Distribution (%) 33.07 66.05 0.88 32.97 66.49 0.54 34.46 64.11 1.43 33.39 65.99 0.62 Void ratio 0.54 0.55 0.54 0.57

4.7. Permeability Coefficient Figure 10 presents the change in soil permeability in different stages of vacuum pressure treatment in the settler. In the logarithmic diagram, permeability in each stage of SVP was nearly linear. The change in the permeability coefficient appeared to decrease by an order of magnitude for each vacuum pressure stage (Figure 10). As vacuum preloading progressed, the permeability coefficient at the center was consistently smaller than that at the edge, and the difference between the permeability coefficient at the center and the edge increased. The difference in the permeability coefficient was attributable to the horizontal displacement of the soil mass toward the central drainage pipe during reinforcement, which rendered the soil sample structure at the center more dense than that at the edge. Meanwhile, under the action of vacuum pressure, fine particles in the soil migrated to the center of the drainage pipe, affecting the permeability of the soil. As permeability decreased, the permeability coefficients of the soil at the center and the edge were always close. At the end of the treatment stage, with a vacuum pressure of 80 kPa, the difference in permeability between the center and the edge was the largest but did not exceed an order of magnitude. The decrease in permeability was attributable to the reduction in porosity. The close permeability coefficient between the center and the edge indicates that the drainage capacity of the center was close to the drainage capacity of the edge. The combined results on the similarity of the grain size distribution and void ratios of soil in various locations reveal that no blockage was observed near the drainage pipe at the center regardless of the displacement of fine particles. This observation confirms that SVP can effectively prevent drainage pipe blockage. J. Mar. Sci. Eng. 2019, 7, x FOR PEER REVIEW 13 of 16

Determination of the grain size distribution and void ratio of the soil was conducted after SVP. The results are presented in Table 6. Desalting was not conducted during the grain size distribution test to measure the actual grain size distribution of soil in various positions. Subsequently, the clay contents of the soil at the center and at the edge after the test were measured. A higher clay content and a lower silt content were found in the soil at the center than at the edge. Moreover, the soil at the center was more compact than that at the edge, as indicated by the void ratios. These differences were attributed to vacuum pressure. However, soil in various locations generally showed highly similar grain size distribution and void ratios, suggesting that after treatment with SVP, the soil mass was relatively even.

Table 6. Grain size distribution and void ratio in different positions of the soil at the end of SVP.

UC UE LC LE Grain Size Clay Silt Sand Clay Silt Sand Clay Silt Sand Clay Silt Sand Distribution (%) 33.07 66.05 0.88 32.97 66.49 0.54 34.46 64.11 1.43 33.39 65.99 0.62 Void ratio 0.54 0.55 0.54 0.57

4.7. Permeability Coefficient Figure 10 presents the change in soil permeability in different stages of vacuum pressure treatment in the settler. In the logarithmic diagram, permeability in each stage of SVP was nearly linear. The change in the permeability coefficient appeared to decrease by an order of magnitude for each vacuum pressure stage (Figure 10). As vacuum preloading progressed, the permeability coefficient at the center was consistently smaller than that at the edge, and the difference between the permeability coefficient at the center and the edge increased. The difference in the permeability coefficient was attributable to the horizontal displacement of the soil mass toward the central drainage pipe during reinforcement, which rendered the soil sample structure at the center more dense than that at the edge. Meanwhile, under the action of vacuum pressure, fine particles in the soil migrated to the center of the drainage pipe, affecting the permeability of the soil. As permeability decreased, the permeability coefficients of the soil at the center and the edge were always close. At the end of the treatment stage, with a vacuum pressure of 80 kPa, the difference in permeability between the center and the edge was the largest but did not exceed an order of magnitude. The decrease in permeability was attributable to the reduction J. Mar.in Sci. porosity. Eng. 2020 The, 8, 38 close permeability coefficient between the center and the edge indicates that the14 of 16 drainage capacity of the center was close to the drainage capacity of the edge.

FigureFigure 10. 10.Permeability Permeability coefficient coefficient in in each each stage. stage.

5. Conclusions On the basis of previous research on step vacuum preloading (SVP), this study presents an experimental study of the SVP (10, 20, 40, 60, and 80 kPa) process in a coastal dredger fill in Tianjin, China. This study evaluated the consolidation effect of this test on a dredger fill with high clay content and analyzed the effective prevention of drain pipe blockage. The loading method was found to effectively consolidate and drain dredger fill with high clay content. The following conclusions are drawn based on this case study:

1. An increase in the vacuum pressure stage can lengthen the process of settlement but can also reduce the differential settlement. In each stage of vacuum preloading, the degree of consolidation can exceed 90%. With SVP, the horizontal displacement of soil can be reduced. As vacuum pressure stages increase, the volume of horizontal displacement decreases. 2. Owing to the different forces and drainage paths of pore water at different positions, the moisture content in soil near the drain pipe is always less than that in soil far away from the drain pipe in the vacuum preloading stage; moreover, moisture content in soil on the upper layer is less than that on the bottom layer. The relationship of moisture content, Atterberg limit, drainage rate, and pore water pressure indicates that free water is discharged from the soil mass in the treatment stage, with a vacuum pressure of 10 kPa. Loosely bound water is discharged from the center in the 20 kPa vacuum pressure stage. The soil mass of the settler is in the stage of loosely bound water discharge, and all loosely bound water is discharged at the end of the treatment stage with a vacuum pressure of 80 kPa. 3. With an increase in the vacuum pressure stage, the dissipation rate of pore water pressure tends to be even, regardless of the position. The pore water pressure decreases during the entire process, and the dissipation at the center is greater than that at the edge. At the later stage of some vacuum pressure treatment, the pore water pressure slightly rises, but no discharge of loosely bound water occurs. This phenomenon can be attributed to the compaction of the soil and the backflow of loosely bound water. 4. SVP is an effective means of avoiding drainage pipe blockage, as verified by the consolidation effect of the soil in each stage, as well as by the increase in drainage rate in the initial stage of vacuum preloading with 80 kPa. Similarities between the center and the edge, with respect to the grain size distribution, void ratio, and permeability coefficient of the soil, indicate that the distance from the drainage pipe causes no horizontal variation in fine particles and that the consolidation effect exerted is even. J. Mar. Sci. Eng. 2020, 8, 38 15 of 16

Author Contributions: Conceptualization, X.Y.; methodology, W.S.; software, J.L.; formal analysis, J.L.; resources, H.C.; data curation, W.S.; writing—original draft preparation, J.L.; writing—review and editing, J.L.; funding acquisition, X.Y. and H.C. All authors have read and agreed to the published version of the manuscript. Funding: The presented work is supported by the National Natural Science Foundation of China (No. 41602285, No. 51890914), Science and Technology Development Program of Jilin Province, China (No. 20180520064JH). Acknowledgments: The authors would like to thank the anonymous reviewers for their constructive comments and suggestions. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Tang, M.; Shang, J.Q. Vacuum preloading consolidation of Yaoqiang Airport runway. Geotechnique 2000, 50, 613–623. [CrossRef] 2. Chai, J.C.; Carter, J.P.; Hayashi, S. Vacuum consolidation and its combination with embankment loading. Can. Geotech. J. 2006, 43, 985–996. [CrossRef] 3. Saowapakpiboon, J.; Bergado, D.T.; Voottipruex, P.; Lam, L.G.; Nakakuma, K. PVD improvement combined with surcharge and vacuum preloading including simulations. Geotext. Geomembr. 2011, 29, 74–82. [CrossRef] 4. Indraratna, B.; Kianfar, K.; Rujikiatkamjorn, C. Radial consolidation model incorporating the effects of vacuum preloading and non-Darcian flow. Geotechnique 2013, 63, 1060–1073. 5. Voottipruex, P.; Bergado, D.T.; Lam, L.G.; Hino, T. Back-analyses of flow parameters of PVD improved soft Bangkok clay with and without vacuum preloading from settlement data and numerical simulations. Geotext. Geomembr. 2014, 42, 457–467. [CrossRef] 6. Quang, N.D.; Giao, P.H. Improvement of soft clay at a site in the Mekong Delta by vacuum preloading. Geomech. Eng. 2014, 6, 419–436. [CrossRef] 7. López-Acosta, N.P.; Espinosa-Santiago, A.L.; Pineda-Núñez, V.M.; Ossa, A.; Mendoza, M.J.; Ovando-Shelley, E.; Botero, E. Performance of a test embankment on very soft clayey soil improved with drain-to-drain vacuum preloading technology. Geotext. Geomembr. 2019, 47, 618–631. [CrossRef] 8. Cai, Y.; Qiao, H.; Wang, J.; Geng, X.; Wang, P.; Cai, Y. Experimental tests on effect of deformed prefabricated vertical drains in dredged soil on consolidation via vacuum preloading. Eng. Geol. 2017, 222, 10–19. [CrossRef] 9. Lei, H.Y.; Lu, H.B.; Liu, J.J.; Zheng, G. Experimental study of the clogging of dredger fills under vacuum preloading. Int. J. Geomech. 2017, 17, 04017117. [CrossRef] 10. Wang, J.; Gao, Z.Y.; Fu, H.T.; Ding, G.Y.; Cai, Y.Q.; Geng, X.Y.; Shi, C.X. Effect of surcharge loading rate and mobilized load ratio on the performance of vacuum–surcharge preloading with PVDs. Geotext. Geomembr. 2019, 47, 121–127. [CrossRef] 11. Zhu, W.; Yan, J.; Yu, G. Vacuum preloading method for land reclamation using hydraulic filled slurry from the sea: A case study in coastal China. Ocean Eng. 2018, 152, 286–299. [CrossRef] 12. Yan, Y.G.; Dong, Z.L.; Yang, C.B.; Cheng, P.S.; Qiu, Q.C. Experimental study on effect of grain size distribution on improvement of dredger fill. Chin. J. Geotech. Eng. 2011, 33, 1775–1779. 13. Wang, J.; Ni, J.F.; Cai, Y.Q.; Fu, H.T.; Wang, P. Combination of vacuum preloading and lime treatment for improvement of dredged fill. Eng. Geol. 2017, 227, 149–158. [CrossRef] 14. Liu, F.Y.; Wu, W.Q.; Fu, H.T.; Wang, J.; Hai, J.; Cai, Y.Q.; Hu, X.Q.; Yuan, W.X.; Lou, X.M. Application of flocculation combined with vacuum preloading to reduce river-dredged sludge. Mar. Georesour. Geotechnol. 2019, 1–10. [CrossRef] 15. Sun, Z.H.; Gao, M.J.; Zhou, W. Dewatering effect of vacuum preloading incorporated with electro-osmosis in different ways. Dry. Technol. 2016, 35, 38–45. [CrossRef] 16. Sun, Z.H.; Gao, M.J.; Zhou, W. Assessment of vacuum preloading of soft sediment applied to various drainage schemes. Soil Mech. Found. Eng. 2014, 50, 255–261. [CrossRef] 17. Zhang, L.; Hu, L.M. Laboratory tests of electro-osmotic consolidation combined with vacuum preloading on kaolinite using electrokinetic geosynthetics. Geotext. Geomembr. 2019, 47, 166–176. [CrossRef] 18. Yuan, X.Q.; Wang, Q.; Sun, Y.; Xia, Y.B.; Chen, H.E.; Song, J. Pore distribution characteristics of dredger fill during hierarchical vacuum preloading. J. Jilin Univ. (Earth Sci. Ed.) 2012, 42, 169–176. J. Mar. Sci. Eng. 2020, 8, 38 16 of 16

19. Song, J. Laboratory Simulation Test and PFC3D Numerical Analysis of High Clay Dredger Fill in the Consolidation Process of Step Vacuum Preloading. Master’s Thesis, Jilin University, Changchun, China, 2011. 20. Wu, Y.J.; Yang, J.B.; Zhang, M.X. Study of impact of vacuum loading mode on dredger fill flow mud consolidation effect and soil particles moving. Rock Soil Mech. 2013, 34, 2129–2135. 21. Yan, H. Aging Characteristics of Material, Structure Changes and Settlement Prediction of Dredger Fill in the Consolidation Process of Step Vacuum Preloading. Master’s Thesis, Jilin University, Changchun, China, 2015. 22. Liu, J.J.; Lei, H.Y.; Zheng, G.; Zhou, H.Z.; Zhang, X.L. Laboratory model study of newly deposited dredger fills using improved multiple-vacuum preloading technique. J. Rock Mech. Geotech. Eng. 2017, 9, 924–935. [CrossRef] 23. Yuan, X.Q.; Wang, Q.; Lu, W.X.; Zhang, W.; Chen, H.E.; Zhang, Y. Indoor simulation test of step vacuum preloading for high-clay content dredger fill. Mar. Georesources Geotechnol. 2017, 36, 83–90. [CrossRef] 24. Fang, Y.G.; Guo, L.F.; Huang, J.W. Mechanism test on inhomogeneity of dredged fill during vacuum preloading consolidation. Mar. Georesources Geotechnol. 2019, 37, 1007–1017. [CrossRef] 25. Indraratna, B.; Rujikiatkamjorn, C.; Sathananthan, I. Analytical and numerical solutions for a single vertical drain including the effects of vacuum preloading. Can. Geotech. J. 2005, 42, 994–1014. [CrossRef] 26. Lei, H.Y.; Wang, X.; Chen, L.; Huang, M.; Han, J. Compression characteristics of ultra-soft clays subjected to simulated staged preloading. KSCE J. Civ. Eng. 2016, 20, 718–728. [CrossRef] 27. Qiu, C.L.; Niu, F.; Yan, S.W.; Ji, Y.C. Laboratory tests on consolidation of dredger fill under vacuum load. J. Tianjin Univ. (Sci. Technol.) 2014, 47, 498–503. 28. Mesri, G.; Khan, A.Q. Ground improvement using vacuum loading together with vertical drains. J. Geotech. Geoenviron. Eng. 2012, 138, 680–689. [CrossRef] 29. Wang, J.; Huang, G.; Fu, H.T.; Cai, Y.Q.; Hu, X.Q.; Lou, X.M.; Jin, Y.W.; Hai, J.; Ni, J.F.; Zou, J. Vacuum preloading combined with multiple-flocculant treatment for dredged fill improvement. Eng. Geol. 2019, 259, 105194. [CrossRef] 30. Qiu, C.L.; Yan, P.W.; Sun, L.Q.; Ji, Y.C. Effect of varying void on consolidation of dredger fill under vacuum preloading. Rock Soil Mech. 2013, 34, 631–638. 31. Tang, L.S.; Chen, H.K.; Song, J. Process of pore pressure diffusion in saturated clay soil and impact of adsorbed water. Geosci. J. 2016, 20, 649–665. [CrossRef] 32. Haigh, S.K.; Vardanega, P.J.; Bolton, M.D. The plastic limit of clays. Géotechnique 2013, 63, 435–440. [CrossRef] 33. Chen, Y.J.; Song, J.; Xia, Y.B.; Wang, Q.; Peng, X.L.; Sang, W.F. Laboratory experimental studies of pore water pressure in dredger fill consolidation with vacuum preloading. J. Eng. Geol. 2010, 18, 703–708.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).