draft · draft 2 abstract: prefabricated vertical drains (pvds) have been used extensively to...

Draft

Improving consolidation of dredged slurry by vacuum preloading using PVDs with varying filter pore sizes

Journal: Canadian Geotechnical Journal

Manuscript ID cgj-2018-0572.R2

Manuscript Type: Article

Date Submitted by the Author: 06-Mar-2019

Complete List of Authors: Wang, Jun; College of Civil Engineering and Architecture, Wenzhou University, Yang, Yongli; Wenzhou UniversityFu, Hongtao; Wenzhou UniversityCai, Yuanqiang; Zhejiang University, Hu, Xiuqing; Wenzhou UniversityLou, Xiaoming; Shanghai Geoharbour Construction Group Co., Ltd. ChinaJin, Yawei; Yixing Xintai Geosynthetics Enterprise Limited

Keyword: gradient ratio, vacuum preloading, dredged slurry, filter, pore size

Is the invited manuscript for consideration in a Special

Issue? :Not applicable (regular submission)

https://mc06.manuscriptcentral.com/cgj-pubs

Canadian Geotechnical Journal

Draft

1

Improving consolidation of dredged slurry by vacuum

preloading using PVDs with varying filter pore sizes

Jun Wang1, Yongli Yang2, Hongtao Fu3, Yuanqiang Cai 4, Xiuqing Hu5，

Xiaoming Lou6，Yawei Jin7

1. Professor, College of Architecture and Civil Engineering, Wenzhou University,

Wenzhou, Zhejiang 325035, P.R. China. Email: [email protected]

2. Master, College of Architecture and Civil Engineering, Wenzhou University,

Wenzhou, Zhejiang 325035, P.R. China. Email: [email protected]

3. Lecturer, College of Architecture and Civil Engineering, Wenzhou University,

Chashan University Town, Wenzhou, 325035, P.R. China. E-mail:

[email protected]

4. Professor, Zhejiang University of Technology, Hangzhou, 310014, China.

Tel: +86-15968795938, Fax: +8657786689611

E-mail: [email protected] (Corresponding author)

5. Associate Professor, College of Architecture and Civil Engineering, Wenzhou

University, Chashan University Town, Wenzhou, 325035, P.R. China. E-mail:

[email protected]

6. Professorate Senior Engineer, Shanghai Geoharbour Construction Group Co.,

Ltd. China. E-mail: [email protected]

7. Engineer, Yixing Xintai Geosynthetics Enterprise Limited, Yixing 214200, P.R.

China. E-mail: [email protected]

Page 1 of 38



Draft

2

Abstract: Prefabricated vertical drains (PVDs) have been used extensively to

accelerate the consolidation rate of dredged slurry. While some fine particles from

dredged slurry can easily squeeze through the filter into the drainage channel, many

cannot. As such, these soil particles deposit on the filter surface causing partial

clogging of the drainage path. Although the pore size of filter is recognized asan

important factor that influences PVD clogging, the standards for determining the pore

size of the filter are lacking. To this end, the traditional gradient ratio tests with four

different filter pore sizes were conducted, and the results show that the permeability

of the filter at a given head increases with the increase in the pore size of filter. To

remove the effect of the difference between static hydraulic gradient and vacuum

pressure, the vacuum preloading tests with varying pore sizes of filters were further

conducted. Through these vacuum preloading test, the degree of vacuum, settlement,

pore water pressure, water content, vane shear strength and other parameters of PVDs

with various filter pore sizes were obtained, and the optimal pore size of filter was

determined.

Keyword: gradient ratio; vacuum preloading; dredged slurry; filter; pore size

Page 2 of 38



Draft

3

Introduction

Many coastal regions are experiencing land scarcity owing to rapid growth in

population and economic development. Land reclamation has been widely used to

overcome land shortages in coastal cities (Sun et al. 2018). The main fill material for

land reclamation is mainly consisted of dredged slurry. Unfortunately, dredged slurry

is characterized by high water content, high compressibility, and low permeability

(Chu et al. 2000; Wang et al. 2016), and hardly possesses any bearing capacity.

Therefore, it is imperative to treat dredged slurry to meet the requirements of

infrastructure construction (Shen et al. 2012).

Vacuum preloading, which was originally proposed by Kjellman (1952), is an

effective and economic method for improving soft soil (Chai et al. 2006). Meanwhile,

sand drains are often utilized for drainage in early vacuum preloading projects.

Nowadays, prefabricated vertical drains (PVDs) are preferred to shorten the drainage

path and accelerate the dissipation of pore water pressure (Cai et al. 2018; Wang et al.

2018b and 2019). Furthermore, PVDs are commonly used for improving the

consolidation process of soft soil owing to their quick installation process and low

cost (Chen et al. 2012).

In general, PVD consists of a core and a filter sleeve. The particles of dredged

slurry are very fine and under vacuum suction, they are forced to enter the drainage

path of the PVDs, thereby clogging these PVDs. Therefore, the filter sleeve of PVDs

has a significant effect on the consolidation rate of subsoil (Wang and Li 2016). Chai

Page 3 of 38



Draft

4

and Miura (1999) conducted discharge capacity tests on two types of PVDs using a

triaxial device and found that the discharge capacity decreased considerably with

elapsed time. It was also concluded that the deformation of the PVD filter contributed

less than 20% toward reducing the discharge capacity, and more than 80% of the

reduction was attributed to filter clogging due to fine soil particles entering the PVD

drainage channels. Le et al. (2014) conducted a set of gradient ratio tests and found

that a new type of PVD, known as NPVD, has higher permeability compared to that

of conventional PVD (referred to herein as CPVD) owing to the use of a hydrophilic

filter in the NPVD. Further, Ali et al. (1993) investigated the effects of filter jacket

and core geometry on the longitudinal permeability of PVDs. The results obtained

show that the discharge capacity was influenced by the flexibility of the filter jacket

and core, as well as the geometrical structure of the core. Furthermore, the influence is

time dependent. Xu et al. (2014) also investigated the effect of vacuum loading

gradient on the drainage capacity of PVD. Their results showed that for a vacuum

load with a large loading gradient, the effects of the vacuum consolidation method on

the reinforcement of newly dredged mud were undesirable, and the filter was easily

clogged under this condition.

In terms of the effect of filter pore size on the consolidation rate of dredged

slurry, the pore size of filters should satisfy the following two requirements. On the

one hand, the pore size of filter should be sufficiently small to prevent fine soil

particles from entering the filter and drain. On the other hand, the pore size of the

Page 4 of 38



Draft

5

filter should not be too small as the filter must have sufficient permeability. A suitable

pore size of filter can aid in the formation of an anti-filtration layer on the surface of

the filter and to alleviate PVD clogging in the process of vacuum preloading. The pore

size of a filter is generally maintained within a certain range to optimize the

performance of PVDs. Some criteria to determine the pore size of filter have been

proposed by several researchers (Christopher and Fischer 1992; Luettich et al. 1992).

Palmeira et al. (2002) conducted different types of tests to determine the pore size of

geotextiles. The authors also evaluated the accuracy of the permeability estimation

formula of geotextiles. Koerner et al. (1994) reported that filters can excessively clog

when permeated with leachate over long periods of time and a design equation for

geotextile filters or graded granular soils used in landfill leachate collection systems

was proposed. Koerner et al. (1982) also conducted gradient ratio test to study the

long-term drainage capability of geotextiles, and concluded that it takes

approximately 200 h to stabilize the gradient ratio for dredged sludge with high clay

content. However, these standards are not uniform. A commonly used criterion

proposed by Carroll (1983) is , where O95 is an average size that is

larger than 95% of the filter pores and D85 is the size of the soil particle that makes up

85% of the mass of the soil sample. However, Chu et al. (2004) concluded that the

above criterion for pore size of filter is too conservative for Singapore marine clay.

They proposed a more relaxed criterion as . Thus, it is

Page 5 of 38



Draft

6

imperative to select a suitable pore size of filter to achieve the goals of reinforcement

effect of dredger slurry.

In this study, the effect of filter pore size on the drainage characteristics of PVDs

was investigated. First, a series of gradient ratio tests were conducted at different

hydraulic gradients using various filter pore sizes to determine the permeability of the

filter sleeve. Then, a set of vacuum preloading tests were conducted to evaluate the

effect of PVD filter on the dredged slurry improvement. During the vacuum

consolidation test, vacuum pressure, pore water pressure, settlement, and discharged

water volume were monitored. The water content and vane shear strength were

measured at the end of the test (Sun et al. 2018). Finally, the optimum pore size of

PVD filter for improving consolidation of dredged slurry was determined based on

the gradient ratio and vacuum preloading tests.

Test materials

Properties of test soil

The dredged slurry utilized in this study was obtained from Oufei tideland

reclamation site in Lingkun, Wenzhou, China at a depth of approximately 1.0 m from

the ground surface. It mainly consists of silt and clay with characteristics of high

water content, low shear strength, and high compressibility. Detailed properties of the

test soil are listed in Table 1. The initial void ratio, water content, and liquid and

plastic limits of the soil samples were 2.32, 88, 49.1, and 28.4, respectively. The test

Page 6 of 38



Draft

7

soil was classified as clay with high plasticity. In addition, the soil was extremely soft

with virtually no shear strength.

Figure 1 shows the particle size distribution curve of the dredged slurry. It can be

observed from the figure that clay particles with diameters less than 5 µm constitute

50% of the total mass of soil samples, and the size of 85% of soil particles (D85) was

approximately 23 μm. The size of 95% of soil particles was smaller than 75 μm. This

indicates that the soil mainly consists of fine particles, which can easily cause

clogging of PVDs during vacuum preloading.

PVDs used in the study

In this study, NPVD as shown in Fig. 2(a) was used. NPVDs usually consist of a

core made of polypropylene and a filter sleeve made of non-woven polypropylene

geotextile, in which the filter sleeve is bonded to the core by heat melting to form an

integrated body. At present, NPVDs are widely used in land reclamation projects in

China.

NPVDs are totally different from CPVDs, which is shown in Fig. 2(b). The filter

sleeve of CPVD is separated from the inner core, and is characterized by an uneven

surface. The convex surface is permeable, whereas the other concave surface is

impermeable. In addition, the CPVD has only one drainage channel; therefore,

clogging occurs easily in this type of PVD. The filter of an NPVD is equipped with

hydrophilic fibres, and has double drainage directions. Figure 3 shows scanning

Page 7 of 38



Draft

8

electron microscope (SEM) images of an NPVD filter with different pore sizes of

filters. The weight per unit area of the O95 80-µm, 120-µm, 160-µm, and 320-µm

filters are 105 g, 100 g, 95 g, and 90 g, respectively. It can be observed that it has a

large filter diameter to ensure a higher drainage rate. In addition, the NPVDs is easier

to degrade to eliminate post-construction settlement.

In addition, the pore sizes of most CPVDs filters vary in the range of 20–40 μm

and the pore sizes cannot be adjusted. However, the pore size of the filter of NPVDs

is adjustable and varies within 60–320 μm. Filters with O95 of 80 μm, 120 μm, 160 μm

and 320 μm were employed in this study based on the criterion proposed by Chu et al.

(2004) to determine the optimum pore size for vacuum consolidation.

Gradient ratio test

Test apparatus and materials

The gradient ratio test apparatus mainly consists of a testing cell and peripheral

components, as shown in Fig. 4. The testing cell is made of polymethyl methacrylate

with an internal diameter of 120 mm. The PVD filter was cut into a circular shape

with a diameter of 100 mm and fixed at the bottom of the testing cell as a geofabric.

Soil was placed in the testing cell and on the PVD filter.

The peripheral components mainly consist of piezometric tubes and a constant

head water device. Tube 1 was used to discharge air in the testing cell. Tube 2 was

connected to a water supply device to provide constant water input, whereas tube 7

Page 8 of 38



Draft

9

was at the bottom of the testing cell and was used to discharge water. Tubes 4 and 5

on the side of the testing cell were connected to a piezometer tube to measure water

pressure at varying depths.

Test procedure

A set of gradient ratio tests were conducted to investigate the permeability of

PVD filter. The testing cell was first filled with dredged slurry to a height of 100 mm.

Then, water was poured into the cell slowly to immerse the test soil. The container lid

was then covered to seal the soil sample. The outlet valve above the glass container

was kept open when the water valve of the constant water head device was closed.

The constant head outlet supplies water until the container has been filled with water,

all bubbles have been discharged, and the outlet valve has been closed. After

completing the above steps, the soil sample was kept for 2 to 3 days to attain stable

precipitation based on the scheduled design and arrangement. Finally, the test was

conducted and the test data were recorded.

The amount of water that flows in the soil would be different depending on the

value of hydraulic gradients. In this study, a uniform hydraulic gradient (i) was

employed for the comparison of four types of PVDs filters. The hydraulic gradient

test (i=5) was first carried out until a steady state was attained. Then, the hydraulic

gradient was increased to i=10 to continue the test until it was completed.

Test results and data analysis

Page 9 of 38



Draft

10

In the gradient ratio test, the gradient ratio (GR) is defined as follows:

(1)

where is the hydraulic gradient of geotextile and soil, is the hydraulic gradient

in soil.

Recalling that the hydraulic gradient (i) is the ratio of head difference along the

flow direction ( ) and the length of the flow path ( ) as follows:

(2)

Therefore, the gradient ratio formula can be rewritten as follows:

(3)

where is the head difference of geotextile and soil, is the length of the flow

path of geotextile and soil, is the head difference in soil, and is the length of the

flow path in soil. A higher GR value indicates the clogging around the geotextile is

more serious.

The GR was determined using measured data and Eq. (3), and the variations in

GR with time for different PVD filters with i=5 and i=10 (after 288 h) are shown in

Fig. 5. The plots for i=5 show an increasing trend in all filters before finally achieving

a steady state. Le et al (2014) also reported similar results. This indicates that the filter

is clogged; the smaller the pore size, the more obvious the clogging.

The hydraulic gradient was changed instantaneously from i=5 to i=10 at 288 h.

The same instantaneous increase in the gradient ratio can be observed in all types of

Page 10 of 38



Draft

11

filters when i=10. This indicates that soil particles were removed and accumulate on

the filter surface at a high speed under a high water head, thereby inducing relatively

severe clogging on the filter surfaces. After about 370 hours, the GR of the filter with

O95 320-μm was less than 3 when i=10, indicating less clogging on such a filter

surface based on the gradient ratio criterion proposed by the U. S. Army Corps of

Engineers (ASTM, 2006). Further, the gradient ratios (GRs) of the four filters initially

decreased and then increased with time when i=10. The reasons for this interesting

phenomenon are discussed as follow.

As the hydraulic gradient i increases from 5 to 10 abruptly, the soil particles

rapidly accumulated and concentrated in the filter hole that increases the head

pressure per unit area on the filter. As time increases, water gradually passes through

the filter that reduces the filter clogging. However, for a constant hydraulic gradient,

the force per unit area of filter hole decreases, soil particles gradually move into filter

holes and then continue to cause different levels of filter clogging. Finally, the

gradient ratio reaches the stable state at about 580 h.

The results of the above tests show that the larger pore size of the filter, the better

permeability behaviour the filter has. The optimum pore size of the filter cannot be

determined by the gradient ratio test. As the vacuum pressure played a dominant role

during vacuum consolidation, the performance of the filter during vacuum

consolidation would be distinct from the ones during the gradient test. The hydraulic

gradient varied with the vacuum pressure cannot reflect the clogging effect of fine

Page 11 of 38



Draft

12

particles entering the drainage path of the PVDs. The influence of different pore sizes

of filters on vacuum consolidation will be further described in the following section.

Vacuum preloading model test

Test apparatus and procedure

The apparatus used in this test mainly consists of testing cylinders and an

improved vacuum preloading system. Four types of NPVDs with different pore sizes

of filters (80 μm, 120 μm, 160 μm, and 320 μm) were used in this test. The testing

cylinders were made of polyethylene and their inner diameter and height were 500

mm and 900 mm, respectively; furthermore, their inner walls were polished to reduce

friction. A PVD specimen was preinstalled at the centre of the cylinders, and the

depth of the PVD was up to 900 mm. Two pore pressure transducers were mounted on

an iron shelf to monitor the pore water pressure at the depths of 30 cm and 60 cm,

respectively. A syringe needle inserted into the core of the PVDs and connected to a

vacuum gauge through a vacuum pipe which was used as a vacuum pressure probe.

Two sets of syringe needles were used; one set was positioned along with the central

PVDs to evaluate the vacuum pressure in the PVDs, whereas the other set was

distributed along with the undrained PVDs to evaluate the vacuum pressure in the soil.

Figure 6 shows the layout of the iron rack in detail.

After placing the slurry, the iron rack was positioned at the bottom of the model

box, and the PVDs were connected to the vacuum pump through vacuum tubes.

Page 12 of 38



Draft

13

Subsequently, one layer of geotextile and two layers of geomembrane were placed on

the surface of the slurry to seal it (Chai et al. 2006). The monitoring points of the

vacuum pressure and pore water pressure are shown in Fig. 6. Vacuum pressure was

applied on the slurries as soon as the vacuum pump starts working, and consolidation

commenced. During vacuum consolidation, the pore water pressure, vacuum suction,

volume of extracted water, and surface settlement were measured. The vacuum was

stopped when the settlement tended to be constant. When the test has been completed,

the vane shear strength and water content of the test soil were measured at different

depths.

Evaluation of anti-clogging performance of filter

Vacuum pressure

The vacuum pressure in the PVDs were measured at different depths and plotted

as shown in Fig. 7. It can be observed that the vacuum pressure remained stable until

the end of the test. The pore size of the filter did not affect the vacuum pressure in the

PVDs. In addition, the loss in vacuum pressure along the depth was estimated as 10

kPa/m. Loss in vacuum pressure occurs because the PVDs not only transfer vacuum

pressure but also discharge water in the vacuum preloading process (Chu et al. 2000);

furthermore, the PVDs bend with consolidation of the soil. Therefore, vacuum

pressure loss along the depth occurs frequently.

Figure 8 shows variation of the vacuum pressure with depth in soil. In Fig. 8(a),

vacuum pressure was not observed in the soil at the beginning of the experiment,

Page 13 of 38



Draft

14

indicating that the soil was in a saturated state at that time. As the test continued, the

vacuum pressure in the soil became measurable when the soil changed from saturated

state to unsaturated state (Ming and Zhao 2008). In particular, the vacuum pressure in

the soil that installed PVD with a O95 320-μm filter was first measurable after 30 h,

indicating that the soil reached an unsaturated state (Qiu et al. 2007). Then, the

vacuum pressure was measured in sequence for PVDs with O95 160-μm, 120-μm, and

80-μm filters. It is possible that the drainage channel is unblocked at the beginning of

the experiment, and the O95 320-μm filter yielded fast drainage. At approximately 150

h, the degree of vacuum pressure for the PVD with O95 120-μm filter was higher than

those of the PVDs with O95 160-μm and 320-μm filters. Figure 8(a) also shows a

similar upward trend for each curve. The vacuum pressure in the soil gradually

increased with time, and finally reached stable state. However, the vacuum pressure

for the PVD with O95 80-μm filter was the lowest, followed by those of the PVDs with

O95 320-μm and 160-μm filters, whereas the vacuum pressure for the PVD with O95

120-μm filter was the highest.

The low vacuum pressure for the PVD with 80-μm filter is probably due to the

small pore size of the filter. A large amount of soil particles accumulated on the

surface of the PVD filter, causing clogging and eventually resulting in low

transmission of vacuum pressure. The highest values occurred when PVD with O95

120-μm filter owing to the reduced clogging of the PVD by using appropriate pore

size of filter and formation of a better anti-clogging system (Wang et al. 2018a).

Page 14 of 38



Draft

15

However, because of the PVDs with large pore size of the 160-μm and 320-μm filters,

slight clogging of the holes occurred in the early stage, which is beneficial to drainage.

Later, soil particles gradually entered into the core of the PVDs and blocked the

drainage channels. In addition, soil particles were partly displaced into the filter

surface and blocked the holes of the filter. The soil around the PVDs cannot form a

good anti-clogging layer owing to significant loss of soil particles, thereby causing

serious clogging in the PVDs and hindering the transmission of vacuum pressure.

From Figs. 8(a) and (b), it can be observed that vacuum transfer in the soil layer

worsens as the soil depth increases. The vacuum pressure decreased by 4 kPa as the

soil depth increased. However, the vacuum pressure in the soil did not develop at the

same rate as vacuum transmission at different depths, which is similar to the result

reported by Khan and Mesri (2014).

Volume of extracted water

The water discharged during the test was collected and measured using an

air–water separation flask. The change in the amount of water discharged with respect

to time was then plotted. Variations in the volume of drainage over time for PVDs

with different pore sizes are shown in Fig. 9. It can be observed that a large amount of

drainage was collected at the early stage, and the drainage volume for the early 48 h

reached 60.9% of the total drainage volume. At the early stage, the drainage increased

with the increase in the pore size of the filter. Then, the drainage when PVD with O95

120-μm filter was used was higher than when O95 160-μm and 320-μm filters were

Page 15 of 38



Draft

16

used at the middle stage. Finally, the drainage volume of the O95 120-μm filter was

significantly higher than those of other filters. The mass of water discharged through

the O95 120-μm filter was 37830 g, which is 16.3%, 4.5%, and 9.4% higher than those

of O95 80-µm, 160-µm, and 320-µm filters, respectively. Besides, the permeability of

the PVD filters gradually decreased, the soil porosity became smaller, and the

drainage rate decreased with time (Chai and Miura 1999).

Settlement

Data of surface settlement at four locations in the model boxes were recorded

during vacuum preloading, and the variations of the settlement over time were plotted

as shown in Fig. 10. The curves show a sharp increase at the initial stage, then the

corresponding settling rates began to fall between 100 h and 250 h. After 250 h, the

surface settlement remained almost unchanged, revealing that further settlement

cannot be achieved (Chu et al. 2000). At the end of the test, the final settlements for

PVDs with O95 80-μm, 120-μm, 160-μm, and 320-μm filters were approximately 15.2

cm, 17.6 cm, 17.1 cm, and 16.3 cm, respectively.

At the early stage, the settlement rate increased as the pore size of the filter

increased. However, it is obvious that the settlement rate of O95 120-μm filter

gradually exceeded those of O95 160-μm and 320-μm filters at the middle stage. Then,

at the advanced stage of the vacuum preloading, the settlement rate and the final

settlement for PVD with O95 120-μm filter were larger than those of other pore sizes

of filters.

Page 16 of 38



Draft

17

The degree of consolidation (DOC) is an important parameter to evaluate the

efficiency of vacuum preloading. Zeng et al. (1959) and Chu et al. (2000) calculated

the final settlement of soil as follows:

(4) )()(

)()(

2312

232123

SSSSSSSSSSS f

where Sf is the final settlement, and S1, S2, and S3 are the settlements at time t1, t2, and

t3 respectively. This must meet the following criteria: t1<t2<t3 and t2-t1= t3-t2 (Zeng et

al. (1959). The definition of DOC is given as follows:

(5) f

tt

SSU

where St is the settlement at time t and tU is the DOC.

The calculated degree of consolidation is summarized in Table 2. The degree of

consolidation of the O95 120-μm filter is the highest, reaching 85.22%, whereas that of

O95 80-μm is 70.5%, which is the lowest. It is clear that the pore water pressure of the

O95 120-μm filter is useful in the dissipation of pore water pressure, which improves

soil consolidation.

The following equation can be derived from the equation by Hansbo (1979) for

the consolidation of PVDs:

(6) tSS

S

tf

f

ln

(7)fD

C

e

h2

8

Page 17 of 38



Draft

18

where is the effective diameter of a unit cell of drain and f is the resistance factor

due to the effects of spacing, smear, and well resistance (Bergado et al. 2002).

The values of s

h

kk and f are calculated as follows ( Bergado et al. 2011):

(8)43)ln(

w

en d

Df

(9) )ln()1(w

s

s

hs d

dkkf

(10)w

hr q

klf 2

32π

(11) rsn ffff

where sd is the diameter of the mandrel, wd is the equivalent diameter of the drain,

hk is the coefficient of horizontal permeability of the undisturbed zone, sk is the

coefficient of horizontal permeability of the smear zone, and wq is the discharge

capacity of the drain at hydraulic gradient of 1. The estimated parameters for

consolidation are listed in Table 3.

The horizontal coefficient of consolidation (Ch) for four PVDs with different

pore sizes of filters calculated based on Eq. 5–11 is shown in Fig. 11. The

back-calculated Ch of 6.17 m2/yr was recorded for the O95 80-μm filter. The value was

the lowest compared to the other pore sizes of filters. The back-calculated Ch of the

120-μm filter was the highest and the maximum Ch is 8.29 m2/yr, indicating that it had

a better compression behaviour and soil consolidation in the radial direction.

Pore water pressure

Page 18 of 38



Draft

19

Figures 11(a) and (b) show variation in the pore water pressure of different pore

sizes of filters at different depths. It can be observed that the dissipation of pore water

pressure decreases with the increase in depth during vacuum preloading. At a depth of

30 cm, the pore water pressure of the 120-μm filter dissipated by approximately 43.4

kPa, whereas it dissipated by only 14.6 kPa at a depth of 60 cm. In contrast, the

dissipation of pore water pressure in the deep soil layer was less than that in the

shallow layer, which is attributed to the high level of vacuum transition in shallow

layers (Cai et al. 2017; Qiu et al. 2007).

Figure 11(a) also shows the variation of pore water pressure of different pore

sizes of filters at a depth of 30 cm during vacuum consolidation of soft soil. It can be

observed that a similar trend of the dissipation of pore water pressure of PVDs

occurred for different pore sizes of filters after vacuum activating. In particular, the

dissipation of pore water pressure for the O95 120-μm filter after 150 h was larger than

those for the O95 160-μm and 320-μm filters. The final dissipation of pore water

pressure of the O95120-μm filter was the largest compared to those of other pore sizes

of filters; that is, the pore water pressure values at the end of the test were -32.1 kPa,

-40.9 kPa, -37.7 kPa, and -35.1 kPa for the O95 80-μm, 120-μm, 160-μm, and 320-μm

filters, respectively. It is known that a higher vacuum pressure always contributes to a

higher rate of drainage, which increases the dissipation of pore water pressure

(Indraratna et al. 2010). Thus, the above results are in good agreement with the

vacuum pressure distribution of the soil (Cai et al. 2017).

Page 19 of 38



Draft

20

Water content

To evaluate the effect of different pore sizes of filters on vacuum consolidation

properties of soft soil, the water contents at various distances from PVDs were

measured and plotted as shown in Figs. 12(a) and (b). It can be observed from Figs.

12(a) and (b) that there is an approximate tendency of the water content distributions

for all PVDs with different pore sizes of filters. First, for the four different pore sizes

of filters, the water content at the soil surface is essentially similar, which is mainly

responsible for the relatively steady vacuum pressure at the soil surface and the

excellent drain performance of the horizontal PVDs. Then, the water contents

increased with the increasing depth; this behaviour is attributed to the fact that the

vacuum pressure transferred to the soil decreases with the increase in the drain depth

(Indraratna et al. 2005).

It can be observed from Fig. 12(b) that theaverage water contents for O95 80-μm,

120-μm, 160-μm, and 320-μm filters were of 56.24%, 53.14%, 54.06%, and 54.92%,

respectively at the end of the tests. At depths of 0 cm to 60 cm, the maximum

increment of water content along the depth occurred in the O95 80-μm filter, which

reached 12.99%, whereas the minimum increment of water content along the depth

occurred in the O95 120-μm filter, which is reached 7.7%. This indicates that the water

content of the soil decreased as the O95 120-μm filter PVD is more uniform along the

depth. Compared to the O95 120-μm filter, the water contents of the soil with O95

160-μm and 320-μm filters had increments of 1.73% and 3.35%, respectively. It can

Page 20 of 38



Draft

21

be observed from Figs. 12 (a) and (b) that there is a distinct difference in the

distribution of water contents of soil strengthened along the radial direction using

different pore sizes of filters along the radial direction. The water content of soil close

to the PVDs is generally low.

Vane shear strength

Figures 13(a) and (b) show variations in the vane shear strength of soil reinforced

using PVDs with different pore sizes of filters along the drain depth. As the depth

increases, a similar decreasing trend was observed in the vane shear strength for all

PVDs with four pore sizes of filters. It is considered that a better consolidation was

achieved in shallow layers because a higher vacuum pressure was maintained.

Figure 13(b) shows plots of the shear strength of the vane at different depths and

at a distance of 15 cm from the plate. Average vane shear strengths of 8.8 kPa, 10.43

kPa, 9.97 kPa, and 9.47 kPa were observed for PVDs with O95 80-μm, 120-μm,

160-μm, and 320-μm filters, respectively. The vane shear strength of the O95 120-μm

filter was the highest, whereas that of the 80-μm filter was the lowest. The vane shear

strength of the O95 120-μm filter increased by 1.63 kPa compared to that of the 80-μm

filter. It can be concluded that PVD with the O95 120-μm filter had the best

reinforcement effect.

Conclusions

Page 21 of 38



Draft

22

The effect of the filter pore size on the permeability of the filter sleeve was

investigated by a series of gradient ratio tests at different hydraulic gradients, and the

effect of different pore sizes of PVD filters on soil improvement was further evaluated

with the vacuum preloading tests. The following conclusions are drawn by comparing

the results from the two sets of experiments:

(1) The gradient test showed that the gradient ratio decreased with the increase in

the pore size of the filter. However, the gradient test may not accurately reflect the

clogging of the PVDs due to the fine soil particles during vacuum consolidation. Thus,

further tests with the varying filter pore size during vacuum consolidation is

necessary.

(2) In the early stage of vacuuming preloading, the vacuum pressure and

discharged water volume of the soil increased with the increase in the filter pore size,

and the pore water pressure dissipation was more obvious. However, as the vacuum

preloading continued, the drainage effect of PVD with O95 120-μm filter became more

obvious compared to those of the other three filters of PVDs, and a higher degree of

consolidation and a larger Ch were finally obtained.

(3) The average water contents for O95 80-μm, 120-μm, 160-μm, and 320-μm

filters were 56.17%, 52.21%, 53.79%, and 55.60% respectively at the end of the tests.

This indicates that a better consolidation can be achieved by using the O95 120-μm

filter.

Page 22 of 38



Draft

23

In summary, the pore size of filter was not entirely determined by the gradient ratio

test due to the difference between static hydraulic gradient and vacuum pressure. Thus,

the optimal pore size of the filter was further determined by vacuum preloading test;

that is, the O95 120-μm filter was determined as equivalent to 5.22D85 and suitable for

vacuum consolidation of dredged slurry. This can provide a design guide for

large-scale application of tideland reclamation projects in Wenzhou, China.

Acknowledgments

This research was supported by the National Key R&D Program of China

(2016YFC0800200), the National Nature Science Foundation Projects of China (No.

51778501, No. 51878512，No. 51778500, No. 51622810, and No. 51620105008),

the National Nature Science Foundation Projects of Zhejiang Province (No.

LR18E080001 and No. LY17E080010), the Key Research and development program

of Zhejiang Province (No. 2018C03038), and Scientific and Technological Innovation

Project for College students in Zhejiang Province (No. 2017R426007), which

collectively funded this project.

References

American Society for Testing and Materials (ASTM). 2006. D5101-01 standard test method

for measuring the soil-geotextile system clogging potential by the gradient ratio. West

Conshohocken: ASTM International.

Page 23 of 38



Draft

24

Ali, F.H. 1993. A method to test the performance of a prefabricated vertical drain. Soils and

Foundations. 33(2): 181-187.

Bergado, D.T., Balasubramaniam, A.S., Fannin, R.J., and Holtz, R.D. 2002. Prefabricated

vertical drains (PVDs) in soft Bangkok clay: a case study of the new Bangkok International

Airport project. Canadian Geotechnical Journal, , 39(2): 304-315.

Bergado, D.T., Teerachaikulpanich, N., Saowapakpiboon, J., Kumar, A., and Artidteang, S.

2011. Enhancement of efficiency of prefabricated vertical drains using surcharge, vacuum and

heat preloading. Geosynthetics International, 18(1): 35-47.

Chu, J., Yan, S.W., and Yang, H. 2000. Soil improvement by the vacuum preloading method

for an oil storage station. Geotechnique, 50(6): 625-632.

Chai, J. C., Carter, J. P., and Hayashi, S. 2006. Vacuum consolidation and its combination

with embankment loading. Canadian Geotechnical Journal, 43(10): 985-996.

Cai, Y., Qiao, H., Wang, J., Geng, X., Wang, P., and Cai, Y.2017. Experimental tests on

effect of deformed prefabricated vertical drains in dredged slurry on consolidation via vacuum

preloading: Engineering Geology, 222: 10-19.

Cai, Y., Xie, Z. , Wang, J., Wang, P., and Geng, X. 2018. A new approach of vacuum

preloading with booster PVDs to improve deep 2 marine clay strata. Canadian Geotechnical

Journal, 55(10) : 1359-1371.

Page 24 of 38



Draft

25

Chai, J.C., and Miura, N. 1999. Investigation of factors affecting vertical drain behavior.

Journal of Geotechnical and Geoenvironmental Engineering, 125(3): 216-226.

Chai J.C., Shen S.L., Long P.V., and Chung, S.G. 2012. Geosynthetics for soft ground

improvement-Prefabricated vertical drains, vacuum/Electro-osmotic/Heat preloading. In

Proceedings of Geosynthetics Asia. 5th Asian Regional Conference on Geosynthetics 13-16

December 2012, Bangkok, Thailand, pp. 109-117.

Chu, J., Bo, M.W., and Choa, V. 2004. Practical considerations for using vertical drains in

soil improvement projects. Geotextiles and Geomembranes, 22(1-2): 101-117.

Christopher, B. R., and Fischer, G. R. 1992. Geotextile filtration principles, practices and

problems. Geotextiles and Geomembranes, 11(4-6): 337-353.

Cao, L., Zhu, X., Pan, R., Chen, J., Yin, J., and Li, W. 2014. Experimental research on

clogging characteristic of two types of PVD filters. Rock and Soil Mechanics, 105(2): 1-8.

Carroll, R.G. 1983. Geotextile Filter Criteria. Transportation Research Record 916, USA, pp.

46-53.

Chen, J., Shen, S. L., Yin, Z. Y., Xu, Y. S., and Horpibulsuk, S. 2016. Evaluation of effective

depth of PVD improvement in soft clay deposit: a field case study. Marine Georesources &

Geotechnology, 34(5): 420-430.

Hansbo, S. 1979. Consolidation of clay by band-shaped prefabricated drains. Ground

Engineering, 12(5): 16-25.

Page 25 of 38



Draft

26

Indraratna, B., Sathananthan, I., Rujikiatkamjorn, C., and Balasubramaniam, A.S. 2005.

Analytical and numerical modeling of soft soil stabilized by prefabricated vertical drains

incorporating vacuum preloading. International Journal of Geomechanics, 5(2): 114-124.

Koerner, G.R., Koerner, R.M., and Martin, J.P. 1994. Design of landfill leachate-collection

filters. Journal of Geotechnical Engineering, 120(10), 1792-1803.

Koerner, R.M., and Ko, F.K. Laboratory studies on Long-Term Drainage capability of

Geotextiles. In Proceedings of the 2nd Int. Conf. Geotextiles, Las Vegas, 1-6 August, 1982.

Kjellman, W. 1952. Consolidation of clayey soils by atmospheric pressure. In Proceedings of

Conference on Soil Stabilization, pp. 258-263.

Khan, A. Q., and Mesri, G. 2014. Vacuum distribution with depth in vertical drains and soil

during preloading. Geomechanics and Engineering, 6(4):, 377-389.

Le C., Xu, C., Wu, X.F., and Jin, Y.W. 2014. Experimental research on clogging

characteristics of two types of PVD filters. Rockand Soil Mechanics 9: 2529-2534.

Liu, Z. Y., Zhang, T. H., and Ma, C.W. 2007. Effect of initial hydraulic gradient on one-dime

nsional consolidation of saturated clays. Rock and Soil Mechanics, 28(3): 467-470.

Luettich, S. M., Giroud, J. P., and Bachus, R. C. 1992. Geotextile filter design guide.

Geotextiles and Geomembranes, 11(4-6): 355-370.

Page 26 of 38



Draft

27

Ming, J. P., and Zhao, W. B. 2008. Mechanism of drainage in vacuum preloading. Chinese

Journal of Geotechnical Engineering, 30(12): 1821-1825.

Palmeira, E. M., and Gardoni, M. G. 2002. Drainage and filtration properties of non-woven

geotextiles under confinement using different experimental techniques. Geotextiles &

Geomembranes, 20(2): 97-115.

Qiu, Q. C., Mo, H. H., and Dong, Z. L. 2007. Vacuum pressure distribution and pore pressure

variation in ground improved by vacuum preloading: Canadian Geotechnical Journal, 44(12):

1433-1445.

Sun, L., Zhuang, D., and Yan, S. 2018. Simulation of the vacuum preloading process on

dredged clay by triaxial test. In Proceedings of GeoShanghai 2018 International Conference:

Fundamentals of Soil Behaviours.

Sun, L., Gao, X., Zhuang, D., Guo, W., Hou, J., and Liu, X. 2018. Pilot tests on vacuum

preloading method combined with short and long pvds. Geotextiles and

Geomembranes, 46(2): 243-250.

Shen, S. L., Xu, Y. S., Han, J., and Zhang, J. M. 2012. A ten-year review on the development

of soil mixing technologies in China. In Proceedings of International Conference on Grouting

and Deep Mixing, pp. 343-356.

Xu, B.B., Yamada, S., and Noda, T. 2014. Influence of bending of plastic vertical drain on

consolidation via vacuum preloading. Applied Mechanics & Materials, 644-650: 4826-4830.

Page 27 of 38



Draft

28

Wang, J., Cai, Y., Fu, H., Hu, X., Cai, Y., Lin, H., and Zheng, W. 2018a. Experimental study

on a dredged fill ground improved by a two-stage vacuum preloading method. Soils and

Foundations. 58(3): 766-775.

Wang, J., Cai, Y., Ma J., Chu, J., Fu H., Wang, P., and Jin, Y.2016. Improved vacuum

preloading method for consolidation of dredged clay–slurry fill. Journal of Geotechnical and

Geoenvironmental Engineering. 142 (11), 06016012

Wang, J., Fang, Z., Cai, Y., Chai, J., Wang, P., and Geng, X. 2018b. Preloading using fill

surcharge and prefabricated vertical drains for an airport. Geotextiles and Geomembranes,

46(5):575-585

Wang, J., Gao, Z., Fu, H., Ding, G., Cai, Y., Geng, X., and Shi, C. 2019. Effect of surcharge

loading rate and mobilized load ratio on the performance of vacuum-surcharge with PVD.

Geotextiles and Geomembranes, 47(2) : 121-127.

Wang, J., and Li, T. 2016. Physical and mechanical properties of core and filter membrane for

plastic vertical drains. Chinese Journal of Geotechnical Engineering, 2016.S1

Wang, Z., and Lu, S. Q. 1991. Discussion on silting standard of geotextile filtration

layer. Journal of Hydroelectric Engineering, (3): 55-63.

Zeng, G. X., and Yang, X. L. 1959. Settlement analysis of sand well foundation. Journal of

Zhejiang University, (3): 34-71. (in Chinese)

Page 28 of 38



Draft

Figure captions

Fig. 1. Particle size distribution of the test soil.

Fig. 2. Structure of NPVDs and CPVDs.

Fig. 3. SEM images of the NPVD filter with different pore sizes (×400)

Fig. 4. Schematic diagram of inlet and outlet components in constant head test.

Fig. 5. Variations in GR for different PVD filters at two hydraulic gradients.

Fig. 6. Apparatus for vacuum preloading model test (in mm).

Fig. 7. Vacuum pressure in PVDs measured at different depths.

Fig. 8. Vacuum pressure in soil measured at different depths.

Fig. 9. Volume of drainage for PVDs with four different pore sizes.

Fig. 10. Surface settlement at four locations in the model boxes.

Fig. 11. Pore water pressure of different pore sizes of filters at different depths.

Fig. 12. Soil water content at different depths and different distances from the plate.

Fig. 13 Shear strength of slabs with different depths and different distances from the

plate.

Page 29 of 38



Draft100 10 1

0

20

40

60

80

100

Perc

ent f

iner

by

wei

ght

Diameter(μm)

(5,50)

(75,95)

(23, 85)

Fig. 1. Particle size distribution of the test soil.

(a) NPVDs (b) CPVDs

Fig. 2. Structure of NPVDs and CPVDs.

Page 30 of 38



Draft

(a) 80 μm (b) 120 μm

(c) 160 μm (d) 320 μm

Fig. 3. SEM images of the NPVD filter with different pore sizes (×400)

671 μm 671 μm

671 μm 671 μm

Page 31 of 38



DraftFig. 4. Schematic diagram of inlet and outlet components in constant head test.

0 50 100 150 200 250 300 350 400 450 500 550 6000

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

Gra

dien

t rat

io(G

R)

Time(h)

80m 120m 160m 320m

i=5 i=10

Fig. 5. Variations in GR for different PVD filters at two hydraulic gradients.

Page 32 of 38



Draft

Fig. 6. Apparatus for vacuum preloading model test (in mm).

0 40 80 120 160 200 240 280 3200

20

40

60

80

100

Vac

uum

pre

ssur

e(kP

a)

Time(h)

0cm depth(80μm)

0cm depth(120μm)

0cm depth(160μm)

0cm depth(320μm)

90cm depth(80μm)

90cm depth(120μm)

90cm depth(160μm)

90cm depth(320μm)

Fig. 7. Vacuum pressure in PVDs measured at different depths.

Page 33 of 38



Draft

0 30 60 90 120 150 180 210 240 270 300 3300

5

10

15

20

25

30

35

40

45

50

55

60

Vac

uum

pre

ssur

e(kP

a)

Time(h)

80μm 120μm 160μm 320μm

0 30 60 90 120 150 180 210 240 270 300 3300

5

10

15

20

25

30

35

40

45

50

55

60

Vacu

um p

ress

ure(

kPa)

Time(h)

80μm 120μm 160μm 320μm

(a) 30 cm depth (b) 60 cm depth

Fig. 8. Vacuum pressure in soil measured at different depths.

0 50 100 150 200 250 300 3500

5000

10000

15000

20000

25000

30000

35000

40000

45000

The

mas

s of d

rain

ge w

ater

(g)

Time (h)

80um(PVD) 120um(PVD) 160um(PVD) 320um(PVD)

Fig. 9. Volume of drainage for PVDs with four different pore sizes.

Page 34 of 38



Draft0 30 60 90 120 150 180 210 240 270 300 330 360

220

200

180

160

140

120

100

80

60

40

20

0

The

surf

ace

settl

emen

t(mm

)

Time(h)

80μm(PVD) 120μm(PVD) 160μm(PVD) 320μm(PVD) 80μm(simulated) 120μm(simulated) 160μm(simulated) 320μm(simulated)

Ch=6.17m2/yrCh=7.30m2/yr

Ch=8.22m2/yrCh=8.29m2/yr

Fig. 10. Surface settlement at four locations in the model boxes.

0 30 60 90 120 150 180 210 240 270 300 330

-50

-45

-40

-35

-30

-25

-20

-15

-10

-5

0

5

Time(h)

pore

wate

r pre

ssur

e(Kp

a)

80μm 120μm 160μm 320μm

0 30 60 90 120 150 180 210 240 270 300 330

-14

-12

-10

-8

-6

-4

-2

0

2

4

6

Time(h)

Pore

wat

er p

ress

ure(

kPa)

80μm 120μm 160μm 320μm

(a) Depth 30 cm (b) Depth 60 cm

Fig. 11. Pore water pressure of different pore sizes of filters at different depths.

Page 35 of 38



Draft

48 50 52 54

70

60

50

40

30

20

10

0

-10

Dep

th (c

m)

Water content (%)

80μm120μm160μm320μm

48 50 52 54 56 58 60 62

70

60

50

40

30

20

10

0

-10

Water content (%)

Dept

h (c

m)


(a) Distance 5 cm (b) Distance 15 cm

Fig. 12. Soil water content at different depths and different distances from the plate.

8 9 10 11 12 13 14 15

80

70

60

50

40

30

20

10

0

Dept

h(cm

)

Shear strength(kPa)


6 7 8 9 10 11 12 13 14

80

70

60

50

40

30

20

10

0

Dept

h(cm

)

Shear strength(kPa)


(a) Distance 5 cm (b) Distance 15 cm

Fig. 13 Shear strength of slabs with different depths and different distances from the

plate.

Page 36 of 38



Draft

Table captions

Table 1. Basic physical and mechanical parameters of the test soil sample

Table 2. Degree of consolidation of filter with different pore sizes

Table 3. Parameters for consolidation

、

Page 37 of 38



Draft

Table 1 Basic physical and mechanical parameters of the test soil sample

Water content

Wet density

Specific gravity

Saturation

Void ratio

Plastic limit

Liquid limit

Permeability coefficient

% g/cm3 % % % cm/s88 1.49 2.64 100 2.32 28.4 49.1

Table 2. Degree of consolidation of filter with different pore sizes

O95 80 μm 120 μm 160 μm 320 μm

195U 70.5% 85.22% 84.97% 71.9%

Table 3. Parameters for consolidation

Parameters

wd

(m) (m)

sd

(m)

s

h

kk hk

(m/s)

wq

(m3/s)

Value 0.05 0.85 0.1 2

Page 38 of 38



draft · draft 2 abstract: prefabricated vertical drains (pvds) have been used extensively to...

Documents