draft · draft 2 abstract: prefabricated vertical drains (pvds) have been used extensively to...
TRANSCRIPT
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:
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:
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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)
80μm120μm160μm320μm
(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)
80μm120μm160μm320μm
6 7 8 9 10 11 12 13 14
80
70
60
50
40
30
20
10
0
Dept
h(cm
)
Shear strength(kPa)
80μm120μm160μm320μm
(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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal
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
https://mc06.manuscriptcentral.com/cgj-pubs
Canadian Geotechnical Journal