processes

Article
Investigation of the Porosity Distribution,
Permeability, and Mechanical Performance of
Pervious Concretes
Rentai Liu *, Haojie Liu, Fei Sha, Honglu Yang, Qingsong Zhang, Shaoshuai Shi and Zhuo Zheng
Geotechnical and Structural Engineering Research Center, Shandong University, Jinan 250061, China;
rlhaojie@163.com (H.L.); shafei_97@163.com (F.S.); yanghonglu321@163.com (H.Y.);
zhangqingsong@sdu.edu.cn (Q.Z.); shishaoshuai@sdu.edu.cn (S.S.); 413708318@qq.com (Z.Z.)
* Correspondence: rentailiu@163.com
!"#!$%&'(!
Received: 18 May 2018; Accepted: 20 June 2018; Published: 26 June 2018 !"#$%&'

Abstract: Pervious concrete is a kind of porous and permeable material for pavements and slope
protection projects, etc. In this paper, a kind of pervious concrete was prepared with Portland cement,
silica fume (SF), polycarboxylate superplasticizer (SP), and limestone aggregates. The performance of
concrete, such as its porosity, pore distribution, permeability coefficients, and mechanical properties,
were studied through laboratory tests. The volumetric porosity was measured by the water
displacement method, and the planar porosity and pore size distribution were determined using
image processing technology. A permeameter with a transparent sidewall and an exact sidewall
sealing method were used to measure the permeability coefficients accurately. The results show that
the segregation index and flow values of pastes increased with the increase of SP and water cement
ratio (W/C). The measured porosity (volumetric porosity and planar porosity) of pervious concrete
with a single-size aggregate was closer to the design porosity than that with the blended aggregate.
Compared with the design porosity selected in this study, aggregate size was the main factor
influencing the pore distribution of pervious concrete. The standard deviation of the permeability
coefficient was less than 0.03 under different hydraulic gradients. It was found that the relationships
between the permeability coefficient and volumetric porosity (or effective pore size d50 ) approximately
obey polynomial function. Based on the test results, the optimized parameters were suggested for
practical engineering: W/C of 0.26–0.30; 0.5% SP content; 5% SF content; 15–21% design porosity;
and aggregate sizes of 4.75–9.5 mm and 9.5–16 mm.

Keywords: pervious concrete; permeability coefficient; porosity; pore distribution characteristics;

strength

1. Introduction
With the rapid development of economy and the increment of urban populations, many large-scale
infrastructures and impermeable roads have been built, which has caused more and more serious
environmental problems, such as urban hot island phenomena and the disappearance of groundwater.
Pervious concrete, also referred to as porous ecological concrete, is a porous media material which
consists of cement, water, aggregates, and admixtures, etc. Pervious concrete has a large porosity and
its permeability can reach up to about 2–6 mm/s [1]. Rainwater can penetrate into pervious concrete
quickly, so groundwater can be replenished and stored effectively. It can also reduce city temperatures
and alleviate urban hot island phenomena [2].
Pervious concrete was first invented in Britain in 1824. It was first introduced into the United States
in the mid-1970s and has since been rapidly developed [3]. The Sponge City was introduced in China in

Processes 2018, 6, 78; doi:10.3390/pr6070078 www.mdpi.com/journal/processes

Processes 2018, 6, 78 2 of 14

approximately 2012. Because of its huge environmental benefits, it was quickly and widely emphasized
and generalized in the construction of China’s Sponge City. To date, the properties of pervious concretes
have been studied in many laboratories, including mixing design, conservation method, porosity
characteristics, strength and durability, etc. [4–6]. Compared with traditional or common concrete,
pervious concrete has an amount of porosity and pore structure features, which have been proven to
have a great influence on the properties of pervious concretes [7,8]. Montes et al. [9] recommended
using the water displacement method, based on Archimedes buoyancy principle, to obtain porosity.
Neithalath et al. [8] used image analysis to analyze the pore structure features of pervious concrete.
The permeability of pervious concrete is high due to the presence of high porosity, and the test methods
of the permeability coefficient are especially important [10,11]. Kayhanian et al. [12] used a National
Center for Asphalt Technology’s (NCAT) falling head permeameter to evaluate the permeability
coefficient of pervious concrete. Montes et al. [13] evaluated permeability by using a falling head
permeameter system. The system could register the total volume flowing out of the permeable material
per second. At the same time, Jiang Zhengwu and Yang et al. [13,14] showed that porosity, W/C,
aggregate cement ratio, and aggregate size were the main factors influencing the permeability and
strength of pervious concrete. William et al. [15] studied the effect of vertical porosity distributions on
the permeability coefficient. Because of the low strength of the pervious concrete prepared by common
materials and methods, various admixtures and fine aggregates were introduced to expand the
application scope of pervious concrete. Yang and Jiang [16] used fine aggregates, organic intensifiers,
and optimized mix proportions to produce a high strength pervious concrete. Huang [17] improved
the compressive strength of pervious concrete by using polymer modification and sand.
Although many researches on pervious concretes have been conducted, studies concerning the
properties of paste, as well as the accurate determination of the permeability coefficient and pore
structure, are still limited. The rheological property of cement paste is one of the key factors responsible
for the preparation of good pervious concrete [18,19]; however, it has not been studied sufficiently.
Besides, when the permeability coefficient of pervious concrete is measured, the fluid leakage through
the specimen-permeameter interface is not given significant attention. In this paper, the influences of
polycarboxylate superplasticizer (SP), silica fume (SF), and W/C on the fluidity of cement pastes and
the segregation index of pervious concretes were discussed in detail. The effect of design porosity and
aggregate size on the porosity distribution of pervious concrete was studied. A visual permeameter and
an exact sidewall sealing method were established to measure the permeability coefficients accurately.
The mechanical behavior, such as the compressive strength, was also measured.

2. Materials and Test Methods

2.1. Materials
P.O 42.5 ordinary Portland cement was provided by Sunnsy Group, Jinan, China, and its chemical
composition ingredients are shown in Table 1. Silica fume (SF) was provided by Elkem Company, Oslo,
Norway, and its chemical composition is presented in Table 1. The polycarboxylate superplasticizer
(SP) was provided by Shandong Academy of Building Research, Jinan, China, and the main properties
of NC-J are shown in Table 2. According to the standard of CJJ/T135-200, the literature [13,20] and
requirements of construction projects, limestone gravel was used as an aggregate, and aggregate sizes
followed those listed in Table 3.
Processes 2018, 6, 78 3 of 14

Table 1. Chemical composition of materials.

Chemical Composition: %
Name
SiO2 CaO Al2 O3 Fe2 O3 MgO K2 O Na2 O SO3 TiO3 LOI Others
Cement 23.52 61.47 7.59 2.42 1.70 0.57 0.27 0.65 0.11 0.84 0.86
Silica fume (SF) 97.2 0.42 0.27 0.08 0.55 0.48 0.19 0.51 - - 0.30
* Chemical compositions of cement and silica fume were tested in the laboratory.

Table 2. Properties of Polycarboxylate Superplasticizer (SP).

Solid Content: % Density: g/mL Na2 SO4 : % Cl : %

pH Water Reducing Rate: %
Al2 O3 Na2 O TiO3 Others
30.25
0.67 0.03
0.42 5.01 1.095 32.1
- 0.30
0.27
* Properties of NC-J came from the manufacturer.

Table 3. Gravel sizes.

A B C D E F G
75% 4.75–9.5 50% 4.75–9.5 50% 4.75–9.5 50% 9.5–16
Gravel sizes: mm 4.75–9.5 9.5–16 16–19
25% 9.5–16 50% 9.5–16 50% 16–19 50% 16–19

2.2. Test Methods

2.2.1. Fluidity
The fluidity of cement paste was tested according to GB/T8077-2012. The truncated cone mold
had a top diameter of 36 mm, a bottom diameter of 60 mm, and a height of 60 mm. Different SP
contents (0.3%, 0.5%, and 0.7%), SF contents (5% and 10%), and W/C (0.22–0.32) were used to prepare
the cement pastes. Then, the paste was poured into truncated cone mold which was placed on a glass
pane. Next, the truncated cone mold was removed from the glass pane. The diffusion diameter of the
paste was measured with a ruler in an orthogonal direction after 30 s. The average flow value was the
initial diffusion diameter of the paste.

2.2.2. Mix Proportions

The mix proportion of pervious concrete was designed by the absolute volume method according
to CJJ/T135-2009. The cement (mc ), water (mw ), and admixture (mj ) were calculated as follows:

mg mc mw mj
+ + + +P=1 (1)
rg rc rw rj

where mg is the mass of the aggregate; r g , rc , rw , r j are the apparent densities of aggregate, cement,
water, and admixture, respectively; and P is the design porosity.
In the segregation tests of pervious concrete, the aggregate size was 4.75–9.5 mm and the design
porosity was 15%. The W/C, SP, and SF were shown in the fluidity test.
In the porosity distribution tests, permeability tests, and mechanical performance tests of pervious
concrete, the aggregate sizes were as shown in Table 3. The design porosities were 12%, 15%, 18%,
21%, 24%, and 27%; W/Cs were 0.24–0.32. The SP content was 0.5%. The SF content was 5%. As an
example, the parameters of pervious concrete with a W/C of 0.28 are shown in Table 4.
Processes 2018, 6, x FOR PEER REVIEW 4 of 14

Processes 2018, 6, 78 Table 4. Mix proportions of pervious concrete. 4 of 14

Aggregate Design Aggrega Cement SF (silica Water SP(polycarboxylate superplasticizer)

Size mm Porosity % Table
te kg/m 3
4.kg/m fume) kg/m
Mix3proportions
3
of pervious concrete.
kg/m 3 kg/m3
12 1489.0 546.1 28.7 161.0 2.87
Aggregate 15
Design 1489.0
Aggregate 496.3
Cement 26.1
SF (silica fume) 146.3
Water 2.61 superplasticizer)
SP(polycarboxylate
Size mm Porosity 3 kg/m3 kg/m3 kg/m3 kg/m3
18 % kg/m
1489.0 446.5 23.5 131.6 2.35
4.75–9.5
21
12 1489.0
1489.0 396.7
546.1 20.928.7 116.9
161.0 2.08
2.87
15
24 1489.0
1489.0 496.3
346.9 18.326.1 146.3
102.2 2.61
1.82
18 1489.0 446.5
4.75–9.5 27 1489.0 297.1 15.623.5 131.6
87.6 2.35
1.56
21 1489.0 396.7 20.9 116.9 2.08
12
24 1518.0
1489.0 527.8
346.9 27.818.3 155.6
102.2 2.78
1.82
27
15 1489.0
1518.0 297.1
478.0 25.215.6 140.987.6 1.56
2.51
18
12 1518.0
1518.0 428.2
527.8 22.527.8 126.2
155.6 2.25
2.78
9.5–16
15
21 1518.0
1518.0 478.0
378.4 19.925.2 140.9
111.5 2.51
1.99
18 1518.0 428.2
9.5–16 24 1518.0 328.7 17.322.5 126.2
96.9 2.25
1.73
21 1518.0 378.4 19.9 111.5 1.99
27
24 1518.0
1518.0 289.4
328.7 15.217.3 85.396.9 1.52
1.73
27
12 1518.0
1529.0 289.4
519.5 27.315.2 153.185.3 1.52
2.73
15
12 1529.0
1529.0 469.7
519.5 24.727.3 138.4
153.1 2.47
2.73
15
18 1529.0
1529.0 469.7
419.9 22.124.7 138.4
123.8 2.47
2.21
16–19 18 1529.0 419.9
16–19 21 1529.0 370.1 19.522.1 123.8
109.1 2.21
1.95
21 1529.0 370.1 19.5 109.1 1.95
24
24 1529.0
1529.0 320.4
320.4 16.916.9 94.494.4 1.69
1.69
27
27 1529.0
1529.0 270.6
270.6 14.214.2 79.779.7 1.42
1.42

2.2.3. Molding
2.2.3. Molding and
and Curing
Curing
Concrete specimen
Concrete specimenpreparation
preparationprocedures
procedures were
were as follows.
as follows. Firstly,
Firstly, the aggregate
the aggregate and 70%andwater
70%
water were mixed for 60 s. Then, the 50% cement and admixture were mixed for
were mixed for 60 s. Then, the 50% cement and admixture were mixed for another 60 s. The remaining another 60 s. The
remaining cement and water were added and mixed for 2 min. In order for the material
cement and water were added and mixed for 2 min. In order for the material to uniformly fill in the to uniformly
fill in the
mold, mold,
fresh freshwas
concrete concrete
dividedwasinto
divided into two
two layers intolayers into the
the mold. Eachmold.
layerEach layer ofwas
of concrete concrete was
vibrated
vibrated for 20 s on the vibration table, and then the concrete was compacted with 0.08 MPa
for 20 s on the vibration table, and then the concrete was compacted with 0.08 MPa compressive stress. compressive
stress.molding,
After After molding, the specimen
the specimen wasinplaced
was placed in abox
a curing curing
at 20box at 20a°C
C with with ahumidity
relative relative humidity
of 97%. The of
97%. The specimen
specimen was removed wasfrom
removed fromafter
the mold the mold
24 h, after 24 h,placed
and then and then
in aplaced
curingin a curing
box box for
for 3 days, 3 days,
7 days, or
7 days,
28 days.or 28 days.

2.2.4.
2.2.4. Segregation Index
The segregation
segregation indexindexofofthe
theconcrete
concretewaswasevaluated
evaluatedaccording
accordingtotoASTMC1610;
ASTMC1610; thethe
test device
test is
device
shown
is shownin Figure
in Figure1. The specimen
1. The specimen waswas
produced
produced in ainpolyvinyl chloride
a polyvinyl (PVC)
chloride tube.
(PVC) TheThe
tube. mass of the
mass of
specimen in theintop
the specimen thesection of the column
top section (mt ) and(mthe
of the column massthe
t) and (mbmass
) in the
(mbottom section
b) in the bottomwere measured
section were
after vibrating
measured afterfor 2 min on
vibrating fora 2vibration
min on atable. The segregation
vibration index (S) was
table. The segregation calculated
index (S) wasascalculated
follows: as
follows: 
mb mt
S = 2 ⇥ m b  mt
 ⇥ 100% (2)
S  2 m b + mt   100% (2)
 mb  mt 

200mm

Top 165mm
PVC Pipe

Mid 330mm

165mm
Vibration Table Bottom

Figure 1.
Figure 1. The
The segregation
segregation test
test device
device of
of pervious
pervious concrete.
concrete.
Processes 2018, 6, x FOR PEER REVIEW 5 of 14
Processes 2018, 6, 78 5 of 14
2.2.5. Porosity
The volumetric porosity was measured by the water displacement method. The specimen was
2.2.5. Porosity
dried in a drying box at 110 °C for 24 h, and then put into water for 24 h, so the volume of the water
The volumetric
(Vd) repelled porosity was
by the specimen was measured by the
acquired. The waterofdisplacement
volume open pores was method. The specimen
calculated was
by subtracting
dried in a drying box at 110 C for 24 h, and then put into water for 24 h, so the
Vd from the sample bulk volume (Vb) [21]. The volumetric porosity of the specimen was calculated as volume of the water
(V d ) repelled by the specimen was acquired. The volume of open pores was calculated by subtracting
follows:
Vd from the sample bulk volume (Vb ) [21]. The volumetric porosity of the specimen was calculated as
 V V 
follows: ✓P   b d ◆100% (3)
Vb VbVd 
P= ⇥ 100% (3)
Vb
2.2.6. Pore Structure Features
2.2.6. Pore Structure Features
The pore structure of pervious concrete was analyzed by Image-Pro Plus 6.0. For a concrete
The pore
specimen, threestructure of pervious
to five cylinders wereconcrete
used, andwas the analyzed by Image-Pro
diameter and height of the Plus 6.0. For
specimen a concrete
were 110 mm
specimen, three to five cylinders were used, and the diameter and height of the
and 160 mm, respectively. In the process of the pore structure test, first of all, the specimen specimen were 110 mm
was
and 160 mm, respectively. In the process of the pore structure test, first of
sectioned into 20 mm thick slices. The surface of the specimen was grinded to obtain a flattened all, the specimen was
sectioned
surface. An into 20 mm
image thick slices.
of surface The surface
was captured of the specimen
by a camera, was region
and the outer grinded of to obtain
this imagea was
flattened
cut to
surface.
avoid edge effects. Finally, the image was processed by Image-Pro Plus 6.0 software, andwas
An image of surface was captured by a camera, and the outer region of this image an
cut to avoid edge effects. Finally, the image was processed by Image-Pro Plus 6.0
appropriate threshold was chosen to obtain the pore size distribution and planar porosity, according software, and an
appropriate
to Referencethreshold
[8]. was chosen to obtain the pore size distribution and planar porosity, according to
Reference [8].
2.2.7. Permeability Coefficient
2.2.7. Permeability Coefficient
The permeability coefficient was measured by the permeameter that was designed with
The permeability coefficient was measured by the permeameter that was designed with
transparent plexiglass by our group. As shown in Figure 2, three outlets at different heights of the
transparent plexiglass by our group. As shown in Figure 2, three outlets at different heights of
permeameter were installed to create different hydraulic gradients. The heights of the three outlets
the permeameter were installed to create different hydraulic gradients. The heights of the three outlets
were 10 cm, 15 cm, and 20 cm. For a concrete specimen with same mix proportions, three cylinders
were 10 cm, 15 cm, and 20 cm. For a concrete specimen with same mix proportions, three cylinders
were prepared, and the diameter and height of the specimen were 110 mm and 120 mm, respectively.
were prepared, and the diameter and height of the specimen were 110 mm and 120 mm, respectively.
In the process of the permeability test, the specimen was put into the permeameter, and the sidewall
In the process of the permeability test, the specimen was put into the permeameter, and the sidewall
space between the specimen and the permeameter was sealed with Vaseline and transparent
space between the specimen and the permeameter was sealed with Vaseline and transparent wrapping
wrapping film, so it was easy to know whether there was sidewall leakage. Each specimen was tested
film, so it was easy to know whether there was sidewall leakage. Each specimen was tested three times
three times under different hydraulic gradients to ensure the accuracy of the results. The permeability
under different hydraulic gradients to ensure the accuracy of the results. The permeability coefficient
coefficient of pervious concrete K was calculated by Darcy’s law as follows:
of pervious concrete K was calculated by Darcy’s law as follows:
Qh
K  Qh (4)
K = HAt (4)
H At
where h is the specimen height (cm); H is the hydraulic gradient (cm); and A is the specimen area
where 2
(cm2). h is the specimen height (cm); H is the hydraulic gradient (cm); and A is the specimen area (cm ).

Outlet valve
Plexiglass tube
S1 Outlet

S2
S3
Outlet Plexiglass water tank
Water
Sealing material containers
Beaker Specimen
Bolts

Water drainage
hole
Figure2.
Figure 2. The
The permeability
permeability test
test device
device of
of pervious
pervious concrete.
concrete.
Processes 6, 78
2018, 6,
Processes 2018, x FOR PEER REVIEW 66 of
of 14
14

2.2.8. Mechanical Strength

2.2.8. Mechanical Strength
The flexural strength and compressive strength of the pervious concrete were tested according
The flexural strength and compressive strength of the pervious concrete were tested according
to GB/T 50081-2002. The size of the specimen used in the compressive strength test was 100 mm × 100
to GB/T 50081-2002. The size of the specimen used in the compressive strength test was
mm × 100 mm. The size of the specimen used in the flexural strength test was 100 mm × 100 mm ×
100 mm ⇥ 100 mm ⇥ 100 mm. The size of the specimen used in the flexural strength test was
400 mm. The mechanical strength of the specimen reached 70% of its peak strength during the test,
100 mm ⇥ 100 mm ⇥ 400 mm. The mechanical strength of the specimen reached 70% of its peak
after which the test was stopped. The compressive strength and flexural strength were tested after 3
strength during the test, after which the test was stopped. The compressive strength and flexural
days, 7 days, and 28 days.
strength were tested after 3 days, 7 days, and 28 days.
3. Results,
3. Results, Analysis,
Analysis, and Discussion
and Discussion

3.1. Fluidity Performance

3.1. Fluidity Performance
Figure 3 shows the measured flow
flow values
values of
of pastes
pastes with
with different
different dosages
dosages of
ofSP,
SP,SF,
SF,and
andW/C.
W/C.

0.3% SP+5% SF
300 0.5% SP+5% SF
0.7% SP+5% SF
280
0.5% SP+10% SF
260
Flow value:mm

240
220
200
180
160
140
120
100
0.22 0.24 0.26 0.28 0.30 0.32
W/C
Figure 3.
Figure 3. Flow
Flowvalues
valuesofofpastes
pasteswith different
with W/C
different (water
W/C cement
(water ratio).
cement Two Two
ratio). to three pastespastes
to three were
prepared for each mix design and the figure shows an average of these together with the
were prepared for each mix design and the figure shows an average of these together with the standard
deviation.deviation.
standard

It can be seen in Figure 3 that the flow values linearly increased with increase of the W/C. SP
It can be seen in Figure 3 that the flow values linearly increased with increase of the W/C. SP
was usually added to improve the fluidity of the paste at a low W/C. The use of SP significantly
was usually added to improve the fluidity of the paste at a low W/C. The use of SP significantly
affected the flow values of the pastes; a small increase of SP could make the flow value of paste reach
affected the flow values of the pastes; a small increase of SP could make the flow value of paste reach
the desired value under a lower W/C. SF was introduced to improve strength of pervious concrete.
the desired value under a lower W/C. SF was introduced to improve strength of pervious concrete.
However, because of the smaller particle size and large specific surface area of SF, the flow values of
However, because of the smaller particle size and large specific surface area of SF, the flow values of
pastes significantly decreased when the content of SF increased.
pastes significantly decreased when the content of SF increased.
Figure 4 shows the effect of SP, SF, and W/C on the segregation index of pervious concretes, for
Figure 4 shows the effect of SP, SF, and W/C on the segregation index of pervious concretes, for
which a 4.75–9.5 mm aggregate size was used.
which a 4.75–9.5 mm aggregate size was used.
In Figure 4, it was obvious that the segregation index increased as the W/C increased. When the
In Figure 4, it was obvious that the segregation index increased as the W/C increased. When
W/C was constant, an increase of SP and a decrease of SF also increased the segregation index. The
the W/C was constant, an increase of SP and a decrease of SF also increased the segregation index.
segregation index of pervious concrete with 0.7% SP and 5% SF was the largest. This was mainly
The segregation index of pervious concrete with 0.7% SP and 5% SF was the largest. This was mainly
related to the flow value of the paste; the increase of the flow value of the paste can increase the
related to the flow value of the paste; the increase of the flow value of the paste can increase the
segregation index. Moreover, it also can be seen from Figures 3 and 4 that the effects of the W/C on
segregation index. Moreover, it also can be seen from Figures 3 and 4 that the effects of the W/C on the
the flow value of the paste and segregation index of pervious concretes were found to be not so
flow value of the paste and segregation index of pervious concretes were found to be not so obvious
obvious compared with those of SP.
compared with those of SP.
Processes 2018,6,6,78
Processes2018, x FOR PEER REVIEW 7 of 14
7 of 14
Processes 2018, 6, x FOR PEER REVIEW 7 of 14
25
25 0.3% SP+5% SF
0.5% SP+5%
0.3% SP+5%SFSF
20 0.7% SP+5%
0.5% SP+5%SFSF

Segregation Index:%
20 0.5% SP+10%
0.7% SP+5%SFSF

Segregation Index:%
0.5% SP+10% SF
15
15

10
10

5
5

0
0 0.22 0.24 0.26 0.28 0.30 0.32
0.22 0.24 W/C 0.28
0.26 0.30 0.32
W/C
Figure 4. Segregation index of pervious concretes. (Three concrete specimens were prepared for each
Figure 4. Segregationindex
indexofofpervious
perviousconcretes.
concretes. (Three concrete specimens were prepared
Figure 4. Segregation
mix design and the figure shows an average of these(Three concrete
together specimens
with the standard were . forfor
prepared
deviation
each
each
mix design and the figure shows an average of these together with the standard deviation
mix design and the figure shows an average of these together with the standard deviation). .
Combined with the above test and strength test, it was found that pervious concrete with 0.3%
SP was Combined towith
difficultwith mold.theTheabove test and strength
paste test, itand
wasgathered
found that in pervious concrete with 0.3%
Combined the above test easily
and flowed down
strength test, it was found that the bottom
pervious of the pervious
concrete with 0.3%
concrete with 0.7% SP. When the SF content increased from 5% to 10%, the flow values of thepervious
SP was difficult to mold. The paste easily flowed down and gathered in the bottom of the pastes
SP was difficult
concrete with to mold.
0.7% SP. The paste easily flowed down and gathered in the bottom of the pervious
drastically decreased, soWhen
there the
wereSFsome
content increasedtofrom
difficulties 5% topervious
molding 10%, theconcrete
flow valueswhen of the
the pastes
W/C
concrete with
drastically 0.7%
decreased,SP. When the SF content increased from 5% to 10%, the flow values of the pastes
was less than 0.28. At so same there were
time, thesome difficulties
strength of the topervious
moldingconcrete
perviousalso concrete when For
decreased. the W/C
the
drastically decreased,
was lessconcrete
than 0.28. so there were some difficulties to molding pervious concrete when the W/C was
pervious withAt0.5% sameSP time,
and 5% theSF,strength
when the of W/C
the pervious concrete
was less than 0.26,also decreased.
the flow values For the
of the
lesspervious
than 0.28. At same
concrete withtime,
0.5%theSP strength
and 5% SF, of when
the pervious concrete
was less also decreased. For the pervious
pastes were relatively small, the cohesive force the
of W/C
the pervious than 0.26, was
concrete the flow
small,values
andofthe
the
concrete
pastes with
were 0.5% SP and small,
relatively 5% SF, thewhen the W/C
cohesive wasof
force less
thethan 0.26, theconcrete
pervious flow valueswas of the pastes
small, and were
the
compressive strength was low; however, when the W/C was greater than 0.30, the pastes easily
relatively small,
compressive the
strengthcohesive
was force
low; of the
however, pervious
when concrete
the W/C was
was small,
greater
flowed down and gathered in the bottom of the pervious concrete, which was unfavorable for water and
than the compressive
0.30, the pastes strength
easily
was low; however,
flowed downThe
permeability. when
andflowgatheredthe W/C
value in thewas
ofthe bottomgreater
cement than
ofpaste
the and0.30,
pervious the pastes
concrete,
segregation easily
which
index flowed
ofwas down
unfavorable
pervious and
concretesforgathered
water
were
insatisfactory
the bottomwith
permeability.of The
the pervious
flow value concrete,
of the which
cement was
paste unfavorable
and for
segregation water
index permeability.
of
a W/C of 0.26–0.30, 0.5% SP, and 5% SF. The W/C of 0.28 was emphatically selected pervious The flow
concretes value
were
offor
the cement
satisfactory paste
with
the following study.a and
W/C segregation
of 0.26–0.30, index
0.5% of
SP, pervious
and 5% SF. concretes
The W/C were
of 0.28 satisfactory
was with
emphatically a W/C
selected of
0.26–0.30, 0.5% SP, and
for the following study. 5% SF. The W/C of 0.28 was emphatically selected for the following study.
3.2. Porosity and Pore Distribution of Pervious Concrete
3.2.3.2.
Porosity and
Porosity Pore
and Pore Distribution
DistributionofofPervious
PerviousConcrete
Concrete
3.2.1. Porosity of Pervious Concrete
3.2.1. Porosity
3.2.1. Porosityofof
Pervious
PerviousConcrete
Concrete
As shown by the results given in Figure 5a, three different sizes of the aggregate were used to
studyAsthe
shown
As effectby
shown ofbythe
the results
the resultsgiven
specimen given in
inFigure
surface Figure
at 5a,
5a, three
different three different
different
locations sizes
on thesizes ofthe
of
planar theaggregate
aggregate
porosity. Figure were
were used
used
5b shows to to
study
studythe effect of
the
the volumetriceffect the specimen
of the
porosity specimen surface
surface
and average at different
at
planar porositylocations
different locations on the
on
of pervious the planarwith
planar
concrete porosity.
porosity. Figure
Figure
different 5b5b shows
shows
aggregate
the the volumetric
volumetric
sizes. porosity
porosity andand average
average planar
planar porosity
porosity of pervious
of pervious concrete
concrete with
with different
different aggregate
aggregate sizes.
sizes.
4.75mm~9.5mm 33 (b) Volumetric porosity
27 (a) 9.5mm~16mm
4.75mm~9.5mm 33 (b) Planar porosityporosity
Volumetric
50%4.75mm~9.5mm+50%9.5mm~16mm
9.5mm~16mm
27 (a) 30 Planar porosity
50%4.75mm~9.5mm+50%9.5mm~16mm
30
24 27
24 27
Porosity:%
Porosity:%

24
Porosity:%
Porosity:%

24
21
21
21
21
18
18 18
18 15
15
15 12
15 0 20 40 60 80 100 120 140 160 180 12 A B C D E F G
0 20 40 Height:mm
60 80 100 120 140 160 180 A B Gravel
C size:mm
D E F G
Height:mm Gravel size:mm
Figure 5. Porosity of pervious concrete. (a) Planar porosity, the design porosity was 21%; (b)
Figure 5. Porosity
Figure of pervious
5. porosity
Porosity concrete.
of pervious (a) Planar
concrete. porosity,
(a) Planar the design
porosity, the porosity
design was 21%;was
porosity (b) volumetric
21%; (b)
volumetric and average planar porosity, the design porosity was 15%, 21%, and 27%. The
porosity and average
volumetric porosity planar
and porosity,
average the design
planar porosity
porosity, the was 15%,
design 21%,was
porosity and15%,
27%.21%,
The figure
and shows
27%. The an
figure shows an average value and standard deviation of three to five specimens.
average
figurevalue
showsand standard
an average deviation
value of threedeviation
and standard to five specimens.
of three to five specimens.
Processes 2018, 6, 78 8 of 14
Processes 2018, 6, x FOR PEER REVIEW 8 of 14

As
As shown
shown in in Figure
Figure 5a,
5a,the
thedifference
differencevalue
valuebetween
between thethe planar
planar porosity
porosity andand design
design porosity
porosity was
was within ±3%, so the planar porosity of the pervious concrete had less fluctuation at
within ±3%, so the planar porosity of the pervious concrete had less fluctuation at each cross-section. each cross-
section. This indicated
This indicated that theofquality
that the quality of the using
the specimen specimen
bothusing both theand
the vibration vibration and method
compaction compactionwas
method was more satisfactory. It can be seen from Figure 5b that the volumetric
more satisfactory. It can be seen from Figure 5b that the volumetric porosity and planar porosity andporosity
planar
porosity
increasedincreased
when thewhendesigntheporosity
design porosity
increased.increased. The difference
The difference value between
value between the volumetric
the volumetric porosity
porosity and planar porosity was within ±3%. Additionally, the volumetric
and planar porosity was within ±3%. Additionally, the volumetric porosity and planar porosity porosity and planarof
porosity of concrete with single-size aggregates were closer to the designed porosity.
concrete with single-size aggregates were closer to the designed porosity. The pervious concretes withThe pervious
concretes with blended
blended aggregates had aggregates had than
higher porosity higher porosity
those than those
with single with
sizes of single sizes
aggregates. The of aggregates.
porosity of the
The porosity
pervious of the with
concretes pervious concretes
50% 9.5–16 mmwith 50% 16–19
and 50% 9.5–16mmmmaggregate
and 50% sizes
16–19was
mmtheaggregate sizes
largest. The was
reason
the largest. The reason for this may be that a loosening effect may have occurred when
for this may be that a loosening effect may have occurred when big aggregates were pushed by small big aggregates
were pushed
aggregates by small
[18], aggregates
and this increased[18],
the and this increased
porosity the porosity
of the pervious of the pervious concretes.
concretes.

3.2.2. Pore
3.2.2. PoreDistribution
Distributionof
of Pervious
Pervious Concrete
Concrete
Figure66shows
Figure showsthetheeffect
effectofofaasingle-size
single-sizeaggregate
aggregateononthe
thepore
poredistribution
distributionofofpervious
perviousconcrete.
concrete.
Figure 6a
Figure 6a shows
showsthe
theeffect
effect of
of aa single-size
single-size aggregate
aggregate on
on the
the pore
pore size
size histogram.
histogram. Figure
Figure 6b
6b shows
showsthe
the
effect of
effect of aa single-size
single-size aggregate
aggregate on on the
the cumulative
cumulative frequency
frequency distribution.

25 (a) 4.75mm~9.5mm
9.5mm~16mm 100
(b)
16mm~19mm 90
20
80
4.75mm~9.5mm
Cumulative frequency:%

70 9.5mm~16mm
Frequency:%

15 16mm~19mm
60
50
10 40
30
5 20
10
0 0
0 2 4 6 8 10 12 14 16 18 20 22 24 0 2 4 6 8 10 12 14 16 18 20 22 24
Pore size:mm Pore size:mm

Figure 6. Effect of a single-size aggregate on the pore distribution of pervious concrete. (a) Frequency
Figure 6. Effect of a single-size aggregate on the pore distribution of pervious concrete. (a) Frequency
distribution of pore sizes; (b) cumulative frequency distribution of pore sizes. The design porosity
distribution of pore sizes; (b) cumulative frequency distribution of pore sizes. The design porosity was
was 15%. The figure shows the average value of three to five specimens.
15%. The figure shows the average value of three to five specimens.

In Figure 6a, the pervious concrete with smaller aggregates had a higher proportion of smaller
pores,Inwhereas
Figure 6a,thethe pervious
pervious concrete
concrete withwith smaller
larger aggregates
aggregates had a had a higher
relatively proportion
large proportion ofof
smaller
large
pores. In Figure 6b, the pervious concrete with larger aggregates had a larger effective pore sizelarge
pores, whereas the pervious concrete with larger aggregates had a relatively large proportion of (the
pores.
pore In corresponding
size Figure 6b, the to pervious
50% of concrete with larger
the cumulative aggregates
frequency had a larger
distribution effective
[8]), and pore size
its cumulative
(the pore size
frequency curvecorresponding
increased more to 50% of the
slowly. Forcumulative
example, whenfrequency distribution
the aggregate size[8]), and its cumulative
of concrete was 4.75–
frequency curve increased more slowly. For example, when
9.5 mm, the pore sizes of pervious concrete were mainly concentrated within 7 mm; 71% the aggregate size of concrete was
of the pores
were smaller than 5 mm and the effective pore size was about 2.4 mm. When the aggregate sizethe
4.75–9.5 mm, the pore sizes of pervious concrete were mainly concentrated within 7 mm; 71% of of
pores were
concrete wassmaller
9.5–16 mm, thanthe5 mm
poreand theofeffective
sizes perviouspore size was
concrete wereabout
mainly 2.4concentrated
mm. When within the aggregate
10 mm
sizethe
and of concrete
effective was
pore9.5–16
size wasmm, the 5pore
about mm.sizes of pervious
However, 40% ofconcrete
the poreswere mainly
in the specimenconcentrated
with 9.5–16 within
mm
10 mm and the effective pore size was about 5 mm. However, 40% of the
aggregates were smaller than 5 mm, 28% of the pores were larger than 10 mm, and the effective porepores in the specimen with
9.5–16
size wasmm aboutaggregates
6.5 mm. were This wassmaller thanthe
because 5 mm, 28%aggregate
smaller of the pores were in
resulted larger thansurface
a larger 10 mm,areaandand the
aeffective pore shape
more regular size was about
of the 6.5 mm.Therefore,
aggregate. This wasas because the smaller
the number of contactaggregate resultedaggregates
points between in a larger
surface area and a more regular shape of the aggregate.
increased, the internal pore size decreased and the pore number increased. Therefore, as the number of contact points
between aggregates increased, the internal pore size decreased and the pore
Figure 7 shows the effect of a blended aggregate on the pore distribution of pervious concrete. number increased.
Figure Figure 7 shows
7a shows the effect
the effect of aof a blended
blended aggregate
aggregate on the
on the porepore
sizedistribution of pervious
histogram. Figure concrete.
7b shows the
Figure 7a shows the effect of a blended aggregate on
effect of a blended aggregate on the cumulative frequency distribution. the pore size histogram. Figure 7b shows the
effect of a blended aggregate on the cumulative frequency distribution.
Processes 2018, 6, 78 9 of 14
Processes 2018, 6, x FOR PEER REVIEW 9 of 14
Processes 2018, 6, x FOR PEER REVIEW 9 of 14
Blended aggregate 100 (b)
25 Blended aggregate
75%4.75mm~9.5mm+25%9.5mm~16mm 100 (b)
25 (a) 75%4.75mm~9.5mm+25%9.5mm~16mm
50%4.75mm~9.5mm+50%9.5mm~16mm 90
(a)
50%4.75mm~9.5mm+50%9.5mm~16mm
50%4.75mm~9.5mm+50%16mm~19mm 90
20 50%4.75mm~9.5mm+50%16mm~19mm
50%9.5mm~16mm+50%16mm~19mm
80

frequency:%
20 50%9.5mm~16mm+50%16mm~19mm
80

frequency:%
70
70
15 60
15 60
Frequency:%

50 Blended aggregate
Frequency:%

50 Blended aggregate
75%4.75mm~9.5mm+25%9.5mm~16mm

Cumulative
10 40 75%4.75mm~9.5mm+25%9.5mm~16mm

Cumulative
10 50%4.75mm~9.5mm+50%9.5mm~16mm
40 50%4.75mm~9.5mm+50%9.5mm~16mm
30 50%4.75mm~9.5mm+50%16mm~19mm
30 50%4.75mm~9.5mm+50%16mm~19mm
50%9.5mm~16mm+50%16mm~19mm
5 20 50%9.5mm~16mm+50%16mm~19mm
5 20
10
0 10
00 2 4 6 8 10 12 14 16 18 20 22 24 0
0 2 4 6 8 10 12 14 16 18 20 22 24 00 2 4 6 8 10 12 14 16 18 20 22 24
Pore size:mm 0 2 4 6 8 Pore
10 size:mm
12 14 16 18 20 22 24
Pore size:mm Pore size:mm
Figure 7. Effect of aa blended
blended aggregate on on the pore
pore distribution of of pervious concrete.
concrete. (a)
(a) Frequency
7. Effect
Figure 7.
Figure Effect of
of a blended aggregate
aggregate on the
the pore distribution
distribution of pervious
pervious concrete. (a) Frequency
Frequency
distribution
distribution of
of pore
pore sizes;
sizes; (b)
(b) cumulative
cumulative frequency
frequency distribution
distribution of of pore
pore sizes.
sizes. The The design
design porosity
porosity was
distribution of pore sizes; (b) cumulative frequency distribution of pore sizes. The design porosity
was
15%. 15%. The
The figurefigure
shows shows the average
the average value
valuevalue of
of threethree to
to five five specimens.
was 15%. The figure shows the average of three tospecimens.
five specimens.

The pore size distribution of pervious concrete with the blended aggregate size was close to that
The pore size distribution of pervious concrete with the blended aggregate size was close to that
of concrete with the single-size aggregate. The addition of smaller aggregates in pervious concrete
of concrete with the single-size
single-size aggregate. The addition of smaller aggregates in pervious concrete
could improve the internal pore size distribution. With the increase of smaller aggregates in concrete,
could improve the internal pore size distribution. With With the
the increase
increase ofof smaller
smaller aggregates
aggregates in
in concrete,
concrete,
the proportion of larger pores decreased; at the same time, the effective pore size also decreased.
the proportion of larger pores decreased;
decreased; at
at the
the same
same time,
time, the
the effective
effective pore
poresize
sizealso
alsodecreased.
decreased.
Figure 8 shows the effect of the design porosity on the pore distribution of pervious concrete.
Figure 8 shows the effect of the design porosity on the pore distribution of of pervious
pervious concrete.
concrete.
Figure 8a shows the effect of the design porosity on the pore size histogram. Figure 8b shows the
Figure 8a
8a shows
showsthe
theeffect
effectofof
thethe
design porosity
design porosityon the porepore
on the size size
histogram. Figure
histogram. 8b shows
Figure the effect
8b shows the
effect of the design porosity on the cumulative frequency distribution.
of the of
effect design porosity
the design on theon
porosity cumulative frequency
the cumulative distribution.
frequency distribution.
Design porosity=12% 100
35 Design porosity=15%
porosity=12% 100 (b)
35 (a) Design porosity=18%
porosity=15% 90 (b)
(a)
30 Design porosity=21%
Design porosity=18% 90
30 80
porosity=21%
Design porosity=24% Design porosity=12%
frequency:%

80
porosity=24% 70 Design porosity=15%
Design porosity=12%
frequency:%

25 Design porosity=27%
25 Design porosity=27% 70 Design porosity=18%
porosity=15%
Frequency:%

Design
60
Frequency:%

Design porosity=21%
Design porosity=18%
20 60 Design porosity=24%
porosity=21%
20 50 Design
50 Design porosity=27%
Design porosity=24%
Cumulative

15 40 Design porosity=27%
Cumulative

15 40
30
10 30
10 20
5 20
10
5 10
0 0
0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 00 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
0 1 2 3 4 5 Pore
6 7size:mm
8 9 10 11 12 13 14 15 16 0 1 2 3 4 5 Pore
6 7size:mm
8 9 10 11 12 13 14 15 16
Pore size:mm Pore size:mm
Figure 8. Effect of the design porosity on the pore distribution of pervious concrete. (a) Frequency
Figure 8. Effect
Effect of the
of the design
design porosity
porosity on
on the
the pore
pore distribution
distribution ofof pervious
pervious concrete. (a) Frequency
Figure 8.
distribution of pore sizes; (b) cumulative frequency distribution of pore sizes.concrete. (a) Frequency
The aggregate size was
distribution
distribution of pore sizes; (b)
(b) cumulative
cumulative frequency
frequency distribution of of pore
pore sizes.
sizes. The
The aggregate size was
was
4.75–9.5 mm.ofThe
pore sizes;
figure shows the average value ofdistribution
three to five specimens. aggregate size
4.75–9.5 mm. The figure shows the average value of three to five specimens.
4.75–9.5 mm. The figure shows the average value of three to five specimens.
In Figure 8b, the difference between all of the cumulative frequencies of pores was relatively low
In Figure 8b, the difference between all of the cumulative frequencies of pores was relatively low
In Figure
for different 8b, the
design difference
porosities. between
The allpore
effective of the cumulative
sizes were within frequencies
2–3 mm. ofIn pores
Figurewas relatively
8a, there was
for different design porosities. The effective pore sizes were within 2–3 mm. In Figure 8a, there was
low for
only different
slight design
increase porosities.
of large pores The effective
in the concrete pore sizes
with thewere within
larger 2–3porosity.
design mm. In Figure 8a, there
This might be
only slight increase of large pores in the concrete with the larger design porosity. This might be
was only slight
attributed to theincrease
fact thatof large
the poresconcrete
pervious in the concrete with the
with a larger larger
design designonly
porosity porosity. Thisinmight
resulted be
a larger
attributed to the fact that the pervious concrete with a larger design porosity only resulted in a larger
attributed to the fact that the pervious concrete with a larger design porosity only resulted in a larger
number of open and connected pores, but had little effect on the pore size of the concrete.
number of open and connected pores, but had little effect on the pore size of the concrete.
number of open and connected pores, but had little effect on the pore size of the concrete.
3.3. Permeability of Pervious Concrete
3.3. Permeability of Pervious Concrete
3.3.1. Accurate Determination of Permeability Coefficient
3.3.1. Accurate Determination of Permeability Coefficient
As shown by the results given in Figure 9, different design porosities and different aggregate
As shown by the results given in Figure 9, different design porosities and different aggregate
sizes were used to study the effect of hydraulic gradients on the permeability coefficient, and the
sizes were used to study the effect of hydraulic gradients on the permeability coefficient, and the
standard deviations of the permeability coefficient under three hydraulic gradients were calculated.
standard deviations of the permeability coefficient under three hydraulic gradients were calculated.
Processes 2018, 6, 78 10 of 14

3.3. Permeability of Pervious Concrete

3.3.1. Accurate Determination of Permeability Coefficient

As shown by the results given in Figure 9, different design porosities and different aggregate sizes
were used to study the effect of hydraulic gradients on the permeability coefficient, and the standard
deviations
Processes 2018,of6,the
Processes 2018, 6, x FORpermeability
x FOR
PEER REVIEW
PEER REVIEWcoefficient under three hydraulic gradients were calculated. 10 of 14
10 of 14

4.75mm~9.5mm 75%4.75mm~9.5mm+25%9.5mm~16mm
0.030 9.5mm~16mm 50%4.75mm~9.5mm+50%9.5mm~16mm
16mm~19mm 50%4.75mm~9.5mm+50%16mm~19mm
50%9.5mm~16mm+50%16mm~19mm
0.025
Standard deviation

0.020

0.015

0.010

0.005

0.000
12 15 18 21 24 27
Design porosity:%
Figure 9.
9. Standard
Standard deviation
deviation of
of the
the permeability coefficient
coefficient of pervious
pervious concrete withwith different
Figure
Figure 9. Standard deviation of the permeability
permeability coefficient of
of pervious concrete
concrete with different
different
hydraulic
hydraulic gradients. The figure shows the average standard deviation values of three specimens.
hydraulic gradients.
gradients. The
The figure
figure shows
shows the
theaverage
averagestandard
standarddeviation
deviationvalues
valuesofofthree
threespecimens.
specimens.

It was obvious that the standard deviation values of pervious concrete with different mixture
It
It was
was obvious
obvious that
that the
the standard
standard deviation
deviation values
values ofof pervious
pervious concrete
concrete with
with different
different mixture
mixture
ratios were no more than 0.03, and the variation of the standard deviation values was not regular.
ratios
ratios were
werenonomore
morethan
than0.03, and
0.03, thethe
and variation of the
variation standard
of the deviation
standard values
deviation was was
values not regular. This
not regular.
This reflected that the dispersion degree of the permeability coefficients under different hydraulic
reflected that the
This reflected dispersion
that degree degree
the dispersion of the permeability coefficients
of the permeability under different
coefficients under hydraulic gradients
different hydraulic
gradients was very small. The use of a visual permeameter and the sealed sidewall method with
was very small.
gradients The use
was very of aThe
small. visual
usepermeameter and the sealed
of a visual permeameter sidewall
and method
the sealed with method
sidewall Vaseline with
and
Vaseline and transparent wrapping film could effectively solve the problem of the permeability
transparent
Vaseline and wrapping film could
transparent effectively
wrapping solve the
film could problem solve
effectively of the the
permeability
problem coefficient changing,
of the permeability
coefficient changing, as the hydraulic gradient changed due to side wall leakage.
as the hydraulic
coefficient gradient
changing, changed
as the duegradient
hydraulic to side wall leakage.
changed due to side wall leakage.
Figure 10 shows the permeability coefficient of pervious concrete measured by the new method
Figure
Figure 1010 shows
shows the
the permeability
permeability coefficient
coefficient of
of pervious
pervious concrete
concrete measured
measured by by the
the new
new method
method
and traditional method (no blocking on the sidewall of the specimen). Figure 10a shows the
and traditional method (no blocking on the sidewall of the specimen).
and traditional method (no blocking on the sidewall of the specimen). Figure 10a shows the
10a shows the
permeability coefficient of pervious concrete with different design porosities. Figure 10b shows the
permeability
permeability coefficient of of pervious
pervious concrete
concretewith
withdifferent
differentdesign
designporosities.
porosities. Figure
Figure 10b10b shows
shows the
permeability coefficient of pervious concrete with different aggregate sizes.
the permeability
permeability coefficient
coefficient of pervious
of pervious concrete
concrete withwith different
different aggregate
aggregate sizes.
sizes.
12 (a)
12 (a) N 10
N 10 (b) N
T (b)
coefficient:mm/s

10 T 9 N
T
coefficient:mm/s

9
coefficient:mm/s

10 T
8
coefficient:mm/s

8
8 7
8 7
6
6 6
6 5
Permeability

5
Permeability

4
Permeability

Permeability

4 4
4 3
3
2 2
2 2
1
1
0 0
0 12 15 18 21 24 27 0 A D E F B G C
12 15 Design18porosity:%
21 24 27 A D E F
Gravel size:mmB G C
Design porosity:% Gravel size:mm
Figure 10. The permeability coefficient of permeable concrete under different test methods. (a) The
Figure 10. The
The permeability coefficient
coefficient of permeable
permeable concrete under
under different test
test methods. (a)
(a) The
The
Figure porositypermeability
design 10. of permeable concrete of with 4.75–9.5 concrete
mm aggregate different
sizes; (b) the methods.
aggregate size of
design
design porosity
porosity of permeable
of permeable concrete with 4.75–9.5 mm aggregate sizes; (b) the aggregate size of
permeable concrete with 21% concrete with 4.75–9.5
design porosity. mm aggregate
The figure sizes;
shows the (b) the
average aggregate
value size of
and standard
permeable
permeable concrete with 21% design porosity. The figure shows the average value and standard
deviation ofconcrete with 21% design porosity. The figure shows the average value and standard
three specimens.
deviation of
deviation of three
three specimens.
specimens.

In Figure 10, the permeability coefficient tested by the traditional method was larger than that
In Figure 10, the permeability coefficient tested by the traditional method was larger than that
tested by the new method. Moreover, the larger design porosity and aggregate size resulted in a
tested by the new method. Moreover, the larger design porosity and aggregate size resulted in a
larger difference value of the permeability coefficient measured by the two methods. For example, in
larger difference value of the permeability coefficient measured by the two methods. For example, in
Figure 10a, the difference value of the permeability coefficient in pervious concrete with 12% design
Figure 10a, the difference value of the permeability coefficient in pervious concrete with 12% design
porosity was 0.35 mm/s, whereas the difference value increased to 2.8 mm/s for pervious concrete
porosity was 0.35 mm/s, whereas the difference value increased to 2.8 mm/s for pervious concrete
with 27% design porosity. In Figure 10b, the difference value of the permeability coefficient in
with 27% design porosity. In Figure 10b, the difference value of the permeability coefficient in
pervious concrete with 4.75–9.5 mm aggregate size was 1.52 mm/s, whereas the difference value
pervious concrete with 4.75–9.5 mm aggregate size was 1.52 mm/s, whereas the difference value
 Processes 2018, 6, 78 11 of 14

In Figure 10, the permeability coefficient tested by the traditional method was larger than that
tested by the new method. Moreover, the larger design porosity and aggregate size resulted in a
larger difference value of the permeability coefficient measured by the two methods. For example, in
Figure 10a, the difference value of the permeability coefficient in pervious concrete with 12% design
porosity was 0.35 mm/s, whereas the difference value increased to 2.8 mm/s for pervious concrete
with 27% design porosity. In Figure 10b, the difference value of the permeability coefficient in pervious
Processes
concrete 2018, 6, x FOR
with PEER mm
4.75–9.5 REVIEW 11 of 14to
aggregate size was 1.52 mm/s, whereas the difference value increased
2.4 mm/s for pervious concrete with 16–19 mm aggregate size. This was because the larger design
sidewall surfaces. Thus, the leakage and permeability coefficient of pervious concrete increased in
porosity and aggregate size resulted in a larger pore size and greater porosity of the sidewall surfaces.
the traditional method.
Thus, the leakage and permeability coefficient of pervious concrete increased in the traditional method.
3.3.2.
3.3.2.Influential
InfluentialFactors
Factorsofofthe
thePermeability
PermeabilityCoefficient
Coefficient
As
Asbe seen from
be seen fromFigure
Figure 10a,
10a, the the permeability
permeability coefficient
coefficient increased
increased as the porosity
as the design design porosity
increased.
increased.
In particular, when the porosity was greater than 15%, a small change of porosity of
In particular, when the porosity was greater than 15%, a small change porosity
significantly
significantly
changed thechanged the permeability
permeability coefficient.
coefficient. This This was
was because the because
perviousthe pervious
concrete withconcrete
a largerwith a
design
larger design porosity resulted in a larger measured porosity. When the measured
porosity resulted in a larger measured porosity. When the measured porosity reached a critical porosity reached
avalue,
criticalthe
value, the resistance
resistance of the internal
of the internal structure
structure of the concrete
of the concrete to water
to water was greatly
was greatly reduced,reduced,
so the
so the permeability coefficient increased rapidly. Figure 11a shows the relationship
permeability coefficient increased rapidly. Figure 11a shows the relationship between the permeability between the
permeability coefficient and volumetric porosity of pervious concrete. It was found
coefficient and volumetric porosity of pervious concrete. It was found that the relationship between the that the
relationship
permeabilitybetween theand
coefficient permeability
volumetriccoefficient andright
porosity was volumetric porosityfunction,
for a quadratic was right
and forthe
a quadratic
correlation
function, and the correlation
coefficient was 0.97. coefficient was 0.97.

10 (a) 8 (b)
K=0.030P2-0.676P+4.615
2
K=0.106X2-0.061X+2.5
R =0.97 R2=0.93
Permeability coefficient:mm/s

Permeability coefficient:mm/s

8 7

6 6

5
4

4
2

3
0
10 12 14 16 18 20 22 24 26 28 2 3 4 5 6 7
Volumetric porosity:% Effective pore size:mm

Figure
Figure11.
11.Permeability
Permeabilitycoefficient
coefficientofofpervious
perviousconcrete.
concrete.(a)
(a)The
Therelationship
relationshipbetween
betweenthe
thepermeability
permeability
coefficient
coefficient and volumetric porosity of pervious concrete with aggregate sizes of 4.75–9.5mm;
and volumetric porosity of pervious concrete with aggregate sizes of 4.75–9.5 mm;(b)
(b)the
the
relationship between the permeability coefficient and effective pore size of pervious concrete
relationship between the permeability coefficient and effective pore size of pervious concrete with with
21%
21% design
design porosity.
porosity.

As can be observed in Figure 10b, the effect of the aggregate size on the permeability coefficient
As can be observed in Figure 10b, the effect of the aggregate size on the permeability coefficient
was significant. The permeability coefficient increased with the increase of single-size aggregates,
was significant. The permeability coefficient increased with the increase of single-size aggregates,
and the permeable coefficient of the pervious concrete with the blended aggregate was slightly lower
and the permeable coefficient of the pervious concrete with the blended aggregate was slightly lower
than that of concrete with the single-size aggregate. As shown in Figure 5b, there was little difference
than that of concrete with the single-size aggregate. As shown in Figure 5b, there was little difference
in the porosity variation of pervious concrete with different mix ratios. Therefore, the permeability
in the porosity variation of pervious concrete with different mix ratios. Therefore, the permeability
coefficient was not sensitive to the porosity. The reason may be related to the internal pore sizes of
coefficient was not sensitive to the porosity. The reason may be related to the internal pore sizes of
pervious concrete; the larger aggregate size in the pervious concrete result in a larger internal pore
pervious concrete; the larger aggregate size in the pervious concrete result in a larger internal pore
size. Figure 11b shows the relationship between the effective pore size and the permeability
size. Figure 11b shows the relationship between the effective pore size and the permeability coefficient.
coefficient. It was found that the relationship between the permeability coefficient and the effective
It was found that the relationship between the permeability coefficient and the effective pore size
pore size approximately obey the polynomial function, and the correlation coefficient was 0.93.
approximately obey the polynomial function, and the correlation coefficient was 0.93.
3.4. Strength of Pervious Concrete
As shown in Figure 12, two different aggregate sizes and three different design porosities were
used in the pervious concrete samples. The compressive strength and flexural strength of the
pervious concrete were measured.
Processes 2018, 6, 78 12 of 14

3.4. Strength of Pervious Concrete

As shown in Figure 12, two different aggregate sizes and three different design porosities were
used in the pervious concrete samples. The compressive strength and flexural strength of the pervious
concrete
Processes were
2018, measured.
6, x FOR PEER REVIEW 12 of 14

Aggregate size=4.75-9.5mm Aggregate size=9.5-16mm

Aggregate size=4.75-9.5mm Aggregate size=9.5-16mm
30 3-day 3-day 5
(a) (b) 3-day 3-day
7-day 7-day
7-day 7-day
28-day 28-day
25 28-day 28-day
4
Compressive strength:MPa

Flexural strength:MPa
20
3
15
2
10

1
5

0 0
15 18 21 15 18 21
Design porosity:% Design porosity:%

Figure
Figure12.
12.Effect
Effectofofthe
thedesign
designporosity
porosityononthe
thestrength
strengthofofpervious
perviousconcrete.
concrete.(a)(a)
Compressive
Compressivestrength;
strength;
(b)
(b) Flexural strength. The figure shows an average value and standard deviationof
Flexural strength. The figure shows an average value and standard deviation ofthree
threespecimens.
specimens.

In Figure 12, it was obvious that the strength of the pervious concrete increased with the decrease
In Figure 12, it was obvious that the strength of the pervious concrete increased with the decrease
of the design porosity. It was also found that the strength of the pervious concrete with 4.75–9.5 mm
of the design porosity. It was also found that the strength of the pervious concrete with 4.75–9.5 mm
aggregate size was larger than that with 9.5–16 mm aggregate size. For the pervious concrete with
aggregate size was larger than that with 9.5–16 mm aggregate size. For the pervious concrete with
4.75–9.5 mm aggregate sizes, the compressive and flexural strengths of the concrete with 15% design
4.75–9.5 mm aggregate sizes, the compressive and flexural strengths of the concrete with 15% design
porosity were about 27.2 MPa and 4.1 MPa, respectively. Meanwhile, they were about 23.5 MPa and
porosity were about 27.2 MPa and 4.1 MPa, respectively. Meanwhile, they were about 23.5 MPa and
3.3 MPa for the pervious concrete with 21% design porosity, respectively. For the pervious concrete
3.3 MPa for the pervious concrete with 21% design porosity, respectively. For the pervious concrete
with 9.5–16 mm aggregate sizes, they were about 19 MPa and 2.4 MPa for pervious concrete with
with 9.5–16 mm aggregate sizes, they were about 19 MPa and 2.4 MPa for pervious concrete with
21% design porosity, respectively. At the same time, the strengths of the pervious concrete were
21% design porosity, respectively. At the same time, the strengths of the pervious concrete were
satisfactory at 0.26–0.30 W/C ratios. For example, the compressive strengths of pervious concretes
satisfactory at 0.26–0.30 W/C ratios. For example, the compressive strengths of pervious concretes
with 4.75–9.5 mm aggregate sizes were about 25.2–27.2 MPa, and the related flexural strengths were
with 4.75–9.5 mm aggregate sizes were about 25.2–27.2 MPa, and the related flexural strengths were
about 3.7–4.1 MPa. The strength of pervious concrete gradually increased with the increase of the
about 3.7–4.1 MPa. The strength of pervious concrete gradually increased with the increase of the
curing time. It could be deduced that the strength was mainly dependent on the bonding strength, as
curing time. It could be deduced that the strength was mainly dependent on the bonding strength,
well as the number of bonding points and bonding areas between cementitious materials and
as well as The
aggregates. the number
number ofof bonding
bondingpoints,
pointscontact
and bonding areas
areas, and between
effective cementitious
bonding materialswith
points increased and
aggregates. The number of bonding points, contact areas, and effective bonding points increased with
the decrease of design porosity and aggregate size.
the decrease of design porosity and aggregate size.
4. Conclusions
4. Conclusions
1. The segregation index of pervious concrete increased with the increase of SP and the W/C ratio,
1. The segregation index of pervious concrete increased with the increase of SP and the W/C ratio,
and the decrease of SF. The use of SP could significantly increase the flow value, while the use
and the decrease of SF. The use of SP could significantly increase the flow value, while the use
of SF signally reduced the flow value. The flow values of pastes and the segregation index for
of SF signally reduced the flow value. The flow values of pastes and the segregation index for
pervious concretes were obtained and found to be very suitable when the W/C of 0.26–0.30, 0.5%
pervious concretes were obtained and found to be very suitable when the W/C of 0.26–0.30, 0.5%
SP, and 5% SF were used.
SP, and 5% SF were used.
2. In the paper, volume porosity, planar porosity, and pore sizes were obtained by the water
2. In the paper, volume porosity, planar porosity, and pore sizes were obtained by the water
displacement method and image processing technology. It was found that the planar porosity
displacement method and image processing technology. It was found that the planar porosity of
of pervious concrete was close to the volumetric porosity. There was little difference between
pervious concrete was close to the volumetric porosity. There was little difference between the
the measured porosity (volumetric porosity and planar porosity) and the design porosity when
measured porosity (volumetric porosity and planar porosity) and the design porosity when a
a single-size aggregate was used. However, the difference was large when a blended aggregate
single-size aggregate was used. However, the difference was large when a blended aggregate was
was used. This may be due to the possible loosening effect of the blended aggregates. The
used. This may be due to the possible loosening effect of the blended aggregates. The aggregate
aggregate size was the main influencing factor of the pore distribution of the pervious concrete,
size was the main influencing factor of the pore distribution of the pervious concrete, and the
and the effective pore size increased with the increase of the aggregate size. The design porosity
effective pore size increased with the increase of the aggregate size. The design porosity had little
had little effect on the pore distribution. The dosage of cementitious material decreased when
the design porosity increased; correspondingly, the number of open and connected pores
increased and the connections became difficult.
3. Based on Darcy’s law, the permeameter and the sealed sidewall method can effectively solve the
problem of sidewall leakage and the permeability coefficient can be accurately determined. The
standard deviation of the permeability coefficient was less than 0.03 under different hydraulic
gradients. The permeability coefficient of the pervious concrete was determined by the porosity
Processes 2018, 6, 78 13 of 14

effect on the pore distribution. The dosage of cementitious material decreased when the design
porosity increased; correspondingly, the number of open and connected pores increased and the
connections became difficult.
3. Based on Darcy’s law, the permeameter and the sealed sidewall method can effectively solve
the problem of sidewall leakage and the permeability coefficient can be accurately determined.
The standard deviation of the permeability coefficient was less than 0.03 under different hydraulic
gradients. The permeability coefficient of the pervious concrete was determined by the porosity
and aggregate size jointly, and the permeability coefficient increased with increase of the
aggregate size and design porosity. It was found that the relationships between the permeability
coefficient and the volumetric porosity (or effective pore size d50 ) approximately obey the
polynomial function.
4. In the pervious concrete material system, the use of SF and SP improved the performance of
pervious concrete, but the dosage of SF and SP should not be excessive. The strength of the
concrete significantly increased with the increase of cementitious material and the decrease
of aggregate size. When 15% design porosity and 4.75–9.5 mm aggregate size were used in
concrete, the 28-day compressive strength was 27.2 MPa and the 28-day flexural strength was
4.1 MPa. Based on the performance of pervious concretes, the W/C of 0.26–0.30; 0.5% SP content;
5% SF content; 15–21% design porosity; and aggregate sizes of 4.75–9.5 mm and 9.5–16 mm
are suggested.

Author Contributions: R.L., H.L. and F.S. conceived of and designed the study. H.L., F.S., and H.Y. performed the
experiments. R.L., F.S. and Q.Z. developed the permeameter. R.L., H.L., F.S., H.Y., Q.Z., S.S. and Z.Z. wrote and
modified the paper.
Funding: This research was funded by the General Program of the National Natural Science Foundation of China,
grand number is 51779133; National Key R&D Plan of China, grand number is 2016YFC0801604; and Joint Funds
of National Natural Science Foundation of China, grand number is U1706223.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Tennis, P.D.; Leming, M.L.; Akers, D.J. Pervious Concrete Pavements; EB302 Portland Cement Association:
Skokie, IL, USA, 2004.
2. Schaefer, V.R.; Wang, K.; Suleiman, M.T.; Kevern, J.T. Mix Design Development for Pervious Concrete in Cold
Weather Climates, Final Report; National Concrete Pavement Technology Center, Iowa State University: Ames,
IA, USA, 2006.
3. Malhotra, V.M. No-fines concrete-its properties and applications. ACI Mater. J. 1976, 73, 628–644.
4. Liang, L. Preparation, Porous Structure and Camouflage Performance of Porous Ecological Concrete; Nanjing
University of Aeronautics and Astronautics: Nanjing, China, 2010.
5. Kevern, J.T. Advancement of Pervious Concrete Durability. Ph.D. Thesis, Iowa State University, Ames, IA,
USA, 2008.
6. Kevern, J.T. Mix Design Determination for Freeze–Thaw Resistant Portland Cement Pervious Concrete.
Master’s Thesis, Iowa State University, Ames, IA, USA, 2006.
7. Hu, J.; Stroeven, P. Application of image analysis to assessing critical pore size for permeability prediction of
cement paste. Image Anal. Stereol. 2003, 22, 97–103. [CrossRef]
8. Deo, O.; Neithalath, N. Compressive response of pervious concrete proportioned for desired porosities.
Constr. Build. Mater. 2011, 25, 4181–4189. [CrossRef]
9. Montes, F.; Valavala, S.; Haselbach, L. A new test method for porosity measurements of Portland cement
pervious concrete. J. ASTM Int. 2005, 2, 73–82.
10. Liu, R.; Li, B.; Jiang, Y. A fractal model based on a new governing equation of fluid flow in fractures for
characterizing hydraulic properties of rock fracture networks. Comput. Geotech. 2016, 75, 57–68. [CrossRef]
11. Liu, R.; Jiang, Y.; Li, B.; Yu, L. Estimating permeability of porous media based on modified Hagen–Poiseuille
flow in tortuous capillaries with variable lengths. Microfluid. Nanofluid. 2016, 20, 120. [CrossRef]
Processes 2018, 6, 78 14 of 14

12. Kayhanian, M.; Anderson, D.; Harvey, J.T. Permeability measurement and scan imaging to assess clogging
of pervious concrete pavements in parking lots. J. Environ. Manag. 2012, 95, 114–123. [CrossRef] [PubMed]
13. Jiang, Z.; Sun, Z. Effects of Some Factors on Properties of Porous Pervious Concrete. J. Build. Mater. 2005, 8,
513–519.
14. Yang, Z.F. Study on Material Design and Road Performances of Porous Concrete; Wuhan University of Technology:
Wuhan, China, 2008.
15. Martin, W.D., III; Kaye, N.B.; Putman, J.B. Impact of vertical porosity distribution on the permeability of
pervious concrete. Constr. Build. Mater. 2014, 59, 78–84. [CrossRef]
16. Yang, J.; Jiang, G. Experimental study on properties of pervious concrete pavement materials. Cem. Concr.
Res. 2003, 33, 381–386. [CrossRef]
17. Huang, B.; Wu, H.; Shu, X.; Burdette, E.G. Laboratory evaluation of permeability and strength of
polymer-modified pervious concrete. Constr. Build. Mater. 2010, 24, 818–823. [CrossRef]
18. Tamai, M. Water permeability of hardened materials with continuous voids. Cem. Assoc. Jpn. Rev. 1990,
446–449.
19. Mishima, N.; Tanigawa, Y.; Mori, H.; Kurokawa, Y.; Terada, K.; Hattori, T. Experimental study on rheological
properties of dense suspension by shear box test. J. Struct. Constr. Eng. 2000, 65, 13–19. [CrossRef]
20. Lian, C.; Zhuge, Y.; Beecham, S. The relationship between porosity and strength for porous concrete.
Constr. Build. Mater. 2011, 25, 4294–4298. [CrossRef]
21. Larrard, D.F. Concrete Mixture Proportioning: A Scientific Approach; E&FN Spon: London, UK, 1999; p. 421.

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).