Journal of Environmental Management 190 (2017) 1e8

Contents lists available at ScienceDirect

Journal of Environmental Management

journal homepage: www.elsevier.com/locate/jenvman

Research article

Leaching and retention of dissolved metals in particulate loaded

pervious concrete columns
Timothy M. Vadas a, b, *, Malcolm Smith a, Hongwei Luan a
a
University of Connecticut, 261 Glenbrook Road, Department of Civil and Environmental Engineering, Storrs, CT 06269, USA
b
University of Connecticut, 3107 Horsebarn Hill Road, Center for Environmental Science and Engineering, Storrs, CT 06269, USA

a r t i c l e i n f o a b s t r a c t

Article history: This study examined metal leaching and retention in pervious concrete with or without embedded
Received 18 August 2016 particulate matter. Particulate matter was collected from an adjacent parking lot and from a nearby
Received in revised form parking garage as examples of weathered and un-weathered particulate matter. Particle size distribu-
17 December 2016
tions were similar, but metal content was 3e35-fold higher and organic matter content was 3-fold higher
Accepted 19 December 2016
Available online 23 December 2016
in the parking garage particulate matter compared to the parking lot particulate matter. Replicate col-
umns were established with either no particulate added as the control, or 20 g of parking lot or parking
garage particulate matter. Synthetic rainwater was passed through the columns at variable rainfall in-
Keywords:
Particulate matter
tensity or fixed intensity to assess leaching. Metals were leached at higher concentrations from the
Pervious parking garage particulate amended column, but from all columns less than 1% of the metal mass
Stormwater leached. Rainfall intensity did not have a large effect on leached metal concentrations, only varying
Urban runoff effluent by about 2-fold. Synthetic stormwater with elevated dissolved Cu, Zn, Cd and Pb concentrations
Heavy metals was passed through the same columns and metal removal efficiencies were on the order of 85e95%, 30
Retention e95%, 60e90%, and 95þ% for each metal, respectively. After loading the column with a year's worth of
Leaching stormwater metal exposure, removal efficiencies in the no particulate and parking lot particulate
amended columns decreased, while parking garage particulate amended columns performed similarly
with a small drop in Cu and Pb removal efficiencies. Generally, columns with no particulate and parking
lot particulate amendments performed similarly, suggesting the pervious concrete is responsible for the
majority of the initial metal retention. The parking garage particulate amended columns retained more
metals from stormwater, perhaps due to an increase in pH that promoted surface precipitation as hy-
droxides or carbonate species on the pervious concrete, or due to complexation in the higher concen-
trations of organic matter and iron oxides in the particulate matter. Overall, metal retention was aided by
the presence of organic matter in the particulate matter, but the pervious concrete itself was more
important than particulates for metal retention. A strategy to increase metal retention and removal from
the environment could involve amending pervious concrete with mixtures of well-defined sorbents to
enhance metal retention.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction increased runoff leading to receiving water body degradation

The built urban environment consists of a large portion of
impermeable surfaces, with transportation related land-uses ac-
counting for up to 70 percent (Legret et al., 2007). Creation of
impermeable surfaces results in decreased infiltration and
* Corresponding author. University of Connecticut, 261 Glenbrook Road,
Department of Civil and Environmental Engineering, Storrs, CT 06269, USA.
E-mail addresses: timothy.vadas@uconn.edu (T.M. Vadas), malcolmsmith18@
gmail.com (M. Smith), hongwei.luan@uconn.edu (H. Luan).
through higher peak flow, decreased base flow, erosion and
contaminant loading (Bunn and Arthington, 2002). Stormwater
best management practices (BMPs) are increasingly implemented
to mitigate surface water body degradation (Booth and Jackson,
1997).
Pervious concrete is an example of a structural BMP that is
meant to provide short-term water storage, increase infiltration
and reduce runoff. The major difference in pervious concrete
compared to conventional concrete is the limitation of fines in the
mixture, leading to the development of a network of

http://dx.doi.org/10.1016/j.jenvman.2016.12.047
0301-4797/© 2016 Elsevier Ltd. All rights reserved.
2 T.M. Vadas et al. / Journal of Environmental Management 190 (2017) 1e8
interconnected void spaces as the concrete cures (Schwartz, 2010).
These void spaces allow the pervious concrete to act as a coarse
filtration media to improve water quality, effectively reducing net
export of particulates typically by greater than 80% (Kuang et al.,
2007; Legret et al., 2007; Sansalone et al., 2008, 2012; Tota-
Maharaj and Scholz, 2010).
While particulate removal has been well established and re-
ductions in total suspended solids (TSS) levels in runoff have been
observed, effectiveness for dissolved contaminant removal is a
concern with respect to water quality standards in receiving water
bodies. Only a few studies have examined dissolved water quality
treatment by pervious concrete (Bean et al., 2007b; Collins et al.,
2010; Haselbach et al., 2014) and shown reductions in total nutri-
ents, sediment bound Zn and dissolved Cu and Zn. Field studies
examining permeable friction courses (Eck et al., 2012) or perme-
able paver systems (Brattebo and Booth, 2003; Fassman and
Blackbourn, 2010) have observed variable effects of pervious
pavements on dissolved water quality, either minimal change or up
to 70% removal of dissolved Cu, Zn, or Pb. Researchers examining
dissolved nitrogen species removal in permeable pavements have
found aerobic conversion of ammonia to nitrate, with little overall
net removal (Collins et al., 2010). While general observations have
been made on water quality improvements with respect to dis-
solved contaminants in field scale pervious pavement systems, the
direct impact of the pervious concrete layer has not been investi-
gated thoroughly.
Pervious concrete is similar to other types of permeable pave-
ments (i.e. plastic grids, porous asphalt, or interlocking pavers) in
that it has pore space that needs to be cleaned out periodically, but
unique in that the calcium hydroxide and calcium carbonate con-
tent of concrete can result in an increase in stormwater pH (Setunge
et al., 2009). The capture of particulates in pervious concrete is a
function of pore space, particle size distribution and sediment
loading (Myers et al., 2011; Sansalone et al., 2012). This can result in
removal of particulate attached metals, but the removal of dis-
solved metals is a function of sorption or precipitation reactions
within the pervious concrete pavement. The capture of particulates
within pervious concrete therefore generates both a filling of the
pores that may affect hydraulic conductivity and new surfaces that
may interact with dissolved metals. Dissolved metals often parti-
tion to suspended solids in runoff and researchers generally
observe a decrease in dissolved metals with increasing total sus-
pended solids (TSS) concentrations (Sansalone and Buchberger,
1997; Stone and Marsalek, 1996). Depending on the characteris-
tics of the TSS, partitioning to the solid phase may occur due to
common mechanisms such as cation exchange, carbonate precipi-
tation, or sorption to Fe/Mn oxides and organic matter (Stone and
Marsalek, 1996). Given the multitude of partitioning processes, it
is reasonable to assume that as particulates accumulate in the
pervious concrete matrix, partitioning of metals between dissolved
and solid phases will be affected in water passing through the
pervious concrete. In addition, the increase in pH or alkalinity of
water passing through pervious concrete may alter metal removal
processes, e.g. increased metal hydroxide precipitation. However, it
is currently unknown whether this particulate matter will act pri-
marily as a source of dissolved metals or as a sink and whether the
pervious concrete material itself or the particulate matter collected
within the pervious concrete are more important for removal of
contaminants. We hypothesize that given the wide variety of metal
partitioning mechanisms in the particulate matter, it has more
control over the fate of dissolved contaminants than the pervious
concrete material itself.
The objective of this study was to examine the impact of par-
ticulate matter captured within the pervious concrete matrix and
the pervious concrete itself on the leaching or removal of dissolved
metal contaminants. Pervious concrete columns were established
ex-situ with either no particulates added or with weathered or un-
weathered roadway particulate matter added. The particulate
matter was characterized for particle size distribution, organic
matter content and initial metal concentrations. Both synthetic
rainwater and stormwater with elevated metals were separately
passed through the pervious concrete columns at specified rainfall
intensities and water quality parameters were measured in the
effluent to assess the impact of particulate loading on leaching or
removal of dissolved metals.
2. Materials and methods
2.1. Pervious concrete column and particulate sample collection
A demonstration pervious concrete plot was installed on the
University of Maryland Baltimore County (UMBC) campus in 2008,
retrofitted within an existing asphalt parking lot. The plot remained
covered, which kept particulates from building up in the matrix,
until a few weeks before ten 9.4 cm diameter cores were cut from
the pavement using an electric powered core drill with a diamond
bit. Cores were cut in length to approximately 11.4 cm with a ma-
sonry blade in a miter saw. In addition, parking lot particulate
material was collected with a fine brush and stainless steel scoop
from both the adjacent asphalt parking lot and from the basement
of a parking garage located on campus. These particulate samples
were expected to differ mainly in characteristics such as metal
content and organic matter accumulation since the parking lot
material is continuously exposed to rainfall and the parking garage
material is not.
2.2. Particle size analysis
Approximately 200 g of each particulate sample was succes-
sively sieved through 4 mm, 2 mm, 1 mm, 0.5 mm, 0.25 mm,
0.125 mm, and 0.063 mm brass sieves to calculate a particle size
distribution by mass. Nested sieves were mechanically shaken for
10 min and the separated samples were weighed on an analytical
balance. Sub-samples for each size fraction were taken for
extractable metals by EPA Method 3050b and a whole sample was
determined for extractable metals and loss on ignition organic
matter analysis (Nelson and Sommers, 1996).
2.3. Column experimental setup
Nine pervious concrete columns were wrapped in two layers of
parafilm, inserted into a 20.3 cm length of 10.2 cm outer diameter
polyvinyl chloride (PVC) pipe (Fig. 1). Each pervious concrete col-
umn was supported at the bottom by three PVC wedges tapped into
place evenly around the edges. The column was sealed at the top
with a bead of silicone placed around the top edge of the column,
set 2.54 cm down from the top rim of the PVC pipe. These condi-
tions ensured that flow passed only through the pervious concrete
matrix and no preferential flow occurred along the side of the
column. Depending on the treatment, either 20 g of particulate
sample or no particulate sample, as a control, were added to 3
replicate columns. Pervious concrete surfaces have widely varying
particulate loads and since this study was meant to compare dif-
ferences in metal leaching or retention with or without particulate
matter, the columns were loaded to fill the surface pores without
significantly reducing infiltration rates. Synthetic rainwater or
stormwater was delivered to each column using a peristaltic pump.
An acrylic disk with eight through holes (one in the center, and
seven evenly spread around the center, halfway to the perimeter of
the disk) was placed on top of the pervious concrete column. An
 T.M. Vadas et al. / Journal of Environmental Management 190 (2017) 1e8
Fig. 1. Diagram of pervious concrete experimental setup. A pervious concrete column
was placed within a 10.2 cm pvc pipe, wrapped in parafilm, and sealed at the top edges
with silicone. Water was delivered using a peristaltic pump with eight simultaneous
lines placed into a 0.64 cm acrylic disk with evenly spaced holes laid over the pervious
concrete column.
eight channel minipuls 2 pump (Rainin Gilson) delivered water
through each hole using tygon pump tubing. With this setup, water
could be evenly spread across the surface of the column while
delivering water at a set rainfall intensity which varied from 0.3 to
10.2 cm/h depending on the experiment.
Synthetic rainwater was prepared using ACS grade chemicals
(Fisher Scientific, Pittsburgh, PA, USA) containing 96 mM NaCl,
10 mM K2SO4, 5 mM CaCl2$2H2O, 6 mM MgCl2$6H2O, 15 mM NH4NO3,
0.1 mM KH2PO4 and pH adjusted to 6.5 following Jones and Edwards
(1993). Synthetic stormwater was prepared containing 80 mg/L Pb,
80 mg/L Cu, 20 mg/L Cd, 600 mg/L Zn, 7 mg/L NO3-N, 2 mg/L NH4-N,
0.6 mg/L PO4-P, and 120 mg/L CaCl2 with chloride salts used for
metals, calcium and ammonium, and sodium salts for nitrate and
phosphate following Davis et al. (2001). The pH was adjusted to 7
for stormwater with either NaOH or HCl.
A treatment consisted of pumping 150 mL of solution through
the column at a fixed flow rate, mimicking common rainfall in-
tensities in the northeast U.S. In the first batch of experiments,
treatments were applied at varying intensity (10.2, 5.8, 3.8, 2.8, 2.3,
1.3, 0.9, and 0.3 cm/h) reflecting average 2 yr storm intensities for
10 min up to 24 h storms, or at a fixed intensity of 2.3 cm/h.
Stormwater treatments were added at a fixed intensity of 2.3 cm/h
to the columns previously exposed to rainwater. The same columns
were loaded with additional metals by pumping an addition
400 mL of synthetic stormwater with a 20-fold increase in metal
concentration through each column. This was meant to represent
about one year’s worth of runoff metal exposure. Final experiments
then pumped synthetic stormwater at a fixed intensity of 2.3 cm/h
to metal loaded columns to examine treatment ability after long-
term metal exposure. Effluent from each column was collected
below in a polypropylene beaker and pH was measured with an
electrode. Each sample was filtered through a 0.45 mm nitrocellu-
lose filter and filtrate was directly collected into polypropylene
tubes for analysis of dissolved metals. Metals samples were acidi-
fied to 1% HNO3 and all samples were stored at 4  C until analysis.
Filters were dried at 70  C until a constant weight was reached and
weighed for total suspended solids. However, after minimal par-
ticulate flushing (<0.1 g) during initial wetting of the columns, total
suspended solids were below detection and are not reported.
3
2.4. Metals analysis
Dissolved metals were analyzed on an Agilent 7700x ICP-MS
with a He collision/reaction cell. Analytical standards were pre-
pared from 1000 mg/L stock solutions from VHG Labs (Manchester,
NH, USA) for metals. Calibration verification standards were pre-
pared from Spex CertiPrep verified standards (Metuchen, NJ, USA).
Quality control samples and spike recoveries were within 10% or
better for ICP-MS measured elements. Minimum detection limits
were 0.1 mg/L for Cu, Zn, and Cd and 0.01 mg/L for Pb. Statistical
comparisons between samples were made based on a student's t-
test with a 95% confidence interval.
3. Results and discussion
3.1. Characteristics of particulate matter
The particulate samples used in this study were initially char-
acterized for particle size distribution, metal content, and organic
matter content. Particle size distribution based on mass (Fig.2) was
similar for both the parking lot and parking garage particulate
matter. Small differences were observed between the samples, for
example, approximately 69% of the particulate matter was finer
than 500 mm for the parking garage material, whereas only 62% was
for the parking lot material. The parking garage particulate matter
had a higher percentage of the finest particulate fraction, 4% less
than 63 mm, whereas the parking lot material only had 1% less than
63 mm. This is important because the finer fractions are more likely
to pass through the pervious concrete. Given the similar particle
size distributions overall, these geographically co-located particu-
late sources were chosen as additives to the pervious concrete
columns assuming that the major difference was chemical due to
the decreased weathering of the parking garage material.
The particulate matter used in this study is similar to those
assessed at other sites across the United States in terms of particle
size distribution and chemical characteristics (Kuang et al., 2007;
Legret and Colandini, 1999; Sansalone et al., 1998; Shaheen, 1975;
Zafra et al., 2008). In particular, Legret and Colandini (1999)
observed that the particles captured within a pervious concrete
field site were about 80% less than 2 mm in size, which is similar to
the 80e85% less than 2 mm in the particulate matter used in this
study. At smaller particle sizes, other studies of urban roadways
have found wide ranges of particulate matter in certain size ranges,
e.g. a range of about 3e40% of the total mass in the less than 63 mm
size fraction (Kuang et al., 2007; Zafra et al., 2008). However, these
Fig. 2. Particle size distribution by mass of parking lot and parking garage roadway
material used in pervious concrete column experiments. Lines are drawn to guide the
eye. Trends may not be linear between points.
4 T.M. Vadas et al. / Journal of Environmental Management 190 (2017) 1e8
are particulate matter fractions generated during runoff events and
are not necessarily the same distribution as will fall on a pervious
pavement which may receive a higher percentage of larger particles
from surface deposition (e.g. from vehicle tires or windblown).
Regardless, the size distributions are more variable at the smaller
size fractions which are the most important due to their potential
to pass through the pavement and likelihood of having more metals
bound to them. Sansalone et al. (2012) suggest greater than 80%
filtration of 300 mm particles and at least 50% filtration of smaller
particles occur in permeable pavements. In our effluent samples,
not enough effluent particulate matter was present for a particle
size analysis, but upon inspection under the microscope, the largest
observed particle sizes were approximately 400 mm, with most
much smaller. The high retention in our study is likely in part due to
the loading conditions (i.e. all at once) and pore structure of the
pervious concrete. Depending on the particulate matter charac-
teristics at a site and the pore structure of the permeable pavement,
the filtration effects differ and researchers have found that filtration
efficiency increases with time (Sansalone et al., 2012).
With respect to chemical characteristics, the parking garage
material was more heavily contaminated with trace metals than the
parking lot material (Table 1). The combination of more braking and
tight radius turning in the parking garage likely contributed to
more vehicle metal sources to the particulate. The difference in Pb
concentration was the highest, with about 35-times more Pb in the
parking garage material, while Cr, Cu, Zn and Ni were at least 10-
fold higher, and Cd about 3-fold higher. Total metal concentra-
tions in the particulate matter were within the lower range
observed by others in roadside particulate material (Jartun et al.,
2008; Sansalone and Buchberger, 1997). The parking garage ma-
terial had higher concentrations of Fe, by about twice as much, and
3 times as much organic matter as the parking lot material. The
higher iron content likely contributed to the higher Cr content of
the parking garage particulate as Cr forms strong surface complexes
with iron oxides. The higher heavy metal concentrations in the
parking garage particulate are likely in part due to the higher
organic matter content (McBride, 1994) and the higher concentra-
tions of Ca and Mg in the parking lot material likely competing for
cation exchange and ligand sites (Tuccillo, 2006).
Given the high retention of particulates in the pervious concrete
cores, greater than 99%, particulate based transport of metals was
not a major factor in this case. However, some smaller particles did
initially pass through the columns. Depending on the storm event,
researchers have found differences in metal content by particle
size. High percent metal fractions (about 60e80%) have been
observed in the >75 mm fraction of particulate matter transported
in low intensity storms with either high or low runoff volumes
(Sansalone et al., 2010), but in high intensity storms, the
largest > 75 mm fraction was only about 40% Pb and the other
metals were typically less than 25%. Assuming the pervious con-
crete is used as a replacement for a parking lot, it is more likely to
receive all size fractions of particulate matter, and not just runoff
transported material as described in Sansalone et al. (2010). How-
ever, given that we observed very low particulate transport through
the columns, pervious pavements have been shown to be effective
filters for TSS (Bean et al., 2007a; Drake et al., 2013; Jiang et al.,
2015; Roseen et al., 2012; Sansalone et al., 2012; Winston et al.,
Table 1
Initial elemental concentrations in roadway particulate matter used in pervious concrete column experiments. Values are averages ± standard deviation of 3 samples.
Sample OM (mass %) Mg (mg/g) Ca (mg/g) Fe (mg/g)
Parking lot 2.5 ± 0.09 76 ± 9.8 200 ± 28 6.5 ± 1.2
Parking garage 7.6 ± 0.47 21 ± 0.46 68 ± 2.1 16 ± 0.51
2016; Yong et al., 2013), and there is typically additional sub-
grade material and/or soil that will filter particles, dissolved con-
taminants are the primary concern for downstream waterbodies. In
systems where over- or underdrains are installed (Schwartz, 2010),
dissolved contaminants are even more relevant. In addition to
dissolved contaminant transport through pervious concrete, the
retention of particles in a pervious concrete pavement represent a
potential source or sink of metals.
3.2. Rainwater exposure
Initial treatments with rainwater were meant to mimic pervious
concrete pavements that treat direct rainfall only. In this manner,
we assess the potential for leaching of metals from the concrete or
particulate matter added to the pervious concrete. In general,
metals were leached from the columns treated with parking garage
particulate matter to a greater extent than columns without par-
ticulate matter or with parking lot particulate matter (Fig. 3). The
largest difference was observed in Cu leaching, with concentrations
leached at 10.2 cm/h rainfall that were about 8-fold higher than no
particulate and 2-fold higher than parking lot particulate amended
columns. Other leaching metal concentrations were typically 2e4
fold higher in parking garage particulate amended columns. This
general trend of more metals leached from the parking garage
particulate matter was expected since it had higher initial con-
centrations of all contaminant metals in the solid phase (Table 1).
As for comparing the no particulate and parking lot particulate
amended columns, effluent concentrations were often similar with
the exception of Ni and Cd for which the no particulate columns
leached more on average. However, for Ni leaching, just one of the 3
columns had higher effluent concentrations, leading to the higher
average and large error bars for the no particulate columns.
The rainfall intensity influenced the leached metal concentra-
tions in some cases, though most often in the columns with par-
ticulate matter added (Fig. 3). There were downward trends in
metal concentrations leached as intensity decreased for almost all
metals, with the exception of Cr in parking lot particulate amended
columns, and for Cu and Ni in parking garage particulate amended
columns. The only leached metal concentrations that seemed to be
influenced by intensity in the no particulate columns was Zn. Most
differences observed with rainfall intensity were not very large
however, typically varying less than 2-fold across the range of
rainfall intensities examined, and with relatively large error bars on
most samples, none were significantly different (p > 0.05). There is
not a substantial amount of research on the effect of rainfall in-
tensity on leaching or runoff concentrations of dissolved metals.
Kayhanian et al. (2003) found that dissolved Cu and Zn concen-
trations decreased with higher maximum rainfall intensity which is
the opposite of our case. Crabtree et al. (2006) also found that metal
concentrations in runoff decreased with increasing rainfall in-
tensity, but with the exception of one site that had a porous asphalt
surface. The reason porous asphalt (Crabtree et al., 2006) and
pervious concrete in our case may have promoted the opposite
trend in many cases is unclear, but could be due to kinetics of
desorption or dissolution from the particulate matter or pervious
concrete as water infiltrates quickly.
In the second scenario of exposure with rainwater, we tested
Cr (mg/g) Cu (mg/g) Zn (mg/g) Ni (mg/g) Cd (mg/g) Pb (mg/g)
4.5 ± 2.1 6.6 ± 5.6 23 ± 7.5 3.0 ± 0.02 0.13 ± 0.02 4.4 ± 1.6
100 ± 2.9 110 ± 3.0 300 ± 7.4 35 ± 0.83 0.34 ± 0.01 160 ± 3.7
 T.M. Vadas et al. / Journal of Environmental Management 190 (2017) 1e8
Fig. 3. Average (±1 SD) effluent metal concentrations from pervious concrete columns amended with either parking lot or parking garage particulate matter and exposed to varying
intensity of rainwater.
whether there would be a difference in metal leaching over time by
assessing the concentration after every 100 mL of rainwater passed
through the column. In all cases, there was a decrease in leached
concentration with increasing volume (Fig. 4). Initial concentra-
tions leached were similar to the same rainfall intensity in the
variable intensity experiment (Fig. 3) for Cu, Zn and Pb, with the
exception of Cd, Cr and Ni, for which concentrations were below the
detection limit. With subsequent volumes, leached concentrations
decreased significantly (p < 0.05) in the case of Pb, in some in-
stances by almost two orders of magnitude, e.g. going from 2 mg/L to
0.03 mg/L in the case of Pb leaching for no particulate columns.
Other cases were not as dramatic and leached Zn concentrations
did not vary much with dosage in the no particulate column.
Overall, this type of trend is consistent with the idea of a first flush,
where a large percentage of the metal pollutants are in the earlier
volumes of the runoff (Davis and McCuen, 2006), and suggesting
that the earlier volumes of rainwater exposure capture the more
easily leached metals.
A variety of reasons could exist as to why the metals leached
from the different materials. The lower pH of the rainwater might
have promoted release of metals since metal solubility generally
increases as pH decreases. However, the pH of the effluent
5
increased in all cases from the initial rainwater value of 6.5,
increasing to an average of 9.2 ± 0.6 in the no particulate columns,
8.4 ± 0.7 in the parking lot particulate amended columns, and
9.0 ± 0.6 in the parking garage particulate amended columns. These
are similar to pH values of DI water passed through a clean column,
which would increase pH to between 9 and 10. Even if the partic-
ulate matter contributed some buffering capacity in the top layers,
interactions as the water passed through the remainder of the
pervious concrete likely led to the increases in pH. Since pH was not
significantly different between columns, other interactions likely
promoted leaching, including ion exchange (e.g. the Ca in the
rainwater exchanged with divalent contaminant metals), and
perhaps some dissolution of organic matter from the un-weathered
parking garage particulate promoted leaching, particularly of Cu
which is known to strongly interact with organic matter. Regard-
less, the mass of metal leached from the columns during both the
variation in intensity and fixed intensity doses, when subtracting
the contribution from the no particulate pervious concrete column,
was often less than 1% of the metal mass in the particulate, sug-
gesting the particulates retain their previously associated metals
well. Based on concentrations of metals in the effluent, some of the
Cu and Cd concentrations are above the acute toxicity aquatic life
6 T.M. Vadas et al. / Journal of Environmental Management 190 (2017) 1e8
Fig. 4. Average (±1 SD) effluent metal concentrations from pervious concrete columns
amended with either parking lot or parking garage particulate matter and exposed to 5
doses of 100 mL rainwater.
criterion for surface waters (EPA, 2016), but would certainly be
diluted below standards once it entered the stream if the effluent
was hydraulically connected.
3.3. Stormwater exposure
In the stormwater exposure scenarios, we dosed each column
with a typical stormwater mixture containing the most common
metals of interest (Cu, Zn, Pb, and Cd) to examine whether these
metals were removed in the pervious concrete and particulate
amended columns, or if leaching still occurred. In all cases, metals
were removed to a large extent, and often more so in the columns
amended with parking garage particulate matter (Fig. 5). Removal
efficiencies were on the order of 85e95% for Cu, 30e95% for Zn,
60e90% for Cd, and 95þ% for Pb. Metal removal in the no partic-
ulate and parking lot particulate amended columns were very
similar, suggesting the pervious concrete was primarily responsible
for metal retention in both cases. The parking garage particulate
amended column was substantially more effective at removing Zn,
Cd and Pb, but not Cu. In general there was little difference as the
number of storm events increased, with only slightly decreasing
effluent Zn and Cd concentrations in the parking lot particulate
amended columns, though not statistically significant.
Following these initial stormwater exposures, the columns were
loaded with a stormwater solution with elevated (20-fold) metal
concentrations meant to expose the material to the mass present in
the average annual rainfall. What this mimics depends on how the
pervious concrete receives stormwater, i.e. this mimics only direct
rainfall volume. Regardless, following additional dosing of storm-
water, removal of metals was generally worse in the case of the no
particulate and parking lot particulate amended columns, and
again effluent metal concentrations were similar between the two
(Fig. 6). The parking garage particulate amended column still per-
formed well for all metals, exhibiting a minor decrease in effec-
tiveness for Cu and Pb. The Cu removal efficiency dropped to
80e90% and Pb removal dropped to 90e98%.
In all stormwater treatments, with or without particulates
loaded, some dissolved contaminant retention was observed. In
general, with the presence of particulate matter, contaminant
retention either remained the same or was enhanced. There is only
one laboratory study and few field studies on pervious concrete
pavement systems to compare against. Haselbach et al. (2014)
observed better than 80% removal of Cu and Zn in pervious con-
crete columns with 20 mg/L and 100 mg/L Cu and Zn, respectively,
exposure concentrations, which are similar to our results in no
particulate columns. Field studies involving several other perme-
able pavement systems, e.g. permeable block (Sansalone et al.,
2010), permeable friction courses (Newman et al., 2002), and
porous asphalt (Legret et al., 1996) have shown good removal of TSS
and total metals, but poor removal of nitrate and dissolved phos-
phorus. Most field studies did not distinguish between total and
dissolved metals, of which dissolved are the most relevant to
toxicity concerns in receiving water. In addition, the location of
effluent sampling varies in field studies, with some in the subgrade,
in the soil below the subgrade, or in a drainage pipe, and thus
represents the influence of multiple sorption site interactions.
Some researchers have found that dissolved Cu concentrations are
not significantly reduced until the runoff reaches the soil layer
(Kwiatkowski et al., 2007), but at this point if soils are burdened
with large metal loads, they cannot be removed from the system
until the entire pavement cross-section is replaced.
The incorporation of sorbents into pervious concrete could add
enhanced contaminant retention capabilities. Interestingly, the
column with no particulate showed nearly as much or more
retention of metals as the parking lot particulate column, sug-
gesting the pervious concrete surface is more important for
retention of metals in those two scenarios. Only with the parking
garage particulate matter was enhanced retention of Zn, Cd and Pb
observed. The pH of the stormwater as it moved through the
pervious concrete columns increased a little, up from 7.0 to at most
7.8 ± 0.1 in the parking garage particulate columns. This increase
relative to the no particulate or parking lot particulate amended
columns, which only increased pH to 7.2e7.4, could have poten-
tially contributed to enhanced surface precipitation of insoluble
metal hydroxides or carbonate species in the parking garage par-
ticulate columns. The parking garage particulate also had higher
organic matter content than the parking lot particulate which could
 T.M. Vadas et al. / Journal of Environmental Management 190 (2017) 1e8 7

Fig. 5. Average (±1 SD) effluent metal concentrations from pervious concrete columns amended with either parking lot or parking garage particulate matter and exposed to
stormwater. Concentration scales are set to show maximum initial concentrations of each element. A * above a group indicates samples within a group were significantly different
(p < 0.05) from the other groups.

Fig. 6. Average (±1 SD) effluent metal concentrations from pervious concrete columns amended with either parking lot or parking garage particulate matter and exposed to
stormwater. Columns were exposed to 1 year's worth of metal contaminated stormwater prior to event 1. A * above a group indicates samples within a group were significantly
different (p < 0.05) from the other groups.

have contributed to enhanced retention of Zn, Cd and Pb, although tend to be relatively slow for Cu-organic complexes (Louis et al.,
enhanced retention of Cu would have been expected as well. Dif- 2009) whereas infiltration is relatively fast through a pervious
ferences in Cu retention could be due to exchange kinetics, which concrete material even with some particulate matter.
8 T.M. Vadas et al. / Journal of Environmental Management 190 (2017) 1e8
3.4. Conclusions
The data presented here indicates that particulate matter,
particularly organic in nature, plays a role in increasing retention of
dissolved metals from stormwater, but does not generally act as a
significant source of metals during rainfall exposure as originally
suggested. The idea that particulate matter has a greater role in
contaminant retention compared to the pervious concrete material
alone was not the case, as the quantities of metals retained in the no
particulate and parking lot particulate amended columns were not
statistically different. However, the additional organic matter pre-
sent in the parking garage particulate amended column did aid in
increasing metal retention in those columns. Increasing metal
retention in pervious concrete through particulate matter sorption
is not ideal since clogging of the pavement is a major concern and
maintenance issue. However, metal retention could be enhanced in
pervious concrete pavements if for example additives such as iron
oxides or fine organic matter could be added to the pervious con-
crete mix design to increase the ability to retain metals as we
observed in the parking garage particulate matter, without clogging
the pores. However, the longevity of the sorption capacity would
need to be tested in the field.
Acknowledgements
The authors thank Stu Schwartz, Senior Scientist at the Center
for Urban Environmental Research and Education at the University
of Maryland, Baltimore County for providing the pervious concrete
columns and for discussions on this project.
References
Bean, E.Z., Hunt, W.F., Bidelspach, D.A., 2007a. Field survey of permeable pavement
surface infiltration rates. J. Irrig. Drain. Eng. 133, 249e255.
Bean, E.Z., Hunt, W.F., Bidelspach, D.A., 2007b. Evaluation of four permeable
pavement sites in eastern North Carolina for runoff reduction and water quality
impacts. J. Irrig. Drain. Eng-ASCE 133, 583e592.
Booth, D.E., Jackson, C.R., 1997. Urbanization of aquatic systems: degradation
thresholds, stormwater detention, and the limits of mitigation. J. Am. Water
Resour. Assoc. 33, 1077e1090.
Brattebo, B.O., Booth, D.B., 2003. Long-term stormwater quantity and quality per-
formance of permeable pavement systems. Water Res. 37, 4369e4376.
Bunn, S.E., Arthington, A.H., 2002. Basic principles and ecological consequences of
altered flow regimes for aquatic biodiversity. Environ. Manag. 30, 492e507.
Collins, K.A., Hunt, W.F., Hathaway, J.M., 2010. Side-by-side comparison of nitrogen
species removal for four types of permeable pavement and standard asphalt in
eastern North Carolina. J. Hydrol. Eng. 15, 512e521.
Crabtree, B., Moy, F., Whitehead, M., Roe, A., 2006. Monitoring pollutants in high-
way runoff. Water Environ. J. 20, 287e294.
Davis, A.P., McCuen, R.H., 2006. Stormwater Management for Smart Growth.
Springer Science, New York.
Davis, A.P., Shokouhian, M., Sharma, H., Minami, C., 2001. Laboratory study of bio-
logical retention for urban stormwater management. Water Environ. Res. 73,
5e14.
Drake, J.A.P., Bradford, A., Marsalek, J., 2013. Review of environmental performance
of permeable pavement systems: state of the knowledge. Water Qual. Res. J.
Can. 48, 203e222.
EPA, 2016. National Recommended Water Quality Criteria - Aquatic Life Criteria
Table.
Eck, B.J., Winston, R., Hunt, W., Barrett, M., 2012. Water quality of drainage from
permeable friction course. J. Environ. Eng. 138, 174e181.
Fassman, E.A., Blackbourn, S., 2010. Urban runoff mitigation by a permeable pave-
ment system over impermeable soils. J. Hydrol. Eng. 15, 475e485.
Haselbach, L., Poor, C., Tilson, J., 2014. Dissolved zinc and copper retention from
stormwater runoff in ordinary portland cement pervious concrete. Constr.
Build. Mater. 53, 652e657.
Jartun, M., Ottesen, R.T., Steinnes, E., Volden, T., 2008. Runoff of particle bound
pollutants from urban impervious surfaces studied by analysis of sediments
from stormwater traps. Sci. Total Environ. 396, 147e163.
Jiang, W., Sha, A., Xiao, J., Li, Y., Huang, Y., 2015. Experimental study on filtration
effect and mechanism of pavement runoff in permeable asphalt pavement.
Constr. Build. Mater. 100, 102e110.
Jones, D.L., Edwards, A.C., 1993. Evaluation of polysulfone hollow fibers and ceramic
suction samplers as devices for the in-situ extraction of soil solution. Plant Soil
150, 157e165.
Kayhanian, M., Singh, A., Suverkropp, C., Borroum, S., 2003. Impact of annual
average daily traffic on highway runoff pollutant concentrations. J. Environ. Eng.
129, 975e990.
Kuang, X.H., Kim, J.Y., Gnecco, I., Raje, S., Garofalo, G., Sansalone, J.J., 2007. Particle
separation and hydrologic control by cementitious permeable pavement.
Transp. Res. Rec. 111e117.
Kwiatkowski, M., Welker, A.L., Traver, R.G., Vanacore, M., Ladd, T., 2007. Evaluation
of an infiltration best management practice utilizing pervious concrete. J. Am.
Water Resour. Assoc. 43, 1208e1222.
Legret, M., Colandini, V., 1999. Effects of a porous pavement with reservoir structure
on runoff water: water quality and fate of heavy metals. Water Sci. Technol. 39,
111e117.
Legret, M., Colandini, V., LeMarc, C., 1996. Effects of a porous pavement with
reservoir structure on the quality of runoff water and soil. Sci. Total Environ.
190, 335e340.
Legret, M., Colandini, V., Pagotto, C., Jullien, A., 2007. The quality of infiltrated water
through porous and conventional asphalt pavements. Road Mater. Pavement
Des. 8, 617e641.
Louis, Y., Garnier, C., Lenoble, V., Mounier, S., Cukrov, N., Omanovic, D., Pizeta, I.,
2009. Kinetic and equilibrium studies of copper-dissolved organic matter
complexation inwater column of the stratified Krka River estuary (Croatia). Mar.
Chem. 114, 110e119.
McBride, M., 1994. Environmental Chemistry of Soils. Oxford University Press.
Myers, B., Beecham, S., van Leeuwen, J.A., 2011. Water quality with storage in
permeable pavement basecourse. Proc. Inst. Civ. Eng.-Water Manag. 164,
361e372.
Nelson, D.W., Sommers, L.E., 1996. Total carbon, organic carbon and organic matter.
In: Page, A.L., Al, E. (Eds.), Methods of Soil Analysis, Part 2, second ed. American
Society of Agronomy, Inc., Madison, WI, pp. 961e1010.
Newman, A.P., Pratt, C.J., Coupe, S.J., Cresswell, N., 2002. Oil bio-degradation in
permeable pavements by microbial communities. Water Sci. Technol. 45,
51e56.
Roseen, R.M., Ballestero, T.P., Houle, J.J., Briggs, J.F., Houle, K.M., 2012. Water quality
and hydrologic performance of a porous asphalt pavement as a storm-water
treatment strategy in a cold climate. J. Environ. Eng. 138, 81e89.
Sansalone, J.J., Buchberger, S.G., 1997. Characterization of solid and metal element
distributions in urban highway stormwater. Water Sci. Technol. 36, 155e160.
Sansalone, J., Koran, J., Smithson, J., Buchberger, G., 1998. Physical characteristics of
urban roadway solids transported during rain events. J. Environ. Eng. 124,
427e440.
Sansalone, J., Kuang, X., Ranieri, V., 2008. Permeable pavement as a hydraulic and
filtration interface for urban drainage. J. Irrig. Drain. Eng-ASCE 134, 666e674.
Sansalone, J., Kuang, X., Ying, G., Ranieri, V., 2012. Filtration and clogging of
permeable pavement loaded by urban drainage. Water Res. 46, 6763e6774.
Sansalone, J., Ying, G., Lin, H., 2010. Distribution of metals for particulate matter
transported in source area rainfall-runoff. J. Environ. Eng. 136, 172e184.
Schwartz, S.S., 2010. Effective curve number and hydrologic design of pervious
concrete storm-water systems. J. Hydrol. Eng. 15, 465e474.
Setunge, S., Nguyen, N., Alexander, B.L., Dutton, L., 2009. Leaching of alkali from
concrete in contact with waterways. Water Air Soil Pollut. Focus 9, 381e391.
Shaheen, D.G., 1975. Contribution of Urban Roadway Usage to Water Pollution.
Office of Research and Development. U.S. Environmental Protection Agency,
Washington DC.
Stone, M., Marsalek, J., 1996. Trace metal composition and speciation in street
sediments: Sault Ste. Marie, Canada. Water Air Soil Pollut. 87, 149e169.
Tota-Maharaj, K., Scholz, M., 2010. Efficiency of permeable pavement systems for
the removal of urban runoff pollutants under varying environmental condi-
tions. Environ. Prog. Sustain. Energy 29, 358e369.
Tuccillo, M.E., 2006. Size fractionation of metals in runoff from residential and
highway storm sewers. Sci. Total Environ. 355, 288e300.
Winston, R.J., Al-Rubaei, A.M., Blecken, G.T., Viklander, M., Hunt, W.F., 2016. Main-
tenance measures for preservation and recovery of permeable pavement sur-
face infiltration rate - the effects of street sweeping, vacuum cleaning, high
pressure washing, and milling. J. Environ. Manag. 169, 132e144.
Yong, C.F., McCarthy, D.T., Deletic, A., 2013. Predicting physical clogging of porous
and permeable pavements. J. Hydrol. 481, 48e55.
Zafra, C.A., Temprano, J., Tejero, I., 2008. Particle size distribution of accumulated
sediments on an urban road in rainy weather. Environ. Technol. 29, 571e582.