lable at ScienceDirect

Journal of Environmental Management 90 (2009) 3574–3580

Contents lists avai

Journal of Environmental Management

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

Leachability of dissolved chromium in asphalt and concrete surfacing materials

Masoud Kayhanian a,*, Akshay Vichare b, Peter G. Green a, John Harvey a

a Department of Civil and Environmental Engineering, One Shields Avenue, Engineering III, University of California, Davis, CA 95616, USAb Chevron EMC, 6111 Bollinger Canyon Road, San Ramon, CA 94568, USA

a r t i c l e i n f o

Article history:Received 11 December 2008Received in revised form30 May 2009Accepted 21 June 2009Available online 14 July 2009

Keywords:Fresh and aged concreteAggregateCementChromiumPavement specimenLeachateHighway runoffWater quality standard

* Corresponding author. Tel.: þ1 530 752 8957; faxE-mail address: mdkayhanian@ucdavis.edu (M. Ka

0301-4797/$ – see front matter � 2009 Elsevier Ltd.doi:10.1016/j.jenvman.2009.06.011

a b s t r a c t

Leachate metal pollutant concentrations produced from different asphalt and concrete pavementsurfacing materials were measured under controlled laboratory conditions. The results showed that, ingeneral, the concentrations of most metal pollutants were below the reporting limits. However, dissolvedchromium was detected in leachate from concrete (but not asphalt) specimens and more strongly in theearly-time leachate samples. As the leaching continued, the concentration of Cr decreased to below orclose to the reporting limit. The source of the chromium in concrete pavement was found to be cement.The concentration of total Cr produced from leachate of different cement coming from different sourcesthat was purchased from retail distributors ranged from 124 to 641 mg/L. This result indicates that thepotential leachability of dissolved Cr from concrete pavement materials can be reduced through sourcecontrol. The results also showed that the leachability of dissolved Cr in hardened pavement materialswas substantially reduced. For example, the concentration of dissolved Cr measured in actual highwayrunoff was found to be much lower than the Cr concentration produced from leachate of both open anddense graded concrete pavement specimens under controlled laboratory study. It was concluded thatpavement materials are not the source of pollutants of concern in roadway runoff; rather most pollutantsin roadway surface runoff are generated from other road-use or land-use sources, or from (wet or dry)atmospheric deposition.

� 2009 Elsevier Ltd. All rights reserved.

1. Introduction

Various studies have linked the sources of dissolved andparticulate contamination in highway stormwater runoff affectingstormwater quality to tire wear, brake wear, vehicular fluids, soilsand atmospheric deposition (Ball et al., 1998; Davis et al., 2001;Councell et al., 2004; Muschack, 1990). Several researchers haveattempt to evaluate the numerical breakdown of contaminatesfrom the contributing sources (Hewitt and Rashed, 1990; Legretand Pagotto, 1999). However, limited research has been performedto evaluate the runoff water quality that is directly attributed to thepavement surfacing materials. The studies that do exist generallyfail to differentiate between portland cement concrete (PCC) andasphalt-surfaced pavements, and the chemical constituents thatare tested usually are limited to a few metals and polycyclicaromatic hydrocarbons (PAHs).

Previous work to evaluate the water quality of leachate frompavement materials was performed by Kriech (1990, 1992), Kriechet al. (2002), Blackburn et al. (1992), Lindgren (1996), Norin and

: þ1 530 752 7872.yhanian).

All rights reserved.

Stromvall (2004), Legret et al. (2005), Birgisdottir et al. (2007),Sanchez et al. (2009). Although the conclusions and analyses werelimited, these initial papers serve as a baseline for identifying theconstituents that pavement materials contribute to highwaystormwater runoff. Kriech tested for five groups of pollutants:PAH’s, polychlorinated biphenyls, semi-volatile organics, volatileorganic compounds (VOCs), and metals. Most constituent concen-trations were below the detection limits except barium, chromiumand naphthalene (a PAH). Vashisth et al. (1997) also looked at theleachability of pavements, but the focus of the research was oncrumb-rubber modifiers (CRM) in hot mix asphalt (HMA) pave-ments. Increasing asphalt temperatures and input water acidity(lowering pH) values were also tested, with the results indicatingan increase in trace metal concentrations due to both of thoseparameters. Brandt and De Groot (2001) investigated PAHs inleachate from various commercially available asphalt binders fromdifferent locations and manufacturers. The constituent concentra-tions among the binder leachate often varied over several ordersmagnitude, leading one to believe that the asphalt binder sourcehas a large effect on the constituents that may be present.

In the above studies, most samples were analyzed at highdetection limits, hence most analytical measurements werereported as not detectable. Therefore, the existing data do not

M. Kayhanian et al. / Journal of Environmental Management 90 (2009) 3574–3580 3575

adequately serve current national and state regulatory waterquality compliance requirements. Lack of sufficient data onpollutant contribution of pavement materials to the runoff waterquality led to a recent investigation performed for the CaliforniaDepartment of Transportation, Division of Environmental Analysisby the University of California, Davis (Signore et al., 2008). This newstudy investigated the water quality of leachate produced from tendifferent modified and unmodified concrete and asphalt pavementsunder controlled laboratory conditions using a much larger suite oforganic and inorganic water quality parameters. From this study itwas determined that an elevated dissolved chromium concentra-tion associated with concrete surfacing pavement materials;especially for leachate samples collected within the first hour of theexperiment. Dissolved chromium, especially Cr (VI) is highly toxicand is a known human carcinogen and mutagen (Kristiansen et al.,1997; Lide, 1998; Frias et al., 1994; Hillier et al., 1999). The toxiceffects of Cr, however, are dependent on the duration of exposureand the dose. Because of the association of Cr (VI) with the cementindustry and concern over human health, in Europe, there isa restriction which prohibits the supply or use of cement which hasa Cr (VI) concentration of more than 2 mg/kg (COSHH, 2007).

The focus of this paper is to present the results of dissolvedchromium concentration leached from different pavements undercontrolled laboratory study as well as identifying the primarysources of chromium in pavement materials.

2. Materials and methods

2.1. Pavement materials

The ten different pavement specimens tested in this studyalong with their respective mix types (portland cement andasphalt binder materials) is summarized in Table 1. The pavementspecimen herewith defined as the mixture of mix type includingthe binders, aggregate and water. Test specimens were preparedat the University of California Pavement Research Center (UCPRC)at the Richmond Field Station (RFS) laboratory. These pavementspecimens were prepared using standard ingot molds and rolling-wheel compaction. Specimens were compacted or placed in steeltrapezoidal specimen molds. The dimension of each specimen(specimen) mold was: bottom (w ¼ 19.12 cm, L ¼ 53.05 cm), top(w ¼ 22.86 cm, L ¼ 57.15 cm) and height ¼ 3.125 cm. The weightand volume of each specimen were approximately 19 kg and9176.3 cm3, respectively. Ingot mold were lined with non-contaminating materials, and form release agents were not used.Portland cement mixes were placed and rodded and vibrated. Theaggregate used in each specimen was primarily obtained froma single source. Triplicates of all pavement type were aged usingoven drying technique that was developed by the Federal

Table 1Pavement specimens mix type and binder materials.

Specimen Mix type Binder

A Rubberized asphalt concreteopen (RAC-O) graded

Rubberized asphalt cement

B Rubberized asphalt concretegap (RAC-G) graded

Rubberized asphalt cement

C, D Open graded asphalt concrete(OGAC)

2 Unmodified performancegrade binder

E, F Open graded asphalt concretepolymer modified (OGAC-PM)

2 Modified performancegrade (PG) binders

G Modified asphalt binder (MB-G) MB4 binderH Conventional dense graded

asphalt concrete (DGAC)Unmodified performancegrade (PG) binder

I PCC-Dense Portland cementJ PCC-Permeable Portland cement

Highway Association and the Transportation Research Boardthrough a national research effort under the Strategic HighwayResearch Program (Bell, 1989) and is widely accepted in pavementresearch. The accelerated aging is representative of approximately15–18 years of in-service pavement life.

2.2. Experimental apparatus and test procedures

The schematic view of the experimental apparatus (set up ina temperature controlled room) is shown in Fig. 1. The majorcomponents of this experimental apparatus include: waterdelivery, auxiliary pumping system, pavement specimencompartment, water distribution system and sample collectingequipment. Further detail on this experimental apparatus can befound in (Signore et al., 2008).

Approximately 30 L of purified (Milli-Q) water produced bya Milli-Q A10 (Millipore Corporation, Billerica, MA) equipment waspumped at a constant rate (0.63 L/h) over each specimen. Thispumping rate was such that the water volume is equivalent to12.7 mm rainfall over the specified specimen area over a 48 hperiod. This simulated rainfall is a typical rainfall in California andwas selected to collect sufficient leachate water to perform thecomplete suite of chemical analysis including the toxicity testing.Selected pavement was triplicated and each pavement specimenwere tested under three different air temperatures of 4, 20 and45 �C representing wide environmental temperature conditions.The water passed through or over the specimen depending on thepavement type permeability. The water was discharged evenly overthe specimen through a small hole on the underside of the spiraledflexible Tygon� tube. The clean water percolated through or passedover the specimens and all water was collected in a compositestainless steel carboy. Additional 12 grab water samples were alsocollected manually for metal analysis during the first 8 h thatinclude five samples during the first hour and one sample everyother hour. These samples were collected to assess the variability ofmetal constituent concentrations in the leachate for each specimen.More detail on experimental apparatus and test procedures can befound in Signore et al. (2008).

To investigate the effect of pH on Cr leachate concentrations,two different methodologies were used. The first approachemployed pH adjusted Milli-Q water (pH of 5.2) to perform theleaching experiment on pavement specimens. To test for potentialcontribution of trace metals due to corrosion of the metal trays, pHadjusted water (pH 5.2) was run through the entire experimentalapparatus without a pavement specimen sample. All other condi-tions were maintained constant for this experimental run. Thesecond method was employed to determine the effect of pH on Crin the particulate state. For this experiment, particulate pavementspecimens were obtained by scraping the concrete and asphaltpavement. Using pavement particles; the metal leachate concen-tration was studied for four different pH’s (7, 6, 5, and 4) with pHadjusted Milli-Q water. The pH was adjusted using trace metalgrade nitric acid. The samples were given a contact time of 2 days toallow for sufficient leaching of chromium. After the contact time,the samples were acidified and analyzed by ICP-MS.

2.3. Field runoff sampling

The filed runoff samples were collected from two separatehighway runoff characterization studies. The first runoff sampleswere collected as part of the statewide highway runoff character-ization in which over 30 urban and non-urban highways weremonitored for metals and other chemical constituents throughoutthe state of California for 3 years during 2000–2003 wet seasons(Kayhanian et al., 2007). These samples were collected by flow-

Composite sample collection carboys

Pavement specimens

Metering pumps

Water supply

Timer switch

Compressed airsolenoid valve

Air filter

Sample collection vials

Pneumatic valves

Ingot trays

Fig. 1. A schematic view of experimental apparatus.

M. Kayhanian et al. / Journal of Environmental Management 90 (2009) 3574–35803576

weighted composited sampler and both dissolved and total Cr andresults were reported as event mean concentration (EMC). Inaddition, samples from six selected sites were also analyzed for Cr(VI). The second field runoff samples were collected during thesame period (2000–2003 wet seasons) from three highly urbanizedhighways in Los Angles as part of first flush highway runoff char-acterization study (Han et al., 2006; Lau et al., 2009). The first flushsamples were collected as grab during the storm event that usuallylasted for about 8 h. In general, five grab samples were collectedduring the first hour with first sample as soon as the flow wasobserved (0 min) and four additional samples were collected forevery 15 min. One grab sample was also collected for every houruntil the end of the rain event. These individual grab samples wereanalyzed for both dissolved and total Cr. Additional detail ofstatewide and first flush highway runoff sample collection can beobtained from Kayhanian et al. (2007) and Lau et al. (2009),respectively.

2.4. Analytical methods

Dissolved Cr was analyzed using an Agilent (Palo Alto, CA) 7500iinductively coupled plasma mass spectrometer (ICP-MS). Thesamples were poured into new plug seal polypropylene centrifuge

tubes (Fisher Scientific, Pittsburgh, PA) and acidified using tracemetal grade nitric acid (Fisher Scientific). The commonly usedreporting limit for Cr by most certified laboratories is 1 mg/L (ng/mL). However, we were able to detect the Cr concentrations downto low parts per trillion (less than 0.01 mg/L) with ICP-MS. Partic-ulate samples were digested in concentrated nitric acid (TraceMetal Grade, Fisher Scientific) with heated (60 �C) sonication for1 h in sealed, new polypropylene centrifuge tubes and diluted foranalysis. Calibration standards were prepared from NIST-traceablestock solutions (Spex Certiprep, Methuen, NJ) from concentrationsof 0.01 mg/L upward with three levels per order of magnitude (e.g.0.02, 0.05, 0.1, etc.) and verified with a standard reference material(SRM) NIST 1643e (NIST, Gaithersburg, MD). Blank and controlsamples were measured and found to have negligible concentra-tions. Relative standard deviations on detected concentrationswere less than 5%, as was agreement with the SRM.

3. Results and discussion

This section of the paper is organized to present the results anddiscussion associated with: (1) dissolved Cr concentration ofleachate from all pavements; (2) variability of Cr concentrationwith respect to time (e.g., first flush effect); (3) impact of pH on Cr

M. Kayhanian et al. / Journal of Environmental Management 90 (2009) 3574–3580 3577

leachability from pavements; (4) the potential source of Cr inconcrete pavements; and (5) Cr concentration with respect to waterquality standards.

3.1. Dissolved Cr concentration of leachate from all pavements

The measured dissolved chromium concentrations (e.g., alldetected values) for all samples tested from fresh and aged pave-ment specimens for three different experimental temperatures arepresented in Figs. 2a,b, respectively. Evaluating data from Fig. 2, it isimportant to note that: (1) the concentration of total suspendedsolids (TSS) in all tested leachate waters were below the reportinglimit of 1 mg/L and filtration of these leachate samples produced nomeasurable solids and hence the results are reported as dissolved;and (2) most dissolved Cr concentrations were observed to be quitelow and could not be depicted meaningfully on a linear scale andhence the concentration data are shown on a logarithmic scale. Asshown in Fig. 2, nearly all of the fresh and aged asphalt pavementspecimens did not produce chromium concentrations above thereporting limit of 1 mg/L. Nearly all elevated Cr concentrations wererelated to the two concrete specimens, which were produced usingthe same cement source. Various parameters such as temperature,aging, specimen permeability were found to impact the leachate Crconcentration. To assess the statistical significance of theseparameters on Cr concentration a t-test was performed. Under thistest a 90% confidence interval (p < 0.1) was used to determine the

Fig. 2. Time series Cr concentrations for all fresh and aged leachate samples tested undspecimens without going through aging process via heat treatment). (b) Aged specimenstreatment).

reliability of the hypothesis that the difference among experi-mental results is real and not due to chance. The results of the t-testrevealed that the change in chromium concentration between freshand aged concrete pavement specimens was statistically significant(p ¼ 0.1). In addition, it was statistically determined that all densegraded (impermeable) concrete specimens had significantly higherchromium concentrations than open graded concrete specimensduring the initiation of the experiment. The higher concentrationmay be explained by the fact that dense graded specimens hadmore surface contact than open graded specimens. However, ingeneral, no significant difference was observed in Cr concentrationof leachate samples obtained under the three different controlledtemperatures.

3.2. Variability of dissolved Cr concentration with respect to time(e.g., first flush effect)

To determine the variability of Cr concentration with respect totime, all 12 samples that were obtained from each concrete spec-imen during the first 8 h of the experiment were separatelyanalyzed and the results for both fresh and aged specimens areshown in Fig. 3. The result presented in Fig. 3 is the average of alltested samples for both triplicated I and J concrete specimens underthe three different temperatures. As shown, the leachate dissolvedCr concentration gradually decreased as the experiment pro-gressed. Similar trend for Cr (VI), with much lower concentration,

er three different temperatures. (a) Fresh specimens (triplicate asphalt and concrete(triplicate asphalt and concrete specimens that went through aging process via heat

0

5

10

15

20

25

30

35

40

45

0 1 2 3 4 5 6 7 8Time (hours)

Co

ncen

tratio

n (p

pb

)

Fresh SpecimensAged Specimens

Fig. 3. Time-variability of average dissolved Cr concentration for leachate producedfrom concrete specimens under controlled laboratory study.

Table 2Dissolved and particulate Cr concentration for ten pavement specimens.

Pavementspecimena

Cr concentration (mg/L)

Dissolved Particulate % differenceb

A 0.10 6.39 98B 0.09 0.54 83C 0.24 0.68 65D 0.18 0.39 54E 0.32 0.39 18F 0.20 0.31 35G 0.18 0.32 44H 0.17 0.28 39I 26.3 33.1 21J 15.3 133.9 89

a For description of specimen mix type and binder see Table 1.b Example calculation: 100 � (6.39 � 0.10)/6.39 ¼ 98%.

M. Kayhanian et al. / Journal of Environmental Management 90 (2009) 3574–35803578

was observed by Takahashi et al. (2007). In general, the highestconcentration of Cr was measured during the early samples and theconcentration of dissolved Cr decreased to about 91% after 8 hcompared to the initial sample. This decreasing trend in Crconcentration may be due to the first flush effect as has beenobserved during actual highway runoff characterization studies(Sansalone and Buchberger, 1997; Han et al., 2006; Lau et al., 2009).Beside the concentration first flush, the California study has alsoshown the mass first flush effect of total and dissolved Cr indicatingthat more of the Cr mass is associated with early portion of therunoff volume (e.g., 10, 20, or 30%) compared to the later portion ofthe storm event volume.

Table 3Total Cr concentration from various cement sources.

Cementbranda

Total Crconcentration (mg/L)b

Ratio relative tocement brand L

K 124 0.19L 641 1.00M 552 0.86O 130 0.20P 135 0.21

a Representative cements type I/II used in construction of concrete pavement.b Average of triplicate samples.

3.3. Impact of pH on Cr leachability from pavements

A statistical analysis was performed to evaluate whethersignificant differences existed between results obtained fromdifferent pH adjusted milli-Q waters. The result of this analysisshowed that there was insignificant (e.g., p > 0.10) difference indetected chromium concentration. However, the change inconcentration of other metals such as Cu, Fe, Pb, Ni and Zn found tobe significant. The increase in concentration for these metals underadjusted pH conditions was mostly attributed by corrosion of thesteel trays since the leachate produced from this experimentshowed elevated concentrations of iron, nickel and to a smallerextent chromium (11.9, 0.32 and 0.10 mg/L, respectively). Theconcentrations of other metals (As, Cd, Cu, Zn, Pb and V) were allbelow the reporting limit. These results confirmed that the majorityof detected iron, nickel and chromium at low pH were due to thecorrosion of the experimental trays. Generally, the concentrationsof chromium at different pHs were observed to be statisticallysimilar (i.e., within a 10% difference in mean for triplicate samples).These results are in good agreement with study performed byErdem and Ozverdi (2008), in which they concluded that theconcentration of Cr was not significantly affected by pH.

The influence of adjusted pH milli-Q water on particulate Crconcentration was also found to be insignificant. Nevertheless, asshown in Table 2, the particulate Cr concentration was higher thandissolved Cr and the difference between them ranged from 18 to98%. In addition, it is important to note that (except for specimen A)the concentrations of all particulate Cr in asphalt pavements werebelow 1 mg/L. A significantly higher particulate Cr concentrationwas also measured under field highway runoff characteristics. Thishigher particulate Cr concentration from field runoff is mostly dueto a much higher TSS concentration (average TSS ¼ 130 mg/L).However, majority of TSS measured from field runoff is not directlyrelated to pavement materials.

3.4. Potential source of Cr in concrete pavements

As shown previously, an elevated Cr concentration was observedin concrete pavement specimens. To determine the source of Cr inconcrete specimens an equivalent ratio of Milli-Q water with sepa-rate samples of the aggregate source and with portland cement usedfor extraction over a period of 48 h. The results of this test revealednegligible Cr associated with aggregate; all dissolved Cr was relatedto the portland cement. To further test the differences in Crconcentration for both hardened cement paste (cement þwaterwithout aggregate) and cement powder, an appropriate ratio ofwater and cement was extracted for analysis. The results showedthat cement paste, on average, produced 13% lower dissolved Crconcentration compared with cement powder.

Based on the results shown in this study, it can be concludedthat the primary source of chromium in leachate produced fromconcrete specimens is the cement. However, all earlier resultspresented in this paper were based on only one cement brand L (seeTable 3) purchased from a major distributor in California. Hence,testing different brands of cement will provide an indication ifthere is a variation in Cr concentration with respect to differentcement sources. Five commercial cement brands type I/II used inCalifornia were tested and the result of total Cr produced from theleachate of pavement surfaces using different cement sources isshown in Table 3. As shown, total Cr concentrations in differentcement brands varied significantly. A review of different cementbrands from Europe have also shown a significant variation in Cr(VI) concentration ranging from 0.1 to 2.8 mg/kg with majority inthe range of 0.2–0.8 mg/kg (Kristiansen et al., 1997). Because largevariation exists among different sources of cement, this fact alonewill present an opportunity to reduce the potential chromiumcontribution in stormwater runoff from concrete paved surfaceswithout using expensively constructed treatment best manage-ment practices (BMPs). For instance, using the results presented inTable 3, the use of cement brand P compared with cement brand Lcan likely reduce Cr discharge to the environment five fold. Furtherreduction of dissolved Cr or Cr (VI) to the environment can be

M. Kayhanian et al. / Journal of Environmental Management 90 (2009) 3574–3580 3579

accomplished through chemical processing. Chemical processing tohinder the release of Cr (VI) was beyond the scope of our study.However, various chemical processes have been used to render therelease of Cr (VI) to the environment including, but not limited to:organosilica sol-gel, ferrous iron, ferrochromium slag, iron sulfate,photoreduction, H2O2 in acidic solution and humic acid (Eary and Rai,1998; Deshpande et al., 2005; Erdema et al., 2005; Fregert et al.,1979;Geelhoed et al., 2003; Kieber and Helz, 1992; Pettine et al., 2002;Peysson et al., 2005; Xu et al., 2004). Among the processes mentionedabove, the addition of ferrous sulfate to cement mix is more commonand seems to be practical and effective (Chou et al., 2008).

3.5. Significance of experimental result with respect to waterquality standards and field data

Dissolved Cr, especially Cr (VI) is a pollutant of concern in pro-tecting the receiving waters in the United States. In California, thedischarge of Cr to receiving water is regulated under the CaliforniaToxic Rule (CTR) (USEPA, 2000). Under CTR, the maximum Cr (VI)concentration is set at 16 mg/L. The CTR also specifies a continuous(4 day average) Cr (VI) maximum concentration at 11 mg/L. Theindividual detected dissolved chromium concentration from theleachate of concrete specimens I and J under this study comparedwith the maximum CTR criterion is shown in Fig. 4. It is importantto note that nearly all higher concentrations of dissolved Cr wererelated to initial leachate sample analysis (e.g., five samples withinfirst hour of the experiment). As shown, the majority of the datapoints for concrete (open and gap graded) pavement specimenswere observed to be above the CTR limit. It is also important to notethat the oxidation state of chromium was not measured, so theexact proportion of Cr (VI) in dissolved Cr is not known. However,according to the pH-pe relationship for Cr it is expected that a largefraction of dissolved Cr measured under this study at near neutralpH and pe ranging between 0 and 11 be in the form of Cr (VI). In onerecent leachate study performed by Takahashi et al. (2007), the Cr(VI) concentration in hardened concrete leachate has been reportedfrom 5.5 to 14.4 mg/L. Nevertheless, it is reasonable to assume thatonly fraction of dissolved Cr concentration measured during thecontrolled ‘initial’ leachate investigation will be measured in actualroadway runoff. It is clear that after one rainfall simulation eventthe dissolved Cr concentration where substantially reduced tobelow CTR limit after the 3 h. It is reasonable to expect that the Crleachability from concrete pavements will gradually diminish witholder roads and additional rainfall events. This statement canpartially be justified by evaluating the individual dissolved Cr event

Triplicate concrete pavement specimens tested

at 4, 20, and 45°C

0.1

1

10

100

4 Degree C 20 Degree C 45 Degree C

Co

ncen

tratio

n (p

pb

)

Specimen ISpecimen JCTR limit

Reporting limit

Fig. 4. Individual dissolved Cr concentration in concrete pavement specimens withcement brand S compared with the CTR limit for Cr(VI).

mean concentration (EMC) obtained from 34 highway sites duringthe 3-year statewide highway runoff characterization (2000–2003)in the state of California (Kayhanian et al., 2007). The result of thisstudy showed that, only 20 out of 460 storm events (less than 5%)produced dissolved Cr concentration exceeding the CTR limit. Inaddition, during the 2001 monitoring season all three forms of Crconcentration (total, dissolved and Cr VI) were measured from foururban highway sites (annual average daily traffic >100,000 vehicle/day) in Los Angles area with concrete paved surfaces. Result of thisfield runoff characterization showed that concentration of Cr (VI)ranged from non-detect (0.01 mg/L) to a maximum value of 1.6 mg/Land the results of both dissolved Cr and Cr (VI) for all four sitesremained below CTR limit. The much lower total and dissolved Crconcentration obtained from actual highway runoff compared withlaboratory testing is another indication of dissipating Cr leach-ability from concrete pavements for older roads and numerousrainfall occurrences. While the majority of field measurement for Crmeet the CTR water quality standard, it is not clear, however, if anyof the small number of dissolved Cr concentration exceeded theCTR values were solely related to pavement materials or insteaddue to other sources. Addressing this question is beyond the scopeof this study and needs to be resolved with further research.

4. Conclusions

Major conclusions drawn from this study include:

1. Compared to other metal constituents, an elevated concentra-tion of Cr was found in leachate of concrete (but not asphalt)pavement materials.

2. Leachability of Cr concentration in concrete pavement speci-mens is influenced by pavement age, permeability, tempera-ture, contact time and to a lesser degree by pH of simulatedrunoff.

3. First flush effect was observed in which the concentration of Crwas much higher during the first hour compared with the restof experimental duration.

4. The source of the chromium in concrete pavement specimens’leachate was determined to be the portland cement, not theaggregate.

5. Not all cements have the same concentration of chromium andthe difference of up to 5 fold have been measured.

6. The concentration of dissolved Cr leached from hardenedconcrete specimens mixed with aggregates was substantiallylower than the Cr concentration measured from the leachate ofraw cement.

7. The concentration of dissolved Cr in leachate of concretespecimens substantially decreased after 8 h of rainfall simula-tions. From this results we can conclude that, except for freshlyconstructed concrete pavement and during the first few eventsas well as the early portion of the rain event, dissolved Cr andCr (VI) is not a constituent of concern with respect to surfacepavement materials alone.

Acknowledgements

This study was partially funded by the California Department ofTransportation, Division of Environmental Analysis (DEA), State ofCalifornia, Business and Transportation Agency. Additional fundingwas provided through a UC Davis Academic Federation ProfessionalDevelopment Award. We are grateful to the staff of the University ofCalifornia Pavement Research Center (UCPRC) for fabrication ofexperimental apparatus and preparing the pavement specimenstested under this study and to Mr. Chris Alaimo for chemicalanalysis and environmental laboratory management.

M. Kayhanian et al. / Journal of Environmental Management 90 (2009) 3574–35803580

DISCLAIMER: The contents of this paper reflect the views of theauthors, who are responsible for the facts and the accuracy of thedata presented herein. The contents do not necessarily reflectthe official views or policies of the State of California. This paperdoes not constitute a standard, specification, or regulation.

References

Ball, J.E., Jenks, R., Aubourg, D., 1998. An assessment of the availability of pollutantconstituents on road surfaces. The Science of the Total Environment 209 (2),243–254.

Bell, C.A., 1989. Summary Report on Aging of Asphalt-aggregate Systems, StrategicHighway Research Program Project, report no. SHRP-A-305.

Birgisdottir, H., Gamst, J., Christensen, T.H., 2007. Leaching of PAHs from hot mixasphalt pavements. Environmental Engineering Science 24 (10), 1409–1422.

Blackburn, R.G., Kriech, J.A, Kurek, J.T., Osborn, L.V., 1992. Detection of coal tarmaterials in asphalt pavements by using chemical and biological methods.Transportation Research Record, paper no. 99-0336, 41–45.

Brandt, H.C.A., de Groot, P.C., 2001. Aqueous leaching of polycyclic aromatichydrocarbons from bitumen and asphalt. Water Research 35 (7), 4200–4207.

Chou, T.-C., Chang, H.-Y., Chen, C.-J., Yu, H.-S., Wu, J.-D., Sheu, S.-C., Shin, T.-S., 2008.Effect of hand dermatitis on the total body burden of chromium after ferroussulfate application in cement among cement workers. Contact Dermatitis59 (4), 151–156.

COSHH, 2007. The Control of Substances Hazardous to Health (Amendment)Regulations. http://www.hse.co.uk/consult/condocs/cd195.htm.

Councell, T.G., Duckenfield, K.U., Landa, E.R., Callender, E., 2004. Tire-wear particlesas a source of zinc to the environment. Environmental Science and Technology38 (15), 4206–4214.

Davis, A.P., Shokouhian, M., Ni, S., 2001. Loading estimates of lead, copper, cadmium,and zinc in urban runoff from specific sources. Chemosphere 44 (5), 997–1009.

Deshpande, K., Chueng, S., Rao, M.S., Dave, B.C., 2005. Efficient sequestration andreduction of hexavalent chromium with organosilica sol-gels. Journal ofMaterial Chemistry 15 (29), 2997–3004.

Eary, L.E., Rai, D., 1998. Chromate removal from aqueous wastes by reduction withferrous iron. Environmental Science and Technology 22 (3), 972–977.

Erdem, M., Ozverdi, A., 2008. Leaching behavior of chromium in chrome shavinggenerated in tanning process and its stabilization. Journal of Hazardous Mate-rials 156 (4), 51–55.

Erdema, M., Altundogan, H.S., Turan, M.D., Tumen, F., 2005. Hexavalent chromiumremoval by ferrochromium slag. Journal of Hazardous Materials B126, 176–182.

Fregert, S., Gruvberger, B., Sandahl, E., 1979. Reduction of chromate in cement byiron sulfate. Contact Dermatitis 5 (1), 39–42.

Frias, M., Sanchez de Rojas, M.I., Garcia, N., Luxan, M.P., 1994. Contribution of toxicelements: hexavalent chromium in materials used in the manufacture ofcement. Cement and Concrete Research 24 (3), 533–541.

Geelhoed, J.S., Meeussen, J.C.L., Roe, M.J., Hillier, S., Thomus, R.P., Farmer, J.G.,Paterson, E., 2003. Chromium remediation or release? – Effect of Fe(II) sulphateaddition on Cr(VI) leaching columns of chromite ore processing residue. Envi-ronmental Science and Technology 37 (14), 3206–3213.

Han, Y., Lau, S.L., Kayhanian, M., Stenstrom, M.K., 2006. Characteristics of highwaystormwater runoff. Water Environmental Research 78 (12), 2377–2388.

Hewitt, C.N., Rashed, M.B., 1990. An integrated budget for selected pollutants fora major rural highway. Science of The Total Environment 93, 375–384.

Hillier, S.R., Sangha, C.M., Plunkett, B.A., Walden, P.J., 1999. Long-term leaching oftoxic trace metals from portland cement concrete. Cement and ConcreteResearch 29 (4), 515–521.

Kayhanian, M., Suverkropp, C., Ruby, A., Tsay, K., 2007. Characterization andprediction of highway stormwater pollutant event mean concentrations. Envi-ronmental Management 85 (2), 279–295.

Kieber, R.J., Helz, G.R., 1992. Indirect photo reduction of aqueous chromium (V1).Environmental Science and Technology 26 (2), 307–312.

Kriech, A.J., 1990. Evaluation of hot mix asphalt forleachability. http://www.asphaltinstitute.org/Upload/Evaluation_HMA_For_Leachability.pdf.

Kriech, A.J., 1992. Leachability of asphalt and concrete pavements. http://www.asphaltinstitute.org/Upload/Leachability_Asphalt_and_Concrete_Pavements.pdf.

Kriech, A.J., Kurek, J.T., Osborn, L.V., Wissel, H.L., Sweeney, B.J., 2002. Determinationof polycyclic aromatic compounds in asphalt and in corresponding leachatewater. Polycyclic Aromatic Compounds 22 (3), 517–535.

Kristiansen, J., Molin, J., Christensen, K., Byrialsen, K., 1997. A DANREF certifiedreference material for chromate in cement. Analyst 122 (10), 1155–1159.

Lau, S.L., Han, Y., Kang, J.-H., Kayhanian, M., Stenstrom, M.K., 2009. Characteristics ofHighway Stormwater Runoff in Los Angeles: Metals and PAHs. Water Envi-ronment Research 81 (3), 308–318.

Legret, M., Pagotto, C., 1999. Evaluation of pollutant loadings in the runoff watersfrom a major rural highway. Science of The Total Environment 235 (1–3),143–150.

Legret, M., Odie, L., Demare, D., Jullien, A., 2005. Leaching of heavy metals and PAHsfrom reclaimed asphalt pavement. Water Research 39 (15), 3675–3685.

Lide, D.R., 1998. Chemical Fact Sheet: Chromium, Hand Book of Chemistry andPhysics, 78th Edition, CRC Press, Boca Raton, FL, pp. 7440–7447.

Lindgren, A., 1996. Asphalt wear and pollution transport. Science of The TotalEnvironment 189/190, 281–286.

Muschack, W., 1990. Pollution of street runoff by traffic and local conditions. TheScience of the Total Environment 93 (4), 419–431.

Norin, M., Stromvall, A.-M., 2004. Leaching of organic contaminants from storage ofreclaimed asphalt pavement. Science of The Total Environment 25 (3), 323–340.

Pettine, M., Campanella, L., Millero, F.J., 2002. Reduction of hexavalent chromium byH2O2 in acidic solutions. Environmental Science and Technologies 36 (5), 901–907.

Peysson, S., Pera, J., Chabannet, M., 2005. Immobilization of heavy metals by calciumsulfoaluminate cement. Cement and Concrete Research 35 (12), 2261–2270.

Sanchez, F., Langley White, M.K., Hoang, A., 2009. Journal of. EnvironmentalManagement 90 (2), 983–993.

Sansalone, J.J., Buchberger, S.G., 1997. Partitioning and first flush of metals in urbanroadway storm water. Journal of Environmental Engineering 123 (2), 134–142.

Signore, J.M., Kayhanian, M., Harvey, J., Troxler, M.P., Jones, D., 2008. An experi-mental apparatus to evaluate the water quality of leachate generated fromvarious pavement surfacing materials. In: International ISAP Symposium onAsphalt Pavements and Environment, August 18–20, Zurich, Switzerland.

Takahashi, S., Sakai, E., Sugiyama, T., 2007. Study of leaching of hexavalent chro-mium from hardened concrete using tank leaching test. Journal of AdvancedConcrete Technology 5 (2), 201–207.

USEPA, 2000. Water quality standards; establishment of numeric criteria forpriority toxic pollutants for the state of California; rule. Federal Register 65 (97),31682–31719.

Vashisth, P., Lee, K.W., Wright, R.M., 1997. Assessment of water pollutants fromasphalt containing recycled rubber in Rhode Island, Transportation ResearchRecord 1626. Paper no. 97-1299, 95–104.

Xu, X., Li, H., Li, X., Gu, J., 2004. Reduction of hexavalent chromium by ascorbic acidin aqueous solutions. Chemosphere 57 (7), 609–613.