Top Banner
Leaching Characteristics of Recycled Asphalt Pavement Used as Unbound Road Base May 2012 Student Investigators: Ryan F. Shedivy & Amara Meier Advisors: Tuncer B. Edil, James M. Tinjum, & Craig H. Benson University of Wisconsin-Madison UNIVERSITY OF WISCONSIN SYSTEM SOLID WASTE RESEARCH PROGRAM Student Project Report
20

Leaching Characteristics of Recycled Asphalt Pavement Used as ...

Jan 22, 2017

Download

Documents

dodiep
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Leaching Characteristics of Recycled Asphalt Pavement Used as ...

Leaching Characteristics of Recycled Asphalt Pavement Used as

Unbound Road Base

May 2012

Student Investigators: Ryan F. Shedivy & Amara Meier

Advisors: Tuncer B. Edil, James M. Tinjum, & Craig H. Benson

University of Wisconsin-Madison

UNIVERSITY OF WISCONSIN SYSTEM SOLID WASTE RESEARCH PROGRAM Student Project Report

Page 2: Leaching Characteristics of Recycled Asphalt Pavement Used as ...

2

Leaching Characteristics of Recycled Asphalt Pavement Used

as Unbound Road Base

Ryan Shedivy1, Amara Meier

1, Junwei Ma

1,2, James M. Tinjum

1, Tuncer B. Edil

1, Craig H. Benson

1,

Jiannan Chen1, Sabrina Bradshaw

1

1Department of Geological Engineering, University of Wisconsin-Madison, Madison, WI, USA

2School of Environment, Beijing Normal University, Beijing, China

1. Introduction

The use of recycled asphalt pavement (RAP) as road base material is an increasing trend in the road

construction business. Use of RAP will reduce the amount of solid waste disposed in landfills and

provide more sustainable construction due to the use of in-situ materials and the lower transportation

cost.

RAP contains natural aggregate and bituminous asphalt, a material that contains heavy metals and

poly-aromatic hydrocarbons (PAHs). Heavy metals and PAHs are pollutants that have been identified

as carcinogenic, mutagenic, and teratogenic. When subjected to rain water, these heavy metals and

PAHs have the ability to leach out of the road base and infiltrate into the water table, potentially

impacting the quality of drinking water. Although its use in road construction projects as an unbound

base is increasing, environmental impacts of its use have not been thoroughly investigated.

In this project, the leaching characteristics of PAHs and heavy metals from five different sources of

RAPs will be investigated.

2. Background

Heavy metals are encountered in various emission sources related to automobiles. Zinc and cadmium

are deposited mainly through tire wear and corrosion of galvanized steel crash barriers, and brake line

wearing constitutes a source of copper (Muschack, 1990; Hewitt and Rashed, 1990). The heavy metal

contamination of highway runoff water and roadside soils has been reported (Warren and Birtch, 1987;

Strecker et al., 1990; Pagotto et al., 2000; Han et al., 2009).

Polycyclic aromatic hydrocarbons are a group of chemicals derived primarily from the incomplete

burning of organics. Some researchers consider sources of PAHs are traffic-related such as vehicle

Page 3: Leaching Characteristics of Recycled Asphalt Pavement Used as ...

3

exhaust, lubricating oils, gasoline, diesel fuel and tire particles (Takada et al., 1990; Baek et al., 1991;

Sadler et al., 1999; Brandt et al., 2001; Kriech et al., 2002). Brantley and Townsend (1999) reported

the leaching of pollutant in six samples of RAP collected from asphalt plants throughout Florida.

Results from batch tests and column experiments indicated that the RAP samples investigated did not

leach chemicals with greater quantities than typical groundwater standards. None of the 16 EPA PAHs

were found to lie above the detection limit that ranged between 0.25 and 5 mg/L. Leachate collected

during column studies did not contain levels of PAHs, VOCs, or selected heavy metals greater than

typical groundwater concentration, except for lead in a RAP sample from an older roadway. This lead

concentration was slightly above the drinking water standards, but this concentration diminished over

time.

Legreta and Odieb (2005) presented the possible leaching of pollutants from RAP procured from a

reconstruction road site located on France’s RN76 highway. Samples were tested in both static batch

tests and column leaching tests. They considered the leaching of pollutants to be rather weak for most

of the parameters studied. Concentrations in solutions from batch leaching tests were generally below

the European community limit values for drinking water. Pollutant concentrations from column

experiments were higher in solutions from the initial leaching stages, but then decreased rapidly and

were at values below the detection limits.

3. Materials and methods

3.1. Sampling

Leaching test for five sources of RAP: Ohio, Wisconsin, California, New Jersey, Colorado, and one

new asphalt material were performed. The new RAP material was acquired from Wingra Stone

Company in Madison, WI. The new material had been crushed to 1.25 inches and picked up at random

from a stockpile using a shovel. The new RAP is termed WG RAP.

For comparative purposes to real RAP, new conventional asphalt material was sampled from an asphalt

lab for batch leaching test and column leaching test. This material consisted of a mixture of natural

aggregates (95%) and bitumen (5%) and was compressed similar to road asphalt (Figure 1).

For batch leaching tests and column leaching tests, each sample was homogenized respectively and

screened at 19.1 mm. New conventional asphalt material were crushed and then sieved through a

19.1mm stainless steel sieve.

Page 4: Leaching Characteristics of Recycled Asphalt Pavement Used as ...

4

Figure 1: New asphalt material was crushed

3.2. Leaching Test

Batch leaching tests and column leaching tests were performed for this project.

3.2.1. Batch leaching experiments

Batch leaching tests were performed to assess chemical leaching potential from RAP according to

existing regulatory protocols. The toxicity characteristic leaching procedure (TCLP) and a deionized

(DI) water leaching procedure were performed. The TCLP test was performed to find out the quantity of

pollutant generated should the material be exposed to extreme conditions. TCLP fluid #1 was used as

leaching fluid. The DI water leaching procedure was conducted in the same manner as the TCLP, but DI

water was used as the leaching fluid.

In accordance with EPA Method-1311, a 140-g specimen of material, preliminarily crushed to a grain

size below 19.1 mm, was exposed to a 24 hour extraction test. These extractions were performed using

deionized water or TCLP solution (fluid #1) under continuous stirring and with a liquid/solid (L/S) ratio

of 20. For NJ RAP, different L/S ratios were performed. 2.5 liter amber glass jars with Teflon lids were

used.

Page 5: Leaching Characteristics of Recycled Asphalt Pavement Used as ...

5

Figure 2: Batch leaching test

3.2.2. Column leaching experiments

Due to time constraints, this test has not been performed yet. Column leaching experiments will be

performed on RAP samples to simulate more realistic environmental conditions in the field. Column

leaching tests are planned to be performed in stainless steel columns. The samples, sieved at a grain size

below 19.1 mm and adjusted to optimal water content according to compaction tests, will be loaded into

columns 15.5 cm in diameter and 12.5 cm high. DI water will be pumped by a multi-channel peristaltic

pump from the storage tank to the columns at a constant flow rate of 301 cm3/day. The columns will be

percolated from the bottom to the top in order to minimize trapping of air bubbles. Synthetic rain water

will be used for some RAPs and the results will be contrasted with DI water. The leachate will be

analyzed for PAHs and heavy metals.

Page 6: Leaching Characteristics of Recycled Asphalt Pavement Used as ...

6

Figure 3: Sketch map of leaching column

Figure 4: Column leaching test apparatus

Filter screen:

S.S. or

fiberglass, 300 µm

Teflon

RAP

Teflon bag:

prevents

exposure to

atmosphere

Teflon bag

Page 7: Leaching Characteristics of Recycled Asphalt Pavement Used as ...

7

3.2. Extraction and clean-up of PAHs

After agitated continuously for 24 hours, the leachate was cloudy. All the leachate (about 2.7L) was

flowed through a filter and the bulk of the aqueous phase from the solid phase was separated.

Filtration was performed through a borosilicate glass funnel with a flat, fritted base and coarse filter

paper. Then two liters of water samples were chosen to be extracted.

Dissolved PAHs in leachate were extracted using solid phase extraction eisks based on Sigma-Aldrich

Corporation application method (Extract Polynuclear Aromatic Hydrocarbons). The sample’s pH

was adjusted to less than two with 6N hydrochloric acid. To avoid adsorption of PAHs upon

glassware, 5% (v/v) methanol or 5% (v/v) isopropanol was added to 2 L of leachate and the solution

was mixed thoroughly. The solution was spiked with 5ml chrysene-d12 (concentration 20ppb) and

aceneph-d10 (concentration 20ppb) separately as surrogate. 90mm glassware was used (flask,

vacuum line, and filtration support). A 90mm ENVITM-18 DSK disk was placed on the apparatus

for support. The disk was cleaned with 10mL methylene chloride and the liquid was drawn through

the disk under moderate vacuum. This process was repeated twice to increase the recovery of PAH.

The SPE disk was conditioned with 15mL of methanol followed by 15mL distilled water. 15mL

methanol was poured and a low vacuum was applied. The vacuum was then released when the

methanol is just above the top surface of the disk. The same procedure was performed with DI

water.

Figure 5 Extraction Test

The 2L sample was percolated through the SPE disk at a flow rate of 100 ml/min. After the entire

sample had been processed, the disk was dried under vacuum. Then the sample collection tube was

inserted and the analyte was eluted with 10 milliliters of acetonitrile twice. This elution procedure

was repeated with methylene chloride and all eluates were then combined in the sample collection

tube. The remaining water was removed from the eluate by passing it through approximately 5g

Page 8: Leaching Characteristics of Recycled Asphalt Pavement Used as ...

8

anhydrous sodium sulfate. The extract was concentrated by a rotary evaporator to 5ml volume for

analysis.

Figure 6 Rotary Evaporation

3.3. Analysis

The leachates collected from the batch tests and column tests were analyzed for a number of chemical

parameters, including heavy metals and PAHs. Heavy metal determinations in leachate were

performed by Inductively Coupled Plasma Optical Emission Spectrometry (ICP-OES). PAH

concentrations were measured after filter and SPE extraction. Experimental procedures for the

extraction, purification and determination of PAHs were adapted from application of extraction of

PAHs from water, using solid phase extraction disk by SUPELCO Company with minor

modifications.

The PAHs were analyzed using high performance liquid chromatograph (HPLC) (Shimadzu) with a

fluorescence (Shimadzu) detector. A 3.2mm x 150mm symmetric C18 column (RESTEK) was used

as the stationary phase. The mobile phase was a mixture of acetonitrile and water. Linear gradient

elution was adapted, mixture of acetonitrile (A)/water (B) at 1 ml/min; the initial composition (40%

(A)) was held for 5 min and then increased to 100% over a period of 27.5 min. The injection volume

was 20 µl. The detection wavelength program (excitation/emission (nm)) and the retention time for

PAH are shown in Table 1.

Page 9: Leaching Characteristics of Recycled Asphalt Pavement Used as ...

9

Table 1: The detection wavelength program (excitation/emission (nm))

and the retention time for PAH

Retention Time

225/340

Naphthalene 8.395

1-Methylnaphthalene 12.29

2-Methylnaphthalene 12.92

Acenaphthene 13.475

Fluorene 14.09

246/368 14.5

Phenanthrene 15.275

Anthracene ? 16.245

280/462 16.6

Fluoranthene 17.45

236/396 17.6

Pyrene 18.075

261/384 19.1

Benzo(a)anthracene 20.465

Chrysene 20.67

280/462 21.8

Benzo(b)fluoranthene 22.35

Benzo(k)fluoranthene 22.905

Benzo(a)pyrene 23.605

290/401 24.4

Dibenzo(a,h)anthracene 24.775

Benzo(ghi)perylene 25.445

Indeno(1,2,3-cd)pyrene 26.21

The QA/QC included:

(a) Laboratory quality control procedures include analyses of sample blanks, reference

material and spiked samples. The EPA Method 8310 PAH Mixture standard solution

was supplied by RESTEK. To affirm the recovery of SPE extraction, chrysene-d12 and

aceneph-d10 were chosen as surrogates. They were bought from Sopelco.

(b) The correlation coefficients for calibration curves of PAHs were all higher than 0.999.

One example (Acenaphthalene) is shown in Figure.6.

(c) The recovery of PAH were between 76% and 100% of the certified values.

Page 10: Leaching Characteristics of Recycled Asphalt Pavement Used as ...

10

Figure 6: PAH calibration curve of Acenaphthalene (Goodness of fit: 0.9992588)

4. Results and Discussion

The results reported include the physical characterization and the results from the TCLP and DI batch

leaching tests.

4.1 Physical characterization result

The RAP samples collected from each site were physically characterized at our lab. The analytical

tests include moisture content, asphalt content, specific gravity, compaction characterization

(modified compaction test), and hydraulic conductivity.

Table 2: Basic properties of RAP

Material Source Asphalt

content

%

Specific

gravity

%

Compaction Void

Ratio, q

(Bulk)

Hydraulic

Conductivity

(cm/sec) Wopt (%) γdmax(kN/m3)

Colorado RAP 5.93 2.23 5.7 20.65 0.06 3.82E-03

Ohio RAP 6.20 2.43 8.8 19.82 0.20 8.32E-03

California RAP 5.70 2.57 6.2 21.16 0.19 2.19E-03

Wisconsin RAP 4.78 2.41 6.1 20.26

New Jersey RAP 5.20 2.37 6.5 20.39 0.14 3.69E-02

Page 11: Leaching Characteristics of Recycled Asphalt Pavement Used as ...

11

4.2 Metal leaching in batching leaching experiments

Batch test experiments were carried out on both the RAP and new conventional asphalt material. pH,

Electrical Conductivity (EC) and Oxidation/Reduction Potential (ORP) of batch leaching test are shown

in Table 3. Minimum Contaminant Level (MCL) and Method Detection Limit (MDL) of metal with

ICP-OES are shown in Table 4. The metal leaching results are shown in Table 5.

Table 3: pH, EC and ORPof batch leaching test

pH EC (µs/cm) ORD (mV)

OH DI 9.48 541 294.9

CA DI 8.59 229 259.9

CO DI 8.77 236 279

NJ DI 8.98 186 396.1

WI DI 9.58 235 104.4

New Asphlat DI 9.57 263 312.9

OH TCLP 7.35 16,000 313.5

CA TCLP 5.16 11,400 353

CO TCLP 7.35 16,000 313.5

NJ TCLP 5.2 12,000 225.4

WI TCLP 6.82 14,200 303.2

New Asphalt TCLP 5.02 12,000 267.7

*DI: DI water as leaching solution

*TCLP: TCLP leaching test

Page 12: Leaching Characteristics of Recycled Asphalt Pavement Used as ...

12

Table 4: MCL and MDL of metal with ICP-OES

Element

MCL(ug/L) in drinking

water MDL(ug/L)

AL 50-200** 2

Fe 0.3** 2

Mg - 1

K - <1000

Si - <1000

Sr 4000*** <1

Ba 2 0.08

Be 4 0.11

Cd 5 0.53

Cr 100 0.3

Cu 1000** 2.7

Pb 15 3.8

Mn 50 0.13

Mo 40*** 3.3

Ni 100*** 1.8

Zn 5000 0.5

As 10 28.6

Page 13: Leaching Characteristics of Recycled Asphalt Pavement Used as ...

13

Table 5: Metal result of batch test carried out on RAP sample

Concentrations (ug/L)

Ag As Cr Cd Cu Pb Ni Zn Mn Mo Al Ba Fe K Mg Sb Si

OH DI 0.30 9.32 Uv Uv 4.93 Uv Uv 30.73 1.50 6.25 22.95 28.31 uv 1679.20 11.00

CA DI 0.25 11.69 Uv Uv 2.43 Uv Uv 10.04 5.01 5.99 100.08 30.49 72.47 2461.15 x 10.92 5233.30

CO DI 0.30 32.40 Uv Uv 4.06 Uv Uv 3.52 6.00 108.43 30.48 71.22 3326.04 877.77 14.05 4121.00

NJ DI 0.29 uv Uv Uv 2.12 Uv Uv 31.30 3.26 5.56 149.69 40.12 85.33 3226.80 965.65 7.48 3317.33

WI DI 0.10 32.17 Uv Uv 0.40 Uv Uv 16.24 0.65 5.28 21.65 21.19 uv 1497.16 x 17.24 x

New Asphalt

DI 0.28 10.47 Uv Uv 2.79 Uv Uv 50.84 3.24 6.15 469.58 16.36 33.61 2950.99 185.06 10.54 2576.21

OH TCLP 0.44 82.79 Uv Uv 6.91 Uv 7.24 12.93 217.75 7.22 8.26 217.43 uv x x 14.97 x

CA TCLP 0.92 73.47 1.61 Uv 14.33 2.53 25.70 107.30 X 6.15 364.55 664.81 52.08 x x 13.78 3884.48

CO TCLP 0.20 12.34 Uv Uv 9.52 0.33 Uv 83.39 0.90 5.50 110.85 25.38 74.78 3607.58 981.30 5.55 3859.60

NJ TCLP 0.30 75.12 0.43 Uv 14.78 11.43 6.56 123.53 756.42 6.12 325.77 558.69 x x x 14.05 2359.76

WI TCLP 0.34 95.66 Uv Uv 5.94 3.83 14.38 38.22 1042.09 5.90 6.36 127.48 uv x x 10.75 x

New Asphalt

TCLP 0.28 82.64 4.15 Uv 29.14 5.25 41.78 48.19 653.75 7.01 471.43 151.19 x x 1056.44 uv 1296.01

*DI: DI water as leaching solution

*TCLP: TCLP leaching test

*x: over top the detection standard concentration

Page 14: Leaching Characteristics of Recycled Asphalt Pavement Used as ...

14

The initial pH for TCLP fluid #1 was 4.93 and approximately 6.0 for DI water. After agitating 24

hours, the leachate pH from the TCLP fluid #1 batch was approximately 5.02 - 7.35. The DI

water leaching solution had a pH between 8.77 and 9.58. Electrical conductivity of the batch

leachate with TCLP fluid #1 was higher than the batch leachate with DI water. The low

conductivity indicated low mineralization of leachates. Coating of mineral particles by bitumen is

capable of preventing these particles from coming into contact with water and is the most likely

reason for such a low conductivity. The concentrations of mostly heavy metals such as silver,

arsenic, copper, nickel, zinc, manganese, barium, aluminum, magnesium leaching from TCLP fluid

were higher than in leachate using DI water. This difference could be due to proton exchange and

solid phase dissolution under more acidic conditions. The concentration of molybdenum and

silicon leaching from the TCLP fluid was the same concentration for the DI water leachate. All

concentrations were below MCL concentration in drinking water, except for manganese and arsenic

whose concentration were significantly higher (manganese = 1042.09 ug/L and arsenic = 95.66

ug/L from WG RAP). The concentration of chromium and cadmium were under detection limits for

leachate from both the TCLP fluid and DI water.

4.3 PAH in batch leaching experiments

Figure 7 presents 16 PAH waves by HPLC of PAH standard solution (with fluorescence

detector). Table 6 presents the groundwater intervention value of PAH. The PAH leaching results

using DI water are shown in Table 7 and using TCLP Fluid #1 are shown in Table 8. The results

show most PAHs concentrations were found to lie close to or below the detection limit, and even

below the groundwater intervention value. PAH leaching using the TCLP fluid was not higher than

leaching from DI water, unlike heavy metal leaching where the TCLP fluid leachate had higher

concentrations than the DI water leachate.

Figure 7: 16 PAHs peaks by HPLC of PAH standard solution (fluorescence detector)

Minutes

0.0

2.5

5.0

7.5

10

.0

12

.5

15

.0

17

.5

20

.0

22

.5

25

.0

27

.5

30

.0

32

.5

35

.0

37

.5

mV

olt

s

0

25

50

75

100

125

150

175

200

mV

olt

s

0

25

50

75

100

125

150

175

200

0

.15

0

0.0

00

0

.21

0

0.0

00

0

.30

0

0.0

00

0

.42

5

0.0

00

0

.50

0

0.0

00

0

.86

0

0.0

00

1

.45

5

0.0

00

Na

ph

tha

len

e

6.9

20

0.0

00

8

.07

5

0.0

00

8

.71

0

0.0

00

(1-M

eth

yln

ap

hth

ale

ne

) 0

.00

0 B

DL

(2-M

eth

yln

ap

hth

ale

ne

) 0

.00

0 B

DL

Ace

na

ph

tha

len

e

12

.03

0

2.0

93

Flu

ore

ne

12

.70

5

41

.89

7

1

3.3

05

0.0

00 1

3.9

70

0.0

00

(Ph

en

an

thre

ne

) 0

.00

0 B

DL

An

thra

ce

ne

15

.26

5

21

.43

6

1

6.3

45

0.0

00

(Flu

ora

nth

en

e)

0

.00

0 B

DL

1

7.7

00

0.0

00

(Pyre

ne

) 0

.00

0 B

DL

1

8.3

95

0.0

00

1

8.9

80

0.0

00

(Be

nzo

(a)a

nth

race

ne

) 0

.00

0 B

DL

(Ch

ryse

ne

) 0

.00

0 B

DL

2

0.8

40

0.0

00

2

1.1

15

0.0

00

2

1.3

40

0.0

00(B

en

zo

(b)f

luo

ran

the

ne

) 0

.00

0 B

DL

(Be

nzo

(k)f

luo

ran

the

ne

) 0

.00

0 B

DL

2

3.3

40

0.0

00

Be

nzo

(a)p

yre

ne

23

.96

0

18

.54

9

2

4.7

10

0.0

00

(Dib

en

zo

(a,h

)an

thra

ce

ne

) 0

.00

0 B

DL

Be

nzo

(gh

i)p

ery

len

e

26

.03

5

5.6

44

(In

de

no

(1,2

,3-c

d)p

yre

ne

) 0

.00

0 B

DL

2

6.7

30

0.0

00

2

7.0

00

0.0

00

3

0.2

00

0.0

00

Fluorescence

Std_5ppb_20111025Name

Retention Time

ESTD concentration

Page 15: Leaching Characteristics of Recycled Asphalt Pavement Used as ...

15

Table 6: Polycyclic aromatic hydrocarbons analyzed

PAH Groundwater intervention value* (µg/L)

Anthracene -

Benzo(a)anthracene -

Benzo(a)pyrene 0.2

Benzo(b)fluoranthene 0.2

Benzo(ghi)perylene -

Benzo(k)fluoranthene -

Dibenzo(ah)anthracene -

Fluoranthene 400

Indeno(123cd)pyrene -

Naphthalene 100

Phenanthrene -

Pyrene 250

Chrysene 0.2

Acenaphthylene --

Acenaphthene -

Fluorene 400

*Wisconsin groundwater quality

Page 16: Leaching Characteristics of Recycled Asphalt Pavement Used as ...

16

Table 7: Results of batch leaching tests carried out on RAP samples (DI water) (µg/L)

OHIO

RAP

California

RAP

NJ RAP WI RAP New asphalt

Material

Naphthalene BDL BDL BDL BDL BDL

1-Methylnaphthalene BDL 0.2555 0.1159 0.1527 0.0857

2-Methylnaphthalene BDL BDL 0.1919 BDL 0.1249

Acenaphthalene BDL 0.1321 0.1342 BDL BDL

Fluorene BDL BDL BDL 0.0615 BDL

Naphthalene 0.0115 BDL BDL BDL BDL

Phenanthrene 0.0894 0.5723 0.9697 BDL 0.0647

Anthracene 0.2200 BDL 3.8343 BDL BDL

Fluoranthene 0.2021 1.5497 2.9548 BDL 0.2893

Pyrene 0.0816 0.6551 1.1956 0.1488 0.1255

Chrysene BDL 0.1938 BDL BDL BDL

Benzo(a)anthracene BDL BDL BDL 0.0831 BDL

Benzo(b)fluoranthene BDL 0.0670 0.0677 BDL 0.1059

Benzo(k)fluoranthene BDL 0.0555 0.0287 BDL 0.0784

Benzo(a)pyrene BDL 0.0338 0.0544 0.0519 0.0300

Dibenzo(a,h)anthracene BDL 0.0952 0.0358 BDL 0.1083

Benzo(ghi)perylene 0.0001 0.0615 0.0926 0.0323 0.2028

Indeno(1,2,3-cd)pyrene BDL BDL BDL BDL 0.0126

Page 17: Leaching Characteristics of Recycled Asphalt Pavement Used as ...

17

Table 8: Results of batch leaching tests carried out on RAP samples (TCLP) (µg/L)

OH

TCLP

CA

TCLP

CO

TCLP

NJ

TCLP

WI

TCLP

New asphalt

material

TCLP

Naphthalene BDL BDL BDL 0.1366 BDL 0.1716

1-Methylnaphthalene BDL BDL BDL 0.0274 0.0676 0.1354

2-Methylnaphthalene 0.1117 0.1562 0.1527 0.0921 0.0939 0.2147

Acenaphthalene BDL 0.0362 BDL BDL BDL 0.0273

Fluorene 0 0.0288 0.0461 0.0061 0.0167 0.021

Phenanthrene 0.0614 BDL BDL 0.0253 BDL 0.1254

Anthracene 0.0668 0.1092 BDL 0.0239 0.1182 BDL

Fluoranthene 0.095 0.0726 BDL 0.0194 0.116 0.2549

Pyrene 0.0459 0.0286 0.0595 0.0279 0.068 0.102

Chrysene 0.0104 0.0021 0.0029 0.0028 0.0162 0.0311

Benzo(b)fluoranthene 0.0045 BDL 0.0051 0.0002 0.0122 0.0309

Benzo(k)fluoranthene 0.0018 BDL BDL BDL BDL BDL

Benzo(k)fluoranthene BDL BDL BDL BDL BDL 0.012

Benzo(a)pyrene 0.0013 BDL 0.0141 0.0019 0.0104 0.0207

Dibenzo(a,h)anthracene 0.0043 BDL 0.0273 BDL 0.0051 0.0011

Benzo(ghi)perylene 0.0014 BDL 0.0363 BDL 0.003 BDL

Indeno(1,2,3-cd)pyrene BDL BDL BDL BDL BDL BDL

BDL: below detection limit.

4.4 Influence of Liquid-solid ratio to PAH leaching

Influence of L/S ratio to PAH leaching was performed with NJ RAP. The results (Table 9) show that

more PAHs (Acenaphthalene, Benzo(a)anthracene, Benzo(b)fluoranthene, Benzo(a)pyrene,

Benzo(ghi)perylene) leach out as the L/S ratio increases. Even though the L/S ratio increases the

PAH leaching, the concentration of PAHs was always below the detection limit and groundwater

intervention value.

Page 18: Leaching Characteristics of Recycled Asphalt Pavement Used as ...

18

Table 9: PAH leaching result with different Solid-Liquid ratio (µg/L)

NJ_1:5 NJ_1:10 NJ_1:15 NJ_1:20

Naphthalene 0.003 BDL BDL BDL

1-Methylnaphthalene BDL BDL 0.137 0.079

2-Methylnaphthalene BDL BDL BDL BDL

Acenaphthalene 0.062 0.034 0.051 0.030

Fluorene BDL 0.029 BDL 0.025

Naphthalene BDL BDL BDL BDL

Phenanthrene BDL BDL BDL 0.054

Anthracene BDL 0.201 BDL 0.172

Fluoranthene BDL BDL BDL BDL

Pyrene 0.009 BDL BDL BDL

Benzo(a)anthracene 0.074 0.008 0.037 0.005

Benzo(b)fluoranthene 0.191 0.027 0.039 BDL

Benzo(k)fluoranthene 0.102 BDL BDL BDL

Benzo(a)pyrene 0.298 0.037 0.040 0.001

Dibenzo(a,h)anthracene BDL 0.002 0.014 BDL

Benzo(ghi)perylene 0.396 0.078 0.045 0.007

Indeno(1,2,3-cd)pyrene BDL BDL BDL BDL

Totals 12.877 2.828 10.344 2.601

5. Conclusion

Within a sustainable development context, the emphasis on recycling will be increasing over time. RAP

can be directly recycled into unbound road base course, decreasing the use of natural aggregates.

Although use of RAP as unbound base course in road construction is an increasing trend, the possibility

of heavy metal and PAH leaching has not been investigated. This study has investigated these

potential problems.

Batch leaching experiments with TCLP fluid and DI water as leaching solution have been performed to

evaluate potential pollutant release from the RAP. The results from this study show that pollutant

leaching concentrations were low for many heavy metals and PAHs studied. The pH and EC were

lower when using TCLP fluid than when using DI water as the leaching solution after agitated 24

hours. The low conductivity indicated low mineralization of leachates. The concentrations of mostly

heavy metals such as silver, arsenic, copper, nickel, zinc, manganese, barium, aluminum, and

magnesium leaching from TCLP fluid were higher than in leachate from DI water, while the

concentration of molybdenum and silicon leaching from TCLP fluid was the same to DI water. This

Page 19: Leaching Characteristics of Recycled Asphalt Pavement Used as ...

19

discrepancy could be due to protonic exchange and solid phase dissolution under more acidic

conditions. All concentrations were below MCL concentration in drinking water, except for Mn and

As whose concentration was significantly higher (Mn 1042.09 ug/L and As 95.66 ug/L from WG

RAP). The concentration of chromium and cadmium were under the detection limit both in leachate

from TCLP fluid and DI water.

PAH leaching tests show most PAH concentrations were found to lie close to or below the detection

limit and below groundwater intervention values. PAH leaching concentrations using the TCLP fluid

was similar to PAH leaching concentration using the DI water.

Influence of L/S ratio to PAH leaching with NJ RAP showed that PAH leaching increased with

increased L/S ratio. The concentrations of PAHs were always below the detection limit or groundwater

intervention value. Column leaching tests still need to be completed along with field experiments so

that both real hydrologic conditions and the scaling effect on leaching can be evaluated.

Page 20: Leaching Characteristics of Recycled Asphalt Pavement Used as ...

20

References

Baek, S.O., Field, R.A., Goldstone, M.E., Kirk, P.W., Lester, J.N., Perry, R., 1991. A review of

atmospheric polycyclic aromatic hydrocarbons: sources, fate and behaviour. Water Air Soil Pollut.

60, 279–300.

Brandt, H. C. A., Groot, P. C. DE, 2001. Aqueous leaching of polycyclic aromatic hydrocarbons

from bitumen and asphalt. Water Research. 35 (17), 4200–4207.

Brantley, A.S., Townsend, T.G., 1999. Leaching of pollutants from reclaimed asphalt pavement.

Environ. Eng. Sci. 16 (2), 105–116.

Han Y., Kang J., Kayhanian M., Stenstrom M. K., 2009. Characteristics of Highway Storm water

Runoff in Los Angeles: Metals and Polycyclic Aromatic Hydrocarbons, Water Environment

Research. 81, 3: 308-318.

Hewitt, C.N., Rashed, M.B., 1990. An integrated budget for selected pollutants for a major rural

highway. Sci. Total. Environ. 93, 375–384.

Kriech A. J., Kurek J. T., Osborn L. V., Sweeney B. J., 2002. Determination of polycyclic

aromatic compounds in asphalt and in corresponding leachate water. Polycyclic Aromatic

Compounds, 22:517–535.

Legret M., Odie L., Demare D., Jullien A. 2005. Leaching of heavy metals and polycyclic

aromatic hydrocarbons from reclaimed asphalt pavement. Water Research. 39, 3675–3685.

Muschack, W., 1990. Pollution of street runoff by traffic and local conditions. Sci. Total

Environ. 93, 419–431.

Pagotto, C., Legret, M., Le Cloirec, P., 2000. Comparison of the hydraulic behaviour and the

quality of highway runoff water according to the type of pavement. Water Res. 34 (18), 4446–

4454.

Sadler, R., Delamont, C., White, P., Connell, D., 1999. Contaminants in soils as a result of

leaching from asphalt. Toxicol. Environ. Chem. 68, 71–81.

Strecker, E.W., Driscoll, E.D., Shelley, P.E., Gaboury, D.R., Sartor, J.D., 1990. The US Federal

Highway Administration receiving water impact methodology. Sci. Total Environ. 93, 489–498.

Takada, H., Tomodo, O., Ogura, N., 1990. Determination of polycyclic aromatic hydrocarbons in

urban street dusts and their source material by capillary gas chromatography. Environ. Sci.

Technol. 11, 1179–1186.

Warren, R.S., Birtch, P., 1987. Heavy metal levels in atmospheric particulates, roadside dust and

soil along a major urban highway. Environ. Technol. 59, 253–256.