1/10 Emissions Characteristics of Volatile and Semi-volatile Organic Compounds and Polynuclear Aromatic Hydrocarbons (PAHs) from Hot and Warm Mix Asphalts Frank Farshidi, University of California, Davis, U.S.A. Isabel Faria, University of California, Davis, U.S.A. Peter G. Green, University of California, Davis, U.S.A. David Jones, University of California, Davis, U.S.A. John Harvey, University of California, Davis, U.S.A. ABSTRACT: Warm mix asphalt (WMA) technologies are now available to reduce the production temperatures of hot mix asphalt (HMA) and rubberized hot mix asphalt (R-HMA) by between 15 C and 55 C (59 F and 131 F). This study evaluates the qualitative and quantitative characteristics of emissions from HMA, R-HMA, and rubberized warm mix asphalt (R-WMA). Using a protocol developed previously by the University of California Pavement Research Center reactive organic gases (ROGs) and particulate matter (PM) were investigated. The ROGs consisted of several volatile organic compounds (VOCs) and semivolatile organic compounds (SVOCs) as well as polynuclear aromatic hydrocarbons (PAHs) both in solid and gaseous phase. Sample emissions were collected at several asphalt plants Emissions from each material were captured in charcoal tubes for VOCs and SVOCs according to NIOSH 1500-1501 and XAD-2 tubes and filters for PAHs according to NIOSH 5515. The effects of material type and temperature on the emissions are investigated and compared. The results indicated that both the mix type and temperature significantly influence the emissions characteristics. Monitoring emissions kinetics over time illustrated that the majority of emissions are volatilized in the first hour after sampling initiation. The method developed in this study is suitable for characterizing emissions in future studies as a reliable and sensitive procedure for accurately quantifying asphalt fumes emissions. 1. Introduction As part of the second phase of the emissions study with WMA technologies, four different asphalt mixes at two different asphalt plants were sampled in the field using the methodology described previously [1]. Besides characterizing the VOCs and SVOCs in asphalt fumes, the second phase of this study also included characterization of PAH compounds from field samples during construction with both HMA and WMA mixes. The PAH chemicals constitute a diverse class of organic molecules and represent components with a wide range of molecular size and structural types [2]. A major health concern from asphalt fumes and PAHs is the potential exposure to carcinogens [3]. Factors such as the temperature and the source of the crude can lead to variations in composition and toxicity of the fumes [4, 5]. Information regarding the asphalt emissions and the effect of WMA technologies on emissions are still limited. Assessing potential exposure and quantifying VOCs, SVOCs and the PAHs in fumes are essentially problematic due to the lack of a methodology to evaluate such a complex mixture. Therefore, one of the major objectives of the second phase of this study was to develop and validate a reliable analytical method for identification and quantification of compounds in this complex mixture. Moreover, understanding the emissions
2/10 kinetics over time is investigated for asphalt plants. various mixes immediately after production at the 2. Experimental Details The second phase of the emissions study included measuring emissions at two different asphalt plants for collecting emissions data from four different mix types for two hours. Field sampling was conducted using the flux chamber discussed previously. Figure 1 illustrates the sampling train used in this study. As shown, the first branch includes a charcoal tube for collecting VOCs and SVOCs at a flow rate of 1.5 L/min for two hours. The second branch is also a charcoal trap at a flow rate of 1.5 L/min. The charcoal tubes at the second branch train are replaced every 30 minutes to understand the kinetics of the emission with time. The third sampling branch employs sampling through a PTFE filter (0.45- followed by an XAD-2 tube for capturing both the particulate and gaseous phase of the PAHs. The fourth and fifth sampling branches consist of XAD-2 trap and PTFE filter respectively as replicates and a check for the verification of the technique. Chamber Pump Charcoal Trap Charcoal Trap PTFE Filter XAD-2 Trap PTFE Filter XAD-2 Trap Figure 1: Diagram of the sampling train for fume collection 2.1 Mix Types Two different asphalt plants in the northern California region were selected for conducting the second phase of sampling. The sampling was conducted on a designated area next to the load-out area of the asphalt plants. Table 1 summarizes the mix properties used for collecting the emissions samples.
3/10 Table 1: Mix Properties during the Second Phase Emissions Sampling Parameter Plant#1 Plant#1 Plant#2 Plant#2 Mix Type 3/4" HMA with 15% RAP 1/2" R-HMA- OGFC 3/4" HMA 3/4" HMA Binder Type PG64-16 PG64-16 PG64-16 PG64-16 Binder Content 5.10% 6.50% 5.50% 4.70% WMA Technology - Advera - Evotherm Load-out Temperature ( C/ F) 160/(320) 147/(297) 150/(302) 123/(253) 2.2 Materials and Instruments Reagents dichloromethane and hexane were acquired from Aldrich (Milwaukee, WI). The PAH standard was purchased from Supelco. XAD-2 tubes [treated with 2- (hydroxymethyl)piperidine] and Poly (tetrafluoroethylene) (PTFE) filters (37 mm, 0.45- pore size) were purchased from SKC Inc. Extraction of asphalt fumes from tubes and filters were performed by ultrasonic extraction. Individual sections of the charcoal tubes were extracted in separate vials using 1.5 ml of carbon disulfide as solvent and 30 minutes of sonication. The PTFE filters and the XAD-2 resin were extracted separately with 5 ml of dichloromethane/hexane (1:1). Ultrasonic extraction was performed for 1 h for filter samples and 1.5 h for XAD-2 samples. 2.3 Sample Purification and Clean-up Asphalt fumes are extremely complex and consist of aliphatic, heterocyclic compounds, polycyclic aromatic hydrocarbons (PAHs) and some nitrogen, oxygen, and sulfur containing compounds. In order to identify and characterize PAH compounds more accurately, a simplified purification method was developed and conducted. In this technique, a column (250mm x 10 mm i.d.) packed with silica cartridge was used to separate the PAHs from saturated and polar compounds. The solvent was dichlromethane/hexane (1:1). The extracts were filtered and preconcentrated by evaporating the solvent under a nitrogen blanket. 2.4 GC/MS Detection for Alkanes The determination of asphalt fume composition was carried out using Electron Impact Ionization (EI) of GC/MS. The high sensitivity of EI was used to quantify the asphalt fume samples. The GC/MS instrument was calibrated by an alkane standard containing 23 n- hydrocarbons (C6 thru C28). An Agilent 6890-5973 series gas chromatograph coupled with a mass spectrometer (Agilent, 5973N) was used. The gas chromatograph, equipped with a 30-5 capillary column, provided satisfactory separation of various compounds. The helium carrier gas was introduced at a rate of 1.0 ml/min) and linear velocity of 20 cm/s. The initial column temperature was maintained at 50 C for 4 minutes, then raised to 125 C at a rate of 2 C/min, and finally to 280 C at a rate of 10 C/min. The extraction efficiencies from activated charcoal tubes are presented in Table 5-2. The table provides information regarding the detection limits as well as the accuracy for the
4/10 alkane compounds. Full scan mode with splitless injection was used as the spectrometric technique for characterizing VOCs and SVOCs. A purge time of 1.5 minute was specified in the method. After the purge time, the splitter vent is opened to purge solvent from the injector. Splitless injection, therefore, concentrates the sample and purges most of the volatile solvent [5]. For this reason, and because of the large amounts of sample that can be injected, splitless injection was used for this study. Table 2: Retention Time, Detection Limit, and Extraction Efficiencies for the GC/MS Analysis of VOCs and SVOCs Compound Retention Time Detection Limit (ng/m 3 ) Recovery from charcoal tubes (%) Decane 11.90 1.5 85 Undecane 14.25 2.0 94 Dodecane 16.35 2.1 90 Tridecane 18.26 1.9 93 Tetradecane 20.05 2.3 94 Pentadecane 21.72 1.8 92 Hexadecane 23.30 2.1 91 Heptadecane 24.79 2.0 90 Octadecane 26.22 1.9 95 2.5 GC/MS Detection for PAHs To detect individual priority PAHs in asphalt fumes at trace levels, a selected ion monitoring (SIM) mass spectrometer was used. In the SIM mode of operation, the detection system was dedicated to monitoring only a few ion currents. In this mode, the mass spectrometer does not spend time scanning the entire mass range, but rapidly changes between m/z values for which characteristic ions are expected [6]. Therefore, the sensitivity was tremendously higher than the scanning mode. A mixture of 16 priority standard PAHs was used for calibration. The initial column temperature was maintained at 100 C for 2 minutes, then raised to 300 C at a rate of 2 C/min. Splitless injection with 1.5 mins purge time was used. The carrier gas helium and a solvent delay of 5 mins were used in the method. To evaluate the recovery of the sample, stable isotope dilution with perdeuterited anthracene as an internal standard was used. The relative recovery of the internal standard can account for losses of the analytes during sample preparation and detection processes. This procedure improved the reliability of the analysis. In these experiments, perdeuterited anthracene was added to the sample at a concentration of 10 ng/l. The recoveries of spiked PAHs from XAD-2 tubes are shown in Table 5-3. Table 5-3 also illustrates the major ions that were monitored in the SIM mode for all the sixteen PAH compounds. Table 3: Retention Time, Recovery, and Major Ion for Priority PAHs Compound Retention Time(min) Recovery from XAD-2 resin (%) Major Ions to be Monitored
5/10 Naphthalene 11.68 95 128, 102, 64 Acenaphthylene 12.45 92 152, 76, 63 Acenaphthene 14.74 91 153, 154 Fluorene 19.05 98 166, 139, 83 Phenanthrene 19.28 99 178, 89, 76 Anthracene 20.26 95 178, 89, 76 Fluoranthene 24.55 100 202, 101, 88 Pyrene 25.53 85 202, 101, 89 Benz(a)anthracene 31.24 92 228, 114, 101 Chrysene 31.39 91 228, 114, 102 Benzo(b)fluoranthene 35.94 85 252, 126, 113 Benzo(k)fluoranthene 36.054 89 252, 126, 114 Benzo(a)pyrene 37.188 91 276, 138, 124 Dibenz(a,h)anthracene 41.29 96 276, 138, 124 Benzo(ghi)perylene 41.44 93 278, 139, 125 Indeno(1,2,3-cd)pyrene 42.18 94 276, 138, 126 3. Results and Discussion 3.1 Emissions of Alkanes The emissions of total alkanes and individual alkanes from the four different mixes collected on activated charcoal tubes are shown in Figure 2 and Figure 3 respectively. It can be observed from Figure 5-2 that the emissions of the mixes without any WMA technologies that had higher temperature at time of the sampling are significantly higher than the mixes with WMA technologies and lower temperatures. The emissions of alkanes as a function of sampling time are presented in Figure 4. The results show that the emissions decreased as time progressed. During the first thirty minutes of sampling, alkane emissions were present in higher concentrations for all the mixes and specially the mixes without WMA technologies. The second thirty minutes produced lower concentrations of alkanes. The emissions in the third and fourth time interval were significantly lower. These results demonstrate that most of the alkane emissions volatilize from the mix during the first hour of sampling and as time progresses the emissions are reduced significantly.
6/10 Alkane Emission Concentration (µg/m 3 ) 3,000 2,500 2,000 1,500 1,000 500 Alkane Emissions Temperature 160 150 140 130 120 110 Mix Temperature ( C) 0 100 Figure 2: Total alkane emissions concentration for four different mixes 450 Alkane Emissions HMA with 15% RAP Alkane Emission Concentration (µg/m 3 ) 400 350 300 250 200 150 100 50 R-WMA with Advera WMA with Evotherm HMA 0 Figure 3: Alkane emissions concentrations for four different mixes
7/10 Figure 4: Alkane emissions versus time for four different mixes 3.2 Emissions of PAHs Emissions of PAHs for all the four mixes are presented in Figure 5. It must be noted that the PAHs are distributed between gaseous and particulate phase. Results in Figure 5 are the concentrations of PAHs collected on XAD-2 tubes (gaseous). It can be observed from this figure that PAH concentrations were significantly decreased as the temperature of the mix was decreased. Results indicate that temperature is an important parameter that influences the PAH emissions. By using WMA technologies to reduce mix temperatures, significant reduction of PAHs is observed. Moreover, in the temperature range for the mixes in this study, the PAH concentrations are only observed at trace levels both for the control and WMA mixes. The particulate phase emissions collected on filters had concentrations below the detection level of the analytical method. The same trend was observed on both the sampling branch in series with the XAD-2 tube and the branch with the filter only. Detailed sample clean-up and SIM method of analyzing ensured the detection of the compounds at ultratrace levels. The Limit of Detection (LOD) and Limit of Quantification (LOQ) were determined as respectively 3 times and 10 times the standard deviation of seven replicates of a low concentration standard. Table 4 lists the LOD and LOQ for all the PAH compounds analyzed. As illustrated in this table, LOD ranged from 0.002 to 0.007 ng/ul (or 2-7 ug/l) while LOQ ranged from 0.005 to 0.02 ng/ul (5-20 ug/l).
8/10 Table 4: LOD and LOQ for the PAH Compounds PAH Compound LOD (ng/ul) LOQ (ng/ul) Naphthalene 0.002 0.008 Acenaphthylene 0.002 0.007 Acenaphthene 0.004 0.013 Fluorene 0.003 0.010 Phenanthrene 0.003 0.010 Anthracene 0.007 0.024 Fluoranthene 0.003 0.010 Pyrene 0.003 0.009 Chrysene 0.002 0.005 Benzo(a)anthracene 0.002 0.008 Benzo(b)fluoranthene 0.001 0.005 Benzo(k)fluoranthene 0.003 0.008 Benzo(a)pyrene 0.003 0.010 Indeno(1,2,3-cd)pyrene 0.002 0.007 Dibenzo(a,h)anthracene 0.006 0.019 Benzo(ghi)perylene 0.004 0.014 120 PAH Emissions 100 PAH Emission Concentration (µg/m 3 ) 80 60 40 20 HMA with 15% RAP R-WMA with Advera HMA WMA with Evotherm 0 Figure 5: Concentration of various PAHs from four different mixes sampled at asphalt plants
9/10 4. Summary and Conclusions The purpose of this study was to understand the emissions kinetics of VOCs and SVOCs over time using the field sampling methodology developed previously and identify/quantify the PAH compounds in asphalt fumes. A sensitive and selective method for identification and characterization of VOCs, SVOCs, and PAHs was developed and verified for various mixes at the asphalt plants. The alkane emissions consisted of n-hydrocarbons ranging from C8 to C18 range. Depending on the mix and the temperature of the mix at the time of sampling, the total alkane emissions ranged from 117 µg/m3 (WMA with evotherm technology) to 2516 µg/m3 for conventional HMA. The kinetic of emissions over time indicated that the majority of alkane emissions are volatilized in the first hour after the sampling initiation. The gaseous phase PAH emissions mainly consisted of low molecular weight PAH compounds and mixes with higher initial production temperatures showed higher PAH concentrations. Both SIM and full mode GC-MS analysis confirmed that the particulate phase of PAHs are present at ultertrace levels and were below the detection limit of used in this study. Limit of detection ranged from 0.002 to 0.007 ng/ul (or 2-7 ug/l) while limit of quantification ranged from 0.005 to 0.02 ng/ul (5-20 ug/l) the mixes at time of sampling ranged from 123 C to maximum of 160 C. The results confirmed that this temperature range is not high enough to initiate significant PAH formation in asphalt emissions. PAH compounds are mainly formed when there is combustion and for asphalt binders that requires temperatures above 500 C. Moreover, depending on the source of the binder and the manufacturing process of the crude oil, many of the lower molecular weight PAH compounds might have been removed during the refinery process. More research is needed to quantify emissions from different mixes at different temperatures, mainly in the range of WMA and R-WMA production temperatures. Simulating asphalt fumes in the laboratory conditions will aid in identifying and characterizing emissions kinetics of different WMA technologies, production temperatures, and binder sources. Determining the optimal temperature range that will minimize the emissions concentrations without undermining performance properties need to be investigated in future studies. 5. References 1. Farshidi, F., Jones, D., Kumar, A., Green, P. and Harvey, J., Direct Measurements of Volatile and Semivolatile Organic Compounds from Hot and Warm Mix Asphalt. Transportation Research Record, 2011. 2. Grosjean, D., Polycyclic Aromatic-Hydrocarbons in Los-Angeles Air from Samples Collected on Teflon, Glass and Quartz Filters. Atmospheric Environment, 1983. 17(12): p. 2565-2573. 3. Vu-Duc, T., C.K. Huynh, and S. Binet, Laboratory generated bitumen fumes under standardized conditions. Clean-up scheme and ion trap GC-MS analysis of VOC, semi-volatile and particulate PAH and PASH. Journal of Occupational and Environmental Hygiene, 2007. 4: p. 245-248. 4. Fernandes, P.R.N., et al., Evaluation of Polycyclic Aromatic Hydrocarbons in Asphalt Binder Using Matrix Solid-Phase Dispersion and Gas Chromatography. Journal of Chromatographic Science, 2009. 47(9): p. 789-793.
10/10 5. Wang, J., et al., Characterization of asphalt fume composition under simulated road paving conditions by GC/MS and microflow LC/quadrupole time-of-flight MS. Analytical Chemistry, 2001. 73(15): p. 3691-3700. 6. Lange, C.R. and M. Stroup-Gardiner, Temperature-dependent chemical-specific emission rates of aromatics and polyaromatic hydrocarbons (PAHs) in bitumen fume. Journal of Occupational and Environmental Hygiene, 2007. 4: p. 72-76.