Environmental Impact of Adding PCSs to HMA

1. Leachability Test for HMA with PCS

The increase in the number of organic contaminants being detected in groundwater as well as in surface water is causing concern because of potential health risks claimed to be associ ated with human exposure to these substances. Soil contaminated with petroleum product is a potential threat to both surface water and groundwater; thus a key issue is whether hydrocarbon will leach out from the asphalt mixture when it is mixed with PCSs and used to pave roads.

The short-term test conducted by Eklund [10] on the environmental effects of paving showed that after allowing a pavement to leach for 7 days with distilled water, the amount of petroleum-based compounds leached out and detected was less than 2 ppm. In another series of experiments by Eklund [10] with 6% waste oil with an appreciable lead content in cold mix asphalts (CMA), when the CMA was leached for a week in an acid rain simulation, the leachate contained less than 3 ppm of lead. Although short-term tests showed no sign of any harmful leachate generation, tests should be conducted to determine the environmental impact of long-term exposure of asphalt produced with contaminated soils on the environment.

A uniform leaching test was used to estimate the quality of leachate that would be produced by the asphalt mixture. A new leachability test was designed to simulate the rate of release of contaminants when HMAs mixed with PCSs are exposed to the actual performance environments when used as pavements. The test methodology and test standards were developed in cooperation with the New Jersey Department of Environmental Protection and Energy (NJDEPE). The experimental protocol was evolved based on the information and experience of EPA toxicity characteristics leaching procedure (TCLP) and waste stabilization/solidification program, the waste solidification program of the U.S. Army Corps of Engineers, and the nuclear waste research program at Brookhaven and Oak Ridge National Laboratories.

The uniform leaching procedure gives an indication of the amount of each organic compound that is leachable under specific experimental conditions. The structural integrity of sample was kept in this test, where particle size reduction was inappropriate, i.e., in those instances where solidification of the waste is needed to meet the best demonstrated available technology provision of environmental law. Grinding may not adequately represent the actual process. Particle reduction alters the physical character of many solidified wastes by destroying the cementitious property of these wastes in such a way as to show an unrealistically high leaching rate.

This water-leachability test was based on the EPA draft on Solid Waste Leaching Procedure (SWLP). This document recommends using laboratory reagent water as the leaching medium. Chemical and physical conditions should be chosen to mimic the environment where the HMA with PCS is used. The leaching medium in this test was chosen as reagent water with a pH value of 6.8 at a temperature of 25°C (77°F). Since the asphalt mixtures are used as paving materials, the reagent water should be more representative of the actual environment than an acidic leaching medium. The reagent water was prepared by boiling deionized water for 15 min. Subsequently, while maintaining the water at about 90°C (194°F), nitrogen gas was bubbled through the water for 1 hr. The HMA specimen was treated as a monolithic waste (i.e., the specimen was not pulverized prior to testing). The asphalt concrete with PCSs were compacted into a specially designed stainless steel mold 2 in. x 2 in. x 2 in. before being tested. Since 2-in. cubes were needed for this test, HMA could not be compacted to the same density as in Marshall tests. However, a loose matrix should produce a higher and conservative leaching rate. HMA specimens were prepared as for the Marshall test, where the blended aggregates were heated to 130°C (268°F). Then heated asphalt cement was added to the dry mix and wet-mixed for 1 min. Then a sufficient quantity of HMA was placed inside the steel molds and compacted with 50 tamps and the HMA specimen was allowed to cool. The compacted HMA specimens with PCSs were placed in 480-mL glass containers and sealed with Teflon septum caps. The volume of the container was at least three times that of the sample, so sufficient space was available in the container and the sample was surrounded on all sides by leaching medium. There was zero headspace to minimize the effects of volatilization. Containers were placed on an Orbit shaker table at 75 rpm for a period of 96 hr. After the end of the shaking period, the containers were opened and a 25-mL water sample was immediately drawn using a gastight syringe. The extracted solution was analyzed using an EPA purge-and-trap method.

The Purge and Trap Method and GC Settings. The analytical system used in this test was an ALS 2016 desorber and Tekmar 2000 purge-and-trap concentrator interfaced to a Varian 3400 gas chromatograph (GC) with a high resolution capillary column and a flame ionization detector (FID).

The parameters of ALS 2016 were as follows:

Preheat 2 min

Purge 15 min, helium flow at 40 mL/min

Dry purge 4 min

Cooldown - 150°C

Desorb preheat 150°C

Desorb 5 min at 180°C

Inject 3 min at 190°C

Bake 10 min at 240°C

The column used for GC/FID analysis was a cross-linked 5% phenyl and 95% methyl silicone gum, 50 m long, 0.2 mm in diameter, and 0.5 p,m film thickness fused silica-bonded high resolution capillary column. Flow rates for the GC were

Hydrogen 30 mL/min Air 30mL/min Carrier gas I mL/min Make-up gas 30 mL/min

The GC parameters were

Initial column temperature

40°C

Initial column hold time

5 min

Program 1 final column temperature

65°C

Program 1 column rate

2°C/min

Program 1 column hold time

0

Program 2 final column temperature

190°C

Program 2 column rate

8°C/min

Program 2 column hold time

5 min

Inject temperature

210°C

Detect temperature

250°C

Table 5 Method Specifications and Regulatory Limits for Fraction Leached (Proposed NJ) in ppb

Compound

MDL

P

±

sP

CV and concentration

Regulatory level

Dichloromethane

0.29

82

±

32%

12.6%/12

270

1,2-Dichloroethane

0.47

76

±

21%

24.8%/26

820

1,1,1-TCE

0.39

73

±

28%

21.4%/8

110

Hexane

0.22

106

+

25%

18.5%/8

N/A

Benzene

0.26

88

16%

14.3%/14

410

Toluene

0.25

94

18%

16.8%/14

350

Perchloroethene

0.13

108

+

24%

19.7%/6

100

Ethylbenzene

0.28

79

±

17%

19.7%/6

360

p,m-Xylene

0.28

95

+

20%

9.6%/5

340

o-Xylene

0.24

91

21%

12.6%/5

340

An external standard was used for the calibration. The standard consisted of a mixture of target compounds in methanol. They were dichloromethane. 1,2-dichloroethane, 1,1,1-trichloroethane, hexane, benzene, toluene, perchloroethylene, ethylbenzene, p,m-xylene, and o-xylene.

Calibration and Quantification of Data. To determine the precision of the measurements, three injections were made for each concentration level, and the results were analyzed to obtain the mean and the standard deviation. Seven standard samples of the same concentration were analyzed to obtain the coefficient of variation (CV) to determine the reproducibility of his analytical method for each compound.

The detection limit of the method (MDL) is defined as the minimum concentration of a substance that can be identified, measured, and reported with 99% confidence when the analytic concentration is greater than zero and is determined from analysis of a sample in a given matrix containing analytic compounds. In this test, reagent water was used to estimate the MDL concentration. Standards (analytic in reagent water) at concentrations equal to 1 to 5 times the estimated MDL were used to calculate the MDL. The MDL was reported in concentration units as the standard deviation (SD) of the replicates multiplied by the appropriate Student's /-value (for a one-tailed test at 99% confidence) for the number of replicates. In this test, the number of replicates was chosen as seven, so MDL was defined as

Each day, three or four samples were spiked with at least 10% of the samples and analyzed to monitor and evaluate the experimental data quality. The percentage recovery was calculated using the equation

where A is concentration after spiking, B is the background concentration, and T is the known true value of the spike.

After the analysis of 10 spiked samples, the average percentage recovery (P) and the standard deviation of the percentage recovery (Sp) values were calculated and expressed as P ± Sp. Table 5 shows those values.

A major source of interference in this test was cross-contamination of samples. To prevent such cross-contamination, reagent water blanks were run twice before running each sample to

Table 6 Leachate Concentration (ppb) from Monolithic HMA Specimens with PCSs

Compound

PCS 1

PCS 2

PCS 3

PCS 4

PCS 5

PCS 6

Dichloromethane

ND

ND

2.96

ND

ND

ND

1,2-Dichloroethane

ND

ND

0.88

0.16

0.18

1.51

1,1,1-TCE

ND

ND

8.01

ND

5.43

5.14

Hexane

0.23

ND

6.46

ND

2.43

0.85

Benzene

ND

ND

3.11

ND

2.19

ND

Toluene

ND

ND

ND

ND

ND

ND

Perchloroethene

ND

ND

1.19

ND

0.97

ND

Ethylbenzene

ND

ND

ND

ND

ND

ND

p,m-Xylene

ND

0.32

0.36

ND

0.33

1.69

o-Xylene

ND

ND

0.37

ND

ND

ND

ND = none detected or less than the detection limit. Source: Meegoda et al. (9],

ND = none detected or less than the detection limit. Source: Meegoda et al. (9], demonstrate that interferences from the analytical system were under control. The following procedure was used to clean the vials and glass containers:

1. Wash all the vials and containers thoroughly in hot water using a detergent to remove the particulate matter and contaminants.

2. Rinse thoroughly using tap water.

3. Rinse three times using deionized water.

4. Place in a vacuum oven at 105°C for 12 hr to bake all volatile compounds.

5. Cool for 30 min, screw the lids tightly, and store in an area not subject to contamination by air or other sources.

Test Results. This method can estimate the concentrations of target chemicals up to a minimum concentration of around one-tenth of 1 part per billion (ppb). Table 6 shows the test results. Ethylbenzene did not leach out from any of the six mixes. Benzene, toluene, and xylene concentrations in all the leachates were less than 8 ppb. For the HMA mix containing PCS No. 3 with an initial petroleum concentration of 6600 ppm in the soil, the total leachate concentration of 10 key chemicals in leachate from solidified and stabilized HMA with PCS No. 3 was less than 25 ppb. The experimental results clearly demonstrate that the HMA with PCSs solidifies and stabilizes most of the petroleum contaminants within the asphalt matrix and the very small concentrations of organic chemicals that may be leaching are from the soil particles that are not completely coated with asphalt cement.

To further evaluate the long-term leaching of contaminants, three specimens of HMA made with PCS No. 3 were prepared, placed on top of the shaker, and shaken for durations of 1 day, 1 week, and 1 month. Those test results are shown in Table 7 and are graphically displayed in Figure 6. Results do not show any significant increase in contaminant concentrations with time for any of the compounds tested.

2. Analysis of Air Quality During the Production of HMA with PCS Czarnecki [4] obtained a permit from the Massachusetts Department of Environmental Quality Engineering (DEQE) to process petroleum-contaminated soils subjected to the following conditions: It should process 95% virgin aggregate and 5% PCS, and 95% of the hydrocarbons should be incinerated during the process. To demonstrate the 95% destruction, a mass balance of pure chemicals was performed for the incineration system. The test used sand contaminated with a 3% concentration of a 50/50 blend of xylene and toluene. Three points in the system

Table 7 Leachate Concentration (ppb) with Time from Monolithic HMA Specimen with PCS No. 3

Compound

One day

One week

One month

Dichloromethane

1.82

2.33

3.12

1,2-Dichloroethane

0.82

0.81

1.03

1,1,1-TCE

5.35

6.29

9.43

Hexane

6.01

6.32

6.87

Benzene

2.19

2.32

2.29

Toluene

ND

ND

ND

Perchloroethene

1.01

1.11

1.12

Ethylbenzene

ND

ND

ND

p,m-Xylene

0.33

1.65

1.80

o-Xylene

0.26

0.28

0.30

ND = none detected or below detection limit for each compound. Source: Meegoda et al. [9].

ND = none detected or below detection limit for each compound. Source: Meegoda et al. [9].

Time (days)

o Diehloromethane

D 1,1,1-TCE

* Hexane

• Benzene

is p,m-Xy!ene

x o-Xylene

Figure 6 Long-term leaching test results. (From Meegoda et al. [9].)

Figure 6 Long-term leaching test results. (From Meegoda et al. [9].)

were selected for analysis: incoming sand from the conveyer belt before the aggregate dryer, sand stripped of contaminants from the storage silo, and air samples from the stack. For three tests, Czarnecki [4] showed 99% destruction of hydrocarbons. Following are the test results from this study; the ait' sample collection method was not documented:

Incoming sand concentration 30,000.0 ppm

Sand coming out from aggregate dryer 0.9 ppm Air samples from the stack 0.2 ppm

The measurement of volatile organic compounds (VOCs) has become an important aspect in understanding photochemical reactions and providing an index of hydrocarbons present in the atmosphere. Knowledge of the levels of such materials in the ambient atmosphere is also required in order to determine human health impacts. However, the determination of toxic organic compounds in ambient air is a complex task, primarily because of the wide variety of compounds of interest and the lack of standards and procedures for analysis. The U.S. Environmental Protection Agency (EPA) has developed several standard analytical methods for the measurement of volatile organic compounds present in ambient air. Sample collection has been reported as the weakest link in the analytical chain for the determination of airborne organics, which is critical to the accuracy of the results. The sampling methods used for measurement of VOCs can be categorized as adsorption by a solid, cryogenic trapping, or whole air collection.

Adsorption by a solid. The solid adsorbents include (1) organic polymers (Tenax, XAD-2), (2) inorganic materials (silica gel, florisil), and (3) carbon (activated carbon, carbon molecular sieves).

In this technique VOCs are collected on a solid sorbent material while the bulk constituents (e.g., nitrogen, oxygen) are allowed to pass through the sorbent. The VOCs adsorbed are then desorbed, and the sample is injected in a gas chromatograph (GC). Stripping of the adsorbed analytes from the adsorbent is typically accomplished by either thermal or solvent desorption. Solvent desorption, although useful in many applications, generally requires sample preconcentration before analysis. This adds to the complexity of the final method and increases the sample handling time and the possibility of contamination. The thermal desorption of analytes has the advantages of reduced sample handling and increased sensitivity because of the transfer of all analytes onto the chromatographic column. The main advantages of the use of a solid adsorbent for organic polymers are that (1) little water is collected in the sampling process and (2) a large volume of air can be sampled relative to other techniques such as cryogenic sampling. The analytes adsorbed onto Tenax, desorbed, and analyzed using GC/MS can determine volatile nonpolar organics (e.g., aromatic hydrocarbons, chlorinated hydrocarbons) having boiling points in the range of 80-200°C [11]. The PUF/XAD-2 adsorption with GC and HPLC detection can be applied to determine polynuclear aromatic hydrocarbons. A major disadvantage of these materials is the breakthrough of volatile compounds. Degradation products of the trapping materials are frequently found in adsorbent tubes such as Tenax. Incomplete desorption is also a problem with these methods.

Inorganic adsorbents include silica gel, alumina, florisil, and molecular sieves. These materials are considerably more polar than the organic polymeric adsorbents, leading to the efficient collection of polar materials. Unfortunately, water is also efficiently captured, leading to rapid deterioration of the adsorbents.

Carbon adsorbents are relatively nonpolar compared to the inorganic adsorbents, and hence water absorption is not a significant problem. Carbon tends to exhibit much stronger adsorption properties than organic polymeric adsorbents, allowing the efficient collection of volatile materials such as vinyl chloride. However, the strong adsorption of carbon adsorbents can be a disadvantage. The desorption of target compounds from the carbon tubes is a common problem. For example, carbon molecular sieves used in EPA Method TO-2 bind aromatic compounds tightly, and a high temperature (=400°C) is required to desorb them. Finally, moisture affects the trapping and desorption efficiency of the charcoal tubes. The adsorption on carbon molecular sieve followed by desorption and GC/MS analysis can be used to determine highly volatile nonpolar organics (e.g., vinyl chloride, vinylidene chloride, benzene, toluene) having boiling points in the range of -15°C to +120°C.

Cryogenic Trapping. The collection of atmospheric organics by condensation in a cryogenic trap is an attractive alternative to adsorption or impinger collection. The primary advantages of this technique include (1) the collection of a wide range of organic materials, (2) avoidance of the contamination problem associated with adsorbents and other collection media, (3) the availability of the sample for immediate analysis without further work, and (4) consis tent recoveries. But the disadvantage of this method is that it is suitable only for volatile and nonpolar organics having boiling points in the range of - 10°C to +200°C. This technique provides quantitative concentrations of identified species of lower molecular weights such as C2-C|0 compounds typically observed in ambient air. However, an important limitation to this technique is the condensation of large quantities of moisture and carbon dioxide and lesser amounts of certain reactive gases.

Whole Air Collection. Collection of whole air samples using stainless steel canisters, evacuated glass bulbs, or similar devices is probably the simplest sampling approach. This approach is most useful for relatively stable volatile compounds such as hydrocarbons and chlorinated hydrocarbons with boiling points below 150°C. The canister samplers have the following advantages compared to solid sorbent tubes: (1) Breakthrough does not occur with canister sampling because the actual air sample is collected; (2) no thermal desorption is required; (3) canister pressure can be used as an indicator of correct sampler operation; (4) analysis of the canister sample can be repeated by using the remainder of the sample in the canister; and (5) the evacuated canisters can be used for sampling without power in the sampling location.

SUMMA passivated canister sampling with GC was developed by USEPA to determine semivolatile and volatile organic compounds. The canister is a sampling device used to collect and store whole air samples. It can be pressurized, thereby increasing the volume of the collected air sample. It has been demonstrated that the SUMMA passivation process, in which a pure chrome-nickel oxide layer is coated on the inner metal surface, increases the stability and the storage life of many organic compounds. However, certain compounds pose stability problems associated with storage by the formation of an oxide coating. Selection of an alternative container material can circumvent these problems in many cases. The most difficult problem associated with this method is the quantity of moisture collected in the canister. Too much moisture can clog the cryogenic trap and the capillary interface.

Due to the complexity of ambient air samples, only high resolution (capillary column) GC techniques are acceptable for most of the above methods. The GC/MS system should be capable of programing subambient temperatures. Unit mass resolution should be better than 800 amu, and GC/MS should be capable of scanning the 30-440 amu region every 0.5-1 sec. The measuring device should be equipped with a data system for instrument control as well as data acquisition, processing, and storage.

Impinger bubblers have traditionally been used for the collection of volatile organics from air. They are charged with a trapping solvent and require much more sampling time for collecting sufficient amounts of the analyte. The growing demand for greater sensitivity and time resolution suggested the development of a system based on solid adsorbent sampling and thermal desorption of VOCs followed by gas chromatographic analysis with cryogenic refocusing techniques.

An air sampling and analysis system for monitoring VOCs being emitted from the stacks of asphalt plants was designed. It was based on solid adsorbent sampling tubes with thermal desorption GC analysis using capillary column separations with cryogenic refocusing techniques. Air samples were collected using the solid adsorbent Tenax in stainless gas collection tubes. Stripping of the adsorbed analytes from the sampling tube was accomplished by thermal desorption and followed by gas chromatographic analysis.

Preparation of Tenax Cartridges. The following procedure was used to prepare a 5/8 in. tube containing Tenax cartridges.

Prior to use, the Tenax resin was subjected to a series of solvent extraction and thermal treatment steps. The operation was conducted in an area where levels of volatile organic com pounds (other than the extraction solvents used) are minimal. All glassware used in Tenax purification and all cartridge materials were thoroughly cleaned by rinsing with water followed by rinsing with acetone and drying in an oven at 250°C. The bulk Tenax was placed in a glass extraction thimble and held in place with a plug of clean glass wool. The resin was then placed in the Soxhlet extraction apparatus and sequentially extracted with methanol and then with pen-tane for 16-24 hr for each solvent for approximately 6 cycles/hr. Glass wool for cartridge preparation was cleaned in the same manner as that for Tenax.

The extracted Tenax was placed immediately in an open glass dish and heated under an infrared lamp for 2 hr inside a hood. Care was exercised to avoid overheating the Tenax with the infrared lamp. The Tenax was then placed in a vacuum oven (evacuated using a water aspirator) without heating for 1 hr. Then it was purged with an inert gas (helium or nitrogen) at a rate of 2-3 mL/min to aid the removal of solvent vapors. Then the oven temperature was increased to 110°C while maintaining the flow of inert gas for 1 hr. Then oven temperature control was shut off, and the oven was allowed to cool to room temperature. Before opening the oven, it was slightly pressurized with nitrogen to prevent contamination with ambient air. The Tenax was removed from the oven and sieved through a 40/60 mesh sieve (acetone rinsed and oven-dried) into a clean glass vessel.

If the Tenax was to be used later, it was stored in a clean glass jar with a Teflon-lined screw cap and placed inside a desiccator. The cartridge used for the monitoring of air was packed by placing a 0.5-1 cm long glass wool plug at the bottom of the cartridge and then filling the cartridge with Tenax up to approximately 1 cm from the top. Then a 0.5-1 cm long glass wool plug was placed on top of the Tenax. The cartridges were thermally conditioned by heating for 4 hr at 270°C while purging with an inert gas (helium at 100-200 mL/min).

The Desorption and GC Settings. The air sampling and analysis system was based on solid adsorbent sampling tubes and thermal desorption gas chromatographic analysis. After collection of air samples, sampling tubes were put into a desorber. Then the VOCs were twice condensed in cryogenic trap to improve capillary column resolution. Before cryogenic refocusing, a purge step served to remove any oxygen that remained in the tube. This eliminated the problem of the solid adsorbent reacting with the oxygen when heated, and it also removed traces of water from the tube.

Collected air samples in Tenax traps were first desorbed using a Tekmar Model 5010 automatic desorber connected to a Varian 3400 high resolution capillary GC with a flame ionization detector. The two systems were interfaced to automate the entire analysis.

The desorption conditions are as follows.

Prepurge

5 min at 10 mL/min

Desorb

8 min, 210°C, 10 mL/min

Cryotrap 1

150°C

Cryotrap 2

150°C

Transfer

10 min, 210°C

Inject

0.5 min, 210°C

The column used for GC/FID analysis was a cross-linked 5% phenyl and 95% methyl silicone gum, 50 m long, 0.2 mm diameter, and 0.5 pm film thickness fused silica bonded high resolution capillary column. Flow rates for the GC were

Hydrogen

30 mL/min

Air

30 mL/min,

Carrier gas

1 mL/min

Make-up gas

30 mL/min

The temperature program was (1) an initial temperature of 15°C for 8 min and (2) temperatures programed up to 210°C at 4°C/min. The detector range was selected as 10. A schematic diagram of the analytical system is shown in Figure 7.

Calibration and Quantification of Data. External standards were used to calculate a response factor for each compound of interest. The process involved the analysis of four calibration levels for each compound during a given day for the determination of the response factor (area! picomole). The linear least squares fit of a plot of picomoles versus area was used for the determination of the response factor. If substantial nonlinearity is present in the calibration curve, a nonlinear least squares fit should be employed. This process involves fitting the data to the equation

where Y is the peak area, X is the quantity of the component in picomoles, and the constants A, B, and D are coefficients to be determined from the regression analysis. If the instrumental response is linear over the concentration range of components, a linear equation (D = 0 in the equation above) can be employed. The system detection limit for each component can be calculated from the calibration standards, where the detection limit is defined as

where DL is the calculated detection limit in picomoles, A is the intercept of the regression analysis equation, and SD is the standard deviation of the samples with replicate determinations of the lowest concentration level.

The standard gas used for calibration consisted of a mixture of chloromethane, hexane, chloroform, 1,1,1-trichloroethane, carbon tetrachloride, trichloroethene, toluene, trichloroet-hylene, benzene, m,p-xylene, and o-xylene. These compounds were injected into an evacuated and clean 13-L stainless steel cylinder, and the cylinder was pressurized with zero grade helium. The preparation and analysis of the standard was done by Alphagaz, Edison, New Jersey. A 2-mL volume sample loop was used for the GC analysis. When the sample loop was flushed, the standard gas mixture passed through the sample loop. A three-port valve with a switch was used to allow helium gas to pass through the loop, flushing the standard gas onto a Tenax cartridge. The compounds in the standard gas were adsorbed by Tenax and quantified by gas chromatography. The following ideal gas equation was used for the calculation of the component concentrations in the standard gas.

where P is pressure (in atm), V is the volume of the sample loop, C is the concentration of each compound expressed as a fraction, R is the universal gas constant (0.082), T is temperature (in kelvins), and n is the amount of each standard gas injected (in moles).

At 25°C, sample concentration in ppb can be expressed as area of sample X C X 24.5 x 109

Sample concentration = -area of the standard X V-

Automatic Desorber

Tenax trap with contaminants

Carrier Helium 80psig, 1 ml/min

Cryogenic trap #1 to collect and then desorb organics

Cryogenic trap #2 to increase the resolution within GC

Automatic Desorber

Tenax trap with contaminants

Carrier Helium 80psig, 1 ml/min

Cryogenic trap #1 to collect and then desorb organics

Cryogenic trap #2 to increase the resolution within GC

Capillary dolumn

Nitrogen, 80psig 30 ml/min ; make up for capillary column

High Resolution Capillary Gas Chromatograph

Figure 7 A schematic diagram of the analytical system used for the air quality analysis. (From Meegoda et al. [8].)

Nitrogen, 80psig 30 ml/min ; make up for capillary column

Capillary dolumn

High Resolution Capillary Gas Chromatograph

Figure 7 A schematic diagram of the analytical system used for the air quality analysis. (From Meegoda et al. [8].)

where C is concentration of the compound in standard gas and V is the volume of the sample in liters. If V is at a different temperature and/or pressure, then the volume of gas should be converted to that at 25° and a pressure of 1 atm.

The air samples were collected by drawing air through the Tenax cartridge, a 5/8 in. stainless steel tube packed with 1.5 g of 60/80 mesh Tenax using a vacuum pump. Figure 8 shows the schematic for extracting air samples from stacks. Samples were drawn at approximately 500 mL/min.

In this experiment, 10 Tenax blank traps were spiked with known quantities of standard and then desorbed into the analytical system. The reproducibility can be expressed by the coefficients of variation (CV) of the target compounds. The CV and DL values for all the tested organic compounds are shown in Table 8.

Tenax Cartridge

Flow Meter

Valve

Vacuum Pump

Vent

Reducing Union

End Cap

Reducing Union

End Cap

Metal Cartridge

Tenax

Figure 8 Schematic diagrams of the setup used for air quality analysis. (From Meegoda et al. [8].)

Metal Cartridge

Tenax

Figure 8 Schematic diagrams of the setup used for air quality analysis. (From Meegoda et al. [8].)

Test Data. An air sampling and analysis system was designed to monitor volatile organic compounds emitting from stacks of asphalt plants. It was based on solid adsorbent sampling tubes with thermal desorption gas chromatographic analysis and used capillary column separations with cryogenic refocusing techniques. Air samples were collected using the solid adsorbent Tenax in stainless gas collection tubes. Stripping of the adsorbed analytes from the sampling tube was accomplished by thermal desorption and followed by gas chromatographic analysis.

During the field study, air samples were drawn for 3 min from the stack in the asphalt plant while the HMA concrete blended with PCS was prepared. The air samples were drawn through a filter to remove the dust going into the Tenax cartridge. The Tenax cartridge was connected to a vacuum pump. Three air samples were obtained for 3 min for test, and the Tenax samples were analyzed for concentrations of target organic compounds. Test results (first day, two soils with four different blends; second day, one soil with three different blends) are shown in Table 9.

Table 8 Method Specifications and Regulatory Limits for Fraction Released (Proposed N.J.) (ppb)

Regulatory level as average flux

Table 8 Method Specifications and Regulatory Limits for Fraction Released (Proposed N.J.) (ppb)

Regulatory level as average flux

Compound

DL (ppb)

CV

(mg/m2 • day)

Chloromethane

0.54

20.5%

270

Dichloromethane

0.13

9.2%

N/A

Hexane

0.03

11.8%

N/A

Chloroform

0.44

9.8%

6

1,1,1 -Trichloroethane

0.22

10.5%

494

Trichloroethylene

0.05

21.4%

13.8

Benzene

0.07

N/A

16.7

Toluene

0.06

6.7%

988

Tetrachloroethylene

0.03

8.7%

9.9

p,m-Xylene

0.05

7.5%

148

o-Xylene

0.06

9.5%

148

N/A = not available.

Areas of all peaks were added to compute the total nonmethane organic carbon (NMOC) emissions. A standard propane/helium gas mixture was used to obtain the GC calibration curve for concentration. Various known amounts of the mixture were injected to the GC and the peak areas were obtain to establish the calibration curve. Then the concentration corresponding to sum of all peak areas was calculated from the calibration curve and reported as total NMOC in Table 9. Figure 9 shows a comparison of total NMOC emission from a regular asphalt plant and that uses PCSs (day 2 test 3 in Table 9).

The concentrations of the target chemicals were less than 1 ppm, and total concentrations of VOCs were lower than the NJDEPE specification for regular asphalt plants (<250 ppm).

Table 9 Air Quality Test Results

Concentration in ppb

Table 9 Air Quality Test Results

Concentration in ppb

Test No.

Toluene

P&M Xylene

O-Xylene

Hexane

Total NMOCs

Day 1 Test 1

52

52

58

8

68,450

Day 1 Test 2

71

28

19

25

11,300

Day 1 Test 3

23

14

24

9

45,590

Day 1 Test 4

90

69

53

11

14,880

Day 2 Test 1

642

205

177

545

86,090

Day 2 Test 2

443

88

99

283

155,230

Day 2 Test 3

366

30

15

93

94,750

Cone, levels

after plant was

11

2

4

149

2,650

stopped

Background

conc. around plant

5

2

2

0

1,210

Stack conc.

from an ordin.

24

16

0

26,700

6

HMA plant

without PSCs

Source: Meegoda et al.,

, 1992 [9],

NMOC

Normal HMA HMA with PCSs

HI ppmC ËH3 lb Carbon/hr

Figure 9 Comparison of non-methane organic carbon (NMOC) emission by a regular asphalt plant and one that uses PCSs. Solid bars, ppm carbon; hatched bars, pounds of carbon per hour. (From Meegoda et al. [9].)

Further details can be found in the 1991 report to NJDEPE on the use of petroleum-contaminated soils in construction material production [12].

0 0

Post a comment