Detecting and Defining Air Pollutants: One ... - ACS Publications

andApproaches. The Texas Air Control Board is the agency responsiblefor maintaining the quality of the ambient air of. Texas. An extensive network of ...
0 downloads 0 Views 7MB Size
Detecting and Defining Air Pollutants: One Laboratory's Experiences and Approaches

1074 A ·

The Texas Air Control Board is the agency responsible for maintaining the quality of the ambient air of Texas. An extensive network of continuous and noncontinuous monitors is used to detect air quality trends in the state. Regulations of the board limit the emission of air contaminants from stationary sources. The agency's well-equipped laboratory analyzes network samples obtained by noncontinuous monitors and samples submitted for analysis by our regional investigators. The samples submitted may be gaseous, liquid, or solid; they are submitted for a variety of reasons. Analyses are sometimes necessary before a permit can be issued to a new source, to check the performance of existing sources, or to answer citizens' complaints. The submitted samples are reviewed by the Laboratory Director (Figure 1) and assigned to the appropriate analyst, depending on his or her discipline. In addition, any special handling or preparation of the sample is discussed at this time. Elemental Analysis

investigative staff of the agency is unaware? • Could the analytical technique benefit the agency in other ways? At that time, atomic absorption (AA) was being used to perform metal analyses. Those acquainted with AA are aware of the digestion and dilution necessary to prepare the sample. If elements to be determined are in a complex matrix, as samples from polluting sources often are, it is necessary to prepare time-consuming internal standards to correct for interelemental interferences. After considering several approaches, a comparison of AA with an emerging technique called energy dispersive X-ray fluorescence spectroscopy (XRF) was undertaken (1). The manpower, cost of chemicals, response time, and capital investment were studied in analyzing for some 12 metals from 10 sampling sites (2). The cost effectiveness of the XRF technique quickly became apparent. The end result has been the employment of the XRF system, so that the agency's 100-plus particulate samples, collected every six days, are each analyzed for some 32 elements.

Early in the history of the Texas Air Control Board, the administration asked the analytical staff several questions: • Can the general public's exposure to some of the more common and toxic elements be determined economically? • Can the analytical results be used to identify polluting sources of which the

A computer controls the XRF during the analysis of the samples, then massages the data and provides a printout of the results in the appropriate units. Some 48 samples are analyzed for 32 elements in a 12-hour period. The quantity of data available on the public's exposure to some 32 elements in Texas is now rather extensive. The cost per determination is far

A N A L Y T I C A L CHEMISTRY, V O L . 5 2 , NO. 9, AUGUST

1980

0003-2700/80/A351-1074$01.00/0 © 1980 American Chemical Society

The Analytical Approach James L. Lindgren Henry J. Krauss John S. Mgebroff Texas Air Control Board 6330 Highway 290 East Austin, Tex. 78723

Sample Review

Elemental Analysis

Organic Analysis

Figure 1. Sample flow and multiinstrumental options available to accomplish the analysis of complex samples

less than can be achieved by other analytical approaches. The second of the questions was also answered early in our search for an alternative to AA. Analysis of the results produced by that first comparative study by the analytical staff, who were not familiar with the industry in the area sampled, indicated a source of Zn. By correlating the analytical results with the meteorology of the sampling day, the location of the suspected source was narrowed and later verified by authorities familiar with local industry. In addition, several previously unsuspected sources were found to be polluting neighboring communities with undesirable levels of certain potentially toxic metals. A smelter was closed when undesirable levels of Mo, Sb, and other metals were detected in its emissions. Several other instances demonstrated that our analytical approach had satisfied the second question. It was only after several years of experience that the third question was adequately answered. The first demonstration was in a court case against an emitter of lead. The defendant claimed that automotive emissions were the source of the lead. Several years of XRF data observation had established that a definite Pb/Cl and Pb/Br ratio existed on the particulate samples collected where automobiles were the only source of lead emissions. The XRF, which is capable of detecting CI and Br adsorbed on the particu-

late matter, failed to establish this definite ratio, though a substantial number of samples near the smelter were collected and analyzed. The XRF data won our case, and since the samples were analyzed nondestructively, they were available for verification had this been necessary. A second answer to question three surfaced when a permit engineer commented to a laboratory staff analyst that there should be an easier way to determine the emissions from a stack than the traditional, time-consuming stack sampling technique. The result of that conversation was a collaborative study with the Source Evaluation Division to determine if XRF analysis of the fuel supply could be a reliable indicator of the sulfur emissions from a stationary source, since SO2 emissions are regulated. The results of the XRF approach consistently agreed with the stack sampling technique (3). The analysis time is only 5-10 min for a sulfur analysis of boiler fuels including the sample preparation, which requires no digestion or dilution. Existing facilities, which propose to change their source of fuel, and new facilities, before they come on line, are routinely sampled by this fuel analysis approach for percent sulfur in the fuel. The permit engineer uses the XRF data to calculate the potential sulfur dioxide emissions from the plant and only if the results are marginal is stack sampling considered. Routinely, the fuel samples, be they oil, coal, or lignite, are analyzed for 32

elements. Sulfur content attracts the most interest, but a 30-min analysis time by the XRF provides the engineer with the elemental concentration of the fuel so that he may decide if toxic metals emissions could be a potential health hazard as well. The XRF approach has been most beneficial to our analyses but is not the complete answer. Our XRF capability covers only an area of the periodic table, and also lacks sensitivity for some of the elements. Flameless AA is the technique of choice for Be since the XRF cannot detect it, and for As since the XRF is much less sensitive. Indeed, AA is used for a number of other elements. There are occasions when an elemental analysis does not explain the source of a pollutant. Optical and scanning electron microscopy are then employed as complements to XRF. Organic Analysis

Gas chromatography (GC) , high pressure liquid chromatography (HPLC), and gas chromatography/ mass spectrometry (GC/MS) are employed in organic analyses as complements to each other. Environmental samples submitted for detection of organic components are seldom clean or simple and often require special concentration steps. HPLC has been effectively used to shorten and simplify the analysis of samples for herbicides. An example of this is the analysis of vegetation for the herbicide Stam (Propanil) (4).

ANALYTICAL CHEMISTRY, VOL. 52, NO. 9, AUGUST 1980 . 1075 A

T h e original GC procedure calls for a 17-h hydrolysis of the vegetation in 25% N a O H followed by steam distilla­ tion into 2N HC1. T h e distillate is washed with benzene, t h e p H adjusted to 10, and again extracted with ben­ zene, passing the benzene t h r o u g h so­ dium sulfate. T h e collected extract is concentrated a n d analyzed by GC for 3,4-dichloroaniline using a halogensensitive detector. T h e substitution of H P L C for GC does not remove the need for t h e hydrolysis a n d steam dis­ tillation, b u t t h e distillate can be m a d e alkaline and extracted with ben­ zene a n d t h e benzene extract concen­ t r a t e d a t this point. It is not necessary to dry t h e extract since t h e Cie re­ versed phase column used in H P L C h a n d l e s t h e water content of t h e ex­ t r a c t with no difficulty, a n d the detec­ tors normally used in H P L C are not h a r m e d by water, as are some electron c a p t u r e detectors. T h e gradient elution capability of t h e H P L C system allows one to choose solvents and gra­ d i e n t conditions which generally will e n h a n c e t h e separation of compounds of close polarity. A dual channel UV detector is t h e m e a n s of monitoring t h e eluting sam­ ple c o m p o n e n t s of an H P L C analysis. A fluorescence spectrophotometer e q u i p p e d with a 10 μΐ flow-through cell follows t h e dual channel UV. W h e n t h e ratio of t h e two UV chan­ nels fails to confirm an eluting peak, it can be stopped in the 10 μΐ sample cell a n d a fluorescence s p e c t r u m can be obtained to further aid in t h e qualita­ tive confirmation. Of course, t h e com­ p o u n d has to have fluorescing proper­ ties. T h i s combination of H P L C and detectors provides confirmation of some polyaromatic hydrocarbons in a m b i e n t air after they have been con­ c e n t r a t e d onto commercially available Cie b o n d e d liquid chromatographic column packing material (5). For t h e concentration of C2-C12 or­ ganic c o n t a m i n a n t s in a m b i e n t air, t h e laboratory has developed a technique which is effective a n d useful, though n o t necessarily unique to our organi­ zation (6). Figure 2 depicts t h e 6" X 6 m m pyrex tubes, which are packed with 3 cm of 60-mesh activated char­ coal a n d 7 cm of 80/100-mesh T e n a x GC. T h e brass caps use a soft front ferrule in order to form a diffusionfree seal. T h e s e absorber t u b e s have been successfully used to concentrate con­ t a m i n a n t s from air samples s u b m i t t e d to t h e laboratory a n d from the a t m o ­ sphere surrounding industrial sites. In a d e m o n s t r a t i o n of t h e latter, a field investigator was provided with a q u a n t i t y of t u b e s for t h e purpose of quickly sampling a specific site for vinyl chloride m o n o m e r (VCM) emis­ sions (7). One of t h e more practical as-

Desorb

Load

Figure 2. Charcoal/Tenax sample concentration tube (top) and tube oven used for desorbing c o l l e c t e d sample (bottom)

pects of using these t u b e s a t remote sites is t h a t only a gas-tight syringe a n d reasonable care in handling are needed for loading. Electrical power or b a t t e r y packs are not required for p u m p s — a n obvious appeal to field in­ vestigators for spot, u n a n n o u n c e d s a m p l e acquisitions. T h e t u b e s can be shipped or, as in this instance, conve­ niently carried to t h e laboratory. T o analyze these t u b e s for VCM, t h e t u b e oven (Figure 2) is m o u n t e d directly on a gas chromatograph, re­ placing t h e s e p t u m cap. T h e analysis is accomplished by interrupting t h e carrier gas flow, inserting a loaded

t u b e into the t u b e oven, retightening the fittings, reconnecting the carrier gas flow, initiating the power t o t h e t u b e oven, a n d initiating t h e t e m p e r a ­ ture program of t h e column oven. T h e sample is desorbed onto an η - O c t a n e on Porasil column, 1/8" X 2 m, held a t 0 °C for 7 min, t h e n increased a t 6 ° C / min for 14 min, a n d finally a 30 ° C / min increase to a final t e m p e r a t u r e of 180 °C. T h e detection of VCM is accom­ plished with a photoionization detec­ tor (PID) and flame ionization detec­ tor (FID) in series. T h e simultaneous trace of t h e P I D a n d F I D responses by

PROSTAGLANDINS U p j o h n Diagnostics n o w offers y o u 9 p r o s t a g l a n d i n c o m p o u n d s —for use as laboratory research chemicals. T h e y are: • PGEj

· PGF20; T r o m e t h a m i n e

• PGE2

· 13,14-diH-15-keto P G F 2 t t

• PGA,

· Thromboxane B2

• PGA2

· 6-keto P G F , a

• PGFia For m o r e information, call or write:

Upjohn diagnostics U p j o h n Diagnostics/Division of T h e U p j o h n C o m p a n y K a l a m a z o o , Michigan 4 9 0 0 1 / ( 6 1 6 ) 3 8 5 - 7 1 1 1 To assure consistent quality, Upjohn prostaglandin chemicals are analyzed prior to release by our Quality Assurance Unit. For details about any batch analysis, contact Upjohn Diagnostics, Kalamazoo, Michigan. Copyright 1979, The Upjohn Company

UD 79-014

CIRCLE 212 ON READER SERVICE CARD

ANALYTICAL CHEMISTRY, VOL. 52, NO. 9, AUGUST 1980 · 1077 A

a two-pen recorder allows t h e two responses t o b e ratioed, t h u s adding confidence to t h e qualitative d e t e r m i nation. T h r e e to five p p b by volume of VCM can confidently be detected with this analytical scheme when only 300 cc of air is sampled. T h e sample t u b e s are versatile enough to be used for t h e concentration of a variety of c o m p o u n d s . Caution should be exercised if t h e t u b e s are to be used for collection of benzene or toluene, because h e a t desorbing of T e n a x routinely results in peaks a t t h e retention times of those c o m p o u n d s . Some investigators have developed rigorous clean-up procedures which prevent this problem (8, 9). T h e r e are m a n y examples of a mult i i n s t r u m e n t a p p r o a c h t o analysis of samples. Our laboratory was involved in sampling a n d analyzing for nitrobenzene, which was suspected of originating from waste oils t h a t h a d been applied to public roads in several communities. Originally, water samples from t h e ditches of the affected area were s u b m i t t e d . T h e s e were r a t h e r dirty a n d could n o t be directly analyzed by G C / M S without clean-up a n d concentration. High pressure liquid c h r o m a t o g r a p h y was employed t o hasten t h e d e t e r m i n a t i o n of nitrobenzene's presence. G C / M S was t h e n used for confirmation of the suspected

peak. After determining t h a t nitrobenzene was indeed p r e s e n t in t h e runoff, t r u e a m b i e n t air samples were obtained by absorption in ethylene glycol a n d by using t h e previously described concentration t u b e s with s u b sequent GC analysis. H P L C became the t e c h n i q u e of choice for t h e analysis of these samples because of the simplicity of the analytical scheme. Monitoring continued until t h e oil h a d been removed from t h e area. T h e complexity of a m b i e n t air samples often requires a m u l t i i n s t r u m e n tal a p p r o a c h if t h e sample is t o be qualitatively a n d quantitatively analyzed so t h a t t h e information can be used for corrective action or t o identify potential problems of air pollution. Competency in staff proficiency m u s t be m a i n t a i n e d , a n d sophisticated ins t r u m e n t a t i o n m u s t be available. T h e need for effective a n d efficient sample concentration techniques c a n n o t be overlooked. It is a n area which needs to be pursued, a n d advantages a n d limitations of a particular procedure should be identified before it is routinely used for t h e collection a n d conc e n t r a t i o n of a m b i e n t air contaminants.

Acknowledgment T h e generous s u p p o r t of the Texas Air Control Board a n d t h e contribu-

tions a n d assistance provided by our staff a n d a d m i n i s t r a t i o n in t h e preparation of this presentation are gratefully acknowledged.

References (1) J. R. Rhodes, Am. Lab., (7), 57 (1973). (2) J. R. Rhodes, A. H. Pradzynski, C. B. Hunter, J. S. Payne, and J. L. Lindgren, Environ. Sci. and TechnoL, 6 (10), 922 (1972). (3) J. S. Mgebroff and J. S. Payne, "The Determination of Trace Elements in Fuel Oil by Energy Dispersive X-ray Fluorescence Analysis," presented at the 29th Annual Meeting of the Pittsburgh Conference on Analytical Chemistry and Applied Spectroscopy. (4) J. L. Lindgren, H. J. Krauss, and J. S. Payne, "Determination of Pesticides in Environmental Samples: Substituting HPLC for GC," presented at the 30th Annual Meeting of the Pittsburgh Conference on Analytical Chemistry and Applied Spectroscopy. (5) J. L. Lindgren, H. J. Krauss, and M. A. Fox, Journal of the Air Pollution Control Association, 30 (2), 166 (1980). (6) W. Bertsch, R. C. Chang, and A. Zlatkis, J. Chromatogr. Sci., 12 (4) 175 (1974). (7) J. L. Lindgren and G. Speller, "Determination of Vinyl Chloride Monomer in the Ambient Air Near Point Source Emissions," presented at the 30th Annual Meeting of the Pittsburgh Conference on Analytical Chemistry and Applied Spectroscopy. (8) E. D. Pelizzari, private communication. (9) L. T. Freeland, private communication.

PHENOLS AND AROMATIC AMINES? We're the specialists in liquid c h r o m a t o g raphy for trace organic determinations. Let us know your problem. We will provide detailed applications support and a selection of instrumentation which will do your job in the most cost effective manner. LC systems based on ultraviolet a n d / o r electrochemical detection are tailor made for environmental analysis. Temperature control, gradient elution, automated trace enrichment, and column switching are options which may be adapted to your needs. Circle the bingo card for general information, but if you want specific answers, write to our Environmental Services Department and tell them what y o u can about your samples.

A N I L I N E S · B E N Z I D I N E S · M O C A · PENTACHLOPHENOL· C R E S O L S · PARABENS· HYDROQUINONES

bas

BIOANALYTICAL

SYSTEMS INC.

P.O. Box 2206 · W. Lafayette. IN 47906 .(317) 463-2505 · telex 276141 BAS WLAF

CIRCLE 22 ON READER SERVICE CARD

1078 A · ANALYTICAL CHEMISTRY, VOL. 52, NO. 9, AUGUST 1980