The Chemical Analysis of Argonne Premium Coal Samples

Edited by Curtis A. Palmer
U.S. Geological Survey Bulletin 2144

Determination of Mercury and Selenium in Eight Argonne Premium Coal Samples by Cold-Vapor and Hydride-Generation Atomic Absorption Spectrometry

By Richard M. O'Leary

ABSTRACT

The methods for the determination of mercury and selenium in whole coal by cold-vapor atomic absorption spectrometry (CVAAS) and hydride-generation atomic absorption spectrometry (HGAAS) are described. The Argonne Premium Coal samples were analyzed in triplicate to determine the precision of the method. The averaged values ranged from 0.01 to 0.39 µg/g for mercury and 0.60 to 6.2 µg/g for selenium. Mercury and selenium were also determined in standard reference materials from the National Institute of Standards and Technology (NIST), the Community Bureau of Reference (BCR) of the Commission of the European Communities, the National Research Council of Canada (NRCC), and the U.S Geological Survey (USGS). Results obtained by these methods were compared with the published values to determine the accuracy of the methods.

INTRODUCTION

Analytical techniques for the determination of mercury and selenium in coal are increasing in importance. With the passage of the 1990 Clean Air Act Amendments (CAAA), the Environmental Protection Agency (EPA) has been given the authority to set emission standards for a number of potentially hazardous air pollutants (HAP's) generated by a number of specific combustion sources. Mercury, selenium, and nine other elements present in coal are among the 189 pollutants identified as air toxins in the CAAA legislation. Although their concentrations in coal are minor, they represent a potentially significant release of mercury and selenium to the environment because of the large tonnage of coal burned in powerplants.

The most common instrumental technique for determining mercury is cold-vapor atomic absorption spectrometry (CVAAS). Some CVAAS techniques preconcentrate the mercury using a gold amalgam, then thermally release the mercury by inductively heating the gold; however, alternative techniques analyze the mercury vapor directly. Other instrumental techniques include graphite furnace atomic absorption spectroscopy (GFAAS), cold-vapor atomic fluorescence spectroscopy (CVAFS), and neutron activation analysis (NAA). Various sample preparation procedures include oxygen bomb combustion, direct coal combustion, microwave digestion, and hotplate acid digestion.

In the procedure described here, a 0.150-g coal sample was decomposed by a heated mixture of nitric and sulfuric acids and vanadium pentoxide. The sample solution was introduced to a continuous-flow CVAAS system, where it was complexed and reduced with a solution of hydroxyl-amine hydrochloride and sodium chloride prior to further reduction with stannous chloride. The mercury vapor was then separated from the liquid in the phase separator before entering the quartz cell for the determination of the mercury concentration.

Like mercury, selenium can be determined by a number of different instrumental techniques such as GFAAS, NAA, and hydride-generation atomic absorption spectrometry (HGAAS). Of the AAS techniques, hydride generation is the technique of choice because it is relatively interference free as compared to GFAAS.

In the method used in this study, a 0.100-g sample of pulverized coal was digested at 150-200°C with a mixture of concentrated sulfuric, nitric, and perchloric acids until a clear to yellow solution was reached. After the addition of 6 M hydrochloric acid, the solution was allowed to set to permit the selenium to reduce the Se +3 state. The diluted solution was then introduced by way of an autosampler and peristaltic pump to a Varian VGA-76 hydride generator coupled to an atomic absorption spectrophotometer. The sample solution was then mixed with concentrated hydrochloric acid and 0.35 percent sodium borohydride, and the resultant selenium hydride was then transported with argon gas to an air-acetylene flame-heated quartz cell for atomization and estimation of the selenium concentration.

PROCEDURE FOR MERCURY

Approximately (scooped) 0.1 g vanadium pentoxide,^1,2 1.5 mL concentrated sulfuric acid, and 3.5 mL concentrated nitric acid were added to 0.150 g of whole coal in a 16 x 150-mm disposable test tube and mixed. The test tube was placed in an aluminum heating block and covered with a watch glass. The temperature was ramped gradually to 150°C over a 2-hour period. The tube was heated overnight at this temperature and then removed and allowed to cool. The sample was diluted to 15 mL with water, capped, and shaken for 5 minutes. It was then centrifuged at 1,000 rpm (revolutions per minute) for 5 minutes, and approximately 12 mL of solution was transferred to a 16 x 100-mm disposable test tube.

The mercury was determined by using a continuous-flow-through CVAAS system (fig. 1) as described by Kennedy and Crock (1987). The test tube containing the sample solution was placed in the autosampler. The sample solution was fed from the autosampler by a peristaltic pump into a continuous-flow system, where it was mixed with a reducing-complexing solution of 3 percent hydroxylamine hydrochloride (reagent grade) and 3 percent sodium chloride (reagent grade) in 10 percent sulfuric acid. Next, the sample was further reduced with a 10 percent stannous chloride in 10 percent hydrochloric acid. The sample then entered a phase separator where the mercury gas passed through the flow-through cell of the AAS for measurement and the liquid was discharged to waste. The absorbance indicating the mercury concentration was recorded on a strip chart, and peak heights were measured. The operating conditions for the AAS are shown in table 1.

The samples were compared against a calibration curve generated by analyzing standards in the 1- to 15-µg/L range. The calibration standards were made by dilution of a 1.47 µg/mL mercury solution (standard reference material (SRM) 1641c) obtained from NIST (National Institute of Standards and Technology). The calibration standards contained a final concentration of 3.7 M nitric acid, 1.8 M sulfuric acid, and 0.5 percent (w/v, weight per volume) sodium dichromate (reagent grade).

PROCEDURE FOR SELENIUM

The method used for determining selenium is a modification of that described by Aruscavage (1977). Twenty milliliters of concentrated hydrochloric acid³ and 2 mL concentrated sulfuric acid were added to 0.100 g of whole coal in a 250-mL Erlenmeyer flask. After the mixture was allowed to stand overnight, 3 mL of perchloric acid (redistilled) was added. A claw refluxer was added to the flask, and the solution was heated on a hotplate at 150-200°C for 30 minutes. The refluxers were removed, and heating of the solution was continued until the development of dense white fumes and a clear to yellow solution. The solution then was removed from the hotplate. When the solution was cool, 25 mL of 6 M hydrochloric acid was added, and the solution was allowed to stand for 30 minutes to permit the selenium to reduce to the Se +3 state. The contents of the flask were transferred to a 60-mL polyethylene bottle and diluted to 55 g with water.

The selenium was determined by the HGAAS system using a Varian VGA-76 hydride generator coupled to a Perkin-Elmer atomic absorption spectrophotometer (fig. 2). The solution was transferred to 13 x 100-mm test tubes and placed in an autosampler. The sample solution was then fed from the autosampler to the hydride-generation system by a peristaltic pump, where it was mixed with concentrated hydrochloric acid and 0.35 percent sodium borohydride. The selenium hydride was then transported with argon gas to the air-acetylene flame-heated quartz furnace of the AAS for atomization. The absorbance indicating the selenium concentration was registered on a strip chart recorder, and peak heights were measured. The operating conditions for the AAS are shown in table 1.

The samples were compared against a calibration curve generated by analyzing standards in the 5- to 20-µg/L range. The calibration standards were made by dilution of a commercially prepared 10-µg/g selenium standard in 10 percent HCl. The calibration standards contained a final concentration of 3 M hydrochloric acid and 0.72 M sulfuric acid.

DISCUSSION

The eight Argonne Premium Coal samples were analyzed for selenium in triplicate on one day and for mercury in triplicate on two nonconsecutive days. All analyses were performed in the U.S. Geological Survey laboratory in Denver, Colo. The averaged values for mercury range from 0.01 to 0.39 µg/g, and the averaged values for selenium range from 0.60 to 6.2 µg/g as shown in table 2. As a measure of quality control, several standard reference materials were also analyzed for mercury and selenium by CVAAS and HGAAS, and the results are reported in tables 3 and 4, respectively. The standards chosen were coal, shale, a plant, and marine sediment. The tables compare the published values of the reference materials with the mean and standard deviation obtained from these methods.

The lower limits of determination for mercury and selenium are 0.01 and 0.1 µg/g respectively, which are based on three times the standard deviation of the blank. The precision of the mercury values is in the range of 3.5 to 12 percent relative standard deviation (%RSD), and the accuracy, based on the percent recovery (%R), which compares this method's values with the recommended values (RV), ranges from 76 to 104 percent (table 3). The precision of the values for selenium is in the range of 8 to 11 %RSD, and the accuracy or %R ranges from 95 to 119 percent of the recommended values (table 4). Both methods offer a technique that is simple and rapid, and both are applicable to a wide range of organically based sample.

¹Unless otherwise noted, all chemicals used in the mercury determinations are of Baker "Instra Analyzed" quality or are labeled "Suitable for mercury determinations," and the water is deionized.
²Some brands of vanadium pentoxide (reagent grade) contain trace amounts of mercury and need to be roasted at 500 °C prior to use.
³Unless otherwise noted, all chemicals used in the selenium determinations are Baker "Instra Analyzed" or of equal purity, and the water is deionized.

REFERENCES

Aruscavage, Philip, 1977, Determination of arsenic, antimony, and selenium in coal by atomic absorption spectrometry with a graphite tube atomizer: U.S. Geological Survey Journal of Research, v. 5, no. 4, p. 405-408.

Govindaraju, K., ed., 1989, 1989 Compilation of working values and sample description of 272 geostandards: Geostandards Newsletter, v. 13, Special Issue, p. 67.

Griepink, B., Colinet, E., and Wilkinson, H.C., 1986, The certification of the contents (mass fraction) of carbon, hydrogen, nitrogen, chlorine, arsenic, cadmium, manganese, mercury, lead, selenium, vanadium and zinc in three coals; gas coal CRM No. 180, coking coal CRM No. 181, and steam coal CRM No. 182: Commission of the European Communities, Community Bureau of Reference (Brussels), Report EUR 10366 EN, 170 p.

Kennedy, K.R., and Crock, J.G., 1987, Determination of mercury in geological materials by continuous-ßow, cold-vapor, atomic absorption spectrophotometry: Analytical Letters, v. 20, p. 899-908.

Lengyel, J., Jr., DeVito, M.S., and Bilonick, R.A., 1994, Interlaboratory and intralaboratory variability in the analysis of mercury in coal: Library, Pa., Consol Inc.

National Bureau of Standards, 1974, National Bureau of Standards certificate of analysis, standard reference material 1632, trace elements in coal (bituminous): Washington, D.C., National Bureau of Standards, 2 p.

------1978a, National Bureau of Standards certificate of analysis, standard reference material 1632a, trace elements in coal (bituminous): Washington, D.C., National Bureau of Standards, 2 p.

------1978b, National Bureau of Standards certificate of analysis, standard reference material 1635, trace elements in coal (subbituminous): Washington, D.C., National Bureau of Standards, 2 p.

[ Contents ]

Created by the EERT WWW Staff.