Link to USGS Home Page

METHODS FOR SAMPLING AND INORGANIC ANALYSIS OF COAL
U.S. Geological Survey Bulletin 1823

Edited by D.W. Golightly and F.O. Simon


ANALYSIS OF COAL ASH BY ATOMIC ABSORPTION SPECTROMETRIC AND SPECTROPHOTOMETRIC METHODS

By Floyd W. Brown and Hezekiah Smith

Abstract

Methods for determining Al, Ca, Fe, Mg, Mn, P, K, Si, Na, and Ti in coal ash are described. A bead formed by fusing a mixture of sample and lithium metaborate--lithium tetraborate flux at l,000° C for 45 min is dissolved in dilute nitric acid. Measurements are made by atomic absorption spectrometry and by spectrophotometry on portions of the resulting solution. Selected silicate standards are used as control samples to assure the quality of measurements. The precision of the methods commonly ranges from 1 to 10 percent relative standard deviation. The methods are both accurate and inexpensive and provide a good alternative to measurements by X-ray fluorescence spectrometry.


INTRODUCTION

The determination of major elements in coal ash requires the analysis of high-temperature ash residues having compositions that resemble those of silicate rocks. Thus, methods used in U.S. Geological Survey laboratories for routinely determining major elements (Al, Ca, Fe, K, Mg, Mn, Na, P, Si, and Ti) in silicate rocks (Shapiro, 1975) are, with certain modifications, applicable to coal-ash residues. The elements Al, P, Si, and Ti are determined spectrophotometrically, and the other six elements (Ca, Fe, K, Mg, Mn, and Na) are determined by flame atomic absorption spectrometry (AAS). The procedures were developed to facilitate the analysis of large numbers of samples.

Each sample, which consists of 100 to 200 mg of coal ash ground to pass through a 100-mesh sieve, is mixed with a reagent-grade lithium metaborate--lithium tetraborate flux and fused at 1,000° C for 45 min in a graphite crucible. Use of a mixed flux (1 part anhydrous lithium metaborate (LiBO2) and 2 parts anhydrous lithium tetraborate (Li2B4O7)) prevents strong adhesion of the final bead to the wall of the graphite crucible. This feature is particularly important for samples having iron concentrations greater than 15 percent. Also, the use of the mixed flux produces more reliable values for silica than do fusions with only lithium tetraborate.

In a practical procedure for determining major elements, the flux-to-sample ratio should be kept low to avoid the introduction of impurities, but the ratio must be high enough to assure complete decomposition of the sample. Concentrations of the major elements in solution must be sufficiently high to accommodate the least sensitive determinations but low enough to prevent silica from precipitating. These requirements are satisfied if 0.2 g of sample is used with 1.2 g of flux. The same flux-to-sample ratio (6 to 1) is used with different sample sizes. The bead that results from the fusion of 0.2 g of sample is dissolved in dilute nitric acid and diluted to 250 mL. Concentrations of major elements in this solution are determined by either spectrophotometry or by AAS.


SPECTROPHOTOMETRIC PROCEDURES

Silicon (as SiO2) is determined spectrophotometrically by measuring the absorbance of a molybdenum complex at 640 nm (Bunting, 1944). In concentrations greater than 200 m g/mL, silica forms polymers that do not react with the molybdate reagent. Formation of these polymers is prevented by the addition of fluoride, as sodium fluoride, to each solution. Because the fluoride solution dissolves glass, thus giving erratic results in determinations of silica, containers made of polyethylene, TeflonTM, or some similar material should be used. The interference caused by phosphate, which forms a phosphomolybdic acid complex, is avoided by adding tartaric acid to destroy the complex.

Aluminum (as Al2O3) is determined spectrophotometrically by measuring the absorbance at 475 nm of the aluminum complex with calcium alizarin red-S (Parker and Goddard, 1950). Iron and titanium also form colored complexes that absorb at 475 nm, but these interferences are eliminated by the addition of potassium ferricyanide and thioglycolic acid as complexing agents.

The titanium determination (as TiO2) is based on the use of disodium-1,2-dihydroxybenzene-3,5-disulfonate (Tiron) as a spectrophotometric reagent. The reagent forms a lemon-yellow titanium complex that has a high molar absorptivity (Yoe and Armstrong, 1947). The absorbance of this titanium complex is nearly independent of pH over the range 4.3 to 9.6. Ferric iron also reacts with Tiron to produce a purple complex, but this interference is eliminated by reducing the ferric iron with sodium dithionate solution buffered at pH 4.7. The absorbance of the yellow titanium complex at 430 nm is measured spectrophotometrically.

Phosphorus (as P2O5) is determined spectrophotometrically by measurement of the absorbance of a heteropoly molybdenum-blue complex that forms upon reduction of heteropoly phosphomolybdic acid with stannous chloride solution. For the conditions used, only the molybdenurn combined with phosphorus is reduced. and the excess molybdic acid is unaffected.


ATOMIC ABSORPTION SPECTROMETRIC METHODS

An aliquot of the original sample solution is diluted with lanthanum solution, and the absorption for Ca, Fe, K, Mg, Mn, or Na is measured by AAS. Operating conditions for the AAS instrument are presented in table 16.


PRECISION OF ANALYSIS

The precision for major-element analyses, which is defined as the reproducibility of replicate determinations of a particular analyte, is described in table 17. The values of relative standard deviation in this table are estimates based on replicate determinations of major elements in both standards and samples.


PREPARATION OF SAMPLE SOLUTION

Reagents and Supplies

1. Flux mixture: 1 part reagent grade lithium metaborate (LiBO2) and 2 parts anhydrous lithium tetraborate (Li2B4O7) are thoroughly mixed and stored in a closed container. Several hundred grams of the flux mixture should be prepared at one time.

2. Standard control samples: Several U.S. Geological Survey silicate standards, such as W-2, BHVO-1, AGV-1, G-2, and RGM-1 and National Bureau of Standards (NBS) 1633a coal fly ash are analyzed along with the samples. Other established standards may be substituted.

3. Graphite crucibles: Cylindrical, 25-mm outside diameter (20-mm inside diameter), high-purity graphite.

4. Dilute nitric acid, 8 M: Prepare several liters.

5. PolYethylene bottles: 250 mL, with caps.

Procedure

1. Weigh 0.2000 g of coal ash sample and mix thoroughly with 1.2 g of flux mixture on glassine paper. Transfer the sample-flux mixture into a graphite crucible.

2. Weigh the standard control samples along with the other samples. Also, take through the procedure a crucible that contains only the flux mixture as blank.

3. Fuse the sample-flux mixtures in a muffle furnace at 1,000° C for 45 min.

4. Remove the crucibles from the furnace and allow them to cool to room temperature. In most cases, the beads produced are easily dislodged from the crucibles by gentle tapping. If the iron content of a particular sample is exceptionally high, a spatula may be needed to remove the bead.

5. Place each bead into a 250-mL polyethylene bottle, and add a TeflonTM-covered, 39-mm magnetic stirring bar.

6. Add 50 mL of boiling distilled water to each bottle from a polyethylene graduated cylinder, and transfer the bottle to a magnetic stirrer. Then, add 5 mL of 8 M nitric acid to each bottle.

7. Stir each solution rapidly for approximately 60 min. Visually examine the bottle to be sure that dissolution of the bead is complete before going to the next step of the procedure.

8. Remove the bottles from the stirrer, and add approximately 100 mL of distilled water to each. Catch the magnetic stirring bar in a funnel while pouring each solution into a separate 250-mL volumetric flask. Rinse the bottle, add water to the volumetric flask to mark, and mix. Pour the solution back into the polyethylene bottle for storage.


DETERMINATION OF SILICON (AS SiO2)

Reagents and Equipment

1. Ammonium molybdate solution: Dissolve 6.0 g of ammonium molybdate {(NH4)6Mo7O24· 4H2O} in 1 L of distilled water.

2. Tartaric acid solution: Dissolve 16 g of tartaric acid {H2C2H4O6} in 1 L of distilled water.

3. Reducing solution: Dissolve 0.28 g of sodium sulfite, 3.6 g of sodium bisulfite, and 0.06 g of 1-amino-2-naphthol-4-sulfonic acid in 1 L of distilled water. Prepare this solution within 48 h of use.

4. Dilute sodium fluoride solution: Dilute 20 mL of a 3-percent sodium fluoride solution and 5 mL of 9 M sulfuric acid to 1 L with distilled water.

5. Ultraviolet-visible spectrophotometer, with 1-cm cell.

Procedure

1. Transfer 0.500-mL aliquots of solutions of the blank, the standards, and the samples into a series of 100-mL polyethylene beakers. Use a high-precision, 0.500-mL piston-type pipet to make these transfers. If the silica concentration is expected to be low, larger aliquots of sample solutions should be used. The following reagents then are added by a pipeting machine.

2. Add 25 mL of the dilute sodium fluoride solution to each beaker and let the mixture sit for 5 min.

3. Add 25 mL of the molybdate solution to each beaker, and let the mixture sit for 10 min.

4. Add 25 mL of the tartaric acid solution to each beaker.

5. Add 25 mL of the reducing solution to each beaker, and let the mixture sit for at least 45 min. The color of the complex is stable for at least 6 h.

6. Set the concentration scale of the spectrophotometer to zero for a wavelength of 640 nm with the blank solution in the absorption cell. Then, set the concentration scale with a solution having a known concentration of silica (standard silicate reference material), and determine directly the percent silica in each of the samples.


DETERMINATION OF ALUMINUM (AS Al2O3)

Reagents, Supplies, and Equipment

1. Complexing solution: To 880 mL of distilled water add 0.3 g of potassium ferricyanide {K3Fe(CN)6}, 40 mL of 10 percent hydroxylamine hydrochloride {NH2OH· HCl} solution, and 80 mL of calcium chloride solution. The calcium chloride solution is prepared by dissolving 14 g of calcium carbonate (CaCO3) in 30 mL of concentrated hydrochloric acid and diluting to 1 L. The complexing solution should be prepared on the same day that it is to be used.

2. Thioglycolic acid solution: Dilute 3 mL of the pure acid to 1 L with distilled water.

3. Buffer solution: Dissolve 80 g of sodium acetate {NaC2H3O3H2O} in 975 mL of distilled water and add 24 mL of glacial acetic acid.

4. Alizarin red-S stock solution, 0.10 percent: Dissolve 1.0 g of the pure dye in 1 L of distilled water and filter the resulting solution.

5. Alizarin red-S, 0.02 percent: Dilute 200 mL of the stock solution to 1 L with distilled water.

6. Ultraviolet-visible spectrophotometer, with 1-cm cell.

Procedure

1. Transfer 0.750 mL of the blank, standard solution, and each of the sample solutions into a series of 150-mL beakers using a high-precision piston pipet. Add the following reagents to each beaker by pipeting machine.

2. Add 25 mL of the complexing solution.

3. Add 25 mL of the thioglycolic acid solution, and let the mixture sit for 5 min.

4. Add 25 mL of the buffer solution, and let the mixture sit for 10 min.

5. Add 25 mL of the 0.02-percent alizarin red-S solution, and let the mixture sit for 45 to 75 min.

6. Set the concentration scale of the spectrophotometer to zero for a wavelength of 475 nm with the blank solution in the absorption cell. Then, set the concentration scale with a solution having a known concentration of alumina (standard silicate reference material), and determine directly the percent alumina in each of the samples.


DETERMINATION OF PHOSPHORUS (AS P2O5)

Reagents and Equipment

1. Ammonium molybdate stock solution: Dissolve 12.5 g of ammonium molybdate {(NH4)6Mo7O24· 4H2O} in 340 mL of distilled water. Then, add 160 mL of 9 M sulfuric acid.

2. Dilute ammonium molybdate solution: Dilute 20 mL of stock ammonium molybdate solution to 1 L.

3. Stannous chloride solution: Dissolve 0.600 g of stannous chloride {SnCl2H2O} in 25 mL of concentrated hydrochloric acid and dilute to 1 L with distilled water. This solution should be freshly prepared just prior to use.

4. Ultraviolet-visible spectrophotometer, with 1-cm cell.

Procedure

1. Transfer 10-mL aliquots of the blank, standards, and the sample solution to a series of 100-mL polyethylene beakers.

2. Add 25 mL of the dilute molybdate solution, and let the mixture sit for 10 min.

3. Add 25 mL of the stannous chloride solution. The developed color is stable for approximately 40 min.

4. Set the concentration scale of the spectrophotometer to zero for a wavelength of 640 nm with the blank solution in the absorption cell. Then, set the concentration scale with a solution having a known concentration of phosphorus pentoxide (standard silicate reference material), and determine directly the percent phosphorus pentoxide in each of the samples.


DETERMINATION OF TITANIUM (AS TiO2)

Reagents and Equipment

1. Disodium-1,2-dihydroxybenzene-3,5-disulfonate (Tiron): Dry reagent powder.

2. Buffer solution: Dissolve 80 g of ammonium acetate {NH4C2H3O2} and 30 mL of glacial acetic acid in 2 L of distilled water.

3. Sodium dithionate {Na2S2O4} (sometimes sold as sodium hydrosulfite): Dry reagent powder.

Procedure

1. Transfer 10-mL aliquots of the blank, each of the standards, and each of the sample solutions to a series of 150-mL beakers.

2. Add 125 mg of Tiron reagent powder to each beaker.

3. Add, by pipeting machine, 25 mL of the buffer solution to each beaker. Then, add 50 mL of distilled water to each beaker.

4. Add 10 to 20 mg of sodium dithionate to the blank and mix gently. Avoid vigorous mixing, which causes sulfur to precipitate. Then, set the concentration scale of the spectrophotometer to zero for a wavelength of 430 nm with the blank solution in the absorption cell.

5. Add sodium dithionate to one of the standard solutions and set the concentration scale of the spectrophotometer to the known concentration value (standard silicate reference material). Determine directly the percent TiO2 in each of the samples. Sodium dithionate may be added to six solutions at a time.


DETERMINATION OF CALCIUM, IRON, MAGNESIUM, MANGANESE, POTASSIUM, AND SODIUM

Reagents and Equipment

1. Lanthanum solution: Transfer 140 g of lanthanum oxide (La2O3, 99.997 percent pure) into a 2-L beaker. Slowly add 300 mL of concentrated hydrochloric acid, allowing time for the reaction to be completed after each addition of acid. Then, add 200 mL of distilled water. Each 4 mL of the final solution contains approximately 1 g of lanthanum.

2. Manganese stock solution: Transfer 0.3872 g of pure manganese metal into a glass beaker, add 20 mL of hot 8 M nitric acid, and gently boil the nitric acid for several minutes. After the resulting solution has cooled to room temperature, transfer the solution to a 500-mL volumetric flask and dilute to volume with distilled water. The concentration of MnO in this solution is 1,000 m g/mL.

3. Stock multiple-element standard solution: Transfer 0.8924 g of CaCO3, 0.9435 g of NaCl, 0.7915 g of KCl. and 2.4556 of FeSO4(NH4)2SO6H2O, (all reagent-grade purity) and 0.3045 g of magnesium ribbon to a 500-mL volumetric flask. (Magnesium ribbon generally is 99 percent magnesium; therefore, the weight of the ribbon includes a 1-percent correction.) Add 50 mL of distilled water and 10 mL of concentrated hydrochloric acid. Boil the dilute acid to dissolve all the constituents. After the solution cools to room temperature, add 50.0 mL of the manganese stock solution (1,000 m g/mL of MnO), dilute to volume with distilled water, and thoroughly mix the final solution. This solution contains the equivalent of 1.00 mg/mL each of Fe2O3, CaO, MgO, Na2O, and K2O, and 0.1 mg/mL of MnO.

4. Working standard solutions: To six 250-mL volumetric flasks, add 0, 6, 12, 18, 24, and 30 mL of the standard stock solution. Then, add 1.2 g of flux mixture, 5 mL of 8 M nitric acid, and approximately 200 mL of distilled water. Agitate the nitric acid solution to dissolve the flux mixture. Then, add distilled water to make the final volume 250 mL and make the solution homogeneous by vigorous mixing. These six solutions represent a blank and 3-, 6-, 9-, 12-, and 15-percent (equivalent in the sample) standard solutions. For MnO, the same six solutions represent a blank and 0.3-, 0.6-, 0.9-, 1.2-, and 1.5-percent standard solutions.

Procedure for Calcium, Iron, Magnesium, and Manganese

1. Transfer 0.750 mL of blank, sample, and standard solutions into small vials or beakers.

2. Dilute 8 mL of lanthanum solution with 200 mL of distilled water. Add 6.5 mL of this solution to all standards, samples, and blanks.

3. Calibrate the atomic absorption spectrometer by setting the concentration scale to zero for the recommended wavelength (table 16) while the blank solution is nebulized into the flame. Then, set the concentration scale with the 6-percent working standard, and verify this setting with solutions of silicate standards. Directly measure the concentrations of calcium, iron, magnesium, and manganese in each of the samples. Most available atomic absorption spectrometers are suitable for these measurements; the optimum operating conditions for each element usually are discussed in the manual provided with the spectrometer. Importantly, the individual measurements of concentration (or, absorbance) for a sample should be "bracketed" between those of standards because the instrumental responses are usually not linear.

Procedure for Potassium and Sodium

1. Transfer 0.200 mL of blank, sample solutions, working standards, and silicate standards into a small vial or beaker.

2. Dilute 1.2 mL of lanthanum solution with 200 mL of distilled water. Add 5.0 mL of this solution to the blank and to each of the standards and samples.

3. Calibrate the atomic absorption spectrometer using the concentration mode with the 6-percent working standard, and check appropriate silicate standards for known values. Measure directly the concentration of samples.


CONCLUSIONS

Methods based on AAS and spectrophotometry provide accurate determinations of 10 inorganic elements in coal ash. Although not as rapid as X-ray fluorescence (XRF) spectrometry, these methods furnish an approach to determining major oxides in coal ash that is both inexpensive and accurate (table 18). The agreement between our measurements and the NBS-certified concentrations for10 elements in NBS 1633a coal fly ash, demonstrated by data in table 18, is quite acceptable. Results from XRF spectrometry for SiO2, Al2O3, and Fe2O3 in ash sample number 1 and for SiO2 in sample number 7 (table 18) are outside the range covered by the standards used for calibration. Thus, extrapolations beyond this range could introduce error into these determinations.


REFERENCES

Bunting, W.E., 1944, The determination of soluble silica in very low concentrations: Industrial and Engineering Chemistry (analytical ed.), v. 16, p. 612-615.

Parker, C.A., and Goddard, A.P., 1950, The reaction of aluminum ions with alizarin-3-sulfonate, with particular reference to the effect of calcium ions: Analytica Chimica Acta, v. 4, no. 5, p. 517-536.

Shapiro, L., 1975, Rapid analysis of silicate, carbonate, and phosphate rocks: U.S. Geological Survey Bulletin 1401, revised ed., 76 p.

Yoe, J.H., and Armstrong, A.R., 1947, Colorimetric determination of titanium with disodium-1,2-dihydroxybenzene-3,5-disulfonate: Analytical Chemistry, v. 19, p. 100-102.


[ Contents | Coal Quality CD ]


Created by the EERT WWW Staff.

[an error occurred while processing this directive]