APPLICATION OF THE DFS-10 QUANTOMETER FOR THE ANALYSIS OF ASH, SLAG, AND BOILER DEPOSITS

A. V. LEVKOVSKII, I. N. BALYCHEV, A. D. GLAZUNOVA

(Presented by the scientific seminar of the Department of Boiler Construction and Boiler Installations)

Chemical analysis of the mineral constituents of power-generating coals under laboratory conditions has until now been carried out by classical methods of silicate analysis. However, methods of chemical analysis of materials of silicate origin—for example, ash from power-generating coals—have a number of disadvantages, the chief of which are long duration, labor intensity, and the use of large quantities of various expensive reagents. At the same time, in recent years instrumental methods for the analysis of ores, metals, and rocks have acquired ever greater importance; among them a prominent place is occupied by the method of spectral analysis, carried out with the use of such instruments as quantometers, styloscopes, spectrometers, etc. The advantages of instrumental methods of analysis, and in particular of spectral analysis, are the wide range of elements determined, the speed of analysis, and the insignificant consumption of the material being analyzed, which, when studying coal-combustion processes with a small volume of selected samples, is of considerable importance. Taking these circumstances into account, we used the DFS-10 quantometer for the analysis of solid products of coal combustion.

The DFS-10 quantometer is a spectral instrument with photoelectric recording of the spectrum, intended for express and grading quantitative analysis of metals and alloys [1].

At present a number of procedures have been developed for the analysis of rocks and minerals on the DFS-10 instrument [2]. However, it had not previously been used for the analysis of ash, slag, and boiler deposits for Si, Ca, Mg, Fe, Al, Mn, and Ti. The chemical composition of combustion products of fossil coals has a wide range of contents of individual elements, from 0 to 80–90%. This considerably complicated the work of selecting suitable analytical lines necessary for carrying out analysis covering such a range. For the analysis we selected the analytical lines given in Table 1.

Every day before the start of work, the positions of the exit slits were adjusted. For these purposes we used the following reference lines:

on the second carriage, the Fe 4404.8 Å line;
on the third carriage, the Fe 3466 Å line;
on the fourth carriage, the Si 2514 Å line.

The process of preparing a sample of ash, slag, and boiler deposits for analysis is as follows. The initial sample is ground until it passes through a 200-mesh sieve in an agate mortar; slags and deposits are first crushed. The sample is then calcined at a temperature of 1000°C for one hour and, after cooling, is diluted with buffer in a ratio of 1 : 9, i.e., 9 parts buffer and 1 part sample. An internal standard was introduced into the buffer. As the internal standard, copper was used

Table 1

Wavelengths in Å

(the comparison line Cu 3247,5 Å), which for coals is a relatively rare element and occurs in insignificant amounts. The buffer is prepared from electrode material with the introduction into it of 10% copper oxide CuO (analytical grade); both materials are ground until they pass through a 200-mesh sieve. The diluted sample is thoroughly ground with alcohol and dried at a temperature of 100–105°C. The samples prepared for analysis are packed tightly into the electrode crater immediately before burning; prior to this they are stored in a desiccator, in cellophane packets.

Quantitative determination of the elements in the sample requires calibration of the instrument and is carried out from calibration graphs, which are presented in Fig. 1. The graphs are constructed in the coordinates \(\lg C, 4n\), where

\(C\) — concentration of the element in the sample,
\(n\) — microammeter reading.

To construct the calibration graphs, we prepared standards in a similar manner from a mixture of oxides. The chemical composition of the standards is given in Table 2.

Table 2

| Standard No. | \multicolumn{8}{c}{Chemical composition, %} |
|---:|---:|---:|---:|---:|---:|---:|---:|---:|
| | SiO₂ | CaO | Fe₂O₃ | Al₂O₃ | TiO₂ | MgO | BaO | MnO₂ |
| 1 | 10 | 60 | 2 | 20 | 0,01 | 6,5 | 1,4 | 0,09 |
| 2 | 10 | 65 | 12 | 6 | 1,45 | 5,47 | 0,03 | 0,05 |
| 4 | 15 | 37,5 | 30 | 2 | 1,0 | 12 | 0,5 | 2 |
| 10 | 40 | 45 | 7 | 4 | 0,75 | 1,98 | 0,02 | 1,25 |
| 13 | 50 | 40 | 4 | 3 | 1,4 | 0,95 | 0,05 | 0,6 |
| 18 | 60 | 15 | 10 | 4 | 0,02 | 9 | 1 | 0,98 |
| 26 | 8,0 | 75 | 2,5 | 8 | 0,1 | 1,5 | 2 | 2,9 |
| 27 | 80 | 2,5 | 1 | 1 | 0,15 | 15 | 0,3 | 0,05 |
| 31 | 30 | 40 | 10 | 10 | 1,2 | 7,9 | 0,6 | 0,3 |
| 34 | 60 | 30 | 3 | 3 | 1,2 | 1,8 | 0,2 | 0,8 |

Comparison of the results of ash analysis obtained on the quantometer with the data of chemical analysis confirms the correctness

of the selected standards (Table 3) and shows sufficiently satisfactory agreement of the results.

Fig. 1. Calibration graphs*

Table 3

* For ease of comparison, the concentrations of the elements are given in the form of oxides.

The choice of the arc-excitation mode (type of current, current strength, voltage, size of the analytical gap, shape of electrode sharpening) substantially affects the accuracy and sensitivity of the analysis [1, 3, 4]. As a result of the studies we carried out, the following mode and analysis conditions were selected.

Operating mode of the GEU-1 generator: arc mode, alternating current, current strength 20 A, voltage 220 V, ignition phase \(90^\circ\), analytical gap 1.5 mm, preburn 5 sec, exposure 45 sec. Carbon electrodes of grade S-2, diameter 6 mm, were used in the work.

The root-mean-square errors of reproducibility, which make it possible to judge the accuracy of the method, are given in Table 4.

Table 4

Root-mean-square errors for each channel

* average of 60 determinations.

One year’s experience in using the DFS-10 quantometer makes it possible to draw the following conclusions:

1. The instrument is convenient to operate. The time and labor required to carry out an analysis are significantly reduced. Thus, 3 min are spent on determining 7 elements in a sample, whereas determining the same elements by a chemical method requires about 5 days.

2. The minimum required amount of sample for analysis is 10–25 mg.

3. The use of the DFS-10 quantometer for the analysis of ash, slag, and boiler deposits gives quite satisfactory results. The average error of determination does not exceed 5%.

4. With the use of appropriate standards, analysis of certain rocks can be carried out.

REFERENCES

1. Photoelectric Methods of Spectral Analysis. Oborongiz Publishing House, Moscow, 1961.
2. Spectral Analysis in Geology and Geochemistry. Nauka Publishing House, Moscow, 1967.
3. A. N. Zaidel, Fundamentals of Spectral Analysis. Nauka Publishing House, Moscow, 1965.
4. A. B. Shlaevich, Methods for Evaluating the Accuracy of Spectral Analysis. Nauka Publishing House, Moscow, 1964.