conditions, with small percentages of methane present, one observer may see a cap distinctly, although another with less acute vision may not. A series of analyses of mixtures examined by safety lamps will show clearly what may be expected from the lamps.19 Since the fact has been recognized that percentages of methane much below that required to form an explosive mixture with air may contribute to disastrous dust explosions, urgent need has been felt for the detection of smaller quantities of methane than are generally

detected by the use of a safety lamp.




The apparatus shown in Figure 6 includes a 21-c. c. burette similar to that used on Haldane's apparatus; the bulb has a capacity of 15 c. c. and the stem a capacity of 6 c. c. The stem is graduated in hundredths of

cubic centimeter. The burette has an attachment that compensates for slight changes in room temperature during an analysis and facilitates the taking of readings. The slow-combus

tion pipette g is similar in FIGURE 6.—Laboratory apparatus for exact deter- construction to the pipette mination of methane. For explanation, see text

shown in Plate III (p. 18). At the laboratory the sample container is set up as shown. To make an analysis proceed as follows: Draw about 10 c.c. of the sample into the burette and discharge it through the three-way cock a, sweeping the air out of the capillary tubing; then draw about 21 c. c. of the sample into the burette and measure it against the pressure in the compensating tube c by means of the liquid in the manometer b. With this apparatus use a mixture of one part of glycerin and three parts of water in the manometer. Such a mixture responds readily to slight movement of the mercury in the burette. Because of its comparatively low density, water alone is unsatisfactory when the burette contains mercury. Although the compensating device is similar in principle to that shown attached to the burette in Plate III, it is operated differently. Put enough glycerin solution in manometer b (fig. 6) to bring the surface of the liquid in both parts of the U tube approximately on a line with the scratch mark shown at d. After the sample has been drawn into the burette turn the burette stopcock e so that communication is made between the burette and the manometer. Slightly raise or lower the leveling bulb f; then movement of the mercury brings the glycerin solution in the U tube exactly to the mark d. Next take the burette reading.

19 Paul, J. W., Ilsley, L. C., and Gleim, E. J. Work cited.

After the gas has been measured in the burette pass it into the slow-combustion pipette and heat the platinum wire. After combustion allow the pipette to cool and return the gas to the burette in order to measure the contraction in volume that results from the burning of the methane. This contraction, divided by 2 and calculated to a percentage basis, represents the proportion of methane in the mine air.

As stated above, methane is the combustible gas commonly found in mines. Traces of gaseous paraffins higher than methane have not been positively identified in normal mine air examined by the authors.

As the sample is drawn through the horizontal capillary tubes leave a small amount therein, but after measurement pass it into the combustion pipette and do not withdraw until combustion is complete. A small amount of gas in excess of that measured in the burette and equal to the amount introduced into the capillary train will therefore remain unburned. This method avoids the small error that would follow if the sample was passed back and forth between the burette and the combustion pipette during combustion. The error would, however, ordinarily be small enough to neglect.

An analysis does not take longer than 10 minutes, and determinations of the same sample agree within 0.01 or 0.02 per cent.



Figure 7 shows a portable apparatus for the determination of methane in mine air. The burette has a capacity of 50 c. c.—the bulb 45 c. C., and the stem 5 c. C.—and is graduated in five-hundredths of a cubic centimeter. To simplify the apparatus, water slightly acidified with sulphuric acid is used in the burette and the combustion pipette. Determinations of methane may be made with an accuracy of 0.1 per cent when the analyses are properly made, and the burette is drained carefully each time before readings are taken.

To make an analysis, draw the sample into burette a (fig. 7) through c, and measure its volume at atmospheric pressure by hold

[ocr errors]

ing the surface of the water in the level bottle of the burette on a line with the surface of the water in the burette. Pass the sample

into the combustion pipette b, and heat the platinum wire a bright yellow. After about 2 minutes the methane in the sample is completely oxidized. Let the pipette cool and transfer the gas to the burette, where the contraction in volume due to the burning of methane is measured in the same way as the sample was originally measured.

The contraction in volume, divided by 2, b

gives the amount of methane originally present in the sample.

Many coal mines normally contain less methane in the workings than a device of the degree of accuracy of this can measure, but, on the other hand, there are mines where the amount of methane, even in the main returns, never falls below 0.2 to 0.5 per cent. The value of the apparatus lies in its ability to detect proportions of methane that are much below those detectable by the safety lamp and are far below the explosive proportion. Because of its ease of manipulation and simplicity the

apparatus may appeal to the officials of FIGURE 7.-Portable appara

mines where the employment of a chemist tus for determination of is not feasible. methane in mine air. Capacity of burette, 50 c. c.;

The following methane determinations capacity of bulb of burette, were obtained with the apparatus by a 45 c. c.; capacity of stem of burette, 5 C. c.; stem

person who was not a chemist and who graduated in 0.05 c. c. For had never before made analyses of gases or explanation, see text

other substances. The sample of mine air actually contained 0.33 per cent of methane, as shown by a more exact laboratory analysis.

Results of methane determinations of a mine-air sample

Sample CH4, per Sample CH4, per
No. cent No. cent


. 35

11 12 13 14 15 16 17 18 19 20


30 39 . 30

36 . 40 .33 .33

4 5 6 7 8 9 10

. 35 . 26 . 30

. 33

24 .30 40

26 . 30

. 26


A storage battery can be used for heating the No. 30 platinum wire to incandescence. A battery that will produce an electric current of 4 amperes at 5 volts is required. Dry cells such as are used with house doorbells may be utilized, but they are not as suitable because they are not intended for continuous service of 2 minutes at a time. Some electric cap-lamp batteries now on the market are suitable. The most satisfactory arrangement will result when a lamp bank or other form of resistance that can carry 4 or 5 amperes is connected to an electric lighting circuit and set up at the headquarters selected for the work; the samples are carried or mailed to this place for analysis. When the air in a large number of mines is being sampled for methane, a resistance bank can be installed at each mine if electricity is used there, and the apparatus for gas analysis can be carried from place to place; a series of samples can then be collected and analyses made at the particular mine where the samples were gathered.

[ocr errors]



FIGURE 8.-Portable appara

tus for determination of carFigure 8 shows a portable apparatus for bon dioxide and methane

in mine air. Water to be methane determinations. It is similar to

used in burette and comthat in Figure 7, except that a caustic pot

bustion pipette. Capacity

of burette, 50 c. c.; capacash pipette, b, is included for the determina- ity of bulb of burette, 45 c. tion of carbon dioxide. Water is used in the c. ; capacity of stem of bur

ette, 5 c. c.; stem graduated burette and the combustion pipette. The


50 C. c. For explana

tion, see text apparatus is operated like that shown in Figure 7, except that the sample after measurement in the burette and before burning is passed into the potassium hydroxide pipette b for the removal of carbon dioxide. The two determinations can be made with an accuracy of 0.1 per cent.

Samples that are to be analyzed for their carbon dioxide content should be collected by air displacement by one of the methods already described. The use of water in the burette of the apparatus shown in Figure 8 does not, however, militate against correct determination of the carbon dioxide, at least within the experimental error of reading the burette. To prove this fact some experimental work was performed in which definite quantities of carbon dioxide were added to air in bottles. The samples were then analyzed, first by means of the apparatus, shown in Plate III, in which mercury is used in the burette, and then by means of the apparatus, shown in Figure 8, in which water is used. Two or more analyses were made of each sample. The results appear in the following table. The percentages of carbon dioxide found with the apparatus shown in Plate III are given in column A; those found with the apparatus shown in Figure 8 appear in column B:

[blocks in formation]



Figure 9 shows a modified Orsat apparatus that is simple in operation and can be used where great accuracy is not desired.20 With this apparatus carbon dioxide and carbon monoxide can be determined to within about 0.2 per cent, and methane and hydrogen can be determined almost as accurately by calculation from the combustion data.


The apparatus consists essentially of the burette e and the four pipettes a, b, c, and d. The burette has a capacity of 100 c. c. and is graduated in 0.2 c. c. The pipettes contain: а, Potassium hydroxide solution for the removal of carbon dioxide; b, alkaline pyrogallate solution for the removal of oxygen; and c, ammoniacal or acid cuprous chloride solution for the removal of carbon monoxide. At d is the slow-combustion pipette for burning the methane; it contains distilled water slightly acidified with sulphuric acid. The pipettes a, b, and c are provided with glass rods (not shown in sketch) to increase the absorption surface of the reagents. The platinum wire inside the combustion pipette is No. 30 (B. & S. gauge), is supported by two pieces of glass tubing, and is sealed to copper wires that pass through the glass tubes.

20 Burrell, G. A., and Seibert, F. M., Gas analysis as an aid in fighting mine fires : Tech. Paper 13, Bureau of Mines, 1912, pp. 14–15.

« ForrigeFortsett »