There are two leveling bottles, f and y, which are manipulated as follows: Before an analysis is begun, bring the solutions in pipettes a, b, c, and d exactly to the marks on the capillary glass tubes of the pipettes by use of the leveling bottle connected to the burette. Close the stopcocks, as shown in the sketch. To draw a sample of gas into the burette, raise the acidulated water in the latter to the point marked h on the capillary tube by raising the leveling bottle, opening the stopcock i to the air, and finally when the water is at the desired point by again closing i. Make a rubber connection between i and the sample container and place the latter in a vessel of water. Open the connections, including the stopcocks or pinchcocks on the sample container and the stopcock i, and lower the leveling bottle until the water in the burette is brought to the lowest division on the burette. Close stopcock i to the outside air and measure the gas sample in the burette at atmospheric pressure by holding the leveling bottle in such a position that the surface of the water in .it is on a line with the surface of the water

urelle. NOW I m ate arver the FIGURD 9.--Modified Orsat gassample is brought into the burette before analysis apparatus. For ex

in ordontomond taking the measurement in order to guard

planation, see text

planai against a small error which would result if time was not allowed for drainage of water down the sides of the burette.




After measurement pass the sample into pipette a in order to remove carbon dioxide. To do this, pass the sample back and forth several times between the pipette and the burette. Bring the solution in the pipette a exactly to the mark on the capillary stem and measure the sample again, allowing 1 minute for drainage. The loss in volume shows the amount of carbon dioxide originally present in the gas sample.


Now pass the sample into pipette b for the removal of oxygen. Manipulate in the same manner as in determining carbon dioxide.

Alkaline pyrogallate solution does not remove oxygen from a gas mixture as rapidly as caustic potash solution removes carbon dioxide; consequently longer contact must be allowed between this solution and the sample.


After the sample in the burette has been measured again and the contraction in volume due to the absorption of oxygen has been recorded, pass it into pipette c for the removal of carbon monoxide.

Cuprous chloride is not as satisfactory an absorbent for carbon monoxide as alkaline pyrogallate solution is for oxygen, especially when the solution is confined in an Orsat pipette and the absorption is conducted under the conditions mentioned.

Absorption of the gas in any pipette may usually be considered complete when no difference in volume of the sample is observed after successive passages of the gas mixture into the pipette.

Attention should be called to the fact that if carbon dioxide is not removed by caustic potash solution it will be absorbed by alkaline pyrogallate solution, and if this solution does not entirely remove the oxygen the latter will be absorbed to some extent by the cuprous chloride solution; hence it is absolutely necessary to remove completely each constituent in the particular pipette in the order provided for its removal before the gas mixture is passed into the next pipette.


After those constituents of a gas sample that are absorbable by the reagents are removed, dilute the sample with air, if it contains less than an explosive proportion of methane, by lowering the leveling bottle connected to the burette, opening the stopcock i to the outside air, and drawing some of the latter into the burette, so that the combined volume of air and residual gas shall be from 98 to 100 c. C. Pass the measured mixture into the slow-combustion pipette d and bring the platinum wire therein to a bright-yellow heat. Allow the wire to remain hot for 2 minutes. After pipette d has become cool to the hand, withdraw the sample into the burette and determine the contraction in volume due to the burning of the methane. Pass the sample into pipette a; the caustic potash solution absorbs the carbon dioxide produced by the burning of the methane. Return the sample to the burette and record the contraction in volume caused by this absorption. The carbon dioxide produced by the burning of the methane should be equal to one-half the contraction. If this relation does not hold good within 0.2 or 0.3 c. c., some other combustible gas is present in addition to methane or an error has been made in the manipulation of the apparatus.

Finally, pass the sample into alkaline pyrogallate pipette 6 to make sure that enough oxygen had been present for complete combustion of the methane.

If 100 c. c. of the sample is taken for analysis and the analysis is made according to the direction given above, the amount of each absorption in pipettes a, b, and c shows directly the percentage of carbon dioxide, oxygen, and carbon monoxide present. The carbon dioxide produced by the burning of the methane is also a direct measure of the percentage of the methane originally present in the sample. When less than 100 c. c. of the sample is taken for analysis, a simple calculation to a percentage basis shows the quantity of each constituent present. If the sample contains more than an explosive proportion of methane, handle the residual gas as follows after the absorbable constituents have been removed by pipettes a, b, and c:

Pass the residual gas directly into pipette d and bring the platinum wire therein to a white heat. Measure enough oxygen or air in burette e to burn the sample completely and pass it into pipette d at the rate of about 10 c. c. per minute. The methane burns as the methane and oxygen come in contact with the hot wire, and an explosion caused by the accumulation of methane and oxygen can not follow.

When combustion analyses are performed with the apparatus, place a wire screen around pipette d, so that, if an explosion shatters the glass, flying pieces will not cause injury. This precaution should be taken whenever combustion analyses are made in any gas-analysis apparatus.


Two or three trials may be necessary before mixtures which contain unknown and large proportions of methane can be handled to the best advantage. The apparatus is accurate to about 0.2 or 0.3 per cent, but good results can be obtained only by the strictest attention to details. Keep solutions fresh, see that the rubber tubing is in good condition and free from leaks, and lubricate all stopcocks well. Never draw the solutions in pipettes a, b, and c into the capillary train above the stopcocks; otherwise the analysis may be worthless.

Carbon monoxide is absent from normal mine air and air typifying mine conditions other than those caused by fires, explosion, or after blasting; consequently, pipette c need not be used in analyzing many samples of mine air.

Rubber bags are provided which are fastened to the reservoir bulbs of the pipettes a, b, and c. These bags should not be removed, and should be examined frequently in order to make sure that the

[blocks in formation]

FIGURE 10.-Copper oxide apparatus for complete analysis of gases. For explanation,

see text

rubber stoppers are tightly in place and that the bulbs do not leak at any point.

It will be found advisable occasionally to run some of the acidified water from burette e through the horizontal capillary train and out through i to neutralize any caustic potash that may have accidentally found its way from pipettes a and 6 into the train.


The Haldane apparatus previously described is intended for use in analyzing gas mixtures that do not contain large enough proportions of any constituents to cause the mercury to rise, during an analysis, to the ungraduated part of the burette where readings can not be made.

The authors have found that an apparatus of the type shown in Plate III can be used for the analysis of many normal or abnormal mine atmospheres. However, many samples containing large proportions of various constituents are analyzed at the laboratory of the Bureau of Mines with an apparatus of the type shown in Figure 9. Such mixtures include illuminating gas, oil gas, mine gas high in methane, natural gas, exhaust gases from gasoline locomotives, and special samples prepared during experimental work in the bureau's laboratory.


The apparatus shown in Figure 10 and Plate V was assembled to contain in one device for general use the desirable features found in several types of gas-analysis apparatus already on the market. These features follow:

1. Mercury in the burette and the combustion pipette.

2. A compensating device, h, allowing small changes in the temperature and pressure of the room during the course of the analysis to be disregarded.

3. A water jacket.
4. Slow-combustion inethod for making combustion analyses.

5. Copper oxide for the fractional combustion of carbon monoxide and hydrogen. This eliminates three pipettes and makes the apparatus more compact. Time is saved because there is no need to prepare cuprous chloride and colloidal palladium solutions, and the effect of solubility of hydrocarbons of high molecular weight is reduced to a minimum. Ethane and propane are soluble to some extent in cuprous chloride solutions. .

6. Adjustable brackets for supporting each pipette. It is easy to make close connections between the train and pipettes if the brackets


« ForrigeFortsett »