without magnification. In reading volumes with the Schellbach type the point where the magnified and nonmagnified portions meet is compared with the calibration marks.

With burettes having numbers increasing from top to bottom (fig. 13, A), the readings of whole and part units are taken in the same order; in other words, the gas space between the 0 mark at the top of the burette and the liquid is measured. With burettes numbered from bottom to top (fig. 13, B), the reverse order is followed; that is, the space between the 0 mark at the bottom and the top of the liquid is measured. A few specific cases will serve to illustrate the manner of reading volumes. In A, each large subdivision (marks extending entirely around the burette) equals 1 c. c. and each small division 0.2 c. c. It follows that meniscus a equals 88 whole units and 1 part unit with a value of 0.2. Consequently, the reading for a is 88.2 c. c.; likewise, that for b is 88.4. In c the meniscus appears in one of the odd or unnumbered whole units; but from the even-numbered unit above (which is 90) it is known that the unnumbered whole unit is 91; and this, with one small division having a value of 0.2, makes the reading 91.2. In d the bottom of the meniscus does not fall exactly on any line but comes midway between, so that the reading is 92 plus one-half of a division whose value is 0.2, or 92.1. The same is true of e; the reading is 92 whole units, 3 full small units, and one-half small unit, the sum of which is 92.7. In Figure 13, B, when the burette reads from bottom to top, ƒ equals 8 whole units from the bottom and 2 small ones, or 8.4. Similarly g, which falls in an unnumbered unit, equals 11.4.

Burettes that are numbered from 100 at the top to 0 at the bottom are designed to give the net volume lost after the gas is subjected to a given reagent by subtracting the total volume as measured before absorption in each successive step in the procedure from the volume after absorption; burettes calibrated from 0 at the top to 100 at the bottom (fig. 13, A) give the net volume lost after absorption by subtracting the volume after absorption from the volume before absorption. For most people this latter type appears the easiest to read as well as the easiest from which to keep records and calculate results, as it permits setting all readings in a vertical column and subtracting one reading from that above to obtain the data for a specific constituent.

PROCEDURE FOR MAKING BURETTE READING

After the gas has been drawn into the burette from either the sample container or one of the pipettes and the stopcocks have been closed, allow one minute for the water to drain down the sides of the burette. Then, by holding leveling bottle ƒ close to burette e (the bottle may be steadied, if necessary, by holding it against the side of the case), bring the level of the water in ƒ to the same height as that in e. Hold

in this position and, with the eye about 15 inches from the burette and at the height of the liquid in f, sight horizontally across the burette and find the graduation mark that coincides with the bottom of the meniscus, at the same time noting the value of the mark as so many units and parts of units. (See fig. 14.) The position of the meniscus should be estimated to the nearest mark or half-division mark, as c or d, h or j, Figure 13. The meniscus and graduation marks can be illuminated by holding an electric light (extension cord), flash light, or cap lamp behind them. The ordinary clear-glass electric-light bulb is too bright for this purpose unless it is screened.

FIGURE 14.-Position of leveling bottle and line of sight in making burette reading

A correct burette reading depends entirely upon (1) having the line of the water in f the same as that in e to bring the gas to atmospheric pressure and (2) having the eye at the same level to make the line of sight in the same horizontal plane as the water levels. Record the volume immediately after making a burette reading.

CONTROL OF TEMPERATURE AND PRESSURE DURING MEASUREMENT

The volume of liquids and solids, such as water, steel, etc., changes comparatively little with ordinary changes in pressure and temperature, and these factors are not usually considered in ordinary measurement. Gases, however, change markedly with changes in temperature

and pressure, and as the ordinary procedure for analyzing them is based on successive comparative measurements of contraction in volume caused by absorption or combustion it is necessary to know or to control the temperature or pressure at which each measurement is made. Otherwise, one could not ascertain whether the differences in readings were due to removal of a constituent or to a change in the conditions under which the gas was measured.

The temperature and pressure are controlled either by measuring them at the time of making each successive reading and transforming each volume by calculation to one standard condition or by devising means for keeping the conditions the same or practically the same throughout the period of analysis. The latter is the usual method employed in gas analysis. The temperature and pressure conditions chosen are those that happen to prevail at the place where the analysis is made.

With the apparatus shown in Figures 10, 11, and 12, the means employed for controlling the temperature over the period of analysis is to inclose the burette in a water jacket. Unless there is a marked change in the room temperature, the temperature of this rather large volume of water will remain practically constant throughout the period of analysis. Also, the gas in the burette is thus protected from the sudden effect of hot or cold drafts of short duration. The water jacket, however, is only intended for ordinary room changes in temperature, and where marked changes do not occur and where an ordinarily comfortable temperature exists a gas apparatus should always be housed. Also, before attempting to make analyses one should allow a gas apparatus to stand in the room long enough for the water in the leveling bottle and water jacket to assume room temperature.

Changes in pressure are sometimes taken care of by compensating devices independent of atmospheric pressure. These are used with the Haldane and laboratory Orsat" types, but with the portable Orsat shown in Figures 10, 11, and 12 the pressure is controlled by bringing the gas within the burette to the existing atmospheric pressure. The barometric-pressure changes over the period of analysis (say, 30 minutes) will seldom be great enough to cause a significant

error.

To make the pressure of the gas in the burette the same as that outside before each measurement is taken, hold leveling bottle f close to burette e and raise or lower f until the liquid in the bottle is the same as that in the burette. This virtually makes a U gage, one arm of which is the burette and the other arm of which is the hose and leveling bottle. When the levels are the same, the pressures on each are the same.

See footnote 6, p. 31.

PREPARATION OF SOLUTIONS

The absorbing solutions are most conveniently prepared in a chemical laboratory where facilities are available; then they are shipped to the field. However, in emergency cases when there is no time for ordering them from a distant laboratory it may be necessary to prepare them in the field. The chemicals used in preparing the solutions for carbon dioxide and oxygen are common and can be obtained from a drug store or photographic supply house. Those required for making the cuprous chloride can be obtained from a university or commercial chemical laboratory.

CARBON DIOXIDE

Potassium hydroxide (KOH) or sodium hydroxide (NaOH) is generally used for absorbing carbon dioxide. As a rule these chemicals are solids marketed in stick form and contained in stoppered bottles to exclude moisture and air. The grade of hydroxide known as electrolytic and not the product purified by alcohol is recommended for gas absorbents, not because the electrolytic grade is better for absorbing carbon dioxide but because it should be used in preparing alkaline pyrogallate solution for oxygen, and because by using the electrolytic for both it is unnecessary to provide two grades in the laboratory.

A considerable amount of heat is produced when potassium or sodium hydroxide is dissolved in water, and if dissolved rapidly the water may even boil. For this reason the solution should be made by dissolving one stick or only a few sticks at a time and by cooling the bottle or flask (an ordinary table dish is suitable if other apparatus is not available) by letting it stand or by holding it under a stream of water, otherwise the heat might cause it to crack. The vessel should never be stoppered while the caustic is being dissolved, but it should be stirred or the contents rotated gently. After the solution has been prepared it should be allowed to cool. A small amount of sediment will settle on standing, and the clear liquid can be poured or decanted off for filling a pipette or can be transferred to storage bottles. Magnesium citrate bottles are convenient containers.

Potassium hydroxide solution.-Dissolve 300 grams in 1,000 c. c. of

water.

Sodium hydroxide solution.-Dissolve 200 grams in 1,000 c. c. of

water.

Distilled water is recommended for preparing absorbing solutions and can usually be obtained from a drug store or garage; if unobtainable, however, rain water or ordinary drinking water may be used.

In preparing either solution a greater or less amount can be made as long as the proportion of each chemical is kept the same; for

example, for 200 c. c. of water 60 grams of potassium hydroxide should be added. Potassium or sodium hydroxide, either in stick form or in solution, is poisonous and corrosive to the skin and clothes. The action is similar to that of ordinary lye.

OXYGEN

Alkaline pyrogallate is used for absorbing oxygen. It consists of a mixture of potassium hydroxide solution and pyrogallic acid (ordinary "pyro," as used in photography) solution. These two solutions are prepared separately and then mixed. In making the potassium hyroxide solution the same precautions given under carbon dioxide absorbents should be taken regarding skin burns, effect on clothing, and heating of solution. No precautions need to be taken with the pyrogallic acid except to avert stains on the skin and clothing.

Potassium pyrogallate solution.-Dissolve 50 grams of pyrogallic acid in 150 c. c. of water. In a separate bottle, flask, or ordinary dish dissolve 1,200 grams of potassium hydroxide (electrolytic grade) in 800 c. c. of water. Smaller amounts of the solutions, even just enough to fill a pipette, may be made by proportionately decreasing the amounts of each chemical; that is, divide each by 2, 5, 10, etc., as desired. After cooling the potassium hydroxide solution add the pyrogallic acid solution, mix, and immediately place in a tightly stoppered bottle to avoid contact with the oxygen of the air. A better and more convenient procedure is to place 250 c. c. of the potassium hydroxide solution in each of a suitable number of bottles of approximately 325 c. c. capacity, such as magnesium citrate bottles. Add 45 c. c. of the pyrogallic acid solution to each and immediately close. These small containers are better than large bottles, as only a small portion of solution is subjected to air if opened, whereas if stored in a large container all of the solution would be exposed to air each time a pipette is filled. Of course, large bottles having siphons or drains for removing the solution would prevent this, but they are not considered satisfactory for field use.

Alkaline pyrogallate solution is poisonous and corrosive to the hands and clothing.

CARBON MONOXIDE

Cuprous chloride solution.-Either ammoniacal cuprous chloride (blue solution) or acid cuprous chloride (straw colored) solution is suitable for absorbing carbon monoxide. However, the acid solution is recommended, as the ammonia vapors given off from the ammoniacal preparation may cause error by making the moisture or drops of water in the manifolds alkaline, thereby prematurely absorbing CO2 from the gases.

Acid cuprous chloride solution is prepared by either the "copper reduction" or "stannous chloride reduction" method.

31650°-29-4