May
ANALYTICAL EDITION
IS, 1943
to complete extraction of the metallic complexes in the following optimum p H ranges: iron, 1.9 to 3.0; aluminum, 4.3 to 4.6; bismuth, 4.0 to 5.2; cobalt, 6.8 or above; nickel, 6.7 or above; and copper, 2.7 to a t least 7.0. At p H values below the lower limits, partial extraction to no extraction will occur. Chloroform solutions containing the equivalents of up to 20 mg. of these metals per liter (or 1mg. of the metal in the aqueous solution extracted) show little or no deviation from Beer's law a t the wave lengths given and thus lend themselves to use in accurate colorimetric determinations. Copper can be determined in the presence of a t least forty times as much cadmium or zinc by extraction a t p H 4.0 and transmittancy measurement a t 410 mp. i l n interesting correlation between p H values for hydrous oxide or hydroxide formation, quinolate precipitation, and quinolate extraction is given.
Literature Cited (1) Alexander, J. W., in "Summaries of Dortoral Dissertations, University of Wisconsin", Vol. 6, p. 205, Madison, Wis., University of Wisconsin Press, 1942. (2) Alten, F., Weiland, H., and Kurmies, B., Angew, Chem., 46, 697 (1933).
349
(3) Alten, F., Weiland, H., and Loofmann, H., Ibid., 46, 668 (1933). (4) Bayerle, V., Rec. trav. chim., 44, 514 (1925). (5) Berg, R., "Die chemische Analyse. XXXIV Band. Das oOxychinolin 'Oxin' ",Stuttgart, Ferdinand Enke Verlag, 1935. (6) Berg, R., Wtdker, W., and Skopp, E., Mikrochemie Emich Festschrzft, 1930, 18. (7) Britton, H. T. S., "Hydrogen Ions", p. 278, New York, D. Van Nostrand Co.. 1929. (8) Fleck, H. R., A h & 62, 378 (1937). (9) Fleck, H. R., and Ward, A. M., Ibid., 58, 388 (1933). (10) Folin, O., and Denis, W., J . Biol. Chem., 22, 305 (1915). (11) Geilmann, W., and Wrigge, F. W., 2.anorg. allgem. Chem., 209, 129 (1932). (12) Gerber, L., Claassen, R. I., and Boruff, C. S., IND.ENcI.CHEM, ANAL.ED., 14, 658 (1942). (13) G o b , H., Sca. Rep. TBhoku Imp. Univ., First Ser., 26, 391 (1937). (14) Hough, W. A., and Ficklen, J . B., J . Am. Chem. SOC.,52, 4752 (1930). (15) Lavollay, M. J., Bull. SOC. chzm. biol., 17, 432 (1935). (16) Moeller, T., IND.ENQ.CHEW,ANAL.ED., 15, 270 (1943). (17) Moeller, T.,J . Am. Chem. SOC.,63, 2625 (1941). (18) Moeller, T., unpublished data. (19) Sazerac, R., and Pouzerques, J., Compt. rend. SOC. biol., 109, 370 (1932). (20) Sheard, C., and States, M. N., J . Optical SOC.A m . , 31, 64 (1941). (21) Teitelbaum, M., 2. anal. Chem., 82, 366 (1930). (22) Yoshimatsu, S . , TBhoku J . Ezptl. Med., 14, 29 (1929); 15, 355 (1930).
An Improved Meter for the Measurement of Gas Flow Rates W. G. L@PLEBY AND W. H. -4VERY Houston Refinery Research Laboratory, Shell Oil Company, Incorporated, Houston, Texas
I
N T H E study of reaction kinetics in flow systems, accurate measurement of the rate of flow of the intake and exit
TUNGSTEN CONTACTS CONNECT TO RELAY AND VEEDER COUNTER
gases is essential. Flow-measuring instruments in common use are of two types: (A) meters that measure some function of the gas velocity-for example, the pressure drop across an orifice or the rate a t which heat is conducted away from a heated wire-and (B) instruments that involve displacement of a liquid by the gas and continuous or discontinuous measurement of the volume displaced in unit time. The familiar wet-test meter is a n example of this type. Both types of instruments, as commercially available, have disadvantages, particularly if flow rates of the order of 25 cc. per minute or smaller are to be measured. Type A requires a separate calibration curve for each kind of gas, hence is not suited for use with mixtures of unknown composition, such as the gaseous product of a thermal reaction. Type B requires estimation of and correction for the effect of the vapor pressure of the displaced liquid and for the solubility of some components of the mixture in the liquid. Meters of this type are-therefore unsuitable for the measurement of intake rates. The flowmeter shown diagrammatically in Figure 1 is an improved form of a meter described by iMarsh ( I ) for measurement of very slow flow rates. The new type of meter is free from the disadvantages of the usual instruments in common use. The operation is clearly shown by the diagram. Referring to the side view, gas enters at A , passes down the vertical tube, and forces a drop of mercury, D,to the right in the horizontal tube. Stopcocks SI and S1 are open; SS is closed. When the drop reaches the end of the tube at E,the gas trap ed behind the drop is discharged. However, as soon as the Z o p
ABOUT ?A+'
TOP VIEW
6MM. TUBING
~
13" ODJJE
1
iil
6.75"
5 TURNS 2MM.TUBING
I.'"" u V.Y.
SIDE VIEW
FIGURE 1. DIAGRAM OF FLOWMETER
350
INDUSTRIAL AND ENGINEERING CHEMISTRY
begins to flow into the 13-mm. bulb a t E, the mercury level is pushed up at C, so that a new drop is forced into the horizontal tube, trapping a new segment of the gas stream. This drop then moves to the right and the cycle is repeated. Thus, each cycle represents the flow of a definite volume of gas and the time between cycles is a measure of the flow rate. For low flow rates the time required for movement of the drop between two fixed points may be a more convenient measure of the rate.
Vol. 1.5, No. 5
The surface of the tubing through which the gas and mercury droplets flow should be free of any oil or water film or dust particles, and the mercury should be pure; otherwise there will be small fluctuations in the rate of gas flow which may give pressure variations as high as 1 or 2 mm. of mercury. This effect can be eliminated in any case, however, by having a surge chamber in the line following the meter. Since the volume displacement per cycle does not depend on the pressure or temperature of the gas, the instrument may be used at subatmospheric pressures. During any particular run, however, the temperature and pressure should have constant values; otherwise i t would be difficult to convert the gas volumes to standard conditions. The meter could also be adapted for use at superatmospheric pressures by using metal parts or suitable plastic tubing and connections.
Acknowledgment The authors are greatly indebted to Harold Gershinowitz for his interest and advice.
Literature Cited (1) Marsh, A.
E. L.,Trans. Faraday SOC.,36, 626 (1940).
Forceps with Platinum-Covered Tips CYCLES PER MINUTE
FIGURE 2 Figure 2 shows a calibration curve obtained with hydrogen and with nitrogen, when gas volumes were measured by collection over water in suitable burets. The points for the gases are seen to fall accurately on the same line. The use of mercury as a displacement fluid eliminates the difficulties caused by use of water or organic liquids in the usual wet-test meters. Some details of the construction and operation of the flowmeter need further discussion. The glass splrals permit small adjustments in the tilt of the tube, which control? the aize of the drop. Since there is a lag between the time at which the drop begins to flow into the bulb at E and the time a t which the mercury rises at C, the drop must be of sufEcient length to ensure that mercury will enter the horizontal tube a t C before the drop a t the right is entirely discharged from the tube. If this condition is not met, a jet of gas will flow through the tube during a part of each cycle. The amount of gas will be a function of the gas viscosity and the calibration curve will then be found to depend on the kind qf gas. The re uirement that there must be a drop of mercury in the horizontal tube a t every part of the cycle suggests the use of a tube of relatively large diameter and as short as possible for the mercury piston, in order to reduce the time lag. However, in such a tube the mercury level tends to oscillate for some time after a drop is discharged a t E, thereby producing a series of dro s at the left instead of one. The constriction in the vertical tug, a t F reduces these oscillations; the long zigzag horizontal tube, shown in the top view, ensures complete damping of the oscillations of one cycle before the second begins and minimizes possible errors from fluctuations in the size of the mercury drops. I n the authors' instrument 2 dro s are formed a t the beginning of every cycle when the flow rate gecomes greater than 15 cc. per minute and as this rate is approached there will be 1 drop one cycle and 2 the next. However, the total volume of mercury remains very nearly constant, as the calibration curve proves. Above 35 cc. per minute 3 drops are formed. The tungsten contacts shown in the top view actuate a counter through a relay, so that the instrument may be used to record total flow as well as to measure the rate.
EARL W. BALIS AND HERMAN A. LIEBHAFSKY Research Laboratory, General Electric C o . , Schenectady, N. Y.
T
HE usual platinum-tipped forceps commercially avail-
able have each platinum tip fastened to nickel-plated steel by means of two rivets in a simple scarf joint. During the past five years, three or four of these expensive tools have failed here, always owing to corrosion or oxidation where the two metals join. Even moderate attack of the steel soon renders the forceps unfit for use because the platinum alone is too weak t o withstand the flexing that occurs at the rivet holes. Forceps that are stronger, cheaper, and more durable than those just described have been made in this laboratory by covering the tips of ordinary steel forceps with 5-mil platinum foil. The tips of the steel forceps were filed t o the shape desired and a piece of foil (0.8 X 2.5 cm.) was fitted to each, so that the seam would fall outside. A close fit was obtained by slitting the lower edge of the foil so that it could be folded around the end of the tip; excess foil was then carefully trimmed away. Spot-welding t o form a practically continuous seam completed the assembly. The welding operation is simple with the proper equipment. I n this case, a low inertia bench welder with synchronous thyratron control was used under the following conditions: 350 amperes for 6 cycles, or 0.1 second; electrode pressure, 20 pounds; flat copper electrodes of diameters a,'/16 inch (lower) and 3/82 inch (upper). The forceps thus constructed have been used constantly during the last six months and have given satisfactory service without deteriorating noticeably. There is no reason why the tips of a wide variety of tongs and forceps could not be covered with platinum foil to great advantage. The implements to be covered should preferably be made of stainless alloys.
