Improved Manostat and Manometer - American Chemical Society

0, 0.2. Be(OH)2 (4.17/ml.) 2. 1. 2.2. 0.2. Be(OH)¡ (1.67/ml.) 2. 1. 2.2. 0.2. Be(OH), (4.1 /ml.) 2. 2. 4. 0.5. Be(OH)! (1.67/ml.) 2. 2. 5. 1.2. Prote...
2 downloads 0 Views 272KB Size
838

ANALYTICAL CHEMISTRY Table 1. Ultrafiltration of Colloids

Ultrafiltrate Colloid concentration yo of initial R.p.m. Hour8 Ml. concn. 2.5 1.0, 0 . 3 2.5 0.0.2 0.2 2.2 0.2 2.2 0.5 5 1.2 2.5 0,o 2.5 0:o Protein (Difco serum) 2.5 0: 0 2 2 1 0,1 Inulin (3.25 mg. %) 1 3 2.2 68,68,92 a Tracer method of analysis used for all Be(0H)n suspensions. Substance Filtered4

Centrifugation Rate/ 1000 Time

Volumo

Plasma was prepared by centrifu ing oxalated beef blood obtained at a local slaughterhouse. T i e 10% Difco serum suspensions were prepared from Difco Bacto dried beef blood serum, All other chemicals used were C.P. grade. Beryllium was determined either by the colorimetric method ( 6 ) or the tracer technique ( 1 ) . Protein was determined by visual comparison of the turbidity obtained upon the addition of 3 drops of 5% sulfosalicylic acid to the test solution and to known dilutions of plasma. Inulin was determined by the method of Kruhoffer (9).

of inulin were required thBt the probable error of the analyses was rather large. Accordingly, the values reported for inulin are of qualitative significanceonly. When solutions of beryllium, in concentrations ranging from a trace to 4.64 micrograms per ml. in 0.001 M hydrochloric acid, were centrifuged a t 1000 r.p.m. for 6 hours, the concentration of beryllium (94 * 1%) in the filtrate was equal to that remaining in the cellophane bag. The consistently low recovery probably is due to a slight adsorption on the cellophane bag. DISCUSSION

The all-glass construction of the new apparatus represents a real advantage in eliminating problems of adsorption and decontamination] permitting accurate studies of extremely dilute solutions. Furthermore, the temperature a t which the filtration is carried out may be conveniently controlled. In spite of a consistently negative absolute error of about 5% (with low concentrations of an adsorbable solute) the ultrafiltration technique described gave very reproducible results. The accuracy seems adequate to permit quantitative studies. Under the experimental conditions employed, the membrane was permeable to dissolved beryllium and a low-molecular weight colloid, inulin, but impermeable to protein and precipitated beryllium hydroxide.

RESULTS

LITERATURE CITED

The force and rate of ultrafiltration are, of course, related to the speed and radius of centrifugation and also are dependent upon the solution being centrifuged. Thus, a t 1000 r.p.m., 2.5 ml. of filtrate were obtained from aqueous solution in 3 hours and from blood plasma in 6 hours. At 2000 r.p.m., the time required for the collection of 2.5 ml. of ultrafiltrate was 1 and 2 hours for aqueous solutions and blood plasma, respectively. The filtration of solutions of Difco dried serum was very slow, probably because of clogging of the membrane. These data are presented in Table I. Table I also shows that the membrane is permeable to small colloidal particles such as inulin but impermeable to beryllium hydroxide and to protein. For this work, such dilute suspensions

(1) Feldman, I., Neuman, W. F., Danley, R. A,, and Havill, J. R.,

Univ. of Rochester Atomic Energy Project, Rept. UR-59 (1948). (2)

(3) (4) (5) (6)

Ferry, J. D., Chem. Reu., 18, 373 (1936). Kruhoffer, P., Acta Physiol. Scad., 11, 1 (1946). Nicholas, H. O., J . B i d . Chem., 97, 457 (1932). Rehberg, P. B., Acta Physiol. S c a d . , 5, 305 (1943). Underwood, A. L., and Neuman, W. F., Univ. of Rochester Atomic Energy Project, Rept. UR-19 (1948); ANAL. CHEM.,

21,1348 (1949). (7) Waard, D. J. de, Arch. n&erZad.physiol., 2, 530 (1918). R E C E I V ~April D 14, 1949. Based on work performed under contract with the United States Atomic Energy Commission &tthe University of Rochester Atomic Energy Project. Rochester, N. Y.

Improved Manostat and Manometer WILLIAM P. RATCHFORD

AND

M. L. FEIN

Eastern Regional Research Laboratory, Philadelphia, Pa.

LTHOUGH many pressure controllers suitable for use in A vacuum distillations have been reported (6),it is believed that the manostat described here is better adapted to precision vacuum fractionation than those previously described. A novel feature is the use of a completely enclosed magneboperated screw, which permits rapid and precise adjustment to any pressure over a wide operating range. The manostat is substantially independent of small variations in room temperature or pump capacity. Moreover, it is small and light in weight, and may be clamped conveniently to the usual laboratory rack or ring stand. Its construction requires the services of a skillful glassblower. A manufacturer of scientific apparatus is now developing a model for commercial production. Because deviations from the controlled pressure depend not only on the pressure-sensitive element, but also on the rest of the system, Figure 1 gives a semischematic representation of the arrangement used to obtain the degree of constancy reported here. The modified 40 X 170 mm. borosilicate glass test tube, 13, has an %nun. sidearm, 10, and a 14/38 ground-glass neck, 9. It holds approximately 40 ml. of mercury. The amount is not critical, but should be enough to fill the central tube completely, in case tube 13 is allowed to come to atmospheric pressure.

The 8-mm. central tube, 12 (soft glass ground to size 14/38), extends to within a few millimeters of the bottom of tube 13 and terminates in a 0.5-mm. hole, 15. The u er portion, 12 X 160 mm., qontains two glass :abs, which h o 8 s t a d e s s steel bracket 7; thw bears two extensions, 25 and 8; 25 serve8 as a bearing for the stainless steel screw, 4 (50 threads per inch), and 8 serves as pivot for the screw. The upper end of the screw terminatas in a soft iron vane, 3. On the screw is threaded stainless steel nut 6 of such size that when the screw is turned the nut is prevented from turning with it b the bracket, and so must advance or retreat alon the screw. Kttached to nut 5 w a stiff tungsten wire lead, 11, w%ch is connected electrically to the outaide of the central tube a t the top through the nut, screw, bracket, and latinum wire 27, which is sealed through the tube. The centr8 tube also contains the platinum lead, 14, which is in contact with the mercury at all tunes and 18 held against the side by a thin glass tube tacked to the wall. Thia lead w also sealed through a t the top, thus making the second of terminals 1, 1 The seals a t the top are protected by thermoplastic cement, which also holds in place the metal guide, 26, on which rests the annular magnet, 2. As the magnet is turned, it turns vane 3 and thus screw 4. moving electrode axially - the ti0 of the tungsten within the tube. The terminals are connected to an electronic relay, 28, which operates solenoid 22 (IO00 ohms). Use of an electronic relay minimizes the probability of sparking at the contact between

639

V O L U M E 22, N O . 6, J U N E 1 9 5 0 electrode 11 and the mercury. A spark is undesirable, especially where flammable materials or explosive vapors may be present in the confined space. The solenoid is connected to the relay, so that when the manostat circuit is broken, plunger 23 (faced off with a soft rubber pad, 24) is drawn up, thus opening orifice 16.

el

.

I

The range 3 to 760 mm. can be covered in five steps. Explosive vapors should be excluded from the central tube, even though the use of an electronic relay minimizes the probability of sparkm is preferable to connect the manostat and leak by a T-type connection near the pump. With an ordinary distilling column no surge chamber between the system and marlostat connection is necessary, but if the controlled system has a small volume or if the pressure is high (150 mm.), a surge chamber will help smooth out the pressure variations caused by the controlling action. It is unnecessary to clamp off the manostat when fractions are changed; admission of air into tube 13, as in evacuating a fresh receiver, causes no permanent change in the system; controlling action resumes when the mercury level reaches the end of the electrode. In some cases of this kind, however, cohesion of the mercury and glass results in a temporary control point 0.1 to 0.2 mm. away from the initial pressure, to which the controller readjusts rather slowly. Hence, it may be advisable to tap the manostat or make an adjustment.

ft

I

‘UYP

Figure 1.

If it is necessary to control pressures higher than about 100 mm., the manostat is tilted on its side, and air is admitted to the central tube while the s stem is near the desired prcssure. If air a t 100 mm. is admittei, the new operating range is 100 to 220 mm.; if air a t 220 mm. is admitted, the range is 220 to 375.

Diagram of Manostat

The flow of air into the system and pump is controlled by a special stopcock, with independent adjustment of a continuous leak through orifice 19, and adjustment of the intermittent leak through orifice 16. The plu of each stopcock is carefully grooved to permit throttling of gas tow. Orifice 16 is a 1-mm. capillary ground to a conical shape and then ground flat on the sealing surface. To place in operation, side arm 10 is connected with pressure tubing to a good pump, and the manostat is tilted to expose hole 15 and thoroughly evacuated. The manostat is then restored to the upright position; the side arm is clamped off, detached from the pump, and attached to the system to be controlled, 18. Pump 17 is started, and with orifice 16 closed by the stopcock, the system is pumped to a few millimeters below the desired p r w u r e by adjustment of orifice 19. At the lower pressures, orifice 19 is completely closed. The pinchclamp on the manostat is next opened, and the position of electrode 11 is adjusted to open the circuit and lift the plunger. The stopcock on the right is then opened slightly to permit the pressure to rise slowly. As soon as the mercury touches electrode 11, the relay permits the plunger to fall on orifice 16. The pressure begins to fall and the mercury is withdrawn from contact 11, whereupon the plunger rises, and the cycle repeats. With proper adjustment of the stopcocks, operat.ion will occur with a slight “breathing” motion of the mercury meniscus. If the equilibrium pressure in the system, measured, for example, by manometer 20, is not that desired, it is quickly brought to the correct oint by turning the magnet (one complete turn of the screw afvances the nut 0.5 mm.). Adjustments can be made to a more precise degree than can be read on the usual U-tube manometer. The manostat prepared in this way is also an absolute manometer. For some purposes, measurement by this manometer may be sufficient, but it is usually desirable to connect a second manometer directly to the system. When the manostat is to be used as a manometer, a millimeter scale is fastened to the side of the tube ,or may be etched directly on tube 13.

The manostat retains all the advantages of an adjustable electrode ( I ) , but avoids the uncertainties and difficulties of leakproof packing ( 8 ) ; moreover, the controlling action in the range 3 to 100 mm. occurs in a vacuum, where fouling of the mercury is less likely to occur. Because the manostat and relay, in effect, constitute a switch, modifications in the leak system could be made to meet various requirements. The relay could be used to switch a pump (2, i‘), operate a breather valve between system and pump (6), or bleed inert gas to a system by means of a breather valve (4). The authors have used the manostat to control the distillation pressures for a series of boiling point measurements (S),and to check reported boiling point measurements. The lowest pressure a t which it was operated was 3.0 mm. Orifice 19 was closed off entirely, and after preliminary adjustment the pressure re mained constant for several hours, without any visible variation in pressure apparent on a Dubrovin gage, which is calibrated in 0.2-mm. intervals. At 144.0 mm., the pressure remained constant within 10.1 mm. for 5 hours. The highest pressure a t which the manostat was operated was 523 mm. At the highest pressure, the capacity of the pump exceeded that of the leaks, and it was necessary to throttle the pump with a pinchclamp. The test was run for 3 hours; the pressure remained a t 523 * 0.2 mm. ACKNOWLEDGMENT

The authors wish to thank J. Leonard Schwartz, Sr., and Hugo Engelhardt of the Philadelphia Thermometer Company for special construction of the magnetic controlling unit, and Harry John of this laboratory for construction of the stopcock. LITERATURE CITED

Cox, H. L., IND.ENG.CHEM.,ANAL.ED.,1, 7-8 (1929). Ellis, L. M., Ibid., 4, 318-19 (1932). Fein, M. L., and Fisher, C. H., J.Org. Chem., 13,749 (1948). Jacobs, G. W., IND.ENG.CHEM.,ANAL.ED., 7, 70 (1935). Morton, A. A., “Laboratory Technique in Organic Chemistry,” pp. 105-9, New York, McGraw-Hill Book Co., 1938. (6) Palkin, S., and Nelson, 0. A., IND. END. CHEM.,ANAL.ED.,6, 386 (1934). (7) Willingham, C. B., Taylor, W. J., Pignocco, J. M., and Rossini, F. D., J . Research Natl. Bur. Standards, 35, 219 (1945).

(1) (2) (3) (4) (5)

RECEIVEDApril 13, 1949. Presented before the Meeting-in-Miniature. Philadelphia Section, AMERICAN CEEMICAL SOCIETY, January 1949.