ANALYTICAL CHEMISTRY
942
as usual. The resistances in ohms (IO)across the polarographic circuit were as follows: Cell
Teat aolution 1mM
2mM
A 1560
1480
B 1230 1170
C .., 1140
D
...
1120
I n comparison, a conventional mercury pool electrode in the otherwise identical system gave resistances of 1310 and 1160 ohms for the 1.00 and 2.OOmM test solutions, respectively. The new reference electrodes were handled frequently during the course of these investigations. Because of this and their excellent stability, they should be particularly suited for routine polarographic analysis. ACKNOWLEDGMENT
The authors wish to express their appreciation to Gertrude E. Wagner for blowing the reference electrodes.
An inert gas-filled dry box affords an oxygen-free atmosphere. In such an apparatus, absolute methanol can be used to dissolve the sodium-potassium directly without an inert-liquid solvent medium. If an adequate dry box is available, this technique is probably the most satisfactory method known for dissolving the alloy. As such a box is not always available, a procedure has been successfully formulated which embodies the principles outlined above. PROCEDURE
The sample of alloy, usually sealed in a glass container (metal containers sealed with an organic: wax are often used), is placed in a beaker containing sufficient n-hexane to cover the sample completely. The beaker rests in a mineral oil bath. The nhexane (boiling point 60” to 65” C.) is less dense than sodiumpotassium, is inert to the alloy a t the temperatures involved, and by floating on the surface of the metal serves as an effective liquid blanket. 4 funnel is inverted over the beaker (as shown in Figure 1) and a blanket of helium or argon placed over the reartion.
LITERATURE CITED
Coggeshall, G. W., 2. p h y s i k . Chem., 17, 62 (1895). Ellis, J. H., J . Ani. Chem. Soc., 38, 737 (1916). Haring, H. E., U. S. Patent 1,865,004 (June 28, 1932). Kolthoff, I. M., and Lingane, J. J., “Polarography,” T’ol. I, New York, Interscience Publishers, 1952. Langer, A., IND.ENG.CHEix., ANAL.ED.,15, 465 (1943). Lewis, G. N., Brighton, T. B., and Sebastian, R. I,., J . A m . Chem. SOC.,39, 2245 (1917). Lingane, J. J., and I,aitinen, H. A . , IXD. E s c . CHEX., .%NAI.. ED.,11,504(1039). Majer, V., Co&ction Czcchoslo~.Chem. Communs., 7, 146 (1936). Meites, Louis, J . S m . Chem. Soe., 72, 2293 (1950). Pesce, h i . R., Knesbach, R. I.., and Ladisch, €1. K., .%SAL. CHEM.,25, 979 (1953). Sauer, L., 2. physik. Chem., 47, 146 (1904). Smith, G. S.,Analyst, 75, 215 (1950). Dissolution of Sodium-Potassium Alloys for Purposes of Analysis. J. C. White, C. K. Talbott, and L. J. Brady, Analytical Chemistry Division, Oak Ridge Kational Laboratory, Oak Ridge, Tenn. ISSOLUTION of
sodium-potassium alloys prior to chemical or instrumental analysis is an extremely hazardous operation because of the great reactivity of these alloys with common aqueous and organic reagents. Water reacts exothermally with the alloy with explosive violence and is obviously not recommended in any form or concentration. Organic reagents are therefore generally used for dissolution. The rate of reaction of sodium-potassium with the lower alcohols is only slightly less than that with water, and the flammability of these reagents presents an additional hazard. Those organic reagents, such as higher alcohols, which react fairly slowly are unsatisfactory because they produce by this reaction alkali metal alcoholates. Being slightly soluble in the alcoholic medium these alcoholates cover the alloy with a protective coating and retard the rate of dissolution. In general, the problem is to use a solvent which yields a controllable reaction with sodium-potassium. The conditions necessary for a rapid yet safe dissolution of sodium-potassium are essentially these:
D.
Exclusion of oxygen. Use of an inert liquid, the specific gravity of which is less than that of sodium-potassium. Controlled addition of reactive reagent. The complete exclusion of oxygen obviously eliminates the possibility of combustion. Submersion of the alloy in an inert liquid serves to prevent contact with air; the liquid also provides a medium in which dissolution can be conducted and the heat of reaction dissipated. Cautious addition of the solvent is necesm r y in order to control the rate of reaction.
I
I Figure 1
Absolute methanol or ethanol is added dropwise, 2 t o 3 drops per addition, until the reaction is complete. The alcoholate formed is appreciably soluble in this medium and does not coat the unreacted sodium-potassium to any great extent. Intermittent stirring with a glass rod eliminates any tendency of the formation of a coating. The dissolution is complete in a relatively short time without any noticeable increase in the temperature of the reaction medium. When the reaction is complete, as noted by the absence of any unreacted sodium-potassium, an excess of alcohol is added as a safety measure. Water is then added to dissolve the alcoholate that has been formed. The two phases are then transferred to a separatory funnel and shaken vigorously, and the heavier water layer is separated.
If required, an aliquot of the aqueous phase may be titrated with standard acid for determination of sample weight. -4s a largr excess of water is present, hydrolysis of the alcoholates formed is essentially complete KO evidence has been noted of alkali metal in the organic phase, Metal impurities almost invariably follow the alkali metals into the aqueous phase. This point can be checked by evaporating the hexane and noting any residue. This procedure has been successfully used to dissolve samples of sodium-potassium in the range of 0.05 to 25 grams. Other precautions which are strongly recommended include the use of a plastic shield in front of the reaction vessel and the wearing of gloves by the operator. Special care must be exercised when handling sodium-potassium that is heavily contaminated with oxides. Potassium is known
V O L U M E 26, NO. 5, M A Y 1 9 5 4
943
to form the superoxide, KOJJwhich is the stable oxide at room temperature, in the presence of normal partial pressures of oxygen. Gilbert [Chem.Eng. News, 26, 2605 (194811 states that the rather frequent explosions which occur in handling potassium can be attributed to the formation of superoxide on the surface of the metal a t room temperature. The cause of this explosion is not completely understood. Such an explosion has been observed in this laboratory with a sample of sodium-potassium heavily coated with oxide. The use of a dry boy is apparently mandatory when denling with such samples. BASEDon work performed for t h e Atomic Energy Commission by Carbide and Carbon Chemicals Corp., a Division of Union Carbide a n d Carbon Corp.
Apparatus for Hot Filtration, Extraction, and Recrystallization with Volatile and Flammable Solvents. Haruo Shiba, Chemical I,aboratory, Facult,y of Engineering, Pniversity of Hiroshima, Hiroshima, Japan. organic solution may be filtered through Rer4. gam1 and Stange’s hot filter (3),or for higher temperature FLAMMABLE
I
long enough to pour the solution onto the filter paper. Even if crystals deDosit on the filter paper during filtration, subsequent refluxing is sufficient to extract the substance out of the filter paper; therefore no loss of crystals is caused. The upper rim of the funnel must be wide enough to prevent the escape of the solvent vapor through the ground-glass joints, Q and R, of the funnel and the condenser. The upper part of the funnel may also be constructed as shown in Fighre 2, if the Figure 2. special H~~ funnel is relatively small. But, it is diffiFilter cult to remove the condenser from the funnel, if the ground-glass joints of the funnel and the condenser become tight, owing to the heat of reflusing liquid. This apparatus may also be uaed for the cont’inuous and automatic extraction of solid substances-e.g., as a substitute for a Soshlet ( 2 , 7 ) . I t is not nereswry to use n Ppecial thimble made of filter paper.
’
1’:iul’s hot filter ( 5 )may he used. Hut in either case, the evaporation of volatile and flammable solvents cannot be prevented. Especially in recrystallization, the deposition of crystals on the filtrr paper owing to evaporation of the mother liquor will cause troublesome manipulation and serious loss in yield. Hot filtration through a preheated Riichner funnel into a preheated suction flask msy not be advimble, as loss of volatile solvent can occur. Houlmi’s device (3)-set,ting a round-bottomed flask upon a stemless funnel and circulating cooling water through the flask:tlthough the same as the following in principle, is provisional and not gas-tight for the rising vapor of the solvent. Extractors of Kulilman and Gerschson ( i ) ,of Rudernian (6), and of Dreschel ( l i , are also similar in principle to the following, but are more r\priiiive or complicated. The author has devised a simple :tiid convenient hot filter. as shown in Figure 1, with which recrystallization with volatile organic solvents can be accomplished with ease and without loss of either solvent or the crystals deposited on the filter paper by ev:iporation of solvent. The glass apparatus includes a conical flask, A , connecting through ground joint, D, to a specially coiistructed funnel, E , having a some what constricted neck and two groundglass joints, E and R. The upper joint, R, has a broad rim which is ground flat to fit the similar joint, Q, ou the umbrellalike “skirt,” S, of the condenser, C. I n order to facilitate the rise of the solvent vapor through the apparatus, holes H and Z< are bored through the stems of the funnel and the condenser. T o manipulate the apparatus, fit a fluted filter paper, P, into the funnel, pour some of the aolvent to be used for recrystallization into the conical flask, and boil gently to preheat the funnel and the filter paper. IVhile the vapor of the solvent is rising among the flutes of the filter paper, take care t h a t the vapor does not reflux over the upper rim, R, of the funnel, because of the flammability of the solvent. As an added precaution, place the condenser on the funnel before refluxing. Then after removing the Bunsen burner flame to prevent possible ignition of the solvent, take off the condenser
Figure 1. Apparatus for Hot Filtration a nd Recrystallization
REFERENCES
( 1 ) Drewhel, J . p r a k t . Chem.. 123, 350 (187;). (2) Fieder, L. F., “Experinletits iii Organic Chemistry,” pp. 37, 41, New York, D. C. Heath Co., 1941. (3) Houben, J.. “Die Methoden der Organisclien Chetiiie,” 3rd ed., T-01. I. pp. 435-8, Leipsig, Georg Thieine. 1925. (4) Kuhlrnari and Gersrhson, 2 . anal. chem., 106, 145 (1936). ( 5 ) Paul, Th.. Ber., 25, 2209 (1892). (6) Kuderrnan, IND. EXG.CHEM..AX.AL.ED., 16, 332 (1944). ( i )Keigand, C.,
“Organisch-Cheniische Experinientierkiinst.” pp.
61-4, Leipsig, Johann Ambrosius Barth, 1938. (8) Weissberger, A,, “Technique of Organic Chemistry,” T’ol. 3. g p . 236-40, New York, Interscience Publishers, 1949.
All-Glass Atomizer for Chromatographic Analysis Procedures. Victor H. Ortegren \Vesteix Regional Research Laboratory, .%lbany. Calif. ail atomhe1 t o a developer solution on paper Tusedu ~ine chromatography is a common procedure requiring ot
IIE
L .
~ p i q
’
equipment that will produce a fine spray with a minimum of larger drops. The atomizer should be resistant t o chemical attack so that it does not contaminate the developer solution, convenient to use and easily cleaned, sturdy to avoid breakage, and inevpensive and relatively simple t o construct. Over 50 atomizere of the design described here have been made h! the author during the past yea1 in the glass-blowing shop in this laboratory. They have pi oved highly satisfactory and popular with the chemists using them. Rate of atomization can ea& be controlled, larger droplets of liquid that may be formed in the atomizing zone (especially on starting) are not discharged a i t h the spray nor allowed to run down on the outside of the apparatus but are automatically returned to the supply vessel, and the atomizer can be used R ith any vessel, even an open beaker 01 test tube, as the reservoir is not under poiitive pressure. Figure 1 represents an elevation in cross section. Figure 2 represents a plan view in cross section taken on AA’of Figure 1. The liquid to be atomized is contained in bottle 1 provided Mith a standard-taper ground joint (19/22), outer 2. The level of liquid in the 250-ml. reagent bottle, 1, is represented by line 3. The atomizer, 4, has a spherical jacket, 6 (about 45 mm. in diameter) provided with a large opening, 7 (about 18 mm. in diameter), for discharge of the atomized liquid. The bottom of a standard-taper ground joint (19/22), inner 5, is attached to tube 8 (11 mm. in outside diameter), provided with a vent, 9. For supplying liquid to the atomizing jet, a capillary tube, 10 (1 mm. in inside diameter), is used, which is rigidly supported in place by rods 11, 12, and 13. Air for atomizing the liquid is introduced via a conduit, 14, which tapers to a small nozzle, 15, as shown. Diversion conduit 16 connects with air conduit 14, and leads to a finger-controlled port, 17, on the outside of jacket 6. The air-supply conduit, 14, is connected to a source of com-
