by assuming use of a 1-megohm input resistor, which produces a counting rate of only about 5000 counts per coulombohm. Under these conditions the coulometer will record 2 X lo-’ coulomb, or 2x microequivalent, per thousand counts. This number probably represents the lower limit of usefulness of this coulometer. Some potentiostats, such as the type used in these determinations, operate with the Forking electrode a t ground potential. The voltage-tofrequency converter cannot be used directly with these potentiostats, because one of the input terminals is grounded through the line cord and the chassis. With these conditions, the converter must be insulated from the chassis of the potentiostat and connected to the line via an isolation transformer. The counter may also have to be floated
in a similar manner. This was employed during the above determinations, and the coulometer continued to operate in a satisfactory manner. The major drawback to this system is its relatively high cost ($650 for the converter). This is compensated somewhat by negligible construction costs: The converter can be used immediately as a coulometer, simply by connecting it to a counter. The ease of operation of the coulometer makes its application to controlled current coulometry (coulometric titrations) attractive. It is free from the usual start-stop errors of mechanical clocks and does not require a constant current for titration. By using a counter with an accurate gate time, the converter-counter combination may also be employed as a digital voltmeter. The coulometer should
prove suitable for automatic analysis, because the output of the counter can operate a digital recorder, and by proper choice of input resistor, it may be made to read directly in microequivalents, grams, etc. LITERATURE CITED
resentation, Division of Analytical Cfemiatry, 142nd Meeting, ACS, Atlantic City, N. J., September 1962. (2) Booman, G. L., ANAL.CHEM.29, 213
(1) Bard, A. J., submitted for
(1957).
(3j-Krarner) K. W., Fischer, R. B., Zbid., 26,415 (1954). (4) Lingane, J. J., “Electroanalytical Chemistry,” 2nd ed., 339-50, 452-60, Interscience, New R r k , 1958. SUPPORTof this investigation by the Robert A. Welch Foundation is gratefully acknowledged.
A Simple Qualitative Ashing Microtest for Organosilicon and Other Organometallic Compounds Leonard Spialter and Manuel Ballester,‘ Aeronautical Research Laboratories, Wright-Patterson AFB, Ohio
of studies on synthetic Ifrequently organosilicon compounds, questions arose as to whether a given N THE COURSE
substance isolated from a reaction mixture contained the desired metallic or metalloidal atom. Ultraviolet absorption spectra proved useless, while infrared spectra were not always unequivocally interpretable, particularly for complex molecules containing certain functional groups. Quantitative microanalysis was not economical or rapid enough for our purpose either. Most of the qualitative tests for silicon in organic compounds are based on chemical cleavage processes which yield silica. These include hydrolyses and/or wet oxidations with fuming nitric acid (C), sulfuric acid alone (8)or in conjunction with potassium permanganate (IS) or with glacial or nitric acid (7), Myo perchloric acid (a good quantitative procedure also) ( 5 ) , hydrochloric acid or sodium hydroxide with subsequent acidification (for silicon orthoesters and siloxanes) (9, II), and fusion with sodium peroxide mixtures in a Parr bomb (12, 16) or a platinum wire loop
face works “reasonably well even with volatile substances.” However, we find that carbonaceous smoke from aromatic silanes obscures this test. Other references (1, 16) cite combustion as a suitable test but give no details or critical evaluation of its scope and reliability. I n our hands, the conventional methods for burning a sample to produce silica-such as holding the material in a flame or furnace with a spatula, wire loop, rod, crucible, or other containerwere not only wasteful of material but also did not yield consistently reliable results. Combustion of a sample a t the open end of a glass capillary tube succeeded in giving a visible ash for some highly and moderately volatile alkylsilanes but failed with aromatic silanes. These latter usually produced a carbonaceous deposit which was difficult to oxidize and readily fused
(10).
holder
Although the concept of burning the unknown substance and examining the residue for silica is well established (14), the literature is vague on a reliable procedure. Eaborn (2) claims that the method of burning a drop of material on a glass rod beneath a cold glass sur-
On leave from the Department of Organic Chemistry, University of Barcelona, Spain.
Substance
Figure 1.
Assembly
into the glass with no sign of the expected silica ash. A simple assembly and a reliable experimental procedure have been developed which gave, with proper combustion technique, a white silica ash from each of more than 40 different organosilicon compounds of varying volatility, structure, composition, and thermal stability. For solids, a 1-mg. sample generally proved more than sufficient, although liquids, particularly the more volatile ones, usually required more material. With greater refinement such as the use of a smaller assembly, a microburner, and examination under a microscope, this method may be successfully used with even less sample. EXPERIMENTAL
Apparatus. T h e assembly (Figure 1) consists of a 5-cm. length of nichrome or platinum wire, about 0.8 mm. in diameter, with a hook of 0.8 mm. radius a t one end. The wire is then passed through a borosilicate glass melting point capillary, about 1.6 mm. 0.d. and 1.5 to 2 cm. long, until the hook contacts the glass, and then about 1 em. into a length of thick-walled glass capillary tubing (1 mm. bore) which serves as a handle. The wire is bent slightly near each end to keep it from slipping out of the handle and also to keep the thin-walled capillary from sliding. Procedure. About 1 mg. of powdered sample or a small drop of liquid is collected in the hook; if a solid, the sample is melted onto the hook by warming it near a flame. For the ashing procedure, a Bunsen burner VOL. 34, NO. 9, AUGUST 1962
1183
?‘
Figure 2.
//
Combustion of sample
should be used with the gas-air mixture so adjusted t h a t the luminosity at the flame tip has just disappeared. This provides an oxidizing yet relatively cool flame. The sample is then cautiously but resolutely burned by bringing the hook near enough to the flame, with the wire and capillary about 30” from the vertical as shown in Position A , Figure 2, so that the material ignites and continues to burn with a flame. After flame from the combustion has disappeared, the full thinwalled capillary is exposed to the flame as in Position B for about 5 seconds and is then removed and allowed to cool for 1 minute. This apparently serves to condition the carbonaceous deposit which has usually appeared on the glass near the hook by this time. The hook and the carbonaceous deposit are then returned to the flame and heated as in Position B with the wire at an inclination of 90” or greater from the vertical until the carbon has been oxidized away. Care must be exercised to avoid heating the glass beyond the area of the deposit. RESULTS AND DISCUSSION
The experimental results, presented in Table I, show that selfsustaining flames are associated with the more volatile compounds, such as the alkylated simple molecules, or with those like
tetracyclohexylsilane and tetrathienylsilane, which readily decompose at the flame temperature to give volatile combustible products. The highly arylated molecules and those of low volatility, appreciable molecular weight, and good thermal stability, such as silicone oil and grease, required continuous heat input during the flaming operation. The formation of carbonaceous deposits with no sign of a gray or white ash just prior to the final oxidation was found usually associated with aromatic groups in the molecule. However, with silicon-rich aromatics such as lJ4-bis(trimethylsilyl) benzene, grayish silica was noted even in the presence of the carbon. Compounds without such aromatic groups gave clean light ash immediately after preliminary combustion. An exception was observed with tetradecyltrichlorosilane which gave some carbon due, evidently, to the presence of the large alkyl group. No difficulty, however, was experienced in burning away such black deposits. The organosilicon compounds all clearly deposited positive white silica on the thin-walled capillary tube near the wire hook. Each of the components of the ashing assembly seems to perform a useful function and none can be eliminated. The hook holds the sample; the wire supports the thin-walled capillary and cools it sufficiently to prevent fusion of the deposit with the glass; the capillary acts to trap and condense volatilized components and smoke from the sample and also provides a disposable substrate uniquely suitable for detection of ash; and the wire, removable from the handle, allows replacement of the thin capillary tube before each run. To test its utility, the present ashing technique was extended to some other organic derivatives of Group IVb elements, as indicated in Table I. All gave positive results: The germanium oxide
Table I. Interpretation of Ashing Characteristics I. Selfsustaining flame“; little or no carbonaceous depositb Silanes and siloxanes of low or moderate molecular weight and containing alkyl, cycloalkyl groups, and/or hydrogen atoms on silicon. Exam les : hexaethyldisiloxane, dioctylsilane, tetradecyltrichlorosilane (eight compounds stuied) 11. Selfsustaining flame”; much carbonaceous depositb Arylsilanes of low molecular weight. Example: phenyltrimethylsilane (three compounds studied) 111. Nonselfsustaining flame”; little or no carbonaceous depositb Silanes or polysiloxanes (silicones) of moderate or high molecular weight and higher silicon content. Examples: silicone oil, bis( trimethylsily1)benzene (four compounds studied) IV. Xonselfsustaining flamea; much carbonaceous depositb Highly arylated silanes and siloxanes of moderate. to high molecular. weight. EXamples: phen ltrichlorosilane, hexaphenyldisilane, biphenylyltriphenylsilane (28 compounds studie4 Note: The arylated germanes, stannanes, and plumbanes (five compounds studied) fall into this class but note differing residue characteristics described in the text Selfsustaining flame means that the compound maintained its own flame outside the burner flame area once it had been ignited (Position A , Figure 2). Nonselfsustaining flame indicates that burning of the sample required continuous heating by the burner flame. b Carbonaceous deposit refers to a dark residue left after preliminary combustion. 0
1 184
ANALYTICAL CHEMISTRY
was gray-white, similar to silica; the tin oxide was bright white, thick, and powdery; the lead oxide was light yellow. Special care was needed with germanium and lead oxides to prevent overheating of the ash which readily fused into the glass. For these, the final oxidations were performed in the hot air zone just above the upper flame tip. Familiarization trials are recommended with germanium and lead compounds. Both the tin and lead oxides seemed to catalyze strongly the oxidation of the carbonaceous deposits from their compounds. The described procedure is probably applicable also to other compounds containing elements whose oxides are relatively nonvolatile and unreactive with the capillary material a t the temperatures used. The chemical identity of the ash residue from a sample can, of course, be readily verified by standard qualitative analytical tests such as x-ray diffraction or by suitable treatment for the appropriate spot tests (3, 6 ) . This ashing procedure has greatly facilitated interpretation and preliminary identification of reaction products in our organometallic research program. Routine use has also increased the efficiency of operations involving the analysis and separation of mixtures, such as recrystallization and both liquid and vapor phase chromatography. LITERATURE CITED
(1) Bazant, V., Chvalovsky, V., Rathousky, J., “Silikony,” p. 115, Statni
Nakladatelstvi Technicke Literature, Praha (Prague), 1954. (2) Eaborn, C., “Organosilicon ComDounda.” w. 503, Butterworth’s, Lonaon, 1960.(3) Feigl, F., “Spot Tests,” Vol. I, Elsevier Publishing Co., New York, 1954. (4) Friedel, C., Crafts, J. M., Ann. 136, 203 (1865). (5) Gilman,’ H., Clark, R.
N.,Wiley, R. E., Diehl, H. J., J . A m . Chem. SOC.
68,2728 (1946). (6) Gilman, H., Ingham, R. K., Gorsich, R. D., J . Am. Chem. SOC.76,918 (1954). (7) Gilman, H., Smart, G. ?;. R., J . Org. Chem. 15, 720 (1950).
(8)Kipping, F. S., Lloyd, L. L., J . Chem. S O C . 79, 449 (1901): (9) Kreshkov, A. P., Bork, V. A., Ti-.
Komis. PO Analit. Khim., Akad. Nauk SSSR, 3, 354 (1953); C.A. 47, 2646
(1953). (10) Kreshkov, A. P., Bork, V. A., Zh. Analit. Khim. 6, 78 (1951); C.A. 45, 6125 (1951). Tr. Komis. PO Analzt. Khim., Akad. Nauk SSSR. 3, 361 (1953). (11) McConville, H. A., Gen. EZec. Rev. 52, No.11, 9 (1949). (12) Marvin, G. G., Schumb, W. C., J . Am. Chem. SOC.52, 574 (1930). (13) Polis, A., Ber. 18, 1540 (1885). (14) Ibid., 19, 1024 (1886). (15) Rochow, E. G., “An Introduction t o
the Chemistry of the Silicones,’’ 2nd ed., p. 160, Wiley, New York, 1951. (16) Whitmore, F. C., Sommer, J. H., DiGiorgio, P. A,, Strong, W. A., Van Strien, R. E., Baile D. L., Hall, H. K., Pietrusza, E. Kerr, G. T., J. Am. Chem. SOC.68, 475 (1946).
%.,
