a six-electron oxidation again with one
0
Table II. Coulometric Oxidation of Methyl Hydrazines Solution: 1.OM in HzSO,. Oxidations carried out at f l . O volt us. SCE Coulometer Amount of nitrogen, Amount, rdg., mmol. meq. n mmol. Methyl hydrazine 0.238 0.916 3.86 0.239 1,2-Dimethyl hydrazine 0.510 3.16 6.20 0.532 0.239 1.42 5.94 0,240 1,l-Dimethyl hydrazine 0.326 1.53 4.7 0,194“ 1.92 5.9 0.270* Electrolysis interrupted for about 1 hour after 0.65 meq. oxidation. Electrolysis carried out continuously.
jacketed under stable laboratory temperature conditions. An electromechanical potentiostat similar to that described by Lamphere and Rogers ( 1 ) was used. Currenttime integration was accomplished with a ball-and-disk integrator attached to a strip-chart recorder. Procedure. I n t’he reduction studies, the solution in the cell was 0.1N in sulfuric acid and about 25 ml. in volume. Initial saturation with hydrogen was achieved by preelectrolysis a t about -0.4 volt us. SCE. Aliquots of standard iron(II1) solution were then added and the electrolysis was performed. For the hydrazine oxidation studies the solution was 25 ml. of 1 , O M HZSOa and was presaturated with nitrogen.
RESULTS
Results for the electrolysis of ferric and hydrogen ions at various potentials are shown in Table I. When the electrolysis of iron(II1) was carried out a t 0.0 volt us. SCE, no hydrogen ion reduction occurred and the measured coulombs agreed with those calculated from the equivalent of iron added. Electrolysis a t -0.4 volt leads to concomitant reduction of hydrogen ion; subtraction of the volume of hydrogen gas evolved, however, again allows calculation of the amount of iron. Results for the oxidation of several methyl hydrazines are shown in Table 11. Methyl hydrazine undergoes a fourelectron oxidation and one mole of nitrogen is formed per mole of hydrazine. 1,2-Dimethyl hydrazine undergoes
mole of nitrogen formed per mole of hydrazine. The oxidation of 1,l-dimethyl hydrazine is coni1Ilicated by slower intermediate chemical reactions. Quantitative nitrogen formation is not observed and the apparant n-value depends upon the rate a t which the electrolysis is carried out. Further study of these systems will be reported elsewhere. I n all cases the gaseous electrolysis product was identified by mass spectrometry as nitrogen. The cell should also find use in constant current coulometry (coulometric titrations), especially for the production of strong oxidants and reductants. Adaption of the cell to electrolysis with a mercury electrode should be straightforward. Further work with this cell is in progress. LITERATURE CITED
(1) Lamphere, R. R., Rogers, L. B.,
ANAL.CHEM.22,463 (1950). (2) Lingane, J. J., “Electroanalytical Chemistry,” p. 453, Interscience, New York, 1958.
DONALD 11,KING ALLENJ. BARD Department of Chemistry The University of Texas Austin, Texas 78712 RECEIVED for review June 2, 1964. Accepted August 28, 1964. Work supported by a grant from the Robert A. Welch Foundation.
I o d ide-T richI oroa c et ic Acid SIR: Numerous good methods for the quantitative determination of azide ion are known. However, there are few methods for the determination of the covalently bound azido group. Ugi (10) has successfully adapted the reaction of aryl azide with arsenite ion (1-3) to the gasometric determination of the aryl azides.
Messmer and Mlinko (6, 7 ) determined the azido group of 1-azido-1-acetyl sugars by ceric ion oxidation of hydrazoic acid liberated by acid digestion. However, except for certain azides, such as triphenylmethyl, acyl and sulfonyl azides, and the 1-azido-1-acetyl sugars, the azido group is not readily hydrolyzed intact by acid or base. The usual result of such attempts is cleavage
2352
ANALYTICAL CHEMISTRY
of the group to form nitrogen plus a complex mixture of imines, amines, olefins, keto compounds, etc. Hoffman, Hoch, and Kirmreuther ( 4 ) determined the azido group in guanyl azide and in carbamyl azide by measuring the nitrogen evolved in the reaction
analyzed benzenesulfonyl azides by two methods 15). Reaction a i t h triphenylphosphine caused quantitative liberation of nitrogen. (CsHJaP
+ CsHsS0211;3 - +
+
C B H ~ S O ~-NP(CEH5)I
+
?rT2
(3)
A potassium iodide-aqueouq acetic acid Rater-insoluble azides were not investigated. Smith (8) reported the reaction of aryl azides with hydroiodic acid to give iodine, nitrogen, and the corresponding arylamine. The determination of the azido group by titration of the liberated iodine was subject to error because of the tendency to form excess iodine by air oxidation of hydroiodic acid (9). The preparation and preservation of stock solutions of iodine-free hydroiodic acid constituted a further difficulty in this analytical approach. Leffler has
combination brought about a quantitative liberation of iodine upon heating. In this laboratory a method was sought for the analysis of various azides, primarily aliphatic. Ugi’s ( I O ) procedure was not suitable for simple alkyl azide. because ot their unreactivity with arsenite ion, Both of Leffler’s methods were also unsuitable because of the poor reactivity of alkyl azides toward triphenylphosphine and potassium iodide-ncetic acid. -4 method has been developed which 1s a modification of the hydroiodic acid reaction. I t involves
Table I .
Azide Determinations
Compound
Moles XZ/azido group X 100 99.5f 0.6a 100.9 102.5 102.8 100.9 102.1 100.9 101.4 99.2 98.6b,c 97 . 8c,d 90.3d,6
104.
99.0
100.6 I C6H5
101.0 94.Of 91.3 97.lC 95.8
85 2/80 99.39 101.9* 101.5 a Average of 7 determinations; thiosulfate titration of iodine yielded 99.7%. * Carried out under nitrogen because of oxygen absorption. c Average of 3 determinations. d Arsenite method gave 10l.OYc for CBH6N3, 100.5% for
Slow reaction. Impure sample. 0 l'ery slow reaction. h Sat. aq. solution; also analyzed gasometrically by the neutral iodine method as 102.5y0. 6
/
the measurement of nitrogen liberated by reaction of the azide with hydniodic acid generated in situ by the action of 90% trichloroacetic acid (loyo water) and sodium iodide. This system will generate hydroiodic acid very rapidly upon demand, but iodine formation by air oxidation proceeds slowly. Trichloroacetic acid also has good solvent properties for many organic substances. With this combination most alkyl and ionic azides react vigorously to give quantitative amounts of nitrogen and iodine. EXPERIMENTAL
Apparatus. T h e equipment employed was similar to t h e standard microhydrogenation apparatus, with mercury leveling bulb, mercury manometer, 100-ml gas buret, and a 100ml. 2-necked flask. One neck of t h e flask contained a glass stopper with a hook for suspending a platinum bucket containing t h e azide. Rotation of t h e stopper dropped t h e bucket directly into the reagent, which was stirred magnetically. Reagents. Trichloroacetic acid (90%) was prepared by heating a mixture of 9 parts acid to 1 part water until a homogeneous solution results. This solution remains liquid a t room temperature. Sodium iodide (B & A granular reagent) was used without grinding. Other forms (crystalline) were not as satisfactory. Procedure. Three grams of sodium iodide and 15 ml. of trichloroacetic acid were placed in the flask. T h e azide (0.2 to 0.6 gram, depending on azide content) was weighed in t h e platinum bucket, which was then positioned on t h e stopper-hook. A few minutes of stirring was used to allow equilibrium before closing the system and dropping the bucket into the reagent mixture. The reaction time varied considerably from a few seconds to a few days depending on the nature of the azide. Since heat is generated in the reaction, 30 to 60 minutes is required to reattain thermal equilibrium if a n air bath is used. The use of a constant temperature bath around the reaction flask greatly speeds up the rate of determination but it was not used for most of the analyses. RESULTS A N D DISCUSSION
Table I shows the results obtained with various azides. Unless otherwise specified the values are for sinqle determinations. An estimate of the degree of precision obtainable by this procedure was made with octyl azide which had been carefully purified by diqtillation. Seven determinations gave an average value of 99.5 f 0.6%. The reaction with alkyl azides liberates iodine quantitatively as well as nitrogen. However, it is inconvenient
to have to neutralize a large quantity of trichloroacetic acid before titration with thiosulfate. Trifluoroacetic acid has also been tried and found effective but, because of its high vapor pressure and pungent odor, was not evaluated further. The mechanism of this reaction is under investigation, but is not yet well underFtood. However, it is obvious that with primary alkyl azides the first step cannot be acid hydrolysis because these azides are quite stable toward trichloroacetic acid in the absence of sodium iodide. Furthermore, displacement of azide ion by iodide ion must be ruled out because in neutral solvents organic azides are quite stable toward iodide ion. This conclusion is based on the fact that alkyl axides can be synthesized in high yield from alkyl iodides and sodium azide in diethylene glycol in which both sodium azide and sodium iodide are very soluble. It is felt that the protonated azido group is highly susceptible to attack by hydroiodic acid, which produces an unstable intermediate which decomposes to the observed products. (By contrast, the quantitative reaction of iodine with azide ion, which requires a sulfurcontaining compound as catalyst, occurs only in neutral or basic solution. [SI I? 2N8- --+ 3N2 21-). Whereas secondary and tertiary azides are less stable in acid than the primary azides, this fact should not affect the analytical value of this method. The decomposition of these azides in acids will produce either nitrogen or hydrazoic acid, which would be further converted to nitrogen and ammonium ion Thus, the quantitative evolution of nitrogen is assured.
+
+
ACKNOWLEDGMENT
The author is indebted to R. A4. Henry, W. P. Sorris, and W. G. Finnegan for many helpful discussions and for several of the compounds studied. LITERATURE CITED
(1) Gutmann, A., Her. 45, 821 (1912). (2) Ibid., 57, 1956 (1924). (3) Gutmann, A., Z. Anal. Chem. 66, 224 (1925). (4) Hoffman, K. A., Hoch, H., Kirmreuther. H..Ann. 380. 131 11911) ( 5 ) Lefflei, j. E., Tii;lo, +.,-7.' Org. C h e m . 28, 190 (1963). (6)llessmer, A,, lllinko, S.,Acta Chirn. Acad. Sci. Hung. 28, 389 (1961). (7)Ibid., 29, 119 (1961). (8) Smith, P. A. S.,J . .4m. Chem. SOC. 73, 2438 (1951). (9) Smith, P.A. S.,University of blichigan, private communication, August 1960. (10)Ugi, I., Perlinger, H., Behringer, L., Her. 91, 2330 (1958).
WAYNER . CARPESTER
Chemistry Division, Research Department U. S. Saval Ordnance Test Station China Lake. Calif. VOL. 36, NO. 12, NOVEMBER 1964
2353
