Sulfur Dioxide Adducts of Some Disubstituted Hydrazines

J. inorg,nucl.Chem..1971,Vol.33, pp. 4051-4055. Pergamon Press. Printedin Great Britain SULFUR DIOXIDE DISUBSTITUTED ADDUCTS OF HYDRAZINES SOME JOSEPH M. K A N A M U E L L E R Department of Chemistry, Western Michigan University, Kalamazoo, Michigan 49001 (Received 3 May 197 1) A b s t r a c t - G a s e o u s sulfur dioxide reacts with I,l-dimethylhydrazine, N-aminopiperidine, N-aminohomopiperidine, and N-aminomorpholine to form 1 : 1 adducts. The adducts are formed between the nitrogen atom of the R._,N group in the hydrazine and the sulfur atom in sulfur dioxide. Analytical. proton NMR and i.r. spectral data are given. INTRODUCTION SULFUR dioxide has been known for some time to form Lewis acid-base adducts with amines [1]. These are considered to be charge-transfer complexes in which the electron pair from the amine nitrogen is donated to the sulfur atom. Interactions of sulfur dioxide with hydrazines, however, have received little study. Michaelis and Ruhl in 1890 reported compounds formulated as [(C6Hs)N HN H212SO2 and (C6Hs)NHNHz.SO2 from the reaction of phenylhydrazine and sulfur dioxide[2]. Ephriam and Piotrowski in 1911 reacted anhydrous hydrazine and sulfur dioxide, obtaining a product formulated as (N2H~)2(SO2NHNHSO21, dihydrazinium hydrazinedisulfinate[3]. Later, two patents were issued based on this process and its extension to substituted hydrazines[4]. For example, l,l-dimethylhydrazine in reaction with sulfur dioxide in chloroform was reported to yield [(CH3)2NHNH2][(CH:02NNHSO2], 2,2-dimethylhydrazinium 2,2-dimethylhydrazinesulfinate. This work reports the formation of 1 : 1 adducts of alkyl disubstituted hydrazines when gaseous sulfur dioxide is passed into an ethereal solution of the hydrazine. The first evidence for such compounds was noted when an ethereal solution of l,l-dimethylhydrazine was reacted with sulfuryl chloride in the presence of triethylamine. This latter reaction, however, is not fully understood. EXPERIMENTAL Materials l,l-Dimethylhydrazine (J. T. Baker Chemical Co.) was distilled and stored over potassium hydroxide (b.p. 63°C). N-Aminopiperidine, N-aminohomopiperidine, and N-aminomorpholine (Adlrich Chemical Co.) were distilled prior to use (b.p. 70°C at 50 torr, 100°C at 50 torr, 64°C at 15 I. Discussions of these adducts are given in the following review articles: W. Karcher and H. Hecht, Chemie in Nichtwiiszrigen Ionisierenden L6sungsmitteln, Vol. 111. Pergamon Press, Oxford (1967); T. C. Waddington, Non-Aqueous Solvent Systems, Chap. 6. Academic Press, N e w York (1965); D. F. Burow, The Chemistry o f Nonaqueous Solvents. (Edited by J. J. Lagowski), Vol. 111. Chap. 2, Academic Press, N e w York (19701. 2. A. Michaelis and J. Ruhl, Chem. Ber. 23, 474 (1890). 3. F. Ephriam and H. Piotrowski, Chem. Ber. 44, 386 (19111. 4. N. W. Dachs, U.S. Pat. 2,803,688 (19571; B. Rudner and M. E. Brooks, U.S. Pat. 2,888,483 (19591. 4051 4052 J.M. KANAMUELLER torr respectively). Sulfur dioxide was obtained from Virginia Chemicals, West Norfolk, Va. Diethyl ether, benzene, and petroleum ether "B" were dried with calcium hydride, chloroform with calcium chloride. A nalyses Elemental analyses were performed by A. Bernhardt Mikroanalytisches Laboratorium, Elbach fiber Engelskirchen, West Germany. Melting points were determined in closed capillary tubes using a Thomas-Hoover melting point apparatus. A Perkin-Elmer 457 spectrophotometer was used to record the i.r. spectra. The proton N M R spectra were obtained on a Varian A-60 spectrometer. Preparation of(CHs)zN N H2"SOs A 500 ml three-necked flask was equipped with a CaCl~-filled drying tube, a magnetic stirring bar and a gas inlet tube reaching to the bottom of the flask. The third neck was stoppered. A solution of 300 ml of anhydrous ether and 7.6 ml of 1, l-dimethylhydrazine (6.0g, 0.10 mole) was prepared in the flask with dry nitrogen gas bubbling into the flask. It was then cooled with an ice bath. The sulfur dioxide gas tank was connected to the nitrogen gas inlet tube through a benzene-filled trap which acted as a flow rate indicator. With cooling and stirring, the sulfur dioxide and nitrogen gas mixture was slowly passed into the solution. The flow of sulfur dioxide was discontinued approximately ten minutes after the formation of a colorless solid was observed. After warming the reaction mixture to room temperature, the nitrogen gas flow was stopped. The gas inlet tube was replaced with a stopper and the drying tube with a gas exit tube containing a stopcock. Two thirds of the solvent was removed through the gas exit tube with the aid of water-aspirator vacuum and a dry-ice filled cold finger trap. All subsequent operations were performed in a nitrogen-filled glove bag. The solid was filtered from the solution. Further removal of solvent yielded more product. Several experiments of this type gave crude yields of 95 per cent. The first crop of crystals was often reasonably pure if moisture was rigorously excluded during the entire procedure. The compound was recrystallized by dissolving it in benzene, then adding petroleum ether "B" until precipitation. In addition, sublimation under reduced pressure can be used for purification. Melting point: 72-74-5°C. Anal. Calcd. for (CHa)2NNHz'SO2: C, 19.35; H, 6.49; N, 22.57; S, 25.82. Found: C, 19.15; H, 6.50; N, 22.37; S, 26.01. The molecular weight was determined cryoscopically in benzene (cryoscopic constant, 5.09 deg/m). Calcd.: 124-16; Found: 128, 131. The same material was obtained when chloroform was used as a solvent. There was no evidence for the formation of 2,2-dimethylhydrazinium 2,2-dimethylhydrazinesulfinate in the reaction mixture nor when a large excess of 1, l-dimethylhydrazine was added to the adduct. An adduct which contained two moles of sulfur dioxide per mole of the hydrazine was not observed. Preparation of other adducts In the manner described above, N-aminopiperidine, N-aminohomopiperidine and N-aminomorpholine were reacted with sulfur dioxide. Analyses, yields, and melting points are given below. (CH~)sNNH2"SO2. Yield, 87%. M.P., 69-72°C. Anal. Calcd.: C, 36"56; H, 7"37; N, 17.06; S, 19.52. Found: C, 36.57; H, 7.23; N, 16.91; S, 19.31. (CH~)tNNH~'SO2. Yield, 88%. M.P., 58-60.5°C. Anal. Calcd.: C, 40-42; H, 7.92; N, 15.72; S, 17.99. Found: C, 40.30; H, 7.71; N, 15.92; S, 18-19. O(CHzCH2)zNNH~.SO~. Yield, 81%. M.P., 94-98.5°C. (decomp.). Anal. Calcd.: C, 28.90; H, 6.07; N, 16.82; S, 19.29. Found: C, 29.16; H, 6.22; N, 16.92; S, 19-33. All of the compounds described above are colorless crystalline solids and are very hygroscopic. Hydrolysis to sulfite and the substituted hydrazinium ions occurs upon contact with water, thus handling procedures in an inert atmosphere are necessary. With storage, even in sealed ampoules, decomposition slowly occurs after several weeks. Selected i.r. spectral data and proton NMR data for the compounds and the hydrazines used are presented in Tables 1 and 2. RESULTS AND DISCUSSION The analytical data indicate that the products formed in the reactions of the hydrazines with sulfur dioxide have a 1:1 mole ratio of reactants. The question arises whether the compounds formed are truly Lewis acid-base adducts or hy- 4053 Sulfur dioxide adducts Table 1. Selected i.r. spectral data* Compound Absorptions in th region Absorptions in the region 3400-3100 cm -1 2850-2100 cm -j (CH 3)2NNHz'SO2t (CH3)zNNH2.SOz ( 10% in CDCI~, 0.1 mm cell) (CH3)2NNHz.SO2 l< 3% in benzene, 0.1 mm cell) (CH3)zNN H2§ (CH3)2NNH2 ( 12% in CDCI3, 0.1 mm cell) (CHz)sNN Hz.SOzt (CH2)sNNHz§ (CH2)sNN H 2'SO2t 3300, s; 3280, sh; 3190, m. 3305, s, 3260, sh; 3190, m. 2280, w, br. Absorptions in the region 1300-1050 cm -j 1280, 1130, 1225, 1090, m, br:~; 1195, s; w; 1100, m; 1075, m. vs; 1065, w; vs:~ 3300, w;3190, shon benzene C - H stretch• 1235, vs:l100, s; 1085, s. 3300, m;3210, w; 3140, m. 3340, s;3300, s; 3200, s. 1245, w;1210, w ; l 1 4 5 , m; 1090, sh;1060, m, br. 1240, w;1205, w:l140, m; 1055, s. 3260, s; 3160, s; 3105, sh. 2720, m, br; 2600, w, br; 2300, vw, br. 3300, m, br;3205, vw;3140, m. 3290, s;3180, s. 2260, vw, br. {CH2)nNNH2§ 3300, m, br; 3210, vw; 3140, m. O(CH2CH2)2NNH~ •SO2t 3290, m; 3225, vw; 3170, m. O(CHzCH2)zNNH2§ 3320, m, br;3225, w; 3140, m. 2750, m, br; 2600, s, br; 2180, m, br. 1280, doublet, m, br:~; 1210, s; 1190, m; 1140, s" 1115, s; 1080, sh; 1075, s. 1265, m; 1250, w; 1150, m; 1120, m; 1095, s; 1060, w. 1285, m, br:~; 1195, s; 1140, w; 1095, vw; 1075, s. 1280, w; 1260, vw; 1235, w; 1180, w; 1160, w; 1120, m; 1090, m; 1050, w. 1300, m; 1250, w; 1215,sh; 1205, s; 1155, w; 1105, s; 1085, s; 1070, sh; 1050, m. 1260, s; 1200, sh; 1180, s; 1105, vs; 1070, s. *Absorptions listed are those not due to solvent or mulling agent, s-strong; m - m e d i u m ; w weak; b r - broad; s h - shoulder; v - very. t F l u o r o l u b e - Mineral oil mull. SThis absorption also appears in mulling agent or solvent, but greater intensity when compound is present. §Capillary film between KBr discs. drazinesulfinates. For the latter, one might consider the structures R2NNHSO2H or R2I~HNHSO2 -. The first of these can be ruled out due to the absence of any identificable O - H absorptions in the i.r. and proton NMR spectra. In addition the compound would be expected to be acidic with proton transfer to unreacted hydrazine. The method of preparation involves bubbling sulfur dioxide into the hydrazine in solution. Thus during part of the reaction there is excess base present, but no hydrazinium hydrazinesulfinate was found. The second structme would indicate that proton transfer occurs intramolecularly. This seems unlikely since the R2N group in the unreacted hydrazine would be expected to be more basic than that in the hydrazinesulfinate. As indicated above, excess hydrazine is present during part of the reaction. 4054 J. M. KANAMUELLER Table 2. Proton NMR data* Concentration (%) r (p.p.m) Assignment (CHa)2NNH 2"SO2 10 (CH3)zNNHz 21 7.43 6.49 7.58 6'89 8"20 multiplet 7.16 triplet 6"67 8"48 multiplet 7.48 triplet 6.48 8.27 multiplet 7.13 multiplet 6.75 8.39 multiplet 7.26 multiplet 6.66 7-19 multiplet 6.88 6.16 multiplet 7.46 multiplet 6.70 6.36 multiplet CH3 NH2 CHa NH2 C-CH2 N-CHz NH~ C-CH2 N-CH2 NH2 C-CHz N-CH2 NH2 C-CH~ N-CHz NH2 N-CH2 N H2 O-CH2 N-CHz NHz O-CH2 Compound (CH2)sNNH2'SO2 (CH~)sNNH2 (CH2)oNNHz'SOz (CH2)6NNH2 O(CH2CH~)2NNH2"SO2 O(CHzCH2)zNNH2 9"0 39 9"5 49 3"0 43 *CDCIz was the solvent for all samples with tetramethylsilane as the internal standard. Since amines are known to form Lewis acid-base adducts, it seems reasonable that similar adducts should form with hydrazines. The hydrazines used potentially contain two basic sites. However, a substituted nitrogen atom, in a hydrazine is known to be more basic than an unsubstituted one [5]. Therefore, it is proposed that the adduct is formed through donation of the unshared electron pair on the R2N group to the sulfur atom in sulfur dioxide. The spectral data in Tables 1 and 2 support this assignment. The N - H stretching vibrations are nearly the same in the adducts as in the hydrazines. Some differences are expected since there are different phases involved. If the nitrogen atom in the NH2 group were the donor atom, the N - H stretching vibrations would be expected to be shifted to lower wave numbers. Included in Table 1 are the observed absorptions in the region 2850-2100 cm -1. In general these are weak, not typical of a quaternary nitrogen atom containing N - H bonds. These absorptions could be due to some intermolecular or intramolecular hydrogen bonding in the solid state. The i.r. spectra of (CH3)2NNHz.SO2 in solution support this, in that they do not show any absorptions in this region not attributable to the solvent. The spectrum of the solid compound however does show a weak absorption in this region. Table 1 also lists the infrared absorptions found in the region 1300-1050 cm -1. 5. P. A. S. Smith, The Chemistry of Open-Chain Nitrogen Compounds Vol. II. p. 126. Benjamin, New York (1966). Sulfur dioxide adducts 4055 The asymmetric and symmetric stretching vibrations due to the SO2 group are generally found in this region and result in strong absorptions [6]. Byrd studied the sulfur dioxide complexes of N,N-dimethylanilines as liquid films. He found the asymmetric stretching vibration at approximately 1280 cm -1 and that for the symmetric stretching vibration at approximately 1120 cm -1 [7]. In deutrochloroform, the adduct (CH3)2NNH2"SO2 shows strong absorptions at 1225 cm -1 and 1090 cm -j and in benzene at 1235 cm -1 and l l 0 0 c m -1. These may be assigned to the asymmetric and symmetric SO2 stretching vibrations respectively. In the spectrum of (CH3)2NNH2"SOz in fluorolube-mineral oil mulls, the absorptions due to these vibrations are found at 1195 cm -~ and 1075 cm -J. The phase difference may be the reason for observing the absorptions at lower wave numbers. In the other adducts prepared, the SOz stretching vibrations are found at approximately the same wavenumbers as those for (CH3)2NNH2"SO2. The absorptions due to the SO2 group then, do appear in the expected region, but are found at lower wavenumbers than those in the compounds studied by Byrd. His compounds, however, contained an aromatic carbon ring system, which may account for the differences between his observations and those reported in this work. The proton N M R spectra (Table 2) do show differences in chemical shift values for the methyl or methylene protons when comparing the adduct with the hydrazine. In the adduct these are always upfield from those in the hydrazine and the N-CH2 protons show the greatest differences in comparison with the hydrazine. Such observations would be expected. One might surmise that smaller differences in the chemical shift values between the hydrazine and adduct would be noted if the donor atom were the nitrogen atom in the NH2 group of the hydrazine. Acknowledgement-This work was supported in part by a Frederick Gardner Cottrell Grant from Research Corporation. 6. L.J. Bellamy, The lnfra-redSpectra of Complex Molecules pp. 360, 404. Wiley, New York (1958). 7. W. E. Byrd, lnorg. Chem. 1,762 (1962).