<<

Reactions of Alkylamino- and Dialkylaminotriphenylphosphonium with and Compounds; Formation of Alkylaminotriphenylphosphonium Polyhalides

Hans Zimmer*, Madhusudan Jayawant, Adel Amer**, and Bruce S. Ault Department of Chemistry, University of Cincinnati, Cincinnati, Ohio 45221, USA

Z. Naturforsch. 38b, 103-107 (1983); received September 20, 1982

Halides, IR Spectra, Raman Spectra

Alkylamino- and cycloalkylaminotriphenylphosphonium halides react with elemental halogens or interhalogen compounds to afford alkylamino- or cycloalkylaminotriphenyl- phosphonium trihalides. The stability of these trihalides depends on the cation as well as the trihalide anion. The assignment of a trihalide structure to these compounds was based on elemental analysis and on IR- and Raman spectroscopic evidence. Most stable are the tribromide and [1X2]° salts. During all reactions involving N-alkylamino- and N-cyclo- alkylaminotriphenylphosphonium halides and elemental halogens an N- of the cation was not observed.

Introduction [Ph3P-N-alk] ®Bre 0 ATtJ >- (2) K2-IN n In a series of papers we demonstrated the syn- Br thetic utility of alkylaminotriphenylphosphonium halides and the corresponding phosphinimines [Ph3P-N-alk] ®Br e [la-d]. It was shown that alkyl- [la] and cyclo- alkyltriphenylphosphinimines could be alkylated NR2 with iodomethane or -ethane to the corresponding strong OHQ e (3) phosphonium salts which upon hydrolysis gave high [Ph3P-N-alk] ®Br yields of secondary amines. Recently this reaction NR2 was extended to synthesize arylalkylamines by Ph3P=0 + R2N-NHalk alkylating aryltriphenylphosphinimines [2]. In or- (R = alkyl) der to further explore the synthetic utility of alkylaminotriphenylphosphonium salts we planned However, instead of N-bromination taking place, to N-brominate these salts in order to obtain the polyhalide formation was observed (eq. (4)). corresponding N-bromo-N-alkylamino-triphenyl- [Ph3P-NHalk] ®Xe + Br2 phosphonium bromides. It was thought that amino- (4) [Ph3P-NHalk]®XBr2® lysis of these salts in analogy to a modified Raschig synthesis [3] would represent a rather convenient While ammonium trihalides have been prepared way to alkylhydrazines via hydrolysis of the anim- and studied in some detail, there have been very ated phosphonium salts (eqs (1-3)) [Id]. few reports of the preparation of quarternary phos- phonium trihalides [4], Also, in earlier reports on [Ph3P-N-alk]©Br® + Br2 - (1) the synthesis of N-alkyl- and N-dialkylaminotri- I phenylphosphonium halides and the corresponding H aryl analogs, no mention of formation of polyhali-

[Ph3P-N-alk]®Br® + HBr des has been made [1, 5-10],

Br Results and Discussion To achieve N-bromination the alkylaminotri- * Reprint requests to Dr. H. Zimmer. phenylphosphonium halides were reacted with ele- * * On study leave from University of Alexandria, Egypt. mental bromide. It was found that tribromides 0340-5087/83/0100-103/$ 01.00/0 derived from N-alkylaminotriphenylphosphonium 104 H. Zimmer et al. • Reactions of Alkylamino- and Dialkylaminotriphenylphosphonium Halides with Halogens

cations form with ease when equivalent quantities showed a positive KI-starch- test. In Table I of salts and elemental are reacted in the synthesized tribromides are compiled. chloroform solution. f-Butylamino-isopropylamino-, For identification purposes we relied on the re- and the unsubstituted aminotriphenylphosphonium sults of the elemental analyses, as well as IR- and cation formed stable tribromides. Raman spectroscopic evidence. In view of the Cycloalkylaminotriphenylphosphonium tribromi- stability of £-butylaminotriphenylphosphonium tri- des were rather unstable and decomposed to a bromide, other trihalides of the /-butylaminotri- certain extent during purification attempts. In the phenylphosphonium cation were synthesized by presence of water, triphenylphosphine oxide and applying the method developed for the tribromide the corresponding alkylammonium bromide were formation. Generally it was found that symmetrical the only isolated products [Id] (eq. (5)). trihalides are more stable than unsymmetrical ones. Thus, when f-butylaminotriphenylphosphonium H20 iodide was reacted with ICI the expected I2C19 salt [(C6H5)3®PNHR]Br3e • (5) was formed initially; but during purification by (C6H5)3P=0 + HaNRHBr successive crystallization from chloroform-ether mixture, the melting point of the salt rose from R= A. 144° to 187 °C, the melting point of pure £-butyl- aminotriphenylphosphonium triiodide. The same The methylamino- and cycloheptylaminotriphenyl- behavior was observed on treating the correspond- phosphonium tribromides could not be purified suf- ing with elemental iodine; the initially ficiently for analysis; both yielded after recrystal- formed Cll2e salt was identical with the first one lization only the corresponding alkylaminotriphe- and during purification attempts it also gave the nylphosphonium bromides, though originally they triiodide. (eq. (6)).

Table I. [(C6H5)3PNHR]©Brö + Br2 [(C6H5)3PNHR]®[Br3]©.

Analysis [%] No. R, Formula Mol.wt. M.p. [°C]a Yield N Calcd Found

1 H Ci8H17Br3NP 518.05 157-158b 84.2 2.70 2.95

c d 2 CH3 Ci9H19Br3NP 532.07 107-108 ' 52.1 -

b 8 C2H5 C20H21Br3NP 546.09 121-122 71.9 2.56 2.60

b 4 ;-C3H7 C21H23Br3NP 560.12 161-162 89.2 2.50 2.57

b 5 2-C4H9 C22H25Br3NP 574.15 146 93.1 2.44 2.52

6 O C21H21Br3NP 558.11 96-97° 65.4 2.51 2.45

C 7 C23H25Br3NP 586.16 130—131 67.7 2.39 O 2.40

8 C24H27Br3NP 600.19 149-150° 63.2 2.33 2.65

c,d 9 C25H29Br3NP 614.21 99-100 52.1 -

a Melting points are uncorrected; b crystallized from CHCls/ether; c crystallized from ethanol/ether; d com- pounds decomposed during purification. 105 H. Zimmer et al. • Reactions of Alkylamino- and Dialkylaminotriphenylphosphonium Halides with Halogens

[(C6H5)3PNHt • C4H9]©I©+IC1 [(C6H5)3PNHtC4H9]®Cl© + IBr n V (6) V (8) X [(C6H5)3PNHt-C4H9]©Cl©+I2 [(C6H5)3PNHt-C4H)]®Br© + IC1 X e recryst. [(C6H5)3PNHt • C4H9]®[ClIBr]

[(C6H5)3PNHt-C4H9]©[ClI2]e —• The polyhalides of the £-butylaminotriphenyl- [(C6H5)3PNHt-C4H9]+[I3-] phosphonium cations which were prepared during If, however, a 1:2 molar ratio of the iodide and IC1 this investigation are listed in Table II. The yield was used, the final product isolated was the in all cases were good to excellent. That trihalide [C1IC1]© trihalide. Its formation could be explained formation is not restricted to the N-£-butylamino- by assuming that the originally formed [CII2]0 anion triphenylphosphonium cation is shown by success- dissociated into Cle and I2; a subsequent reaction ful synthesis of a few trihalides derived of other between the Cle and IC1 led to the isolated N-alkyl- and N,N-dialkylaminotriphenylphosphon-

[(C6H5)3PNHt.C4H9]®[ClICl]e. The formation of ium cations (Table III). All products were stable in this salt is in agreement with the fact that mixed the solid state and could be kept for years provided polyhalide anions with iodine as the central atom moisture was excluded. are generally fairly stable. Our observation about Attempts to obtain the desired N-bromo com- the stability of the trihalides derived of the f-butyl - pounds by reacting the triphenyl4-butylamino- aminotriphenylphosphonium cation parallels the phosphonium brimide with N-bromosuccinimide order of stabilities of alkali trihalides estab- yielded a rather unstable colorless compound which lished by Ephraim [11]. Other mixed trihalides were gave a positive test with KI-starch reagent. At- prepared by reacting halides with interhalogen tempts to purify this compound by crystallization compounds or halogens as illustrated by the follow- or thin-layer chromatography resulted only in ing reactions (eqs (7) and (8)). formation of 5.

Spectroscopical Investigation [(C6H5)3PNHt • C4H9]®I© + Br2 V The infrared spectra, in the region 200-4000 cm-1, [(C6H5)3PNHtC4H9]©Br© + IBr x were similar for all compounds and showed usually -1 [(C6H5)3PNHt-C4H9]©[BrIBr]ö (7) a rather broad peak in the 3240-3350 cm region

Tab. II. [(C6H5)3PNH-f-C4H9]©X© + Y2 -> [(C6H5)3PNH-*-C4H9]®[XY2]©a.

Analyses [%] No. XY2- Formula Mol.wt. M.p. Yield C H N Cl Br I [oCja,b Calcd Calcd Calcd Calcd Calcd Calcd Found Found Found Found Found Found

1 BrCl2 C22H25BrCl2NP 485.23 144-145 85.2 - - 2.89 14.61 16.47 - 2.85 14.37 16.19

2 IC12 C22H25C12INP 532.22 157 84.1 - - 2.63 13.32 23.84 2.69 12.96 _ 23.85

3 Br3 C22H25Br3NP 574.15 146 93.1 46.03 4.39 2.44 - 41.77 - 45.99 4.42 2.52 41.78

4 IBr2 C22H25Br2INP 621.14 160 90.2 - - 2.25 - 25.73 20.43 2.28 25.38 20.34

5 I3 C22H25I3NP 715.14 188-189 95.5 36.95 3.52 1.96 - - - 37.01 3.56 2.27

6 ClIBr C22H25BrClINP 576.68 160 86.7 - - 2.43 6.15 13.86 22.01 2.34 6.08 13.92 22.03

7 Brl2 C22H25BrI2NP 668.14 170 81.4 — — 2.09 — — — 2.06

a Compounds crystallized from CHCl3/ether; b melting points are uncorrected. 106 H. Zimmer et al. • Reactions of Alkylamino- and Dialkylaminotriphenylphosphonium Halides with Halogens

•Ri /RI

Table III. (C6H5)3P-N: X© + Y2 (C6H5)3-PN: [XY2]©Z. \R2 \R2

Analyses [%] [°C]a,b No. RI R2 XY2- Formula Mol.wt. M.p Yield N Calcd Found

1 (CH3)2CH H Brl2 C2iH23BrI2NP 654.11 142- 143B 92.3 2.14 2.12

2 CH3 (CH3)3C IBr2 C23H27Br2INP 635.17 167 90.5 2.20 2.13

3 CH3 (CH3)3C IS C23H27I3NP 729.17 209- 210 93.0 1.92 1.93

a Melting points are uncorrected; 13 crystallized from CHCl3/ether. characteristic of the NH-stretching mode of the responding to the first overtone of the symmetric cation. For the Raman spectroscopic investigation stretching mode, indicating some resonance en- compounds 1,3 (Table II) and [Ph3PNH-t-C4H9]©Br© hancement of the signal. The Raman spectrum of [la] were selected. The Raman spectra of com- the BrCl2e salt showed two weaks lines, at 163 and pounds 1 and 3 (Fig. 1), in the low frequency region 275 cm-1 while the Br© salt showed no Raman lines in the low energy region. The Raman spectra strongly thus support the existence of the trihalide anions in these salts. The Raman spectra also provide some informa- tion as to the local environment of the anion. For Br3©, an intense Raman line was observed for vi. the symmetric stretching mode, while no hint of a line was observed near 190 cm-1, where v3, the anti- symmetric stretching mode should be observed. This mode is Raman inactive [13] if the anion main- tains a center of symmetry, but should be activated if the anion is distinctly perturbed in the crystal. The Raman spectrum of the BrCl2© salt was less intense, and somewhat less definitive, showing lines at 163 and 275 cm-1. The upper line is in good agree- ment with the symmetric stretching mode of the centrosymmetric BrCl2© anion, but again only one vibrational mode is anticipated for this species. The Raman line at 163 cm-1 may indicate that the anion is not centrosymmetric, but distorted or even -centered, BrClCl©; this species should have Raman stretching modes in both of these spectral 100 300 regions. Alternatively, the 163 cm-1 line might be RAMAN SHIFT, cm-1 attributed to a small amount of Br3© impurity. 100-500 cm--1 showed spectra characteristic of each Further support for lack of N-bromination but anion. The Raman spectrum of the Br3e salt was formation of trihalide anions instead during these dominated by an intense line at 167 cm-1, within reactions comes from 31P-investigations of 1 cm-1 of the literature value for the symmetric [Ph3P-NHC(CH3)3]®Br© and compoundsl (Table II). stretching mode of the Br3e anion [12]. In addition, The observed values were d —31.156 and —31.196 a weak line was observed at about 330 cm-1, cor- with H3PO4 as external standard. These values are 107 H. Zimmer et al. • Reactions of Alkylamino- and Dialkylaminotriphenylphosphonium Halides with Halogens well in the range of 31P chemical shifts of amino- phonium tribromides were precipitated directly triphenylphosphonium salts [14], thus showing that from chloroform solutions by addition of dry ether. They were purified by crystallization from chloro- no structural changes in close proximity of the form-ether. P-atom occurred. In case of methyl-, cyclopropyl-, cyclopentyl-, cyclohexyl-, and cycloheptyl-aminotriphenylphos- Experimental phonium tribromides, however, chloroform was Infrared spectra were recorded in Nujol mulls on removed as completely as possible before crystal- a Beckman IR-12 infrared spectrophotometer, while lization either from ethanol or from ethanol-ether Raman spectra were taken of the powdered material mixtures was attempted. Crystallizations of these in capillary tubes. These spectra were recorded on a compounds proceeded very slowly and decomposi- Spex Ramalog spectrometer, after excitation by tion to the corresponding phosphonium bromides or either the 4880 A or the 5145 Ä line of an argon the primary amine hydrobromides plus triphenyl- laser (Coherent Radiation). Slight decomposition of phosphine oxide was extensive. the Br3e and BrCl2e salts was observed when 5145 Ä t - Butylaminotriphenylphosphonium polyhalides; excitation was employed, and considerable decom- general procedure: These polyhalides were synthe- position was noticed when the 4880 Ä line was used. sized according to the above procedure. The follow- The BrCl2e was particularly susceptible to decom- ing halogens or interhalogen compounds were used position, and consequently less intense Raman lines for the reaction with t- butylaminotriphenylphos- were obtained. The 31P NMR spectra were obtained phonium chloride, bromide or iodide: CI2, Br2, I2, with a Bruker HX-90 model in DMSO-d6 solution. IC1, and IBr. Reactions with chlorine were carried Melting points were uncorrected. The micro- out by bubbling a slow stream of dry chlorine gas analyses were done by Galbraith Laboratories, into an ice-cold 10% solution of the appropriate Knoxville, TN. phosphonium halide in chloroform. The amount of chlorine absorbed was measured by gain in weight. Alkyl- and cycloalkylaminotriphenylphosphonium The trichloride decomposed readily. Secondary tribromides- general procedure reactions observed in the case of £-butylaminotri- To a magnetically stirred, ice-cold solution of phenylphosphonium iodide - , cycloalkyl or alkylaminotriphenylphosphonium bro- and t-butylaminotriphenylphosphonium iodide - mide in chloroform an equimolar 10% solution of chlorine reactions were the result of increased bromine in chloroform was added dropwise. Tem- proportions of or interhalogen compounds. perature was not allowed to exceed 10 °C during They are described in the discussion section. addition and an inert atmosphere was maintained N,N-Dialkylamino- and i-propylaminotriphenyl- over the reaction mixture. When the reaction was phosphonium polyhalides Table III: Conditions as over, chloroform was evaporated at temperatures above were used to react N-methyl,N-£-butylamino- below 50 °C and the residue was triturated with dry triphenylphosphonium iodide with bromine or iodine ether. In the cases of low melting solids, an orange and of isopropylaminotriphenylphosphonium bro- viscous oil was obtained after removal of the solvent mide with iodine. To the reaction mixture in chloro- and this was triturated with ether. form, 50 ml dry ether was added to precipitate the Aminotriphenylphosphonium tribromide and corresponding polyhalides, which were filtered off, ethyl-, isopropyl- or £-butylaminotriphenylphos- dried and purified by recrystallization.

[1] a) H. Zimmer and G. J. Singh, Org. Chem. 28, 483 J. Gen. Chem. USSR 32, 2544 (1962). (1963); [6] D. F. Clemens, W. Woodford, E. Dellinger, and b) H. Zimmer and G. Singh, Angew. Chem., Int. Z. Tyndall, Inorg. Chem. 8, 998 (1969). Ed. Engl. 2, 395 (1963); [7] R. Appel, K. Kleinstück, D. Ziehn, and F. Knoll, c) H. Zimmer and G. Singh, J. Org. Chem. 29, Chem. Ber. 103, 3631 (1970). 3412 (1964); [8] R. Appel, Angew. Chem. 85, 863 (1975); Angew. d) H. Zimmer, M. Jayawant, and P. Gutsch, Chem. Int. Ed. Engl. 14, 801 (1970). J. Org. Chem. 35, 2826 (1970). [9] K. Fukui and R. Sudo, Bull. Chem. Soc. Jpn. 43, [2] E. M. Briggs, G. W. Brown, J. Jiricny, and M. F. 1160 (1970). Meidine, Synthesis 1980, 295. [10] H.-J. Cristeau, A. Chene, and H. Christol, Syn- [3] L. F. Audrieth, U. Scheibler, and H. Zimmer, thesis 1980, 551. J. Am. Chem. Soc. 78, 1852 (1956). [11] F. Ephraim, Ber. 50, 1069 (1917). [4] a) G. Wittig and G. Geissler, Justus Liebigs Ann. [12] W. Gabes and H. Gerding, J. Mol. Struct. 14, 267 Chem. 580, 44 (1953); (1972). b) D. Kanai, T. Hashimoto, H. Kitano, and K. [13] K. Nakamoto, "Infrared and Raman Spectra of Fukui, Nippon Kaguku Zasshi 86, 534 (1965); Inorganic and Coordination Compounds" 3rd ed. C. A. 63, 6586 c (1965); Wiley-Interscience, New York 1978. c) G. P. Schiemenz, Chem. Ber. 98, 65 (1965). [14] P. Beck, in G. M. Kosolapoff and L. Maier (eds.): [5] I. N. Zhmurova, A. A. Kisilenko, and A. V. Organic Phosphorus Compounds, Vol. 2, Chapter 4, Kirsanov, Zh. Obschch. Khim. 32, 2580 (1962); Table 16, Wiley-Interscience, New York 1972.