...!le . . 02.; ~ OYQ. ...l I Do, ' .l uh -’\‘.§ “v.0. all. .. an? . ¢ ‘ n a tilt... . .V . 0 fin It . -. 'O\. p‘ ;._...n.I «3.? .... 1. It‘d: ‘ t '0. In! - o 9:”. 1‘ . .T I aux» . .c t 1.. .0» 0 . .. u u . :H... : . . u .. Va . . .\ C ‘ ..‘I... . l.l| fl .4. T o . ... ... o.‘ .. u I. J\' . II A L n t p I 0;“ r v“ . h S ‘ . V O . \Uboo .0. .0 O... . ’~. I C . o. ’ p .... 05.5.“ I o o . h ‘u . . . DO _ .,o. a . . Of: . . .A .0 Ir. '0 J V.- o 1 n ‘ .. . n a l o a... l a... H ~ OI... \0-0 0 l . .4 .. a .f ....s. r‘ v o'\.I\ ' 0-0“ ‘ u 0" ‘0 ., . .I a... .. b . . .0. I . I b..\.. h s I t o .l .v. o u '0. o ‘ ...- L 6.. (I. . .. a p U ... . . s» . . . r. ‘ o . ‘ ... o . . 6.... I0: — .‘rb ...w n '9. . . u . . . o . d 4 3.... n . . .. 2‘ .. o. . . u u ‘ ....... ..q. r . . . . . . I 5 o n ..U ..."..m Wflb. 4 .o . Q 3 . fi II I | . .. s 09H. t. . 1 9| . o . ¢ . at. .. I. w .m. .. t .v\ ‘3. A v -s . ‘ . . o 4.. 10.. v ..I‘. .8 .v-po. .. ... ‘ 9...; . . .n a a .N In - .c.. c J ‘ NCPIL I. , 1 ..an. . to. I! p . 0‘0: . .- .4 A b or v .. \ o b , f r ‘4. o .d f ....o. K: a‘ .n ‘- ...‘O. ' . . I . ‘Q ‘- \ v I. .‘ ‘0‘ p u o .5 .- . ~ I. ‘0‘ -. t.-. , o I <1 . F. . I. t...“. Vx 0/ Z . no ... ..Z n ' ... o 0.0 . .. ... C . .0- 4 Oil. I .... 5.! Ila: .I £1.37 .. .A 6 ‘u. .‘u... . A . . r.“ . \o.... .w o?» u . I . .ol 5‘ a !. yo.“ "I n '5'. to c t '!713 a... no...- - w to. I I O. a. \O'.. .. a . "(k . v ‘ _. . . . . . . . “a a: .I... ‘ k 109.: 0.....‘ o “ . .. v. '0'... L a. .u 0. s.l ... ~ tlb [kill 7285/63 ABSTRACT THE INTERACTIONS OF DIMETHYLAMINOTETRAFLUOROPHOSPHORANE WITH AMMONIA AND HYDROGEN CHLORIDE: ATTEMPTED SYNTHESIS OF PHOSPHINOFLUOROARSINES, DIMETHYLAMINOPHOSPHINE, AND DIFLUOROAMINOPHOSPHINE By Edward Robert Falardeau The interactions of dimethylaminotetrafluorophosphorane with hydrogen chloride and ammonia were investigated. The product of the reaction in which hydrogen chloride was used was tentatively identified as PF4Cl-NH(CH3)2 by the use of nmr spectral and stoichiometric data. The interaction of PF4N(CH3)2 with ammonia produced diaminotrifluoro- phosphorane which was identified by the use of nmr spectral data. The syntheses of phosphinofluoroarsines were attempted in two ways: first, by allowing phosphine to interact with arsenic trifluoride in the presence of a tertiary amine; second, by allowing arsenic trifluoride to interact with LiAl(PH2)4. Neither method was successful. The synthesis of dimethylaminophosphine by reduction of dimethylaminodichlorophosphine with LiH, LiAlH4, and NaAlH2(OCHZCH20CH3)2 was attempted. There was no reaction when LiH was used. The results obtained from reactions in which LiAlH4 and NaAlH2(OCH2CH20CH3)2 were employed suggest the formation of an aluminium-amine complex. The synthesis of difluoroaminophosphine was attempted by allowing phosphine to interact with tetrafluorohydrazine. The spectral data obtained on the compounds which resulted from this reaction could not be unequivocally interpreted. THE INTERACTIONS OF DIMETHYLAMINOTETRAFLUOROPHOSPHORANE WITH AMMONIA AND HYDROGEN CHLORIDE: ATTEMPTED SYNTHESIS OF PHOSPHINOFLUOROARSINES, DIMETHYLAMINOPHOSPHINE, AND DIFLUOROAMINOPHOSPHINE By Edward Robert Falardeau A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE l97l ACKNOWLEDGMENTS The author is indebted to Dr. Kim Cohn for his guidance and personal interest throughout the course of this research. 11 II. III. TABLE OF CONTENTS INTRODUCTION EXPERIMENTAL A. B. General The Interaction of Dimethylaminotetrafluorophosphorane with Anhydrous Hydrogen Chloride C. The Interaction of Dimethylaminotetrafluorophosphorane with Ammonia The Interaction of LiAl(PH2)4 with Arsenic Trifluoride The Interaction of Phosphine with Arsenic Trifluoride The Interaction of Phosphine with Arsenic Trifluoride in the Presence of Triethylamine G. The Interaction of Phosphine with Arsenic Trifluoride in the Presence of Trimethylamine H. The Interaction of Dimethylaminodifluorophosphine with Lithium Hydride I. The Interaction of Lithium Hydride with Dimethylamino- dichlorophosphine J. The Interaction of Lithium Tetrahydroaluminate with Dimethylaminodichlorophosphine K. The Interaction of Dimethylaminodichlorophosphine with "Redal" L. The Interaction of Tetrafluorohydrazine with Phosphine DISCUSSION A. Reactions of PF4N(CH3)2 with HCl and NH3 l. Reaction with HCl 2. Reaction with NH3 B. Attempted Synthesis of Phosphinofluoroarsines 1. Interaction of LiAl(PH2)4 with AsF 2. Interaction of AsF3, PH3, and NR3 3 iii 10 1O 10 12 13 13 14 18 18 18 20 20 20 20 C. Attempted Synthesis of PH PX N(CH 2 3)2 D. Attempted Preparation of PHZNF BIBLIOGRAPHY 2 iv N(CH3)2 by Reduction of 2 22 23 25 LIST OF TABLES Table l. Reactants, Conditions, and Observations for the Gas Phase Reactions 4 Table 2. Mass Spectral Data of AsF3-N(CH3)3 Reaction Products ll Table 3. Mass Spectral Data of PH3-N2F4 Reaction Products 17 Figure 1. Figure 2. Figure 3. Figure 4. LIST OF FIGURES 1 H NMR of PF4N(CH3)2-HC1 Reaction Products F19 NMR of PF N(CH3)2-HC1 Reaction Products 4 31 P NMR of PF N(CH3)2-HC1 Reaction Products 4 l H1 NMR of Pure "Redal" vi H NMR of PC12N(CH3)2-”Reda1” Reaction Products: 15 INTRODUCTION Many workers have examined the physical and chemical properties of aminophosphines. Some of the results have been rationalized by suggesting that the N-P bond arises from hybridized atomic orbitals forming a sigma molecular orbital. The bond is supplemented by additional pi molecular orbitals which arise from the delocalization of the lone pair of electrons on nitrogen into the empty d orbitals 1’10 which have employed of phosphorous. Although the number of papers multiple bond formation involving d orbitals to rationalize experimental results is large, there have been few systematic attacks in which conclusive evidence as to the consequences and extent of (p+d)n bonding has been obtained. Experiments can not establish whether d orbitals are really used because the bonding models employed are primitive. A (p+d)n bond model is useful, however, because it enables many of the observed properties of aminophosphines to be brought into a coherent scheme. In this thesis a report on the attempted synthesis of a number of compounds containing the N-P or the As-P bond is presented. It was hoped to use the data obtained to establish a descriptive framework which would delineate the extent to which d orbital participation could be used to predict the physical and chemical properties of these compounds. The syntheses which were attempted are: first, the synthesis of the mixed aminodialkylaminofluorophosphorane, aminodimethylaminotrifluorophosphorane, ((CH3)2NPF3NH2), by allowing l 2 dimethylaminotetrafluorophosphorane, PF4N(CH3)2, to interact with ammonia; second, the synthesis of the previously reported chlorotetrafluorophosphorane,28’31’33 PF4C1, by allowing dimethyl- aminotetrafluorophosphorane to interact with hydrogen chloride; third, the synthesis of phosphinofluoroarsines, Ast(PH2)3_x, by allowing LiAl(PH2)4 to interact with arsenic trifluoride or by allowing arsenic trifluoride to interact with phosphine in the presence of a tertiary amine; fourth, the synthesis of dimethylaminophosphine, PH2N(CH3)2, by allowing dimethylaminodihalophosphine, PX2N(CH3)2, to interact with several reducing agents; fifth, preparation of difluoroaminophosphine, PHZNFZ’ by allowing phosphine to interact with tetrafluorohydrazine in the presence of light. Experimental A. General Standard high vacuum techniques were employed throughout. Proton and fluorine nmr spectra were obtained on a Varian Model 56/60 nuclear magnetic spectrometer operating at both ambient and low temperatures. Phosphorous nmr spectra were obtained on a Varian DP-6O nuclear magnetic spectrometer. Tetramethylsilane, fluorotrichloromethane, and trimethoxyphosphine were used as external standards by the use 1, F19, and P3] nmr of the tube interchange technique for the H respectively. The ir spectra were obtained on a Perkin-Elmer 301 spectrophotometer. A gas cell with a 7.5 cm path length and KBr or AgCl windows was employed for volatile samples. Mass spectra were obtained on RMU-6 Hitachi mass spectrometer. The gas phase reactions 28’33 The reaction were carried out in a manner previously described. conditions, the amounts used, and the observations obtained for reactions in which the gas phase was employed are summarized (Table l). B. The Interaction of Dimethylaminotetrafluorophosphorane with Anhydrous Hydrogen Chloride Samples of anhydrous hydrogen chloride (Matheson Co.) were purified by distillation in_yagug_through traps held at —78°, -78°, and -l96°. Samples of PF4N(CH3)2 were purified by distillation ifl_yagug_through traps held at -45°, -78°, and -l96°. The nmr spectral data obtained 32 The vapor 32 on PF4N(CH3)2 was identical to that previously described. pressure at 0° was found to be 42.0 mm (literature value, 44.3 mm). 3 Table l Reactants, Conditions, and Observations for the Gas Phase Reactions Reactants mmol Time Temp Observations PF4N(CH3)2 2.9 immediate formation of 30 sec 23° HCl 5.6 finely divided white solid PF N(CH ) 3.6 clear liquid deposited 4 3 2 . o 5 min 23 NH3 7.6 in droplets on sides of reaction bulb AsF3 3.0 mirror-like brown 12 hr 23° PH3 6.0 coating formed on sides of reaction bulb over a 12 hr period AsF3 3.0 immediate formation N(CHZCH3)3 5.0 20 sec 23° of brownish-white PH3 6.0 cloud AsF3 5.0 immediate formation N(CH3)3 5.1 20 sec 23° of brownish-white cloud, PH3 6.0 brownish droplets were deposited on the walls of the reaction bulb 5 Dimethylaminotetrafluorophosphorane and anhydrous hydrogen chloride were allowed to interact in the gas phase (Table l). The volatile products were distilled ig_vggug_through traps held at 0°, -78°, and -l96°. No volatile materials could be recovered from the 0° or -78° traps. The contents of the -l96° trap were shown to be unreacted HCl with traces of POF3 and SiF4 by the characteristic ir spectra of 30’3] The contents of the -l96° trap were then these compounds. distilled jfl_vagug_through traps held at -78°, ~127°, and -l96°. The contents of the -78° and -127° traps combined was less than 0.1 mmol of volatile materials. The -l96° trap held 2.1 mmol of unreacted HCl. This indicated an approximate lzl reaction of HCl with PF4N(CH3)2. The nonvolatile materials in the reaction bulb were investigated. Approximately 3 ml of acetone was condensed into the bulb. The bulb was removed from the vacuum line and allowed to warm to room temperature. The contents of the bulb were then transferred to a nmr tube. The 1 19 31 sample was a clear yellow liquid. The H , F , and P nmr spectra of the sample were obtained (Figures 1, 2, 3). These results are presented and discussed in a subsequent portion of this paper. C. The Interaction of Dimethylaminotetrafluorophosphorane with Ammonia Ammonia (Matheson Co.) was dryed over sodium prior to use at -60°. Dimethylaminotetrafluorophosphorane was purified as previously described.32 A gas phase reaction (Table 1) resulted in some volatile products. These were distilled jn_vacuo through traps held at -45° and -l96°. All of the volatile products remained in the —45° trap. These products 1 19 were solid at 23°. The H and F nmr spectral data obtained on these products was identical to that previously described for diaminotri- fluor0phosphorane.29 do.m IN fir. __3L_.. m.m IN 1+... a u m.m nus 43m d mdmcxm A. I 22m om vmponxwvmnxnd zmmnfiso: vxoacofim mpaauoca ecwuaaam _u=-~Amzuvz¢aa co mZZ m_a .N atamwa am_uu sag _.mo+ u a _ _ . __ . m. _ : = M g m _. _. . m m u I.“ _ m a; mm N: 0mm 5 a +20.l ppm from P(OCH3)3 31 Figure 3. P NMR of PF4N(CH3)2-HC1 Reaction Products 9 D. The Interaction of LiAl(PH2)4 with Arsenic Trifluoride Arsenic trifluoride was purified by distillation in_vagug_through traps held at -45°, -78°, and -l96°. The AsF3 in the -78° trap was used. The LiAl(PH2)4 was prepared from phosphine (Rocky Mountain Research Co.) and LiAlH4 (Alpha Inorganics).16 Stoichiometry, based on the amount of phosphine recovered, indicated 73% conversion of LiAlH4 to LiAl(PH2)4. This mixture was used without further purification. The reaction of AsF3 and LiAl(PH2)4 was carried out in a 200 ml round bottom flask equipped with magnetic stirrer. The flask was immersed in a 0° temperature bath and attached to the vacuum line. The reaction was monitored with a manometer. A 4 mmol sample of LiAl(PH2)4 (75% LiAl(PH2)4, 27% LiAlH4) in 18 ml of diglyme was placed in the reaction flask. A 20 mmol sample of AsF3 was condensed jn_yagug_at -l96° into the reaction flask. The reactants were allowed to warm to 0° and then stirred for 1 hr. The reaction flask became coated with a brownish yellow solid. The volatile products were distilled ig_yagug_through traps held at -78° and -l96°. The nmr spectral data obtained on the contents of the -l96° trap were identical to those previously described for arsine.38 The sample which remained at -78° was identified as diglyme by comparison of its H1 nmr spectrum with that of an authentic sample 19 of glyme. The F nmr spectra of both the ~78° and -l96° fractions displayed no signals. E. The Interaction of Phosphine with Arsenic Trifluoride The reaction of PH3 and AsF3 was carried out in the gas phase (Table l). The volatile products were distilled ifl_yagug through traps held at -78° and -l96°. The nmr spectral data obtained on the sample held at -78° were identical to that of an authentic sample of 10 AsF The nmr spectral data obtained on the contents of the -196° 3. trap were identical to that of an authentic sample of PH3. F. The Interaction of Phosphine with Arsenic Trifluoride in the Presence of Triethylamine Triethylamine was dried over sodium at 23° for 24 hr prior to use. A gas phase reaction was employed (Table l). The volatile products were distilled jn_yagu9_through traps held at -45°, -78°, and -l96°. The nmr spectral data obtained on the -78° sample was identical to that of authentic samples of AsF3 and N(CHZCH3)3. The nmr spectral data of the -l96° sample was identical to that of an authentic sample of PH3. The nonvolatile material left in the reaction bulb was not investigated. G. The Interaction of Phosphine with Arsenic Trifluoride in the Presence of Trimethylamine Trimethylamine was dried over sodium at -60° for 48 hr prior to use. A gas phase reaction was employed (Table l). The volatile products were distilled in yagug_through traps held at -45°, -78°, and -l96°. The -45° trap held a liquid which exhibited a vapor 19 nmr spectrum consisted of a broad pressure of 8 mm at 23°. The F singlet at +53 ppm from CCl3F. The H1 nmr spectrum consisted of a broad singlet at +1.7 ppm from TMS. The sample melted fairly uniformly from 13° to 15°. The compound was purified by another distillation jg_yggug_through traps held at -45° and -196°. The melting point was found to be 10° to 12°. Anal, Found: C, 23.11, 21.43; H, 5.62, 5.28. The mass spectral data is summarized in Table 2. H. The Interaction of Dimethylaminodifluorgphosphine with Lithium Hydride LiH (Metal Hydrides Inc.) was used without further purification. Dimethylaminodifluorophosphine was prepared and purified as previously Table 2 Mass Spectral Data of AsF3-N(CH3)3 Reaction Products m/e Relative Intensity Assignment 155 1 ? 96 9.1 ? 81 40.5 ? 66 1 ? 59 29 N(CH ) + 3 3 + 58 80 N(CH3)2CH2 56 2 ? 47 8 2 .... 42 20 N(CH3)2 30 15 NCH3+ 15 9 CH T 3 12 20 A 24 mmol sample of LiH was placed in a 200 ml round described. bottom flask which was equipped with a magnetic stirrer. The flask was connected to the vacuum line and a 12 mmol sample of PF2N(CH3)2 were condensed into the flask jg_yggug_at -l96°. The reactants were allowed to warm to 23° and stirred for 3 hr. No evidence of reaction was observed in the flask. The nmr spectral data obtained on the volatile products of the reaction were identical to that of an authentic sample of the reactants, PF2N(CH3)2 and diglyme. I. The Interaction of Lithium Hydride with Dimethylaminodichlorqphosphine Dimethylaminodichlorophosphine was prepared by reaction of 20 It was purified by phosphorous trichloride with dimethylamine. distillation jn_yagug_through traps held at -45° and -196°. The fraction held in the -45° trap was used. A 100 mmol sample of LiH in approximately 10 ml of di-n-butyl ether was placed in a two necked 200 ml round bottom flask. The flask was attached by means of an adapter equipped with a stopcock to the vacuum system. A 50 mmol sample of PC12N(CH3)2 was placed in a 50 ml bulb equipped with a stopcock. The storage bulb was attached to the reaction flask by means of an adapter which was fitted with a sintered glass disc. The reaction flask was evacuated and cooled to -78°. The stopcock which connected the storage bulb and the reaction flask was slowly opened. The PC12N(CH3)2 then dripped through the filter at a slow rate. After 1 hr all of the PC12N(CH3)2 had been added and no visible reaction had taken place. The reaction system was then allowed to warm to room temperature over a period of 2 hr. The nmr spectral data of the volatile materials were identical to that of authentic samples of PC12N(CH3)2 and di-n-butyl ether which indicated that no reaction took place. 13 J. The Interaction of Lithium Tetrahydroaluminate with Dimethylamino- dichlorophosphine LiAlH4 (Alpha Inorganics) was used without further purification. Dimethylaminodichlorophosphine was prepared and purified as previously described.20 The reaction equipment and procedure were the same as those described in section I. After the reaction was allowed to proceed for 1 hr, the volatile products were distilled ifl_vagug_ -through traps held at -78° and -l96°. The nmr spectral data obtained on the volatile products were identical to that of PC12N(CH3)2 and di-n-butyl ether. This reaction was attempted several times with and without solvent and at temperatures ranging from -78° to 23°. In every attempt a yellow nonvolatile solid remained in the reaction flask after the volatile materials were removed. The volatile materials were always identified as PC12N(CH3)2 and solvent. K. The Interaction of Dimethylaminodichlorophosphine with "Redal" "Redal" (Aldrich Chemical Co.) is a 70% solution of sodium dihydro- bis(2-methoxyethoxy)a1uminate, NaAlH2(OCH2CH20CH3)2, in benzene. PC12N(CH3)2 was prepared and purified as previously described.20 A 60 mmol sample of Redal was placed in a 200 ml flask. The flask was attached to the vacuum line and a 57 mmol sample of PC12N(CH3)2 was distilled 1.112292 into the flask. The reaction mixture was allowed to warm to 0° and was stirred constantly. After the mixture warmed a violent reaction took place and a gas was liberated. The gas could not be condensed at -l96°. Yellow solid was present everywhere in the reaction system. The volatile products were distilled jg_vacuo through traps held at -45°, -78°, and -l96°. The nmr spectral data of the contents of 14 the 445° trap were identical to that of authentic samples of benzene and PC12N(CH3)2. The PC12N(CH3)2 was present in extremely small amounts. Due to the violence of the reaction another procedure'was tried.20 A 60 mmol sample of Redal, diluted 50% by volume with toulene, was placed in a dropping funnel which was attached to a 500 ml three necked flask. A 60 mmol sample of PC12N(CH3)2 was placed in the reaction flask and the other two necks were fitted with a stirrer and a dry ice condenser. The Redal-toulene mixture was added dropwise over a period of 2 hr. The nitrogen flow rate through the system was adjusted to about 20 cc/min. When a drop of Redal hit the PC12N(CH3)2, a white cloud formed, and the pressure of the system decreased. The reaction mixture turned yellow and thickened. After the addition of "Redal" was complete, the reaction flask was allowed to warm to room temperature. The dry ice condenser was replaced by a connection to a U-trap held at -78°. The volatile products were carried into the U-trap by the nitrogen stream. The nmr spectral data obtained on the contents of the U-trap were identical to that of authentic samples of benzene and toulene. A large amount of yellow nonvolatile solid remained in the reaction flask. The solid was insoluble in pentane and carbon disulfide. In water it formed a gel. The solid dissolved in methylene chloride to give a clear yellow solution. The H1 nmr of the solid dissolved in CHZCl2 is shown in Figure 4. L. The Interaction of Tetrafluorohydrazine with Phosphine Tetrafluorohydrazine (Air Products) and PH3 (Rocky Mountain Research) were used without further purification. The reaction was carried out in the gas phase in a 500 ml reaction bulb. 15 1 z 1 .a 1 3. 1: 1 :1 A 1 1 11 1‘ {I 1‘ 1 1 :’ 51 1 '3 1 i1 1. ,1 1 .1 ‘11 1'1 11111 1111141111111111111111111 Figure 4. H1 NMR of PC12N(§:3:$;"Redal" Reaction Products: 16 In a typical reaction a 4 mmol sample of phosphine and a 4 mmol sample of tetrafluorohydrazine were condensed jg_vggug_at -l96° into the reaction bulb. The bulb was then removed from the vacuum system and placed behind an explosion shield in a fume hood. The reactants were allowed to warm to 23° and irradiated at intervals with an ultra-violet light source. Total irradiation was 310 min over a period of 50 hr. The inner surface of the reaction bulb turned light green during the course of the reaction. At the end of 50 hr, the reaction bulb was reconnected to the vacuum system and the entire contents transferred to a nmr tube. The H1 nmr, obtained at -lOO°, exhibited two doublets of approximate 2:1 areas. The larger doublet was identified as phosphine by comparison of the chemical shift and coupling constant with an authentic sample of phosphine. The smaller doublet (6 = -0.5 1 0.2 ppm from PH 19 3, JP-H = 177 :_2 Hz) was not nmr spectra consisted of a broad singlet at 21 identified. The F +4.7 ppm from CC13F; N2F4 lit. 4.7 ppm from CC13F. The ir spectra in the gas phase at 20 mm pressure showed the spectra of PH3 and N2F4 superimposed.2] The mass spectral data is summarized in Table 3. Upon freezing the reaction mixture to -l96° a violent explosion accompanied with a loud report occurred. 17 Table 3 Mass Spectrum of PH3-N2F4 Reaction m/e Relative Intensity Assignment 104 1 N254+ 88 3 953+ 85 5 P1121152+ or N253+ 69 4 PF2+ 66 2 PHZNF+ or ~ze+ 53 12 NF2H+ 52 110 NF2+ 47 3 PH2N+ or N21:+ 46 3 PHN+ 34 125 9H3+ 33 180 PH2+, NF+ 32 120 PH+ 31 80 P+ 28 62 N + I‘llill'llllllllll'l Ill DISCUSSION A. Reactions of PF4N(CH3)2 with HCl and NH3 1. Reaction with HCl The reaction between dimethylaminotetrafluorophosphorane and anhydrous hydrogen chloride has been reported to yield chlorotetra— 28,33 fluorophosphorane, PF4Cl, and (CH3)2NH2+C1' as products. In the present investigation the reaction was carried out under identical 28’33 Hydrogen chloride conditions to those previously described. was the only volatile product. The reaction stoichiometry suggested that one mole of HCl reacted with one mole of PF4N(CH3)2. The nonvolatile products of the reaction were investigated by the use of nmr spectroscopy. These data suggest that an adduct, PF4Cl-HN(CH3)2, was formed when HCl and PF4N(CH3)2 were allowed to interact. In the following discussion of the nmr data the methyl protons are referred to as the B protons and the other proton as the A proton. The H1 nmr spectrum (Figure 1) showed a quintet or triplet of doublets of doublets at 6 = +2.5 ppm from TMS. The splitting can be rationalized as the signal split into a doublet, JP-H = 10.5 Hz, each member of which is further split into a doublet, JHA'HB = 5.5 Hz, which in turn is split by the fluorines into a triplet or quintets, JF-HB = 1.6 Hz. A signal that could be attributed to the A proton was not observed. The fact that the samples were dilute and that the A proton was attached to a nucleus with a large quadrapole moment suggests that the signal was too weak and broad to be observed. 18 19 The F19 nmr spectrum (Figure 2) showed a doublet of doublets at 5 = +69.1 ppm from CC13F. The splitting can be rationalized as the signal split into a doublet, JP-F = 796 Hz, each member of which is further split into a doublet, JHA-F = 53 Hz. Splitting due to the methyl protons was not observed. The P31 nmr spectrum showed a quintet each member of which was split into a doublet further split into at least a quintet or a septet with the outer two members unresolved at 6 = +201 ppm from P(0CH3)3. The spectrum can be rationalized as the four equivalent fluorines splitting the signal into quintets, JP-F = 802 Hz, each member of which is further split by the A proton into a doublet, JP-H = 53 Hz, each member of which is split by the B protons into a A septet, JP-H = 10.5 Hz. 8 3] chemical The magnitude of the P-F coupling constant and the P shift are characteristic of tetrafluorophosphorane compounds.32 The interpretation of the nmr data supports the formation of PF4C1-HN(CH3)2. However, these results are equivocal because the compound was impure and attempts to purify the compound failed. Impurities present in the nmr spectra could not be identified but were less than 20% when compared to characterized absorptions. Other workers have shown that the reaction between PF4N(CH3)2 and HC1 does not always yield PF4C1 and (CH3)2NH2+C1'.40’]8 0n the basis of results obtained in this study, it can be suggested that the first step in the reaction may be described by equation 1. (CH3)2NPF4 + HC1 = PF4C1-HN(CH3)2 (1) This adduct will, under conditions not fully understood, spontaneously decompose according to equation 2. _ + - HC1 + PF C1 HN(CH3)2 - PF4C1 + (CH3)2NH2 C1 (2) 4 20 2. Reaction with NH3 The first amino or mixed amino-dialkylamino derivitives of phosphorous pentafluoride were reported in 1970. Roesky and Lustig prepared diaminotrifluorophosphorane, PF3(NH2)2, from the interaction of PF5 29 Their results differed from an earlier investigation of with NH3. the PF5-NH3 system which reported the formation of various solids of general composition xPF5-yNH3 where 5 5_x, y :_1.35 We attempted to prepare a mixed amino-dialkylamino trifluorophosphorane, HZNPF3N(CH3)2, by the interaction of PF4N(CH3)2 and NH3. The nmr spectral data indicated PF3(NH2)2 to be the major product. The formation of PF3(NH2)2 indicates that this compound is more stable than either PF4NH2 or HZNPF3N(CH3)2. B. Attempted Synthesis of Phosphinofluoroarsines 1. Interaction of LiAl(PH2)4 with AsF3 Norman used LiAl(PH2)4 to prepare phosphinosilanes. ‘3 This suggested that LiA1(PH2)4 could interact with AsF3 to produce phosphinofluoroarsines. The only volatile product obtained was arsine. The formation of arsine can be attributed to the interaction of AsF3 with the impurity, LiA1H4, which was present in the reaction system.25 No AsF3 or PH3 remained at the end of the reaction. This indicated that all of the fluorine and phosphorous were tied up as nonvolatile brown solids. If the formation of phosphinofluoroarsines had taken place, they probably would have been volatile. No attempt to identify the solid was made. 2. Interaction of AsF3, PH3 and NR3 The formation of ionic amine-hydrogen halides is a strong driving 34 force in many reactions. Thus, syntheses directed at producing 21 phosphinofluoroarsines by use of reaction mixtures of AsF3, NR3, and PH3 were attempted. This reaction was expected to proceed in a manner similar to that summarized by equation 3. + PH + NR AsF3 3 3 = Ast(PH + (3-x)HNR3+F' (3) 2)3-x The tertiary amines were employed in this reaction to scavenge the HF which would form (equation 4). AsF + PH3 = PHZASFZ + HF (4) 3 It was found that when a mixture of AsF3 and N(CH3)3 were allowed to interact with PH3 for 20 sec in the gas phase, a brownish white cloud formed. When the volatile products were removed by distillation in vacuo. a slightly volatile liquid (vp = 8mm; mp = 10-13) and unreacted PH3 were obtained. Although samples of the liquid obtained in several different runs melted over a constant, relatively narrow, temperature range, attempts to obtain consistent elemental analysis 1 were unsuccessful. The H nmr spectrum of the liquid consisted of a broad singlet at +1.7 ppm from TMS. The F19 nmr spectrum of the liquid consisted of a broad singlet at +53 ppm from CCl3F. The interpretation of the mass spectral data suggested that the predominate ions in the fragmentation pattern arose from N(CH3)3. The liquid can not be unequivocally identified. However, the mass spectral data suggests the presence of a N(CH3)3 group. Also, interpretation of the F19 nmr spectrum indicates fluorine(s) to be in a magnetic environment similar to those of the fluorines in AsF3 (AsF3, broad singlet at 6 = 52.5 ppm from C13F, liquid, broad singlet at 6 = +53 ppm from CCl3F). Since AsF3 and N(CH3)3 do not react in the gas phase, further investigation to ascertain the role of phosphine in the reaction and to establish the identity of the liquid is needed. 22 C. Attempted Synthesis of PH2N(CH3)2 by Reduction of PX2N(CH3)2 Alkyl phosphinesj‘lnz’m 1‘ 14 13,14,15 aryl phosphines, phosphinosilanes, and phosphinogermanes have been prepared and characterized by a number of workers. Recently iodophosphine, the first monohalophosphine known, was prepared.39 However, no preparations for aminophosphines, RZNPHZ’ have been reported. Alkyl and aryl phosphines have been synthesized by the action of the dihydrophosphide ion, PHZ', on 11,12 alkyl or aryl iodides. Various PHZ' donating compounds have been employed. The first PHZ' species used was tetrasodium diphosphide. 11 12.16 Later potassium dihydrophosphide and LiAl(PH2)4 were used. Also, alkyl phosphines have been prepared by reduction of alkyl- ]2 Consideration of these reactions dichlorophosphine with LiAlH4. suggested that dimethylaminodichlorophosphine could be reduced to dimethylaminophosphine by the use of an appropriate reducing agent. No reaction occured between LiH and either PF2N(CH3)2 or PC12N(CH3)2. In the reaction of PC12N(CH3)2 with LiAlH4 the predicted products might be PH2N(CH3)2, or if LiAlH4 broke the P-N bond, PH3. Neither process took place. A yellow nonvolatile solid was the major product. The reduction of PC12N(CH3)2 with "Redal" also produced an unidentified yellow solid. The similarity of the H1 nmr spectrum of the solid dissolved in CH2C12 to the spectra of "Redal" (Figure 4) suggests that an aluminium complex similar to that of "Redal“ was formed. However, in any case the desired reaction to produce PH2N(CH3)2 does not take place when PC12N(CH3)2 and either LiAlH 4 or "Redal" are allowed to interact. 23 D. Attempted Preparation of PHZNF2 Difluoroaminodifluorophosphine, PFZNFZ’ was recently prepared 22 from the gas phase interaction of PFZI and N2F4. Tetrafluorohydrazine 23 is known to form stable NF2 radicals. This property has been exploited by other workers in the preparations of ethyldifluoroamine and methyldifluoroamine.23 Phosphine, when irradiated with uv light, decomposes to red phosphorous and hydrogen. A free radical process has been proposed for this decomposition which suggests that the PH2 radical is the chain carrier. Later studies of the PH3 decomposition and also of the PH3-O2 system support this view.27 The willingness of N2F4 to undergo free radical reactions and the evidence of PH2 radical formation prompted the investigation of the PH3 N2F4 reaction system. It was hoped the formation of PHZNF2 would result. The interpretation of the H1 nmr spectrum of the reaction products indicates proton(s) attached to phosphorous in a compound other than phosphine. This was concluded by the similarity of the coupling constant and chemical shift to those of phosphine. The mass spectral data of the sample can be interpreted in two ways, as indicated by the dual assignments of m/e = 85 and 66 (Table 3). Reported mass spectral data for N2F4 in most cases do not indicate 21 m/e values greater than the NF2+ ion, m/e = 52. However, Colburn reported peaks at m/e = 104, 85 and 66 to be present in low relative abundance.26 The gas phase ir gave the spectrum expected for a mixture of PH3 and N2F4. The violent explosion of a mixture of PH3 and N2F4 when cooled to -196° indicated a definite sensitivity to thermal shock not possessed by either PH3 or N2F4. The interpretation of the H1 nmr spectra and the explosive nature of the mixture indicate the 24 the formation of a new compound. Unfortunately, the difficulty in the interpretation of the mass spectral data, and the equivocal results of the ir leave the identity of the compound open to speculation. Further investigation of the reaction system is required to make any positive identification of products. BIBLIOGRAPHY N d—l—J—l—l—l—J Chm-hum... 17. 18. 19. 20. 21. OSOCDNOSU'l-Pw BIBLIOGRAPHY A. B. Burg and P. J. S1ota, J. Am. Chem. Soc., 80, 1170 (1958). G. Ewart, D. S. Payne, A. L. Porte, and A. P. Lane, J. Am. Chem. Soc., 3984 (1964). R. R. Holmes and R. P. Carter, Inorg. Chem., 2, 1146 (1963). W. A. Hart and H. H. Sisler, Inorg. Chem., 3, 617 (1964). . Schmutzler, Inorg. Chem., 3, 415 (1965). . H. Cowley and R. P. Pinnell, J. Am. Chem. Soc., 87, 4654 (1965). . H. Cowley and M. H. Hnoosh, J. Am. Chem. Soc., 88, 2545 (1966). . Cohn and R. W. Parry, Inorg. Chem., 7, 46 (1968). . F. Nixon and M. D. Sexton, J. Chem. Soc., (A), 1089 (1969). . M. Kren and H. H. Sisler, Inorg. Chem., 9, 836 (1970). . J. Malone and R. W. Parry, Inorg. Chem., 6, 176 (1967). . 0. Norman, J. Am. Chem. Soc., 90, 23 (1968). . Fritz, Z. Anorg. Allgem. Chem., 280, 332 (1955). . E. Drake and W. L. Jolly, Chem. Ind., 1470 (1962). . E. Finholt, Chelling, Vimhof, L. Nielson, and E. Jacobson, R A A K J R W. L. Jolly, Inorganic Synthesis, XI, 124 (1968). L A G J A Inorg. Chem., 2, No. 3 (1963). N. L. Paddock, Nature, 167, 1070 (1951). G. N. Whitesides, Personal Communication. R. Cass and G. E. Coates, J. Chem. Soc., 2347 (1952). J. G. Morse, K. Cohn, R. W. Rudolph, and R. W. Parry, Inorg. §y3,, X, 147 (1967). C. B. Colburn, Advances in Fluorine Chemistry, 3, 92 (1963). 25 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 26 J. Smith and K. Cohn, J. Am. Chem. Soc., 92, 6359 (1970). J. W. Frazer, J. Inorg. Nucl. Chem., 16, 63 (1960). J. P. Freman, A. Kennedy, and C. B. Colburn, J. Am. Chem. Soc., 82, 5304 (1960). N. G. Gaylord, Reduction with Complex Metal Hydrides, (1956). C. B. Colburn and F. A. Johnson, J. Chem. Phys., 33, 1869 (1960). A. H. Cowley, R. P. Pinnell, L. D. Quin, H. Halmann, D. S. Payne, G. Mattin 1y, 0. Talibudren, K. Kenard, Topics in Phosphorous Chemistry, 4, 49-77 11967). R. Rogowski and K. Cohn, Inorgnghem., 7, 2193 (1968). . Lustig and H. W. Roesky, Inorg. Chem., 9, 1289 (1970). . P. Carter, and R. R. Holmes, Inorg, Chem., 4, 738 (1965). . R. Holmes, J. Chem. Phys., 46, no. 10 (1967). M R R D. H. Brown, G. W. Fraser, and D. W. Sharp, J. Chem. Soc., 171 (1966). R. Rogowski, Thesis, Michigan State University (1968). J. Smith and K. Cohn, J. Am. Chem. Soc., 92, 6385 (1970). S. Johnson, Ph.D. Thesis, Purdue University (1953). F . A. Cotton and G. W. Wilkinson, Advanced Inorganic Chemistry, 447 (1966). H. A. Szymanki, Develop. Appl. Spect., l, 227 (1962). E. A. V. Ebsworth and G. M. Sheldrick, Trans. Faraday Soc., 63, 1071 (1967). M. Schmidt and H. J. Schroder, Agnew. Chem., 9, 808 (1970). D. DesMarteau, Personal Communication. ”11111112111111 [1 1111111111111111111“