Citation
Studies in phosphorus chemistry

Material Information

Title:
Studies in phosphorus chemistry (I) Reactions of triethylaluminum with some hydrazinophosphines; (II) Reactions of tertiary phosphines with chlorophosphines
Added title page title:
Reactions of triethylaluminum with some hydrazinophosphines
Added title page title:
Reactions of tertiary phosphines with chlorophosphines
Creator:
Spangenberg, Stanley F., 1942-
Publisher:
[s.n.]
Publication Date:
Language:
English
Physical Description:
vii, 127 leaves. : illus. ; 28 cm.

Subjects

Subjects / Keywords:
Phosphorus ( lcsh )
Chemistry thesis Ph. D ( lcsh )
Dissertations, Academic -- Chemistry -- UF ( lcsh )

Notes

Thesis:
Thesis - University of Florida.
Bibliography:
Bibliography: leaves 123-126.
General Note:
Manuscript copy.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
021562313 ( ALEPH )
18114922 ( OCLC )

Downloads

This item has the following downloads:


Full Text

PAGE 1

STUDIES IN PHOSPHORUS CHElVIISTRY: (I) REACTIONS OF TRIETHYLALUMINUM WITH SOME HYDRAZINOPHOSPHINES; (II) REACTIONS OF TERTIARY PHOSPHINES WITH CHLOROPHOSPHINES By STANLEY F. SPANGENBERG A DISSERTATION PRESENTED TO THE GRADUATE COUNCil.. OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1968

PAGE 2

ACKN OWLEDG I'ilil-f l' S The author sincerely thanks his research direc tor, Dr. Harry H. Sisler, for g ivin g guidance and encour agement during the course of this research. Dr. Sisler' s suggestions and discussions have been very valuable in completing this work. The author also thanks the other members of his committee for their valuable assistance. The author is greatly indebted to Dr. Wallace S. Brey and Dr. K. N. Scott for their help in obtaining the phosphorus nuclear magnetic resonance spectra and for their assistance 1n interpreting nuclear magnetic reso nance data. The author also appreciates the help of the mem bers of his research group, Dr. K urt Utvary, Dr. Robert Kren, Dr. s. R. Jain, Dr. Joe Hoffman, Dr. Donald F. Clemens, Dr. Ronald Highsmith, Dr. Hari Prakash, Y~. Larry Briel, Mrs. John Soderstrum, and Miss Catherine Conner. The suggestions and advice of Dr. Larry K. Krannich are especially appreciatea. Acknow!edgment is made to the N ational Scienc e Foundation for partial support of this research throu g h grant number NSF-GP-7863 with the University of Florida. The author expresses his gratitude to his wife, Constance, for her encouragement and especially for her assistance in preparing the fi g ures in this manuscript. ii

PAGE 3

TA BLE OF CO NTEN TS Page ACKNO W L ED GMENTS 1.1. L I ST OF TABLES tv LI S T OF F IGUR E S v C HAPTE R I. THE REACTIONS OF TRIE 1 'f fY LAL UMlN UM W ITH SC} :E HYDJAZINOPBOSPBINES .1 Introduct1.on 1 Expe ri me ntal 3 Discussion 29 Su mma ry 39 II. THE REAC"rIONS 01<., SOMR TER'l'IARY PhOSPHlNES WIT H SO ME C HL O ROPHOSPHINE S 41 Introduction 41 :ri:x!")erir::ental J.~5 Di s cussion 11 1 Sum mar y .. .121 BI B LI OG RA PHY 123 BIOGRAPHI CAL SKETCH 127 111

PAGE 4

LIST OF TABLES Table Page l. INFRARED ABSORPTION DATA. 5 2. NUCLEAR MAGNETIC RESO N ANCE DNI:A 8 3. INFRARED ABSORPTION DAT'A 47 4. NUCLEAR MAGNETIC RESONANCE DATA ,51 5. MASS SPECTRAL DATA FOR (OH 3 P)5 62 6. 3lp NUCLEAR MAGNETIC RESO NAN GE DA1'A FOR THE REACTIONS OF [ ~C H 1 ) 2N]1P WITH PCl3, (CH3) PC12, (C6H.5 ) zPCl, AND (C6H5)PCl2 90 7. SUMMARY OF THE REACTIONS OF SOME TERTIARY PHOSPHINES WITH so~~ CHLOROPHOSPHINES 112 iv

PAGE 5

LIST OF FIG u.RES Figure Page 1. Infrared Spectrum of P[(cH 3 ) N1-I (CH 3 )] 3 P 9 2. NMR Spectrum of P[(cH 3 ) N N(CH 3 )] 3 P 10 3. Infrared Spectrum of (C 6 H 5 ) 2 P[ (CH 3 )1 -::: N(CH 3 )] P(C 6 H 5 ) 2 13 4. NMR Spectrum of ( C6H5) 2P [( CHJ) NN ( CH 3 )] P( C6H5) 2 .14 5. Infrared Spectrum of (C 6 H 5 )P[(CH)NN(C H 3 )] 2 P(C 6 B 5 ) 16 6. NMR Spectrum of (C 6 H5)P[(CH 3 )NN(CH 3 )] 2 P(C 6 H5) 17 . .19 8. NMR Spectrum of P[(cH 3 )NN(CH 3 )] 3 P{Al(C 2 H 5 ) 3 ] 2 20 9. Infrared Spectrum o~ (c 6 H5)P[(cH 3 )NN(CHJ~ 2 P(C 6 H 5 ) [Al(C 2 H5)~ 2 23 10. 11. 12. 13. 14. 15. N MF t Spectrum of (C6H5)P[(CH 3 ) N N(CH 3 )] 2 P(C6H5) {Al(C 2 H5) 2 24 Infrarec Spectrum of (C6H5) 2 P[(cH 3 )N N (CH 3 )] P(C6H5)2[Al(C2H5) Jl 2.27 N !v ffi Spectrum of (C6H5)2P[(CH3)NN(CH3U P(C6 H 5)2[Al(C2H5)~ 2-28 Molecular Hodel of P[(C H ) N N(CH 3 )]_! 30 .32 V

PAGE 6

16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. Infrared Spectru ffi of (C 2 H 5) 3 PC12 55 m m Spectr u m of (C2H5) 3PCl2 56 Infrared Sp e ctrum of (C H 3 P) 5 -~0 NMR Spectrum of (CH 3 P) 5 .......... 61 Infrared Spectrum of [(c 2 H .5) 3 PP(CH 3 ) 2 ] Cl 63 NMR Spectrum of [(C 2 H5) 3 PP(CH) Cl 64 Infrared Spectrum of (c 3 H 7 ) 3 Pcl 2 67 NMR Spectrum of (c 3 H 7 ) 3 Pcl 2 68 Infrared Spectrum of [(C-3H 7 ) 3 PP(CH 3 ) 2 ] Cl 72 NMR Spectrum of [(c 3 H 7 ) 3 PP(cH 3 ) 2 :kc1 73 Infrared Spectrum of [(c 4 H 9 ) 3 PP(CH 3 ) J Cl 76 NMR Spectrum of [(c 4 H 9 ) 3 PP(CH 3 ) 2 ] Cl 77 Infrared Spectrum of [(C4H 9 ) 3 PP(C 2 H5) J Cl 79 NMR Spectrum of [(c 4 H 9 ) 3 P P(C 2 H.5) J Cl 80 Infrared Spectrum of (C8H 17 ) 3 Pcl 2 82 NMR Spectrum of (C 8 H 17 ) 3 Pcl 2 83 Infrared Spectrum of [(CsH 17 ) 3 PP(CH 3 ) 2 1 Cl 87 NMR Spectrum of [(C8Hl 7 ) 3 PP(CH 3 ) 2 ]cl 88 Infrared Spectrum of (CH 3 )2NP(CH3)2~ ( CH 3) 2PC 1 9 1 NMR Spectrum of (CH 3 ) 2 NP{CH 3 ) 2 (CH 3 ) 2 PC1 92 Infrared Spectrum of [(C2H5)3P]2PC1 3 .9 5 NMR Spectra of [ ( C 2H 5) 3 P] 2 PC l 3 9 6 38. Infrared Spectrum of [(c 2 H5) 3 P] 3 Pcl 3 97 39. NMR Spectrum of [(C2H5) 3 P] 3 Pc1 3 98 40. Ni-"i.R Spectra of (C 2 H 5) J~P(C6H5)Cl 2 102 41. Ini'rared Spectrum of (C 2 H.5) 3 PP(CH 3 ) Cl 2 10$ vi

PAGE 7

42. NMR Spectr Ll.n o f (C 2 H5) 3 P P ( CE 3 )cl 2 10 6 43. Infrared Sp e ctrum of (C 4H 9) 3 PP(C H 3 )c1 2 109 44. NHR Spectrum of (C4H 9 ) 3 PP(CH 3 )c1 2 110 vii

PAGE 8

CI-L\Pl'E R I THE REACTIONS OF TRIETHYL A LU MIN U M WITH SO M E HYDRAZINOPBOSPHI NE S Introduction The reactions of organoaluminum compounds with hydrazines, biphosphines, bis(diphenylphos ph ino)amines and aminophosphines have been recently studied in this laboratory (1, 2, 3). These studies have established the existence of several additional compounds in which the aluminum atom is believed to be pentacoordinate. The studies also indicate that the aluminum atom reacts preferentially with the phosphorus atom rather than the nitrogen atom in aminophosphines and bis(diphenylphos phino)amines. These past studies su gg ested to us that the reactions o f triethylaluminum with a series of hydra zinophosphorus compounds mi g ht also provide insight into the manner in which aluminum alkyls react with polybasic group V compounds as well as to provide p roducts in which the aluminum atom is pentacoordinate. The first objec tive of this study was to synthesize a series of struc turally related hydrazinophosphorus compounds. The second objective was to study the reactions of t ri ethylaluminum with these hydrazinophosphorus compounds and to examine the resulting products for the pr esenc e of pentacoordinate aluminum atoms. 1

PAGE 9

2 H i s t or i en 1 B ae k r. ( r o u nc : P e ntacoor di n a t e a l u m i n um h a s b o on ~ cfi n ite ly established by the X-r a y cry s t a l st r u ct ure do t e r mi n a t l on of bis(trimethylamine)alane ( 4 ). T h is co mp o u nd is m nomeric and has a linear N-Al-N axis. T he coordination number of five for aluminum has also been established for the adduct of aluminum hydride with N, N ,N 1 ,N 1 tetramethyl methylenediamine. The structure was a tri g onal bipyramid with the hydrogen atoms in the equatorial plane (5). Pentacoordinate aluminum is beli e ved to exist in adducts of trialkylaluminum compounds with nitro g en biden d ates (6, 7, 8), tetramethyltetrazene (9, 10), and tetrameth y hydrazine (2, 11) as well as i n certa i n a m ine complexes with aluminum hydride (12, 13). T e traphenylbiphosphine reacts with triethylaluminum to form e n ad d uct in which aluminum is postulated to be p ent a coordinate (1, 14); however, tetramethylbi p hosphine for m s the product [(C2H5)3A1] 2 (cH3)2PP(CH 3 ) 2 (1). Bis(di p h e nylphosphino)methylamine and bis(dip h enylphos ph ino)ethylamine ad d ucts with triethylaluminum are also reported to contain penta coordinate aluminum whereas the product (CH 3 ) 2 N P(CH 3 ) 2 Al(C 2 H5) 3 is believed to be a mixture in which approxim a tely 80 per cent of the aluminum atoms are bonded to phosphorus and 20 per cent are bonded to nitro g en (1). The syntheses of hydr a zino p ho s phorus compound s by the reactions of chloro p hos ph in e s with various hydra zine derivatives in the pr e sence of triethyla m in e a re

PAGE 10

3 reported by Nielsen a n d Sisl e r (1 ~ ) a n d Kanamuel ler and Sisler {15). The react i on of h yd r azin o ph os ph or us com pounds with methyl iodi de or chlo ramine r es ults in al ky la tion or chlora mi nation o f the pho sphorus atoms rath e r than the hydrazino nitro g en atoms (15, 16). Experimen-ta1 Manipulation of Materials All organo-aluminum compounds, hydrazines and phosphines were handled in an atmosphere of nitrogen or in an all glass vacuum line. A Vacuum Atmosphere s Model HE-43 inert-atmosphere box equipped with a M odel HE-93B Dri-Train was used for manipulation and stora g e of all reagents, All materials were degassed before re action on the vacuum line by freezing with liquid nitro gen, evacuating the reaction flask, closing the reaction flask stopcock, and allowing the material to warm to room temperature. This procedure was repeated three times for each reaction. Toluene, hexane, diethyl ether, benzene, and pe troleum ether were obtained as rea g ent g rade materials and were dried and stored over calcium hydride. Triethyl aluminum was obtained from the E thyl Corporation. It was fractionally distilled under nitrogen and the fraction boiling at 66 C (0.4 mm) retained. Phosphorus trichloride, obtained from the J. T. Baker Chemical Company, was used as received. Diphenylchlorophos phin e and phenyldichloro

PAGE 11

4 phos p hine were obtai ne d fro m th e Vic ~ or Chemical Com pa ny. These rea g ents were distilled un de r nitro g en and t he frac tions boilin g from 212-214 C (760 mm) an d 110 C (0.5 mm), respectively, were ke p t f or u se 1,2-Dimethylhydra zine dihydrochloride was obtained from the Aldrich Chemi cal Company and used as obtained. Dimethylamine was ob tained from the Matheson Company, Incorporated. Analyses Elemental analyses were done by Galbraith Labora tories, Incorporated, Knoxville, Tennessee, and by Schwarzkopf Microanalyt1cal Laboratory, W oodside, New York. Melting points were obtained using a Thomas-Hoover capillary melting point apparatus and are uncorrected. Infrared Spectra Infrared spectra were recorded usin g a Beckman IR-10 Spectrometer. The spectra of solids were obtained using Kel-F and/or Nujol Mulls of the solids supported between KBr plates. A summary of the infrared data is found in Table 1. Nuclear Ma g netic Resonance Sp ec tr a The proton magnetic resonance spectra were re corded using a Varian Model A-60-A nmr spectrometer. Tetramethylsilane was used as an internal re fe rence when it was possible and as an external reference in all other cases. The phosphorus nuclear magnetic resonance spectra were obtained using a Varian high-resolution spectrometer,

PAGE 12

5 TA BLE la INFRARED ABSO R PTIO N DATA (cm1 ) P [ ( CH 3 ) NN ( CH 3 ) ] 3 P, M ll 11 2860 (m) 2780(m) 1410 (m) 1210(m) 1180(s) 113o(w) l085(s) 1oi5 ( s) 990(w) 910(vs) 760(s) 745(s) 620 (m) 5 O(s) 520(s) 430 (m) (C6H5) 2 P[(cH 3 )NN(CH 3 )] P(C6H5) 2 Mu ll 304o(m) 2960(w) 2905(w) 2 860 ( w ) 27 8o ( w ) 1940(vw ) 1870(vw) 1800 ( vw) 1750 ( vw) 1645(vw) 1575( w) 14 70 ( m ) 1455(sh) 1425(s) 130 5 ( m) 1220 ( m) 1175(m) 1150(s) llOO(sh) l080(s) l060(s) 1020( m ) 990 ( m ) 960 ( m) 910(w) 84o(vw) 7 JO (VS) 690(vs) 615(w) 600(w) 530(s) 490(s) 480(s) 440(s) (C 6 H 5 )P[(cH 3 ) NN (CH 3 )] 2 P(C 6 H 5 ), Mu ll J060(w) 304o(w) 2960 ( m) 2920 (m) 2880 ( m ) 2 850 ( m) 2780(w) 1960(w) 1900(w) 1 8JO(w ) 1735(w) 1675(w) 15eo(w) 1560(w} 1460 (m) 1 4 3o(s) 1 4oo ( w ) lJOO( w ) l215(sh) 1200 (m) 1175(w) 11 4 0 ( m) 1115 ( m) 1090 ( s) 1020 ( m) 995(m) 940(s) 930 (m) 860(w) 745(vs) 710 ( vs) 680(w) 670(sh) 615(w) 590 (m) 570 ( s) 490(s) 430(s) P [( CH 3 ) NN( CH3)] 3P -[Al ( C2H5) J2, N ujol Mull 2780(s) 2700 (m) 1400 (m) 1300(sh) 1220(w) 11 80 (s) 1085 (m) 1040 (m) 970 ( m) 920(s) 770 ( m) 755 ( sh) 715(w) 640(s) 590 ( s) 4 9.5 ( m) 475(sh) (C6H5)P[(cH 3 )N1f(CH 3 )]2P(C6H5)[Al(C2H.5)3]2, N ujol M ull 3090 (w) 3070(w) 2800 (m) 27 30 (w) 1980(vw) 1920 (vw) 1 8 35(vw) 1790 ( vw) 15 80 ( vw) 1 445( s) 1 4 1 5 (m ) 1 30 5 (w) 1235(m) 1210 (m) 11 85 ( rr.) 11 4..5 ( w) 1115(sh) 1105(m) l095(m) 1030 (w) 990 ( sh) 96_5(s) 950 ( s) 920(w) 850(w) 770 ( s) 750(vs) 710 ( s ; 640(s) 625(sh) 610 (m) 570(s) 495(s) 440 (m)

PAGE 13

2910(m) 1J60(vw) 995(w) 73o(s) 510 (m) a 2880 ( m) 1310(w) 96S(w) 69S(s) 465(m) 6 TABLE 1 (continued) 28SO(m) 1225(w) 950 ( m) 64S(s) lli75 (w) 1180(w) 920(sh) 635(sh) lLi-30 (m) 1090(s) 8L1-0 ( vw) 620(s) 1400 ( vw) 1025(m) 7Li-5(s) 540(s) s, strong; m, medium; w, weak; v, very; sh, shoulder.

PAGE 14

7 Model V-4300-2, equipp e d with a fi e ld homo ge n e ity control, ma 8 not insulation a nd field stab i liz e r. The se spectra were recorded at 19.J M c usin g 85 p e r cent pi osp h oric acid as an external reference. A summary of nmr data is found 1n Table 2. This compound was synthesized by the method of Payne, Noth, and Henniger (17) accordin g to the equation > + The infrared and 1 H nmr spectra of P[(cH 3 )NN(CH 3 )] 3 P are shown 1n Figures land 2, respectively. The 3lp nmr spectrum consisted of a single peak at -89.2 p p m relative to 85 per cent phosphoric acid. The tris(di methylamino)phosphine was prepa~ed by the reactiori of phosphorus trichloride with dimethylamine according to standard procedure (18). Preparation of 1,2-Dimethylhydr az i ne 1,2-dimethylhydrazine was prepared by the reac tion of an aqueous solution of 1,2-dimethylhydrazine

PAGE 15

TABLE 2 NUCLEAR MAGNETIC RESONANCE DATA 1 H Chemical Shifts in Values Compound C6H5 NCH 3 CH 3 P [(cH 3 ) NN( CH 3 )] 3P 7.9 (x 7.8) P[(cH 3 )NN(CH 3 )] 3 P[Al(C 2 H5) 3 ] 2 7.95 9.23 ( d 14. 7) (t 7 .6) (c 6 tt 5 )P [(cH 3 )NN(cH 3 )] 2 P(c'6R5) 2.4L1(ave.) 6.96 ( c) ( d 1~ 5) (c 6 tt 5 )P [(cH 3 )NN(CH 3 )] 2 P(C6H5)2.54 (ave.) 7.02 8.75 [Al(C 2 H5)] 2 {c) (d 13.0) ( t 7. ?) (c 6 H 5 ) 2P[(cH 3 )N N (CHJ)] P(C6H5) 2 2.6l(ave.) 7.32 ( C) (x 1.0) (C6H5) 2 P[(cH 3 )NN(CH 3 ~ P(C6H5)2 2.60(av e .) 7.40 R.70 [ A 1 ( C 2 H 5) 3j 2 ( C) ( d 6. 5) (t 7.9) (C6H5) 2P[(CH3)NN(CH3)] P(C6H5) 2 + 7 .t12 8.57 Al2(C2H5)6 (d 5.0) (t 8.0) (c 6 tt 5 ) 2 P[(ctt 3 )NN(CH 3 )] P(C 6 H 5 ) 2 + 7.40 8.78 2 Al 2 (c 2 H.5) 6 (d 6.5) (t 7.7) CH 2 10.37 (q 7.6) 9.86 (q 7.7) 9.65 (q 7.9) 9.57 (q 8.0) 9 .68 (q 7.7) 3lp Shifts in ppm Relative to 85 % H 3 Po4 -89.2 -11.8 -29.9 -79.2 -62.5 -61.2 -61.8 d = doubl e t, t = tripl e t, q = quartet, c = compl e x, x = distort e d tripl e t, () = couplin g con s tant in cps. CD

PAGE 16

3000 2500 2000 1000 500 ( crn :.. 1 ) Fig. 1.--Infrared Sp e ct:rum of P[(C E }NN(CH 3 )] 3 P (Mull).

PAGE 18

11 dihydrochloride with a hot aq u e o us s ol u t i on of s od ium hydroxide. The resultin g m ix tu r e of t h is r ea ct i on w as distilled and the fraction boil i n g f rom 8 29 5 C (7 6 0 mm) collected. This fraction wa s a llow e d to st a n d ov e r sodium hydroxide pellets and t he n r e distill e d. A fin a l distillation was carefully done from calcium hydride un der anhydrous conditions and the fraction boiling from 82-83 C (760 mm) retained and stored under dry nitrogen. The Preoaration of 1 2-Bis{di h e n 1phosphino -1,2-Dimethylhydr a zine This hydrazine was prepared by the reaction of diphenylchlorophosphine with 1,2-dimethylhydrazine using triethylamine as a hydrogen chloride acceptor according to the equation In a typical reaction, 1,2-dimethylhydrazine (2.58 g, 0.0431 mole) was added to a solution of di p henylchloro phosphine (19.l g 0.0861 mole) and triethylamine (10.8 g 0.107 mole) in 200 ml of benzene. The solution was stirred as the hydrazine was added. The resultin g mixture was allowed to stand for 12 hours. 'I'he mixture was th e n filtered and the solvent removed under vacuum to yield 16.81 g of a clear liquid. This liquid was crystallized by dissolving it in boiling hexane and allowin g t h e solu tion to cool. The resulting white crystals were filtered

PAGE 19

12 and after dryin g under vacuu m the y mel ted fro~ 8 284 C. The infrared spectrum obta i n e d u si n g Ke l-F an d N ujol mulls is shown in Fi gu re 3. The 1 H nm r s pe ctru m (Fi g u re 4) was obta i ned u s in g c 6 n6 as a solvent a nd t e tra m ethyl silane as an internal standard. P e ak A (12.61 ave.) is assigned to the ph e nyl protons and p eak B (17.32) to the methyl protons. The I values and couplin g constants are listed in Table 2. The area ratio of peaks A:B is 3.37:l.00 (calculated for (C6H5)2P[(cH 3 ) N N(CH38 P(C6H5)2: 31 3.33:l.00). The P nmr spectrum exhibits a sin g le peak at -62.5 ppm relative to 85 per cent phosphoric acid. Anal. Calculated for (C6H5)2 P (CH3)N N (CH3)P(C6H.5)2: C, 72.88; H, 6.12; P, 14.45; N, 6._5_5. Found: C, 72.82; H, 6.20; P, 14.21; N, 6.41. Yield of the product was 10.09 g of recrystallized material (56% of theory based on the amount of the hydrazine used). The Preparation of 2,),S,6-Tetra me t h y l 2,3,5,6-Tetraaza-l,4-Diphenyl-l, 4 Diphosphacyclohexane The solid, (C 6 H.5)P[(CH 3 ) NN (CH)] 2 P(C 6 H.5), was prepared by the reaction of phenyldichlorophosphine with 1,2-dimethylhydrazine using tri e thylamine as a hydrogen chloride acceptor accordin g to the equati?n 4(C 2 H 5 ) 3 N + 2(C 6 H.5)PC1 2 + 2H(CHJ! NN (CH 3 H CH 3 CHJ /N~N~ (C6H5) p ~N/ /P(C6H5)

PAGE 20

3000 1500 1000

PAGE 21

14 A l B l ~ I

PAGE 22

15 In a typical reaction, a benz e ne solution of ph e n y l dich loro phosphine (35.7 g 0.161 mol e ) and a benz e n e solution of 1,2-dimethylhydrazine {9.10 g 0.149 mol e ) we r e adde d s i m ul taneously over a period of six hours to a stirr e d solution of 100 ml of triethylamine (72.3 g 0.716 mole) in 1200 ml of benzene. The resulting mixture was allowed to stand for 48 hours and then filtered. The solid amine hydrochloride was discarded. The solvent was evaporated under vacuum to yield a mixture of white crystals and yellow oil. The crystals were filtered from the oil and the oil filtrate then dissolved in hot benzene and allowed to cool. Additional white crystals precipitated from this solution. These crystals were filtered and recrystallized from benzene together with the ori g inal crop of crystals. The recrystallized product melted from 222-223 c. The molecular weight determined cryoscopically in benzene was 332 {calculated for (C6H5)P[(cH 3 ) N N(CH 3 )]2P(C6H5): 333). The infrared spectrum is shown in Fi g ure 5. The 1 H nmr spectrum (Figure 6) was obtained usin g CDc1 3 as a solvent. Peak A (1"2.44 ave.) is assi g ned to phenyl protons and peak B ( i' 6.96) is assi g ned to methyl protons. The area ratio of peak A to peak Bis 1.00:1.22 (calculated ratio for (C6H5)P[(cH 3 ) N N(CH 3 )] 2 P(C6H5): 1.00:1.20). The 31 P nmr spectrum of a benzene solution of this ring compound was a single peak at -29.9 ppm relative to 85 per cent phosphoric acid. Anal. Calculated

PAGE 23

-----~ ~----------L..' --------+, -----------.L..--..J 1500 100 0 500 (c m1 ) 3000

PAGE 24

17

PAGE 25

18 P, 18.64; N, 16.86. Found: C, 58.00; H, 6.78; P, _18.46; N, 16.64. The yield of product was 3.77 g of recrystal lized material (15% of theory based on the amount of hydra zine used). The Reaction of 2,1,5,6,7.8-Hex ame th yl 2,3,5,6z7,8-Hex a aza-1,4-Diphos chabic yclo 2.2.2 Octane with Triethylaluminu m Al 2 { c 2 H5) 6 + P [( CH 3 ) NN ( CH 3 )] JP --"'7 P [( CHJ) NN ( CH 3 )] 3P{Al ( C2H5) J 2 In a typical reaction triethylaluminum (0.64 g, 5.6 mmole) was transferred inside the inert-atmosphere box into a double-neck, round bottom, reaction flask containing 5 ml of toluene. The P[(cH 3 )NN(cH 3 fl 3 P (0.66 g, 2.8 mmole) was weighed into a tipping tube side arm. The reaction flask with the side arm attached and fitted with a vacuum stopcock adaptor was transferred to the vacuum line. The solution was then degassed as previously described and finally allowed to warm to -78 C. By tipping the sidearm, the P[(cH 3 )NN(CH 3 )] 3 P was added to the solution and the resulting mixture stirred for two hours as the mixture was allowed to warm to room tempera ture. The solid dissolved as the temperature of the mixture approached room temperature. The toluene was dis tilled out of the reaction flask on the vacuum line leav ing a dry white solid residue. This solid after recrystal lization from hexane melted from 153-155 C. The infra red spectrum is shown in Figure 7 and the peak values a re listed in Table 1. The 1 H nmr (Figure 8) was obtained

PAGE 26

1500 1000 ~o

PAGE 27

A C I\) 0

PAGE 28

21 usin g C6D6 as a solvent. Peak A ( 1'7. 9 5) is assi g ned to the N CH 3 nr oton s, p eak B (19.2 3 ) to t he CH-:i .) p roto ns of the ethyl g rou p and peak C ( r 10.31) to the CH 2 protons of the ethyl g roup. The a r ea ratios of peak s A: B :C are 3.0:3.2:2.0 (calculated for [(C 2 H5) 3 A 1] 2 P[(c H 3 ) NN (CH 3 )] 3 P: 3.0:3.0:2.0). The 31 P nmr spectrum consisted of a sin gle peak at -11.8 ppm relative to 85 per cent ph osphoric acid. Anal. Calculated for [(c 2 H.5) 3 A1] 2 P[(cH 3 )NN(CH 3 )] 3 P: C, 46 .54; H, 10.41; N, 18.09; P, 13.34; Al, 11.62. Found: C, 46.30; H, 10.67; N, 17.81; P, 13.13; Al, 11.87. The yield of product was 1.20 g of unrecrystallized material (92% of theory based on the amount of triethylaluminum used). This reaction was also carried out with 1:1 and 3:l mole ratios of Al(C 2 H.5) 3 to P[(CH 3 )NN(CH 3 )] 3 P. Both reactions resulted in t n e isolation of the 2:1 adduct described above. These results were confirmed by com pa ri son of the 1 H nmr and infrared spectra of the products obtained to the spectra of the 2:1 adduct. The Reaction of 2,,z..2.. 7 6 -Tetr amethyl 2, 3, 5, 6-Tetra a z a-1 )_ : ~ '. c:t e ny l-1 IiDi chos pha cyclohexan 8 wi th Triethylaluminum Al 2 (c 2 H.5) 6 + (C 6 H.5)P[(cH 3 ) NN (CH 3 )] 2 P(c 6 H.5) (C6H5)P[(CH3)NN(CH3)]2P(C6H5) [Al(C2H.5) iJ 2 Triethylaluminum (0.35 g, 3.0 mmole) was trans ferred inside the inert-atmosphere box int o a reaction

PAGE 29

22 flask containing 5 r :l of tolu e ne. Tho (C6H5)P[(c H 3 ) N (cH3)]2P(C6H5) (0.50 g 1.5 rnrnole) was w e i gh ed into a tippin g tube side arm. The reaction fl a sk with attached side arm was closed to the atmosphere with a stopcock adaptor and transferred from the inert-atmosphere box to the vacuum line. The solution was degassed as described previously and allowed to warm to -78 C. The above phos phinohydrazine was added to the triethylaluminum solution at this temperature. The resulting mixture was stirred for two hours as the mixture was allowed to warm to room temperature. The solid dissolved as the temperature of the mixture approached room temperature. The toluene was distilled out of the reaction flask on the vacuum line leaving a white residue. This solid, after recrystalliza tion from hexene, melted from 181-184 C. The infrared spectrum is shown in Figure 9. The 1 H nmr spectrum (Figure 10) was obtained using C6D6 as a solvent and tetramethyl silane as an external reference. This spectrum consisted of a complicated multiplet (peak A, T 2.54 ave.) assigned to c 6 H 5 protons, a doublet (peak B, 1 7.02) assigned to the NCH 3 protons split by phosphorus, & triplet (peak C, 18.75) assigned to the CHJ protons of the ethyl group and a quartet (peak D, 'I 9.86). The area ratios of peaks A:B:C:D are l.OO:l.19:1.78:1.10 (calculated for [(C 2 H5) 3 Al] 2 (C6H5)P[(CH 3 )NN(CH 3 )] 2 P(C 6 H.5): 1.00:1.20:1.80:1.20). The 3lp nmr spectrum consisted of a sin g le peak at -79.2 ppm relative to 85 per cent phosphoric acid. Ana l.

PAGE 30

3000 1500 1000 I f\) \.,.J

PAGE 32

25 Calculated !'or [(C 2 H.5) JA1] 2 (C6H5)P[(CH 3 )NN(CH 3 )] 2 P(C6H5): C, 59.96; H, 9.35; N, 10.00; P, 11.05; Al, 9,64. Found: C, 59.65; H, 9,0J; N, 10.33; P, 11.56; Al, 9,27. The yield of unrecrystallized product was 0.85 g (loo% of theory b9sed on the amount of triethylaluminum used). This reaction was also conducted with 1:1 and J:l mole ratios of Al(C 2 H.5)J to (C6H5)P[(cH 3 )NN(CH 3 )] 2 P(C6H5). Both reactions resulted in the isolation of the 2:1 ad duct described above. Thege results were confirmed by comoarison of infrared and ltt nmr spectra with those of the known compound. The Reaction of l,2-Bis(diphenyl ohosnhino)-1,2-Dimethylhydrazine with Triethylaluminum (C 6 H.5) 2 P[(cH 3 )NN(CH 3 )]P(C6H5) 2 + Al 2 {c 2 H5)6 ( C 6 H .5) 2 P[ ( CH J) NN (CH J} ] P ( C 6 H .5) 2 [ A 1 ( C 2 H .5) 3 ] 2 ) Triethylaluminum (0.53 g, 4.7 mmole) was trans ferred inside the inert-atmosohere box into a reaction flask containing 5 ml of toluene. The (c 6 H 5 ) 2 P[(cH 3 )NN (CH3)]P(C6tt5)2 (1.00 g, 2.33 mmole) was weip;hed into a tipping tube side arm. The assembled reaction flask was then transferred to the vacuum line. The solution was degassed as described oreviously and allowed to warm to -78 C. The phosphinohydrazine was added to the triethyl aluminum solution at this temperature. The resulting mix ture was allowed to stir for two hours as it warmed to room temoerature. The solid dissolved in the solution

PAGE 33

26 as the temperature of the mix t~ re appr o ached room t e mpe r a ture. The toluene was distill ed out o f the r ea ction f lask on the vacuum line l ea vin g a white re s idu e After re crystallization from hexane, thi s co mp o und melted from 79-80 C. The infrared spectrum is s hown in Figure 11. The 1 H nmr spectrum (Figure 12) was obtained using c 6 n 6 as a solvent and tetramethylsilane as an external refer ence. This spectrum consisted of a com p licated multiplet (peak A, 1"' 2.60 ave.) assigned to the c 6 H.5 protons, a triplet (peak B, 1 8.70) assigned to the CH 3 protons of the ethyl group and a quartet {peak D, 1' 9.65) assigned to the CH 2 protons of the ethyl group. The area ratios of peaks A:B:C:D are 3.59:1.00:3.33:2.00 (calculated for ( C 2 H 5) 3 A 1 2 ( C 6 H5 ) 2 P [ ( CH 3 ) NN ( CH 3 ) ] P ( C 6 H .5) 2 : 3 3 3 : 1 0 0 : 3.00:2.00). The 3lp nmr spectrum consisted of a sin g le peak at -61.2 ppm relative to 85 per cent phosphoric acid. Anal. Calculated f or [(c 2 H 5 ) 3 A1] 2 6 H 5 ) 2 P [(cH 3 ) KN( CH 3 )] P(C6H5) 2 : C, 69.49; H, 8.59; N, 4.27; P, 9.43; Al, 8.21. Found: C, 69.90; H, 8.25; N, 4.59; P, 9-44; Al, 8.30. The yield of unrecrystallized material was 1.28 g (85 % of theory based on the amount of phosphinohydrazine used). This reaction was also car rie d out in 1:1 and J:l mole ratio~ of Al(C 2 H 5 ) 3 to (C6H5) 2 P[(cH 3 )NN(CH 3 )]P(C6 H 5) 2 The reaction in which a 3:1 mole ratio was used resulted in the isolation of the 2:1 adduct described above. It was identified by melting point, infrared spectrum and 1 H nmr spectrum. The reaction in which a 1:1 mole rat i o

PAGE 34

...._-~-----/v-__J_ ___ _________ _._ 3 000 1 500 1 000 f \) --.l

PAGE 35

B I I I I I I D Vt I\) CD

PAGE 36

29 was used resulted i ~ the isolation of a cloudy oil. This oil would not cry st a 11 i ze. 'l he 31 ? nmr spectrum of this oil consisted of a single peak at 6 1. 8 ppm rela tive to 85 pe r cent phosphoric a cid (found for (C6H5) 2 [(cH3)NN(CH3)] P(C 6 tt 5 ) 2 : -62.5 ppm; found f or [(C 2 H5) 3 A1] 2 (C6H5)2P[(CH3) :NN (CH3)]P(C6H5) 2: -61.2 ppm). The 1 H nmr spectrum of this oil was similar to that observed for the 2:1 adduct described above exceptin g the area ratios of the peaks. The oil was dissolved in hot hexane and al lowed to cool, however, the material separated again as an oil. Discussio n Molecular models of P[(cH 3 )NN(CH 3 )] 3 P have a cage structure in which the phosphorus atoms are rigidly held in contact or near contact by three hydrazine bridges. The phosphorus lone pair electrons are directed in op posite directions relative to each other and away from the molecule. Although several conformations of this compound are possible, one of the least sterically hindered has the methyl groups staggered (Fi gure 13). This struc ture has diagonal nitrogen lone pairs in spacial proximity. An aluminum alkyl adduct of this compound could result in the formation of pentacoordinate aluminum if a tri alkyl aluminum atom would associate with both nitrogen lone pairs. The preparation of (C6H5)P[(cH 3 )N N (CH 3 )] 2 P(C6H5) illustrates that the general synthe s is of hydrazinophos

PAGE 37

30 w I I H H Top View Side View

PAGE 38

31 phorus compounds ~ y th s reaction of chlo r o o hos p hines with hydrtlzines (14, 15) can be extended to cyclic com pounds. The reactants were added to the reaction mixture simultaneously and slowly in order to minimize the forma tion of polymeric compounds. The low yield of (c 6 H 5 [(cH3)NN(CH3)] 2 P(C 6 H 5 ) probably results from the forma tion of polymeric species. Assuming staggered methyl groups on th~ hydrazino~bridges, there are two possible geometrical isomers of this compound. Figure 14 shows the isomer having cis phenyl groups and Figure 15 repre sents the isomer in which the phenyl groups are tr a ns to each other. The cis isomer (Figure 14) would allow pentacoordination of aluminum by formation of a bicyclic compound having an aluminum bridge between the two phos phorus atoms. This would not be possible with the tr a ns isomer (Figure 15) since the phosphorus lone pairs extend in divergent directions. There are other possible g ometrical isomers of this compound which involve non staggered methyl groups, however, the isomers which seem to have the least steric hindrance are those shown in Figure 14 and Figure 15. The 1 H nmr spectrum of (C 6 H5)P[(CH3)NN(CH3)]2P(C6H5) (Figure 6) exhibits a single doublet assigned to the NCH 3 protons. This fact and the rela tively sharp melting point indicate that this material is probably not a mixture of isomers. The doublet observed for the NCH 3 protons of this compound is in contrast to the triplets observed for the NCH 3 protons in the

PAGE 39

32 Top View Side View Fi g

PAGE 40

33 Top View Side View Fig. 15.--Molec ular Mode l pf th G tr8ns isomer of (C 6 H 5 )P~(CH 3 )KN(CH 3 )] 2 P( C6H5).

PAGE 41

31+ 1 H nmr spectra of P (cn 3 ) NN (CH 3 ) 3 P ( F igure 2) and (C 6 H 5 ) 2 P (CH 3 ) NN (CH 3 ) P (C 6 H 5 ) 2 (Figure 4). This phenome non is discussed below. Molecule1, models of (C 6 H 5 ) 2 P (CH 3 ) NN (CH 3 ) P(C6H5) 2 show that by assuming rotation about the phosphorus-nitro ge n and nitrogun-nitrogen bonds, we can rationalize a single alu m inum atom associating with both phosphorus atoms simultaneously. As noted above a triplet ls ob served for the NCH 3 protons in the 1 H nmr spectrum of (C 6 H 5 ) 2 P (CH 3 )N N (CH 3 ) P(C 6 H 5 ) 2 According to D.S. Payne et~!. (17), the tr1.plet observed in the 1 H nmr spectrum of P (CH 3 ) NN (CH 3 ) 3 P ma y be tentatively attributed to ad ditional phosphorus-proton coupling or to methyl-lone pair interactions and/or methyl protons in several con formations. The me thyl proton resonances of (c 6 H 5 ) 2 (CH3)cH2cH2(cH3}P(C6H5)2 I 2 appear in the 1 H nmr spec trum a s a mu lti p l e t c ons isti ng of e s h a r p doublet e nclos ing a broad d iff 11se peak ( 19) Brophy and Gallagher (19) attribute this triplet nature to virtual coupling in which the CH 3 protons are influenced by both 31 P nuclei. Virtual coupling can be used to explain the NCH 3 triplets obs erv ed in the 1 H mnr spectra of P (cH 3 )NN(cH 3 ) 3 P and (C6H5) 2 P (CH 3 )NN(CH 3 ) P(C6H5) 2 if it is assumed that the phosphorus nuclei are strongly coupled. Virtual coupling in thes~ compounds implies that the NCH protons will act 3 as if they were coupled to both 3lp nuclei. The shape of the triplet depends on the difference in chemical

PAGE 42

3 5 31 shift of these P nuclei. The e ffe c t is a mgx i mum whe n the d ifference in c h e m ic al s hi ft i s zer o ( 20) The st ro n g phosphoru s -phos oh orus co upl i ng c ou l d occur t hrough t he bon d s of th ese co mp o unds or t hrou g h a m e c ha nis m not in volvin g t he b on di n g sy ste m s i n c e in the above c omp o u n ds the phos ph orus atom s a re or co u ld be in s pa c ia l p roxi m ity (21). Alternatively direct couplin g can b e used to ex plain the observed triplets if the couplin g constants of the two Jlp nuclei relative to a g iven me thyl g rou p are nearly the same causing an overla p of peaks res u ltin g in a triplet. Direct coupling, however, would i nvol ve c o up lin g over four bonds whic h is not pro bab ly or co up lin g through a non-bond mechanism which also is no t pr ob a b l e since the NCH protons are not in s p aci a l p ro x i m i t y to 3 the second Jlp nucleus in P[(cH 3 ) NN (CH 3 )] 3 P. I n contr a distinction, the NCH 3 protons of (c 6 H 5 )P[(c H 3 ) NN ( C H 3 D 2 P(c6Hs) are not observed as a triplet in t h e 1 H n mr sp e trum. The doublet observed can be e xp lain e d by t h e fact that the bond an g les may not be a pp ropriate for couplin g of both 3lp nuclei. If this is tru e the N CH 3 p rotons will couple with only one phosphorus nucle u s g ivin g a doublet. The effect that bond an g l e s may ha v e on th e ph nomenon of virtual couplin g has b ee n obs er v ed in t he 1 H nmr spectra of some transition me t a l comp lexe s havi n g phosphine ligands. Virtual couplin g w as observed in tr a ns-square planer, octahed r al and tri g on a l bi p yramidal complexes, but the ph e nomen on was no t found for ci s iso:ners

PAGE 43

36 of these complexes or for compl e x e s of tetr ahed r a l ge ometry (22, 23). If the tr a ns iso me r of (C 6H 5)P[(c H 3 NN(CH3J2P(C6H5) (Fi g ure 15) pr e dom i n a t e s, the p h os ph orus atoms could not approach one anoth e r closely, t h ereby reducing phosphorus-phosphorus cou p lin g and t h e effect of virtual coupling. In general the evidence g iven by D. F. Cle me ns et al. (1) for the existence of pentacoordinate alumi num in aluminum alkyl adducts with phosp h orus-nitro g en compounds consisted of: (1) the stoichiometry of the adducts, (2) equivalence of the phosphorus atoms as shown by 3lp spectra, (3) equivalence of alkyl or aryl groups on phos phorus as shown by 1 H nmr spectra, and (4) absence of the peak commonly assi g ned to tetra coordinate phosphorus at 1120 cm1 to which at least one phenyl group is attached in the infrared spectrum. The reaction of P[(cii 3 ) NN (CH 3 )] 3 P with triethyl aluminum in 1:1, 1:2, and 1:3 mole ratios resulted in isolation of P[(cH 3 )NN(CH)] 3 P{Al(C 2 H5) 3 ] 2 The fact that only a 1:2 adduct was isolated may indicate that the aluminum atoms are bonded to phosp h orus rather than to nitrogen since three equivalent sets of nitro g en lone pairs are available in P[(ctt 3 ) NN (CH 3 )] 3 P. Since bondin g to either nitrogen or phosphorus could result in equiva lent phosphorus nuclei, the sin g le peak observed in the

PAGE 44

37 Jlp n:nr spectrum do e s not i..n c;:i. c at e the coor d in nt i o n n u mber of the alumi n u ~ atom. Tje s in g l e tri p l et o b s e rv e d in nmr sp e ctrum of this a d d uc t ( Figu r e 8) s ugp;e sts that aluminum is bond e d to ph os ph oru s I f a lumin um w er e bonded to nitro g en, equivalent N C H 3 p ro t ons wo uld not be observed unless a ra p id exchan ge of bo ndi r.. g s it es w as oc curring. Since the infrared band at 1120 cm1 is postu lated to be a result of the presence of tetracoordinate phosphorus compounds having a phos phorus-phenyl bond (24), the absence of this band does not imply the absence of tetracoordinate phosphorus. The fact th at a 1:2 ad. duct was isolated even when the mole ratio of reactants was 1:3 may be attributed to steric hindr a nce and/or to electronic considerations. M odels of P[(c H 3 ) N"'N (CH 3 )] 3 P indicate some steric hindrance to attack on nitro g en by the methyl groups is possible. The availability of the nitrogen lone pairs may also be reduced by nitrogen to phosphorus pn' -d11 bonding. Althou gh a 1:2 adduct was isolated when the mole ratio of reactants w as 1:1 or l:J this is not proof that a 1:1 or l:J adduc t did not form, but that the 1:2 adduct was the preferred product for the experimental conditions used. The reaction ~f (C6H5)P[(CH 3 )N N (CH 3 )] 2 P (C6 H 5) with triethylaluminum in 1:1, 1:2, and 1:3 mole r a tios resulted in isolation of (C6H5)P[(C H 3 ) NN (CH 3 )] 2 P(C6H5) [Al(C2H5) ;J 2 Steric co n siderations favor the attack of aluminum on phosphorus in the t r 8ns isomer of the

PAGE 45

38 hydrazinophosphorus compo u n d. Althou g h th e 1:2 stoichi ometry does not support a struct11re in which alumi .n um is pentacoordinate, it does not eli m inate t h at possibility. Equivalence of the 3lp nuclei and the NCH 3 protons as shown by the 3lp nmr and 1 H nmr (Fi g ure 10) spectra, re spectively, do not indicate the bonding in this case. Since the phosphorus starting material has a peak at 1115 cm1 in the infrared spectrum, the shoulder at 1115 cm1 in the infrared spectrum of the 1:2 adduct cannot necessarily be assigned to tetracoordinate phosphorus. Again the isolation of a 1:2 adduct when the mole ratio of hydrazinophosphorus compound to triethylaluminum was l:3 may be attributed to steric and/or electronic con siderations. The reaction of (C6H5) 2 P[(CH 3 )NN(CH 3 )] P(C6H5)2 with triethylaluminum in 1:2 and 1:3 mole ratios resulted in isolation of the 1:2 compound, (C6H5) 2 P[(cH 3 )NN(CH 3 P(C6H5)2[Al(C2H5)3]2. As in the case of the adducts described above, the stoichiometry, 31 P nmr spectrum and 1 H nmr spectrum ( Figure 12) of this adduct do not ind cate its structure. The absence of a peak at 1120 cm1 in the infrared spectrum of the adduct could indicate the absence of tetracoordinate phosphorus (1). The nature of this peak in hydrazinophosphorus compounds is not well determined, however, and therefore its absenc e can not be regarded as proof t h at phosphorus is not tetra coordinate. The reaction of (C6H5)2P[(CH3)NN(CH3)]P(C6H5)2

PAGE 46

39 with triethylalu minum in a 1:1 m ole ratio resulted in isolation of an oil which was not ch a ract e rized as a defi nite compound or mixture. The 31 P nmr spectra of benzene sol u t ions of (Cb H 5 ) 2 P [ ( C H 3 ) NN (CH 3 )] ? ( C 6 H 5) 2 ( C 6 H _s ) 2 P [ ( CH 3) N N (CH 3 )] P(C6H5) 2 {Al(C 2 H5 ) 3 ] 2 and (C6H5)P[(cH 3 ) NN (CH)] P(C 0 H5)2Al(C2H5)3 exhibit sin g le p e a k s at 6 2.5 ppm, -61.2 ppm, a nd 6 1. 8 ppm res pec ti vely T hese data suggest that there is a lar g e degree of dissociation in sol tion since the che, r.i cal shift observed is nearly in de nendent of the concentratio n o f triethylaluminum. All o f the 1 H nmr spectra of benzene solutions having (C6H 5 ) 2 [(cH3)NN(CH3)J P(C 6 H 5 ) 2 to Al(C 2 H5) 3 ratios of 1:1, 1:2, l:3, 1:4, and 1:5 exhibited sin g le peaks for each type of proton. The values for a given g roup of protons wer e nearly independent of the mole ratio of reactants used. These facts support the postu la t ion that the adduct is largely dissociated in solution. Summsrv Two new hydrazinophosphorus co mpounds (C 6 H 5 ) 2 p [( CH3) NN ( CH3)]P( C6H5) 2 and ( C6 H5) P [ (CH3 ) N:,r ( C H3 )] 2P ( c 6H5) were prepared an d were c ha racteriz ed by elementa l analy ses, infrared s p ectra, 1 H n mr spectra, 3 1 p n m r s pe ctra, and molecular wei gh t. Cis and trans isomeric structur es nomenon of virtual cou p ling may explain differences in the 1 H nmr spectra of.' these c c ,i : p ounds. Three new molecul 3 r a ddit io ~ pr~ducts P[(cH 3 )

PAGE 47

40 N N ( CH J)] JP [A 1 ( C 2 H 5 ) J] 2 ( C 6 H 5) P [ ( C H J) NN ( C H J) J 2 P ( C 6 H 5) [Al(C 2 H 5 ) 3 ] 2 ar. d (C 6 H_5 ) 2 P[(CHJ) NN (C H 3 )] P(C 6 H_5) 2 [Al(C 2 E_5 ) 3 ] 2 were prepared and w e re c hara c te rized by e l emen tal ana lyses, infrared spectra, 1 H nmr spectra, and 3lp nmr ~pectra. Th e experim e ntal evidence su gg ests th at th ese adducts are the r e sult of phosphorus-aluminum bonding in which al umi num is tetracoordinate. The 31 P nmr and 1 H nmr spectra of benzene s o 1 u t ions of ( C 6 H 5) 2 P [ ( CH J) NN ( CH J)] P ( C 6 H 5) 2 [Al(C 2 H5) 3 ] 2 suggest this adduct is lar g ely dissociated in solution.

PAGE 48

CHA P TER II THE REACTIONS OF SO ME T ER'r IA R Y P H OSPHI N ES WITH SOME CHLOROPHOSPHINES Introduction It has been known for more than 90 years that tertiary phosphines react with chlorophosphines. However, the early reports did not correctly ch a racterize the oroducts of these reactions. There have been only a few recent studies involving reactions of this type. Our present interest was stimulated by the contrast of products reported in these investigations. The reactions of tri-n butylphosphine with phosphorus trichloride, phenyldichloro phosphine, and diphenylchlorophosphine have been studied in our laboratory (25) and found to give products repre sented in the following equations: 3 ( C 4 H9) 3 P + 2 PC l 3 3 ( C 4 H9) 3 PC 12 : 2 11 ? 11 4(C4H 9 ) 3 P + 4(C6H5)PC1 2 >4(C4H 9 ) 3 Pc1 2 + (C6H5)P 4 (C4H 9 ) JP + 2(C6H:;) 2PCl -> (C4H9) 3PCl2 + (C 0 H5) 2PP(C6H5) 2 On the other hand, t h e reaction of triethylphosphine with diphenylchlorophosphine is reoorted to give [(C6H5) 2 PP {c2H5) .i} Cl (26). The fact that triethylphosphine and tri-n-butylphosphine r c -:;t with diphenylchlorophospbine to give different products was of interest since these phosphines generally have very similar chemical pro ~e rties. 41

PAGE 49

42 A systemati c s tudy of the r eacti o n3 of tertiary phosphines with c hlorophosphines seemed i: .;) o:rtont bc~cause it not only promised to incr ease the basic k n owled g e of phosphorus c hemistry but als o l oad to useful synthet;c routes to phosphorus-phosp h o rus bonded com pounds Such a study also promised n e w insi gh t to possib l e si d e r e actions in the synthe$is o f organ o phosph oru s co mp o u nds by the reaction of a Grignard or or gan olithium rea g ent with a halophosphine. K. D. Berlin et a l. have noted that although the scope of the reactions of tertiary ph os phines with chlorophines is unexplored, it could account in part for the variable yields of tr i alkylphosphines by Grignard or organolithium synth eses (27). The objectives of this systematic study of the reactions of some ~e rtiary phosphines with some chloro phosphines were, therefore, (1) to permit como o risons of reaction pr oducts for various tertiary phos ph ines and chloro ph os phin es, (2) to investi ga te t he p ossibility o f extendin g to new systems th e synthetic me thod of chlorine abstrac tion by tertiary phosphines, and ()) to investigate t he possible relationship of the chlorine abstraction behavior of tertiary phos ph ines to adduct or ohosphonium halide f ormation. Hi s torical Backgro u nd The reaction of phenyldimethylphosphine with phenyldichlorophosphine was re p orted by H. K ohler and

PAGE 50

4 ') ) A. Michaelis in 1877 ( 28 ). T he Drodu ct3 they repor~ are describe d incorrectly ( 25) Althou g h F. C h J l len g er and F. Pritchard reported th a t triphenylphosphine will react with phosphorus trichloride a ccordi n g to the equa tion (C6H.5) 3 P + 2PC1J ---;,i(C6H.5) 3 rc1 2 + P 2 Cl4 ( ?) they isolated only triphenylphosphine oxide and a second product containing phosphorus and chlorine (29). A cording to J. N. Collie, triethylphosphine reacts with phosphorus trichloride or phosphorus oxychloride to form red phosphorus (JO). This author did not report an y other products. It is also report i d by J. N Collie that tri ethylphosphine reacts with a diethyl ether solution of phosphorus oxychloride to form a white soli d that decom poses to yellow phosphorus; however, no characterization of the white solid or experimental evidence for yellow phosphorus was given. The r eacti on of tris(biphen y phosphine with phosphorus trichloride has been examined but only phosphine oxides were isolated (Jl). Trimethyl phosphine has been reported to react w ith liquid phos phorus trichloride at o C and a bromobenzene solution of phosphorus pentachloride at o C with the forma tion of the solid nd ducts [(cH 3 ) 3 P] 2 Pcl 3 and [(CH 3 ) 3 P] 2 PC15, respectively (32). ~-R. Hol me s and E. F Be rtaut stated that the structures of the s e addu cts have not been de termined and that they may possibly be mixtures The following analogous reactions were also reported by R .R.

PAGE 51

L tL ~ Holmes and E. F. B 0 rtaut (33). 3 ( CH J) 3 Sb + 2 PC 1 3 -;;,> 3 ( CH 3 ) 3 S b C 1 2 + 2 P, 5(CH 3 ) 3 sb + 2PC1 5 ----3? 5(CH 3 ) 3 sbC1 2 + 2 P a nd (CH J) 3 As + PC 15 ( CH J) 3 As C 1 2 + P C 1 3 Tertiary phosphines react with ha lod i m e t hy l a rsin e s form ing dimethyl a rsinophosphonium salts accordin. g to the general reaction (34) (CH 3 ) 2 AsX + RR 2 P ----:7 (CH 3 ) 2 AsPR;H X I where X = C,I, R = C2H5, and R = C2H5 or C6H5. Additional reactions of this g eneral type are re p orted by J.M. F. Braddock and G. E. Coates (35). As noted abov e V. W. Seid~l reported the reactions of triethylp h os p h ine with a number of dior g anohalo p hosphines resulting in t h e formation of [(C 2 H 5 ) 3 PP(C 2 H 5 ) 2 ]x where X = Cl,I, (C 2 H 5) 3 FP (C4H9)2Br, and (C 2 H5) 3 PP(C6H5) 2 Cl (26). This author also noted that t h ere is no reaction between triethyl phosphine and diethylaminoethylchloro p hos p hine or between diethylaminodiethylphosphine and diethylchlorophos p hine. Recent work in this laboratory has shown that v a rious haloarsines react with tri-n-butylphosphine according to the equations (36) 2AsX 3 + J(C 4 H 9 ) JP~ 3(C 4 H 9 ) 3 Px 2 + 2As X = Cl,I 2(C6H5) 2 AsX + (C4H 9 ) 3 P ~(C4H 9 ) 3 Px 2 + (C 6 H 5) 2 AsAs(C6H5) 2 X = Cl,I (C6H5)AsCl2 + (C4H9) 3P >[(C6H5) (Cl)AsP( C 4H9) 3Jc1 (CH 3 ) 2 AsI + (c 4 H 9 ) 3 P --:? [(cH 3 ) 2 AsP(c 4 H 9 ) I ( CH J) As x 2 + ( C 4 H 9 ) y ;_:. >[ ( CH J ) ( X) A s P ( C 4 H 9 ) 3 J X X = C l I

PAGE 52

4 5 It has been reported that tris(dim e thylamino) phosphine will r ea ct with dimethylchlor am ine to yield ~cH 3 ) 2 N] 3 Pcl 2 (37, 38). Tris( diethy la mino)phosphine is reported to r ea ct with phosphorus tric hl oride a ccord in g to the equation (39) [(c 2 H 5 ) 2 N] 3 P + PClJ > [(c 2 :n 5 ) 2 1{] 2 PC1 + Qc 2 H 5 ) 2 N ] PC1 2 This paper also reported other analo g ous reactions of aminophosphines with chlorophosphines. E.xoerim.ent a l Manipulation of Mate rials The experimental techniques used were t he same as those described previously. Acetonitrile was obtained as re agen t g rade ma. terial and was distilled from phosphorus (V) oxid e Chloro form was obtained as rea ge nt g rade material and dried over Linde Molecular Sieve Type 4A. Diethyl ether was obtained as anhydrous reagent grade material and was dried and stored over calcium hydride. Triethylphosphine and diethylchlorophosphine were obtained from K & K Labora tories and were used as obtained. Me thyldichloro ph os p hine was obtained from the FMC Company and was distilled a t 79-83 C. Tri-n-butylphosphine was obtained from the Carlisle Chemical Works, Incorporated, and the fraction boilin g at 40 C (0 .05 m.rn) was re tained for use. Tri-n octylphosphine was obtained from the M atheson, Coleman, and Bell Company and used as obtain e d. Deuterochloroform

PAGE 53

46 and hexadeuterob e n zene wer e obtain e d :~ om Sto h l e r Is to p e Chemicals and used as obt s in 2 : Thiophos p hory l Chloride was obtained from Alpha Inor ga nic, Incorporated, and used as receiv ed An a lyses Elemental analyses were done by Galbraith Labora tories, Incorporated, and by Sch wa rzkopf Microanalytical Laboratory. Melting points were obtained usin g a Thomas Hoover capillary melting point apparatus and are uncor rected. Infrared Spectra Infrared spectra were obtained using a Beckman IR-10 Spectrometer. The spectra of solids were obtained using Kel-F and/or Nujol mulls of the solids supported between KBr plates. The spectra of liquids were obtained using a demountable liquid infrared cell with KBr windows and a .05 mm teflon spacer The spectral data are sum marized in Table J. Nuclear M agnetic Resonance Spectr a The proton ma g netic resonance spectra were re corded using a Varian Model A-60-A n m r spectro me ter quipped with a variable temperature probe. Tetramethyl silane was used as an internal stand a rd. The phosphorus nuclear magnetic resonance spectra were obtained using a Varian high resolution s? ectrometer, M odel V-4300-2, equipped with a fi 0:d ho m ogeneity control, ma gne t i~s ula

PAGE 54

1+7 m1 D B T 1? 3 a .... ~ 1.....1' I NFRArtED ABS 0 RPTI01 \/ DATA ( cm -l) (CH3)2NP{CH3)2(CH3)2PCl, Nujol Mul l 1710(vw) 1610(vw) 1315(m) 1305(s) 12 8 0 ( m) 1175(m) l065( m) 990(s) 965 ( m) 950 (m) 89 0 ( m) 770 ( sh) 760(m) 710(w) [( C 2 H 5) 3 P] 2 Pc 1 3 Nujol :Mull 1580(vw) l u. 05 (m) 1280 (m) 1255 (m) 1155(vw) 1 050(s) l0l0( m) 980 ( sh) 880(vw ) 775(s) 7S5(s h ) 720(sh ) 640 (m) 625(sh) 540(w) 520(w) soo(m) 400 (m) [( C 2 H _5) JP] J PC 1 3 M ull 2970(w) 2 880 ( m ) 2 80 0 ( sh) 1 4 6,S(sh) 1h50 ( 1.-:r) 1405(w) 1375(w) 127S(sh) 1260(m) 1 050 ( s) 10 3S(s) 1 015 ( sh ) 1000 ( sh ) 77 5 ( s) 7 so ( s) 710 (m ) 685(vw) / 75 I 0 \ vw J 650 ( m) 625(m) 540 ( vw) 400 ( vw) 520 ( vw) 490(s) 43o(vw) [(C2H5) 3PP(cH 3 ) (Cl)] Cl, Nuj ol Mu ll 1410 (m) 1275 ( w) 12 1.!. 0 (w) 10 _5 0 ( sh ) l 0h0( s) l0l0(sh) 980(sh) 89 0 ( sh) 880 ( m) 870(m) 76_S ( m) 7.50 (m) 735(s) 710 ( sh) 665(w ) 62 5(w) 61_5(w) (C4H9)3P + ( CHJ) PCl2 Neat 2960(s) 2920(s) 2 86 0 ( m) ll.i60(s) l~ 00 (s) 1 38 0 (s ) 134o(w) 1305(w) 12 T~, ) 1220( sh) 1210 ( m) 11 8_5 ( m) I_::; \ W 1155(vw) 1090(s) 107 0 ( sh ) l 050(sh) 990 ( m) 960 ( m) 910(sh) 880(s) 795(m) 770 ( sh) 715(s) 670 ( m ) 46 0(w) 440 ( sh) ( C 2 H 5) J PC 1 2 11, ~ 7 1 ...... 1..A...i.. 29 8 0 ( m) 29 50 ( sh ) 2890(vs) 2 810 ( sh) 1 4 6 o(s) I ~ j ( ) f I' -,; <:t ... ... ~-... i.J 1405(m) 13 90 ( m) 1280 (m ) 12 50 ( m) 1 1 5 0 ( v w) /"' ,.. 0 ( l. V )' S) 1005 ( m) 9 80 ( m) 77 5( v s ) 7 .S S (sh ) 7 3o(s 1 1.) 720 (mo 640 ( s) 5 40( w) 520 ( m) 50 0 ( m) 400 (m)

PAGE 55

48 T ABLE 3 ( c onti nued ) [ (cH 3 ) P] s~ Nea t 29 6 0(s) 292 0 ( m ) 2900(s) 2 2!.J. 0 (w ) 2 80 0 ( m) 1 440 ( sh) 141o(s) 1275 ( s) 860 ( s) 6 8 0 ( sh ) 665 ( s) [(C 2 H .5) 3 PP(CH 3 ) 2 ] Cl, Nu jol 1 1 ull 1 4 25 (m) 12 90(w ) 1275( m) 12 6 5(m). 12 u.0 (w ) 1060 ( sh) 1045(s) l0 00 (v w ) 9 5 5(m) 90 5( s) 7 .:~ O ( s) 760(s) 720 (m) 705( w ) 6 80 ( w) 660 ( m ) 620 {w ) 450(vw) ( CJ H ?) J PC 1 2 Mul l 29 6 0(s) 2 930 (m) 2 8 70 ( s) 27 8 0 ( s h) 1 46 5 ( s) 1 4 55 ( sh ) 1 4 0 5(w ) 1 375(m ) 13 4 0 (w) 1305(w) :;-2 6 .S ( w) 12 h0 ( sh ) 122_5 ( s ) 10 80 ( sh ) l 070(s) 1 045 ( m ) l0J5(sh) 9 8 5(w) 960 ( s h) 900 ( m ) 850 (m ) 84 0 ( sh) 820 (sh) 7 8 0 ( m ) 7JS(s) 710(sh) 650(vw ) 560 ( sh ) 530 ( m ) 510 (s) 4 70( s h ) 455(w ) [(c 3 H 7 ) 3 PP( C H 3 ) Cl Neat 29 6 0(s) 2 9 J 0 ( sh ) 286 0 (s) 2800 ( sh) l h60 ( s) 1 h15 ( s ) 1375( m ) 1 34o(w ) 1 300 (w) 1 23 5(m) 11 4 5( vw) li lO(m ) 1075(s) l 0J 0(s h ) 9S0 ( m ) 905(s) 850 ( :n) 775(sh) 760 (m) 720(s) 705 (s h) 655(w) 50 0( vw ) 48o( v w ) [(c 4 H 9 ) 3 PP (C HJ) 2 ] Cl Neat 29 6 0(vs) 2 9J 0 ( v s ) 2 900 ( sh ) 2870 (v s) 2800 ( sh) 1 455 (s) 14 4 5(sh) 1 425 ( sh ) 1 4 15 (m) 138o(m ) lJ L~ 0 ( vw ) 12 90( w) 1225(w) lll 0(sh ) 1 090 ( m ) 1050(sh ) 995(w ) 955(m ) 900 ( s) S oo (w) 715 ( m) [(c 4 H 9 ) 3 : C 2 H _5) 2 ]Cl, Neat 2 960 (vs) 2 9 3 0 ( vs ) 2 87o ( vs ) 2 800 ( s h) 1 46 0(s) 1 41o(s ) 1 38 o(s) 13 4 o( w) 1 310 ( sh ) 1~ c; 1 ) c:. j \ s n. 12 80(w) 123o( m ) 121 5( m ) 11 90(sh ) 11 6 0 (m ) l090(s) l Oh O ( m ) 1 020 ( m) 1000 (m) 960 ( m ) 9 05 (;:, ) 790 ( sh ) 750(sh ) 7 20 ( s) 66 0 ( m ) 620 ( sh ) 505 (w) 460 (m )

PAGE 56

29 6 0(s) 1295(w) 1095(m) 2960(sh) 14 75 ( sh) 1225 ( m) 1230 ( sh) a 292 0 (s) 12 6 0 (w) 1025(m) 710 ( s) 2920(s) 1460 ( s) 1 205 ( sh) 995(w) 49 TABLE 3 (continu e d) (C sH 1 7 ) 3 Pc l 2 Mul l 28_50(s) 2780(sh ) 122 0 (m) 1 190(w) l0 00( sh) 920(w) 650(sh) 540(sh) 2870 ( s h) 1 410 \ s ) 11 9 0 (sh ) 955 (m) 675(s) 2 850 ( s) 13 80 ( :n) 1155(w) 89_5(s ) 4 30 ( vw) 146o(m) 11 60 (sh) 840 ( m) 515(s) 2800 ( sh) 13 !1 0 (w) 10 9 0 ( s) 790 ( m) 1375(w) 110_5(sh) 80 0 ( m) 1820 ( V1.-l ) l290(w) 10 1..i.o ( sh ) 7i5{s) s, stron g ; m, med i um; w, weak ; v very; sh, s ~o ulder.

PAGE 57

50 tion and fi e l d s t sb iliz e r. The se spect r a were r e c orde d at 19.J Mc usin g 85 per cent phosnhor ic a c id n ~ a n e ternal reference. A s urr.mary of n mr data i s found in Tabl e 4. :Mas s S p e ctr a The ms ss spectra were obt ained usin g a Hitachi Perkin-Elm e r RMU-6E Ma ss Spectrom e ter run at an ioniz ing voltage of 70 ev. The Preoar a tion of Bis(dimethyl amin chloroohosphine Bis(dimethylamino)chloro ph osphin e was prepa red by the reaction of dimethyla m ine with phosphorus tr i~ ~ lor ide according to the method of E. M Evle th, Jr., L. D. Freeman, and R. I. Wagner (40) The pr o duc t was distilled from 90-92 C (34-36 mm) (published boilin g point: 93970 C (47-49 mm) (40)). The 1 H nmr ar:d infrared spectra were identical with those of a kno~n s ample The Preoaration of Tri-n-oropyloh osohine Tri-n-propylphosphine was prepared by the reaction of n-propyl magnesium bromide with phosnhorus trichloride in a manner analogous to that r ep ort e d by W C Dav ies and W. J. Jon e s ( 4:) and by W C. Davies, P. L. Pearse, and W. J. Jones (42). The Gri gnar d re age nt was prepared by the standard method. A diether ether solution of phosphorus trichlorida was e dded slowly to t he ether s o lution of n-propyl magnesium bro ~:d e which w as being

PAGE 58

TABLE 4 NUCLEAR MAGN ETIC RESONANCE DATA __ ., 1 H Chemic al Shifts in Valu es 3lp Shifts in ppm Relativ e t o Com pou n d -x B C D 85 fo H 3 Po 4 ( G 2 H5) 3 1 c 1 2 6.611 ( ave .) 8 56(ave. ) -11).j ( qq ) (t t ) ( HH 7.~) ( HH 7.4) (p H 9.0) ( pH 22. 5 ) ( C 3 H? ) 3 PC 1 2 6.72( ave .) 8.23 ( ave. ) 8.Bl+ ( avc .) -103 \.ft ( C) (c) (c) I-' ( C5H17 ) 3PCl2 6.78 ( av e ) 8.03 to 9.39 -103 ( C) (c) [(c 2 B 5 ) 3 P P ( CH 3 ) ; ]Cl 7.37(a ve .) 8.60(ave. ) +1+2. 2 a 1:1d 33. 7 { C) ( C ) [(c 3 H 7 ) 3 rP(CH 3 ) iJ Cl 7 .1~8( av e ) R 31 8.82( ave .) +47.9 and -22 .s (c) ( c ) ( C) [( C4H 9 ) 3 PP ( CH 3 ) 2] C 1 7. 115 ( av e ) 8.31 9.0l( e v c .) +51.0 and -21.2 (c) (c) { c ) [< caH17 > 3PP(C H ) 2 Jc1 8 .L ~ O to 9. 21 +25.9 an d llt. 5 ( C) [(cl 1 H 9 ) 3 PI)(C 2 H.5) 2 ] Cl 8.33 to 9.29 ( c )

PAGE 59

TABLE 4 (continued) 1 H Chemical Shifts in Values Compou n d A B C D -------------------------L(c 2 H 5) 3 PJ 2 rc1 3 ( -2 0c) [(C2H5)3J:j 2PC l3 (250c) 6.42 to 9.06 6.64(av e .) ?.52( avo .) (qq) (c) ( HH 7. S) (pH 10.0) 6.62(ave.) 7.6l( ave. ) (c) {c) [( C 2 H _s ) 3 I'P ( C 6H 5 ) CJ] C 1 2. 0 6 ( ave ) ( 25c after standing(c ) 6.63(av c .) (c) 12 hOUl'S ) [sc 2 n 5 ) 3 PP(cH 3 )c1Jc1 r 'C q \ 1 1 1 -' ( ( IT )0 1 7 0] 4 L 9 3 ,[ '3 ,_, U v ( crr 3 ) 2 NP (C H 3 ~. (C H 3 ) 2 Cl ?. 56 ( avo. ) ( C) ?.69( avc. ) 7. ll+ < a 10. 5) 8.62 ( ave .) (c) 8 5)+ ( f We ) 7.6 6 (d lJ.6) 8.56 (tt) (HH 7.7) { pH 22 .5) 8.6J(ave.) (c) ?.JB( a ve.) 8.73(av0.) (c) (c) 9.1 2 ( ave .) 31 P Shift s 1.n ppm Relati.ve to 85 '.'fo H3P04 +Hf.7 and 1 6.S -6 6 .L t c = c omplex, qq = do ub le set of qu ar t ets tt -double s et of t rip lets, HH = i> l"O to n--protcm cou p lin ['; pH = p h osp h-1r us-p :!.' oton cou p lin e; \J 1. I\)

PAGE 60

53 stirred in an ice bath Afte r the complet ion o f thi s addition, tho rea ction m i xture was r e f lux ed for two hours. A satur ate d aqueous so lu tion of ammonium chloride whi ch had been previously boiled while passi n g ni tro g en throu g h it was slowly added to the r ea ction mix ture. The ether layer was decanted from th e resulting aq ueous layer and dried over calcium sulfate in a nitro g en atmosphere. The ether solution was then distilled. The product boiled from 183-185 C (760 mm) (published boiling point: 187.5 C (760 mm) (42)). The Preparation of Tetramethyldiohosphine Disulfide T~tramethyldiphosphine disulfide was prepared by the reaction of thiophosphoryl chloride with methyl magnesium bromide analogous to the procedure of H. Niebergall and B. Lagenfeld (43) and S. Frazier (44). The thiophos phoryl chloride was slowly added to a diethyl ether solu tion of methyl magnesium bromide which was being stirred in an ice bath. After the thiophosphoryl chloride had been added, the resulting mixture was allowed to stand 12 hours at room temperature. A 10 per cent solution of sulfuric acid was slowly added to the reaction mixture which was cooled with an ice bath. The insoluble product was washed with water, methanol, and finally diethylether and then dried under vacuum. The Preparation of Dimethvlchlorophosohine Dimethylchlorophosphine was prepared by the reaction

PAGE 61

54 of tetra ~ o th yld i ~ ~ osphi ne with phcny:d ~chl oro ph o sph in e accordin g to t ho procedure of G. W. Pa ~ sha ll ( 45) The product boiled from 69-75 C (760 mm ) (pu blishe d value: 77 C (760 mm) ( 4 5)) The Reacti on of Triethylnhosnhine with Phosphorus Trichloride 2PC 1 3 + 3P( C 2 H5) 3 ----3 2 11 P 11 + 3 ( c 2 H5) 3 Pcl 2 Triethylphosphine (0.50 g, 4.2 mmole) was added to a solution of phosphorus trichloride (0.39 g, 2.8 mmole) in 20 ml of benzene. A yellow solid precipitated from the resulting solution. This solid became more orange-red in color upon standing. I t was filtered, washed with boiling acetonitrile, filtered a g ain and dried under vacuum. This orange-red material did not melt up to 330 C; O' it reacted violently with concentrated nitric acid. Anal Calculated for red phosphorus: P, 100. Found: P, 88.79; Cl, .45; C, 7.25; H, 1.37. The benzene filtrate combined with the acetonitrile washings was evaporated to dryness. A white residue remained that after recrystallization from toluene melted from 242-247 C with decomposition (literature value for (C 2 H 5 ) 3 Pcl 2 : 240-250 C with de composition) (46). The infrared spectrum of this compound is shown in Figure 16. 1 The H nmr spectrum shown in Figure 17 was obtained using CDC1 3 as a solvent. It consists of a multiplet (peak A, 1' 6.64) that may be two overlapping quartets and a multiplet (peak B, 1' 8.56) that appears to be two sets of triplets. The ar G a ratio

PAGE 62

3000 1500 1000 Fig. 1 ( .--Infrared Spectrum of (c 2 tt 5 ) 3 Pcl 2 (Mull). \ Jl. \J\

PAGE 63

A B

PAGE 64

57 of A:B is: 1.00:1.57 (calculated for (C 2 H 5 ) 3 Pcl 2 : 1.00: 150). The 3lp nmr spectrum exhibited a sin g le peak at -114 ppm in acetonitrile. Anal. Calculated for (C 2 H5) 3 PC12: C, 38.12; H, 8.00; P, 16.38; Cl, 37.50. Found: C, 37.92; H, 8.25; P, 16.13; Cl, 38.17. The yield of oran g e in soluble material was 0.08 g (89 % of theory based on t h e amount of phosphorus trichloride used assumin g this prod uct to be phosphorus). The Reaction of Triethyl nh osphin e with Phenyldichloroohosphine Triethylphosphine (0.50 g, 4.2 fil~ole) was added to phenyldichlorophosphine (0.76 g, 4.2 mmole). An exo thermic reaction followed resulting in the formation of a solid mixture within a few seconds. This solid was carefully washed with cold (-5 C) acetonitrile; the in soluble material was filtered, dried under vacuum, and recrystallized from acetonitrile. It melted from 1531560 C (literature value for tetraphenylcyclotetraphos phine: 154-5-156 C) (47). The infrared spectrum of this insoluble solid is identical with that published for tetraphenylcyclotetraphosphine (48). The 31 P nmr spec trum exhibited a single peak at +2.8 ppm using hexane as a solvent (literature values for tetraphenylcyclotetra phosphine are +9 (49) and +4.6 (50)). The yield of tetra phenylcyclotetraphosphine was 0.37 g (80% of theory based' on the amount of phenyldichlorophosphine used). The ace

PAGE 65

58 tonitrile filtrate w as eva p orat ed l ea vin g a white s o lid. This material exhibited a m e ltin g ran g e of 240-250 C with d e composition (literature v ~ lue for (C 2 H5) 3 Pc1 2 : 240-250 C) (46). The infrared and 1 H nmr s pe ctra were identical with the correspondin g spectra of (C 2 H 5 ) 3 Pcl 2 shown above in Fi g ures 16 and 17, res p ectively. The 3lp nmr spectrum consisted of a single peak at -111 ppm in acetonitrile. The Reaction of Triethylohosphine with Methyldichloronhosphine 5(CH 3 )PC1 2 + 5(C 2 H5) 3 P--7(CH 3 )P5 + 5(C 2 H5) 3 Pcl 2 Triethylphos p hine (3.02 g 25.6 mmoles) was added to methyldichlorophos p hine (3.00 g 2.56 mmole). An exothermic reaction resulted in the formation of a solid over a period of two to three minutes. The solid was washed with cold (-5 C) hexane, filtered and purified by sublimation. This insoluble material exhibited a melting range of 239-245 C (literature for (c 2 H 5 ) 3 Pcl 2 : 240-250 C) (46). The infrared and 1 H nmr spectra were identical with the correspondin g spectra of (C 2 H 5 ) 3 Pcl 2 shown in Figures 16 and 17, respectively. The 3lp nmr soectrum consisted of major p eak at -114 ppm and a small peak at -37 ppm ( p robably an impurity). The yield of (C 2 H5) 3 Pc1 2 was 3.91 g (81 % of theory based on the amount of triethylphosphine used). The hexane filtrate was evaporated at about .05 mm for 30 minut e s leavin g a clear liquid. The infrared spectrum obtained as the neat l i quid

PAGE 66

59 is shown in Figure 18. The 1 H nrnr sp e ctr u m {Figure 19) obtained using C6D6 as a solvent consisted of a sin g le broad peak (8._56 T' ave.) (literature for (CH 3 P).5is a broad peak at 8.5 T) (51). The liquid exhibited a boil ing point range of 65-70 C (.0_5 mm) {literature value for (CH 3 P) 5: 110-112 (1mm)) (49). The 3lp nmr spectrum obtained in hexane consisted of a single peak at -14.5 ppm (literature value for (CH 3 P).5: -21 ppm) (49). The mass spectrum exhibited a parent peak at a m/e value of 230 (molecular weight of (CH 3 P) 5 : 230) as well as numerous other peaks attributable to fragments of this cyclic phosphine. Five of these peaks were listed by A.H. Cowley and R. P. Pinnell (52) and these peaks are listed in Table 5. The yield of undistilled (CH 3 P)5 was 0.97 g (82% of theory based on the amount of methyldichlorophos phine used). The Reaction of Triethylphosohine with Dimethylchlorophosphine Triethylphosphine (0.61 g, 5.2 mmole) was added to dimethylchlorophosphine (0.50 g, 5.2 mmole) result ing in a vigorous exothermic reaction that may have vola tized part of the starting materials. The product, a white solid, after recrystallization from hexane melted from 87-89 C. The infrared spectrum of this material is shown in Figure 20. The 1 H nmr s pectrum obtained in CDC1 3 (Figure 21) consists of a broad peak (peak A,~

PAGE 67

. I 3000 1500 1000 Fig. 18.--Infrared Sp e ctrum (CH 3 P).5 (Neat). O' 0

PAGE 69

62 TABLE 5 MASS SPECTRAL DATA FOR (CH 3 P)5 m/e Relative Intensity Assignment Reported peaks for (CH 3 P) 5 (52) 230 53.6 ( CH 3 p) 5-f" 215 39.0 (CH3) 4P5+ 184 4.5 (CH 3 P) 4 + 122 100.0 (CH3)4P2+ 61 60 .1 (CH3)2p+ Corresponding peaks found for (CH 3 P) 5 230 64 (CH 3 P)5+ 215 42 (CH3)4P5+ 184 20 (cH 3 P) 4 + 122 53 (CH3)4p2+ 61 100 (CH3)2p+

PAGE 70

1500 1000 500 Fi g 20.--I n frar e d Spectrum [(c 2 H 5 ) 3 PP(CH 3 ) J Cl (Nujol Mull).

PAGE 72

65 7.37 ave.) and multiplet (peak B, 'I 8.60 ave.). The area ratio of A:B is 1.00:1.75. The 3lp nmr obtained using acetonitrile as a solvent consisted of peaks at +42 ppm and -34 ppm of approximately 1:1 area ratio. Anal. Calculated for (C 2 H5) 3 PP(CH 3 ) 2 Cl: C, 44.76; H, 9.86; P, 28.86; Cl, 16.52. Found: C, 39.85; H, 9.99; P, 29.35; Cl, 16.66. The yield of unrecrystallized ma terial was 0.36 g (35% yield based on the amount of tri ethylphosphine and dimethylchlorophosphine used). The Reaction of Tri-n-propylphosphine with Phosphorus Trichloride 3(C 3 H 7 ) 3 P + 2PC1 3 2 11 P 11 + J(C 3 H 7 ) 3 Pcl 2 Tri-n-propylphosphine (1.00 g, 6.25 mmole) was added to a solution of phosphorus trichloride (0.57 g, 4.2 mrnole) in 20 ml of benzene. The solution became yellow-orange in color within a few seconds and after about two minutes an or.ange:..red precipitate formed. The solid was filtered, washed with boiling acetonitrile, filtered again and dried under vacuum. The solid exhib ited no melting point up to 330 C; it reacted violently with concentrated nitric acid. Anal. Calculated for elemental red phosphorus: P, 100. Found: P, 92.42; C, 3.82; H, .80; Cl, .90. The yield of orange solid was 0.13 g (100% of theory base~ on the amount of phosphorus trichloride used assuming this product to be elemental phosphorus). The benzene filtrate combined with the ace tonitrile washings was evaporated to dryness leaving a

PAGE 73

66 white solid residue. The solid, recrystallized from hexane, melted from 141-148 c. The infrared spectrum of this compound is shown in Figure 22. The 1 H nmr spec trum shown in Figure 23 was obtained using CDC1 3 as a solvent. It consisted of a multiplet (peak A,~ 6.72), a broad peak (peak B, 8.23 ave.) and a multiplet (peak C, 8.84). The area ratios of A:B:C are 1.00:1.00:1.56 (calculated for (c 3 H 7 ) 3 Pcl 2 : 1.00:1.00:1.50). The 3lp nmr spectrum exhibited a single peak at -103 ppm. Anal. Calculated for (c 3 H 7 ) 3 Pcl 2 : C, 46.77; H, 9.15; P, 13.40; Cl, 30.68. Found: C, 46.66; H, 9.09; P, 13.63; Cl, 30.61. The yield of (c 3 H 7 ) 3 Pcl 2 was 1.41 g (98% of theory based on the amount of tri-n-propylphosphine used). The Reaction of Tri-n-prooylohosphine with Phenyldichloroohosphine Tri-n-propylphosphine (1.00 g, 6.24 mmole) was added to phenyldichlorophosphine (1.12 g, 6.24 mmole). An exothermic reaction followed resulting in the formation of a solid in less than a minute. The mixture was washed with about 4 ml of cold (-5) acetonitrile; the insoluble material was filtered, dried under vacuum, weighed and recrystallized from acetonitrile. This off-white ma terial melted from 154-156 C {literature value for tetra phenylcyclotetraphosphine: 154-5-156 C) (47). The in frared spectrum of this solid is identical with that published for tetraphenylcyclotetraphosphine (48). The

PAGE 74

3000 1500 1000

PAGE 76

69 Jlp nmr spectrum consisted of a sin g le peak at +3.9 ppm in benzene (literature values for CTc6H5P~4: +9 ppm (49) and +4.6 ppm (50)). The yield of [(C6H5)P]4 was 0.67 g (98% of theory based on the amount of phenyldichlorophos phine used). The acetonitrile filtrate was evaporated to dryness leaving a white solid residue. After recrystal lization from hexane, this solid melted from 138-146 C (found above for (C 3 H 7 ) 3 Pcl 2 : 141-148 C). The infrared and 1 H nmr spectra are identical with corresponding spectra of (C 3 H 7 ) 3 Pcl 2 shown in Fi g ures 22 and 23, respec tively. The 3lp nrnr spectrum consisted of a single peak at -104 ppm (found above for (C3H7)3PCl2: -103 ppm). The yield of unrecrystallized material was 1.44 g (100% of theory based on the amount of tri-n-propylphosphine used) The Reaction of Tri-n-prooylphosphine with Diphenylchlorophosphine (CJH?)JP + 2(C6H5) 2 PC1 ------;;..(c 3 H 7 ) 3 Pcl 2 + (C6H5) 2 PP(C6H5) 2 Tri-n-propylphosphine (0.50 g, 3.1 mmole) was added to diphenylchlorophosphine (l.38 g, 6.25 mmole). A viscous oil re~ulted that slowly crystallized over a period of 12 hours. This solid was washed with 5 ml of cold (-10 C) acetonitrile and the resulting mixture filtered. The insoluble material melted from 120-123 C after recrystallization from acetonitrile (literature value for (C 6 H 5 ) 2 PP(C 6 H 5 ) 2 : 120 .5 C) (53). The infra red spectrum is identical with that published for

PAGE 77

70 (C6H5) 2 PP(C 6 H 5 ) 2 (53). The 31 P nmr spectrum consisted of a single peak at +14.8 ppm (literature value for (C 6 H 5 ) 2 PP(C 6 H 5 ) 2 : +15.2 ppm) (53). The yield of (c 6 H 5 ) 2 PP(C6H5)2 was 0.81 g (70% of theory based on the amount of diphenylchlorophosphine used). The acetonitrile fil trate waa evaporated to dryness leaving a white solid resi due. This solid melted from 134-141 C after recrystal lization from hexane (found above for (c 3 H 7 ) 3 Pcl 2 : 1411480 C). The infrared and 1 H nmr spectra are identical with corresponding spectra of (c 3 H 7 ) 3 Pcl 2 shown in Figures 22 and 23, respectively. The 3lp nrnr spectrum consisted of a single peak at -104 ppm (found above for (c 3 H 7 ) 3 PC1 2 -103 ppm). The yield of unrecrystallized material was 0.47 g (65% of theory based on the amount of tri-n-propyl phosphine used) The Reaction of Tri-n-propylphosphine with Methyldichloroohosphine 5(CH 3 ) PC1 2 + 5(C 3 H 7 ) 3 P ((cH 3 ) P] 5 + 5(C 3 H 7 ) 3 Pcl 2 Tri-n-propylphosphine (4.11 g, 25.7 mrnole) was added to methyldichlorophosphine (3.00 g, 25.8 mmole) and the resulting oil warmed to approximately 100 C for 10 minutes. The oil crystallized on standing over a 12-hour period. This solid was washed with 5 ml of cold (-5) hexane and the resulting mixture filtered. The isolated solid melted from 138-146 C after drying under vacuum (found above for (C3H7)3PCl2: 141-148 C). The infrared and 1 H nmr spectra are identical with the corresponding spectra of (C 3 H 7 ) 3 Pc1 2 shown in Figures 22

PAGE 78

71 and 23, respectively. The 3lp nrnr contained a sin g le peak at -98.5 ppm in acetonitrile (found above for (c 3 H 7 ) 3 Pcl 2 : -103 ppm). The yield of unrecrystallized solid was 5.85 g (99% of theory based on the amount of tri-n-propylphosphine used). The hexane solvent was re moved from the filtrate leaving a clear liquid. The infrared and 1 H nmr spectra of this liquid are identi cal with the corresponding spectra of ~(CH 3 )'P] 5 shown in Figures 18 and 19. The 31 P nmr spectrum was a single peak at -13 .4 ppm in hexane ( found above for [( CH 3 ) P] 5: -14.5 ppm). The yield of this liquid was 0.95 g (80% of theory based on the amount of methyldichlorophosphine used). The Reaction of Tri-n-prooylphosphine with Dimethylchlorophosphine (C 3 H 7 ) JP + (CH 3 ) 2 PC1 --;.. [(c 3 H 7 ) 3 PP(CH 3 ) Cl Since the reaction of these materials resulted in the formation of an oil, diethylether solutions of the starting materials were allowed to react in order to separate the product from possible unreacted start ing materials. Tri-n-propylphosphine (0.83 g, 5.2 mmole) in 5 ml of diethylether was added to dimethylchlorophos phine (0.50 g, 5.2 mmole) in 5 ml of diethylether result ing in the formation of a liquid two-phase mixture. The ether layer was decanted from the viscous oil and the remaining solvent removed under vocu um The infrared spectrum is shown in Figure 24. The 1 H nmr spectrum (Figure 25) consisted of a very broad peak (peak A,'I

PAGE 79

3000 1500 1000

PAGE 81

74 7.48 ave.), a broad sin g let (pea k B, ~ 8 .31) and a multiplet ( p eak C, 'i' 8.82). Since th e se peaks all overla p ped, only approximate are a ratios could be obtain e d; for the ratios A:B:C these were 1.00:2.00:1.95. Th e Jlp nmr spectrum consisted of two peaks at -22.5 pp m and +47.9 ppm of approximately 1:1 area ratio. An a l Calculated for [(C3H7)3PP(CH3);Jc1: C, 51.46; H, 10.60; P, 24.13; Cl, 13.81. Found: C, 51.16; H, 10.53; P, 24.23; Cl, 13.99. The yield of this liquid was 1.01 g (78% of theory based on the amount of both starting materials used). The Reaction of Tri-n-butylohosphine with Methyldichloroohosphine Tri-n-butylphosphine (5.18 g, 25.6 mrnole) was added to methyldichlorophosphine (3.00 g 25.6 mrnole) and the resulting oil heated to 150 C for about 10 minutes. This oil crystallized on standin g over a peri od of 12 hours. The solid was washed with 5 ml of cold (-5 C) hexane, filtered and dried. It melted from 127134 C (literature value for (C4H 9 ) 3 Pcl 2 : 134-137 C) (25). The infrared spectrum was identical with that reported for trl-n-butyldichlorophosphorane (44). The 31 P nmr spectrum consisted of a sin g le peak at -100 ppm (literature value for ( C4 H 9 ) 3 Pc 1 2 : -106 ppm) ( 54) Evaporation of the hexane resulted in the isolation of a liquid and solid mixture. A second washing with hexane also resulted in isolation of a liquid solid mixture.

PAGE 82

75 The infrared spectrum of this mixtur e consi s t e d of an overlap of the infrared spectra of (c 4 H 9 ) 3 Pcl 2 a nd [(cH 3 ) P] 5 Th e liquid was decanted from the solid and the 3lp nmr spectrum of this decantate consisted of a sin g le peak at -14.4 ppm (found above for [(CH 3 )PJ5: -14.5}. Unsuccessful attem p ts to isolate pure QcH 3 P~ 5 included washin g the mixture with hexane, washin g the mix ture with acetonitrile, liquid liquid extraction with a chloroformhexane system, sublimation and distillation. The Reaction of Tri-n-butylohosphine with Dimethylchlorophosphine (C4H9) 3P + (CH3) 2PCl [(C4H9) 3 PP(CH3) 2] Cl Tri-n-butylphosphine (0.52 g 2.6 mmole) in 3 ml of diethylether was added to dimethylchlorophosphine (0.25 g, 2.6 rnrnole) resulting in the separation of an oil. The ether layer was decanted from the oil layer and the remaining solvent removed under vacuum. The infrared spectrum is shown in Figure 26. The 1 H nmr spec trum (Figure 27) consists of a broad peak (peik A, 'Y 7.45), a broad multiplet (peak B, 'Y 8.31) and a second broad multiplet (peak C, 'Y 9.01). The approximate area ratios of these three overlappin g peaks, A:B:C, are 1.00:2.57: 1.36. The 31 P nmr spectrum in benzene consisted of two peaks at -21.2 ppm and +51.O ppm of 1:1 area ratio. Anal. Calculated for nc4H9) 3 PP(CH 3 )~Cl: C, 56.27; H, 11.14; P, 20.73; Cl, 11.86. Found: C, 55.99; H, 11.11; P, 21.03; Cl, ll.83. The yiel d of this liquid

PAGE 83

3000 1.500 1000

PAGE 85

78 was 0.62 g (80% of theory based on the amount of both reactants used). The Reaction of Tri-n-butylohosohine with Diethylchloronhosohine Tri-n-butylphosphine (1.63 g 8 .05 mmole) was added to diethylchlorophosphine (1.00 g, 8.05 mmole) and the resulting solution warmed to 80 C for 12 hours. After cooling, the solution was extracted with diethyl ether and the ether solution decanted from the oil. The remaining solvent was removed under vacuum. The infra red spectrum is shown in Figure 28. The 1 H nmr (Figure 29) consists of a complicated pattern of peaks in the region of 1 8. 331' to 9 .29 1 Anal. Calculated for I [(C4H9)3PP(C2H5)2]Cl: C, 58.79; H, 11.42; P, 18.95; Cl, l0.84. Found: C, 58.73; H, 11.43; P, 19.04; Cl, l0.83. The yield of liquid was 2.33 g (89% of theory based on the amount of both reactants used). The Attempted Reaction of Tri-n butylphosohine with Bis(dimethylamino) chlorophosphine Tri-n-butylphosphine (1.31 g, 6.49 mmole) was added to bis(dimethylamino)chlorophosphine (1.00 g, 6.48 mmole) and warmed to 80 C for about 10 minutes and al lowed to stand at room temperature for 12 hours. The 31 P nrnr spectrum of the resulting solution consisted of a peak at +34.8 ppm (literature value for (C4H9) 3 P: +32.3 ppm) (S5 ) and a peak at -155 ppm (literature value

PAGE 86

5000 1500 1000

PAGE 88

81 for ((cH 3 ) 2 N] 2 PC1: -160 ppm) (56 ) The infrared spec trum ~f this liquid consisted of an overlap of the in frared spectra of the starting ~~terials. The Reaction of Tri-n-octylohosphine with Phosphorus Trichloride 3(C8Hl 7 ) 3 P + 2PCI) 2"P" + 3(C8Hl 7 ) 3 Pcl 2 Tri-n-octylphosphine (3.00 g, 8.09 mmole) was added to phosphorus trichloride (0.74 g, 5.4 mmole) in 20 ml of benzene. 'I'he solution turned yellow-orange in color and gradually became darker in color until a red-orange precipitate formed after about five minutes. The solid was filtered, washed with boiling acetonitrile, filtered again, and dried under vacuum. The solid ex hibited no melting point up to 330 C; it reacted vio lently with concentrated nitric acid. Anal. Calculated for phosphorus: P, 100. Found: P, 87.85; C, 7.63; H, 1.48; Cl, 1.92. The yield of orange solid was 0.15 g (90% of theory based on the amount of phosphorus trichlor1de used). The benzene filtrate combined with the ace tonitrile filtrate was evaporated to dryness leaving a white solid residue. After recrystallization from hexane, the solid melted from 88-90 c. The infrared spectrum of this compound is shown in Figure 30. The 1 H nmr spec trum obtained in CDc1 3 (Figure 31) consists of a broad peak (peak A, r 6.78 ave.) and a broad pattern of peaks (peak B) from 8.031' to 9.39 'I he area ratio of A:B is 100:11.6. The 3lp nmr spectrum exhibited a single

PAGE 89

CD f\.> --'--------,/ "v-----__._ ___________ .,L_ ________ __ L_ _J 1500 1000 50 0 ( cm 1 ) 3000

PAGE 91

84 peak at -103 ppm in acetonitrile. A n a l. Calculated for (CsH 17 ) 3 Pcl 2 : C, 6$.29; H, 11.64; P, 7.01; Cl, 16.06. Found: C, 63.63; H, 11.37; P, 6.69; Cl, 15.62. The yield of the unrecrystalled product was 3.59 g (100 % of theory based on the amount of tri-n-octylphos p hine used). The Reaction of Tr i -n-octvlphosphine with Phenyldichlorophosphine Tri-n-octylphosphine (2.00 g, 5.39 mmole) was added to phenyldichlorophosphine (.96 g 5.36 mmole) resulting in an exothermic reaction and formation of a solid. This solid mixture was separated with acetoni trile as described above. The insoluble material melted from 154-156 C after recrystallization from acetonitrile (literature value for rrc6H5PTI4: 154.5-156 C) (47). The infrared spectrum of this solid is identical with that published for tetraphenylcyclotetra p hosphine {48). The 31 P nmr spectrum consisted of a sin g le peak at +3.5 ppm (literature values for liC6H5P~ 4 are +9 ppm (49) and +4.6 (50)). The yield of unrecrystalliz ed CTc 6 H 5 )P] 4 was 0.53 g (91% of theory based on the amount of phenyl dichlorophosphine used). The material soluble in ace tonitrile was isolated by evaporating the filtrate to dryness. This solid melted from 88-91 C after recrys tallization from hexane (found above for (C 8 H 17 ) 3 Pcl 2 : 88-90 C). The infrared and 1 H nmr spectra are identical

PAGE 92

85 with the correspondin r, s pe ctra of (c 8 H 17 ) 3 P cl 2 sh o w n in Fi g ures 30 and 31, res p ectively. Th e 31 P nm r sp e trum consisted of a sin g le pe ak at -1 0 4 ppm (found above Th e Re a ction o f T r i -n-o c tyl ph o snhine with Di n h e nylchloroo h o s phin e (CsH17)3P + 2(c6H5)2PCl~{C6H5)2PP(C6H5)2 + (CsH17)3PCl2 Tri-n-octylphosnhine (1.50 g 4.04 mmole) was added to diphenylchlorophosphine (1.79 g, 8.08 rnmole) resulting in an oil that slowly crystallized over ape riod of two to three days. This solid mixture was sepa rated using acetonitrile as described above. The insoluble material melted from 120-123 C after recrystallization from acetonitrile. The infrared spectrum of this solid is identical with that published for tetraphenyldiphos phine (53). The 3lp nmr spectrum consisted of a sin g le peak at +14.8 (literature value for {C6H5) 2 PP(C6H5) 2 : +15.2) (53). 'l'he yield of tetraphenyldiphosphine was 1.36 g (90 % of theory based on the amount of diphenyl chlorophosphine ). The material soluble in acetonitrile was isolated by evaporating the solvent. This material melted from 82-85 C after recrystallization from hexane. The infrared and 1 H nmr spectra are identical with cor responding spectra of (C 8 H 17 ) 3 Pcl 2 shown in Fi g ures 30 and 31. The 31 P nmr spectrum consisted of a sin g le peak

PAGE 93

86 The Re a ctio n of Tri-n-octvlohosnhine with Methvldichloro o hos o h i ne Mixin g these phosphines in a one to one mole ratio resulted in the-formation of a homogeneous liquid. This liquid was warmed to 150 C for one hour and then allowed to cool over a period of 12 hours. The oil did not crystallize. It was soluble in diethyl ether, hexane, benzene, chloroform, and acetonitrile. The 31 P nmr spec trum of the neat liquid consists of a single extremely broad peak at 0 ppm. The Reaction of Tri-n-octylohosohine with Dimethylchloroohos oh ine Tri-n-octylphosphine (2.93 g, 7.91 mmole) was added to dimethylchlorophosphine (0.75 g, 7.8 mrnole) resulting in the formation of a viscous oil. This oil was soluble in diethyl ether, hexane, chloroform, benzene, and acetonitrile. The infrared spectrum of this oil is shown in Figure 32. The 1 H nmr spectrum (Figure 33) consists of a complex series of overlappin g peaks from 8 .40 f to 9 .21 r 'I'he 3lp nmr spectrum consisted of two peaks at +25.9 ppm and -14.5 ppm of area ratio 1.0: 1.0. The 3lp nmr Investigation of the Re actions of Tris(dimethvlamino)phosphine with Phoschorus Trichloride, Diohenyl chlorophosphine, Phenyldichloroohosphine, and Methyldichlorophosohine [(CH 3 ) 2 N] 3 P + PClJ ~[(CH 3 ) 2 N] 2 PC1 + [(cH 3 ) 2 N]PC1 2 [(CH3)2N]3P + (C6H5)2PCl >KCH3)2NJ2PCl + (C6H5)2P[N(CH3)i}

PAGE 94

3000 1500 1000 Fi g 32.--Infrar e d Sp e ctrum of [(CeH17) 3 PP(CH 3 ) 2Jc1 (N eat ).

PAGE 95

88 1 d I

PAGE 96

89 [(CH 3 ) 2 N] 3 P + (C 6 H5)PC1 2 ---"?UCH 3 ) 2 N] 2 PC1 + (C 6 H 5 )P(Cl)[N(CH 3 ) 2 ] [(CH 3 ) 2 N] 3 P + (CH)PC1 2 ) [(cH 3 ) 2 N] 2 PCl + (CH 3 )P(Cl)[N(CH 3 ) 2 ] In a typical reaction the tris(dimethylamino) phosphine was added to the chlorophosphine resultin g in the evolution of heat and the formation of a clear solll tion. The solution was allowed to stand for 12 hours and then transferred into a 31 P nmr tube with a ground glass joint sealed onto the open end to p ermit stopper ing. The 31 P nmr spectrum was obtained on the neat liquid. The results of these experiments are summarized in Table 6. The Reaction of Tris(dimethvlamino) ohosohine with Dimethylchloroohosphine 2(CH 3 ) 2 PC1 + [(cH 3 ) 2 N] 3 P ~rrcH 3 ) 2 N] 2 PC1 + (CH 3 ) 2 PN(Cn 3 ) 2 (CH3) 2 PC1 Tris{dimethylamino)phosphine {0.42 g, 2.6 mmole) in 5 ml of diethylether resulting in the formation of a white precipitate. The mixture was filtered and the solid dried under vacuum. The melting point of this solid was not obtained since it sublimed out of the warm por tion of the melting point tube. The infrared spectrum is shown in Figure 34. The 1 H nmr spectrum (Figure 35) consists of a doublet (peak A, r 7.14) and a second doub let (peak B, r 7.66) with an area ratio for A:B of 2.04: 1.00. On standing in CDC1 3 the material apparently de composed or reacted precipitatin g a small amount of uni dentified orange solid. Anal. Calculated for (cH 3 ) 2 PN (CH3)2(CH3)2PCl: C, 35.74; H, 9.00; P, 30.73; N, 6.95;

PAGE 97

TABLE 6 3lp NUCLEAR MAGNETIC RESONANCE DATA FOR THE REACTIONS OF (CH 3 ) 2 N 3 P WITH Pc1 3 (CH 3 )PC1 2 (C6H5) 2 PC1, AND (C6H5)Pc1 2 3lp Shifts Relative Reported 31 P Shifts for the Reactants to 85~ H 3 P04 Area Ratio Products Products [( CH 3 ) 2 N] JP -157 1/1 [(CH3) 2NJ2PCl -160 (55) Pc1 3 -163 [(CHJ) 2 NJPC1 2 -166 (55) [{ CH 3 ) 2 N] 3 P -159 1/1 [(CH3) 2NJ2 F C l -160 (55) (cH 3 )PC1 2 -149 [( CH 3 ) 2 N JP{ CHJ) Cl -150. 7 ( 55) [(cH 3 ) 2 N] JP -159 1/1 [( CH3) 2N]2PC 1 -160 < 55) (C6H5) 2PCl 64.8 [(CH3) 2 N ] P (C6H5) 2 64 (57) U CH3) 2N] 3P -159 1/1 [(CH3) 2N]2 PC l -160 (55) (c 6 n 5 ) PC1 2 -147 8CH 3 ) 2 N] P (C6H5)Cl "' 0

PAGE 98

1500 1000 soo

PAGE 99

92 I '' i' A

PAGE 100

93 Cl, 17.58. Found: C, 35.51; H, 8.80; N, 7.27; Cl, 17.08. The phosphorus analysis by difference is 31.34. The actual analysis for phosphorus was not obtained because of an error by the analytical laboratory. The diethyl ether was removed from the filtrate resultin g in isola tion of a liquid. The 1 H nmr spectrum of this liquid is similar to that of bis(dimethylamino)chlorophosphine with some small impurity peaks. The 31 P n.111r spectrum of the reaction mixture before separation in chloroform consisted of two peaks at -66.4 ppm and -164 ppm (litera ture value for ncH3ND2PCl: -160 ppm) (55). The yield of this liquid was 0.43 g (104% of theory based on the amount of tris(dimethylamino)phosphine used if all of the liquid was [(cH 3 ) 2 N] 2 PC1). The Reaction of Triethylphos oh ine with Phosphorus Trichloride in Diethyl Ether (2:1 Mole Ratio) 2(C2H5) 3P + PCl3 >[(C2H5) 3PJ2PCl3 Triethylphosphine (0.50 g, 4.2 m.111ole) in 5 ml of diethyl ether and phosphorus trichloride (0.29 g, 2.1 mmole) in 5 ml of diethyl ether were added simultane dusly and slowly to 20 ml of diethyl ether cooled to -20 C. A white precipitate immediately formed and was isolated by filtration. This solid was kept at -20 C since at room temperature it became yellow-orange in color and tacky. The white solid did not exhibit a true melting point up to 330 C but did become deep bro w n-red in color. The infrared spectrum taken at room temperature is shown

PAGE 101

94 in Figure 36. The 1 H nmr spectra of this solid o b tained at -20 C and at room tem p erature are shown in Fi g u re 37. At -20 C the peaks are bro a d a n d overlaopin g from 6.42 'Y to 9.06T'. At room temperature the em e r g ence of peaks at 6.64 j" and 8.56 T'. They may be co mp ared with the 1 H nmr soectrum of (C 2 H 5) 3 Pc1 2 (Fi g ure 17). Anal. Calculated i'or [(c 2 H 5 ) 3 P] 2 Pc1 3 : C, 3 8 .57; H, 8.09; P, 24.87; Cl, 28.47. Found: C, 3 8 .40; H, 8.29; P, 24.70; Cl, 28.42. The yield of' this solid was a 0.36 g (57% oi' theory based on the total amount of react an ts used). The product described above was also obtained ii' the reactants were mixed in a 1:1 mole ratio. The Reaction of Triethylphos o hin e with Phosphorus Trichloride in D iet hy l Eth er Phosphorus trichloride (0.29 g, 2.1 mmole) in 5 ml of diethyl ether was slowly added to a stirring solution of triethylphosPhine (0.75 g, 6.3 m.~ole) in 20 ml of diethyl ether at room tem p erature. The white precipitate thRt resulted was filtered and dried under vacuum. This solid which became only slightly yellow on standing at room temperature i'or several hours melted from 138-141 C. The infrared spectrum of this solid is shown in Figure 38. The 1 H nmr spectrum (Figure 39) obtained in CDc1 3 consists of a series of peaks (peak A, t' 6.62), a broad band of pe .:.k s (oeak B, T 7.61 ave.) and a series of peaks (peak C, 'I 8.63 ave.). The area

PAGE 102

1500 1000 500

PAGE 104

, 3000 1500 1000

PAGE 106

99 ratios of oeaks A:B:C are l.00:2.46:5.5 4 An al. Calc ~ 1 ate d for ( C 2 H .5) 3 P] 3 PC 1 3 : C 4 3 9 6 ; H, 9 2 2 ; P, 2 5 19 ; Cl, 21.63. Found: C, 43-73; H, 9.52; P, 24.90; Cl, 21._55. The yield was 0.86 g (83 % of theory based on the amount of both reactants used). The Re a ct ion of [ ( C 2 H 5:) 3 P ]-3. PC 1 w it h Phosnhorus Trichloride at F lev a ted Temperature [(C2H5) 3P] 3 PC13 + PCl3~ 2 11 P 11 + 3(C2H5) 3PCl2 Phosphorus trichloride (0.08 g, 0.6 mmole) was added to [(c 2 H 5 ) 3 P] 3 Pc1 3 (0.29 g, 0.59 mmole) in 10 ml of toluene. The resulting solution was warmed to the boil ing point for one and one-half hours resulting in the formation of a red-orange precipitate. This solid was filtered, washed with boiling acetonitrile, filtered again, and dried under vacuum. The red-oran g e solid did not exhibit a melting point up to 330 C; it reacted vi olently with concentrated nitric acid. Anal. Calculated for phosphorus: P, 100. Found: P, 92.40; C, 4.32; H, 1.27; Cl, 1.49. The yield of this solid was 0.03 g (75% of theory based on the amount of both reactants used and assuming the product to be elemental phosphorus). The toluene and acetonitrile filtrates were combined and resulting solution evaporated to dryness leaving a white solid residue. This solid melted from 240-246 C after drying under vacuum (literature melting point for (C 2 H 5 ) 3 Pcl 2 : 240-250 C) (46). The ini'rared and 1 H nmr spectra are identical with the corresponding spectra

PAGE 107

100 of (c 2 H 5 ) 3 Pc1 2 The yield was O.JO g (91 % of theory based on the amount of both react an ts used). The Attempt ed Reaction of Diohenyl chloronhos oh ine with [( CsH5) 2 PP ( C 2 H5) 3 ] C 1 at Elev2ted Temoerature The addition of 1:1 mole ratio of diphenylchloro phosphine to triethylphosphine results in the formation of the white crystalline compound [(c 6 H 5 ) 2 PP(C 2 H5)) Cl as reoorted by Seidel ( 26) Di p henylchlorophos ph ine (0.93 g, 4. 2 rrnnole) was added to Qc 6 H 5 ) 2 PP(C 2 H 5 ) 3 ] Cl (1.43 g, 4 .22 mmole) in 20 ml of toluene and heated to the boiling point for one and one-half hours. After cooling, the mixture was filtered and the collected solid dried. The 1 H nmr and infrared spectra indicated this solid to be Rc6H5) 2 PP(C 2 H5)_i]c1 with a possible trace of (C 2 H5) 3 Pcl 2 The toluene solvent was re m oved from the filtrate by evaporation. The infrared spectrum of the resulting liquid is identical with the infrared spec trum of diphenylchlorophosphine. The Reaction of Triethyl ph o sohine with Phenyldichlorophos phi ne in :Qiethyl Ether Triethylphosphine (0.25 g 2.1 mmole) in 5 ml of diethyl ether cooled to -20 C was added to phenyl dichlorophosphine (0.38 g, 2.1 ::n.rnole) in 10 ml of diethyl ether also maintained at -20 C. A white solid immedi ately formed. The mixture was then evaporated to dryness

PAGE 108

101 maintaining the tem pe rature at -20 C. The r esulting dry white crystals were dissolved in cold (-20 C) CDc1 3 and transferred into an nmr tube. The 1 :n mar sp e c trum of this solution at -20 C (Fi gure 40) consist ed of two overlapping peaks (peak A, 12.0 6 ave .), a broad peak (peak B, 1' 7.38), and a second broad peak ( oeak D, i' 8.73). Peak A is assi gned to phenyl pr otons and peaks Band D to ethyl protons. The area ratio of peak A to peak B plus peak Dis 1.0:3.2 (calculated for [ ( C6H5)(Cl) PP(C2H.5)3Jc1: 1.0:3.0). The CDC1 3 solution was allowed to warm to +2.5 C and a 1 H nmr spectrum obtained immedi ately. The spectrum (Fi g ure 40) exhibits the gr owth of a multiplet at 6.63 'I (peak C) as well as two sets of triplets superimposed on a broad peak at 8.,561" (found above for (C 2 H.5) 3 Pc1 2 : multiplet at 6.6 4 i' and two sets of triplets at B .,561' ) The phenyl peak is also changed in shape. After standin g 12 hours the 1 H nmr s pe ctrum (Figure 40) consists of peaks very simil ar to those found for (C 2 H.5) 3 Pc1 2 (Figure 17) as well as ph enyl peaks simi lar to those found for (c 6 H.5P) 4 The reaction was repeated as described above except the produ ct was filtered from the diethyl ether solvent and dried under vacuum. This compound melted over a range from 80 -235 C. Anal. Calculated for [(c 6 H.5) (Cl) PP(C 2 H.5) j Cl: Cl, 2J.86. Found: Cl, 24.28. The yield was 0.43 g {68% of theory based on the total amount of both reactants used).

PAGE 109

(C2H5) 3 PP(C6H5)C1 2 at +25 C after sta 1 : r 1ing 12 hour s B C D D t-' 0 f\)

PAGE 110

103 The Th e rmal D ecomoosition of [(C 6 H5) (Cl)P P (C 2 H s)yCl 4 "[1c 2 H5) 3 PP(C6H5) (Cl)] Cl~ 4(C 2 H_5 ) 3 Pcl 2 + [(c6H5) ~4 The crystalline solid [(c6H5) (Cl) PP(C 2 H5) 3 j Cl (0.45 g, 1.5 rnmole) was added to 20 ml of hexane and the resultin g mixture was maintained at the boilin g point for one and one-half hours. The mixture was eva po rated to dryness and washed with 10 ml of cold (-5 C) acetoni trile. The resulting mixture was filtered and the in soluble material dried under vacuum. The meltin g point of this material was 150-156 C (literature value for (C6H5P)4: 154.5-156 C) (47). The infrared spectrum of the solid is identical with that published for tetraphenyl cyclotetraphosphine (48). The yield was 0.13 g (81 % of theory based on the amount of [(C2H5) 3PP(C6H5) qJ Cl used). The acetonitrile filtrate was evaporated to dryness leav ing a white solid residue. After recrystallization from toluene the solid melted from 242-246 C (literature value for (C 2 H5) 3 Pcl 2 : 240-250 C) (46). The infrared spec trum of this solid is identical with the infrared spec trum of (C 2 H 5 ) 3 Pcl 2 shown in Figure 16. The yield was 0.27 g (93% of theory bas ed on the amount of [(c 2 H5) 3 PP( C 6 H 5 ) C 1] Cl used) The React i o n of Tri e thv1ohosph ine with Me thyldichlorophoso hin e in Diet h yl E ther Methyldichlcrophosphine (0._50 g, 2.3 mmole) in

PAGE 111

104 3 ml of diethyl ether was added to triethylphosphine (0.51 g, 2.3 mmole) in 5 ml of diethyl ether at o C. The resulting white precipitate was filtered and dried under vacuum. 'l'he solid melted from 76-78 C. The infrared spectrum is shown in Figure 4 1. The 1 H nmr s pe ctrum (Fi ~ ure 42) obtained in CDc1 3 consist ed of a broad peak (peak A, f 7.56) and a broad g rou p of peaks (peak B, I 8.62 ave.). Anal. Calculated for [(c 2 H5) 3 PP(Cl) (cH 3 )] Cl: C, 35.77; H, 7.72; P, 26.35; Cl, 30.16. Found: C, 35.54; H, 7.98; P, 26.08; Cl, 29.85. The yield of this compound was 0.82 g (81 % of the6ry based on the amount of both reactants used). The Thernrnl Decomoosition of (C2H5) 3 PP(Cl) {C HJ) Cl 5 [(c 2 H 5 ) 3 PP(CH) (Cl)] Cl > [(cH 3 )Pj 5 + 5(C 2 H5) 3 PC1 2 The crystalline solid (C 2 H 5 ) 3 PP(Cl) (CH 3 ) Cl (1.07 g, 4.55 mrnole) was added to 20 ml of hexane and heated to the boilin g point of the mixture for one hour. After cooling to -5 C, the solution wa s filtered. After recrystallization from toluen e this so li d melted from 239-248 C (literature value for (C 2 H5) 3 Pcl 2 : 240-250 C) (46). The infrared and 1 H nmr spectra are identical with the corresponding spectra of (C 2 H5) 3 Pc1 2 shown in Figures 16 and 17. The yield of (C 2 H 5 ) 3 Pcl 2 was 0.67 g (77% of theory based on the amount of starting material used). The hexane solvent was removed by ev aporat ion from the filtrate yielding a mixture of a solid and a

PAGE 112

1500 1000 soo t-' 0 \n

PAGE 114

107 liquid. The liquid was isol a ted by extraction with cold hexane and evaporation of th e hexane. The inf rar ed and 1 H nmr spectra of the liquid ar e identical with the cor responding spectra of ~cH 3 PTI 5 shown in Figures 1 8 and 19. The yield of ~cH 3 P~5 was 0.14 g (67% of theory based on the amount of startin g material used. The Reaction of Triethvloho so h ine with Dimethylchloroohosnhine fn75' iethyl Et ~er (C2H5)3P + (CH3)2PCl --?1"'ITC2H5)3PP(C H 3)2l Cl Triethylphosphine (O.Jl g, 2.6 mmole) in 5 ml of diethyl ether was added to dimethylchlorochosphine (0.25 g, 2.6 mmole) in 5 ml of diet hyl ether resulting in the precipitation of a white solid. The infrared and 1 H nmr spectra of this solid are identical with the cor responding spectra of [(cH 3 ) 2 PP(C 2 E5) 3 ] Cl shown in Figures 20 and 21, respectively. The yield of unrecrystallized material was 0.47 g (84% yield based on the amount of both reactants used). The Attemoted Reaction of Di me thvl chloronhosohine with [ (CH ) "' PPt,C....,E ,-,) Cl Dimethylchlorophosphine (0.11 g, 1.1 mmole) was added to [(cH 3 ) 2 P P (C 2 H 5 )il C in 10 ml of hexane and main tained at 68 C for two hours. The mixture was filtered and the insoluble material dried under vacuum. The 1 H nmr spectrum of this solid was very simil a r to that of [(cH 3 ) 2 PP(C 2 H 5 ) 3 ] Cl. Small impurity peaks at 6.64 r ma y indicate the formation of a small amount of (C 2 H5) 3 Pc1 2

PAGE 115

108 Repeating t he reaction w it h no solvent at 100 C resulted in loss of some volatile products. The 1 H n rn. r spectrurn of the remaining soli d was similar to that of [(C H3 )2PP (C2H5) 3 1c1, however, the peaks which may c e assi g ned to (C 2 H5) 3 Pc1 2 had increased in relativ e area. Inte g ratio n of these areas indicates approximately a 29 per c en t con version of [{CHJ) 2 PP(C 2 H 5 ) 3 ]cl to (C 2 H 5 ) 3 Pcl 2 The vola tile (CH 3 ) 2 PP(CH 3 ) 2 was not recovered and may have evap orated from the reaction mixture. The Reaction of rri-n -b utylphosohi~~ with M ethyldic hl oropho sohine in P e troleum Rther Since the product of this reaction is soluble in diethyl ether, petroleu -:: ~ ether was used as the solvent. Tri-n-butylphosphine (1.72 g 8.54 rnmole) was slowly added to methyldichloroohosphine (1.00 g 8.54 rn..~ole) in 20 ml of petroleum ether. Two liquid phases resulted. The petroleum ether layer was decanted from the more dense viscous layer and remainin g petroleum ether removed under vacuum. The infrared sp e ctrum is sh o w n in Fi g ure 43. The 1 H nmr spectrum (Figure 44) consists of a b r oad peak (peak A,~ 7. 69 ), a second broad peak (pe ak B,~ 8.54) and a group of peaks (peak C, 9.12). The area ratio of peak A to peaks Band C is 1.0:2.8. The 31 P nmr spectrum consists of two ma jor peaks of approximately 1:1 area ratio at +14.7 ppm and -16.5 ppm as well as three minor peaks at -77.9 ppm, -33.7 ppm, a nd -1.7 ppm.

PAGE 116

--___/\,-----.._ _..__ ____ ---1\, -----...___ 30 0 0 1500 1000

PAGE 117

C ,_, ,_, 0

PAGE 118

111 Calculated for [(C ,.H 9 ) ,.,P?(CH,.,) (Cl)] Cl: C, 48 .91; H, 9 4 7 ; 4 .) .) P, 19.41; Cl, 22.21. Fo .:.nd : C, 50 10; E, 9 99 ; P, 1 8 .71; Cl, 21.25. This analysis gives an emnirical formula of P 2 c1 2 c 13 9 H 33 3 (c alculated for [ (C;/i 9 ) 3 PP(CH 3 ) (Cl)] Cl: P 2 c1 2 c 13 H 30 ). The additional carbon and hydro g en found may be due to incompletely removed solvent. The infra red and 1 H nmr spectra of the liquid product of this re action are identical with the corresoondin g spectra of the product obtained by the reaction of these phosphi nes without solvent and before heating. The re acti on of this above. D1scuss1on The products of the reactions of some tertiary phosphines with some chloro ph osphines are summarized in Table 7. In general, analogous reaction products are formed by the reaction of any chlorophosphine with the tri-n-alkylphosphines. Ex ce ption s to this are the reactions of (C 2 H 5 ) 3 P with (C 6 H 5 ) 2 PC1 and (C 8 H 17 ) 3 P with (C6H5) 2 PC1. The r ea ction of (C 2 H 5 ) 3 P with (C 6 H5) 2 PC1 results in the formation of a phosphinophosphonium chloride, [(C 2 H 5 ) 3 PP(C 6 H 5 ) 2 ]c l (26). The reaction of (c 8 H 17 ) 3 P with CH 3 Pc1 2 gives an oil tha t exhibits a single extremely broad peak at O ppm in the 31 P nmr spectrum. Attempts to separate this oil fr om possible unre s ~ ~ e d starting ffiB terial were unsuccessful. 'l he infrared spectrum of this oil does not have a peak in the r eg ion expected for a com

PAGE 119

E t P 3 TAB LE 7 SU MMARY OF T HE REA CT I O N S O F S OME T ER TIA R Y P HO SP HINE S W1 TH SO ME C H L OR O PH O SP HI NE S Pc1 3 q) 2 PC 1 (p P C1 2 Me PC1 2 II p fl + E t 3 P~ C l ( q) P) 4 + ( Me P)5 + Et 3 Pcl. 2 ( 20 ) I Et 3 PC1 2 I Et 3 P C12 I II P" + PRP2 + (Q) P ) 4 + ( MeP ) 5 + Pr 3 PC1 2 I Pr 3 PC12 Pr 3 Pcl 2 Pr 3 Pc 1 2 .,. I I II pl! +


PAGE 120

113 pound havin g a P-Cl bond (CH 3 Pcl 2 has a stron g peak at 480 cm1 and (c 8 n 17 ) 3 Pc1 2 has a strong peak at 510 cm1 ). The 1 H nmr spectrum has ch ara cteristic octyl peaks ::..; s well as some broad overlappin g peak s in the re g ion from 7.5~ to 9.Jr. This oil was not further characterized. Althou g h analo g ous products were isolated in reactions of any chlorophos phi ne with the tri-n-alkylphos phines, there were differences in the reactivity of t he various tri-n-alkylphos p hines observ e d. The difficulty in isolating products in the re a ction of (c 8 tt 17 ) 3 P with CH 3 Pcl 2 was not completely unexpected since the reactions of RJP compounds (R c 2 H 5 > c 3 H 7 c 4 H 9 and C 3H 17 ) with CH 3 Pcl 2 required increasin g ly hi gh tem pe ratures as the alkyl group is chan g ed from ethyl to nropyl to butyl to octyl. The reaction of CH 3 Pc1 2 with (C 2 H5) 3 P was exo thermic and the product crystallized over a pe riod of two to three minutes. It was necessary to hea t the reaction mixture of CH 3 Pcl 2 with (c 3 H 7 ) 3 P to 100 C in order to com plete the reaction. Another increase in reaction te mp era ture (150 C for 10 minutes) was required for the reac tion of CH 3 Pcl 2 with (c 4 H 9 ) 3 P to occur. Kinetic or thermo dynamic factors may account for t he re a ction te mpe ratures. It should also be noted that the compl ete s ep aration of (c 4 H 9 ) 3 Pcl 2 and [(cH 3 ) P] 5 was not achieved bec 2c. se of the increased solubility of the (c 4 H 9 ) 3 Pc1 2 in non polar solvents. There were also observed differences in the re

PAGE 121

114 actions of Pc1 3 with the various tri-n-alkyl p hosnhines were also observed. The reactio~ of PCl3 with (C 2 H 5 ) 3 P usin g benzene as a solvent resulted in formation of a yellow precipitate whtch on standin g beca m e re d -or ange in color. It is possible that the yellow solid was a reaction intermedi ate ; however, onl y nptt and (C 2 H 5 ) 3 Pc1 2 were isolated. When (c 3 H 7 ) 3 P and (c 4 H 9 ) 3 P were allowed to react with a benzene solution of ?Cl~, a red-oran g e .., solid prec i pitated immediately. Precipitation did not occur for s everal minutes in the re action of (CsH 17 ) 3 P with Pc1 3 in benzene. After about five minutes, a red orange solid did precipitate from the red solution. The red-oran g e solids isolated in these reactions were not pure elemental red phosphorus. The percentage of phosphorus in these products varied from 88.79 per cent to 92.42 per cent. The other elements pre s ent were orinci pally carbo n hydrogen, and small percenta g es of chlorine. The presence of car bon hydrogen, and chlorine may be a result of end groups on phosphorus networks. Similar substances have been reported as products of the photo polymerization of white phosphorus in the presence of terti ary phosphines. The insoluble red polymers obtained in these reactions were characterized as red ph os phorus networks having organic end grou ps (58). The reactions of (c 6 H 5 )PC1 2 and (c 6 H 5 ) 2 PC1 with R 3 P were observed to be very sim il s r with the exception of the reaction of (c 2 H 5 ) 3 P with (C6H 5 ) 2 PC1 as noted a b ove.

PAGE 122

115 The reaction of (C 2 H 5 ) 3 P w ith (C H 3 ) 2 P C1 r e sulted in the formation of ~c 2 H 5 ) 3 P P (c F. 3 ) 2 Jc1. A lt h o ug h th e structure is written as a phos ph ino oh os ph on iLl m c h lori d e, the actual structures .of this co m poun d and t h e oth e r compounds of this type have not be e n d eter m ined. T he structures are written similarly to those to co m pounds reported by V. W. Seidel (26) and S. R Jain a nd H. H. Sisler (36). In addition, these co mp ounds w hose analyses correspond to a general em p iric a l formul a of R 3 PPrt 2 Cl have F reater solubility in polar solv en ts than i n non polar solvents. These compounds, ho w ever, ma y h a ve dif ferent bonding and structur e s. The reactions of (C 3 H 7 ) 3 P, (c 4 H 9 ) 3 P, and (C 8 H 17 ) 3 P with (CH 3 ) 2 PC1 resulted in liquid products. This may be a result of t h e lon g er alkyl chains causing decreased ionic behavior as evidenced by the fact that [(csH 17 ) 3 PP(CH 3 ) 2 ]Cl is soluble in hexane whereas [(C 2 H5) 3 PP(CH 3 ) 2 ]Cl is much less soluble in nonp olar so 1 vent s S inc e p ur if i c at i on of [ ( C 8 H 1 7 ) 3 PP ( CH 3 ) 2 ] C l was unsuccessful, elemental analyses do n ot d ifferentiate a comoound formation fro m a homo g ene o u s m ixture of start ing materials and/or products. Th e evidence of compound f i 1--k b h Jl ormat on 1.s tue two pea .s o served 1n t e P nmr spectrum at +25.9 ppm and -14.5. These pe aks do not corres pond to those of the starting material and are similar to the peaks observed in the 31 P nmr spectra of other com pounds of the type [R 3 PP(CH 3 ) Cl ( T able 4). Tri-n-butylphos ph ir.e reacts with (C 2 H 5 ) 2 PC1 yield

PAGE 123

116 ing an adduct analogous to the p roduct of the reaction of (c 4 H 9 ) 3 P with (CE 3 ) 2 PC1. Tri -n-b u tylphos p hine was found not to react with [(cH 3 ) 2 N ] 2 PCl even at 80 c. These reactions are not summarized in 'I'able 7. Althou g h it was reported th a t e xch ange of amino g roups with chlorine atoms occurs in the r ea ctions of some aminophosphin es with some chloro ph os ph ines (3 9 ) the reactions of [(cH 3 ) 2 N ] 3 P with t he above chloro ph osphin e s were carried out because of the unusual c hlorine abstrac tion behavior of [(cH 3 ) 2 N J 3 P when it is allowed to re a ct with dimethylchloramine (37, 38). These reactions result in exchan ge of g roups as shown in Table 7. In each case liquid products were obtained with the exception of a solid-liquid mixture isolated in the reaction of [(cH 3 ) 2 N ] 3 P with (CH 3 ) 2 PC1. The elemental ana ly ses cf the solid correspond to an empirical for m ula of (cH 3 ) 2 N P(cH 3 ) 2 (cH 3 ) 2 PCl. This adduct is re porte d but not c hara cteriz ed by A. B. Burg and P. J. Slota, Jr., in a different reaction (18). The single peak attributable to this adduct in its nmr spectrum sup po~ ts a structure ha vin g equivalen t 3lp nuclei such as [~cH 3 ) 2 PJ 2 l (cH 3 ) 2 }c1 in which nitro g en is quarternized. However, the 1 H nmr spectrum e xhibits two doublets of 2:1 area ratio. If t he above structure is maintained in solution, ihe 1 H nmr spectrum woul~ be expected to exhibit a triplet assi g ned to the NCH 3 pro tons since these protons would be co uple,d t o t w o 3lp nu clei. The C H 3 P proton~ mi g ht be a doublet, triplet, or

PAGE 124

117 quartet deoendin g on the coupling of the 31 P nuclei. A very rapid equilibrium between (CH 3 ) 2 PN(CH 3 ) 2 8nd (CE 3 ) 2 PC1 can cause the nmr measurements to 11 see 11 the a vera g e en vironment of the nuclei. 'I he reported 31 ? nmr shifts for (CH 3 ) 2 PC1 and (CH 3 ) 2 NP(CH 3 ) 2 are -93 ppm (45) and -39 ppm (59), respectively. The mole weighted average of these shifts is -66 ppm (found for {CH 3 )2PN(CH 3 ) (CH 3 )2PCl: -66.4 ppm). This equilibrium also explains the observed doublets in the 1 H nmr spectrum. Because of this rapid exchange, the structure of the solid adduct cannot be determined from these data. Attempts to avoid the forma tion of the adduct and isolate (CH 3 ) 2 PN(CH 3 ) 2 and [(cH 3 ) 2 N] 2 PCl by slowly adding (CH) 2 PC1 to z. n excess of [(CH) 2 N] 3 P were unsuccessful. The fact that : [(CH) 2 N] 3 P does not react analogously to tri-n-alkylphosphines and form [(cE: 3 ) 2 N] 3 Pc1 2 may reflect nitrogen to phosphorus pi( d1Y bonding. Although the mechanism of chlorine abstrac tion by tri-n-alkylphosphines has not been elucidated, it is not improbable that phosphorus a-orbitals may be involved in the mechanism. Assuming appreciable pt( -ct1'1 bonding in [(cH 3 ) 2 N] 3 P, the availability of phosphorus a-orbitals in the aminoohosphines will be reduced by this effect. According to J. R. Van Wazer and L. Maier, the exchange of amino groups and chlorine atoms on phos phorus which results in nonrandom mixtures is probably attributable to considerable rehybridization of the elec tronic orbitals of phosphorus when a chlorine atom is

PAGE 125

118 exchanged for an amino grou p or vice versa (39). These authors repor t th at the crit ical f a ctor is probably not one of steric hindrance. The driving force of these re actions is difficult to s pecify usin g the available thermo dynamic data. Trends were also observed as the chlorophosphines were va fi ed for any of the above tri-n-alkylphosphines. Under th ~ conditions of our experiment, the reactions of Pc1 3 with tri-n-alkylphosphines are hi g hly exotherm i c and rapid, and benzene ~as used as a solvent to mode r ate these reactions. It was found that phenyldichl oro ph os phine reacts with tri-n-alkyl ph o s p hines to form crys ta line products quickly. In general, several hours are re quired to form crystalline products in the reactions of (C6H5) 2 PC1 with tri-n-alkylp hosphines. Heatin g mos t of the r~action mixtures of (CH 3 ) FC1 2 and tri-n-alkylphos phines was necessary to complete the reaction. Dimethyl chlorophosphine did not react with tri-n-alkyl ph os phines to form (cH 3 ) 2 PP(CH 3 ) 2 and R 3 Pcl 2 Adducts were isolated as noted above. The 1 H nmr spectra of these adducts all exhibit broad peaks. The broadness of the pe a ks may re sult from protons cou p lin g with protons a nd with the two nonequivalent phos ph orus nuclei. The reaction of (cH 3 ) 2 PCl with (C 2 H 5 ) 3 P is sufficiently exoth e rmic so that consider able amounts of startin g moterials are volatilized. In part this accounts for the low yield of adduc t by t he direct addition of the phosphines. The yield of this

PAGE 126

119 reaction was increased from 32 p e r cent to SL~ pe r cent by using diethyl ethor as a solvent to m oderate t h e h e L~ in g of the mixture. Possible ex p lanations for the ob served difference in reaction behavior o f (CH 3 ) 2 PCl a s compared to PC1 3 (C 6 H 5 )PC1 2 (C 6 H 5 ) 2 PC1, and CH 3 Pcl 2 are discussed below. The reactions of (C 2 H 5 ) 3 P with Pc1 3 (C6H 5 )PC1 2 and CH 3 Pcl 2 were carried ou t usin g diethyl ether as a solvent in an attempt to isolate reaction intermediates. The reaction of Pc1 3 with (C 2 H 5 ) 3 P at -20 C using a 1:2 mole ratio in diethyl ether solvent resulted in iso lation of [(c 2 H 5 ) 3 P] 2 Pc1 3 '1 1 his solid gave evidence of decomoosition at room temperature by chan g in g color from white to orange as well as becomin g tacky. The product is soluble in polar solvents suc h as chloroform and ace tonitrile. The 1 H nmr s r, .::. t rum. at -2 0 C { Fi g ure 37) contains peaks typical of the adducts described abo ve. The 1 H nmr spectrum obtain e d at room temperature (Figure red precipitate occurred at room temperature sug5esting that the adduct further re acts to form (C 2 H 5 ) 3 Pcl 2 and 11 P.'' When ~c1 3 and (C2H5) 3 P were allowed to react in a 1:3 mole ratio at room tem perature in diethyl ether, [(C 2 H 5 ) 3 P] 3 Pcl 3 was isolated. This solid did not c han g e color or physical state at room temperature over a ~2 riod of several hours. It was soluble i n acet o nitrile and chloroform. Its 1 H n m r s pe ctrum shows ch aracteristically

PAGE 127

120 broad peak s. T his compoun d w as mixed 0 ith the stoi c hi om et ric a m ount v f Pc 1 3 in boilin r tolu e n e to form (c 2 H 5 ) 3 PC12 and 11 P." These are the products o f the r e action of Pc1 3 and (C 2 H5) 3 P in be nzene at room temperature. Attemots to isolate (C 2 H 5 ) 3 PPC1 3 at -20 C in diethyl ether u sin g a 1:1 mole ratio of (C 2 H 5 ) 3 P to Pc1 3 were u n suc c e s sf ul. The adduct [(c 2 H 5 ) 3 PP Cl(C 6 H 5 )]Cl isolated in di ethyl ether at -20 C was decomposed to form ~C6H5)~4 and (C 2 H 5 ) 3 Pcl 2 The 1 H nmr spectra (Fi g ure 40) show this decomposition. T he products o f this de co mposi ti on are again the same as the p roducts of the direct reac tion of (C 6 H 5 )PC1 2 and (C 2 E 5 ) 3 P using no solvent. Analo gous to the above reactions [(c 2 H5) 3 PPCl(CH 3 )]Cl was isolated and decomposed to [(cH 3 ) P ] 5 and (C 2 H 5 ) 3 Pc1 2 The reaction of (c 4 H 9 ) 3 P with CH 3 Pc 1 2 using p etroleum ether solvent resulted in formation of [(c 4 H 9 ) 3 PPCl(CH 3 )] Cl which decom p osed to (c 4 H 9 ) 3 Pc1 2 and [(c H 3 )P] 5 at 150 C. These reactions have s h o wn that addu ct formation may be an intermediate step in chlor ine a bstra ct ion A tt empts at elevated temperatures res u lt ed in only a s ma ll con version to (c 2 H5) 3 Pc1 2 Higher temperatures or other experimental conditions may result in chlorine abstrac tion to a greater degree. As noted abo ve the kineti cs of these reactions have not been determined and several mechanisms c a n be used to explain the products. On e mechanism involvin g

PAGE 128

121 an adduct intermediate is represented by the followin g series of equations. + R 3 P + R 2 PC1 > R 3 PPR 2 Cl~ R 3 P-PR 2 Cl Cl [ Cl + ] + RJP PR2' + R 2 1 PCl :;.,. R P ? P I Cl __ :--..c,. R P,..,l Cl + :::> 1 PDR 1 ---:, 3 PR~ 7 3 V .:. '2 .. 2 R = n-alkyl, R 1 = C6H5. This mechanism can be extended to the re a ction of RPC1 2 I R PC1 2 and Pc1 3 with R 3 P. In this mechanism the stability of the adduct determines tte oroducts isolated. Possibly electron withdrawing R 1 grou p s (C6H5> Cl) de stabilize the ionic structure of the adduct whereas elec tron donating R' groups (CH 3 c 2 H,) stabilize this struc ture. Relative sizes and lattice ener g ies may also be important in stabilizing the intermediate. By lowering the temperature, intermediates not stable at room tempera ture might be stabilized as observed in several cases. Other mechanisms are possible and additional kinetic and thermodynamic data are required in order to defini tively establish the mechanism. Sur. .:.rn a ry A systematic study of the reactions of some terti ary phos rihines with s c : :: chlorophos phines has shown that generally chlorine abstraction occurs in the reactions of tri-n-alkylphosphines with Pcl 3 (C6H5) 2 PC1, (C6H5)Pc1 2 and CH 3 Pc1 2 A number of new phosphine adducts were iso lated in the reactions of (CH 3 ) 2 PC1 with tri-n-alkylphos

PAGE 129

122 phines. Tris(dimethylamino)p hosphine undergoes ex chan ge of amino groups and chlorine atoms with any of the above chlorophosphines. This study has extended the syntheti c method of chlorine abstraction by tri-n-butylphosphine to three other tri-n-alkylphosphines end has shown that [(cH 3 )P] 5 can be syr. the sized ty this mettod. The isolation of a number of new phosphine addu cts at low tempera ture in diethyl ether and their subsequent decomposition to chlorine abstraction products su gg est that adduct formation may be an intermediate step in chlorine abstrac tion.

PAGE 130

BIBLIOGRAPHY 1. D. F. Clemens, H. H. Sisler, and W. s. Brey, Jr., Inorg. Chem., .,2, 527 (1966). 2. D. F. Clemens, W. s. Brey, Jr., snd H. H. Sisler, Inorg. Chem., 2 1251 (1963). _, 3. R. P. Nielsen and H. H. Sisler, J. Chem. Soc., 3819 (1962). 4. C. W. Heitsch, C. E. Nordman 1 and R. W. Parry, Inorg. Chem., 2, 508 (1963). 5. G. J. Palenik, Acta Cryst., 11,, 1573 (1964). 6. W. Brueser, K. H. Thiele, and H.K. Mueller, z. Chem., 2, 342 (1962). 7. K. H. Thiele, H.K. Mueller, and W. Brueser, z. Anorg. Allgem. Chem., lbJ2, 194 (1966). 8. N. R. Fetter and D. W. Moore, Can. J. Chem., bt_g, 885 ( 1964) 9. N. R. Fetter and B. Bartocha, Can. J. Chem., b!Q, 342 (1962). 10. N. R. Fetter, F. E. Brinclnnan, and D. W. Noore, Can. J. ~hern., .!_, 2184 (1962). 11. N. R. Fetter, B. Bartocha, F. E. Brinckman, and D. W. Moore, Can. J. Chem., ~' 13.59 (1963). 12. E. Wiberg, H. Graf, and R. Uson, z. Anorg. Allgem. Chem., 272, 221 (1953). 13. J. K. Ruff and M. F. Hawthorne, J. Am. Chem. Soc., l, 535 ( 1961) 14. R. P. Nielsen and H. H. Sisler, Inorg. Chem., 2, 753 (1963). 15. J.M. Kanamueller and H. H. Sisler, Inorg. Chem., 6 1765 (1967). 123

PAGE 131

16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 124 R. P. Nielsen, J. F. Vincent, and H. H. Sisler, Inorg. Chem., 2, 760 (1963). D. s. Payne, H. Noth, and G. Henniger, Chem. Commun., !., 327 (1965}. A. B. Burg and P. J. Slota, Jr., J. Am. Chem. Soc., 80, 1107 (1958). J. J. Brophy and M. J. Gallagher, Aust. J. Chem., 20, 503 (1967). F. A. L. Anet, Can. J. Chem., 12,, 2262 (1961). w. s. Brey, Jr., Private Communication. J.M. Jenkins and B. L. Shaw, J. Chem. Soc., 770 ( 1966) R. B. King, Inorg. Chem.,, 936 (1963). w. A. Hart and H. H. Sisler, Inorg. Chem., J, 617 ( 1964) S. E. Frazier, R. Pu Nielsen, and H. H. Sisler, Inorg. Chem., l, 292 (1964). V. W. Seidel, z. Anorg. Allgem. Chem., .12Q, 141 ( 1964) 27. K. D. Berlin, T. H. Austin, M. Peterson, and M. 28. 29. 30. 31. 32. 33. 34. Nagabhudranam, 1.'opics in Phosphorus Chemistry, Volume I, John Wiley and Sons, Inc., New York, 1964, p. 20. H. Kohler and A. Michaelis, Chem. Ber., 10, 814 ( 1877) F. Challenger and F. Pritchard, J. Chem. Soc., 864 (1924). J. N. Collie, J. Chem. Soc., 964 (1925). P. E. Worrall, J. Am. Chem. Soc.,, 2933 (1930). R.R. Holmes and E. F. Bertaut, J. Am. Chem, Soc., Bo, 2983 (1958). R. R, Holmes and E. F. Bertaut, J, Am. Chem, Soc., Bo, 2983 (1958). G. E, Coates and J. G, Livingstone, Chem. and Ind., 1366 ( 1958)

PAGE 132

125 35. J.M. F. Braddock and G. E. Coates, J. Chem. Soc., 3208 ( 1961) 36. S. R. Jain and H. H. Sisler, Inorg. Chem., In Press (1968). 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. s. R. Jain, L. K. Krannich, R E Hi g hsmith, and H. H. Sisler, Inorg. Chem., 6, 1058 (1967). D. B. Denne) and s. M. Felton, Inorg. Chem., 1, 99 (1968 J. H. Van Waz er and L. Maier, J. Am. Chem. Soc., 86, 811 (1964). E. M. Evleth, Jr., L. D. Freem a n, a nd R. I. Wagner, J. Org. Chem., g]_, 2192 (1962). W. C. Davies and W. J. Jones, J. Chem. Soc., 33 (1929). W. C. Davies, P. L. Pearse, and W J. Jones, J. Chem. Soc., 1262 (1929). H. Niebergall and B. Lagenfeld, Chem. Ber., .2.2, 64 (1962). S. E. Frazier, Master's Thesis, University of Florida (1963). G. W. Parshall, J. Inorg. Nucl. Chem., 12, 372 (195960) G. J. w. w. M. u. M. Kosolapoff, Or ga noohosohorus Co m pounds, John Wiley and Sons, Lnc., New York, 1950, p. 73. W B. Reesor and G. F. W ri g ht, J. Org. Chem., 22, 386 (1957). Kuchen and H. Buchwald, Chem. B e r., 91, 2296 (1958). A. Henderson, Jr., M Epstein, and F. s. Seichter, J. Am. Chem. Soc., .., 2462 (1963). L. Nielsen, J. V. Pust i n ge r, and J. Strobel, J. Chem. Eng. Data, 9, 167 (1964). Schmidt, R. Schroer, and li A ch e nbach, Angew. Chem., Intern. Ed.,~' 316 (1 9 66 ). A.H. Co w ley and R. P. Pinnell, Inorg. 1459 (1966). Chem., .2,,

PAGE 133

53. 55. 56. 57. 126 W~ Kuchen and H. Buchwald, Chem. Ber., 2.!., 2871 (1958). G. A. Wiley 8 nd W. R. Stine, Tetr~ahedron Lett., 2u., 2321 {1967). s. o. Grim and W6 M cFarlane, Nature, ~0-~, 995 (1965). K. Moedritzer, L. Maier, and L. C. D. Groenweghe, J. Chem. Eng. Data, 1, 307 (1962). L. Maier, Helv. Chim. Acta, 46, 2667 {1963). 58. V. P. Parlni and M. Y. Kraft, Sbornik State: Obschei Khimi, Akad. Naulc SSSR, 1., 716 (1953). 59. J. R. Van Wazer, c. F. Callis, J. N. Shoolery, and R. C. Jones, J. Am. Chem. Soc., ]!}_, 5715 ( 1956).

PAGE 134

BIOGRAPHICAL S IIB TCH Stanley F. Spangenberg was born October 27, 1942, in New Ulm, Minnesota. He is the son of M r. and Mrs. Stanley M. Spangenberg. He graduated from Fairmont Senior High School, Fairmont, Minnesota, in June, 1960. He completed the Bachelor of Arts deg ree from Au g sburg Col lege, Minneapolis, Minnesota, in June, 1964. He enrolled as a graduate student at the University of Florida in September, 1964. Mr. Spangenberg is married to the former Constance Lynne Gildseth. He is a member of the American Chemical Society and of Sigma Xi. 127

PAGE 135

Th1.s dissert atic n wa.s n r e pared under the d1rect1. o n of the chairman of t h e candid ate 's supervis o ry cormn1ttee and has been a p proved by all m e mber s of that committee. It was submitted to the De a n of t h e Colle g e of Arts and Sc 1ences and to the Graduate Council, and was approved as partial fulfillment of the requirements for the de gree of Doctor of Philosophy. December, 1968 Dean, College of Arts and Sciences Dean, Graduate School Supervia ory Com.'lli t tee: Cha rman


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID ESWBUUMPH_I32ZZO INGEST_TIME 2018-07-05T17:52:37Z PACKAGE AA00062990_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES