摘要:
764 J.C.S. DaltonConformational Effects on PNCH, PNC, and PNSi Spin Coupling inTe rva I en t P hos p h o r us-N it rog en Co m po u n dsBy Gordon Bulloch, Rodney Keat," and David S. Rycroft, Department of Chemistry, University of Glasgow,Glasgow GI 2 8QQThe l H and 13C n.m.r. spectra of a series of tervalent phosphorus-nitrogen compounds having a known conform-ation about the P-N bond have been obtained. These measurements confirm previous assumptions that thecoupling constants J(PNCH) and J(PNC) are relatively large and positive for a methyl group having a cis relationwith the lone pair of electrons on phosphorus. J(PNCH) is small, and J(PNC) small and negative when themethyl group is trans to the lone pair. The results are discussed by reference to compounds of unknown conform-ation.The coupling constant, J(PNSI'), in a series of silylaminophosphines, Ph,P-NR.SiMe, (R = Me, Et. Pr',or But), also appears to be related to the conformation adopted by the P-N bond.IT has been recognised l-* for several years that the spincoupling constant, J(PNCH) , in tervalent phosphorus-nitrogen compounds has a marked stereochemicaldependence. This was clearly demonstrated by studiesof the low-temperature lH n.m.r. spectra of the amino-phosphine, PPhCl(NMe,), and related compounds ~ ~ 9 7 inwhich rotation about the P-N bond became slow on theFIGURK 1 Thc dihedral angle, 8, for an aminophosphine,P(NMe,) (X)Yn.m.r. time scale and two distinct P-N-C-H couplingswere measurable. Both couplings had the same signbut a problem was posedl in relating the individualcouplings to the position of the methyl groups relative tothe lone pair of electrons on phosphorus.A planar, ornear planar, distribution of bonds about nitrogen isgenerally observed in aminophosphines 9-15 so that theassignment of couplings may be considered by referenceto Figure 1. The dihedral angle 0 for the aminophosphine,P(NMe,)(X)Y, is that between the plane containing theP, N, and Cn (in MeA) atoms and the plane f whichbisects XYY along the P-N axis. After consideration of-t This plane is assumed to bisect the lone pair, but when X #Y this is not strictly correct.A. H. Cowley, M. J. S . Dewar, W. R. Jackson, and W. B.Jcnnings, J . Amer. Chem. Soc., 1970, 92, 5206 and refs.therein.M.-P. Simonnin, R.-M. Lequan, and F. W. Wehrli, J.C.S.Chenz. Conzm., 1972, 1204.J. P. Albrand, A. Cogne, D. Gagnaire, and J. B. Robert,Trtvahedron, 1972, 28, 819.J . Nelson, R. Spratt, and B. J. Walker, Chem. Comm., 1970,1509.S. DiStefano, H. Goldwhite, and E. Mazzola, Org. MagneticResonance, 1974, 6, 1.J. Burdon, J. C. Hotchkiss, and W. B. Jennings, TetrahedronLettevs, 1973, 4919.J. Burdon, J. C. Hotchkiss, and W. B. Jennings, J.C.S.Pevkzn I?, 1976, 1052. * H. Boudjebel, H. Gonsalves, and F. Mathis, Bull. Soc. chim.Fvance, 1975, 628.the lH n.m.r. spectra of P(CF,),(NMeH) it was proposedthat J(PNCHI is relatively large (k19.2 Hz) for methylgroup A (0 ca. 0') and relatively small ( 5 6 . 7 Hz) formethyl group B (0 ca. 180') in PPhCl(NMe,).The samerelation has been assumed in making assignments to the13C spectra of PPh(NMe,)X (X = C1 or OMe); J(PNC)for X = C1 is relatively large and positive (k33.9 Hz)when 0 ca. O", but smaller and negative (zf11.4 Hz)when 0 ca. 180". A similar relation is suggested forJ(PNC) and J(PNCC) in PR(C1)(NPri,) (R = Me orPh) ,6 and for J(PNP) in diphosphinoamines.16In order to extend and, if possible, confirm theseproposals for J(YNCH) and J(PNC), we have examinedthe lH and 13C n,m.r. spectra of several tervalentphosphorus-nitrogen compounds of known crystalstructure (or preferred conformation established byelectron diffraction), and assumed the preferred structurein solution is the same. The angle 0 has not been repor-ted in most structure determinations and, where possible,we have used the fractional co-ordinates to obtain thisangle.RESULTSHydrogen-1 and 13C n.m.r.data for the N-methyl groupsof compounds of known structure are listed in Table 1. Therelative signs of the coupling constants were established bylH-{ 13C}, 1H-{31P), and 13C-( l H ) double-resonance experi-ments. J(PNCH) had the same sign as J ( C H ) in selectedcases (see Tables 1 and 2) and is always assumed to bepositive. In those cases where 13C satellites were too lowin intensity to obtain relative sign information from the lHspectra the relative signs of J(PNC) and J(PNCH) wereE. D. Morris and C. E. Nordmann, Inovg. Chem., 1969, 8,1673.lo G. C. Holywell, D. W. H. Rankin, B. Beagley, and J.M.Freeman, J . Chem. Soc. ( A ) , 1971, 785.l1 D. E. J. Arnold, E. A. V. Ebsworth, H. F. Jessop, and D. W. H.Rankin, J.C.S. Dalton, 1972, 1681.12 P. Forti, D. Damiani, and P. G. Fevero, J . Amer. Chem.Sac., 1973, 95, 756.l3 E. Hedberg, L. Hedberg, and K. Hedberg, J . Amev. Chem.Soc., 1974, 96, 4417.l4 L. Vilkov and L. S. Khaikin, Topics Current Chem., 1975,53, 25.l5 K. M. Ghouse, R. Keat, H. H. Mills, J. M. Robertson, T. S.Cameron, K. D. Howlett, and C. K. Prout, Phosphorus, 1972, 2,47, and unpublished work.l6 R. J . Cross, T. H. Green, and R. Keat, J.C.S. DaEton, 1976,14241978 765obtained by off-resonance 13C-{lH) experiments. The signs couplings [for example (1-5) < (3-7) and (2-6) <of the two J(PNC) couplings relative to J(PNP) in Ph,P* ( P - 8 ) shows that J(PNC)/J(PNCH) and J(HCNPF)/NMe*P(S)Ph, were obtained by 13C-{31P} experiments under J(FPNC) are both positive.TABLE 1Hydrogen- 1 and 13C n.m.r.data for tervalent phosphorus-nitrogen compounds of known conformationCompound J(PNCH) "/Hz J(PNC)/Hz Gc/p.p.m. Structure ref.PC1, (NMe,) f 12.9 b*c f21.2 f 1 b 38.3 b 1, 12, 21PCl (NMe,) , 11.8 h c f 1 6 . 5 f l b 38.8 dP(NMeJ3 18.8 c,e & 19.0 f 0.5 e*f 38.3 22, 23F,P.NMe.PF, f 3 . 0 g f2.25 f 0.1 g 21.4 g 13C1PPU But-PCl-NMe f l l . l Q,c f15.0 f 1 (NMe) 28.7 cf. 25(ClPNBut) , 6.6 f 0.5 54.1 g 25CdNMeCH,CH,kMe & 15.1(Me) e 5 1 8 . 8 f 0.28 33.1[J(HCNPF) f 1.51 [J(FNPC) f 7.G]7 - 7[J(PNCC) 6.11& 7.2(CH2) h FlO.8 f 0.2 48.8 24,IClPNMeNMeP(C1) NMe 12.3(PNMeP) 3517.3 (PNMeN) i(from ref.28)Ph,P-NMe.P(S)Ph, f 1.8 r 5 . 0 f 0.5 32.1 15[J(PvNC) f 5.0,J(PNP) rt 991P(NMeNMe) ,P f 15.4 c~ ~ P J f12.8 & 0.2 e * i f j 37.3 33ClP(NMeNMe) ,PC1 16.8 i 34p 4 (NMe) 6 f 16.3 e i j f36.4 f 0.1 e v f 37.3 37(from ref. 28)(see text)a fO.l Hz, except for data from ref. 28. Solutions in CH,Cl,.a' IJ(PNCH) + J(PNNCH)I or IJ(PNC) + J(PNNC)(.Same sign as J ( C H ) . d N. M. Zaripov, N. A. Nauniov, andf I n good agreement with the results of R. D. Bertrand, F. B. Ogilvie, andI n C6D6. h Mean value,L. L. Tuzova, Phosphovus, 1974, 4, 179.J . G. Verkade, J . Amev. Chem. Soc., 1970, 92, 1908; G. A. Gray and T. A. Albright, ibid., 1976, 98, 3857.see ref. 3.I n CDC1,.j Complex triplet, see text.conditions of lH noise decoupling.The case of F,P*NMe*PF, is complicated by the coupling to four fluorine nuclei;however, a single experiment (Figure 2) is sufficient t o ob-tain the relative signs. In Figure 2(b) the proton irradiation--vFIGURE 2 Carbon-13 n.m.r. spectra of F,P*NMe.PF,: ( a ) withproton noise decoupling; (b) with proton irradiation to highfrequency of N-methyl signals (see text). Both spectra wererecorded with the same offsetfield is set to high frequency of the N-methyl proton signals(the intensities are distorted because of population-transfereffects). Comparison of appropriate residual 13C - * 1HThe 13C n.m.r. spectra of the phenyl groups of thediphosphinoamines, Ph,P-NR.PPh, (R = Me, Et, or P r i ) ,are of the AA'X type (or ABX in the event of a small l3C:isotope shift).Of the six-line pattern expected for theX(W) spectrum,l7 we observed a deceptively simple triplet(R = Me or Et), the separation of the outer components ofwhich is IJ(PC") + J(PNPCn)I. The triplet structureshows that I J ( P N P ) I 2 IJ(PCn) $- J(PNPC") I. However,when R = Pri, C1 and C2 were doublets and C3 hacl tripletstructure. This places a value of ca. 10 Hz on IJ(PNP)I(i.e. between 6.3 and 13.3 Hz). The small value of IJ(PC1)+J(PNPC1)I when R = Me, relative to R = Et or Pri [orTABLE 2Hydrogen-1 and 13C n.m.r. data for selected tervalentphosphorus-nitrogen compoundsSCtCompound J(PNCH) "/Hz J(PNC)/Hz p.p.m.Cl,P.NMe*PCl, 1 3.0 1.6 & 0.3 28.8Cl,P.NMe.P(O)Cl, f 1.6 1 2 .8 f 0.2 31.5[J(PVNCH) [J (PVNC) rk 1.81f 15.51PC1, (NMePh) &4.8 T 3 . 8 f 0.3 35.3 bPPh,(NMePh) &1.7 -i8.7 f 0.3 b 35.7 bCl,P.NMe-SiMe, f 6.8 eye ~ 4 . 2 & 1 " 30.5 ePh,P*NMe-PPh, f 3.0f 5.4 f 0.2 f 32.6 fa fO.lHz. b I n C,D,. Same sign as .J(CH). d -2.0 +0.5 Hz in I. J. Colquhoun and W. McFarlane, J.C.S. Favaday 11,1977, 722. I n CH,Cl,. f I n CDC1,.J(PC1) in PPh, (Table 3)] suggests that J(PNPC1) is sig-nificant. lH-{ 3lP) INDOR results l6 on Ph,P*NR*PPh,(R = Et or Pri) indicated that IJ(PNP)I is < ca. 50 Hz.17 S. Aime, R. K. Harris, E. M. McVicker, and M. Fild, J.C.S.Dalton, 1976, 2144766 J.C.S. DaltonThe 13C spectrum of P(NMeNMe),P was a deceptively simpletriplet (Table l), but for reasons which are not clear thecentre component of the triplet was much broader than theouter components.Similar effects were apparent in the13C spectra of P4(NMe), (Figure 3). The results of spectrasimulations of this AX,X', spin system do not give an upperestimate of J(PNP), but they do show that the inclusion ofa four-bond coupling, J(PNPNC), doubles up the outercomponents of the triplet. The observed separation of theouter sharp lines (72.8 Hz) is therefore a close approxim-ation to 2J(PNC).DISCUSSIONvalues of n in the series PCla_,(NMe,), (12 = 1-3).Both ~(pNc-1 and J(pNC) decreased with increasingThistrend has previously been reported for J(PNCH).20 InOrder to to the approximate con-formations adopted about the P-N bonds, about whichthere has been some dispute for PCl,(NMe,) 1*21 andP(NMe,)3,22,23 it is necessary to establish whether thesum of the two 90" couplings is greater than the sum ofthe couplings for 8 = 0 and 180".This is not possible,theseTABLE 3Carbon- 13 data a for diphenylphosphinoamines, Ph,P.NR*PPh,W C ) J(PNC) Wl) J(PC') W2) JW2) W3) J(PC3) -- -R p.p.m, Hz p.p.m. Hz p.p.m. Hz p.p.m. HzMe 32.6 -5.4 138.5 6.5 132.4 20.7 128.2 6.3Et 47.2 +11.3 139.6 12.5 132.6 22.2 128.0 6.0Pri 51.9 +10.3 139.gb 13.3 132.9 b 22.8 128.0 6.3PPh, 6 - 12.51 + 19.65 +6.800 In CDC1, a t ambient temperatures. C', C2, and C3 signals appeared as deceptively simple triplets, and the separation of theouter components of these triplets gave I J(PCn) + J(PNPC") I ( f 0.7 Hz) ; see text for interpretation in terms of individual couplings.C4 was asinglet at 6 128.6 f 0.1 in all three cases.S. Splrensen, R. S. Hansen, and H, J . Jakobsen, J . Amer. Chem.SOC., 1972, 94, 5900.Doublet.lH-{ ,OSi) and 1H-{ 31P) double-resonance experiments onthe ,OSi satellites of the SiMe, signals of the compoundslisted in Table 4 showed that J(PNSi) and J(SiCH) haveopposite signs (taking into account 18 a negative magneto-gyric ratio for 2OSi). The latter coupling is known 19 to bepositive in SiMeCl,H, so we deduce that J(PNSZ) has anegative sign, The four-bond coupling J(PNSiCH) is alsoFIGURE 3 13C-(1H} n.m.r. spectrum of P,(NMe),, obtained with-out exponential weighting of Free Induction Decaypositive. It was possible to check these assignments in thecase of PCl,[NMe(SiMe,)] since 2gSi satellites of the N-methyl signals were observable and the relative signs ofJ(PNCH) and J(PNSi) could be established by 1H-{20Si}double-resonance experiments.W.McFarlane, Ann. Rev. N.M.R. Spectroscopy, 1968, 1,135.lg W. McFarlane, J. Chem. SOC. ( A ) , 1967, 1275.2o G. Mavel, Ann. Rev. N.M.R. Spectroscopy, 1973, B5, 1 andL.V. Vilkov and L. S. Khaikin, DokZady Akad. Nauk S.S.S.R.,refs. therein.1966, 168, 810.but there is little doubt that the larger couplings, forn = 1, most closely represent the latter combination ofangles in so1ution.lIn a recent electron-diffraction study l3 it was estab-lished that F,P*NMe*PF, adopts a preferred conformationin which both angles 8 = 180".Also, the crystal struc-ture of the diphenylphosphino-derivative, Ph,P*NMe*P(S)Ph,,15 shows that 8 = 160". Whilst correlatioiisinvolving these compounds require an extrapolationbetween the vapour or solid phase and solution, the rela-tively small PNCH couplings are consistent with theproposals for PPh(C1) (NMe,).l The negative 2°CTABLE 4N.m.r. data for silylaminophosphines a8(31P) 6(,*Si) J(PNSi) J(PNSiCH) dsf -- Compound p.p.m. p.p.m. Hz HzCl,P.NMe*SiMe, 173.5 15.8 - 44 +2.8Ph,P.NMe*SiMe, 48.5 12.1 - 32 + 1.4Ph,P*NEt.SiMe, 45.8 10.7 - 33 + 1.3Ph,P.NPri SSiMe, 42.1 9.0 - 10 +0.8C1( Ph) PSNMe-SiMe, 137.6 15.9 - 34 +1.9Ph,P*NBut.SiMe, 46.9 8.3 ] < l o / I <0.21a In CH,Cl, solution a t ambient temperatures, measuredusing 1H-{31P} and lH-(%i} double-resonance methods.b Downfield from 85% H,PO,.Downfield from internal Si-Me,. d Sign assumes J(SiCH) and J(PNCH) positive.J(SiCH) was ca. 8 Hz in each case. e &2 Hz. f f O . l Hz.coupling for Ph,P*NMe*P(S)Ph, is also to be expected onthis basis, but it is surprising that F,P*NMe*PF, has apositive PNC coupling, albeit small.22 L. V. Vilkov, L. S. Khaikin, and V. V. Evdokimov, J. Struct.Chem., 1972, 15, 4.23 A. H. Cowley, M. J. S. Dewar, D. W. Goodman, and J. R.Schweiger, J. Amer. Chem. SOC., 1973, 95, 6506; M. F. Lappert,J. B. Pedley, B. T. Wilkins, 0. Stelzer, and E. Unger, J.C.S.Dalton, 1975, 1207; A. H. Cowley, D. W. Goodman, N. A. Kueb-ler, M. Sanchez, and J. G. Verkade, Inorg. Chem., 1977, 16,854197810 N'fs zQ-lv767-Intermediate values of 8 are to be found in the diaza-ph~spholan,~~ ClbNMeCH,CH,NMe which probably hasan envelope conformation in the vapour phase fromwhich angles 8 of 70 & 20 and 150 & 20" may be estim-ated for N-Me and N-CH, groups respectively.In thecyclodiphosphazane CIP*NBut*PC1*NMe, 8 is ca. 70",provided that the compound retains a conformationsimilar to that of CIP*NBut*PC1*NBut .25 The PIIINCHcouplings for N-methyl groups which form part of af our-,26 f i ~ e - , ~ ~ 27928 or six-membered 29-31 ring generallyoccur in the range 10-19 Hz, with some of the largestcouplings observed in 2-phospha-l,3-diazacyclo-hexane~.,~ Apparently, the only exceptions are to befound with the dimet h ylamino-derivatives Me,N-POCH,CH(Me)NMe and Me,N*P(NMeNMe),P*NMe,,where IJ(PNCH)I 7.5 (ref.32) and an apparent 6.9 Hz(see below) 28931 respectively.The crystal-structure data 33 for the cage diphosphineP(NMeNMe),P may be used to give 8 = 49.5" for anapparent P * H coupling of 15.4 Hz.~* In C1P-(NMeNMe),PCl 8 is 65",34 and in ClPNMeNMeP(C$Me8 is 42.6 (PNMeP) and 68.1" (PNMeN),35 for whichcouplings of 16.8,35 12.3, and 17.3 Hz (ref. 28) respec-tively have been obtained. The couplings which relateto the PNMeNMeP group are IJ(PNCH) + J(PNNCH)Iand may not bear a straightforward relation to 8 ifJ(PNNCH) is significant. It is worth noting thatJ(PNNCH) is zero in NN-dimethylhydrazino-derivativesof pho~phines.~~ The apparent 16.3 Hz coupling ob-served in the lH spectrum of P,(NMe), is also compli-cated by long-range coupling effects.Its crystalstructure has been e~tablished,~' and this indicates that8 is ca. 0".The separation of the outer components of the tripletin the 13C n.m.r. spectrum of P,(NMe), gives J(PNC)+36.2 Hz (see above); this is good evidence for a rela-tively large and positive 8 ca. 0" coupling. The +12.8Hz separation of the outer components of the tripletobtained in the 13C spectrum of P(NMeNMe),P isIJ(PNC) + J(PNNC)I. Trial simulations show that,whilst J(PNNC) affects the intensity and position of thecombination lines (which were not seen), it could assumevalues approaching that of J(PNC) and still leave thespectrum as a triplet.7 7'1-r 77In Figure 4 J(PNCH) is plotted against estimatedvalues of 0, and this shows that there is no doubt that theassignment 1,2 applied to PPh(C1) (NMe,) and relatedcompounds is correct.Indeed, the lH and 13C n.m.r.data obtained1,2 for this molecule would fit that inFigures 4 and 5 quite well making the approximationthat 8 = 0 and 180". A scatter of points is to be expec-ted in view of (i) the changes in electron-supplying power0 90 1800/ OGraph of J(PNCH) against 8 for (a) P,(NMe),, ( b )P(NMeNMe),P, (c) ClhNMaCH,CH,dMe, (d) ClP(NMeNMe) 2-PC1, (e) ClPNMeNMeP(Cl)NMe, (f) ClPNMeNMeP(Cl)NMe, (g)ClP-NBut-PC1-NMe, (h) ClbNMeCH,CH,&Me, (i) Ph,P.NMe*P(S)Ph,, and ( j ) F,P*NMe.PF,FIGURE 4 ~ _ _ _ _I . 1 I II Iof the phosphorus and nitrogen substituents, (ii) theapproximate values of 8 in some cases, and (iii) theuncertainties involved in extrapolating solid-state datato solution. It is clear, however, that the conformationadopted by the P-N bond is the most important factordetermining the magnitude of J(PNCH).WhenJ(PNC) is plotted against 0 (Figure 5) the scatter ofpoints is as great as that for J(PNCH), but the generalform of the relation with 8 is established. Less sig-nificance may be placed on point (b) P(NMeNMe),P (seez4 V. A. Naumov, N. A. Gulyaeva, and M. A. Pudovik, Dolkady25 K. W. Muir, J.C.S. Dalton, 1975, 259.26 G. Bulloch and R. Keat, J.C.S. Dalton, 1976, 1113.27 J,-P. Albrand, A. Cogne, D. Gagnaire, J . Martin, J.-B.Robert, and J. Verrier, Ovg. Magnetic Resonance, 1971, 3, 75;Yu.Yu. Samitov, M. A. Pudovik, A. I. Khayarov, and L. K .Kibardina, J. Gen. Chem. (U.S.S.R.), 1973, 43, 42; L. V. Vilkov,L. S. Khaikin, A. F. Vasilev, T. L. Italinskaya, N. N. Melnikov,and V. V. Shvetsov-Shilovsky, Doklady Akad. Nauk S.S.S.R.,1969, 187, 1293.28 H. Noth and R. Ullmann, Chern. Bev., 1976, 109, 1942.29 R. 0. Hutchins, B. E. Maryanoff, J. P. Albrand, A. Cogne,D. Gagnaire, and J . B. Robert, J. Amer. Chem. Soc., 1972, 94,9151.Akad. Nauk S.S.S.R., 1972, 203, 590.above), and on points (d), ClPNMeCH,CH,NMe, and (f)3O H. Noth and R. Ullmann, Chem. Ber., 1974,107, 1019.31 M. D. Havlicek and J. W. Gilje, Inorg. Chem., 1972, 11,32 J. Devillers, M. Cornus, J. Roussel, J. Navech, and R. Bur-33 W. Van Doorne, G. W. Hunt, R.W. Perry, and A. W. Cordes,34 H. Noth and R. Ullmann, Chem. Ber., 1976, 109, 1089.35 R. Ullmann and H. Noth, Chem. Ber., 1976, 109, 2581.36 L. I(. Peterson, G. L. Wilson, and K. I. Thk, Canad. J. Chern.,37 F. A. Cotton, J. M. Troup, F. Casabianca, and J . G. Riess,1624.gada, Org. Magnetic Resonance, 1974, 6, 205.Inorg. Chern., 1971, 10, 2591.1969, 47, 1025.Inovg. Chirn. Acta, 1974, 11, L33768 J.C.S. Daltonthe P-N bond as a result of the relatively electronegativephosphorus substituents 41 is difficult to predict. Thereis only a very general correlation between J(PNC) andJ(PNCH) for the data in Tables 1 and 2.The PNP coupling constants in bis(dipheny1-phosphino) amines, Ph,P*NR*PPh,, have been related tothe approximate conformations adopted by the P-Nbonds.16 The large positive P N P coupling observedwhen R = Me suggests that the preferred angles 0 arenear MOO, a result also expected from the small value ofJ(PNCH) and negative J(PNC).The PNC(alky1)coupling in Ph,P*NR*PPh, is much more positive whenR = Et or Pri than when R = Me. For R = Et orPri this suggests a small angle 0, which is in good accordwith the relatively small PNP couplings (< ca. 50 Hz)observed for these compounds.The lH n.m.r. spectra of a series of silylamino-phosphines 42y43 (Table 4) were examined to try to estab-lish whether the coupling constant J(PNSi) also has aconformational dependence. This type of couplinginvolving tervalent phosphorus has been reported for(Me,Si),N*P:N*SiMe,,& where IJ(PNSi) I is greater for the:N*SiMe, group (26.8 Hz) than for the *N(SiMe,), group(9.1 Hz).During the course of this work the crystalstructure of the dimeric form of this phosphazene, namely[(Me,Si),N*PNSiMe,],, was reported 45 and this shows thatthe N(SiMe,), groups have a planar distribution of bondsabout nitrogen and that the SiMe, groups lie cis (0 0') andtrans (0 180') to the lone pairs on phosphorus. Insolution, rotation about the exocyclic P-N bonds is slowon the n.m.r. time scale, and two couplings, IJ(PNSi) +J(PNPNSi)I of 3.4 and 31.0 Hz, are observed. Severalexamples of PNSi coupling involving quinquevalentphosphorus have been r e p ~ r t e d . ~ ~ . ~ 'It is clear that J(PNSi) is also dependent on the stericbulk of the R group in the series Ph,P*NR*SiMe,.It isnot clear whether the more negative result for X = C1than Ph in X,P*NMe*SiMe, can be related to the electro-negativity of the phosphorus substituents. The trendfor the diphenylphosphino-derivatives suggests thatJ(PNSi) will be most negative if a preferred conform-ation in which 0 is small is adopted. Increasing size ofthe R group would be expected to increase the import-ance of those conformations where 0 approaches 180".The trend to less negative PNSi couplings is paralleledby the trend to less positive PNSiCH couplings. Theconformational dependence of the latter coupling hasrecently been recognised in P(CF,),[N(SiMe,),] and in[(Me,Si),N*PNSiMe,J, 45 (see above), where the onlyPNSiCH coupling (2.8 Hz) observed in the former com-pound was assigned to the SiMe, group cis to the lone45 E.Niecke, W. Flick, and S. Pohl, Angew. Chem. Internat.Edn., 1976, 15, 309.48 E. Niecke and W. Flick, Angew. Chew. Internat. Edn., 1974,15, 134.47 W. Buchner and W. Wolfsberger, 2. Naturforsch., 1974, B29,328; W. Wolfsberger, ibid., 1975, B30, 900, 904, 907; E.-P.Elindt, H. Rose, and H. C. Marsmann, 2. anovg. Chem., 1977,430, 155.48 R. H. Neilson, R. Chung-Yi Lee, and A. H. Cowley, J . Amer.Chem. SOC., 1975, 97, 5302; Inorg. Chem., 1977, 16, 1455.(ClPNBd),, which refer to CH2- and (CH,),C-carboncouplings respectively.An attempt was also made to relate the position of N-methyl groups to the phosphorus lone pair in PPh(C1)-(NMe,) (ref.1) and in the cyclodiphosphazane Me,N-P*NBut*PCl-NBut (ref. 38) by the Nuclear OverhauserEffect . Homonuclear double-resonance experimentsinvolving irradiation of the phenyl and t-butyl protonsrespectively at low temperatures had no discernibleeffects on the N-methyl signals, which were non-equiv-alent because of slow rotation about the P-N bond.I 13 4-10 7 I I I0 90 180o/FIGURE 5 Graph of J ( P N C ) against 0 for ( a ) P,(NMe),, ( b )P(NMeNMe),, (c) ClbNBu t*PCFkMe, (d) ClbNIMeCH,CH,?d-M e , ( e ) ClPNMeCH,CH,NMe, (f) (ClPNBut), [ J ( P N C ) as-sumed positive], (g) Ph,P*NMe-P(S)Ph,, and (12) F,P-NMe-PF,_______77Hydrogen-1 and l3C n.m.r. data for other tervalentphosphorus-nitrogen compounds are listed in Table 2,where, in general, structural information is lacking.Inthe light of the foregoing discussion, Cl,P*NMe*PCl, (ref,39) would be expected to have a preferred conformationclose to that of its fluorinated analogue (0 ca. 180').13 Asimilar preferred conformation may be expected for thePIII-N bond in Cl,P*NMe*P( 0) C1,.40 The relativelysmall changes in J(PNCH) and J(PNC) on passing fromCl,P*NMe*PCl, to Cl,P*NMe*P(O)Cl, could be the resultof conformational or electronic effects (which are doubt-less interdependent anyway). The more negativeJ(PNC) in (PPh,) (NMePh) relative to (PCl,)(NMePh)could be viewed in terms of steric interactions resultingin a more preferred 0 ca. 180" conformation in the formermolecule. However, the effect of more s character in38 G.Bulloch, R. Keat, and D. G. Thompson, J.C.S. Dalton,1977, 99, 1044.40 R. Keat, J . Chem. Soc. ( A ) , 1970, 2732.41 H. A. Bent, Canad. J . Chem., 1960, 38, 1235.42 R. Jefferson, J. F. Nixon, T. M. Painter, R. Keat, and L.43 R. Keat, J . Chem. SOC. ( A ) , 1970, 1795.44 E. Niecke and W. Flick, Angew. Chew. Internat. Edn., 1973,J. F. Nixon, J . Chew. SOC. ( A ) , 1968, 2689.Stobbs, J.C.S. Dalton, 1973, 1414.12, 5851978 769pair on phosphorus (0 ca. 0').weight to this assignment.Our findings add furtherEXPERIMENTALContinuous-wave lH and pulsed-Fourier-transform 13Cn.m.r. spectra were obtained on JEOL C6OHL (60 MHz) andVarian XL- 100 ( 2 5 . 2 MHz) spectrometers respectively.Values of J ( P N C ) were obtained from spectra with a digitalresolution of 0.25 Hz or better. lH-(13C}, 1H-(31P}, and1H-(29Si) experiments on the C6OHL were carried out aspreviously described.l6 13C-( 1H,31P} triple-resonanceexperiments were performed using the XL-100; the de-coupler transmitter coil was double-tuned to accept 100MHz from the power amplifier of the Gyrocode decouplerand 40 MHz direct from the output of a ND-IOOM Schomandlfrequency synthesiser. Spectral simulations were calcul-ated using the Simeq I1 program of C. W. F. Kort andM. J . A. De Bie on the Varian computer of the XL-10049 W. Gerrard and H. R. Hudson, ' Organic Phosphorus Com-pounds,' eds. G. M. Kosolopoff and L. Maier, Wiley-Interscieiice,New York, 1973, vol. 5, pp. 222-223.50 F. Ramirez, A. V. Patwardhan, H. J. Kugler, and C. P.Smith, J . Amer. Chem. SOC., 1967, 89, 6276.spectrometer. The compounds PCl3_,(NMe2), (n = 1-3) ,49C12P.NMe.PC1,,3g F2P*NMe*PF2,39 Cl,P*NMe-P( 0) C12,40PCI,(NMePh) ,49 C1,P*NMe-SiMe3,42 Cl(Ph) P*XMe*SiMe,,4ZPPhCl(NMe,) ,I ClPNMeCH,CH,NMe,60 P(XMeNMe)3P,31y517 7P4( NMe) 6, 52 CIP*NBut *PCl*kMe, 63 (ClPN But) 2 , 42 Me,N-P*NBU~*PC~*NBU~,~~ PPh2(NMePh),49 Ph,P*NR-SiMe,[R = Me, Et, Pri, or But: b.p. 55-60 "C (0.1 mmHg)],43Ph,P.NR*PPh, (R = Me,64 E t , S 4 or Pri 16), and Ph,P.NMe*P(S)Ph, 43 were prepared by literature methods.We thank the S.K.C. for the award of a studentship (toG. B.) and assistance with the purchase of n.m.r. equipment,and Dr. P. Mallinson for the calculation of dihedral anglesfrom fractional co-ordinates.[7/1559 Received, 31st August, 1977151 R. Goetze, H. Noth, and D. S. Payne, Chem. Ber., 1972,52 R. R. Holmes, J . Amev. Chem. Soc., 1961, 83, 1334.53 G. Bulloch and R. Keat, J.C.S. Dalton, 1974, 2010.5 4 G. Ewart, A. P. Lane, J. McKechnie, and D. S. Payne, J .105, 2637.Chern. SOC., 1964, 1543
ISSN:1477-9226
DOI:10.1039/DT9780000764
出版商:RSC
年代:1978
数据来源: RSC