摘要:

J. CHEM. soc. DALTON TRANS. 1986 1333Reactions of Halogenohydridobis(triethylphosphine)platinum(~~) withTrihalogenophosphines: Some Unusual Complexes of Pt" containing Five-co-o rd i na t ed PhosphorusChristopher W. Allen, E. A. V. Ebsworth," Steven G. Henderson, David W. H. Rankin.Heather E. Robertson, Bryan Turner, and John D. WhitelockChemistry Department, Edinburgh University, West Mains Road, Edinburgh EH9 3JJ,'P N.m.r. spectra show that [PtCIH( PEQ,] reacts with PCI, in CH,CI, at 180 K to give[PtCI,H,(PEt,),], (5), and the previously unknown [PtCI(PEt,),(P'CI,)], (4). On warming to 240 K,complex (5) decomposes and (4) is reversibly protonated to give [PtCI( PEt,),( P'CI,H)] +. Above240 K, (4) is converted into the novel [PtCI(PEt,),(P'CI,H,)], (8). This complex is stable insolution up to 260 K, but decomposes irreversibly at higher temperatures.Changes in the n.m.r.spectra on adding HCI or BCI, are interpreted in terms of dissociation of (8) into[PtCI(PEtJ,(P'CIH,)]+ and [HCI,]- or [BCIJ-. Reaction of [PtBrH(PEtJ,] with PBr,at 160 Kgives [ PtBr,H (PEt,),( P'Br,)], which loses H Br at 1 90 K, giving [ PtBr( PEt,),( P'Br,)] and[PtBr,H,( PEtJ,]. At 240 K, [ PtBr( PEt,),( P'Br,H,)] is produced. This persists for short periodsat room temperature, and is stable at 170 K; addition of HBr or BBr, leads to the formation of[PtBr( PEtJ2( P'BrH,)] + by loss of bromide ion. All these species were identified by n.m.r.spectroscopy .We have recently'.' prepared complexes containing PCl, orPF, groups bound to six-co-ordinated 19" by the oxidativeaddition of PCl, or PClF, to tr~ns-[Ir(CO)Cl(PEt,)~]. Thesecomplexes are unusual in that they contain three-co-ordinatedphosphorus ligands, and so can be regarded as phosphines withone substituent of exceptional steric and electronic character;they can be used to prepare mixed-metal species with PF,bridges, and they can be converted into complexes containingunusual ligands such as PH,Se by reaction at the PCl, or PF,groups.' Analogous complexes of Pt' would have additionalmodes of reaction open to them, through the possibility ofaddition to the metal centre; their synthesis and chemicalproperties are therefore also of interest.Unfortunately, how-ever, reaction between PCIF, and trms-[PtClH(PEt,),] gives 'the binuclear complex [(Et3P)2C1Pt(PF,)Pt(PEt3)C12] con-taining a PF, bridge in which P is four-co-ordinated.We havetherefore investigated the reactions between PCI, or PBr, and(3), to try to prepare platinum complexes of three-co-ordinatedP and to study their reactions.CPtC1H(PEt,),f, (I), CPtBrH(PEt,),l, (21, or [PtHI(PEt,),I,ResultsReaction between PCI, and complex (1) in CH,Cl, is rapid at180 K, and several products are formed as the solution isallowed to warm. The products depend on the initial ratios ofreagents taken. With equimolar proportions, the process maybe represented by Scheme 1. Most of the products are unstablein the reaction mixture and have not yet been isolated, but theyhave been identified by their n.m.r.spectra, using variationsin the reaction conditions. Additional information has beenobtained from reactions between PCl, and complex (2) or (3),and between PBr, and (2). The system will be considered instages.The Initial Reaction.-The ,lP-(lH} n.m.r. spectrum of anequimolar mixture of complex (1) and PCl, in CH2Cl, showedno trace of (1) after an hour at 190 K. Peaks were observed dueto the novel complex tr~ns-[PtCl(PEt~)~(P'Cl~)], (4), and to[PtCl,H,(PEt,),], (5) [equation (l)]. The n.m.r. parameterstrans-[PtClH(PEt,),] + PCl, - trans- [PtCl PEt 3)2( P'CI )](4)+ CPtC12H,(PEt,),l (1)(5)of complex (5) are well known.4 The spectrum of (4) showedtwo resonances. One was a triplet C2J(PP) = 59.1 Hz] with Ptsatellites ['J(PtP') = 465.4 Hz) at very high frequency (6 =31 1.1 p.p.m.), a region associated with metal-P'CI, group^.^.^The other was a doublet (6 = 20.4 p.p.m.) with the samecoupling as the triplet splitting in the high-frequencyresonance, and with platinum satellites.The chemical shift istr~ns-[PtClH(PEt,)~] + PCl,CH,Cl, 180K+HC1. c240 K [ PtCl( PEt,),( P'Cl,H)] +(4) (Wi [PtCI,H,(PEt,),] + tr~n~-[PtCl(PEt,),(P'Clz)] . -HC,(7) (8)(5) 1 2 4 0 ~trans-[ PtCl,( PEt ,)J + cis-[ PtCl,( PEt ,),I + [ PtCl( PEt 3)2( P'Cl,H,)](6)Scheme 1J. CHEM. SOC. DALTON TRANS. 1986 1334T a k N.m.r. parameters for complexes obtained in this work, with standard deviations in parenthesesW)3 1 1.1 (5)307.4( 1 )298.0(1)320.2( 1)277.0(1)62.7( 1)42.2(2)-71.0(10)- 16.14(2)- 128.0(10)- 53.5( 1)W )204 1 )18.3( 1)13.4(1)17.7( 1)23.0(2)15.5( 1)12.48(2)17.34(2)4.22(2)11.83(2)4.3(1) -W )----15.04(2)8.69(2)8.29(2)6.22(2)6.38(2)5.83(3)5.66(2)‘J(PtP)2 638.5(3)2 630(2)2 622(2)2 123(2)1741(2)1 943(5)2 083(5)2 135(2)1989(2)2 168(3)1974(2)* J(PtP‘)465.4( 2)245.8(3)185.6(2)167.0(2)4 750( 10)4 830(10)4 725( 10)370(2)4 121(10)5 002( 10)4 112(5)Shifts are given to high frequency of SiMe, (‘H) or 85% H3P04 (3’P) for solutions in CD,CI,.‘J(P’H) ’J(PP’) ’J(PtH) 3J(PH)- 59.1(2) - -- 59.1(3) - -- 59.6(2) - -- 69.9(2) - -- 44.9(2) - -550.2(2) 23.2(2) 235(1) 9.5( 1)soO.5(2) 23.9(2) 218(5) 10.5(5)555(1) 30.5(5) 109.2(4) 8.4(2)487(2) 29.5(5) 112(1) 9.2(4)569(1) 27.7(2) 117.3(2) 8.0(2)484.7(5) 27.2(2) 112.8(2) 8.8(3)TIK180180180176176180200230230250180characteristic of PEt, complexes, and the value of ‘J(PtP),2 638.5 Hz, shows that the complex contained four-co-ordinated Pt.The resonance due to P was not affected whenproton coupling was restored, and in the proton resonancespectrum the only PtH peaks were those due to complex (5).These observations show that (4) was [PtCl(PEt,),(PCl,)]. Asthe solution was allowed to warm, this complex reacted further(see below), but it has been obtained as the only Pt-containingproduct from reactions between (1) and a large excess of PCl,,or from (l), E l , , and NMe, in 1 : 1 : 1 molar ratio. We havebeen able to obtain this complex as an oil, and will report onits reactions in more detail elsewhere.Its n.m.r. parameters aregiven in the Table.The products of the initial reactions between PCl, andcomplex (2) or (3) were very unstable, and we have not been ableto characterise them so fully. The 31P-(1H) n.m.r. spectrum ofan equimolar solution of (2) and P a , in CH2Cl, showed threesinglet resonances with platinum satellites in the PEt, region;we assign these on the basis of coupling constants and chemicalshifts4 to [PtCI,H,(PEt,),], [PtBrCIH,(PEt,),], and[PtBr,H,(PEt,),]; the relative intensities were roughly1:40:40. In addition there were two triplet resonances withplatinum satellites in the high-frequency region, each associatedwith a doublet with platinum satellites in the PEt, region.Oneset of doublets and triplets corresponds with the spectrum ofcomplex (4). The parameters of the other set, which was initiallyabout twice as strong, were similar, but the chemical shifts andcoupling constants were rather smaller. We identify this species,(9), as trans-[PtBr(PEt,),(P’CI,)]. The chemical shifts wereconsistent with PtBr(P’C1,) rather than with PtCI(P’Br,),particularly when taken with those of the products of thereaction between (2) and PBr, (see below), and the value of*J(PtP) shows clearly that Pt was four- and not six-co-ordinated; no additional PtH resonance was observed in theproton resonance spectrum. It appears that the initial reactionis like that between complex (1) and PCl,, but that halogenexchange has taken place, favouring PtBr in six-co-ordinatedcomplexes [equation (2)].The close correspondence of thet runs-[ Pt BrH( PEt3)2] + PCI , + trans-[ PtCl( PE t j ) Z -( P’C1 ,)I + trans- [ P t B r ( P E t , ) ( P ’ C 1 ,)I + [ Pt C1 , H , ( P E t , ) ,]+ [PtBrClH,(PEt,),] + [PtBr,H,(PEt,),] (2)parameters of one of the products to those of complex (4), andthe observation of five sets of resonances, show that anyexchange of halogen at Pt or P’ must be slow on the n.m.r. time-scale at this temperature.Reaction between complex (3) and PCI, under the sameconditions gave a similar mixture of products, except that in thissystem there was about ten times as much [PtH,I,(PEt,),] as[PtClH,I(PEt,),] and we did not detect (5).As before, therewere two sets of high-frequency triplets with associated PEt,doublets; one, the stronger, was assigned from its parameters to(4), and the other to trans-[PtI(PEt,),(P’Cl,)], (10). The n.m.r.parameters are given in the Table.Reaction between PBr, and complex (2) was somewhatdifferent; by working at very low temperatures we were able todetect an unstable platinum(1v) complex formed by oxidativeaddition of PBr, to (2). If the reaction was allowed to occur inCH,Cl, at temperatures above 200 K, we observed no P’resonances in the high-frequency region. If the solution was notallowed to warm above 176 K, however, there were two tripletresonanm at about + 300 p.p.m., each with platinum satellitesand associated PEt, doublets.We assign one set of resonancesto trans-[PtBr(PEt,),(P’Br,)J, (1 l), on the basis of its n.m.r.parameters. The chemical shifts of the second species, (12), weresimilar, but the couplings to Pt were much smaller, suggestingthat the Pt might be six-co-ordinated. In the proton resonancespectrum there was an additional PtH triplet with platinumsatellites at - 15.04 p.p.m., a shift associated with H transto halogen4W6 in complexes of six-co-ordinated Pt”. We wereable to show by heteronuclear double resonance that the P’resonance, the PEt, resonance, and the new hydride resonancewere due to the same species; though ’J(P’PtH) was notobserved, irradiation at the PEt, frequency collapsed the tripletin the PtH resonance. We therefore identify complex (12) as[PtBr,H(PEt,),(P’Br,)], formed by oxidative addition of PBr,to (2) [equation (3)].The only other resonances were due to (2)[PtBrH(PEt,),] + PBr, ---, [PtBr,H(PEt,),(P’Br,)] ( 3 )and to [PtBr,H,(PEt,),]. The proportion of (12) formed wasreduced if a large excess of PBr, was taken. As the solution wasallowed to warm to 200 K, the peaks due to (12) shrank andvanished; those due to (11) persisted to 200 K, but disappearedat higher temperatures.The Second Stage (Protonation of[PtCl(PEt,),(P’Cl,)]}.-Ifthe equimolar solution of PCl, and complex (1) from which (4)was obtained was allowed to warm to temperatures between180 and 240 K, the triplet due to P’ broadened greatly andshifted to lower frequency; at 40 MHz the high-frequencysatellite could still be observed, but the low-frequency satellitebecame too broad to detect.The PEt, resonances changedrelatively little. As the temperature approached 240 K, thechanges in the P’resonance began to be reversed: the main peaJ. CHEM. soc. DALTON TRANS. 1986 1335moved back towards 300 p.p.m., and the satellites moved with it.These peculiar changes can be understood from the results ofadding HCl. If a two-fold molar excess of HCl was added to thesolution at 180 K, the P resonance shifted to 62.7 p.p.m. andbecame a relatively sharp triplet with platinum satellites; whenproton coupling was restored, this peak showed a wide doubletsplitting ['J(P'H) = 550.2 Hz]; the changes in chemical shiftand coupling constant for the PEt, resonance were muchsmaller.There is clearly an equilibrium between (4), HCl, andthe protonated species (4.) [equation (4)]. At low temperatures[PtCl(PEt,),(P'Cl,)] + HCl G+(4) [PtCl(PEt,),(PCl,H)] + + C1- (4)in the presence of excess of HCl the equilibrium lies to the right.With a deficit of HCl there is a fast exchange between (4) and(4a), giving average positions for the P resonance and for eachof its satellites; as the temperature is raised, the equilibriumshifts to the left, perhaps through evaporation of HCl from thesolution. Addition of a 1:l mixture of HCl and BCl, to asolution of complex (4) gave a solution whose ,IP-{ 'H} and 'Hn.m.r. spectra corresponded with those of (4a); the 'B spectrumshowed the sharp (w - 2 Hz) line at 7.6 p.p.m.associated with[BCl,]-. The n.m.r. parameters for complex (4a) are given inthe Table.Solutions of PBr, and complex (2) at 200 K gave resonancesthat resembled those due to (&), and these have been assignedby analogy to protonated (ll), i.e. (11a); the parameters aregiven in the Table. The amount of (lla) in these solutions wasalways small, however, and was not significantly increased byadding HBr; further reaction occurred much more readily in thebromide system, giving products described in the next section.Formation of Species containing Five-co-ordinatedPhosphorus.-If the equimolar solution of PCl, and complex(1) was allowed to warm to temperatures between 240 and 260K, the resonances due to (4), (&), and (5) slowly disappeared,and new resonances grew in their place; this happened faster andat lower temperatures if the PCI, was not rigorously freed fromHCI.Among the new resonances, peaks due to complexes (6)and to (7) were identified from their chemical shifts and from'J(PtP). In addition to these, there were peaks that we assign toa most unusual and unexpected complex, (8). These consisted ofa doublet with platinum satellites in the PEt, region, and atriplet with platinum satellites at -71 p.p.m. The very lowchemical shift suggests2*' Pv rather than P". When protoncoupling was restored, the resonance at - 7 1 p.p.m. split into awide triplet ['J(P'H) = 555 Hz]; in the proton resonancespectrum, there was a wide doublet of triplets with platinumsatellites [S = 6.22 p.p.m.'J(PtH) = 109.2; ,J(PH) = 8.4 Hz].The P resonance was reasonably sharp if the temperature waskept between 200 and 260 K; at higher temperatures itbroadened significantly, but sharpened again on recooling if thesolution was not kept at temperatures above 260 K for long. Theproton resonance was sharpest at 260 K, and broadened as thetemperature was lowered. If the solution was kept at highertemperatures, peaks due to complex (8) shrank and vanished,and the end-products were (6), (7), and [PEt,H]+.The n.m.r. spectra show the structure of the co-ordinationskeleton of the complex. The value of 'J(PtP) (2 135 Hz) showsthat the platinum atom is four-co-ordinated,, and the proton-coupled P' resonance shows that P is bound to Pt and to two Hatoms and is at least four-co-~rdinated;~ the value of 'J(P'H) israther large for four-co-ordinated P bound to two H atoms anda metal (though not impossibly so), but it seems quite consistentwith five-co-ordinated P'.There are two possible structures for(8); the complex might be molecular, with five-co-ordinated P',or cationic, (8a), with four-co-ordinated P'. Experiments withl+HPEt3 CI I I HCI-Pt-P" CI-Pt -P'/-CI CI''" I \H I CIPEt3 PEt3I "LI '" I P'BCl, and with HCl lead us to conclude that the complex is bestrepresented by the molecular form (8). If PCl, and complex (1)are allowed to react together in the presence of an equimolarproportion of BCl, or a four-fold excess of HCl, the reactionstops at (4a); the further reaction to give species based on (8) isinhibited.If, however, PCl, and (1) are allowed to react until (8)has been formed, and an equimolar amount of BCl, is thenadded, there is a marked change in the n.m.r. parametersassociated with (8). The resonance due to P shifts from - 7 1 to- 16.1 p.p.m., and there are large reductions in 'J(PtP') and in'J(PH), with smaller changes in other parameters. The "Bresonance is at 7.4 p.p.m., close to the value for [BCl,] -, thoughthe line is broader (w - 20 Hz) than usual for this species. Wesuggest that BCl, has removed C1- from the equilibriumbetween (8) and (&), generating the cation (&). Addition of atwo-fold molar excess of HCl to a solution containing complex(8) leads to similar but smaller changes; for instance, S(P') is- 36 p.p.m.; we suppose that HCl acts as a weaker acceptor forC1-, forming [HClJ-, and so does not shift the equilibriumfully to the right.Addition of a large excess of NPr,Cl to asolution of PCl, and complex (1) before formation of (8) alsoinhibits formation of (8), but addition to a solution alreadycontaining (8) does not lead initially to significant changes inthe mm.r. parameters of (8), though on standing at 220 K thepeaks due to P broaden and on prolonged standing at thistemperature the spectrum changes further. We conclude that, inthe absence of a chloride-ion acceptor, complex (8) is present inCH,C12 in the molecular form, with little or no dissociation into(&) and Cl-. The parameters in the Table for the molecularspecies are those taken from such a solution, and those for (&)are taken from the solution to which an equimolar proportionof BCl, had been added.The chemical shift for P' in molecular(8) varied by 1-2 p.p.m. from one solution to another, and byabout the same amount with a change in temperature of 40 K.Experiments with PBr, and complex (2) supported theseconclusions. When the solution containing PBr, and (2) wasallowed to warm above 220 K, peaks due to a complexanalogous to (8), i.e. (13), appeared and rapidly grew stronger.They remained relatively sharp even at 280 K for short periods,suggesting that (13) is more stable than (8). Addition of HBr at200 K did not alter the general pattern of the spectrum, butshifted the resonance due to P'; with a two-fold proportion ofHBr the resonance shifted from - 128 to -65.6 p.p.m., and witha four-fold proportion it shifted to -60.9 p.p.m., with largedecreases in 'J(PtP') and smaller changes in other parameters1336 J.CHEM. SOC. DALTON TRANS. 1986We presume that added HBr has led to the formation of thecation (1%) and [HBr,]- [equation (5)]. Addition of an[PtBr(PEt,),(P'Br,H,)] + HBr e[PtBr(PEt3)2(PBrH2)] + + [HBr,] - ( 5 )equimolar amount of BBr, (based on Pt initially taken) to thesolution of PBr, and complex (2) at 200 K gave a solution withseveral unidentified resonances in the "P-('H) n.m.r. spectrum,but peaks due to (13a) were strong and well defined; that due toP' had shifted to -53.5 p.p.m., with a corresponding drop in'JfPtP') and in 'J(P'H) and smaller changes in otherparameters.The 'B n.m.r. spectrum showed a fairly sharp line( w - 7 Hz) at - 23.3 p.p.m., assigned '*lo to [BBr,]-. Additionof a four-fold molar proportion of NPr,Br (based on Pt initiallytaken) did not perturb the n.m.r. parameters of complex (13)significantly. We conclude that (13) is present in CH,C1, in themolecular form unless a bromide-ion acceptor is added; theparameters for (13) in the Table are from the solution inCH,Cl,, and those for (13a) are from a solution to which BBr,had been added.Reacting Ratio.-It is hard to determine the stoicheiometryof the reactions that give complex (8), because of paralleldecomposition reactions forming (6) and (7).In the bromidesystem, PBr, is a difficult material to transfer quantitatively, butcareful experiments using a standard solution of PBr, inCH,Cl, showed that excess of PBr, was present after prolongedreaction at 250 K unless the starting ratio of PBr, : (2) was lessthan 112. This implies that the overall reaction to form (13) maybe written as in equation (6). However, complex (13) wasPBr, + 2[PtBrH(PEt,),)] -[PtBr2(PEt3),] + [PtBr(PEt,),(P'Br,H,)] (6)unstable at 260 K in the absence of excess of PBr,.Reactions with an Excess of [PtHX(PEt,),].-Reactionbetween PCI, and an excess of complex (1) in CH,Cl, gave (4),(5), and (1) at temperatures up to 300 K; no further productswere detected on standing. The reaction between PBr, and anexcess of complex (2) was more complicated.At 175 K,resonances were detected due to (2), to [PtBr,H,(PEt,),], to(ll), and to (12); in this system, the resonances due to (12) fadedquickly at 180 K, but those due to (11) persisted to 250 K. Whenthe solution was shaken for about a minute at roomtemperature and then cooled again to 250 K, peaks due tocomplex (1 1) became very weak indeed; resonances wereobserved due to cis- and trans-[PtBr,(PEt,),], to (2), to[PtBr,H,(PEt,),], to [Pt(PEt,),H]+, and to a small amount ''of [( Et,P),BrPt(P'H,)PtBr( PEt,),] + .Attempts to Fluorinate P' in [ PtBr( PEt 3)2( P'Br,H,)].-Inorder to try to demonstrate directly that P' in complex (13) wasbound to two halogen atoms as well as to two H atoms and Pt,we tried to fluorinate it with GeMe,F, a mild fluorinating agentthat was unlikely to react in other ways with the rest of themolecule.Treatment of a solution in dichloromethanecontaining (13) with GeMe,F in equimolar proportions or inexcess led to very complicated reactions at low temperatures;HF was eliminated, and many platinum complexes wereformed. The results did not help to confirm the formulation of(1 3).DiscussionThe most unexpected process in the systems we describe here isthe formation of the species (8) and (13) which we formulate asplatinum(r1) complexes of five-co-ordinated phosphorus. Theinitial reactions between (1) or (2) and the correspondingphosphorus(rr1) halide present no surprises. The detection of thesix-co-ordinated platinum species (12) at very low temperaturesin the reaction between (2) and PBr, confirms that the initialstep involves oxidative addition of P-Br to Pt", as in theanalogous reaction between PX, and [Ir(CO)X(PEt,),], butwith Ir the product is stable to reductive elimination, and thereaction stops at this stage; in the platinum systems, the initiallyformed six-co-ordinated complex eliminates HX, which is takenup by excess of [PtHX(PEt,),] to give [PtH,X,(PEt,),].Thisshows that the substitution of H by PX, reduces thesusceptibility of Pt" to oxidative addition. The stability of[PtY(PEt,),(P'X,)] in the reaction solutions is very sensitive toboth X and Y; indeed, the only product of this type stableenough for us to isolate it is (4). We intend to study the reactionsof complex (4) in more detail; here we are concerned with thebehaviour of (4) and its analogues in the reaction system.It is hard to see how species of the type [PtX(PEt,),(PX,)]are transformed into [PtX(PEt,),(P'H,X,)].The transform-ation must involve HX, yet addition of an excess of HCl to thereaction mixture of (4) and (5) inhibits the formation of (8). Wesuggest that this is because an excess of HCl protonates P', andit follows that protonation is not involved in transforming P'CI,into P'Cl,H,. Moreover, [Ir(C0)C1,(PEt,),(PCl2)] does notreact with HCl below 240 K, and at higher temperatures thereaction involves displacement of PCl, with formation of IrHrather than protonation of P'Cl,. While P'CI, is likely to bemore electron-rich when bound to Pt" than to Ir'", we believethat 'P-{ 'H) n.m.r.spectra of equimolar solutions of complex(1) and PCI, show that the PCl, group of (4) is protonatedreversibly between 200 and 250 K with no sign that this is a stepin the formation of (8). It seems more likely that the metal atomis involved in this reaction. It is possible that HCI might addreversibly to (4) if the concentration of HCI were maintained atthe right level; this addition could be followed by elimination ofPC1,H and rapid readdition (see below). We should have tosuppose that the bound PClH added HCI so fast that we couldnot detect PC1,H either bound or free. We know thatplatinum(rv) complexes such as (5) lose HX reversibly, but thatdissociation is slight; the presence of (5) might thus impose aclose control on the concentration of free HCl (Scheme 2).It is worth noting that complex (13) is formed much morereadily than (8) is; this implies that the metal is likely to beinvolved in the conversion of P'X, into P'H,X,, since PtBr ismuch more readily formed than is PBr.Since we have not beenable to isolate either (8) or (13), our formulation of them ascomplexes of five-co-ordinated phosphorus must rest on adetailed analysis of the n.m.r. parameters. There is no doubtthat the ligands contain two P'-H bonds and 'J(P'H) is so largethat P' must be either four- or (more probably) five-co-ordinated; there is no direct evidence to show which it is, but webelieve that our experiments with BCl, make it very probablethat the molecular formulation is correct.As would have beenexpected on ionisation, P' shifts to high frequency, and 'J(P'H)drops to a value fully consistent with four-co-ordinated P'. Wealso believe that the formation of [(Et,P),BrPt(P'H,)PtBr-(PEt,),]' in the system containing PBr, and an excess of (2)is more consistent with the formation of molecular (13). It ispossible to envisage transfer of Br, from P' to (2); the resulting[PtBr(PEt,),(P'H,)] could then displace Br- from [PtBr,-(PEt,),] to give the PH,-bridged species. Loss of Brf from thecation (13a) seems much less likely, particularly if (13a) hadbeen formed by loss of Br- from (13).The observations we have made show that in the species (8)and (13) the unique phosphorus atom P' is at least four-co-ordinated.It is conceivable that (8) and (13) contain five-co-ordinated Pt rather than P', and should be written aJ. CHEM. soc. DALTON TRANS. 1986 13371 - I /ctI H’ I ICI-Pt-P‘CIZ + HCI + CI-Pt-P’Clz P‘ClzH +IPtCI2(PEt3)2]PEt3 PEt3Scheme 2.PEt3[PtX,(PEt,),(P’H,X)]. Such species would be expected to losehalide ion from Pt on the addition of halide-ion acceptors.However, we do not believe that loss of halide ion from themetal would lead to the very large changes in P’ n.m.r.parameters on the addition of BX, or HX, and so we regard thisformulation as improbable. We cannot exclude the possibilitythat these species are anionic and contain six-co-ordinated P’,being of the form [PtX(PEt,),(P’H,X,)]-. There are noanalogous compounds on whose n.m.r.parameters we couldbase a distinction between these anions and the molecular (8)or (13). The value of ‘J(PH) drops slightly l 2 from PF,H toPF,H-; ’J(P’H) is relatively large in (8) and in (13), whichseems rather more in keeping with the molecular representation.Moreover, there is no obvious cation that could be present as acounter ion. But until we isolate either complex (8) or (13), thispossibility is likely to remain, though in the context of generalchemistry of Pv we regard it as improbable.N.M. R. Parameters.-The very high-frequency shifts for P X ,ligands in these platinum complexes are fully in keeping withwhat has been observed in related complexes of other metals.The P’ resonance moves to lower frequency when for a given Xthe halogen on Pt is changed from C1 to Br to I; there is,however, a substantial shift to high frequency between (4) and(11).Oxidation at Pt shifts P’ to low frequency. The mostunusual parameters in this group of complexes, however, are thevalues of ‘J(PtP’), which are exceptionally small. In mostcomplexes of Pt” in which P is also four-co-ordinated, ‘J(PtP)is over 2 OOO Hz; here, with three-co-ordinated P’, ‘J(PtP’)ranges from 465.4 to 185.6 Hz. For P’CI, species, this couplingconstant drops as the halogen trans to P’ changes from C1(465.4) to Br (370) to I (245.8 Hz); since the two ligands involvedshare a common orbital, so large a change is understandable.There is a further drop in ‘J(PtP’) as Pt is oxidised, from 185.6in complex (11) to 167.0 Hz in (12).This drop is smaller thanexpected for a change in co-ordination from four to six, imply-ing that in these systems at least ‘J(PtP’) is affected byother factors. In platinum(I1) complexes with four-co-ordinatedP trans to halogen, ‘J(PtP) is around 3 W OOO Hz; in ourcompounds, ‘J(PtP’) with P’ trans to halogen is around 4 OO@-5 OOO Hz, whether P’ is formulated as four- or five-co-ordinated.The parameters associated with P’ in complexes (8), (13), @a),and (13a) have already been discussed in some detail. The valuesfor the proton chemical shifts are reasonable for such species,and ‘J(PtP’), though large, is not exceptionally so in any ofthese species. It is worth noting that the chemical shift of P‘ inthe chloride species (8) and (8a) is to high frequency of its valuein the corresponding bromide; the shift from chloride tobromide is in the same direction l 3 in PX, or POX,, but in theopposite direction in PX, and in PtP’X,, emphasising thatspecies derived from (8) and (13) contain oxidised phosphorus.ExperimentalVolatile compounds were manipulated using vacuum systemsfitted with greased glass or greaseless Sovirel taps; air-sensitiveinvolatile materials were transferred under dry N, in glove-bagsor in a Vacuum Atmospheres glove-box; purity was checkedspectroscopically and (where appropriate) by measurement ofvapour pressure.N.m.r. spectra were recorded using JEOLFX60Q (for 31P), Bruker WP80A (for ‘H and 19F), WP200 andWH360 (for H and 3 1 P) spectrometers.Platinum starting materials were prepared by standardmethods.Boron and phosphorus trihalides were purchased.Phosphorus trichloride was purified in the vacuum system andmeasured by direct weighing. Phosphorus tribromide was notvolatile enough for this; it was purified by bubbling dry N,through it for 5 min, followed by freeze-degassing, andmeasured using a standard solution in CD,CI,: roughly therequired amount was transferred to CD,Cl, in a calibrateddispenser vessel in a glove-bag, using a syringe, and the con-centration determined gravimetrically by precipitation of AgBrafter hydrolysis. Typical experiments are described below.(a) Reaction of Complex (1) with PCI,.-An n.m.r. tubecontaining complex (1) (0.05 mmol) was attached to a vacuumsystem by a ground-glass joint and CD,CI, was added, followedby a weighed amount of PCl,.The solution was kept at 90 K,and allowed to melt in the probe of the n.m.r. machine; spectrawere obtained over a range of temperatures.(b) Reaction of Complex (1) with PC1,followedby HC1.-Thereaction mixture ofcomplex (1) and PC1, was prepared as in ( a ) ,and kept at 200 K (30 min) before a measured amount of HCIwas added; the tube was sealed, and spectra recorded.( c ) Reaction of Complex (8) with HCI.-An n.m.r. tube wasmade with a glass breakseal, as well as the usual ground-glassjoint. A reaction mixture of complex (1) and PCl, was preparedas in (a), and the tube was sealed. The spectrum was monitoredas the solution warmed until strong resonances due to (8) wereobserved.The contents of the tube were then frozen and theappropriate amount of HCl added through the breakseal. Thetube was then resealed and the spectrum of the productsobtained.( d ) Reaction of Complex (8) with NPr,Cl.-A solution ofcomplex (1) and PCl, was prepared in an n.m.r. tube with 1338 J. CHEM. soc. DALTON TRANS. 1986breakseal as described in (c), and the reaction was allowed tocontinue until strong resonances due to (8) were observed. Aweighed amount of NPr,Cl was then added through thebreakseal under vacuum, giving a ratio NPr,Cl: PLI, of 4: 3,and the tube was resealed.(e) Reaction of Complex (2) with PBr,.-The contents of ann.m.r. tube containing complex (2) (0.05 mmol) in a 1 : 1 mixtureof CD,Cl, and Et,O were frozen and PBr, added either inrough quantity using the vacuum system or in accuratelymeasured amounts using the standard solution and thedispenser. Spectra were recorded at temperatures from 150 K.AcknowledgementsWe are grateful to Messrs. Johnson Matthey for lending uschemicals. C. W. A. thanks the Department of Chemistry,University of Vermont, for leave.References1 E. A. V. Ebsworth, N. T. McManus, D. W. H. Rankin, and .I. D.Whitelock, Angew. Chem, 1981, 93, 9.23456789101112E. A. V. Ebsworth, R. 0. Could, N. T. McManus, N. J. Pilkington,and D. W. H. Rankin, J . Chem. SOC., Dalton Trans., 1984, 2561.E. A. V. Ebsworth, D. W. H. Rankin, and J. D. Whitelock, J . Chem.SOC., Dalton Trans., 1981, 840.D. W. W. Anderson, E. A. V. Ebsworth, and D. W. H. Rankin, J.Chem. Soc., Dalton Trans., 1973, 854.W. Malisch and R. Alsmann, Angew. Chem., Int. Ed. Engl., 1976, 15,769.I. M. Blacklaws, L. C. Brown, E. A. V. Ebsworth, and F. J. S. Reed, J.Chem. SOC., Dalton Trans., 1978, 877.N. J. Pilkington, Ph.D. Thesis, Edinburgh, 1984.J. R. Granada, G. W. Stanton, J. H. Clarke, and J. C. Dore, Mol.Phys., 1979, 37, 1297.R. R. Holmes and R. N. Stores, Inorg. Chem., 1966, 5, 2146.R. K. Harris and B. E. Mann, ‘N.M.R. and the Periodic Table,’Academic Press, London, 1978.E. A. V. Ebsworth, B. J. L. Henner, and F. J. S. Reed, J. Chem. Soc.,Dalton Trans,, 1978, 272.P. M. Treichel, R. A. Goodrich, and S. B. Pierce, J. Am. Chem. Soc.,1967, 89, 2017; A. H. Cowley, P. J. Wisian, and M. Sanches, Inorg.Chem., 1977,16,1451.Phosphorus Chem., 1976,5, 227.13 V. Mark, C. H. Dungan, M. M. Crutchfield, and J. R. van Wazer, Top.Received 24th July 1985; Paper 5 J 126

ISSN:1477-9226    

DOI:10.1039/DT9860001333

出版商:RSC    

年代:1986

数据来源: RSC