摘要:
J. CHEM. SOC. DALTON TRANS. 1991 949Stability of (Chloromethyl)platinum(ii) ComplexesRobert McCrindle," Gilles J. Arsenault, Anuradha Gupta, Mark J. Hampden-Smith,Richard E. Rice and Alan J. McAleesGuelph- Waterloo Centre for Graduate Work in Chemistry, Guelph Campus, Department of Chemistry andBiochemistry, University of Guelph, Guelph, Ontario N I G 2 W7, CanadaThe stabilities of [Pt(CH,Cl),(cod)], [Pt(CH,CI)Cl(cod)] (cod = cycloocta-l,5diene) and a range ofphosphine-containing mono- and cis-bis- (chloromethyl)platinum( 1 1 ) complexes have been investigatedin deuteriochloroform at room temperature. Some of t h e bis(chloromethy1) derivatives appear to beindefinitely stable (cod and chelating arylphosphines), others suffer very slow decomposition to thedichlorides (non-chelating arylphosphines), and the remainder decompose relatively rapidly, andcleanly, to the dichlorides plus ethylene (alkylphosphines, non-chelating faster than chelating).Rapiddecomposition of the arylphosphine complexes can be induced by adding hexafluoroisopropyl alcoholto the deuteriochloroform solutions. Attempts to generate [Pt(CH,CI),{P(C,H,,),),1 by addition ofP(C6Hl1), to [Pt(CH,Cl),(cod)] resulted in the formation of cis- [Pt-{CH,CH,P+(C6Hl,),)CI,(P-(C6H,l)3}]; a mechanism is proposed. All cis-mono(chloromethy1) derivatives studied appear tobe indefinitely stable. In contrast, the trans-mono(chloromethy1) complexes, although stable in verydry solvent, undergo decomposition in the presence of moisture to the corresponding hydrides plusformaldehyde; a mechanism is proposed.The hydrides undergo subsequent conversion into a mixtureof cis and trans dichlorides.In the course of our studies of the preparation of (halogeno-methyl)platinum(rI) complexes it became clear that some ofthe products showed a much greater tendency than others todecompose when left standing in solution. Thus, as a part of ageneral investigation of the chemistry of these complexes, astudy of their solution stabilities at room temperature wasinitiated in the hope of determining the mode(s) of suchdecomposition. Prior to the 1980s cr-halogenomethyl derivativesof transition metals had the reputation of being 'notoriouslyunstable',2 although many members of this class are nowknown3 Some of these compounds have been reported to belabile, readily losing the CH, moiety to give the correspondingmetal halides.In most of these cases there is no mention of thefate of this moiety but in one case5 (Pt") it was presumed tobe lost as ethylene while in another6 (Ir') it was shown thatpolymethylene was the major product. The present investigationhas dealt with both mono- and bis-(chloromethyl)platinum(II)complexes, the latter representing, to our knowledge, the onlyknown derivatives of this type for the transition metals.Results and DiscussionDuring this investigation certain experimental results provideda forceful reminder that the presence of small amounts ofimpurities may have a marked effect on the apparent stabilityof a transition-metal compound. In several instances, chloro-methyl complexes freshly generated in reactions with diazo-methane underwent fairly rapid decomposition in the reactionmedium but decomposed much more slowly in a similar solventafter purification. In general, cleaner products, more amenableto further purification, are generated by ligand displacementfrom [Pt(CH,Cl),(cod)] 1 or [Pt(CH,Cl)Cl(cod)] 2 (cod =cycloocta-1,5-diene) both of which can be purified readily andare quite stable when pure.This route was used to produce thephosphine-containing complexes studied and, indeed, providesaccess to compounds which are either inaccessible by, or onlyminor products of, direct insertion of methylene into a Pt-CIbond by diazomethane. The stability of the complexes wasexamined for samples stored in the dark at room temperature,usually in deuteriochloroform solution.67891011192024n R1 R' = R2 = CH2CI 3 2 Ph2 R' = CHzCI, R2 = CI 4 3 Ph5 4 Ph12 2 C6Hll13 3 C6Hll14 4 C ~ H I ~n15 216 317 418cis-B~s(chloromethyl) Cornplexes.-All but one of the com-plexes of this type studied were bis(phosphine) complexes, theother being [Pt(CH,Cl),(cod)] 1.The last, when monitored by'H NMR spectroscopy, showed no evidence of decompositio950 J. CHEM. SOC. DALTON TRANS. 1991t,Pt, + C2H4 - CI- \ ,CI I+ CIScheme 1the bis(chloromethy1) complexesProposed mechanism for the production of ethylene fromwhen kept in solution for a month in acid-free solvent. In someexperiments slow conversion into [Pt(CH,Cl)Cl(cod)] 2 wasobserved, but this could be avoided by running the solventthrough a short column of anhydrous potassium carbonatebefore making up the solution.The three chelating aryl phosphine complexes 35, whencarefully purified, are also quite stable in solution at roomtemperature, remaining apparently intact for three months(sealed tubes in uacuo).When stored under air in capped tubes[Pt(CH,Cl),(PPh,),] 6 yielded a small amount (less than 1%)of the corresponding dichloride 7 during 2 weeks. In analogousexperiments the corresponding p-methoxy- and p-fluoro-phenylderivatives, 8 and 10, also produced small amounts of thecorresponding dichlorides, 9 and 11 slowly (ca. 5 and 2% during1 week). The fate of the CH2 moiety has not been investigatedin these cases.The three chelating alkyl phosphine complexes12-14 are substantially less stable than the compoundsdiscussed above, undergoing complete conversion into thedichlorides 15-17 during about 2 weeks. No significant peaksattributable to products from the CH, moieties could be foundin the 'H NMR spectra of these solutions, but analysis of thehead-space gases by GLC showed ethylene accompanied by alittle methane. Since in this (see below), and earlier work wehave found that some reactions in deuteriochloroform resultfrom the presence of water in the solvent, the decomposition ofcomplex 12 was monitored for a solution in deuteriochloroformthat had been saturated with water. However, no obviouschange in either the rate or the course of the reaction wasobserved.When tetrabutylammonium chloride (4 molarequivalents), rather than water, was added to a solution of 12 therate of decomposition to 15 almost doubled. Even more rapiddecomposition of 12 was observed when 1,1,1,3,3,3-hexafluoro-propan-2-01 (4 molar equivaledts) was substituted for theammonium salt, conversion into 15 being complete within 1 d.Indeed, the three chelating aryl phosphine complexes >5 alsosuffered quantitative conversion into the corresponding di-chlorides ('H NMR spectroscopy tubes sealed in vacuo) within1 d under similar conditions [CDC13-(CF3),CHOH, 9: 11.Exposure of a solution of complex 12 to room light had noapparent effect on its rate of decomposition; however, irradi-ation with 300 nm light accelerated the production of 15 andyielded a new product, the mono(chloromethy1) derivative 18.The bis(triethy1phosphine) complex 19 proved to be the mostlabile of all the compounds of this type studied, producing thedichloride 20 within 1 d.When this decomposition was allowedto proceed in a sealed NMR tube the 'H spectrum contained asignal attributable to ethylene. An estimate of the ethylenepresent in the tube was made based on the intensity of the signal,the volume of both the solvent and of the free space in the tube,and the solubility of ethylene in the solvent.8 This indicated thatessentially all of the methylene moiety lost in the production of20 could be accounted for as ethylene.The decomposition of these bis(chloromethy1)phosphinecomplexes to the dichlorides and ethylene may proceed uia areaction pathway (Scheme 1)* analogous to that proposed forthe conversion of [PtMe(CH,Cl)(cod)] into [PtEt(Cl)(cod)],viz.by ionisation to a cationic carbene intermediate, followed bymigration of the remaining CH2Cl moiety, and then p elimin-ation of chloride.? Certainly, some of the observations reportedhere could be taken as evidence for the involvement of such apathway. The more strongly electron-donating alkyl phosphineligands lead to faster decomposition than their aryl counter-parts, presumably by facilitating ionisation of the halide ion(and possibly also migration of the chloromethyl group), whilethe acceleration in the rate of decomposition effected by theaddition of 1,1,1,3,3,3-hexafluoropropanol would result fromthis compound's ability to assist heterolysis of the CH2-Clbond.With this latter effect in mind, [Pt(CH,Cl),(cod)] wasallowed to stand in CDCl,-(CF,),CHOH. Disappearance ofthe starting complex occurred slowly, and after 3 weeks a'H NMR spectrum showed that about half of the substratehad reacted to give three major products, [PtMe(Cl)(cod)],[Pt(CH,Cl)CI(cod)] and chloromethane, along with a minorquantity of [PtEt(Cl)(cod)]. The formation of the productsfound can be rationalised on the basis of initial carbene-cationgeneration if this undergoes attack by adventitious water ratherthan chloromethyl migration, as indicated in Scheme 2. Theabsence of significant amounts of the dichloride [PtCl,(cod)] inthe product mixture indicates that if a cationic carbeneintermediate is formed (Scheme 2) then the chloromethyl groupis reluctant to migrate.A new type of product was formed" rapidly when[Pt(CH,Cl),(cod)] was treated with 2 molar equivalents oftricyclohexylphosphine, P(C,H, 1)3, in deuteriochloroform inan attempt to generate the phosphine complex 24.Proton and,'P NMR spectra of the resulting solution revealed the presenceof essentially a single platinum-containing species, the phos-phonioethyl complex 21, th6 proposed structure of which wassupported by conversion into the hexafluorophosphate 22whose structure has been established by X-ray crystallography.' 'Thus, tricyclohexylphosphine promotes coupling of the twomethylene moieties in the chloromethyl groups of [Pt(CH,Cl),-(cod)] but, in contrast to the outcome reported above for someother alkylphosphines, in this case the CH,CH2 fragmentremains attached to the metal rather than being lost as ethylene.Although the mechanism of the reaction remains in doubt, thefollowing observations may be relevant to it. In deuteriochloro-form the reaction is so rapid that, when solutions of thereactants are mixed, a 'H NMR spectrum run immediatelycontains resonances attributable to only free cod and 21.Whena similar experiment was carried out using only 1 molarequivalent of the phosphine per platinum, 'H and 31P NMRspectra revealed the presence of only unreacted [Pt(CH,CI),-(cod)], 21, and free cod.The reaction is slower in perdeuterio-benzene requiring more than a day to reach completion.Phosphorus-31 NMR spectra run during the course of thisreaction revealed the presence of three major phosphorus-containing species, 21, cis-[Pt(CH,Cl),{ P(C,H ' ,),},] 24, andthe ylide complex 23. The production of 24 is perhapsunexpected in light of the report l 2 that [PtR,(cod)] (R = Phor Me) does not react with bulky monodentate tertiaryphosphines. A possible mechanism for the formation of 21 isoutlined in Scheme 3. The loss of one or other P(C,H,,), from[Pt(CH2C1),{ P(C,H, 1)3}2] is reasonable, given the stericdemands of these two cis-phosphine ligands and the presenceof an alkyl-type ligand trans to each of them. A 1,2 shift of achloride group from CH2CI into the co-ordination site available* We proposed earlier9 that this type of pathway may explain why cis-[PtCI,{P(C,H, ,),),] acts as an effective catalyst for the conversionof diazomethane into ethylene.Results for the tricyclohexylphosphinederivatives reported here appear to rule out such an explanation.Indeed, a number of (chloromethy1)platinum complexes, including 1and 12, catalyse the conversion of diazomethane into ethylene. When[2H2]diazomethane is used the 1 and 12 recovered do not containdeuterium. It thus appears likely that the catalytic activity arises simplyfrom the presence of a Lewis-acid site in the complex.7 P-Halide elimination is a well-known lo process in organopalladiumchemistry but we do not know of any examples in platinum chemistryJ.CHEM. SOC. DALTON TRANS. 1991 95 I2 1+2Scheme 2 Possible pathways from complex 1 in CDC1,-(CF,),CHOH to the observed productsr 12223R27 Ph28 Et30 Ph31 Et33 Et34 Et36 Ph29 C6H1132 C6H1135 C6HI1XCH2CICH2CICH2CIHHHCICH20MeCH20MeCIn25 226 3on platinum could then take place, followed by attack of anexternal phosphine on the resulting carbene. The production ofa similar ylidic species by a direct S,2 attack of a P(C6H11)3molecule on a bis(chloromethy1) complex appears unlikely. Theylidic intermediate can then suffer a Wagner-Meerwein shift ofthe ylide fragment to give 21. A mechanism involving a cationiccarbene intermediate is also possible, but in this case migrationof (or external attack by) phosphine would have to be faster thanmigration of CH2Cl (see Scheme 3) otherwise dichloride andethylene might be expected to be major products.In summary, the behaviour of the bis(chloromethy1) com-plexes suggests that both carbene formation, by ionisation ofchloride, and subsequent migration of the remaining CH2Clmoiety are favoured by the presence of good o-donating ligandson platinum.In contrast, x-acceptor ligands would be expectedto make both of these steps more difficult. Thus, the alkyl2423Scheme 3 Possible mechanism for the formation of complex 21phosphine complexes form ethylene readily, while their arylcounterparts do so only in more polar media. In the case of[Pt(CH,Cl),(cod)] the olefinic ligand is probably both aweaker CT donor and a stronger 7c acceptor than the phosphineligands. Decomposition of 24 gives a unique type of product.Although we cannot discount the possibility that this is formeduia initial ionisation of a chloride, we suggest that loss of aphosphine is an easier process in this case.cis-Chloro(chloromethyl) Complexes.-All four complexesexamined, namely 2, 18, 25 and 26, when pure, survived insolution in deuteriochloroform under vacuum for severalmonths, even in one case 25 in the presence of a drop of water.This stability contrasts sharply with the results reported belowfor trans-chloro(chloromethy1) complexes.trans-Chloro(chloromethyl) Complexes.-The three com-plexes studied, 27-29, were relatively stable when dissolved indeuteriochloroform from freshly opened vials.However, theyall decomposed in solution in moist deuteriochloroform.Monitoring by ‘H and 3’P NMR spectroscopy revealed thatthe tricyclohexylphosphine complex 29 reacted more rapidlythan the other two, being converted almost quantitativelyinto the relatively stable hydride 32 during 3 d. In the course of4 weeks, appreciable amounts of 27 and 28 were lost (ca. 85and 40% respectively, P NMR spectroscopy) giving mainlythe corresponding hydrides 30 and 31 together with smallquantities of both the cis and trans dichlorides. In these, andsimilar experiments, ‘H NMR spectra of the product mixturesdid not provide clear evidence for the fate of the ‘lost’chloromethyl moiety, although it was suspected that somebroad multiplets at about 6 5 arose from oligomeric formal-dehyde species. However, formaldehyde was recovered fromseveral of these product mixtures as the 2,4-dinitrophenyihydra-zone.The two alkylphosphine complexes are soluble enough inC6D, to allow NMR monitoring of their stability in solution i952 J. CHEM. SOC. DALTON TRANS. 1991PEt3 + H'I - HCIPEt3 PEt3J PEt3 PEt31 - H2CO I HCII ICI-Pt-CH,OH - CI-Pt-H - (Et3P)2PtC12?Et3 PEt3Scheme 433 from 28 and of 28,31,20 and 33 from 34Possible pathways for the formation of complexes 31,20 andthat solvent. Both showed little evidence of decomposition in1 week. However, addition of a drop of water to each sampleinduced 29 to decompose to the hydride 32 (ca.25% conversionin 3 d) while 28 gave 31 and finally 33 during several months.When 28 and 29 in C6D6 were exposed to water in the presenceof 1 molar equivalent of ethyldiisopropylamine the rates of theirdecomposition were reduced markedly.Reasoning that hydroxymethyl species are likely inter-mediates in the production of these hydrides (see Scheme 4which treats 28 as a representative example), complex 28 wasexposed to methanol in deuteriochloroform in the hope thattruns-[Pt(CH,OMe)Cl(PEt,),l 34 would be formed. However,decomposition proceeded as before, first to hydride and thento the cisltrans dichlorides. Monitoring the progress of thereaction by 31P NMR spectroscopy failed to reveal signalsattributable to 34.On the assumption that this failure may haveresulted from the reaction of any 34 formed with adventitiouswater and hydrogen chloride, the reaction was repeated in thepresence of a scavenger for these two compounds, trimethylorthoformate. However, reaction of the complex proceeded asbefore, yielding the hydride and the dichlorides. In addition,the 'H NMR spectrum showed the presence of dimethoxy-methane (from formaldehyde liberated by decomposition of theCH2Cl group), and peaks which could be ascribed to methylchloride and methyl formate. We then turned to the moreaggressive scavenger dimethylformamide dimethyl acetal. Thisdid, indeed, lead to the formation of the (methoxymethyl)complex 34 in essentially quantitative yield in less than a week.The analogous reaction for the tricyclohexylphosphine deriv-ative 29 gave 35 in 1 d.With 34 now in hand it was possible toevaluate its reactivity towards HC1 (cu. 0.8 mol equivalent,generated from methanol and acetyl chloride) in deuterio-chloroform. Phosphorus-3 1 NMR spectroscopy revealed thatwithin 1 d, complex 34 suffered substantial conversion intomainly the hydride 31 along with, in progressively smalleramounts, the (chloromethyl) complex 28, the cis dichloride 20,and trans dichloride 33. This outcome appeared to indicate thatthe chemical reactivity of the Pt(CH,OMe) moiety is more akinto that of an acetal than that of a simple ether and is in apparentaccord with the mechanism outlined for the decomposition of34 and 28 in Scheme 4.Certainly our evidence does not rule outa variety of other possible mechanisms, including some basedon associative or other dissociative processes. The mechanismsuggested has the advantage of simplicity while accounting forthe following observations, if carbene formation is rate-limiting.The decomposition is facilitated by a more polar solvent, by thepresence of protons in the solution, and of bulky tricyclohexyl-phosphine ligands in the co-ordination sphere. It may be notedthat the formation of a carbene trans to a halogen is expectedto be more facile than that trans to a phosphine.'ExperimentalMelting points were determined on a Kofler hot-stage, and areuncorrected. Proton (internal tetramethylsilane reference) and31P (external 85% H,P04) NMR spectra were recorded fordeuteriochloroform solutions (unless otherwise stated) on aBruker WH-400 spectrometer.Elemental analyses were carriedout by Galbraith Laboratories, Knoxville, Tennessee. Platesfor thin-layer chromatography were spread with Kieselgel G(Merck).Mono(chloromethyl)platinum(rI) complexes were preparedfrom chloro(chloromethyl)(cycloocta-1,5-diene)platinum(11) 2,and bis(chloromethyl)platinum(II) complexes from bis(ch1oro-methyl)(cycloocta-l,5-diene)platinum(11) 1 in essentially quanti-tative yield by ligand exchange and were purified by preparativeTLC and/or crystallisation unless noted otherwise. Most ofthe compounds have been prepared earlier but the followinghave been characterised for the first time: [ 1,2-bis(diphenj*f-phosphino)ethane]bis(chloromethyl)plutinum(r~) 3, colourlesscrystals from dichloromethane [m.p.> 280 "C (change ofcrystalline form at ca. 270 "C)], 6," 2.21 (m, 4 H, PCH,) and3.96 (dd, 4 H, PtCH,, ,JPH = 7.4 and 2.8, ',IplH = 48.4), 8,43.59(lJptp = 1893 HZ) (Found: c , 48.95; H, 4.25. C~8H,8C12P,Ptrequires C, 48.55; H, 4.1%); [ 1,4-bis(diphenylphosphino)-butune]bis(chloromethyl)platinum(~r) 5, fine, colourless needlesfrom dichloromethane-pentane [m.p. > 280 "C (decomp. from200 "C)] BH 1.79 (br m, 4 H, PCH,CH,), 2.51 (vbr s, 4 H, PCH,),and 3.58 (d, 4 H, PtCH,, ,JPH = 7.9, 'JptH = 44.8), 15.77('JpIp = 1913 Hz) (Found: C, 50.15; H, 4.5. C,,H,,Cl,P,Ptrequires C, 50.0; H, 4.5%); [ 1,2-bis(dicyclohexylphosphino)-ethane]chloro(chlorornethyl)pplatinum(~~) 18, colourless crystalsfrom dichloromethane-pentane, crumbled on drying [m.p.>280 "C (decomp.from 230 "C)], 8, 3.78 (t, ,2 H, PtCH,,,JPH = 2.8, 'JptH = 37), 6p 56.09 (lJpIp = 4025) and 63.68('Jptp = 1832 Hz) (Found: C, 44.2; H, 7.15. C2,HS0-C12P2Pt~0.5CH2Cl, requires C, 44.3; H, 6.9%); trans-chloro-(chloro~ethyZ)bis(triethylphosp~ine)pZatinum(~I) 28, colourlesscrystals from pentane (at - 78 "C) (m.p. 62-68 "C), 8~(C6D6)0.96 (18 H, CH,), 1.80 (12 H, PCH,) and 3.88 (t, 2 H, PtCH,,,JPH = 8.5, ',IptH = 55.6), 6, 16.0 (lJptp = 2793 Hz) (Found:C, 30.05; H, 6.0. C13H,,Cl,P,Pt requires C, 30.25; H, 6.25%).In general, the stabilities of complexes (0.054.1 mmol) weremonitored by 'H and, for all but 1 and 2, 31P NMR spectros-copy for solutions in deuteriochloroform in 5 mm tubes whichhad been sealed under vacuum, or capped with rubber septaunder nitrogen, and were kept at ambient temperature in thedark.Approximate relative concentrations of complexes inmixtures resulting from substrate decomposition were estimatedonly by comparison of relative peak heights in 31P NMRspectra. Acid-free deuteriochloroform refers to material thathad been passed through a column of anhydrous potassiumcarbonate. Head-space gas analyses were carried out with aVarian Aerograph 1200 gas chromatograph [two stainless-steelcolumns (4 in x 5 ft), one containing 3% SE-30, the otherPorapak Q, in series; nitrogen as carrier gas with a flow rate of25 cm3 min-'1.cis-Bis(chloromethy1) Complexes.-Bis(chloromethyl)(cyc/o-octa-l,5-~iene)plutinum(rr) 1.Proton NMR spectra of thiscomplex in acid-free deuteriochloroform showed no changesduring 1 month. In deuteriochloroform-1 ,1,1,3,3,3-hexafluoro-propanol (9:l) ca. one half of the solute 1 decomposed in3 weeks giving 'H NMR signals for the complexes 1,9[PtMe(Cl)(cod)], [Pt(CH,CI)Cl(cod)], and [PtEt(Cl)(cod)]and for MeCl (6, 3.02, confirmed by spiking). In C6D6-(CF,),CHOH (9: 1) less than 20% of 1 was lost during 3 weeksto give [PtMe(Cl)(cod)], [Pt(CH,Cl)Cl(cod)] and MeCl; thehead-space gas contained large amounts of CH, along withsmaller amounts of C2H6, C,H, and MeCl.Chela t ing A rylphosphine Complexes 3-5.-T hese threecomplexes showed no changes in their 'H and 31P NMR* For brevity, 'H resonances for the phenyl rings are not reported forthis and other complexes; couplings quoted are observed valuesJ.CHEM. SOC. DALTON TRANS. 1991 953spectra in acid-free deuteriochloroform (tubes sealed in oacuo)during 3 months. When dissolved in CDC13-(CF3),CHOH(9: l), 'H and ,'P NMR spectra run as soon as possiblethereafter contained resonances for only the correspondingdichlorides and, in the 'H NMR spectra, a signal at 6 5.42 forethylene.Monodentate Arylphosphine Complexes 6, 8 and 10.-Solutions of the first of these in CDCl, (capped tubes), preparedeither from crystalline material ' or in situ by adding PPh, (1.9rnol equivalent) to a solution of 1 in that solvent, decomposedvery slowly to the dichloride 7 (ca.0.5% farmed in 2 weeks, ,'PNMR spectroscopy). The p-OMe and p-F analogues 8 [S, 20.8('JPtp = 2001 Hz)] and 10 [S, 22.2 ('Jptp = 1964 Hz)] preparedin situ also decomposed slowly to the analogous dichlorides(ca. 5% of 9 and 2% of 11 in 1 week, 'P NMR spectroscopy).Chelating Alkylphosphine Complexes 12-14.-Samples of thelast two complexes were prepared in situ by dissolving a mixtureof 1 and either 1,4-bis(dicyclohexylphosphino)butane or 1,3-bis-(dicyclohexy1phosphino)propane (0.9 mol equivalent) in CDC1,under N, (capped tubes). Complex 14 had 6 , 3.86 (d, 4 H,PtCH,, ,JPH = 7.1, ,JPtH = 43 Hz) and 6, 18.3 (lJPtP = 1971Hz): during 10 d the signals due to 14 were replaced by those forthe dichloride 17 [S, 21.5 (lJptp = 3621 Hz)]. Complex 13 had6, 3.86 (d, 4 H, PtCH,, ,JPH = 5.7, 2JptH = 41 Hz) and 6, 4.1('Jptp = 1909 Hz): within ca.2 weeks the signals due to 13 werereplaced by those for 16 [S, 8.7 ( 'Jhp = 3464 Hz)]. When solid12 was dissolved in CDCl, the ,'P NMR signal arising fromthis compound was gradually replaced by one for 15 during 2-3weeks. After decomposition, analysis of the head-space gasesby GC showed the presence of a large amount of ethylene anda minor amount of methane. A sample of 12 prepared in situfrom 1, by treatment with the phosphine ligand in CDCl,,decomposed to 15 at essentially the same rate (i.e. in thepresence of free cod). A sample of solid complex 12 (180 mg)was dissolved in CDCI, (5 cm3) and the solution dividedamong six NMR tubes (5 mm).One was kept (under nitrogen),as a standard, in the dark and its rate of decompositionwas compared (,'P NMR spectroscopy) with the other fivesamples which were treated as follows: (i) left on the bench(fluorescent/daylight), no rate difference: (ii) drop of wateradded, no rate difference; (iii) tetrabutylammonium chloride(4 mol equivalents) dissolved, rate ca. double; (iv) 1,1,1,3,3,3-hexafluoropropan-2-01 (30 pl) added, decomposition to 15complete in 1 d; (v) exposed to 300 nm light for 15 min, ca.75% decomposition to approximately equal amounts of 15and 18.The Bis(triethy1phosphine) Complex 19.-This complex,which was prepared ' from the dichloride 20 by treatment withdiazomethane, decomposed rapidly (overnight) back to thedichloride when dissolved in CDCI,.The resulting solutioncontained ethylene ('H NMR integration indicated 42% of thatexpected on the basis of stoichiometry in solution, while thepresence of a further 45% was calculated to be present in the gasphase assuming that the solubility of ethylene in chloroformis similar to its solubility in chlorobenzene'). Ethylene anda trace of methane were detected in the head-space gas byGC.The Bis(tricyclohexy1phosphine) Complex 24.-When com-plex 1 and tricyclohexylphosphine (2 mol equivalents) weredissolved in CDCI, under nitrogen, only the phosphonioethylcomplex 21 could be detected '' by a 31P NMR spectrum runimmediately. When this reaction was repeated using C,D6,initially at 10 OC, and subsequently at ambient temperature(25 "C), peaks due to 21 were found right from the beginning.However the most prominent peaks appeared at 6 19.4 ('Jptp =2774) and 45.2 ( 'JPtp = 67.5 Hz), and are assigned to the ylidecomplex 23.These peaks gradually decreased in intensity onstanding, and finally disappeared after about 1 d at 25 "C. Theinitial spectrum also showed a prominent signal at 6 19.0('JPtp = 1904 Hz), which disappeared within a few hours at25 "C, and is ascribed to the cis-bis(chloromethy1) complex 24.After 1 d the major component present was 21, although severalweaker resonances were observed. The overall reaction was lessclean in benzene than in chloroform. When the same reactionwas monitored by 'H NMR spectroscopy, signals ascribable toCH2C1 resonances of 23 and 24 were observed at 6 4.5 (,JpH =respec ti vely .6.8, 'JptH = CU.60) and 4.42 (,JPH = 8.5, 'JptH = Ca. 35 HZ)cis-Chloro(chloromethy1) Complexes.-Solutions of all fourcomplexes 2, 18, 25 and 26 in CDCI, (sealed tubes) wereprepared from crystalline material. These showed no evidence(,lP NMR spectroscopy) of decomposition during 3 months atambient temperature. An identical result was obtained for asolution of 25 in CDCl, containing a drop of water.trans-ChZoro(chZoromethy1) Complexes.-trans-Chloro-(chloromethyl)bis(triphenylphosphine)platinum(Ii) 27. Thiscomplex gave similar, small, amounts (2-3%, ,'P NMR peakheights) of 30 and 7 during 3 weeks, when dissolved in CDC1,(capped tubes) from a freshly opened vial. This experiment wasrepeated in the presence of a drop of water.After 4 weeks some27 (ca. 15%, 31P NMR spectroscopy) remained along with thehydride 30 (ca. 60%) and the two dichlorides 36 (ca. 25%) and 7(ca. 10%).trans-Chloro(chloromethyl)bis( tricyclohexy1phosphine)-platinum(rr) 29.-Treatment of complex 1 with tricyclohexyl-phosphine (2 mol equivalents) in benzene gave 29 which wascrystallised from benzene-pentane. It had 6, 3.63 (t, 2 H,PtCH,, ,JPH = 8.0, ' JptH = 5 1.6) and 6, 19.4 ( lJptp = 278 1 Hz).When a solution of this complex in CDCl, (from a bottle thathad been opened frequently) was sealed under vacuum a 31PNMR signal for the hydride 32 gradually increased in intensityduring 4 d. Thereafter no change in ,'P NMR spectra wasapparent (29:32 ca.3:2, based on peak height). This reactionwas repeated with a drop of water added and gave, within 3 d,essentially pure 32; ijH - 18.82 (t, 1 H, PtH, 2JpH = 12.3, 'JptH =1293) and 6, 39.1 ('Jptp = 2802 Hz). In C6D6, 29 gave ,'PNMR signals for only this compound even after 1 week butwhen a drop of water was present a signal arising from 32gradually increased in intensity (ca. 30% of maximum heightduring 3 d). When both water (1 drop) and ethyldiisopro-pylamine (1 mol equivalent) were added to 29 in C6D, only asmall amount (< 10%) of 32 was produced in 3 d and some 29survived for 6 months (29:32, 4:7, ,'P NMR peak height).Exposure of 29 in CDC1, to MeOH (10 mol equivalents) and(MeO),CHNMe, (5 mol equivalents) gave the (methoxymethyl)derivative 35, quantitatively, within 1 d.Upon crystallisationfrom CH,Cl,-hexane this trans-chloro(methoxymethy/)bis-(tricyclohexylphosphine)platinum(u) gave colourless crystals[m.p. 22G223 "C; 15~3.20 (s, 3 H, OMe) and 3.68 (t, 2 H, PtCH,,,JPH = 6.8, ,JPtH = 67.3); 6,21.6 ('JPtp = 2934 Hz) (Found: C,55.35; H, 8.7. C3*H,1C10P2Pt.0.2C6H14 requires C, 55.15; H,8.7%).trans-Chloro(chloromethyl)bis(triethylphosphine)plutinum( 11)28.-(i) A solution of complex 28 in CDCI, from a freshlyopened vial was stable (,'P NMR spectroscopy) for 7 d but,when the tube was opened and a drop of water added, totaldecomposition into the hydride 31 and the dichlorides 20 and 33occurred during the next 7 d (31:20:33 cu. l : l : l , ,'P NMRspectroscopy). Thereafter, the concentration of the transdichloride 33 increased at the expense of 20 and 31.The reactionin the presence of water was repeated in a sealed tube. When allthe substrate had been consumed (,'P NMR spectrum) thetube was opened and the contents were treated with excess ofethanolic 2,4-dinitrophenylhydrazine reagent. Most of th954 J. CHEM. SOC. DALTON TRANS. 1991solvent was then removed in vacuo and the residue waspartitioned between water and dichloromethane. The materialdissolved in the organic layer was subjected to preparative TLC(chloroforrn-hexane, 1 : 1) to give formaldehyde 2,4-dinitro-phenylhydrazone (identified by TLC and 'H NMR spectralcomparison with an authentic specimen).(ii) Complex 28 was also stable in dry C,D, (3 months, 'Hand ,'P NMR spectra) but decomposed slowly, upon theaddition of a drop of water to this solution, giving mainly 31 and33 (after 10 weeks, 33:31:28:20 18: 16:6: 1, 31P NMR peakheights).(iii) Upon repeating the last experiment in the presence ofethyldiisopropylamine (1 mol equivalent) a 'P NMRspectrum run after 7 months showed that 28 had largelysurvived, the only other peaks present arising from thehydride 31 (ca.5%).(iv) When complex 28 was dissolved in deuteriochloroformcontaining methanol (10 mol equivalents) and the resultingsolution monitored by 31 P NMR spectroscopy, conversioninto the hydride and dichlorides proceeded at a similar rate tothat observed in (i) with water added.( v ) A similar outcome and rate of reaction was observed whenexperiment (iu) was repeated using trimethyl orthoformate(5 mol equivalents) in addition to the methanol. Proton NMRspectra of the final solution contained resonances arising fromdimethoxymethane (6 3.37 and 4.58), methyl formate (6 3.77and 8.1 l), and methyl chloride (6 3.02).(vi) Complex 28 reacted cleanly with MeOH-(MeO),CHN-Me, (10 and 5 mol equivalents respectively) in CDCl, (within1 week) and in C,D, (within 1 month) to give 34.Thismethoxymethyl complex was purified by preparative TLC(CH,Cl,-MeOH, 397:3) giving a clear gum which failed tocrystallise, but which showed only peaks for 34 in its 'H and 31PNMR spectra; 6, 1.14 (m, 18 H, CH,CH3), 1.93 (m, 12 H,PCH2), 3.31 (s, 3 H, OMe) and 4.12 (t, 2 H, PtCH,, ,JPH = 6.2,2Jp,H = 62.1 Hz), 6, 18.3 ('JRP = 2999 Hz). Complex 34 inCDCl, was treated with methanol-acetyl chloride (0.8 molequivalent of each) and gave (31P NMR spectra), after 1 d,mainly the hydride 31 and the chloromethyl complex 28(34:31:28:20 19: 17: 17:2) and, after 10 d, mainly 28 and thecis dichloride 20 (28:20:31:33:34 54:36: 12:6: 1, "P NMRpeak heights).AcknowledgementsFinancial support from the Natural Sciences and EngineeringResearch Council of Canada is gratefully acknowledged.References1 R.McCrindle, G. J. Arsenault, R. Farwaha, M. J. Hampden-Smith,R. E. Rice and A. J. McAlees, J. Chem. SOC., Dalton Trans, 1988,1773.2 N. J. Kermode, M. F. Lappert, B. W. Skelton, A. H. White andJ. Holton, J. Chem. Soc., Chem. Commun., 1981, 698.3 See R. McCrindle, G. Ferguson, G. J. Arsenault, A. J. McAlees, B. L.Ruhl and D. W. Sneddon, Organometallics, 1986,5,1171; H. Werner,L. Hofmann, W. Paul and V. Schubert, Organometallics, 1988, 7,1106; C. A. Ghilardi, S. Midollini, S. Moneti, A. Orlandini and J. A.Ramirez, J. Chem. SOC., Chem. Commun., 1989,304, M. Huser, M. T.Youinou and J. A. Osborn, Angew. Chem., Int. Ed. Engl., 1989, 28,1386.4 See, for example, A. N. Nesmeyanov, E. G. Perevalova, E. I.Smyslova, V. P. Dyadchenko and K. I. Grandberg, Zzv. Akad. Nauk.SSSR, Ser. Khim., 1977, 2610; 0. J. Scherer and H. Jungman,J. Organomet. Chem., 1981, 208, 153; C. Botha, J. R. Moss andS. Pelling, J. Organomet. Chem., 1981,220, C21; D. G. Harrison andS. R. Stobart, J. Chem. SOC., Chem. Commun., 1986,285.5 G. B. Young and G. M. Whitesides, J. Am. Chem. Soc., 1978, 100,5808.6 F. D. Mango and I. Dvoretzky, J. Am. Chem. SOC., 1966,88,1654.7 R. McCrindle, G. Ferguson, M. A. Khan, A. J. McAlees and B. L.Ruhl, J. Chem. SOC., Dalton Trans., 1981, 986; R. McCrindle, D. K.Stephenson, A. J. McAlees and J. M. Willson, J. Chem. Sac., DaltonTrans., 1986, 641.8 J. H. Hildebrand and R. L. Scott, in The Solubility of Nonelectrolytes,Dover Publications, New York, 1964, p. 243.9 R. McCrindle, G. J. Arsenault, R. Farwaha, M. J. Hampden-Smithand A. J. McAlees, J. Chem. SOC., Chem. Commun., 1986,943.10 See, for example, R. F. Heck, J. Am. Chem. SOC., 1968, 90, 5531;R. McCrindle, E. C. Alyea, G. Ferguson, S. A. Dias, A. J. McAleesand M. Parvez, J. Chem. SOC., Dalton Trans., 1980,137.11 R. McCrindle, G. Ferguson, G. J. Arsenault, M. J. Hampden-Smith,A. J. McAlees and B. L. Ruhl, J. Organomet. Chem., 1990, 390,121.12 G. K. Anderson, H. C. Clark and J. A. Davies, Inorg. Chem., 1981,20,3607.13 A. J. L. Pombeiro, S. S. P. R. Almeida, M. F. C. G. Silva, J. C. Jeffreyand R. L. Richards, J. Chem. SOC., Dalton Trans., 1989,2381.Received 28th August 1990; Paper 0/03891
ISSN:1477-9226
DOI:10.1039/DT9910000949
出版商:RSC
年代:1991
数据来源: RSC