摘要:
Synthesis, characterisation and substitution reactions of gold(@C,N-chelatesPierre A. Bonnardel, R. V. Parish * and Robin G. PritchardDepartment of Chemistry, UMIST, PO Box 88, Manchester M60 1 QD, UKNew complexes of the type [AuCl,L] have been prepared, where L is a chelate consisting of a phenyl groupbearing an N-donor substituent (oxazoline and/or dimethylaminomethyl). The structures of two of these,together with that of [AuCl{ C6H3(CH,NMe,),-2,6)]2[Hg2C16], have been determined by X-ray crystallography;the gold atoms exhibit strict square-planar geometry in all cases. The chloride ligands undergo readysubstitution by other halides, thiocyanate, acetate or diethyldithiocarbamate. The monodithiocarbamatecomplexes [Au(S,CNEt,)L] + contain chelated S,CNEt2 groups whereas in [Au(S,CNEt,),L] the L ligand ismonodentate (through C), one S,CNEt, is monodentate, the other bidentate; in solution the two S,CNEt,ligands appear equivalent on the NMR time-scale, indicating a rapid equilibrium between the two possibleforms.Following the discovery and exploitation of the antitumourproperties of platinum(I1) complexes, many new compoundshave been synthesised and screened, but the square-planarcomplexes cisplatin [cis-diamminedichloroplatinum(~~)] andcarboplatin [cis-diammine(cyc1obutane- 1,l -dicarboxylato)-platinum(r~)] remain amongst the most successful.Since gold(111) also usually shows square-planar co-ordinationwe have established a programme to investigate complexesstructurally related to the effective platinum compounds, inthe hope that some might show antitumour properties.Theformation of neutral complexes with a cis-MX, configuration ismost readily achieved using a mononegative bidentate ligandto occupy the other two co-ordination positions. We havepreviously reported the synthesis of complexes containingN,O - ligands,, which proved to be unstable under biologicalconditions. However, more recently we showed that theN,C --bonded [AuCl,(dmamp)] [dmamp = 2-(N,N-dimethyl-aminomethy1)phenyll possesses in vitro antitumour activity-and has in vivo activity against the human breast tumour(ZR-75- 1) parallel to that of cisplatin. The correspondingdiacetato complex [Au(O,CMe),(dmamp)] is more solubleand more active.6We now report the preparation, characterisation andreactions of a variety of new gold(w) complexes containingother mononegative N,C- ligands together with two containinga mononegative N,C-,N ligand.X XMe0XMe2Nl a X=CIl b X = B rl c X = lId X = SCN2a X = CI2b X = Br2CX=I2d X = SCN3a X = CI3bX = Br3 c X = I3dX = SCNMe0 Me2N Me04 5a X=CI5 b X = 02CMe6a X = CI6b.X = 02CMe/ \ Au-CI&;Me2Ei-C, I\ / OMe7 aa 9Results &L I Chelated organogold(rI1) complexes can readily be preparedby transmetallation from the corresponding organomercury(r1)We have used this route with the newarylmercury(i1) compounds recently reported by us.l o Reactionwith [NMe,][AuCl,] in dry acetone or acetonitrile, in thepresence of one molar equivalent of NMe,Cl, gives thechlorogoId(r1r) complexes la, 2a, 3a, 4, 5a, 6a and 7 in fairlygood yield, as yellow solids.The function of the NMe,Cl isto encourage precipitation of [NMe,],[Hg,Cl,], shifting theequilibrium in favour of the organogold complex. Complexes8a, 9 and the ionic 10a, containing a [Hg,Cl6I2- anion, wereobtained similarly, but without adding NMe,Cl, whichappeared to inhibit the reaction. Complex lob, containing the[AuCl,] - anion, was obtained in aqueous solution; in thismedium the mercury remains predominantly as HgCl, andthe anion [Hg,C16]2- is not formed. The transmetallationCl Cl1 2 R = H13 R = E t10a Y = 1 / 2 ~ ~ 2 ~ ~ 6 1110b Y = AuC14reaction appears to be sensitive to the solvent chosen, sincearylmercury compounds containing 2-SO2NMe, or 2-CONHX-Et, - ( x = 0-2) substituents showed no reaction in acetoneor acetonitrile but gave good reaction in dimethyl sulfoxide(dmso) (NMR evidence; the spectra show that reaction iscomplete only after several hours).Unfortunately, we havenot yet been able to isolate these new gold complexes ingood condition from dmso, but the NMR spectra (see below)J. Chem. SOC., Dalton Trans., 1996, Pages 3185-3193 318indicate that they are analogous to the other products, andwe formulate them as 11-13. The reactions of HgR(C1) [R =2,5-(N,N-dimethylaminomethyl)phenyl, 2-(dimethylamino)-6-(N,N-dimethylaminomethyl)phenyl or 2-rnethoxy-6-(N-methyl-aminomethy1)phenyll gave only deposits of gold metal in allsolvents tried.Since the substitution chemistry of gold(rr1)-dmampcomplexes has been well e~plored,~."-'~ we have concentratedour efforts on complexes of the oxazoline-substituted ligands.Simple substitution reactions of the chloro-complexes la-3awith stoichiometric amounts of alkali-metal salts gave thecorresponding bromo- (lb-3b), iodo- (lc-3c) or thiocyanato-complexes (ld-3d).Yields were low, presumably because of thesusceptibility of the oxazolinyl group to nucleophilic attack;this group is often used in organic synthesis as a protectinggroup for carboxylic acids because of the ease with which itcan be removed under relatively mild conditions. ' A similarobservation was made during reactions with silver acetate (seebelow).It also proved possible to substitute a single chloride ligandin complex 3a by reaction in dichloromethane with a smallamount of dmso in the presence of silver perchlorate.Theresulting complex, 3e, was a 1 : 1 electrolyte in acetone (96 Smol-'). A similar reaction occurred with 1 molar equivalentof triphenylphosphine to give le and 3f (the amount ofphosphine has to be carefully controlled to avoid sidereactions). Other substitution reactions will be describedbelow. Analytical data for all the new materials are given inTable 1.Spectroscopic propertiesThe infrared spectra of the compounds (Table 2) are consistentwith the presence of the expected ligands. For the chloro-complexes la-3a, 4, 5a, 6a, 7, 8a and 9 bands are observed atabout 350 and 305 cm-' which may be assigned, on thebasis of the trans influence, to the stretching of Au-CI bondstrans to the nitrogen and carbon atoms of the chelaterespectively.For 10a and 10b a single gold-chlorine stretchingfrequency is observed for the organogold cation, at 3 14 and 307cm-' respectively; additional bands are present at 360-340 cm-'due to the metalkhlorine vibrations of the inorganic anions.The v(C-N) bands of ld-3d suggest that the thiocyanate ligandsare S-bonded, as would be expected; the C-S bands wereobscured by absorptions due to the organic ligands. Gold-carbon stretching frequencies are observed at 430440 cm-'for the phenyl compounds, and at 424-429 cm-' for thenaphthyl derivatives 8a and 9.l eCI-I fFor la-lc, 2a-2c and 3a-3c the typical bands of theoxazoline group occur at significantly lower frequencies thanfor the free aromatics [v(C-N) 1649-1 66 1, 1585-1 61 2 and1562-1589 cm-' for the three molecules], indicating that thisgroup is co-ordinated to gold.This is in marked contrast to thecorresponding mercury derivatives, HgR(Cl), which are two-co-ordinate and show values very close to those of the freearomatics. ' OThe infrared spectrum of the dmso complex 3e shows a singleAu-Cl stretching frequency at 356 cm-', consistent withchloride trans to nitrogen. Substitution has therefore occurredat the kinetically controlled site, trans to carbon. The S-0stretching frequency of the co-ordinated dmso appears at 1 186cm-l, which indicates co-ordination through sulfur.' A bandat 429 cm-I is more intense than expected for v(Au-C), and maybe v(Au-S). '' The triphenylphosphine derivatives le and 3falso show a single v(Au-Cl), but at markedly lower frequencies(ca. 310 cm-'). This suggests the configuration which isnormally found for monosubstitution in this type of system,where the incoming ligand (eventually) occupies the positiontrans to nitrogen. It is generally expected that the two softestligands will be cis to each other in the thermodynamically moststable isomer.The NMR spectra (Tables 3-5, numbering as in Scheme 1)are very much as e~pected,~ and also indicate co-ordination ofthe oxazoline group: there are strong downfield shifts for C7relative to the free aromatics and to the corresponding mercurycomplexes.The same applies to the C H , and CH, resonances ofthe CH,NMe, groups ofcomplexes 4,5a, 6a, IOa and lob. As wehave shown in the case of gold(rII)-dmamp complexes, suchdownfield shifts are diagnostic of co-ordination of the nitrogenatom.For all the complexes studied, C' and, to a lesser extent, C6also show strong downfield co-ordination shifts, which mostlyincrease as X changes from C1 to Br to I (Table 5); C2 wouldbe expected to show similar shifts but, as for the mercuryderivatives," these are masked by the effects of the substituentgroups.The 13C NMR spectra of solutions in (CD,),SO ofHgR(C1) [R = C6H4S0,NMe,-2 or C6H4C(0)NH,Et2 - ,-2( x = 1 or 2)] changed dramatically when NaCAuCl,] wasadded. New sets of signals appeared rapidly and after somehours were the only peaks in the spectra. The aromatic regionthen strongly resembled that of the other chelated gold(r1r)complexes (Table 6).In particular, there were markeddownfield shifts for the carbon atoms of the S02NMe,,C(O)NH, and C(0)NHEt groups, indicating co-ordination ofthe nitrogen atoms. Thus, although they were not isolated, it is2e Y=BPh42f Y = PF6 2gR= "3<L N3186 J. Chem. Soc., Dalton Trans., 1996, Pages 3185-319Table 1 Analytical data" and yieldsAnalysis (%)Compound Cla 30.2 (29.9)l b 24.6 (24.9)l c 21.2 (21.2)Id 32.3 (32.0)le 49.4 (49.4)If 28.6 (28.9)2a 30.8 (30.5)2b 25.9 (25.7)2c 22.2 (22.0)2d 32.7 (32.4)2e 56.6 (56.7)2f 29.1 (29.4)2g 38.2 (37.9)3a 32.0 (32.3)3b 27.5 (27.2)3c 23.1 (23.4)3d 34.0 (34.3)3e 29.2 (28.7)3f 49.5 (49.8)3g 30.8 (30.5)4a 27.5 (27.8)5a 29.9 (29.7)5b 36.9 (36.6)6a 28.3 (28.6)6b 35.1 (35.4)6c 33.5 (33.7)7a 27.6 (27.8)8a 34.8 (34.5)9 32.9 (32.9)1 Oa 20.0 (19.7)lob 18.5 (18.9)H2.7 (2.7)2.3 (2.3)1.7 (1.9)2.4 (2.5)3.8 (3.8)3.1 (3.3)2.9 (3.0)2.5 (2.5)2.3 (2.7)4.8 (5.1)2.0 (2.1)3.5 (3.5)4.9 (4.9)3.2 (3.5)3.0 (3.0)2.4 (2.5)3.2 (3.2)4.6 (4.3)3.8 (3.8)3.3 (3.2)3.8 (3.8)5.0 (4.7)4.6 (4.3)3.1 (3.2)3.2 (3.1)2.7 (2.7)2.2 (2.6)2.2 (2.5)3.7 (3.7)3.4 (3.5)3.4 (3.5)N3.2 (3.2)2.6 (2.6)2.2 (2.2)8.7 (8.6)2.1 (2.0)4.2 (4.2)3.2 (3.0)2.5 (2.5)8.4 (8.1)3.2 (3.2)4.0 (4.0)6.0 (6.0)5.8 (5.8)4.8 (4.9)4.1 (4.2)1 0.6( 10.6)4.5 (4.5)3.4 (3.7)5.9 (5.9)5.7 (5.7)2.1 (2.1)3.5 (3.2)6.2 (6.3)3.0 (3.0)2.8 (2.7)5.4 (5.6)3.2 (3.2)3.4 (3.1)3.1 (3.2)3.9 (3.8)3.4 (3.7)Xb15.9 (16.1)30.3 (30.1)40.2 (40.6)13.2(13.1)4.8 (4.4)10.0 (9.6)14.7 (15.0)28.1 (28.5)38.7 (38.8)12.5 (12.4)7.0 (7.4)9.2 (9.2)18.4 (1 8.4)14.8 (14.6)28.0 (27.8)38.5 (38.0)12.2 (12.1)5.5 (5.1)4.1 (4.1)9.5 (9.1)16.4 (1 6.4)15.8 (16.0)15.9 (1 5.4)16.5 (16.4)16.2 (15.7)16.2 (1 6.2)18.8 (19.4)22.9 (23.3)Au44.4 (44.6)29.9 (29.7)41.5 (41.7)40.5 (40.6)45.6 (45.6)44.2 (44.3)42.4 (42.6)45.8 (45.6)43.5 (43.6)44.1 (45.0)26.8 (27.0)50.9 (51.7)" Calculated values in parentheses.' C1, Br, I, P or S as appropriate.M - C1. M+. M - AuCI,.mlz47248 5397462402'423 'Yield (%)854533656882483271666863834234676387687585737862588379Table 2 Selected infrared bands (cm-') for the gold complexesv(Au-Cl) ~(Au-C) v(C-N) v(C-0)Complex _____3f3g4a5a6a7a8a91Oa10b" 21la 354 305 431l bl cIdl e 312If2a 355 308 4372b2c2d2e2f2g3a 356 300 4403b3c3d3e 356310354 303 439355 304 435354 304 439353 303 441355 307 424354 307 429351 314 418360' 3073, 2075 cm ' for SCN liganc1620 1576 - 10371622 1572 - 10371619 1570 10381619 1585"1620 15761616 15771607 1585 I0281609 1589 10351611 1585 I0301605 1589'1616 15771610 15871643 1585 10351597 1529 13651595 1524 13601597 1529 13651597 1538'1599 15851593 15151595 15351271 10781340 10201271 10781245 1078107410781089I09321 16, 2073 cm-' for SCN liganc' 21 17, 2073 cm1 186; v(Au-S) 429 cm-'.For [AuCl,] - .for SCN ligand. Also C10,- at 1091, 623; v ( W )certain that the chelated dichlorogold(Ir1) complexes 11-13 wereformed in solution.11 12/-"Me2Scheme 1?NM&-"MezCrystal structuresThe structures of the chloro-complexes la, 6a and 10a wereconfirmed by X-ray crystallography. Selected bond distancesand angles are given in Tables 7 and 8. Views of the molecularstructures are displayed in Figs. 1-3.In complexes l a and 6a the gold atoms have essentiallysquare-planar CNC1, co-ordination, and 10a CN,Cl, with Auand the four ligand atoms being closely coplanar in each case(mean deviations from the planes: 0.0076,0.0550 and 0.0071 Afor la, 6a and 10a, respectively). The Au-C, Au-N and Au-Clbond distances are very similar to those reported for othergold(rI1) complexes.5~"~'2~1s For l a and 6a, the two Au-C1J. Chem.SOC., Dalton Trans., 1996, Pages 3185-3193 318Table 3 Proton NMR data (6, CDCI, solutions unless otherwise noted)Complexl al bl c2a2b2c3a3b3c45a5b"6a8a b,c10alobbR5-OMe5-OMe5-OMe5-NMe25-NMe25-NMe25-OMe5-NMe25-NMe25,6-(OMe),H37.4 (d)7.35 (d)7.4 (d)7.3 (d)7.3 (d)7.35 (d)7.1 (d)7.1 (d)7.1 (s)7.0 (d)6.95 (d)6.8 (d)6.8 (d)7.85 (d)7.25 (d)7.05 (d)H47.4 (dd)7.35 (dd)7.3 (dd)6.8 (dd)6.75 (dd)6.75 (dd)6.4 (dd)6.4 (dd)6.4 (dd)6.75 (d)6.6 (d)6.5 (d)6.85 (d)7.95 (dd)7.85 (dd)7.3 (dd)H5 H67.4 (dd) 7.95 (d)7.35 (dd) 8.2 (d)7.3 (dd) 8.55 (d)7.55 (d)7.8 (d)8.15 (d)7.2 (d)7.35 (d)7.5 (d)7.0 (s)7.05 (s)6.3 (s)7.5 (d) 7.5 (d)7.25 (d)7.05 (d)" For MeCO, group: 6 2.1 (s) and 1.95 (s).In (CD,),SO. ' For NMe,, 6 4.9 (s).H7 H84.8 (s)4.55 (s)4.5 (s)4.6 (s)4.5 (s)4.45 (s)4.5 (s)4.45 (s)4.45 (s)4.3 (s)4.25 (s) 3.2 (s)4.1 (s) 2.85 (s)4.3 (s) 3.1 (s)7.3 (dd) 7.8 (d)4.95 (s) 3.4 (s)4.85 (s) 3.35 (s)H'O1.75 (s)1.8 (s)1.85 (s)1.8 (s)1.8 (s)1.8 (s)1.7 (s)1.7 (s)1.75 (s)3.25 (s)R3.9 (s)3.85 (s)3.85 (s)3.05 (s)3.05 (s)3.05 (s)3.2 (s)3.1 (s)3.8 (s), 3.85 (s)3.3 (s)3.75 (s)Table 4 Carbon- I3 NMR data (6, CDCI, solutions unless otherwise noted)Complexl al bl c2a2b2c3a3b3c45a5b"6a10a10b111213C'147.7148.7150.2149.2150.2151.9153.7150.4150.2148.6148.5148.5149.0162.4162.5142.0140.3141.9C2126.5127.4127.811 7.8118.6120.6110.1110.4110.8135.7131.7130.1136.0146.8146.9146.1145.8147.9c3128.5127.4127.2114.7114.2113.2109.9109.5109.41 15.4113.0112.2112.9126.0126.3138.5135.7135.2c4128.5128.8128.9115.5117.9122.0129.6129.6130.3115.81 14.61 1 1.7118.5133.3133.6130.4131.8132.3a For MeCO, groups: 6 24.4, 22.0; 177.3, 174.8.In (CD,),SO.c5135.0135.3136.3163.6163.7163.9149.1153.6153.6157.4149.5137.7153.8126.0126.3136.7132.6132.6C6130.6132.8135.5129.9130.0130.4131.0152.3133.2123.9124.0122.8153.9146.8146.9133.8132.5132.4c7179.7180.5182.4179.3180.0182.0179.3179.5179.875.675.574.676.281.281.442.0183.4182.0C8 C9 C'O R84.8 69.0 27.384.5 69.7 27.983.5 70.2 28.784.6 68.6 27.3 56.084.3 68.2 27.8 55.983.6 70.0 29.0 56.084.4 68.0 27.5 40.584.1 68.3 27.7 40.483.9 68.6 27.9 40.353.8 55.740.9 53.640.3 52.753.6 56.3, 62.658.158.244.5 20.1Table 5 Co-ordination chemical shifts, A6*/ppmComplex C' C6 C7 C82a2b2c3a3b3c3f45a5b6a1 Oa10b19.4 16.7 17.3 4.820.4 16.8 18.0 4.522.1 17.2 18.0 3.824.0 1.3 16.7 5.620.7 2.6 16.9 5.320.5 2.5 17.2 5.124.2 23.0 21.3 7.918.5 10.2 12.1 8.818.2 11.8 11.7 8.418.4 10.6 10.8 7.528.0 5.2 12.32 8.532.7 10.1 17.1 13.032.8 8.45 17.3 13.1* A6 = G(comp1ex) - G(aromatic ligand).bond lengths are sensibly different with that trans to the phenylgroup being the longer, as expected on trans-influence grounds.For 10a the Au-Cl bond length is closely similar to the longerof the two bonds in 6a, suggesting that the carbon donor atomof the tridentate ligand in 10a has a similar trans influence tothat of the analogous bidentate ligand in 6a.However, thetridentate co-ordination results in a slightly shorter Au-Cbond, while the Au-N bonds are a little longer.Acetate complexesThe complex [AuCl,(dmamp)] reacted readily with silveracetate to give a diacetato complex with considerably enhancedaqueous solubility.5 , 6 For the present compounds, stableproducts were obtained from complexes 5a and 6a, containingthe dmamp analogues (the reaction of 4a was not attempted).These products, 5b and 6b, like [Au(O,CMe),(dmamp)], wererather light sensitive. The oxazoline complexes la-3a alldecomposed to metallic gold in the presence of silver acetate,even when light was rigorously excluded. Similar decompositionoccurred in other reactions of these complexes (see above), andwas attributed to loss of the oxazoline group. The most likelyreaction is conversion of the oxazoline into a carboxylic acidderivative.In principle, this group could bind to gold, forminga five-membered ring; it appears that this system is not stable,at least when the other ligands are hard (e.g. halide, acetate).Complex 6b was prepared only on a very small scale, and nospectroscopic data were recorded. For 5b, the carboxylateinfrared vibrations were observed at 1599, 13 13 and 1 502, 1370cm-', which are assigned to acetate ligands trans to nitrogenand to carbon, respectively. Distinct NMR signals were alsoseen for the two acetate groups (Tables 3 and 4). As for[Au(O,CMe),(dmamp)], the pairs of I3C signals were ofdifferent intensity; the pair with the higher chemical shifts (624.4, 177.3) was broader and less intense than the second set (622.0, 174.8).A detailed study of the dmamp system ' showed3188 J. Chem. SOC., Dalton Trans., 1996, Pages 3185-319Table 6 Carbon- 13 NMR data for sulfonamide and amide derivatives in (CD3),S0c3 c4 c5 C6 c7 c8 c9 Compound C' C2[HgCI(C6H4S02NMe,-2)] 144.3 155.9 136.5 142.5 132.9 133.0 44.811 [AuC12(C6H4S02NMe2-2)] 142.0 146.1 137.2 138.5 130.4 134.6 51.8[HgC1{C6H4C(o)NH2-2}1 155.1 141.5 141.4 131.5 135.4 131.5 174.612 [AuC12(C6H4C(0)NH,-2}] 145.7 140.3 135.7 131.8 132.6 132.5 183.4[HgCI( C,H,C(O)NHEt-2)] 155.6 141.8 140.9 131.4 131.5 135.1 172.3 38.4 18.513 [AuCI,(C,H,C(O)NHEt-2}] 147.7 141.9 141.7 131.2 132.4 134.6 181.8 44.3 19.9Table 7and 6aSelected bond lengths (A) and angles (") for complexes l al a 6aAu-CI( 1) 2.350(3) 2.367(8)Au-CI( 2) 2.277(3) 2.269(8)Au-C( 1) 2.040( 8) 2.08( 2)Au-N 2.051(8) 2.05(3)C1( 1 )-Au-C1(2) 90.2( 1) 87.8(3)C1(2)-Au-C( 1) 92.0(3) 97.6(8)C1( 1 )-Au-N 96.1(2) 95.0(7)N-Au-C( 1) 81.7(3) 80(1)Table 8 Selected bond lengths (A) and angles (") for [AuCI-( c 6H 3 (CH 2 NMe 2 ) ,-2,6}] 2 [ Hg2 CI 61 10aCation Anion1 )-N( 1 ) 2.1 l(2) Hg( 1 )-Cl(2) 2.367(6)1 W ( 2 ) 2.12(1) Hg(l)-C1(3) 2.366(7)Au( 1 )-Cl( 1 ) 2.369(6) Hg( 1)-C1(4) 2.651(6)Au(l t-C( 1 1.96(2) Hg( l)-C1(4 *) 2.643(5)C1( I)-Au( 1 )-N( 1) 97.8(4) Hg( l)-C1(4)-Hg( 1 *) 88.4(2)N(1)-Au(1)-C(1) 82.0(7) C1(2)-Hg(l)-C1(4*) 103.7(2)C1( 1 )-Au( 1 )-N(2) 98.3(4) C1(2)-Hg( 1 )-C1(3) 137.2(2)N(2)-Au( I )-C( 1) 82.0(7) C1(2)-Hg( 1)-C1(4) 107.6(2)C1(3)-Hg( 1 )-C1(4 *) 106.9(2)C1(3)-Hg( 1 )-C1(4) 100.6(2)C1(4)-Hg( 1 )-C1(4 *) 9 1.6(2)* The asterisk denotes the symmetry-related position 1 - x, 2 - y ,1 - 2 .Q 8Fig.2 Molecular structure of complex 6a1Fig. 3 Molecular structure of complex 10acomplex 6b reacted readily with 2 molar equivalents ofpyridinium perchlorate to give the dicationic complexpyridine) for which spectroscopic data were not obtained.c~~(~,~,~C~,~~e,-2~~O~~~,-5,6}~PY~,lC10, 6c (py =Fig. 1 Molecular structure of complex l athat this behaviour is due to hydrolysis by adventitious traces ofwater in the CDCI, solvent used. The acetate group trans tocarbon undergoes exchange with water at a rate comparable tothe NMR time-scale.In a reaction parallel to that of its dmamp analogue,IgDi thiocarbama te complexesReaction of the oxazoline complexes la, 2a, 3a or theaminomethylnaphthyl complex 8a with one molar equivalentof sodium diethyldithiocarbamate, followed by addition ofsodium hexafluorophosphate or tetraphenylborate, gave solidsIf, 2e, 2f, 3e and 8b.These materials behaved as 1 : I electrolytesin acetone (AM = 120, 124 and 122 S mol-' for the PF,- saltsIf, 2f and 3g). The infrared spectra (Table 2) confirm thatthe oxazoline groups remain co-ordinated to gold. The C-Nstretching frequencies of the S,CNEt2 ligands are similarto those of other gold(m) complexes containing chelatedJ. Chem. SOC., Dalton Trans., 1996, Pages 3185-3193 318dithiocarbamate ligands (Table 9) Monodentate dithiocar-bamates, such as the gold(1) complexes [Au(S,CNR,),] -, givelower frequencies (1488-1506 cm-', R = Me, Et, Pr" or Bun).,'Nevertheless, as in the corresponding dmamp complexe~,~ it isexpected that the two Au-S bonds would be different in length,that trans to the carbon atom being longer; the two C-S bondsare presumably also inequivalent.The lack of planes of symmetry in these molecules, and inthe S,CNEt, units in particular, is reflected in the I3C NMRspectra, where the ethyl groups are not equivalent (Table 10).These non-equivalences are barely resolved in the 'H spectra.The presence of an S-donor ligand trans to C1 increases the co-ordination chemical shift values (owing to the change in solventthe closest values for comparison are those for 11-13, Table 6).The chemical shifts for the carbon atoms of the oxazoline rings,C7 and Cs, also confirm that the oxazoline group remains co-ordinated, as does the NMe, group in 8b.Complex 2a reacts with 2 molar equivalents of NaS2CNEt,to give 2g.The spectroscopic properties of this complex aremarkedly different from those of the monodithiocarbamatecomplexes. In the infrared spectrum (Table 2) the typical C-Nband of the oxazoline group occurs at a frequency (1 643 cm-')comparable to that of the free aromatic and of the mercury(r1)derivative (1649, 1645 cm-' respectively) and about 30 cm-'higher than for the other gold(rI1) complexes reported here. Thisstrongly suggests that the oxazoline group is no longer co-ordinated to gold.The S,CNEt, ligands show a single broadC-N stretching band at about 1531 cm-', about 30 cm-l lowerthan the corresponding band for 2e.The 13C NMR spectrum also indicates that the oxazolinegroup is not co-ordinated: the chemical shifts of its carbonatoms occur at markedly lower chemical shifts than for any ofthe other oxazoline derivatives (Table 10). On the other handthe chemical shift for C' has decreased substantially comparedto those of 2e and 2f, which also suggests a change in the modeof co-ordination of the aryl ligand. Only a single set of signalsis found for the two S,CNEt, ligands; there is no sign of non-equivalence for the ethyl groups, and the CS2 carbon atomappears as a single sharp signal. While it is possible that thiscould indicate five-co-ordination for the gold, with twobidentate S,CNEt, ligands, it is more likely that one dithio-carbamate is monodentate.In the case of the correspondingTable 9 The v(C-N) bands (cm-') for dithiocarbamate complexesCompound v(C-N) Compound V( C-N)l e 1552 [AuMe,(S,CNMe,)] ' 15532e 1558' [AuMe,(S,CNEt,)] ' 15383e 1535 [A~(S,CNMe,)(dmarnp)]BPh~~ 1568a Ref. 20. ' The PF6- and BPh4- salts gave the same value. Ref. 5.dmamp complex it was shown crystallographically that, inthe solid state, one S,CNEt, was monodentate. The solutionbehaviour was explained by fast alternation of the twoS,CNEt, ligands between mono- and bi-dentate binding.DiscussionThe stability of the five-membered AuC,N chelate ringdemonstrated by Vicente and co-workers for the dmampcompIexes7*8,'1-14 extends to the other complexes shown here,in the ring-substituted analogues of dmamp and its tridentateanalogue, and the oxazoline ligands.Two six-membered-ringsystems have also been obtained, in the aminomethylnaphthylderivatives 8 and 9. The susceptibility to nucleophilic attackof the oxazoline group, which makes it an easily removedprotecting group in organic chemistry, is still evident and mayeven be enhanced by co-ordination. Substitution reactionswhich, in other complexes, are rapid and quantitative resultedin the production of gold metal; the organic products of thesereactions were not identified.The chloro-complexes were obtained by transmetallationfrom the corresponding arylmercury compounds.The originalpreparation of [AuCl,(dmamp)] employed [AuCl,(tht)](tht = tetrahydrothiophene) as the gold-containing startingmaterial and acetone as the ~ o l v e n t . ~ We have found[NMe,][AuCl,] to be better.6b The reaction is also markedlysolvent dependent, being more efficient in polar solvents such asacetonitrile and dimethyl sulfoxide. Although the mechanism ofthe transmetallation reaction seems not to have been studied, itis likely in this case to proceed via formation of a bimetallicintermediate, by co-ordination of the N-donor group to gold(Scheme 2). The role of the solvent may be to inhibit associationof the N-donor group with the mercury atom. This would alsobe consistent with the failure to obtain successful metallationwith [AuClJ and lithium aryls, when the aryl carries asubstituent which is a potential ligand;7 presumably the bindingof the ligand group to gold is inhibited by its binding to thelithium cation.The most interesting reactions of the chloro-complexes arethose with dithiocarbamates. Even when the stoichiometry isstrictly controlled to be 1 : 1, both chloride ligands are replaced;the single S,CNEt, ligand chelates to gold(rIr), giving square-planar CNS,-co-ordinated cations (Scheme 3).The softS,CNEt, nucleophile binds to the gold directly, and does notattack the oxazoline group, which remains co-ordinated.Addition of a second S,CNEt, ligand gives products which alsoretain the oxazoline group but it is no longer co-ordinated.Although the two S,CNEt, ligands appear equivalent in theNMR spectra, it is very likely that one is bi- and the othermono-dentate, with a rapid exchange between them.This non-Table 10Complexle2e2f2g3e8bbCarbon- 13 NMR data for diethyldithiocarbamate complexes in (CD,),SOR C'152.35-OMe 154.15-OMe 154.15-OMe 147.25-NMe, 153.9143.9152.8CZ132.7124.0124.1123.1115.3135.9152.2c3 c4132.8 132.0117.5 118.5118.5 117.6112.4 116.7113.8 114.1133.9 132.5128.7 132.0c5133.0168.1166.0161.4147.3129.9131.7C6139.2135.1135.1130.7134.0130.9132.9c7183.9183.7183.8163.8183.9133.977.1c8 c9 c10 cll87.1 70.7 30.887.1 70.7 31.1 60.387.2 70.8 31.2 60.479.2 67.7 28.4 55.686.7 70.1 31.2 58.4134.6 138.2 130.7 68.856.6DithiocarbamateCS, CH, CH,197.1 51.2 16.150.7 15.9183.7 51.3 16.450.9 16.1197.3 51.4 16.450.9 16.2200.9 47.1 12.5197.6 51.2 16.450.8 16.1196.1 52.1 16.651.1 16.3195.5 48.4 12.747.1 12.4a For NMe,, 6 57.4.' [A~(S,CNEt,)(drnamp)]BPh~.~3190 J. Chem. SOC., Dalton Trans., 1996, Pages 3185-319equivalence has been demonstrated crystallographically for thedmamp complex,' and is similar to the behaviour of the bis--dmamp complex [AuCl(dmamp),]. 21 In the latter complexthe dmamp ligands switch rapidly between mono- and bi-dentate co-ordination at room temperature (Scheme 4) but theexchange slows sufficiently at 230 K to allow observation ofseparate NMR signals for the two ligands.This exchange ispostulated to occur by dissociation and rebinding of thechloride ligand; the cation [Au(dmamp),] + has been isolatedas the perchlorate. The S,CNEt, case might be analogous,involving an intermediate (14, Scheme 3) with two monodentateS,CNEt, ligands and a chelating N, C ligand. However, the I3CNMR spectrum of [Au(S,CNEt,)(dmamp)] at 230 K showsbroadening of the S,CNEt, signals but none for those of theCH,NMe, group;' equally there is no change in the positionof these latter signals. It is therefore probable that the N,Cligands remain monodentate and that the S,CNEt2 exchangeoccurs through a five-co-ordinate intermediate containing twobidentate S,CNEt, ligands (15).ExperimentalElemental analyses were carried out by the UMIST ChemistryDepartment Microanalytical Service and the positive-ion fastatom bombardment mass spectra were recorded by the UMISTCentre for Mass Spectrometry; m/z values are quoted for ,*Cl.Infrared spectra (4000-300 cm-') were recorded on a Nicolet5PC Fourier-transform spectrometer in Nujol mulls betweenKBr plates, 'H and I3C NMR spectra on a Bruker AC-200spectrometer at respectively 200 and 50.3 MHz in CDCI, or(CD,),SO at 25 "C using the solvent signal as internal standard.The 'H and 13C spectra of complex la were fully assigned byuse of homo- and hetero-nuclear correlated two-dimensionalspectra.It was assumed that the chemical shifts are in the sameorder for the substituted gold complexes.SynthesesThe chloro-complexes la-7 were prepared as follows: the ap-propriate organomercury(I1) precursor l o ( 5 mmol), tetramethyl-ammonium tetrachloroaurate(II1) (5 mmol) and tetramethyl-cIHg I CI+ctCI, IAu I 'CIClammonium chloride were stirred in anhydrous acetone oracetonitrile (40 cm3) at room temperature for 2 d, during whicha white precipitate was formed, [NMe,][Hg,Cl,].The reactionmixture was filtered, and the solid carefully washed withanhydrous acetone or acetonitrile (2 x 15 cm3). The solventwas removed from the combined extracts under reducedpressure, and the remaining solid extracted into CH2C12(3 x 40 cm3). The solution was left to evaporate slowly,depositing 3443 mmol of yellow crystals.Complexes 8a-10a were prepared in a very similar manner.Sodium tetrachloroaurate(r1r) was used instead of thetetramethylammonium salt and no NMe,CI was added to thereaction mixture.The complexes 8a and 9 did not crystallise,but white cubic crystals were obtained from the slow evapor-ation of an acetone solution of 10a.Complex 10b was obtained from [2,6-bis-(N,N-dimethyl-aminomethyl)phenyl]chloromercury(~r) dihydrochloride (5mmol) and Na[AuCl,] ( 5 mmol), which were dissolvedand stirred in water (40 cm3) for 1 h. The bright yellowprecipitate obtained was filtered off, washed with water (2 x 15cm3), dried in vacuo and crystallised from acetone (3.9 mmol).Complexes 5b and 6b were prepared by stirring for 2 ha suspension of the appropriate cis-dichloroorganogold(II1)complex 5a or 6a (0.5 mmol) and silver acetate (1 mmol) in dryacetone (30 cm3).The solution was filtered, the solventevaporated under reduced pressure and the residue extractedinto dichloromethane (3 x 20 cm3). Evaporation of the solventleft 0.37 mmol of a white solid.Complex 6c was synthesised by stirring a suspension of6b (0.3 mmol) and pyridinium perchlorate (0.7 mmol) indichloromethane (20 cm3) for 3 h. The excess of pyridiniumperchlorate was filtered off and the solution left slowly toevaporate, depositing 0.1 3 mmol of white crystals. CAUTION:perchlorates are potentially explosive, and should be handledwith care and only in small quantities.Complexes lb-ld, 2b-2d and 3b3d were prepared as follows:A suspension of the appropriate cis-dichloroorganogold(II1)complex la-3a (0.4 mmol) and either lithium bromide (0.8mmol), potassium iodide (0.8 mmol) or potassium thiocyanate(0.8 mmol) in acetone (30 cm3) was stirred for 3 h.The solventwas removed under reduced pressure, and the remaining solidNI T"' CICICIIHg I CI+ClScheme 2,s1415Scheme 3 ( i ) NaS,CEt,Scheme 4J. Chem. SOC., Dalton Trans., 1996, Pages 3185-3193 319Table 11 Experimental data for the crystallographic analysesFormulaMCrystal systemSpace groupalAblACIA$ i 3 zDJMg m-3No. reflections for lattice parameters8 RangerflowCrystal size/mmplmrn-'Absorption correction (minimum, maximum)o-Scan speed/" min-'o-Scan width/"8 Rangerh,k,l RangesNo. measured reflectionsNo. reflections used in refinement [I > 30(1)]Goodness of fitMaximum shift/errorMinimum, maximum height infinal A F maple A-3RR'c in w = [02(F,) + cFO2]-'Rintl aC1 ,Hl,AuC12N0442.09MonoclinicP2,lc (no.14)8.37 l(4)14.624(3)1 0.630( 3)96.28(3)1294(1)42.2702526.33-35.928240.20 x 0.20 x 0.20117.330.91, 1.088.01.13 + 0.30 tan 80-25.5255314250.0641.0360.080-9,0-16, -12 to 12-0.59,0.640.0280.0300.030extracted into dichloromethane (3 x 15 cm3). The solvent wasevaporated from the combined extracts to give the substitutedproducts as dark orange (lb, 2b, 3b), dark red (lc, 2c, 3c) andpale yellow solids (la, 2d, 3d).Complexes le, 2e, 2f and 3e were prepared from theappropriate cis-dichloroorganogoId(1n) (0.4 mmol) dissolvedin methanol (40 cm3).To this solution was added sodiumdiethyldithiocarbamate (0.4 mmol). The mixture was stirred for2 h, and a solution of ammonium hexafluorophosphate (0.5mmol) or sodium tetraphenylborate (0.5 mmol) in methanol (1 5cm3) was added. The solution was stirred for 30 min, filteredand the solvent evaporated under reduced pressure. The residuewas extracted into acetone (30 cm3) and the solution was leftto evaporate, depositing approximately 0.25 mmol of whitecrystals.Complex 2g was prepared by stirring 2a (0.5 mmol) andNaS,CNEt, (1 mmol) for 1 h in acetone-methanol (1 : 2, 50cm3). The solvent was removed under reduced pressure and theremaining solid extracted into dichloromethane (30 cm3). Thesolution was concentrated to 10 cm3 and complex 2g wascrystallised by addition of diethyl ether (5 cm3).Complexes l e and 3f were prepared by adding PPh, (0.5mmol) in dichloromethane (10 cm3) to the dichloro-complex laor 3a (0.5 mmol) in the same solvent (40 cm3) and stirring themixture for 1 h.The solution was then concentrated to about 15cm3 and diethyl ether added (ca. 10 cm3); the products slowlycrystallised.CrystallographySuitable crystals of complexes la, 6a and 10a were obtainedby slow evaporation of the solvent from a dichloromethane(la and 6a) or acetone solution (10a). X-Ray diffractionmeasurements were performed at room temperature on aRIGAKU AFC6S diffractometer using graphite-monochroma-tised Mo-Ka radiation. The structures were solved by directmethods using SHELXS.22 The structures were subjected to6aC1 ,HI ,AuCl,NO,462.13MonoclinicP2,ln (no.14)9.46 1 (2)8.895(3)16.604(4)92.00( 3)1397(1)42.198198720.30 x 0.30 x 0.20108.770.97, 1.038.01.18 + 0.30 tan 80-20.25144975 10.1403.3021.0714.4 1-20.3 30-6,0-8, - 15 to 15-1.17, 1.070.0460.0570.020I OaC24H38Au2C18Hg2N41461.32MonoclinicP2Jn (no. 14)9.848( 3)18.778(7)9.894( 3)92.95(3)1827(2)22.26562513280.25 x 0.25 x 0.25169.700.94, 1.082.01.21 + 0.30tan80-250-10,o-22, -11 to 11347418720.0641.8951.851 6.0 1-23.75- 1.40, 1.850.0450.0540.030full-matrix least-squares refinement based on F. Non-hydrogenatoms were treated anisotropically and hydrogen atoms wereconstrained to chemically reasonable positions. No reflectionswere recorded above 40" for 6a because of its poor crystal-linity, which also led to a high Rint value and necessitatedisotropic refinement of the nitrogen and several carbon atoms.Crystallographic data are given in Table 1 1.Atomic coordinates, thermal parameters and bond lengthsand angles have been deposited at the Cambridge Crystallo-graphic Data Centre (CCDC). See Instructions for Authors,J.Chem. Soc., Dalton Trans., 1996, Issue 1. Any request to theCCDC for this material should quote the full literature citationand the reference number 1 86/ 123.AcknowledgementsWe are grateful to Johnson Matthey plc for the loan of goldsalts .References1 B.Rosenberg, L. Van Camp and T. Krigas, Nature (London), 1965,205,698.2 N. Farrell, in Transition Metal Complexes as Drugs andChemotherapeutic Agents, Kluwer Academic Press, London, 1989.3 For a review see D. C. H. McBrien and F. T. Slater (Editors),Biochemical Mechanisms of Platinum Antitumour Drugs, IRL press,Oxford, 1986.4 A. Dar, K. Moss, S. M. Cottrill, R. V. Parish, C. A. McAuliffe,R. G. Pritchard, B. Beagley and J. Sandbank, J. Chem. Soc., DaltonTrans., 1992, 1907.5 R. V. Parish, B. P. Howe, J. P. Wright, J. Mack, R. G. Pritchard,R. G. Buckley, A. M. Elsome and S. P. Fricker, Inorg. Chem., 1996,35, 1659.6 (a) R. V. Parish, J. Mack, L. Hargreaves, J. P. Wright, R. G.Buckley, A. M. Elsome, S. P. Fricker, and B. R. C. Theobald,J. Chem. Soc., Dalton Trans., 1996, 69; (b) R. G. Buckley, A. M.Elsome, S. P. Fricker, G. R. Henderson, B. R. C. Theobald,L. Kelland, B. P. Howe, and R. V. Parish, J. Med. Chem., submittedfor publication.3192 J. Chem. SOC., Dalton Trans., 1996, Pages 3185-3197 J. Vicente, M. T. Chicote and M. D. Bermudez, J. Organomet.8 J. Vicente and M. T. Chicote, Inorg. Chim. Acta, 1981,54, L259.9 E. C. Constable and T. A. Leese, J. Organomet. Chem., 1989, 369,419.10 P.-A. Bonnardel and R. V. Parish, J. Organomet. Chem., 1996,515,221.1 1 J. Vicente, M. D. Bermudez, M. T. Chicote and M. J. Sanchez-Santano, J. Organomet. Chem., 1989,371, 129.12 J. Vicente, M. T. Chicote, M. D. Bermudez, P. G. Jones, C. Fittschenand G. M. Sheldrick, J. Chem. SOC., Dalton Trans., 1986, 2361.13 J. Vicente, M. T. Chicote, M. D. Bermudez and M. Garcia-Garcia,J. Organomet., Chem., 1985,295, 125.14 J. Vicente, M. T. Chicote, M. D. Bermudez, M. J. Sanchez-Santano,P. G. Jones, C. Fitschen and G. M. Sheldrick, J. Organomet. Chem.,1986,310,401.15 A. I. Meyers, D. L. Temple, D. Haidukewytch and E. D. Mihelic,J. Org. Chem., 1974,39, 2787.Chem., 1984,268, 19 1 .16 F. A. Cotton, R. Francis and W. D. Horrocks, J. Phys. Chem., 1960,64, 1534; F. A. Cotton and R. Francis, J. Am. Chem. Soc., 1960,82,2986.17 R. A. Potts, J. Inorg. Nucl. Chem., 1972,34, 1749.18 J. A. J. Jarvis, A. Johnson and R. J. Puddephatt, J. Chem. Soc.,Chem. Commun., 1973,373.19 J. Vicente, M. T. Chicote, M. D. Bermudez, P. G. Jones andG. M. Sheldrick, J. Chem. Rex, 1985, ( S ) 72; (M) 954.20 H. J. A. Blaauw, R. J. F. Nivard and G. J. M. van der Kerk,J. Organomet. Chem., 1964,2,226.21 J. Vicente, M. D. Bermudez, M. J. Sanchez-Santana and J. Paya,Inorg. Chim. Acta, 1990, 174, 53.22 G. M. Sheldrick, in Crystallographic Computing 3, eds. G. M.Sheldrick, C. Krueger and R. Goddard, Oxford University Press,1985, pp. 175-189.Received 4th March 1996; Paper 6/01 5401J. Chem. SOC., Dalton Trans., 1996, Pages 3185-3193 319
ISSN:1477-9226
DOI:10.1039/DT9960003185
出版商:RSC
年代:1996
数据来源: RSC