The First Complexes of Tetragoldmethane CAu4 H. Schmidbaur* and O. Steigelmann Anorganisch-chemisches Institut der Technischen Universität München, Lichtenbergstraße 4, D-W-8046 Garching Z. Naturforsch. 47b, 1721-1724(1992); received September 4, 1992 Organogold Compounds, Tetragoldmethane, (Phosphine)gold(I) Complexes, Borylmethanes Treatment of C[B(OMe)J4 with four equivalents of a (rerr-phosphine)gold(I) chloride LAuCl with L = P(oC6Hn)3 or (2-MeC6H4)3P in the presence of CsF in hexamethylphosphoric triamide (HMPT) affords the first complexes o f the type (LAu)4C. The product is formed to­ gether with salts (LAu )5C+BF4_ as the by-product for L = tricyclohexylphosphine, from which it can be separated by fractional crystallisation. With Ph,(2-MeCH4)P only the hexacoordinate species (LAu)6C2+2 BF4~ is obtained, while with Ph(2-MeC6H4)2P a mixture of (LAu)6C2+ and (LA u )5C+ salts is generated. The title compounds are stable, colourless solids readily identified by standard analytical and spectroscopic techniques, including l3C NMR of l3C-enriched ma­ terial.

All previous attempts to generate a species CAu4 bulky substituents at phosphorus appeared to be or its complexes, e.g. C(AuL)4 with L a donor li­ the most obvious concept. gand, have met with failure [1-3]. Initially these Since the reaction of (tri^/zewj/phosphine)- negative results came as a surprise, since 1) analo­ gold(I) chloride with tetrakis(dimethoxyboryl)- gous mercury compounds of the type C(HgX)4 methane in the presence of CsF in hexamethyl­ have been known in the literature for a long time phosphoric triamide [6] leads to [(Ph3PAu)6C]2+, in [4, 5], and 2) since aurated ammonium cations of a first series of experiments the phenyl substituents the type N(AuL)4+ are well established [6-10]. were gradually exchanged by sterically more de­ Careful investigation of the auration of suitable manding o-tolyl groups. While the mono(o-tolyt)~ functional methane precursors like poly(boryl)- diphenylphosphine ligand (2-MeC6H4)Ph2P did methanes has recently shown, however, that com­ not yet change the course of the reaction, both plete auration of carbon does in fact take place, (2-MeC6H4)2PPh and (2-MeC6H4)3P induced the but is transgressing the stage of C(AuL)4 molecules expected reduction in the coordination number of leading instead to hypercoordinate cationic species carbon in the product from six to five, and finally of the types C(AuL)5+ and even C(AuL)62+. These to four: cations contain penta- and /zexacoordinate carbon Ph3(2-MeC6H4)PAuCl C[B(OMe);]4. CsF. HMPT atoms with trigonal bipyramidal and octahedral geometry, respectively [11-15]. Molecules [Ph2(2-MeC6H4)PAu]6C2+ 2 BF4~ C(AuL)4 were thus recognized to be “auriophilic” Ph(2-MeC6H4)2PAuCl -> and function as strong nucleophiles towards LAu+ [Ph(2-MeC6H4)2P Au ]5C+BF4", units in the reaction medium. This unprecedented [Ph(2-MeC6H4)2PAu]4C phenomenon has meanwhile been rationalized on (2-MeC6H4)3PAuCl [(2-MeC6H4)3PAu]4C the basis of sophisticated theoretical calculations including relativistic effects [16-20], The preparation yields a mixtures of products Notwithstanding, there remained the challenge with different stages of carbon auration for L = to synthesize the elusive C(AuL)4 species. The in­ J/(o-^o/y/)phenylphosphine, which are difficult to troduction of reagents which prevented auration separate due to facile exchange of LAu+ units. of carbon beyond tetra-coordination through Though identification of the components of the mixture is readily accomplished, mainly by NMR spectroscopy (Exp. Section), isolation of analyti­ * Reprint requests to Prof. Dr. H. Schmidbaur. cally pure samples is extremely tedious. In further experiments, therefore, tri cyclohexyl- Verlag der Zeitschrift für Naturforschung, D-W-7400 Tübingen phosphine was used as a ligand, which owing to its 0932-0776/92/1200-1721/$ 01.00/0 still larger cone angle should favour more strongly 1722 H. Schmidbaur-O. Steigeimann • The First Complexes of Tetragoldmethane CAu4 the tetracoordination of carbon, and reduce ligand gether. It should be remembered that tetra(auro)- exchange rates. arsonium salts have been shown to have a square (c-C6H n)3PAuCl — [(c-C6 Hn)3PAu]5C+BF4- pyramidal Au4As+ core [10]. This violation of the [(c-C6H u )3PAu ]4C 1 classical Le Bel/vant’Hoff and Lewis concepts (which require arsonium cations with an octet of P(c-C6Hn )3 electrons at arsenic to be tetrahedral) is a conse­

Au quence of strong intramolecular Au--Au attrac­ I tions which lead to a reduction of the A u-A s-A u / c x Au I Au angles from the tetrahedral 109° to ca. 80° in the (LAu)4As+ cation. Tetra(auro)methanes with ic6h„i3p/' T Plt- W 3 P(c-C6Hn »3 small ligands L are also expected to be strongly distorted, and this unusual structural feature is to 1 be taken as one of the reasons for the hypercoordi- nation phenomena observed in the corresponding Though penta-auration was again not fully sup­ system. pressed, and the reaction mixture contained both the penta- and tetra-aurated species, separation of the components presented no problem in this sys­ tem. Experimental Pure tetrakis(tricyclohexylphosphinegold)meth- All experiments were routinely carried out in an ane (1) was obtained in a 23% isolated yield as a atmosphere of dry, purified nitrogen. Glassware colourless, polycrystalline solid, which is soluble in was oven-dried and filled with nitrogen, and sol­ di- and trichloromethane, tetrahydrofuran, ben­ vents were purified, dried and saturated with ni­ zene, and toluene, but only very sparingly soluble trogen. NMR: Jeol GX 270 and GX 400 spectro­ in diethylether and pentane. For solutions in meters; MS: Finnigen MAT 90 spectrometer, in CD2C12 the 31P resonance appears at S = 48.4 ppm. field desorption or fast atom bombardement The 'H NMR spectrum shows only a multiplet at mode. The tertiary phosphines and their AuCl com­ S = 1.14-2.07 ppm for the cyclohexyl groups, but plexes were obtained using standard methods with the CH quartet resonance clearly distin­ [21-24]: guished at Ö = 1.47. For 13C NMR spectroscopy, a (2-MeC6H4)Ph,PAuCl: m.p. 201-205 °C, 31P sample 13C-enriched in the center of the molecule N M R (CD2C12): ÖP 26.9 ppm. was prepared following a procedure described else­ (2-MeC6H4)2PhPAuCl: m.p. 202-205 °C, ÖP = where [14] for other poly(auro)methanium com­ 17.5 ppm. pounds. This labeled material shows a resonance (2-MeC6H4)3PAuCl: m.p. 212-215 °C, ÖP = for the interstitial carbon atom at S = 99.3 ppm as 8.7 ppm. a 1:4:6:4:1 quintet with 2/(CAuP) = 61.0 Hz. (c-C6H u)3PAuC1: m.p. 221-223 °C, SP = (The same splitting can be extracted from the satel­ 54.9 ppm. lites in the 31P spectrum of the enriched material.) C1-C4 of the C6Hn groups are located at 34.5 (.J = 23.7 Hz), 27.9 (J = 12.2 Hz), 26.7, and General synthetic procedure 30.9 ppm. The field desorption mass spectrum A suspension of the (phosphine)gold(I) chloride shows a parent peak at m/e = 1921.5. and an excess of cesium fluoride in hexamethyl- All attempts to grow single crystals of one of the phosphoric triamide (ca. 30 ml) is treated at room tetragoldmethane complexes were unsuccessful. temperature and with stirring with a solution of the stoichiometric amount of tetrakis(dimethoxy- For the present cases there is little doubt that the boryl)methane in 10 ml of HMPT. The reaction CAu4 cores of the molecules have a tetrahedral mixture is stirred for 24 h and then filtered. Pen­ structure. However, this prediction is only valid tane is slowly added to the filtrate to give a precipi­ for complexes with bulky ligands, where inter- tate, which is washed with pentane and recrystal­ ligand repulsion prevents any structural reorgani­ lized from dichloromethane/pentane or tetra- zation which could bring the gold atoms closer to­ hydrofuran/diethylether. H. Schmidbaur-Q. Steigeimann • The First Complexes of Tetragoldmethane CAu4 1723

Hexakis[ ( o-tolyl) diphenylphosphine-gold( I) ]- NM R (CD2C12): S = 23.0. 13C NM R (CD2C12): methanium (2+)bis( tetrafluoroborate) 23.1 ( / = 9.8, Me), 128.0 (J = 49.4), 143.4 (J = 1.05 g (2.06 mmol) 2-MeC6H4Ph2PAuCl; 0.16 g 12.2), 132.4 (J = 7.9), 131.5, 126.8 (J = 9.8), 134.1 (0.53 mmol) [(MeO)2B]4C; 5.10 g (33.7 mmol) (J = 9.2), (C1-C6 of o-tolyl). >H NMR (CD2C12): CsF. Yield 0.20 g (20%), m. p. 155 °C (decomp.). 2.15, Me; 6.86-7.35, Aryl. CI15HW2Au6B2F8P6 (3025.34) Pentakis[ tricyclohexylphosphine-gold(I) ]- Calcd C 45.66 H 3.40%, methanium( +1) tetrafluoroborate and tetrakis- Found C 44.97 H 3.51%. [tricyclohexylphosphine-gold(I) ]methane (1) MS (FAB, /7-nitrobenzylalcohol): m/e = 1425.1 1.42 g (2.77 mmol) (c-C6Hn)3PAuCl; 0.22 g [10%, M2+(cation)]. 31P NMR (CD2C12): ö = (0.72 mmol) [(MeO)2B]413C (20% enriched in the 23.4 ppm. ,3C NM R (CD2C12): 22.5, Me; 128.6 position indicated), 5.22 g (34.5 mmol) CsF. Yield (J = 56.1), 141.7, 132.6 (J = 7.0), 132.4, 127.0 0.80 g of a mixture of both products. The meth- (J= 3.0), 133.6 (J = 3.8) (C l- C 6 of o-tolyl); 129.2 anium salt has the following spectroscopic data (J = 57.2), 134.4 (J = 1.5), 129.7 (J = 3.0), 132.3 taken from solutions of this mixture: 3IP NMR (C1-C4 of phenyl). 'H NMR (CD2C12): 2.04, Me; (CD-Gl?): Ö = 50.9, doublet (7 = 60.3). 13C N M R 6.79-7.44 (o-tolyl, phenyl). (CD2C12): Ö = 185.3, quintet (J - 60.3). MS (FD): m/e = 2398.3 [100%, M+(cation)]. Extraction of the mixture with benzene and reprecipitation with Pentakis[ di( o-tolyl)phenylphosphine-gold( I) ]- pentane gives pure tetraauromethane 1, yield methanium(+l)tetrafluoroborate (M) and tetrakis- 0.31 g (23.3%). [di (o-tolyl) phenylphosphine-gold (I)] methane (M') 1.00 g (1.91 mmol) (2-MeC6H4)2PhPAuCl; 0.15 g C73Hi32A u4P4 (1921.62) (0.49 mmol) [(MeO)2B]4C, 4.21 g (27.9 mmol) CsF. Calcd C 45.63 H 6.92%, Yield 0.31g of a mixture of both products. 31P Found C 44.89 H 6.88%. NM R (CD,Cl-,): Ö = 24.8 and 25.4. !H N M R MS (FD): 1921.5 (45%, M +). 3,P NM R (CD2C12): 2.20 and 2.32 (Me); 6.82-7.63, Aryl. (CD2C19): 8 = 48.4, doublet (J = 61.0). I3C NM R MS (FÄB): m/e = 2445.2 [59%, M+(cation)]; (CDiC12): 99.3, quintet (./ = 61.0); 34.5 (J = 23.7), 1961.2 [18%, (AT)+]. 27.9'(/ = 12.2), 26.7, 30.9 (C 1 -C 4 of cyclohexyl). 'H NMR (CD2C12): 1.14-2.07, m, cyclohexyl.

Tetrakis[ tri( o-tolyl)phosphine-gold( I) ] methane This work was supported by Deutsche For­ 1.2 g (2.24 mmol) (2-MeC6H4)3PAuCl; 0.17 g schungsgemeinschaft (Leibniz-Programm), by (0.56 mmol) [(MeO)2B]4C; 4.82 g (31.9 mmol) Fonds der Chemischen Industrie, and through the CsF. Yield 0.26 g (23%), m.p. 162 °C (decomp.). donation of chemicals by Hoechst AG, Degussa MS (FD, CH2C12): m/e = 2018.2 (7%, M+). 3IP AG, and Heraeus GmbH.

[1] H. Schmidbaur, Gold Bull. 23, 11 (1990). [8] A. Grohmann, J. Riede, and H. Schmidbaur, Na­ [2] H. Schmidbaur, in Gmelin Handbook of Inorganic ture 345, 140(1990). Chemistry, Organogold Compounds, Springer Ver­ [9] A. Brodbeck and J. Strähle, Acta Crystallogr. A 46, lag, Berlin (1980). C-232 (1990). [3] W. Kreß, Diss., Univ. Erlangen-Nürnberg (1983). [10] E. Zeller, H. Beruda, A. Kolb, P. Bissinger, J. Riede, [4] D. K. Breitinger and W. Kreß, J. Organomet. and H. Schmidbaur, Nature 352, 141 (1991). Chem. 256, 217 (1983); D. K. Breitinger, W. Kreß, [11] F. Scherbaum, A. Grohmann, B. Huber, C. Krüger, and R. Sendelbeck, J. Organomet. Chem. 243, 245 and H. Schmidbaur, Angew. Chem., Int. Ed. Engl. (1983). 27, 1544(1988). [5] D. Grdenic, M. Sikirica, and B. Korpar-Colig, J. [12] F. Scherbaum, A. Grohmann, G. Müller, and H. Organomet. Chem. 153, 1 (1978); D. Grdenic, M. Schmidbaur, Angew. Chem., Int. Ed. Engl. 28, 436 Sikirica, B. Korpar-Colig, and G. Javanovski, J. (1989). Chem. Soc. Chem. Commun. 1974, 646. [13] O. Steigeimann, P. Bissinger, and H. Schmidbaur, [6] E. G. Perevalova, E. I. Smyslova, V. P. Dyadchen- Angew. Chem., Int. Ed. Engl. 29, 1399 (1990). ko, K. I. Grandberg, and A. N. Nesmeyanov, Izv. [14] H. Schmidbaur, B. Brachthäuser, and O. Steigel- Akad. Nauk SSSR, Ser. Khim. 1980, 1455. mann, Angew. Chem., Int. Ed. Engl. 3 0 ,1488 (1991). [7] Y. L. Slovokhotov and Y. L. Struchkov, J. Organo­ [15] H. Schmidbaur, B. Brachthäuser, O. Steigeimann, met. Chem. 277, 143 (1984). and H. Beruda, Chem. Ber., in press. 1724 H. Schmidbaur-O. Steigeimann • The First Complexes of Tetragoldmethane CAu4

[16] P. Pyykkö, Chem. Rev. 88, 563 (1988). [21] H. Schindlbauer, Monatsh. Chem. 96, 1936, 2051 [17] N. Rösch, A. Gerling, D. E. Ellis, and H. Schmid­ (1965). baur, Angew. Chem., Int. Ed. Engl. 28, 1357 (1989). [22] M. A. Bennett and P. A. Longstaff, J. Am. Chem. [18] A. Görling, N. Rösch, D. E. Ellis, and H. Schmid­ Soc. 91,6266(1969). baur, Inorg. Chem. 30, 3986 (1991). [23] H. Fritzsche, U. Hasselrodt, and F. Körte, Chem. [19] P. Pyykkö and Y. Zhao, Angew. Chem., Int. Ed. Ber. 97, 1988(1964). Engl. 30, 604 (1991); Chem. Phys. Lett. 177, 103 [24] K. C. Dash and H. Schmidbaur, Chem. Ber. 106, (1991). 1221 (1973); C. Kowala and J. M. Swan, Aust. J. [20] D. M. P. Mingos, T. Slee, and L. Zhenyang, Chem. Chem. 19, 547(1966). Rev. 90, 383 (1990); see also: D. M. P. Mingos, J. Chem. Soc. Dalton Trans. 1972, 1035.