2616
Organometallics 1995, 14,2616-2617
Ruthenium-Mediated Hydrogenation of Carbon Dioxide by NaBH4 via the Formation of the Formate Complex C5MesRu(PCys)(q2-OCHO) Chae S. Yi* and Nianhong Liu Department of Chemistry, Marquette University, Milwaukee, Wisconsin 53233 Received January 13, 1995@ Summary: Reaction of C5Me&u(PCydCl (1) with 1-2 a t m of carbon dioxide in the presence of NaBH4 initially gave the new formate complex C&fe$u(PCyd(+-OCHO) (2). The complex 2 in the reaction mixture was gradually converted into formic acid and the previously known C5Me&u(PCyd(CO)H (3). Deuterium labeling experiments suggest that the acid proton of formic acid came from the solvent. A possible mechanism of the formation of 2 and subsequent formation of formic acid is discussed. Remarkable progress has been made in the transformation of carbon dioxide employing organometallic catalysts in recent years.lP2 A few notable recent examples include Nicholas’s rhodium-catalyzed hydrogenation of carbon dioxide,2aKubiak’s electrocatalytic reduction of carbon dioxide in a nickel cluster,2b and Noyori’s (PMe3)4RuH2-catalyzedformation of formic acid in supercritical C02 medium.2c To get some understanding of the mechanisms involving the reactions of transition-metal complexes with carbon dioxide, we have been investigating the reactions of ruthenium complexes with carbon dioxide, and herein we report a mild hydrogenation reaction of carbon dioxide using the organoruthenium complex CgMegRu(PCy3)Cl(lI3 and NaBH4. Our initial plan was t o prepare a coordinatively unsaturated ruthenium hydride species of the type CgMegRu(PCya)(H)and use it as a precursor for the carbon dioxide reaction. Several preliminary attempts to generate CgMegRu(PCy3)(H)from the reactions of 1 with @Abstractpublished in Advance ACS Abstracts, May 1, 1995. (1)For recent reviews, see: (a) Behr, A. Carbon Dioxide Activation by Metal Complexes; VCH: New York, 1988. (b) Darensbourg, D. J.; Kudaroski, R. A. Adv. Organomet. Chem. 1983,22,129. (c) Sneeden, R. P. A. In Comprehensive Organometallic Chemistry;Wilkinson, G., Stone, F. G. A., Eds.; Pergamon Press: New York, 1982;Vol. 8. (d) Braunstein, P.; Matt, D.; Nobel, D. Chem. Rev. 1988,88, 747. (e) Catalytic Activation of Carbon Dioxide; Ayers, W. M., Ed.; ACS Symposium Series 363;American Chemical Society: Washington, DC, 1988. (0 Denise, B.; Sneeden, R. P. A. CHEMTECH 1982,12,108. For recent selected transition-metal carbon dioxide complexes, see: (g) Calabrese, J. C.; Heerskovitz, T.; Kinney, J . B. J . Am. Chem. Soc. 1983, 105,5914. (h) Alvarez, R.; Carmona, E.; Marin, J. M.; Poveda, M. L.; Gutierrez-Puebla, E.; Monge, A. J . A m . Chem. Soc. 1988,108, 2286. (i) Fu, P.-F.; Khan, M. A.; Nicholas, K. M. J . A m . Chem. SOC.1992, 114,6579.(j) Komiya, S.;Akita, M.; Kasuga, N.; Hirano, M.; Fukuoka, A. J . Chem. SOC., Chem. Commun. 1994,1115. (k) Fu, P.-F.; FazlurRahman, A. K.; Nicholas, K. M. Organometallics 1994,13,413. (1) Sakamoto, M.; Shimizu, I.; Yamamoto, A. Organometallics 1994,13, 407. (m) Pinkes, J. R.; Steffey, B. D.; Vites, J. C.; Cutler, A. R. Organometallics 1994,13,21. (n) Gibson, D. H.; Ye, M.; Richardson, J. F.; Mashuta, M. S. Organometallics 1994,13,4559. (2)(a)Tsai, J.-C.; Nicholas, K. M. J . A m . Chem. Soc. 1992,114,5117. (b) Morgenstern, D.A.; Wittrig, R. E.; Fanwick, P. E.; Kubiak, C. P. J . A m . Chem. SOC.1993,115,6470.(c) Jessop, P. G.; Ikariya, T.; Noyori, R. Nature 1994,368, 231. (d) Jessop, P. G.; Hsiao, Y.; Ikariya, T.; Noyori, R. J . A m . Chem. SOC.1994,116,8851. (3)(a) Campion, B. K.; Heyn, R. H.; Tilley, T. D. J . Chem. SOC., Chem. Commun. 1988,278. (b) Campion, B. K.; Heyn, R. H.; Tilley, T. D. J . A m . Chem. SOC.1988,110, 7558. (c) Luo, L.; Nolan, S.P.; Fagan, P. J. Organometallics 1993,12,4305.
hydride reagents, however, gave disparate results. Reaction of 1 with LiAlH4 exclusively gave the previously known trihydride complex CgMegRu(PCy3)(H)3, whereas reaction with LiBH4 gave both the borohydride species C ~ M ~ ~ R U ( P C Y ~ ) ( , Uand - H )CsMegRu(PCy3)~BH~ (H)s.~No apparent reaction was observed in the reaction of 1 with NaBH4, and none of these hydride complexes were reactive toward carbon dioxide under our reaction conditions. Surprisingly, however, when 1 was treated with carbon dioxide in the presence of NaBH4, a new species was observed by NMR. In a sealed NMR tube, a mixture of carbon dioxide (1-2 atm), 1 (10 mg, 0.0368 mmol), and 5 equiv of NaBH4 (7 mg, g, 0.184 mmol) in THF-ds was vigorously shaken a t room temperature for 1-2 h. The deep blue solution turned red-brown, and a set of new resonances appeared which was subsequently characterized as the new formate complex CgMegRu(PCy3)(v2-OCHO) The ‘H NMR of 2 exhib-
ited a downfield-shifted formate hydrogen signal a t 6 8.50, which became a doublet when 13C02was employed (JCH= 209.6 Hz). The formate carbon resonance a t 6 164.1 (d, 3 J p = ~ 3.7 Hz) and the characteristic formate IR stretching frequencies (voco(asym) = 1618 and voco(sym) = 1452 cm-l) were also consistent with the y2-formategeometry.6 On a preparative scale, complex 2 was isolated as a brown-red solid in approximately 80%yield, but we were not able to obtain an analytically pure complex due t o its thermal instability. Several attempts to synthesize 2 independently from the reaction of 1with NaOzCH were unsuccessful. The initially formed 2 in a sealed NMR tube in THFd8 was further converted to a mixture of the previously known carbonyl hydride C ~ M ~ ~ R U ( P C ~ ~ )(3)7 ( C and O)H formic acid in approximately a 1:2 ratio. In the IH NMR, a set of new peaks at 6 11.34 (HC02H) and 6 7.90 (HC02H) due to formic acid appeared at the expense of (4)(a) Suzuki, H.; Lee, D. H.; Oshima, N.; Moro-oka, Y. Organometallics 1987, 6, 1569. (b) Lee, D. H.; Suzuki, H.; Moro-oka, Y. J . Organomet. Chem. 1987,330,C20. ( 5 )For 2: ‘H NMR (C&, 300 MHz) 6 8.50(s, OZCH),1.76 (s, CgMe5), 1-2 (m, PCys); 13C{lH}NMR (C6D6, 75 MHz) 6 164.1(d, 3 J ~ = p 3.7 Hz, OzCH), 88.2 (C5Me5), 31.2,28.2,27.4,and 26.4 (PCys), 11.7 ( C a e 5 ) ;31PNMR (C&k3,121.6 MHZ) 6 54.6(PCY,); IR (C&) 1618 (S, vocdasym)), 1452 (s, vocdsym)) cm-’. (6) (a) Nakamoto, K. Infrared and Raman Spectra of Inorganic and Coordination Compounds, 4th ed.; Wiley: New York, 1986. (b) Silverstein, R. M.; Bassler, G. C.; Morrill, T. C. Spectrometric Identification of Organic Compounds, 5th ed.; Wiley: New York, 1991.
0276-733319512314-2616$09.00/00 1995 American Chemical Society
Communications
Organometallics, Vol. 14, No. 6,1995 2617
nium and the chloride, making the chloride ligand more the resonances of 2. The formate hydrogen peak at 6 susceptible to the NaBH4 reduction. Further intramo7.90 became a doublet (JCH= 210.3 Hz) when 13C02 lecular insertion of carbon dioxide would give the for(99%, Cambridge Isotopes) was employed in the reacmate complex 2. Carbon dioxide insertion into the trantion, indicating its attachment t o formic acid. Eventusition-metal-hydride bond has been well documented.la,b ally after 5 h, all formic acid was decomposed to Also, a reversible carbon dioxide insertion into RuNaOzCH, 3, and other unidentified species. In a (PR3)dHz (R = Ph, Me) has been previously reported.'l preparative reaction, complex 3 was isolated as a pale The detailed reaction mechanism of the conversion white solid in 85% yield based on 1. The choice of from 2 to 3 and formic acid is not yet clearly understood. solvent appeared to be important to the success of the In a preliminary experiment, initially formed 2 and reaction; similar reactions in MeOH led to the exclusive formation of the trihydride complex C ~ M ~ S R U ( P C ~ ~ ) ( Hformic ) ~ , acid were extensively deuterated at the formate hydrogen when NaBD4 in THF was employed in the while no reaction was observed in Et20 or CHzClz. reaction, but the hydride of 3 contained approximately 50% deuterium as estimated by NMR. Similarly, when the reaction was carried out with NaB& in THF-ds, the acid proton of formic acid contained nearly 70%deuterium. These results suggest that the possible source of the acid proton of formic acid is from the solvent and not from NaBH4 and that the substantial deuterium scrambling may have occurred during the subsequent formation of formic acid and 3. Wilkinson previously proposed a mechanism involving proton transfer from a coordinated THF to a coordinated hydroxy ligand.12 In summary, the formation of formic acid from the The apparent lack of reactions between 1 and NaOzhydrogenation of carbon dioxide using the coordinatively CH indicates that the possible formation of complex 2 unsaturated ruthenium complex 1 and NaBH4 has been from the reaction of 1 with a preformed formate ion is described. Preliminary studies indicated that the prenot likely.8 The fact that complex 1is unreactive toward coordination of carbon dioxide to the complex 1 proNaBH4 in the absence of carbon dioxide suggests that moted the hydrogenolysis of chloride ligand to give the carbon dioxide was precoordinated to 1 before the formate complex 2, and the subsequent decomposition hydrogenolysis of chloride? These results are consistent of 2 led to the formation of formic acid and 3. Studies with a reversible precoordination of carbon dioxide to 1 on the scope and the detailed reaction mechanism to produce the 18 e- intermediate 4 and subsequent involving the complex 1 are currently being pursued. reduction by NaBH4 to give the more stable formate complex 2.1° Although carbon dioxide is a labile ligand Acknowledgment. Financial support from Marand, therefore, the concentration of the intermediate 4 quette University is gratefully acknowledged. should be quite low at relatively low pressures of carbon dioxide, the precoordination of carbon dioxide to 1would OM950025K reduce the n-bonding interactions between the ruthe(7) Complex 3 was previously prepared from the reaction of 1 with CO followed bv NaBHd reduction: Arlieuie. T.: Border. C.: Chaudret. B.; Devillers, i.;Poilblanc, R. Organomgallfcs 1989,8, 1308. Spectral data for 3: 'H NMR (C&6,300 MHZ) 6 1.98 (8,CsMes), 1-2 (m, PCY~), -11.90 (d, J p " = 33.8 Hz, Ru-H); 13C11H} NMR (CsDs, 75 MHz) 6 210.4 (d, JCP= 18.3Hz, CO), 95.1 (CsMes),38.3,30.8, and 28.1 (PCyd, 12.1 (Cae5); 31PNMR (&De, 121.6 MHz) 6 75.0 (Jcp = 18.3 Hz, PCy3); IR (&He) 1892 (s) cm-l; FAB MS M+ at mlz 545. (8) The reaction of NaBH4 with C02 (5 atm) in THF-ds did not produce any detectable amount of NaOzCH even in the presence of Alc13. (9) Reference 3a described the reversible formation of the adduct C&MesRu(PCy3XCHz=CHz)(Cl)from the reaction of 1 with CH2-CH2. In support of the mechanism, we also found that the reaction of 1with CH2-CH2 in the presence of NaBH4 gave CsMesRu(PCy3)(CHz-CH2)(HI, which may have formed in a fashion similar to that in 2.
(10) Another possible mechanism involves NaBH4 reduction of 1 in the first step to generate the reactive CsMesRu(PCy3)(H),which then inserts carbon dioxide to give 2. Although we cannot rule out this mechanism, the apparent lack of reactivity of NaBH4 toward 1 in the absence of carbon dioxide and the general tendency of forming a stable and unreactive CsMesRu(PCys)(H)asuggest that this route is unlikely. Also, as a reviewer pointed out, carbon dioxide could react first with NaBH4 to form a formate ion, which would then react with 1 to form 2. The facts that complex 1 did not react with NaOzCH and that we did not observe any evidence of reaction between carbon dioxide and NaBH4 in the absence of 1suggest that this route also seems unlikely. (11)(a) Komiya, S.;Yamamoto, A. J. Orgunomet. Chem. 1972, 46, C58. (b) Kolomnikov, I. S.; Gusev, A. I.; Aleksandrov, G . G.; Lobeeva, T. S.; Struchkov, Y. T.; Vol'pin, M. E. J . Organomet. Chem. 1973,59, 349. (12) Chaudret, B. N.; Cole-Hamilton, D. J.;Nohr, R. S.; Wilkinson, G. J. Chem. SOC.,Dalton Trans. 1977, 1546.
