Photo-induced tautomerisation of methyltrioxorhenium(VII): the intermediate in olefin metathesis?

Abstract

Matrix-isolated [CH₃ReO₃] tautomerises to [H₂C[double bond, length half m-dash]Re(O)₂OH] under the influence of UV light; the carbene has been characterised in its normal and ²H- and ¹³C-enriched isotopic forms by its IR spectrum with results well replicated by quantum chemical calculations.

---

Since methyltrioxorhenium(VII), [CH₃ReO₃] 1, was first described, its rôle in promoting and catalysing numerous organic reactions has been explored in detail.^1,2 In fact, 1 is probably the most widely active organometallic catalyst reported to date. This activity encompasses two general areas: (i) oxidation reactions (including olefin epoxidation, Baeyer–Villiger and aromatic oxidation);^2–5 and (ii) olefin isomerisation and metathesis.⁶ The peroxo derivatives of 1 active in many reactions of type (i) have been isolated and characterised both structurally and spectroscopically.⁷ By contrast, the species responsible for catalytic activity of type (ii) have eluded direct detection, although a tautomer of the form [H₂C[double bond, length half m-dash]Re(O)₂OH] 2, has long been presumed to be the active form of 1.^2,8 There are numerous reports of rhenium–carbon multiple bonds in the literature generally, but not exclusively, involving rhenium in formal oxidation states <+7.^8,9

Photolysis of 1 in solution appears to result in homolysis of the Re–C bond.¹⁰ Our studies reveal, however, that the matrix-isolated molecule exhibits altogether different behaviour, initially tautomerising to the previously unknown carbene derivative 2 under the influence of UV light at wavelengths near 254 nm (Scheme 1). Tautomer 2 is also photolabile, broad-band UV–VIS irradiation (λ = 200–800 nm) causing it to decay to a product containing an Re–CO fragment, 3, possibly [H₂Re(O)(OH)CO].

Scheme 1

Exposure of 1 isolated in an Ar matrix at 14 K to UV radiation with λ = ca. 254 nm for several minutes results in the decay of the IR absorptions due to 1 and the simultaneous appearance and growth of new absorptions apparently due to a single product 2 (Fig. 1). Irradiation of the matrix with broad-band UV–VIS light (λ = 200–800 nm) was observed to cause the evolution from 2 of at least one further product 3 which could not be conclusively identified by its IR spectrum. The IR bands identified on the evidence of their growth-decay patterns enable 2 to be characterised as [H₂C[double bond, length half m-dash]Re(O)₂OH], the conclusions being underpinned (i) by the observed effects of ²H- and ¹³C-enrichment of the products derived from the species [CD₃ReO₃] 1-d₃, and [¹³CH₃ReO₃] 1-¹³c, (ii) by parallels with the spectra of related carbene and Re[double bond, length half m-dash]O derivatives, e.g. CoCH₂¹¹ and ReO₂F₃,¹² and (iii) by comparisons with the results of Density Functional Theory (DFT) calculations. (Calculations were carried out in Gaussian 98¹³ with geometries optimised at the BPW91/LANL2DZ level of theory; standard 6-31G(d,p) basis sets were used for C, O and H, whilst the Re basis set was augmented with an additional f-type polarisation function.) Prominent among its IR absorptions were those at 3650.0, 992.2, 963.3 and 668.4 cm⁻¹ which are identifiable by their frequencies, intensities and responses to ²H- and ¹³C-enrichment with the modes ν(O–H), ν_s(Re[double bond, length half m-dash]O), ν_as(Re[double bond, length half m-dash]O) and ν(Re–OH), respectively. The presence of the Re[double bond, length half m-dash]CH₂ unit is signalled by bands at 3079.6, 2985.8, 1320.9, 778.8, 756.4 and 627.9 cm⁻¹ which we associate with the modes detailed in Table 1; each assignment is attested by analogy with the corresponding mode of CoCH₂¹¹ and by the ²H and ¹³C isotopic shifts, although the description of the motion is sometimes less than exact. The spectrum is well simulated by a scaled force field computed for 2 on the basis of DFT calculations (Table 1), the 36 frequencies measured for 2, 2-d₃ and 2-¹³c being matched with an r.m.s. deviation of only 1.66%. It is also evident that mixing of the ν(Re[double bond, length half m-dash]C), ρ(CH₂) and δ(OH) motions complicates the interpretation of the spectrum in the region 600–800 cm⁻¹, preventing the identification of any one feature with the ν(Re[double bond, length half m-dash]C) mode. The relative intensities of the bands due to the well defined modes ν_as(Re[double bond, length half m-dash]O) and ν_s(Re[double bond, length half m-dash]O) imply an O[double bond, length half m-dash]Re[double bond, length half m-dash]O angle in the order of 116°, in good agreement with the optimum geometry computed for 2.

Fig. 1 (a) IR spectrum of [CH₃ReO₃] 1 isolated in an argon matrix at 14 K; (b) IR spectrum showing the effect of irradiation at λ = 254 nm for 15 min (↓ indicates a feature associated with [H₂C[double bond, length half m-dash]Re(O)₂OH] 2); and (c) the spectrum of 2 based on the results of DFT calculations.

Table 1 Observed and calculated fundamental vibrational frequencies for 2, 2-¹³c, and 2-d₃ under C_s symmetry. Observed and calculated intensities are in parentheses^a

H2^12CReO2(OH)		H2^13CReO2(OH)		D2^12CReO2(OD)
Obs.	Calc.^b	Obs.	Calc.^b	Obs.	Calc.^b	Description of mode
a Frequencies in cm^−1; all intensities normalised to that of the most intense band set equal to 100 (in parentheses).b Calculated frequency scaled by a factor of 0.9740. The r.m.s. deviation between observed and scaled calculated frequencies is 1.66%.c Indicates a feature too weak to be observed.
3650.0 (100)	3669.6 (87)	3650.4 (100)	3669.6 (87)	2694.2 (73)	2672.3 (63)	ν(O–H)
2985.8 (3)	2994.4 (5)	2980.2 (4)	2989.0 (5)	2195.1 (5)	2171.4 (6)	νs(CH2)
1320.9 (2)	1295.0 (5)	1311.8 (2)	1286.7 (5)	1009.9 (2)	1018.8 (6)	δ(CH2)
992.2 (25)	988.4 (30)	991.9 (26)	988.3 (30)	992.2 (31)	987.1 (42)	νs(ReO2)
799.2 (2)	815.4 (8)	798.7 (3)	812.4 (9)	622.6 (3)	631.0 (3)  [upper bond 1 end]
778.8 (34)	792.0 (10)	768.3 (53)	770.4 (14)	578.9 (43)	556.8 (81)	δ(O–H) + ρ(CH2)
756.4 (30)	741.9 (73)	739.6 (19)	736.5 (67)	685.9 (10)	698.2 (6)	+ ν(Re[double bond, length half m-dash]C) + ν(Re–OH)
668.4 (83)	664.2 (100)	668.8 (87)	664.2 (100)	660.0 (69)	664.7 (75) [lower bond 1 end]
^c	286.4 (2)	^c	286.4 (2)	^c	286.0 (2)	ReO2 wag
^c	255.6 (1)	^c	255.6 (1)	^c	252.7 (2)	ReO2 scissor
^c	239.8 (3)	^c	239.8 (3)	^c	217.6 (3)	δ(CReOH)
3079.6 (3)	3091.9 (0.6)	3068.4 (1)	3079.5 (0.6)	2315.2 (1)	2299.0 (0.3)	νas(CH2)
963.3 (64)	967.2 (81)	963.4 (80)	967.1 (81)	961.8 (100)	965.0 (100)	νas(ReO2)
627.9 (3)	644.2 (1)	625.7 (5)	639.3 (2)	500.2 (2)	504.1 (0.3)	CH2 scissor
^c	526.4 (0.2)	^c	525.8 (0.2)	^c	386.1 (0.01)	CH2 wag
321.2 (33)	315.9 (42)	320.7 (38)	315.2 (43)	239.3 (19)	253.4 (6)	δ(O–H)
^c	266.8 (15)	^c	266.8 (15)	^c	237.3 (15)	δ(CReO2)
^c	233.2 (0.03)	^c	233.2 (0.03)	^c	208.5 (16)	d(CReO2)

---

Similar experiments with [CH₂DReO₃] 1-d₁ give rise not only to [H₂C[double bond, length half m-dash]Re(O)₂OD], but also to the isotopomer [H(D)C[double bond, length half m-dash]Re(O)₂OH] 2-d₁. Here the ν(C–H) fundamental is isolated from all other modes in the molecule and so gives access, in principle, to relatively precise estimates of the dimensions of the CH₂ unit.¹⁴ The measured value of ν^is(C–H) (3035.4 cm⁻¹), taken together with ν(¹²C–H), ν(¹³C–H) and ν(¹²C–D) data for the other isotopomers, affords values that tally with the results of the DFT optimisations given in parentheses: r₀ = 1.088 Å (r_e = 1.099 Å), ∠HCH = 119 ± 4° (∠_eHCH = 115.8°).

Although 2 was formed almost exclusively when matrix-isolated 1 was photolysed at wavelengths near 254 nm, exposure to broad-band UV–VIS light gave rise to a secondary change. The sole detectable product 3 formed from 2 under these conditions, but always in the presence of an abundance of 1 and 2, could be identified by a single IR band at 2051.4 cm⁻¹. Assignment to the ν(C–O) mode of an Re–CO moiety is strongly urged by a minimal change of frequency when 3 is formed from 1-d₃ or 1-d₁ but by a shift to 2003.8 cm⁻¹ when it is formed from 1-¹³c. The circumstances preclude positive identification, but a possible candidate for 3 is the novel rhenium(V) compound [H₂Re(CO)(O)OH] formed by photoisomerisation of [H₂C[double bond, length half m-dash]Re(O)₂OH] in a change that would parallel the conversion of [H₂COSi] to [H₂Si∶CO].¹⁵ DFT calculations provide some support for [H₂Re(CO)(O)OH], finding a potential energy minimum 126 kJ mol⁻¹ above that of [H₂C[double bond, length half m-dash]Re(O)₂OH] with a structure approximating to a square-based pyramid having the unique oxide ligand at the apex, and a calculated ν(C–O) frequency of 2040 cm⁻¹ (¹²C/¹³C shift = 44.5 cm⁻¹).

[CH₃ReO₃] is active in olefin metathesis only when activated by a co-catalyst (S₄N₄/AlCl₃), or when supported on silica or alumina.⁸ DFT calculations indicate that 2 lies ca. 89 kJ mol⁻¹ higher in energy than 1, and hence is inaccessible under normal thermal conditions. Model calculations on [CH₃ReO₂{(η²-OSiH₂) ₂O}] 4, the product formed by condensation of [CH₃ReO₃] with disilanol ([H₂Si(OH)]₂O), show the tautomeric H-atom transfer to occur preferentially to an Re–O–Si bridging oxygen atom rather than to an Re[double bond, length half m-dash]O unit. The resulting carbene species [H₂C[double bond, length half m-dash]ReO₂{(η²-OSiH₂ )(O)(SiH₂OH)}] 5 is effectively stabilised by ca. 48 kJ mol⁻¹ relative to complex 4. This alternative H-atom transfer to the Re–O–Si bridge would seem to be a more realistic mechanism for carbene formation on supported [CH₃ReO₃].

References

1. W. A. Herrmann, J. Organomet. Chem., 1995, 500, 149.
2. C. C. Romão, F. E. Kühn and W. A. Herrmann, Chem. Rev., 1997, 97, 3197.
3. W. A. Herrmann, R. W. Fischer and D. W. Marz, Angew. Chem., Int. Ed. Engl., 1991, 30, 1638.
4. W. A. Herrmann, R. W. Fischer and J. D. G. Correira, J. Mol. Catal., 1994, 94, 213.
5. R. A. Sheldon, Top. Curr. Chem., 1993, 164, 21.
6. W. A. Herrmann, W. Wagner, U. N. Flessner, U. Volkhardt and H. Komber, Angew. Chem., Int. Ed. Engl., 1991, 30, 1636.
7. A. M. Al-Ajlouni and J. H. Espenson, J. Org. Chem., 1996, 61, 3969; M. M. Abu-Omar and J. H. Espenson, Organometallics, 1996, 15, 3543.
8. D. M. Hoffman, in Comprehensive Organometallic Chemistry II, ed. C. P. Casey, Pergamon, Oxford, 1995, vol. 6, p. 231..
9. J. M. O’Connor, in Comprehensive Organometallic Chemistry II, ed. C. P. Casey, Pergamon, Oxford, 1995, vol. 6, p. 200..
10. H. Kunkely, T. Türk, C. Teixeira, C. de Meric de Bellefon, W. A. Herrmann and A. Vogler, Organometallics, 1991, 10, 2090.
11. W. E. Billups, S.-C. Chang, R. H. Hauge and J. L. Margrave, J. Am. Chem. Soc., 1995, 117, 1387.
12. I. R. Beattie, R. A. Crocombe and J. S. Ogden, J. Chem. Soc., Dalton Trans., 1977, 1481.
13. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakrewski, J. A. Montgomery, R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz, B. B. Stefanov, G. Lui, A. Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe, P. M. W. Gill, B. G. Johnson, W. Chen, M. W. Wong, J. L. Andres, M. Head-Gordon, E. S. Replogle and J. A. Pople, Gaussian 98, Revision A.3, Gaussian Inc., Pittsburgh PA, 1998..
14. D. C. McKean, Chem. Soc. Rev., 1978, 7, 399; Croat. Chem. Acta, 1988, 61, 447..
15. G. Maier, H. P. Reisenauer and H. Egenolf, Organometallics, 1999, 18, 2155.

---