J. Am. Chem. Soc. 2001, 123, 7469-7470
7469
Reactivity of [MoX(η3-allyl)(CO)2(N-N)] Complexes
with Simple, Nonstabilized Carbanions
Scheme 1
Julio Pe´rez,*, ‡ Luc´ıa Riera,‡ V´ıctor Riera,‡
Santiago Garc´ıa-Granda,§ and Esther Garc´ıa-Rodr´ıguez§
Departamento de Qu´ımica Orga´nica e Inorga´nica/
I.U.Q.O.E.M.
Departamento de Qu´ımica F´ısica y Anal´ıtica
Facultad de Qu´ımica, UniVersidad de OViedo-C.S.I.C.
33071 OViedo, Spain
ReceiVed January 12, 2001
ReVised Manuscript ReceiVed May 2, 2001
Scheme 2
In 1982 Trost reported that [MoCl(η3-allyl)(CO)2(L-L)]
(L-L ) bidentate chelating ligand) complexes react with
stabilized carbanions to yield olefins.1 It remains unknown
whether the nucleophilic attack takes place (a) directly to the allyl
ligand or (b) to the metal, followed by a metal-to-allyl migration.
In general, the site of primary nucleophilic attack in the reactions
of η3-allylmetal complexes is dictated by the nature of the
nucleophile: hard nucleophiles prefer the metal, whereas soft,
stabilized ones attack the allyl group.2
(bipy)](2a) has the geometry shown in Scheme 1, with the methyl
and allyl groups in mutually trans positions.
When methyllithium was used as alkylating reagent, traces of
other unidentified species were obtained together with 2a; thus,
more selective reagents were sought. Methylmagnesium iodide
afforded spectroscopically pure 2a; however, it transformed into
a second, orange complex upon standing in solution at room
temperature or during attempted isolation. The same result was
found using lithium dimethylcuprate, prepared in situ by reaction
of CuI and MeLi,5 as carbanion source. The orange complex was
found to be [MoI(η3-allyl)(CO)2(bipy)].6 Dimethylmagnesium,
selective and halide-free,7 was found to be the reagent of choice.
Slow formation of the chlorocomplex 1a was observed in
dichloromethane solutions of 2a; thus, non-chlorinated solvents
had to be used for workup and NMR studies.
Pseudooctahedral [MoCl(η3-allyl)(CO)2(L-L)] complexes have
been known for more than three decades.3 However, their
reactivity toward simple, nonstabilized carbanions remains so far
unexplored.4 We have initiated the study of this reactivity and
here we report our preliminary results.
The orange complex [MoCl(η3-allyl)(CO)2(bipy)] (bipy ) 2,2′-
bipyridine) (1a)3a-c reacts in THF solution with methylating
carbanionic reagents (vide infra) to give a single blue product 2a
with two C-Ã stretching bands of similar intensity in the IR
spectrum, diagnostic of a cis-M(CO)2 unit, at frequencies lower
1
than the starting chlorocomplex. The H NMR shows the three-
signal pattern of a static, symmetrical, untouched η3-allyl ligand,
and a low-frequency singlet typical of a metal-bound methyl
group. The pattern of four multiplets for the bipy hydrogens,
maintained at low temperature (180 K), reveals the existence of
a mirror plane in the molecule. Hence, [Mo(CH3)(η3-allyl)(CO)2-
To test the generality of our synthetic procedure and the
stability of the alkyls, we prepared similarly the ethyl 2b and
benzyl 2c derivatives, as well as the three analogues with
N-N ) 1,10-phenanthroline (phen) (compounds 3a-c).9 The
alkylations were specific, and neither allylic alkylation nor acyl
products were formed. The same result was found using [MoBr-
(η3-allyl)(CO)2(N-N)] complexes as starting materials.
* Correspondence author: (e-mail) japm@sauron.quimica.uniovi.es.
‡ Departamento de Qu´ımica Orga´nica e Inorga´nica/I.U.Q.O.E.M.
§ Departamento de Qu´ımica F´ısica y Anal´ıtica.
The higher solubility of the complexes with the phen ligand
(1) Trost, B. M.; Lautens, M. J. Am. Chem. Soc. 1982, 104, 5543-5545.
Trost found that the reaction can be made catalytic in metal: molybdenum
carbonyl complexes catalyze the reaction of allylic acetates with stabilized
carbanions, such as the one obtained by deprotonation of diethyl malonate.
For studies of the influence of the ligands on this reaction, see: Trost, B. M.;
Merlic, C. A. J. Am. Chem. Soc. 1990, 112, 9590-9600. Sjo¨gren, M. P. T.;
Frisell, H.; Akermark, B.; Norrby, P.-O.; Eriksson, L.; Vitagliano, A.
Organometallics 1997, 16, 942-950. For the asymmetric version of this
reaction, see: Trost, B. M.; Hachiya, I. J. Am. Chem. Soc. 1998, 120, 1104-
1105.
allowed their 13C NMR spectra to be acquired. The spectral data
(5) Caldarelli, J. L.; Wagner, L. E.; White, P. S; Templeton, J. L. J. Am.
Chem. Soc. 1994, 116, 2878-2888.
(6) See Supporting Information for details.
(7) Actually, the dimethylmagnesium we used, prepared from MeMgI and
1,4-dioxane (see ref 6), contains some halide. A method to prepare truly halide-
free dimethylmagnesium is given in ref 8. However, the extremely hazardous
nature of HgMe2 deterred us from using it.
(2) Consiglio, G.; Waymouth, R. M. Chem. ReV. 1989, 89, 257-276. Trost,
B. M.; Van Vranken, D. L. Chem. ReV. 1996, 96, 395-422. The terms hard
and soft are used here as defined in these reviews.
(3) (a) Hull, C. G.; Stiddard, M. H. B. J. Organomet. Chem. 1967, 9, 519-
525. (b) tom Dieck, H.; Friedel, H. J. Organomet. Chem. 1968, 14, 375-
385. (c) Brisdon, B. J. J. Organomet. Chem. 1977, 125, 225-230. (d) Faller,
J. W.; Haitko, D. A.; Adams, R. D.; Chodosh, D. F. J. Chem. Soc. 1979, 101,
865-876. (e) Baker, P. K. AdV. Organomet. Chem. 1995, 40, 45-115 (review
of halocarbonyl Mo(II) and W(II) compounds, with a section on allyl
complexes).
(4) To our knowledge, the only such reaction reported is that of [MoX-
(η3-C3H5)(CO)2(DAB)] (X) Cl, Br; DAB) N, N′-dialkylethanediimine) with
MeLi or MeMgI, which afforded unidentified decomposition products: Hsieh,
A. T. T.; West, B. O. J. Organomet. Chem. 1976, 112, 285-296. The
possibility of nucleophilic attack to the metal was suggested: Curtis, M. D.;
Fotinos, N. A. J. Organomet. Chem. 1984, 272, 43-54. In the study of the
Mo-catalyzed allylic alkylation, Trost found results suggesting attack to the
metal by the less bulky malonate anions: Trost, B. M.; Lautens, M. J. Am.
Chem. Soc. 1987, 109, 1469-1478. For another example of at-the-metal attack
by a soft nucleophile, see: Faller, J. W.; Linebarrier, D. Organometallics 1988,
7, 1670-1672.
(8) Cotton, F. A.; Wiensinger, K. J. Inorg. Chem. 1990, 29, 2594-2599.
(9) 3a: To a solution of 1b (0.10 g, 0.24 mmol) in THF (20 mL) was
added MgMe2 (0.26 mmol, 0.9 mL of a 0.29 M solution in Et2O). After the
mixture was stirred for 15 min, the solvent was removed under vacuum. The
residue was extracted with toluene (4 × 10 mL) and filtered through Celite
to yield a blue solution. The solvent was then removed under vacuum to afford
0.063 g of 3a as a blue solid. IR (CH2Cl2): 1922, 1830 (νCO). 1H NMR (CD2-
Cl2): 8.96, 8.42, 7.97, and 7.71 [m, 2H each, phen], 3.05 [m, 1H, CH η3-
C3H5], 2.77 [d (6.4), 2H, Hsyn], 1.49 [d (9.3), 2H, Hanti], -0.69 [s, 3H Mo-
CH3]. 13C{1H} NMR (CD2Cl2): 235.60 [CO], 151.69, 144.36, 136.02, 130.49,
127.45, and 124.29 [phen], 73.56 [ C2 η3-C3H5], 51.82 [ C1 and C3 η3-C3H5],
12.51 [Mo-CH3]. 3b: A similar procedure using 1b (0.10 g, 0.24 mmol)
and MgEt2 (1.13 mL of a 0.21 M solution in ether, 0.24 mmol) gave the ethyl
1
complex 3b (0.057 g) as a violet solid. IR (CH2Cl2): 1919, 1827 (νCO). H
NMR (CD2Cl2): 1.52 [t (7.9), 3H, CH3], -0.29 [q (7.8), 2H, CH2]. 13C{1H}
NMR (CD2Cl2): 30.67 [CH3], 19.82 [CH2]. (3c). From 1b (0.10 g, 0.24 mmol)
and MgBz2 (0.70 mL of a 0.34 M solution in THF, 0.24 mmol) 3c (0.95 g)
1
was obtained as a dark blue solid. IR (CH2Cl2): 1925, 1827 (νCO). H NMR
(C6D6): 6.14 and 5.14 [m, C6H5], 3.00 [s, 2H, CH2]. 13C{1H} NMR (CD2-
Cl2): 152.07, 151.62, 144.27, 135.59, 130.19, 127.30, 126.26, 124.27, 124.15,
and 118.57 [m, C6H5 and phen], 36.65 [CH2].
10.1021/ja015538z CCC: $20.00 © 2001 American Chemical Society
Published on Web 07/07/2001