Ruthenium alkoxycarbene complexes from an acetal function by C-O bond cleavage and alcohol elimination

Article abstract of DOI:10.1021/om9603171

Heating solutions of acetal complexes {CpRu-(CH₃CN)₂[η¹-P-2-(Ph ₂P)[CH(OR)₂]C₆H₄]}OTf (3; R = Me, Et) at 60-95 °C results in loss of one CH₃CN ligand and the alcohol R

Full text of DOI:10.1021/om9603171

2860
Organometallics 1996, 15, 2860-2862
Ru th en iu m Alk oxyca r ben e Com p lexes fr om a n Aceta l
F u n ction by C-O Bon d Clea va ge a n d Alcoh ol
Elim in a tion
D. B. Grotjahn* and H. C. Lo
Department of Chemistry and Biochemistry, Box 871604, Arizona State University,
Tempe, Arizona 85287-1604
Received April 26, 1996X
Summary: Heating solutions of acetal complexes {CpRu-
(CH3CN)2[η1-P-2-(Ph2P)[CH(OR)2]C6H4]}OTf (3; R )
Me, Et) at 60-95 °C results in loss of one CH3CN ligand
and the alcohol ROH, with concomitant formation of
carbene complexes CpRu(CH3CN)[η2-C,P-2-(Ph2P)[C(OR)]-
C6H4]}OTf (4) in high yield. Kinetic and reactivity
studies suggest that the rate-determining step of the
conversion of 3 to 4 is oxidative addition of an acetal
C-O bond to the ruthenium center, which occurs under
neutral, mild conditions.
plexes 4. In preliminary studies, the carbene complex
4a shows intriguing ability to isomerize allylic alcohols
to saturated aldehydes.
On the basis of the known stability of CpRu alkoxy-
carbene complexes to boiling alcohols,9 the conversion
of 3 to 4 (Scheme 1) was chosen for initial study.
Benzaldehyde acetals with a 2-diphenylphosphino sub-
stituent (1) were synthesized by adapting known meth-
ods.10 The ruthenium component 2 was prepared in
82% yield by a modification of the published method for
the PF6- salt.11 Addition of 1 to 2 in CDCl3 resulted in
the immediate formation of 3,12,13 which could be
isolated in g90% yield but was usually used directly in
subsequent reactions. Heating a solution of 3 in CDCl3
at 60 °C for 3 h (3a ) or 1 d (3b) led to carbene complex
4,14 CH3CN, and ROH, all in g90% yield as determined
by NMR integration; 4 was isolated as air-stable red
solid in g78% yield after chromatography over SiO2
using CH2Cl2-CH3CN mixtures and recrystallization
from CH2Cl2-Et2O. The formation of a carbene ligand
in 4 was indicated by a downfield doublet in 13C NMR
spectra:15 for 4a and 4b, δ 300.25 (d, J ) 7.6 Hz) and
297.34 ppm (d, J ) 7.6 Hz), respectively. Other spectral
changes accompanying the transformation of 3a to 4a
Metal-carbene complexes are exceedingly versatile
stoichiometric reagents in organic synthesis1 and highly
active catalysts for alkene metathesis.2 A common
preparation of late transition metal carbene complexes
relies on a combination of strongly basic, nucleophilic,
and electrophilic reagents with a metal carbonyl1,3 or
on the action of metal complexes on reactive groups such
as cyclopropenes2,4 or diazo compounds,5 both of which
are of limited accessibility. Because acetals are stable
compounds,6 easily made from widely available carbonyl
compounds, we examined a new reaction for aldehyde
acetals, summarized in eq 1. This metal-induced net
(7) For organic routes to free alkoxycarbenes, see: Heydt, H.; Regitz,
M. Hydroxy-, Organooxy-, Silyloxy-carbene. In Houben-Weyl Methoden
der Organischen Chemie; Thieme: Stuttgart, Germany, 1989; Band
E19b, pp 1628-1682.
R-elimination of an alcohol is a new route to an alkoxy-
carbene complex.7 The closest precedents might be net
R-elimination of H2 from ethers (double C-H acti-
vation)8a-d or net elimination of Me2NH from an aminal
on an osmium cluster,8e in mechanistically uncharac-
terized processes. Here, we report that eq 1 has been
realized in what has the characteristics of a C-O bond
activation process, leading to ruthenium carbene com-
(8) (a) Boutry, O.; Gutie´rrez, E.; Monge, A.; Nicasio, M. C.; Pe´rez,
P. J .; Carmona, E. J . Am. Chem. Soc. 1992, 114, 7288-7290. Gutie´rrez,
E.; Monge, A.; Nicasio, M. C.; Poveda, M. L.; Carmona, E. J . Am. Chem.
Soc. 1994, 116, 791-792. (b) Luecke, H. F.; Arndtsen, B. A.; Burger,
P.; Bergman, R. G. J . Am. Chem. Soc. 1996, 118, 2517-2518. (c)
Werner, H.; Weber, B.; Nu¨rnberg, O.; Wolf, J . Angew. Chem., Int. Ed.
Engl. 1992, 31, 1025-1027. (d) Li, Z.-W.; Taube, H. J . Am. Chem. Soc.
1994, 116, 11584-11585. (e) Aminals to aminocarbene ligands on
clusters: Adams, R. D. Chem. Rev. 1989, 89, 1703-1712.
(9) Review: Bruce, M. I.; Swincer, A. G. Adv. Organomet. Chem.
1983, 22, 59-128, especially pp 69-73.
X Abstract published in Advance ACS Abstracts, J une 1, 1996.
(1) (a) Do¨tz, K. H. Angew. Chem., Int. Ed. Engl. 1984, 23, 587-608.
(b) Transition Metal Carbene Complexes; Seyferth, D., Ed.; VCH:
Weinheim, Germany, 1983. (c) Wulff, W. D. In Comprehensive Organic
Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon: New York, 1991;
Vol. 5, pp 1065-1113. (d) Wulff, W. D. In Advances in Metal-Organic
Chemistry; Liebeskind, L. S., Ed.; J AI Press: Greenwich, CT, 1989;
Vol. 1, pp 209-393.
(10) Hoots, J . E.; Rauchfuss, T. B.; Wroblenski, D. A. Inorg. Synth.
1982, 21, 175-179.
(11) (a) Gill, T. P.; Mann, K. R. Organometallics 1982, 1, 485-488.
(b) In the photolysis of [CpRu(benzene)]+TfO- in CH3CN used to make
2,11a,12 it was found that careful monitoring of the reaction was required
to minimize formation of Ru(CH3CN)62+(TfO-)2.11c (c) Rapaport, I.;
Helm, L.; Merbach, A. E.; Bernhard, P.; Ludi, A. Inorg. Chem. 1988,
27, 873-879.
(2) Wu, Z.; Nguyen, S. T.; Grubbs, R. H.; Ziller, J . W. J . Am. Chem.
Soc. 1995, 117, 5503-5511 and references therein.
(12) Experimental procedures and spectral data not mentioned in
the text or footnotes, derivations of rate laws, and kinetic traces appear
as Supporting Information. All compounds were characterized by 1H,
13C, 31P{1H} NMR, IR and (with the exception of 3) elemental analysis.
(13) Partial data12 for 3a : 31P{1H} NMR (161.9 MHz, CDCl3) δ 46.30.
Data for 3b: 1H NMR (400 MHz, CDCl3) δ 5.63 (s, 1H), 4.33 (s, 5H,
(3) Imwinkelried, R.; Hegedus, L. S. Organometallics 1988, 7, 702-
706. Dvora´k, D. Organometallics 1995, 14, 570-573.
(4) Semmelhack, M. F.; Ho, S.; Cohen, D.; Steigerwald, M.; Lee, M.
C.; Lee, G.; Gilbert, A. M.; Wulff, W. D.; Ball, R. G. J . Am. Chem. Soc.
1994, 116, 7108-7122.
Cp), 3.27 [quintet, J ) 7.6 (2J HH
)
3J HH), 2H], 2.84 [quintet, J ) 7.4
(5) (a) Seitz, W. J .; Saha, A. K.; Hossain, M. M. Organometallics
1993, 12, 2604-2608. (b) Mu¨ller, P.; Baud, C.; Ene´, D.; Motallebi, S.;
Doyle, M. P.; Brandes, B. D.; Dyatkin, A. B.; See, M. M. Helv. Chim.
Acta 1995, 78, 459-470 and references therein.
(2J HH ) 3J HH), 2H], 2.13 (s, 6H, CH3CN), 0.93 (t, J ) 7.6, 6H); 31P{1H}
NMR (161.9 MHz, CDCl3) δ 46.44.
(14) Additional data12 for 4a : 31P{1H} NMR (161.9 MHz, CDCl3) δ
76.38. Partial data for 4b: 1H NMR (400 MHz, CDCl3) δ 5.32 (qd, J )
7.0, 10.6, 1H), 5.01 (qd, J ) 7.0, 10.6, 1H), 4.88 (s, 5H), 1.86 (s, 3H),
1.71 (t, J ) 7.0, 3H); 31P{1H} NMR (161.9 MHz, CDCl3) δ 76.97; correct
analyses for C, H, N, and S.
(6) Schmitz, E.; Eichhorn, I. Acetals and Hemiacetals. In The
Chemistry of the Ether Linkage; Patai, S., Ed.; Interscience: London,
1967; Chapter 7, pp 310-351.
S0276-7333(96)00317-2 CCC: $12.00 © 1996 American Chemical Society
+
+

Communications
Organometallics, Vol. 15, No. 13, 1996 2861
Sch em e 1
Sch em e 2
F igu r e 1. Rate plot for reaction of 3 generated from 1 and
2 in CDCl3, 60 °C. The concentration of 3 is represented
by signal integration (1H NMR) in arbitrary units. For each
data point, a single transient was acquired with a pulse
width of 90°, acquisition time <3 s, with d1> 5T1. Reactions
were followed for at least 3 half-lives. For 3a , k ) 1.9933
((0.0763) × 10-2 min-1 (R 2 ) 0.998); for 3a -d1, k ) 1.4654
2
((0.1148) × 10-2 min-1 (R ) 0.992); for 3b, k ) 2.3583
2
((0.1311) × 10-3 min-1 (R ) 0.995). All deviations are
expressed at the 99% confidence level.
would transfer a proton20 to the ruthenium-bound
alkoxide to give 4 and ROH.
A series of kinetic and other experiments were
performed to shed light on the mechanism of the new
included downfield shifts of the signals for the Cp
protons (from δ 4.36 to 4.89) and -OCH3 protons (from
3.00 to 4.84 ppm). The presence of a CH3CN ligand was
shown by a three-proton singlet at δ 1.84 ppm in the
1H NMR spectrum, by weak IR absorption at 2284
21
transformation. Solutions of 3a in CDCl3 were moni-
1
tored by H NMR spectroscopy. At 60 °C, resonances
for species other than 3a , 4a , CH3CN, or CH3OH were
not detected. Because 2 undergoes CH3CN exchange
by a dissociative pathway,22 and because equilibrium
between added CD3CN (10 equiv) and bound CH3CN
in 3a was reached within 80 s at room temperature, we
propose that CH3CN loss from 3a is more facile than
conversion to 4a , which requires several hours at 60 °C
for completion. If the first step toward 4 is a rapid
equilibrium between 3 and 5, lying on the side of 3,23
before the rate-determining step, the rate law for
disappearance of 3 should be first-order in 3.12 For all
complexes examined this was shown to be the case over
at least 3 half-lives (Figure 1).12 If mechanism a were
cm-1 16
and by correct combustion analysis for C, H, N,
,
and S. Homolog 4b exhibited similar spectral features,
with diastereotopic methylene protons on the ethoxy
group.14
Three mechanisms for the formation of 4 were con-
sidered (Scheme 2). All begin with coordination of an
oxygen to the metal, giving 5, although direct conversion
of 3 to 6, 7, or 8 is also imaginable. Mechanism a would
give 6 by oxidative addition of the methine C-H bond
to the ruthenium center17 and 4 by subsequent elimina-
tion of ROH.18 Alternatively, in mechanism b, oxidative
addition of the C-O bond19 would produce 7, which
would lose ROH to form the metal-carbene bond. In
the final mechanism c, the metal in 5 would act as Lewis
acid, facilitating formation of a carbocation (8), which
(20) Deprotonation of dialkoxycarbonium ions by hindered amines:
Olofson, R. A.; Walinsky, S. W.; Marino, J . P.; J ernow, J . L. J . Am.
Chem. Soc. 1968, 90, 6554-6555. See also ref 7.
(21) Purification of CDCl3 was effected by initial distillation from
K2CO3 and subsequent redistillation from P4O10 onto K2CO3 for storage
in the dark in a glovebox. Addition of HOTf (ca. 0.1 equiv) or Me3-
SiOTf (ca. 0.1 equiv) to reactions of 3a led to faster but considerably
messier reactions, and the non-coordinating base 2,4,6-tris(tert-butyl)-
pyridine (1 equiv) had little effect on the rate. The negligible influence
of added base shows that traces of acidic impurities are not responsible
for the reactions described.
(15) Leading references: Gamasa, M. P.; Gimeno, J .; Mart´ın-Vaca,
B. M.; Borge, J .; Garc´ıa-Granda, S.; Perez-Carren˜o, E. Organometallics
1994, 13, 4045-4057. Pilette, D.; Ouzzine, K.; Le Bozec, H.; Dixneuf,
P. H.; Rickard, C. E. F.; Roper, W. R. Organometallics 1992, 11, 809-
817.
(16) Weak IR absorption of coordinated nitriles: Rouschias, G.;
Wilkinson, G. J . Chem. Soc. A 1967, 993-1000.
(17) C-H activations of alkanes and arenes: J ones, W. D.; Feher,
F. J . Acc. Chem. Res. 1989, 22, 91-100. Crabtree, R. H.; Hamilton, D.
G. Adv. Organomet. Chem. 1988, 28, 299-338. Ryabov, A. D. Chem.
Rev. 1990, 90, 403-424.
(18) Treatment of R-alkoxyalkyl complexes with acid or other
electrophiles can lead to carbene complexes with loss of the alkoxy
group: J olly, P. W.; Pettit, R. J . Am. Chem. Soc. 1966, 88, 5044-5055.
Review: Brookhart, M.; Studabaker, W. B. Chem. Rev. 1987, 87, 411-
432.
(22) Luginbu¨hl, W.; Zbinden, P.; Pittet, P. A.; Armbruster, T.; Bu¨rgi,
H.-B.; Merbach, A. E.; Ludi, A. Inorg. Chem. 1991, 30, 2350-2355.
(23) Heating a solution of 3c prepared in situ from 1c and 2 (60 °C,
24 h) leads to a mixture containing 3c (25%) and a new, chromato-
graphically unstable product (32%) with at least one stereogenic center,
as revealed by the number of signals for the CH2CH2 bridge (four one-
proton multiplets between δ 3.2 and 4.0). On the basis of the
unexceptional chemical shift of an associated one-proton singlet (5.28
ppm), 6c is ruled out, leaving tentative formulation as 7c or 8c.
Decomposition ensues on further heating. Additional data for the
intermediate: 31P{1H} NMR δ 39.41 ppm. Dissolution of 2 and
(2-(methoxymethyl)phenyl)diphenylphosphine, an ether analog of 1a ,
in CDCl3 gives the corresponding η1-P-monophosphine complex and
CH3CN, a mixture which shows no sign of forming a η2-(O,P)-chelate
and a second mole of CH3CN on heating at 60 °C for several days.
(19) This is apparently unknown for acetals: Yamamoto, A. Adv.
Organomet. Chem. 1992, 34, 111-147.
+
+

2862 Organometallics, Vol. 15, No. 13, 1996
Communications
responsible for the formation of 4, a substantial primary
isotope effect might be observed if conversion of 5 to 6
was rate-determining. However, the isotopomer of 3a
featuring a deuterium at the acetal methine site (3a -d)
was used to determine kH/kD ) 1.36(16). This observed
isotope effect is at the lower end of the range of reported
values for primary isotope effects in C-H activation
processes;24 alternatively, the isotope effect is at the
upper end of values reported for secondary isotope
effects in acid-catalyzed acetal hydrolysis, considered to
be a model for mechanism c, 5 f 8.25 A value for the
secondary isotope effect in C-O bond activation (mech-
anism b) could not be found for comparison.
To probe the role of charge dispersal during the
reaction course, 3a was heated in CD3NO2, a polar yet
nondonating solvent.26 The resulting profound reduc-
tion in reaction rate required elevating the temperature
by 35 °C to achieve a rate similar to that seen in CDCl3.
Although this result must be interpreted with caution,
it seems inconsistent with the localization of charge
presumed to accompany conversion of 5 to 8. Further-
more, the ratio of observed rate constants k3a /k3b ) 8.5-
(8) shows a pronounced steric effect on conversion to 4,
which seems too large to be accommodated by a rate-
determining C-H bond activation.27 Taken together,
the available evidence seems most consistent with
approach of the ruthenium center to the C-O bond as
the key step (mechanism b, 5 f 7), which would
represent a new reaction for acetals.
To our knowledge, 4 is the first CpRu-carbene
complex with a potentially labile CH3CN ligand.9,15
Preliminary studies of the reactivity of 4a show that
the acetonitrile ligand is displaced by PMe3 (CDCl3,
room temperature, 3 h; 91% yield).28 Surprisingly,
however, cationic complex 4a was inert to either O-
demethylation29 or ligand substitution by NaI in CD3-
NO2 (60 °C, 3 d). Exchange of the carbene OCH3
substituent for OCD3 by heating with CD3OD (ca. 1
equiv in CDCl3 or as solvent, 60 °C, 12 h) did not occur,
but 4a catalyzes the isomerization of prop-2-en-1-ol to
propanal at room temperature, a reaction which we are
investigating further.
This work establishes the first transformation of an
acetal to an alkoxycarbene complex. Future and ongo-
ing explorations involve the extension of this chemistry
to other aldehyde derivatives and the applications of the
resulting chiral carbene complexes.
Ack n ow led gm en t. We thank Engelhard Corp. for
a generous gift of RuCl3. Dr. Ron Nieman and Camil
J oubran assisted with NMR kinetics experiments.
Su p p or tin g In for m a tion Ava ila ble: Text giving spectral
data, preparations of 12 compounds, and derivations of kinetic
equations and rate plots (23 pages). Ordering information is
given on any current masthead page.
(24) Low isotope effects in C-H bond activation processes have been
attributed to nonlinear hydrogen transfer or an early transition state;
see ref 17.
(25) Shiner, V. J ., J r. In Isotope Effects in Chemical Reactions;
Collins, C. J ., Bowman, N. S., Eds.; Litton Educational Publishing:
New York, 1970; pp 135-136. See also: Reaction Rates of Isotopic
Molecules; Melander, L., Saunders, W. H., J r., Eds.; Wiley: New York,
1980; pp 172-174.
(26) Reichardt, C. Solvents and Solvent Effects in Organic Chemistry;
VCH: Weinheim, Germany, 1988.
(27) For example, kinetic selectivities of Cp*M(PMe3) (M ) Rh, Ir)
for primary C-H bonds of propane and hexane are the same within a
factor of 2: J ones, W. D. In Activation and Functionalization of
Alkanes; Hill, C. L., Ed.; Wiley: New York, 1989.
OM9603171
(28) Partial data for 4 (L ) PMe3, R ) Me): 1H NMR (400 MHz,
CDCl3) δ 5.18 (s, 5H), 4.59 (s, 3H), 0.82 (d, J ) 10.0 (J PH), 9H); 13C
NMR (CDCl3) δ 290.43 (dd, J ) 7.9, 15.8, RudC); 31P{1H} NMR (161.9
MHz, CDCl3) δ 77.35 (d, J ) 35.0), 8.24 (d, J ) 35.0, PMe3).
(29) Davison, A.; Reger, D. L. J . Am. Chem. Soc. 1972, 94, 9237-
9238.
+
+