Mechanistic studies on reaction of [ReH4(η2- H2)(Cyttp)]+ with ketones to give the hydrido-oxo complex [ReH2(O)(Cyttp)]+ (Cyttp = PhP(CH2CH 2CH2PCy2)2)

Article abstract of DOI:10.1016/j.jorganchem.2004.10.027

Mechanistic studies were conducted on reaction of [ReH₄(η ²-H₂)(Cyttp)]OTf (1(OTf); Cyttp = PhP(CH ₂CH₂CH₂PCy₂)₂, OTf = O₃SCF₃) with ketones, both neat and in solution. Treatment of 1(OTf) with excess acetone at 60-65°C affords [ReH₂(O)(Cyttp) ]OTf (2(OTf)) in high yield, nearly 1 equiv. of H₂, 2 equiv. of 2-propanol, 1 equiv. of each of 4-hydroxy-4-methyl-2-pentanone (B) and 4-methylpent-3-en-2-one (C), and smaller amounts of other organic products derived by condensation or related reactions of acetone. The presence of C, apparently arising by dehydration of B, points to the formation of 1 equiv. of H₂O in the reaction system. Use of acetone-d₆ in conjunction with 1(OTf) gives 2(OTf) containing no deuterium, as well as 1 equiv. of each of (CD₃)₂CHOH/OD and (CD₃) ₂CDOD/OH. Reactions of 1(OTf) with cyclohexanone, including cyclohexanone-2,2,6,6-d₄, under comparable conditions, give analogous results. The ketones cyclopentanone, 2-butanone, and 3-pentanone also convert 1(OTf) to 2(OTf) upon heating, as does isobutyraldehyde, but only in the presence of the stabilizer BHT. In contrast, the more robust ketones 2,4-dimethyl-3-pentanone, 2,6-dimethylcyclohexanone, and 2-adamantanone, which do not undergo condensation, failed to effect this transformation. Other organooxygen compounds, i.e., methanol, cyclohexanol, 1,2-butene oxide, cyclohexene oxide, DMSO, and Me₃NO, also are ineffective. A mechanism is proposed which begins with loss of H₂ by 2 to give a 16-electron [ReH₄(Cyttp)]⁺ which, depending on the experimental conditions, binds a solvent or ligand molecule. A [ReH ₄(R₂CO)(Cyttp)]⁺ intermediate generated in this manner reacts spontaneously by elimination of R₂CHOH (containing methine hydrogen even when deuteriated ketone is used), which results from transfer of two hydride ligands to coordinated ketone. Continued reaction leads to the formation of 2 and another molecule of R₂CHOH (containing methine deuterium when deuteriated ketone is employed), with the added hydrogens coming from H₂O, which derives from solvent/reactant ketone.

Full text of DOI:10.1016/j.jorganchem.2004.10.027

Journal of Organometallic Chemistry 690 (2005) 862–873
www.elsevier.com/locate/jorganchem
Mechanistic studies on reaction of [ReH4(g2-H2)(Cyttp)]+
with ketones to give the hydrido-oxo complex
[ReH2(O)(Cyttp)]+ (Cyttp = PhP(CH2CH2CH2PCy2)2)
1
Dean E. Rende , Andrew Wojcicki
*
Department of Chemistry, The Ohio State University, 100 W. 18th Avenue, Columbus, OH 43210-1185, USA
Received 2 September 2004; accepted 8 October 2004
Available online 19 November 2004
Abstract
Mechanistic studies were conducted on reaction of [ReH4(g2-H2)(Cyttp)]OTf (1(OTf); Cyttp = PhP(CH2CH2CH2PCy2)2,
OTf = O3SCF3) with ketones, both neat and in solution. Treatment of 1(OTf) with excess acetone at 60–65 °C affords [ReH2(O)-
(Cyttp)]OTf (2(OTf)) in high yield, nearly 1 equiv. of H2, 2 equiv. of 2-propanol, 1 equiv. of each of 4-hydroxy-4-methyl-2-penta-
none (B) and 4-methylpent-3-en-2-one (C), and smaller amounts of other organic products derived by condensation or related
reactions of acetone. The presence of C, apparently arising by dehydration of B, points to the formation of 1 equiv. of H2O in
the reaction system. Use of acetone-d6 in conjunction with 1(OTf) gives 2(OTf) containing no deuterium, as well as 1 equiv. of each
of (CD3)2CHOH/OD and (CD3)2CDOD/OH. Reactions of 1(OTf) with cyclohexanone, including cyclohexanone-2,2,6,6-d4, under
comparable conditions, give analogous results. The ketones cyclopentanone, 2-butanone, and 3-pentanone also convert 1(OTf) to
2(OTf) upon heating, as does isobutyraldehyde, but only in the presence of the stabilizer BHT. In contrast, the more robust ketones
2,4-dimethyl-3-pentanone, 2,6-dimethylcyclohexanone, and 2-adamantanone, which do not undergo condensation, failed to effect
this transformation. Other organooxygen compounds, i.e., methanol, cyclohexanol, 1,2-butene oxide, cyclohexene oxide, DMSO,
and Me3NO, also are ineffective. A mechanism is proposed which begins with loss of H2 by 2 to give a 16-electron ‘‘[ReH4(Cyttp)]+’’
which, depending on the experimental conditions, binds a solvent or ligand molecule. A [ReH4(R2C@O)(Cyttp)]+ intermediate gen-
erated in this manner reacts spontaneously by elimination of R2CHOH (containing methine hydrogen even when deuteriated ketone
is used), which results from transfer of two hydride ligands to coordinated ketone. Continued reaction leads to the formation of 2
and another molecule of R2CHOH (containing methine deuterium when deuteriated ketone is employed), with the added hydrogens
coming from H2O, which derives from solvent/reactant ketone.
Ó 2004 Elsevier B.V. All rights reserved.
Keywords: Mechanism; Rhenium complexes; Hydrido-oxo complexes; Ketone complexes; Oxygen transfer; Isotope effects
1. Introduction
H2(O)(Cyttp)]+ (Cyttp = PhP(CH2CH2CH2PCy2)2) (2)
[1], which displays a rich chemistry, including transfer
of both hydrogen and oxygen to unsaturated substrates
such as CO and nitriles [2]. This chemical behavior is un-
ique among known mononuclear transition-metal hydr-
ido-oxo complexes [1–8].
Complex 2 was prepared as 2(SbF6) or 2(OTf)
(OTf = O3SCF3) by reaction of [ReH4(g2-H2)-
(Cyttp)]SbF6 (1(SbF6)) or –OTf (1(OTf)), respectively,
with neat acetone or cyclohexanone on heating
We previously reported the synthesis and characteri-
zation of the rhenium(V) hydrido-oxo complex [Re-
*
Corresponding author. Tel.: +1 614 292 4750; fax: +1 614 292
1685.
E-mail address: wojcicki@chemistry.ohio-state.edu (A. Wojcicki).
1
Present address: UOP, 50 East Algonquin Road, P.O. Box 5016,
Des Plaines, IL 60017-6016, USA.
0022-328X/$ - see front matter Ó 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2004.10.027

D.E. Rende, A. Wojcicki / Journal of Organometallic Chemistry 690 (2005) 862–873
863
Scheme 1.
(Scheme 1A) [1]. The net chemical transformation may
2. Experimental
be described as loss of g2-H2, replacement of two
hydrido ligands with the ketone-derived oxygen, and
retention of the oxidation number of +5 by rhenium
(Scheme 1B). It is unprecedented in synthetic method-
ology for transition-metal hydrido-oxo complexes.
Known preparative routes to such compounds include
replacement of methylene ligand in a metal hydrido
complex with oxide from H2O [3], substitution of ha-
lide or alkoxide ligand in appropriate metal oxo com-
plexes by hydride from BH3 Æ THF or (n-Bu)3SnH
[5a,7], thermal decomposition of a metal oxo-formato
complex [5a], and migratory rearrangement of a metal
hydroxo complex [6]. The only reported reaction
based on replacement of hydrido ligand with oxide
is that between ½Cp0 TaH2ꢀþ (Cp0 = g5-C5H4Bu-t) and
OHꢁ to afford Cp022TaðOÞH [4]; however, the source
of entering oxygen there is very different from that
in Scheme 1.
2.1. General procedures and measurements
Reactions and manipulations of air-sensitive com-
pounds were conducted under an atmosphere of dry
argon by use of standard procedures [10]. Solvents
were dried [11], distilled under argon, and degassed
before use. Elemental analyses were carried out by
M-H-W Laboratories, Phoenix, AZ. IR and NMR
2
(1H, H, 13C, 19F, and 31P) spectra were obtained as
previously described [12,13]. GC-MS-IR experiments
were performed on a HP5890A Series II GC con-
nected to
a
HP5970B Mass Selective Detector
(MSD) and a HP5965A Infrared Detector (IRD).
Injections were usually accomplished by using a
7673A Automatic Injector. The columns leading to
MSD and IRD were Altech RSL300 with the station-
ary phase consisting of crosslinked 35% phenyl and
50% methyl silicone. The IRD column was 30-m long
and 0.33-mm ID. The MSD column was 15-m long
and 0.25-mm ID linked to a 1-m long and 0.1-mm
ID deactivated fused silica tube.
Oxygen atom transfer from ketones and related or-
ganic compounds to metal was studied by Mayer for
tungsten(II) complexes [9]. Reactions of WCl2L4
(L = PMePh2) with ketones (R2C@O) result in
substitution of L by R2C@O followed by conversion
to tungsten(IV) or tungsten(VI) products, WCl2(O)L3
or WCl2(O)(CR2)L2, respectively. These reactions
2.2. Materials
certainly proceed by
a
mechanism substantially
different from that expected for the transformation
of 1 to 2 by the action of acetone or cyclohexanone,
where the oxidation number of metal remains
unchanged.
In this paper we report our studies directed at
gaining mechanistic insight into the formation of
2(OTf) from 1(OTf) and ketones. Isotopically
labeled reactants were employed, and organic products
were systematically analyzed. Reactions of 1(OTf)
with other organooxygen compounds were also
examined.
Most of the reagents were procured from
various commercial sources. Organic liquids were
dried [11] and distilled, whereas solids were
generally used as received. The complexes Re-
H5(Cyttp) [14] and [ReH4(g2-H2)(Cyttp)]OTf (1(OTf))
[1b,15] weꢁre synthesized by published procedures.
The BPh4
salt of 1, [ReH4(g2-H2)(Cyttp)]BPh4
(1(BPh4)), was obtained from 1(OTf) and NaBPh4 in
methanol. The complex [ReH2(O)(Cyttp)]OTf
(2(OTf)) was prepared and completely characterized
earlier [1b].

864
D.E. Rende, A. Wojcicki / Journal of Organometallic Chemistry 690 (2005) 862–873
2.3. Preparation of [ReH4(L)(Cyttp)]n (L = MeCN,
n = +1; OTfꢁ, n = 0)
2.4.2. Reaction with acetone
2.4.2.1. Characterization of noncondensable gas.
A
flask fitted with a Solv-seal adapter and a valve stop-
cock was charged with 1(OTf) (1.190 g, 1.282 mmol),
excess acetone (10.0 ml, 7.91 g, 136 mmol), and a stir
bar. It was then connected to a vacuum line, and after
three freeze–pump–thaw cycles the stopcock was
closed to isolate the solution. The flask was placed
in an oil bath at ca. 60 °C, and the contents were stir-
red for 15 h. The solution was then cooled in a liquid
nitrogen bath, and the noncondensable gas was re-
moved by use of a Toepler pump. The gas was meas-
ured by using a calibrated vacuum manifold. Yield of
H2, 0.821 mmol (64% based on 1(OTf) used in the
reaction; however, since the reaction was only 80%
complete (by 31P{1H} NMR spectroscopy), the yield
based on 1(OTf) consumed in the reaction was
80%). MS: m/z 2 (M+).
2.3.1. Reaction of [ReH4(g2-H2)(Cyttp)]OTf
(1(OTf)) with MeCN
To 1(OTf) (0.213 g, 0.230 mmol) in benzene was
added 5.00 ml (3.96 g, 95.7 mmol) of MeCN, and
the resulting mixture was heated at 60 °C for 6 h.
Concentration of the solution to ca. 2 ml and addition
of 10 ml of Et2O afforded a white solid which was
filtered off, washed with 10 ml of Et2O, and dried
overnight under vacuum. The yield of [ReH4(MeCN)-
(Cyttp)]OTf (3(OTf)) was 0.19 g (86%). The product
was characterized by comparison of its 1H and
31P{1H} NMR spectra with those reported for
3(BF4), obtained from ReH5(Cyttp), [C7H7]BF4, and
MeCN [15].
2.3.2. Reaction of ReH5(Cyttp) with MeOTf
2.4.2.2. Characterization of organic products. A screw-
cap NMR tube was charged with 1(OTf) (0.050 g, 0.053
mmol) and 0.500 ml (0.396 g, 6.81 mmol) of acetone.
The NMR tube was placed in an oil bath at 60 °C,
and after 15 h the contents were transferred to a distilla-
tion apparatus. The volatiles were collected on a vac-
uum line in a cold trap using liquid nitrogen.
To a solution of ReH5(Cyttp) (0.200 g, 0.257 mmol)
in 10 ml of benzene at room temperature was added an
excess of MeOTf (0.120 ml, 0.0828 g, 0.505 mmol),
resulting in the evolution of a gas, presumably CH4.
After stirring for 30 min, the solution was concentrated
to 1 ml, and vigorous agitation resulted in the forma-
tion of a brown solid. The solid was collected on a
filter frit and washed with a small amount (ca. 1 ml)
of benzene. It was then dried under vacuum overnight
to yield 0.12 g (50%) of white ReH4(OTf)(Cyttp) (4).
Selected spectroscopic data. 19F{1H} NMR: d ꢁ77.2
(referenced to internal CFCl3, s, OTf) ppm. 31P{1H}
C3H8O (A): Yield 1.86 equiv. IR: m(OH) 3652 (w),
1
m(CH) 2975 (s), 1377 (m), 1142 (m), 951 (w) cmꢁ1. H
NMR: d 3.8 (m, CH), 1.1 (d, Me) ppm. MS: m/z 59
(M+ ꢁ H), 45. C6H12O (D): Yield 0.05 equiv. IR:
m(CH) 2968 (s), m(C@O) 1730 (vs), 1384 (m), 1182 (m)
cmꢁ1 1H NMR: d 0.81 (d, Me) ppm. MS: m/z 100
.
2
(M+), 85, 58, 57, 43. C6H10O (C): Yield 1.07 equiv.
IR: m(CH) 2927 (s), m(C@O) 1701 (vs), 1630 (vs), 1446
NMR (C6D6): d 19.5 (t, JPcPw = 12.9 Hz, Pc), 17.0
(d, JPcPw = 12.9 Hz, Pw) ppm. Anal. Found: C,
2
1
(m), 1365 (s), 1164 (s) cmꢁ1. H NMR: d 6.07 (s,@CH)
47.91; H, 6.89. Calc. for C37H65F3O3P3ReS: C, 47.99;
H, 7.07%.
ppm. MS: m/z 98 (M+), 83, 55, 43. C6H12O2 (B): Yield
0.92 equiv. IR: m(OH) 3566 (w), m(CH) 2980 (s),
1
m(C@O) 1735 (vs), 1370 (s), 1185 (s), 950 (w) cmꢁ1. H
2.4. Reactions of [ReH4(g2-H2)(Cyttp)]OTf
(1(OTf)) with ketones
NMR: d 3.70 (s), 2.52 (s), 1.10 (s) ppm. MS: m/z 101
(M+ ꢁ Me), 83, 59, 43. C9H18O2 (E): Yield 0.02 equiv.
IR: m(CH) 2987 (s), 1458 (w), 1376 (s), 1181 (s), 1002
(m) cmꢁ1. MS: m/z 143 (M+ ꢁ Me), 101, 83, 59, 55, 43.
A similarly conducted reaction of 1(BPh4) with ace-
tone gave comparable results.
2.4.1. General considerations
Reaction progress was monitored by 31P{1H} NMR
spectroscopy. Unless stated otherwise, [ReH2(O)(Cytt-
p)]OTf (2(OTf)) was the only major (and generally the
only) rhenium-containing product observed. It was iso-
lated from the nonvolatile reaction residue as described
previously and was characterized by comparison of its
spectroscopic properties with those already reported
[1b]. The volatile organic products were characterized
by GC-MS-IR and 1H NMR spectroscopy. Selected
data are provided for each product. In most cases, spec-
tra were compared with those recorded for authentic
samples. Yields of organic products (given as no. of
equivalents) were obtained by GC analysis and are
based on 2(OTf) formed.
2.4.3. Reaction with acetone-d6
The reaction and workup were carried out as in Sec-
tion 2.4.2.2 by using 1(OTf) (0.030 g, 0.032 mmol) and
acetone-d6 (0.500 ml, 0.436 g, 6.80 mmol).
C3H8 ꢁXDXO (A): Yield 1.80 equiv. 1H NMR: d 3.85
(s, CH) ppm. MS: m/z 49 (M+ ꢁ CD3). C6H12 ꢁXDXO
(D): MS: m/z (112, 111, 110 equal intensity) (M+), (93,
92, 91), 78, 46. C6H10 ꢁXDXO (C): MS: m/z 108 (M+),
90, 62, 46. C6H12 ꢁXDXO2 (B): IR: m(CD) 2221 (m),
m(C@O) 1725 (vs), 1240 (s) cmꢁ1. MS: m/z 109
(M+ ꢁ Me), 65, 46. C9H18 ꢁXDXO2 (E): IR: m(CD)

D.E. Rende, A. Wojcicki / Journal of Organometallic Chemistry 690 (2005) 862–873
865
2220 (m), 1260 (m), 1185 (s), 1020 (m) cmꢁ1. MS: m/z
158 (M+ ꢁ Me), 94, 66, 62, 46.
Yield 0.06 equiv. MS: m/z (185, 184 equal intensity)
(M+), (99, 98), 84, 71.
2.4.4. Reaction with acetone-2-13C
2.4.7. Reaction with cyclopentanone
The reaction and workup were conducted as in Sec-
tion 2.4.2.2 by using 1(OTf) (0.030 g, 0.032 mmol) and
acetone-2-13C (0.500 ml, 0.402 g, 6.80 mmol).
The reaction and workup were conducted as de-
scribed in Section 2.4.2.2 by using 1(OTf) (0.030 g,
0.032 mmol) and cyclopentanone (0.015 ml, 0.014 g,
0.17 mmol) in 0.5 ml of benzene.
12C213CH8O (A): Yield 1.84 equiv. H NMR: d 3.8
1
(dm, CH), 1.05 (dd, Me) ppm. 13C NMR: d 63.5 (d,
1JCH = 140 Hz, CH) ppm. MS: m/z 46 (M+ ꢁ Me).
12C413C2H12O (D): Yield 0.09 equiv. IR: m(CH) 2964
C5H10O (I): Yield 1.60 equiv. IR: m(OH) 3655 (w),
m(CH) 2981 (s), 1450 (w), 1343 (w), 1177 (w), 1005 (w)
1
cmꢁ1. H NMR: d 4.1 (m, CH), 1.1 (m, CH2) ppm.
(s), m(C@O) 1660 (s), 1359 (m), 1153 (m) cmꢁ1
.
1H
MS: m/z 85 (M+ ꢁ H). C10H14O (J): MS: m/z 150
NMR: d 0.80 (dd, Me) ppm. MS: m/z 102 (M+), 87,
59, 44. 12C413C2H10O (C): Yield 0.80 equiv. IR: m(CH)
2927 (s), m(C@O) 1660 (s), 1586 (s), 1363 (s), 1150 (s),
(M+), 149, 84, 79.
2.4.8. Reaction with each of 2-butanone and 3-pentanone
A solution of 1(OTf) (0.020 g, 0.022 mmol) in 2-buta-
none or 3-pentanone (5.00 ml) was heated at 65 °C while
being monitored by 31P{1H} NMR spectroscopy. After
6 h, only [ReH2(O)(Cyttp)]OTf (2(OTf)) was detected.
Organic products were not analyzed.
1
954 (m), 614 (m) cmꢁ1. H NMR: d 6.1 (@CH) ppm.
13C NMR: d 198 (s, C@O), 154 (s,@C) ppm. MS: m/z
100 (M+), 85, 56, 44. 12C413C2H12O2 (B): Yield 0.87
equiv. IR: m(OH) 3566 (w), m(CH) 2980 (s), m(C@O)
1
1682 (s), 1366 (s), 1165 (s), 936 (w) cmꢁ1. H NMR: d
2.6 (t), 1.2 (d) ppm. 13C NMR: d 209 (s, C@O), 71 (s,
COH) ppm. MS: m/z 103 (M+ ꢁ Me), 85, 60, 44.
12C613C3H18O2 (E): Yield 0.06 equiv. MS: m/z 146
(M+ ꢁ Me), 103, 85, 60, 44.
2.4.9. Attempted reactions with 2,4-dimethyl-3-
pentanone, 2,6-dimethylcyclohexanone, and 2-
adamantanone
A mixture of 1(OTf) (0.026 g, 0.028 mmol) and 2,4-
dimethyl-3-pentanone (0.500 ml, 3.54 mmol) was heated
at 65 °C for 6 h. 31P{1H} NMR spectroscopy showed
some uncharacterized Cyttp-containing species, but no
2(OTf). Under similar conditions, there was also no for-
mation of 2(OTf) from 1(OTf) (0.020 g, 0.022 mmol)
and each of 2,6-dimethylcyclohexanone (0.030 ml, 0.22
mmol) and 2-adamantanone (0.028 g, 0.187 mmol) in
THF (0.5 ml).
2.4.5. Reaction with cyclohexanone
The reaction and workup were conducted as in Sec-
tion 2.4.2.2 by using 1(OTf) (0.030 g, 0.032 mmol) and
cyclohexanone (0.500 ml, 0.474 g, 4.82 mmol).
1
C6H12O (F): H NMR: d 3.7 (m, CH) ppm. MS: m/z
100 (M+), 82, 67, 57, 44. C12H18O (G): MS: m/z 178
(M+), 149, 135, 107, 93, 79, 67.
This reaction was also carried out in benzene-d6 (0.5
ml) with 0.030 g (0.032 mmol) of 1(OTf) and 0.020 ml
(0.19 mmol) of cyclohexanone at 60 °C for 16 h. The
workup was the same as in Section 2.4.2.2.
C6H12O (F): Yield 1.91 equiv. IR: m(OH) 3651 (w),
m(CH) 2955 (s), m(CH) 2885 (m), 1455 (s), 1370 (w),
1085 (m), 966 (w) cmꢁ1. MS: m/z 100 (M+), 82, 67, 57,
44. C12H20O (H): Yield 0.08 equiv. IR: m(CH) 2942 (s),
m(CH) 2864 (m), 1663 (m), 1451 (w), 1363 (w), 1266
(w), 1176 (s), 1054 (w) cmꢁ1. MS: m/z 180 (M+), 98,
83, 70, 55, 41.
2.5. Reaction of [ReH4(g2-H2)(Cyttp)]OTf (1(OTf))
with isobutyraldehyde
A solution of 1(OTf) (0.026 g, 0.028 mmol), isobutyr-
aldehyde (0.050 ml, 0.482 mmol), and 2,6-di-t-butyl-p-
cresol (BHT) (0.030 g, 0.136 mmol) in 0.5 ml of THF
was heated at 60 °C for 16 h. It was then transferred
to a distillation apparatus, and the volatiles were col-
lected on a vacuum line in a cold trap using liquid nitro-
gen. The inorganic product in the residue was identified
spectroscopically as [ReH2(O)(Cyttp)]OTf (2(OTf)).
C4H10O (K): MS: m/z 74 (M+), 59, 56. C8H16O (L):
IR: m(CH) 2988 (s), m(C@O) 1890 (m), 1468 (w), 1388
(w), 1171 (s), 1049 (w) cmꢁ1. MS: m/z 128 (M+), 72,
57. C8H16O2 (M): IR: m(CH) 2975 (s), m(C@O) 1769
(s), 1426 (m), 1387 (w), 1254 (m), 1188 (s), 1076 (m),
999 (w) cmꢁ1. MS: m/z 144 (M+), 129, 101, 89, 71, 56,
55. C12H26O2 (N): IR: m(CH) 2987 (s), m(CH) 2883
(m), 1891 (w), 1472 (m), 1388 (m), 1053 (s) cmꢁ1. MS:
m/z 159 (M+ ꢁ CHMe2), 129, 103, 73, 57.
2.4.6. Reaction with cyclohexanone-2,2,6,6-d4
The reaction and workup were carried out as detailed
in Section 2.4.2.2, but in THF solution (0.5 ml), by using
1(OTf) (0.030 g, 0.032 mmol) and cyclohexanone-
2,2,6,6-d4 (0.020 ml, 0.019 g, 0.19 mmol).
C6H12 ꢁXDXO (F): Yield 1.92 equiv. IR: m(OH) 3653
(w), m(CH) 2936 (s), m(CH) 2869 (m), m(CD) 2197 (w),
m(CD) 2110 (w), 1454 (m), 1064 (m), 911 (w) cmꢁ1. H
1
2
NMR: d 3.5 (s, CH), 1.3 (s, CH2) ppm. H NMR: d
3.5 (s, CD), 1.2 (s, CD2) ppm. MS: m/z (104, 103 equal
intensity) (M+), (86, 85), 58, 46. C12H18 ꢁXDXO (G):

866
D.E. Rende, A. Wojcicki / Journal of Organometallic Chemistry 690 (2005) 862–873
2.6. Attempted reactions of [ReH4(g2-H2)-
(Cyttp)]OTf (1(OTf)) with other organooxygen
compounds
of any gaseous products evolved, detection and/or isola-
tion of possible inorganic reaction intermediates, char-
acterization of the organic compounds produced,
isotopic labeling, and use of various ketones and other
organooxygen compounds to determine generality of
this reaction. Essentially all of the work was done on
1(OTf); limited previous [1b] and present studies on
1(SbF6) and 1(BPh4) showed that the counterion has lit-
tle, if any, effect on the formation of 2 from 1. Acetone
and cyclohexanone were employed in most experiments.
Generally, reactions required heating at 60–65 °C; no
progress was observed at room temperature within sev-
eral hours. Experiments were carried out in neat organic
reactants as well as in benzene, toluene, or THF
solution.
Reactions were carried out between 1(OTf) and an
excess (P10-fold) of each of the following: 1,2-butene
oxide, cyclohexene oxide, Me3NO, DMSO, methanol,
and cyclohexanol either neat or in THF solution gener-
ally at 65 °C, with monitoring by 31P{1H} NMR spectr-
oscopy. No [ReH2(O)(Cyttp)]OTf (2(OTf)) was
observed. Either no reaction took place or uncharacter-
ized Cyttp-containing products were formed. From
1(OTf) and Me3NO in THF at 25 °C, ReH5(Cyttp)
[14] was isolated in 84% yield.
2.7. Reactions of [ReH4(MeCN)(Cyttp)]OTf
(3(OTf)) with ketones
3.1. Gas analysis
A solution of 3(OTf) (0.030 g, 0.031 mmol) in cyclo-
hexanone (0.500 ml, 0.474 g, 4.82 mmol) was heated at
95 °C for 1 h. The solution was transferred to a distilla-
tion apparatus, and the volatiles were collected in a cold
trap using a liquid nitrogen bath.
A solution of 1(OTf) in acetone was heated at ca.
60 °C for 15 h on a vacuum line. Proper precautions
were taken to remove any noncondensable gas prior to
the start of the reaction (cf. Section 2.4.2.1). The gas
evolved was collected and characterized as H2 by MS.
Based on the amount of 1(OTf) consumed (or 2(OTf)
produced), 80% of the H2 expected from dissociation
of g2-H2 in 1(OTf) during the reaction was collected.
1
C2H3N: H NMR: d 0.70 (s) ppm. MS: m/z 41 (M+),
40. C6H12O (F): 1H NMR: d 3.82 (m) ppm. MS: m/z 100
(M+), 82, 67, 57, 44. C12H18O (G): MS: m/z 178 (M+),
149, 135, 107, 93, 79, 57.
A 31P{1H} NMR spectrum of the nonvolatile residue
showed the presence of [ReH2(O)(Cyttp)]OTf (2(OTf)),
along with a minor impurity.
3.2. Studies directed at detection and/or isolation of
possible inorganic reaction intermediates
Use of acetone in place of cyclohexanone also resulted
in good yields of 2(OTf) when lower concentrations of
3(OTf) and longer reaction times were employed.
No inorganic intermediates could be detected by
31P{1H} NMR spectroscopy during the conversion of
1(OTf) to 2(OTf) by acetone. Loss of H2 on heating
1(OTf) in acetone may result in the formation of a coor-
dinatively unsaturated ‘‘[ReH4(Cyttp)]+’’ species, which
could then be trapped by the solvent/reactant or OTfꢁ.
To learn more about such possible behavior, we exam-
ined chemical composition of 1(OTf) solutions in ben-
zene, toluene, and THF at elevated temperatures.
2.8. Reactions of ReH4(OTf)(Cyttp) (4) with ketones
A solution of 4 (0.030 g, 0.032 mmol) in cyclohexa-
none (0.500 ml, 0.474 g, 4.82 mmol) was heated at
90 °C for 1 h. The yield of [ReH2(O)(Cyttp)]OTf
(2(OTf)) as determined by 31P{1H} NMR spectroscopy
was less than 60%. Another Cyttp-containing product
was observed, but could not be isolated or characterized.
Heating benzene or toluene solutions of 1(OTf) at 60
°C leads to the formation of a Cyttp-containing product
2
with 31P{1H} resonances at d 19.5 (t, JPcPw = 12.9 Hz,
PC) and 17.1 (d, JPcPw = 12.9 Hz, PW) ppm. There is
1
2
Cyclohexanol was detected by H NMR spectroscopy.
Similar reactions were carried out with acetone in
place of cyclohexanone. 31P{1H} NMR spectroscopy
showed formation of 2(OTf), generally in less than
50% yield, and 2-propanol was observed by H NMR
spectroscopy.
also another phosphorus-containing species with a very
weak set of signals at d ca. 19.3 (partially masked) and
2
18.5 (d, JPcPw ꢂ 13.1 Hz, PW) ppm (cf. Scheme 2).
1
2
These patterns and values of JPcPw are consistent with
a meridional arrangement of the Cyttp ligand and the
presence of a weak ligand opposite PC [15]. A similar
31P{1H} NMR spectrum was obtained for a heated solu-
tion of 1(OTf) in THF, except that the species with the
signals at d 19.3 and 18.5 ppm now occurs in signifi-
cantly higher concentration. The 19F{1H} NMR spectra
of thermolyzed solution of 1(OTf) in all three solvents
show singlets at d ꢁ77.2 and ꢁ77.3 ppm which are as-
signed to coordinated and free triflate, respectively
3. Results
The following approaches were employed in an at-
tempt to elucidate the mechanism of conversion of
[ReH4(g2-H2)(Cyttp)]+ (1) to [ReH2(O)(Cyttp)]+ (2) by
the action of ketones (cf. Scheme 1): characterization

D.E. Rende, A. Wojcicki / Journal of Organometallic Chemistry 690 (2005) 862–873
867
Scheme 4.
Scheme 2.
[16]. From their relative intensities in these solvents, it
appears that the product with the 31P{1H} resonances
at d 19.5 and 17.1 ppm is [ReH4(solvent)(Cyttp)]OTf,
whereas that with the resonances at d 19.3 and 18.5
ppm is ReH4(OTf)(Cyttp) (4).
All attempts at isolation of the aforementioned com-
plexes from thermally treated solutions of 1(OTf) led
either to a mixture of products or to decomposition.
Therefore, we turned to more stable analogs of such
complexes or to other synthetic routes. A complex of
the type [ReH4(solvent)(Cyttp)]OTf could be isolated
pure when solvent = MeCN (Scheme 3), a better ligand
than THF. This previously reported compound [15] was
characterized spectroscopically as [ReH4(MeCN)(Cytt-
p)]OTf (3(OTf)), with its 19F{1H} resonance at d
ꢁ77.3 ppm implicating ionic triflate [16]. Complex 4
was prepared as shown in Scheme 4 and was character-
ized by 31P{1H} and 19F{1H} NMR spectroscopy and
elemental analysis (Section 2.3.2). The spectroscopic
data are essentially those recorded for the thermally
treated 1(OTf) in each of benzene, toluene, and THF.
When solvent was removed from a solution of 4 in ben-
zene and the residue was dissolved in THF, the two
NMR spectra changed accordingly. They reverted to
the original ones upon removal of THF and dissolution
of the solid in benzene.
Attempts were made to convert 3(OTf) to 2(OTf) by
the action of acetone. Initially, however, a solution of
1(OTf) (1 equiv.) and a small quantity of MeCN (ca.
0.2 equiv.) in acetone (large excess) was heated at 60
°C. After 15 h, the mixture was shown by 31P{1H}
NMR spectroscopy to contain 2(OTf) and 3(OTf).
On raising the temperature or lengthening reaction
time, only 2(OTf) was discernible. This result suggests
that MeCN is competing for a coordinatively unsatu-
rated ‘‘[ReH4(Cyttp)]+’’ intermediate. Reaction of
Scheme 5.
Scheme 6.
authentic 3(OTf) with neat acetone at reflux tempera-
ture for 15 h afforded good yields of 2(OTf), and
2-propanol was detected by 1H NMR spectroscopy.
Heating more concentrated solutions of 3(OTf) in
cyclohexanone at 95 °C for only 1 h achieved similar
conversion to 2(OTf) (Scheme 5), with a minor
Cyttp-containing side product also being formed. The
organic reaction products characterized by 1H NMR
and GC-MS were MeCN, cyclohexanol, and the enone
C12H18O (G) (cf. Section 3.3).
In parallel experiments, heating 4 in neat acetone or
cyclohexanone at 90 °C for 1 h afforded 2(OTf) in
<50% and ca. 60% yield, respectively. Other uncharac-
terized Cyttp-containing products also were observed
by 31P{1H} NMR spectroscopy. 2-Propanol and cyclo-
hexanol, respectively, were detected by 1H NMR spectr-
oscopy in the two reactions (Scheme 6).
In summary, reactions of 3(OTf) and 4 with each
of acetone and cyclohexanone proceed to 2(OTf) and
the appropriate alcohols, as do the corresponding reac-
tions of 1(OTf). However, in general, and especially
for 4 and acetone, they do not appear to progress
as cleanly as those of 1(OTf) under comparable
conditions.
Scheme 3.

868
D.E. Rende, A. Wojcicki / Journal of Organometallic Chemistry 690 (2005) 862–873
were detected. Again, 2 equiv. of alcohol (F) were ob-
tained based on 2(OTf) generated. The IR and 1H
NMR spectra of F (Section 2.4.5) closely match those
of cyclohexanol, and the MS shows a parent ion peak
at m/z 100. The fragmentation pattern and retention
time on the GC-MS column compare well with those ob-
served for an authentic sample. Similarly, the MS of G
shows good agreement with the spectrum of the authen-
tic enone.
3.3. Characterization of organic products from reactions
of 1(OTf) with ketones
Reactions of 1(OTf) with various ketones were con-
ducted as described in the appropriate parts of Section
2.4. Volatile organic products were analyzed by GC-
IR, GC-MS, and NMR spectroscopy, and selected data
are presented with the description of each reaction.
Treatment of 1(OTf) with neat acetone at 60 °C for
15 h yields the organic compounds A–E. Formation of
1
When the reaction in question was run in ben-
zene solution under similar conditions with a 5-fold
excess of cyclohexanone over 1(OTf), excellent
conversion to 2(OTf) was achieved. This is in contrast
to the analogous reaction with acetone, which requires
ca. 100-fold excess of the ketone over 1(OTf) in a
rather concentrated benzene solution to proceed
cleanly. Analysis of the organics from the cyclohexa-
none reaction revealed the formation of 2 equiv. of
cyclohexanol (F) and a much smaller amount (0.08
equiv.) of the vinyl ether H. The latter may be
formed by acid-catalyzed reaction of cyclohexanone
with F.
Good yields of 2(OTf) were also realized when a solu-
tion of 1(OTf) and cyclopentanone (neat or in benzene)
was heated at 60 °C for 15 h. Cyclopentanol (I) and the
enone J were detected as the major organic products
(Section 2.4.7). Again, approximately 2 equiv. of I were
obtained.
2-propanol (A) was inferred from the H NMR spec-
trum which showed a multiplet at 3.8 ppm (CH) and a
doublet at 1.1 ppm (Me). Acetone and A could not be
completely separated by GC-MS; however, by subtract-
ing out the fragmentation pattern of acetone, the MS of
A was obtained that agrees well with the spectrum of
authentic 2-propanol. The amount of A generated was
determined by integration of the CH proton signal of
A and the rhenium hydride signals of 2(OTf). Approxi-
mately 2 (experimental, 1.86) equiv. of A were obtained
based on 2(OTf) produced. The remaining organic com-
pounds B–E represent addition/condensation and re-
lated products derived from acetone rather than
primary products of the conversion of 1 to 2. Their for-
mation may be promoted by metal complexes in the
reaction system. The presence of B (diacetone alcohol)
is indicated by 1H NMR spectroscopy (singlets at d
3.70, 2.52, and 1.10 ppm in acetone solution), and its
MS fragmentation pattern matches that of an authentic
sample. The IR spectrum reveals m(OH) and m(C@O)
absorptions that are in complete accord with those of
authentic 4-hydroxy-4-methyl-2-pentanone. Compound
1
C, derived by loss of H2O from B, shows H NMR,
IR, and GC-MS spectra that are consistent with the
spectra of this enone. As will be considered again later,
it is significant that approximately one (experimental,
1.07) equiv. of C forms with each equiv. of 2(OTf).
The remaining two compounds, D and E, were obtained
in low yields and are assigned structures based on the
data given in Section 2.4.2. Compound E and the
authentic cyclic ketal have the same retention time on
the GC-IR and GC-MS columns. The source of E
may be reaction of acetone with 2,4-dihydroxy-2-meth-
ylpentane, formed by hydrogenation of B [17]. Com-
pound D could result from hydrogenation of C (recall
that H2 was lost by 1(OTf) on heating).
Several other ketones were examined, but in less
detail than the foregoing three, with regard to reactiv-
ity toward 1(OTf). Thus, neat solutions of 1(OTf) in
each of 2-butanone and 3-pentanone showed conver-
sion to 2(OTf) in 6 h at 72 °C. Organic products of
these reactions were not analyzed. In contrast, none
of 2,4-dimethyl-3-pentanone, 2,2-dimethylcyclohexa-
none, and 2-adamantanone produced detectable
2(OTf) on heating either neat or in THF solution.
The latter three ketones are characterized by their
more robust properties and inability to undergo con-
densation reaction.
Reaction between 1(OTf) and neat cyclohexanone,
also at 60 °C for 15 h, affords compounds F and G as
the major organic products. No hydrocarbon products

D.E. Rende, A. Wojcicki / Journal of Organometallic Chemistry 690 (2005) 862–873
869
3.4. Isotopic labeling experiments
Reactions of 1(OTf) with isotopically labeled acetone
and cyclohexanone were investigated to gain additional
insight into mechanism. Deuterium labeling will be con-
sidered first.
Use of acetone-d6 in conjunction with 1(OTf) under
the same conditions as those for the analogous reaction
of unlabeled acetone afforded 2(OTf) that contained no
incorporation of deuterium into the Re–H. This was
determined by comparing the intensities of the inte-
grated signals at d ꢁ1.0 and ꢁ12.4 ppm of the two hy-
drides and at d 8.3 ppm of two Cyttp phenyl protons in
the NMR spectrum of isolated 2(OTf). As expected for
the absence of isotopic exchange, an intensity ratio of
1:1:2 was observed. Also of significance in the 1H
NMR spectrum at completion of the reaction is the rel-
ative intensity of the signal of the CH proton of A at d
3.85 ppm. It indicates that only 1 equiv. of 2-propanol
was generated which contains unlabeled hydrogen. This
species must result from reduction of the ketone by
1(OTf). Since a GC analysis of the volatile products
of the reaction reveal that a total of 2 equiv. of 2-pro-
panol was formed, the methine group of the other
equivalent of the alcohol must contain deuterium.
The possible sources of this deuterium are acetone-d6
and D2O formed in condensation reaction. Organic
products of condensation similar (except for isotopic
composition) to those obtained in the reaction of
1(OTf) with unlabeled acetone, viz., B–E, were ob-
served by GC-MS-IR (cf. Section 2.4.3) in support of
formation of D2O in the reaction system. The GC-
Scheme 7.
ilar to those with the unlabeled acetone, gave the organ-
ic products A–E in comparable yields. Again, 2 equiv. of
2-propanol were obtained. Its 13C NMR spectrum
showed a strong doublet signal at d 63.5 ppm
(1JCH = 140 Hz), as expected for the methine carbon.
The other two compounds obtained in high yield, viz.,
B and C, displayed intense signals for what had been
the carbonyl carbon-13 atoms of the precursor ace-
tone-2-13C. These resonances were observed at d 209
and 71 ppm and at d 198 and 154 ppm and are assigned,
respectively, to the C@O and COH of B and the C@O
and @CMe2 of C.
1
MS-IR data also show some incorporation of H (from
(CD3)2CHOH) into B–E, as well as an increase in the
residual acetone-d5 signal, all probably occurring
through reversible addition/condensation and related
reactions with H/D scrambling. Also to be noted is
the absence of the hydroxy proton signal of 2-propanol
owing to exchange with deuterium.
No experiments were conducted with 18O-labeled ace-
tone, since they would not provide useful information.
Analysis of the organic products in this study has indi-
cated formation of H2O in the conversion of 1(OTf) to
2(OTf), and the oxygen of H2O was previously found
to exchange with that of 2 under the experimental con-
ditions employed [1]. Involvement of O2 can be ruled
out, since the reactions were conducted under argon
and proceed with good reproducibility. Besides, simple
calculations reveal that a very significant amount of
air would be needed in the reaction vessel to provide
the oxygen needed for conversion of 1(OTf) to 2(OTf)
in high yield.
Analogous results were obtained by using cyclohexa-
none-2,2,6,6-d4 in place of acetone-d6, either neat or in
benzene-d6 solution. Again, the methine proton of the
alcohol integrates to 1 equiv., and a corresponding sig-
2
nal is observed in the H NMR spectrum. Consistent
with the foregoing results, the GC analysis shows 2
equiv. of cyclohexanol. The GC-MS-IR data reveal for-
mation of the organic products I and J and, as for the
1
acetone-d6 reaction, some incorporation of H into the
organic products. The most important results of the
deuterium labeling experiments are summarized in
Scheme 7.
Isotopic labeling experiments were also conducted
using 13C. Acetone-2-13C was employed, since it was ex-
pected to provide information that would aid in the
characterization of the organic products. The reaction
of 1(OTf) with acetone-2-13C, under the conditions sim-
3.5. Reaction with other organooxygen compounds
In addition to ketones, a number of other organoox-
ygen compounds were used as potential sources of oxy-
gen in the conversion of 1(OTf) to 2(OTf). Initial
experiments with isobutyraldehyde showed considerable
decomposition of 1(OTf) and no detectable formation
of 2(OTf). To circumvent this behavior, attributed to

870
D.E. Rende, A. Wojcicki / Journal of Organometallic Chemistry 690 (2005) 862–873
4. Discussion
A balanced equation for the reaction of 1(OTf) with
ketones is presented in Scheme 9. The formation of
dihydrogen and alcohol according to the indicated stoi-
chiometry was established analytically, and the presence
of sufficient H2O in the reaction mixture was inferred
from the relative yield of the enone C, presumably
formed by dehydration of the keto-alcohol B. With the
exception of the alcohol derived from the starting ke-
tone, all characterized organic products are formed by
condensation and various related reactions. Apart from
the B-to-C transformation, these side reactions, which
may be catalyzed by Re complexes, do not appear to
have an effect on the conversion of 1 to 2 and are not
considered further.
On the basis of the results presented in Section 3 we
wish to propose the following mechanism for the reac-
tion of 1(OTf) with ketones (Scheme 10). A similar
mechanism would also apply to the reaction with isobu-
tyraldehyde and, probably, other aldehydes.
Scheme 8.
instability of the aldehyde to radical decomposition
upon heating, the stabilizer BHT [18] was used in subse-
quent reactions. Two control experiments were carried
out. In one, a solution of 1(OTf) and BHT in THF
was heated at ca. 60 °C. No reaction was observed. In
the other, 1(OTf), cyclohexanone, and BHT in THF
were allowed to react at 60 °C for 16 h, and the products
were analyzed. They were found to be identical with
those obtained when the reaction was performed in the
absence of BHT.
The first step (a) of the overall transformation ap-
pears to be loss of H2 from 1. Approximately 1 equiv.
of gaseous dihydrogen was collected and characterized
by MS. Some of the ‘‘missing’’ H2 could have been con-
sumed in the reduction of organic compounds generated
by side reactions (e.g., C to D); however, the amount
used up in this manner would not be significant. Also,
this experiment demonstrates that the dissociated dihy-
drogen was not involved in the reduction of the ketone
to the corresponding alcohol.
Heating a THF solution of 1(OTf), isobutyralde-
hyde, and BHT as in the experiment with cyclohexa-
none afforded 2(OTf) in high yield, as well as the
organic products K–N (Scheme 8) which were analyzed
by GC-MS-IR (cf. Section 2.5). The IR and MS data
are consistent with the assigned structures. The forma-
tion of the alcohol K and the vinyl ether L parallels the
formation of analogous products from 1(OTf) and
cyclohexanone. The ester M results from the dispro-
portionation of two molecules of isobutyraldehyde in
a side reaction. The ability of metal hydrides to cata-
lyze such a disproportionation of aldehydes has been
reported [19].
Since alcohols are observed during the reactions lead-
ing to the formation of 2(OTf), they were examined as
potential sources of oxygen for 1(OTf) in the absence
of a ketone. However, neither methanol nor cyclohexa-
nol was effective in that capacity. The epoxides 1,2-bu-
tene oxide and cyclohexene oxide also were ineffective.
With DMSO, 1(OTf) afforded uncharacterized Cyttp-
containing species, but no 2(OTf). A clean reaction oc-
curs between 1(OTf) and Me3NO; however, the latter
behaves solely as a proton abstractor to furnish Re-
H5(Cyttp). Finally, H2O without added ketone failed
to convert 1(OTf) to 2(OTf).
Dissociation of H2 from 1 is expected to result in the
formation of a 16-electron ‘‘[ReH4(Cyttp)]+’’ (step (a)),
which then can bind a solvent or ligand molecule, or
even triflate ion. Participation of a ketone or another
solvent/ligand in the displacement of dihydrogen (i.e.,
SN2 mechanism) is considered less likely than dissocia-
tion, since loss of H2 occurs essentially at the same tem-
perature for a variety of entering Lewis bases. Although
[ReH4(R2C@O)(Cyttp)]+ intermediates could not be de-
tected in the course of conversion of 1(OTf) to 2(OTf),
the related complexes [ReH4(MeCN)(Cyttp)]OTf
(3(OTf)) and ReH4(OTf)(Cyttp) (4) were prepared and
isolated as described in Section 3.2. Each reacts with
acetone or cyclohexanone to yield 2(OTf) and the
Scheme 9.

D.E. Rende, A. Wojcicki / Journal of Organometallic Chemistry 690 (2005) 862–873
871
Coordination of ketone to rhenium in ‘‘[Re-
H4(Cyttp)]+’’ leads to a migration of hydride to the
C@O carbon to give an alkoxide-hydride intermediate
(step (c)). Such a migratory insertion reaction is prece-
dented for complexes containing these two types of
ligand [22]. The next step (d) involves reductive elimina-
tion of the alkoxide formed in step (c) in conjunction
with another hydride ligand as a secondary alcohol,
R2CHOH; this reaction is assisted or immediately fol-
lowed by ligation of another molecule of ketone. Similar
reductive elimination reactions of alkoxide and hydride
are known [23]. It is to be noted that the R2CHOH re-
leased in step (d) contains the CH and OH hydrogen
that originated from Re-H. We have shown in this study
that the hydride ligands do not exchange with deuterium
when acetone-d6 or cyclohexanone-2,2,4,4-d4 is used for
the conversion of 1(OTf) to 2(OTf). Therefore, 1 equiv.
of the alcohol R2CHOH produced in the experiments
with deuterium-labeled ketones should contain C1H,
as was indeed observed. In contrast, the CO1H proton
of the alcohol, readily exchanges with the deuterium of
the D2O molecule that results from the condensation
of the deuteriated ketones. It should also be mentioned
that during the formation of R2CHOH in step (d),
reduction of the metal occurred from rhenium(V) to
rhenium(III).
The final part of the mechanistic pathway involves
conversion of a rhenium(III) species, [ReH2(R2C@O)-
(Cyttp)]+, to the rhenium(V) product
2
with
Scheme 10.
release of another molecule of alcohol. For the mecha-
nism to be consistent with the deuterium-labeling data,
the source of the CH hydrogen in that second equivalent
of R2CHOH must be the R groups of the ketone. Since
ketone condensation was inferred to occur during the
formation of 2(OTf), water seems to be the most plausi-
ble supplier of the alcohol hydrogens. We propose that a
second molecule of ketone is ligated to rhenium in an g2
mode. Such an attachment would be more favored in
[ReH2(R2C@O)(Cyttp)]+ than in [ReH4(R2C@O)-
(Cyttp)]+ owing to a lower oxidation number and a re-
duced coordination number of the metal in the former.
Intermediate [ReH2(R2C@O)(Cyttp)]+ then undergoes
nucleophilic attack by H2O (step (e)) at the carbonyl
carbon which results in cleavage of the ketonic car-
bon-oxygen linkage. Protolysis of the Re–C bond would
afford a second molecule of alcohol and 2 (step (f)). It is
to be noted that this pathway does not lead to a scram-
bling of the H2O and Re–H hydrogens. Thus, the gener-
ated alcohol should contain deuterium in the methine
and, initially, hydroxy positions, with both atoms deriv-
ing exclusively from D2O, when acetone-d6 or cyclo-
hexanone-2,2,6,6-d4 is employed in the reaction. In
contrast, any water protons participating in bond for-
mation with the rhenium center would be expected to
lose their identity through scrambling with hydride lig-
ands, since high-coordination number transition-metal
appropriate alcohol; however, some of these reactions
are not as clean as those of 1(OTf), possibly owing to
decomposition of 3(OTf) or 4. Nonetheless, over longer
reaction times and/or under more forcing conditions,
comparable yields of 2(OTf) from 3(OTf) as from
1(OTf) were achieved. In a related experiment, when a
mixture of 1(OTf), MeCN, and acetone was heated un-
der the conditions that mirror those for the complete
conversion of 1(OTf) to 2(OTf) by acetone, both
2(OTf) and 3(OTf) were produced. Longer heating or
higher temperatures afforded only 2(OTf). These results
support competition among potential donors for coordi-
natively unsaturated ‘‘[ReH4(Cyttp)]+,’’ with the rela-
tively good ligand acetone, present in excess,
eventually driving the reaction to completion by forming
a spontaneously reactive [ReH4(Me2C@O)(Cyttp)]+.
The ketone in [ReH4(R2C@O)(Cyttp)]+ can be either
g1-bonded through the oxygen or g2-bonded through
the C@O. Both modes of attachment are known for a
large number of reported transition-metal ketone com-
plexes [9,20], and, in some cases, equilibrium exists be-
tween the two [21]. Owing to the robust nature of the
Cyttp ligand, it is assumed here that the ketone is coor-
dinated to rhenium in an g1 fashion. However, g2 bond-
ing would not alter the proposed steps of this part of the
mechanism.

872
D.E. Rende, A. Wojcicki / Journal of Organometallic Chemistry 690 (2005) 862–873
[2] Y. Kim, D.E. Rende, J.C. Gallucci, A. Wojcicki, J. Organomet.
Chem. 682 (2003) 85.
hydrides are susceptible to rapid intramolecular rear-
rangements [24].
[3] (a) A. van Asselt, B.J. Burger, V.C. Gibson, J.E. Bercaw, J. Am.
Chem. Soc. 108 (1986) 5347;
Other mechanisms were considered for the formation
of 2. Since equilibria between ketones and H2O on the
one side and gem-diols on the other are rapidly estab-
lished in solution [25], the latter also seemed to be poten-
tial sources of a second molecule of alcohol. However, it
is not apparent to us how a gem-diol intermediate,
[ReH2(R2C(OH)2)(Cyttp)]+, would generate 2 and R2-
CHOH. Another mechanism, involving g2-ketone coor-
dination in step (b) and/or (d) followed by spontaneous
cleavage of the C@O bond appears improbable. Rhe-
nium(V) or rhenium(III) complexes, unlike tungsten(II)
complexes [9], are too electron-poor to undergo such a
reaction. Besides, that pathway would likely produce
alkenes derived by coupling of the alkylidene groups
CR2 from R2C@O and/or hydrogenation products of
the CR2. Such compounds were not observed among
the organics isolated in this study. Any viable mecha-
nism must also account for the observation that only ke-
tones and aldehydes convert 1 to 2; other organooxygen
sources such as alcohols, epoxides, DMSO, and Me3NO
are completely ineffective.
Carbonyl compounds that participate in the reaction
in Scheme 5 include, besides acetone and cyclohexa-
none, also cyclopentanone, 2-butanone, 3-pentanone,
and the aldehyde isobutyraldehyde. All of them generate
H2O by condensation. Ketones that do not undergo
condensation, viz., 2,4-dimethyl-3-pentanone, 2,6-
dimethylcyclohexanone, and 2-adamantanone, failed to
yield 2. It appears that the inability to produce H2O
and the robust nature of the latter three compounds
are the important factors.
(b) G. Parkin, A. van Asselt, D.J. Leahy, L. Whinnery, N.G.
Hua, R.W. Quan, L.M. Henling, W.P. Schaefer, B.D. Santasiero,
J.E. Bercaw, Inorg. Chem. 31 (1992) 82.
[4] (a) H. Brunner, M.M. Kubicki, J.-C. Leblanc, C. Mo¨ıse, F.
Volpato, J. Chem. Soc., Chem. Commun. (1993) 851;
(b) P. Sauvageot, A. Sadorge, B. Nuber, M.M. Kubicki, J.-C.
Leblanc, C. Mo¨ıse, Organometallics 18 (1999) 2133.
[5] (a) E. Spaltenstein, T.K.G. Erikson, S.C. Critchlow, J.M. Mayer,
J. Am. Chem. Soc. 111 (1989) 617;
(b) R.R. Conry, J.M. Mayer, Organometallics 10 (1991) 3160.
[6] S.H. Tahmassebi, R.R. Conry, J.M. Mayer, J. Am. Chem. Soc.
115 (1993) 7553.
[7] (a) Y. Matano, S.N. Brown, T.O. Northcutt, J.M. Mayer,
Organometallics 17 (1998) 2939;
(b) Y. Matano, T.O. Northcutt, J. Brugman, B.K. Bennett, S.
Lovell, J.M. Mayer, Organometallics 19 (2000) 2781.
[8] D.E. Rende, Y. Kim, C.M. Beck, A. Wojcicki, Inorg. Chim. Acta
240 (1995) 435.
[9] J.C. Bryan, J.M. Mayer, J. Am. Chem. Soc. 112 (1990) 2298.
[10] D.F. Shriver, M.A. Drezdzon, The Manipulation of Air-sensitive
Compounds, second ed., Wiley, New York, 1986.
[11] D.D. Perrin, W.L.F. Armarego, D.R. Perrin, Purification of
Laboratory Chemicals, Pergamon Press, Oxford, 1991.
[12] R.R. Willis, M. Calligaris, P. Faleschini, J.C. Gallucci, A.
Wojcicki, J. Organomet. Chem. 193–194 (2000) 465.
[13] M.W. Baize, J.C. Gallucci, A. Wojcicki, Inorg. Chim. Acta 259
(1997) 339.
[14] Y. Kim, J. Gallucci, A. Wojcicki, Organometallics 11 (1992)
1963.
[15] Y. Kim, H. Deng, J.C. Gallucci, A. Wojcicki, Inorg. Chem. 35
(1996) 7166.
[16] G.A. Lawrance, Chem. Rev. 86 (1986) 17.
[17] A. Streitwieser, C.H. Heathcock, E.M. Kosower, Introduction to
Organic Chemistry, fourth ed., Macmillan Publishing Co., New
York, 1992, pp. 389–398.
[18] E.T. Denisov, I.V. Khudyakov, Chem. Rev. 87 (1987) 1313.
[19] (a) T. Ito, H. Horino, Y. Koshiro, A. Yamamoto, Bull. Chem.
Soc. Jpn. 55 (1982) 504;
Finally, it is of interest to note that cyclohexanone re-
acts faster than acetone with 1(OTf) under comparable
conditions. This difference in reactivity probably results
from the relief of ring strain in cyclohexanone on addition
of hydride to give cyclohexanol (step (c)). Cyclohexanone
is over 10 times more reactive than acetone toward the
nucleophiles CNꢁ, HOCH2CH2Sꢁ, and SO23ꢁ [26].
(b) G.A. Slough, J.R. Ashbaugh, L.A. Zannoni, Organometallics
13 (1994) 3587.
[20] Selected examples: (a) Y.-H. Huang, J.A. Gladysz, J. Chem.
Educ. 65 (1988) 298;
(b) R.H. Crabtree, G.G. Hlatky, C.P. Parnell, B.E. Segmuller,
¨
R.J. Uriarte, Inorg. Chem. 23 (1984) 354;
(c) J.L. Caldarelli, L.E. Wagner, P.S. White, J.L. Templeton,
J. Am. Chem. Soc. 116 (1994) 2878;
(d) R.L. Cicero, J.D. Protasiewicz, Organometallics 14 (1995)
4792;
Acknowledgements
(e) L.E. Helberg, T.B. Gunnoe, B.C. Brooks, M. Sabat, W.D.
Harman, Organometallics 18 (1999) 573, and references therein;
(f) S.H. Meiere, W.D. Harman, Organometallics 20 (2001)
3876.
The authors acknowledge the donors of The Petro-
leum Research Fund, administered by the ACS, for par-
tial support of this research. We also thank Dr. Gordon
Renkes for help with GC-MS-IR analyses.
[21] (a) W.D. Harman, M. Sekine, H. Taube, J. Am. Chem. Soc. 110
(1988) 2439;
´
(b) D.P. Klein, D.M. Dalton, N.Q. Mendez, A.M. Arif, J.A.
Gladysz, J. Organomet. Chem. 412 (1991) C7;
(c) D.W. Powell, P.A. Lay, Inorg. Chem. 31 (1992) 3542;
´
(d) N.Q. Mendez, J.W. Seyler, A.M. Arif, J.A. Gladysz, J. Am.
References
Chem. Soc. 115 (1993) 2323.
[22] (a) R.A. Sanchez-Delgado, O.L. DeOchoa, J. Mol. Catal. 6 (1979)
[1] (a) Y. Kim, J. Gallucci, A. Wojcicki, J. Am. Chem. Soc. 112
(1990) 8600;
303;
(b) R.A. Grey, G.P. Pez, A. Wallo, J. Am. Chem. Soc. 103 (1981)
7536.
(b) Y. Kim, D.E. Rende, J.C. Gallucci, A. Wojcicki, Inorg. Chim.
Acta 352 (2003) 171.

D.E. Rende, A. Wojcicki / Journal of Organometallic Chemistry 690 (2005) 862–873
873
[23] (a) L.J. Newman, R.G. Bergman, J. Am. Chem. Soc. 107 (1985)
5314;
[25] (a) M.G. Hutchings, J. Gasteiger, J. Chem. Soc., Perkin Trans. II
(1986) 455;
(b) D. Milstein, J. Am. Chem. Soc. 108 (1986) 3525.
[24] R.H. Crabtree, in: G. Wilkinson, R.D. Gillard, J.A. McCleverty
(Eds.), Comprehensive Coordination Chemistry, 2, Pergamon
Press, Oxford, 1987 (Chapter 19).
(b) K.B. Wiberg, K.M. Morgan, H. Maltz, J. Am. Chem. Soc.
116 (1994) 11067.
[26] F.A. Carey, R.J. Sundberg, Advanced Organic Chemistry, third
ed., Plenum Press, New York, 1990 (Chapter 8).