Perhydrocarbyl ReVII complexes: Comparison of molecular and surface complexes

Article abstract of DOI:10.1021/ja020136s

The molecular complex [Re(≡C^tBu)(=CH^tBu) (CH₂^tBu)₂] (1) reacts with a silica partially dehydroxylated at 700 °C to give syn-2,[(≡SiO)Re(≡C^tBu)(=CH^tBu) (CH₂^t

Full text of DOI:10.1021/ja020136s

Published on Web 12/12/2002
Perhydrocarbyl ReVII Complexes: Comparison of Molecular
and Surface Complexes
Mathieu Chabanas,† Anne Baudouin,† Christophe Cope´ret,*,† Jean-Marie Basset,*,†
Wayne Lukens,‡ Anne Lesage,§ Sabine Hediger,§ and Lyndon Emsley*,§
Contribution from the Laboratoire de Chimie Organome´tallique de Surface, UMR 9986 CNRS -
ESCPE Lyon, 43 bd du 11 NoVembre 1918, F-69626 Villeurbanne Cedex, France,
Chemical Sciences DiVision, Lawrence Berkeley National Laboratory, Berkeley, California
94720, and Laboratoire de Ste´re´ochimie et des Interactions Mole´culaires, (UMR 5532 CNRS/
ENS), Laboratoire de Recherche ConVentionne´ du CEA (23V), Ecole Normale Supe´rieure de
Lyon, 46 Alle´e d’Italie, 69364 Lyon Cedex 07, France
Received January 25, 2002 ; E-mail: basset@cpe.fr, coperet@cpe.fr
t
Abstract: The molecular complex [Re(tCtBu)(dCHtBu)(CH2 Bu)2] (1) reacts with a silica partially dehy-
t
droxylated at 700 °C to give syn-2, [(tSiO)Re(tCtBu)(dCHtBu)(CH2 Bu)], as a single isomer according to
mass-balance analysis, IR, and solid-state NMR spectroscopy. 1D and 2D solid-state NMR (HETCOR and
long-range HETCOR) on a 13C-labeled-2 has allowed us to observe the chemical shifts of all carbons
(including those that are not labeled) and ascertain their assignments. Moreover, EXAFS data are consistent
with the presence of two carbons at a relatively short distance (1.79 Å), which cannot be deconvoluted, but
which are consistent with the presence of alkylidene and alkylidyne carbons along with two other first
neighbors at a longer distance (2.01 Å), the alkyl carbon and the O atom by which the Re is attached to
the surface. Moreover, the data also suggest the presence of a siloxane bridge of the silica surface at 2.4
Å in the coordination sphere of the Re center. Thermal and photochemical treatment allow us to observe
the anti isomer, which was also fully characterized by 1D and 2D solid-state NMR. This behavior parallels
1
the reactivity of molecular Re complexes, and their respective H and 13C chemical shifts match those of
the corresponding molecular analogues syn- and anti-2m and n. Finally, the grafting of 1 onto silica involves
the reaction of both the alkyl and the alkylidene ligand with an equiprobability, leaving the alkylidyne as a
t
spectator ligand. Noteworthy is the formation of 2 [(tSiO)Re(tCtBu)(dCHtBu)(CH2 Bu)], rather than the
corresponding trisneopentyl-neopentylidyne Re complex, monografted on silica, [(tSiO)Re(tCtBu)(CH2-
tBu)3], which would have been expected from the reactivity of 1 with various molecular Bro¨nsted acids and
which also suggests that a proximal siloxane bridge forces the R-H abstraction process, leading to syn-2a.
Scheme 1. Grafting of 1a onto Silica Partially Dehydroxylated at
700 °C
Introduction
In our continuous effort to study the interaction of organo-
metallic complexes with oxide supports such as silica, we have
recently disclosed the outcome of the reaction of 1a [Re(tC-
tBu)(dCHtBu)(CH2 Bu)2]1 with a silica partially dehydroxylated
t
at 700 °C, that is, the formation of a well-defined surface
alkylidene complex obtained as a single isomer: [(tSiO)Re-
(tCtBu)(dCHtBu)(CH2 Bu)] (syn-2a) (Scheme 1).2 The relative
t
and rather unexpected simplicity of the coordination sphere
obtained on a silica support is worth pointing out because
organometallic Re complexes have a large palette of potential
reactivities and structural isomerisms; it is therefore of great
interest to analyze the differences and similarities between 2a
and molecular Re complexes. We decided (i) to further
investigate the structural identity of 2a and its stability, and (ii)
to understand the mode - mechanism - of grafting of 1a onto
silica to compare the reactivity of organometallic complexes
with surfaces and that of molecular complexes in solution.
* To whom correspondence should be addressed.
† Laboratoire de Chimie Organome´tallique de Surface.
‡ Lawrence Berkeley National Laboratory.
Results and Discussion
§ Laboratoire de Ste´re´ochimie et des Interactions Mole´culaires.
(1) For the preparation of 1a, see: (a) Edwards, D. S.; Biondi, L. V.; Ziller,
J. W.; Churchill, M. R.; Schrock, R. R. Organometallics 1983, 2, 1505.
(b) Toreki, R.; Schrock, R. R.; Davis, W. M. J. Am. Chem. Soc. 1992,
114, 3367.
Mass Balance Analysis. The preparation of 2a involves the
reaction of 1a in solution in pentane for 2 h with a silica partially
dehydroxylated at 700 °C (SiO2-(700)). Typically Re is grafted
in 3.9-4.7%wt depending on the batch of silica used (Table 1),
(2) Chabanas, M.; Baudouin, A.; Cope´ret, C.; Basset, J.-M. J. Am. Chem. Soc.
2001, 123, 2062.
9
492
J. AM. CHEM. SOC. 2003, 125, 492-504
10.1021/ja020136s CCC: $25.00 © 2003 American Chemical Society

Perhydrocarbyl ReVII Complexes
A R T I C L E S
Table 1. Grafting of 1a onto Silica Partially Dehydroxylated at 700
°C
Scheme 2. Some Hypothetical Structures for the Surface Species
Formed upon Reaction of 1a with SiO2-(700)
a
NpH/Re
(grafting)b
C/Re
Np#/Red
(hydrogenolysis)
a
c
impregnation
% Rea
% C
(Np#/Re )
w
w
1
2
3
4
5
6
4.71
4.16
4.24
4.03
4.00
4.75
3.95
3.52
3.50
1.03
4.54
4.55
15.0 (3.00)
17.0 (3.40)
1.10
1.08
1.08
0.79
1.07
1.27
1.03
2.95
2.83
2.89
3.74
4.55
4.01
3.51
3.33
14.9 (2.98)
15.8 (3.16)
15.5 (3.09)
14.8 (2.96)
7
8e
9e
a Elemental analysis. b Neopentane (NpH) released during the grafting,
quantified by GC. c Number of neopentyl-like ligands (Np#) present on
average in the coordination sphere of Re, determined by elemental analysis.
d Average number of neopentyl-like ligands (Np#) around Re, determined
by quantifying CH4 produced during the hydrogenolysis of the solid, Np#/
Re ) (CH4/Re)/5. e Surface not saturated due to a lack of 1a in the pentane
solution.
which corresponds to the consumption of 0.21-0.26 mmol/g
of silanol groups and to the amount of silanols accessible for
bulky complexes on a silica partially dehydroxylated at 700 °C.3
Moreover, the amount of neopentane evolved during grafting
is on average 1.06 ( 0.13 mol per grafted Re on the surface, in
agreement with the replacement of one neopentyl ligand of 1a
by a siloxy ligand of the surface. The corresponding grafted
organometallic compound contains 15.5 ( 0.8 carbons per
grafted Re according to elemental analysis, that is, the equivalent
of 3.1 ( 0.2 “neopentyl-like” ligands (Np#).4 Additionally, the
hydrogenolysis of 2a at 250 °C for 2-4 days liberated 14.5 (
0.3 methane/Re along with the complete disappearance of typical
alkyl stretching frequencies in the 2800-3000 cm-1 region
according to in situ-IR spectroscopy. This is accompanied with
the partial recovery of silanol groups, which probably implies
that some Re-O bonds with the support are cleaved under these
conditions to form Re particles. Nonetheless, the amount of
methane formed can be related to the number of carbons or
“neopentyl-like” ligands around the Re center, that is, an average
of 2.9 ( 0.1 Np#/Re, which is in close agreement with elemental
analysis data. Therefore, this set of data is consistent with the
loss of approximately 1 equiv of neopentane during the grafting,
leaving on average three “neopentyl-like” ligands around the
metal center. On the basis of these data alone, it is possible to
propose several surface complexes, as well as mixtures of
surface complexes (Scheme 2).
a Structures involving three bonds to the surface are omitted because
the probability of forming such a species on SiO2-(700) is very low.
Figure 1. 1H-13C HETCOR solid-state NMR spectra of the Re surface
species 2b using a contact time for the cross-polarization of 0.5 ms.
t
assigned to CH3, CH2 Bu, and dCHtBu resonances, respec-
1D and 2D Solid-State NMR Spectroscopy. We have
already reported the H NMR solid-state NMR of 2, which
tively. An extra signal was observed at 292 ppm, when using
direct carbon excitation, which was attributed to the (tCtBu)
resonance (vide infra Figure 4b). These assignments were
exclusively based on the comparison with chemical shifts
obtained from a molecular analogue 2m [Ph3SiORe(tCtBu)-
1
shows signals at 11.0 and 1.0 ppm along with two other signals
of less intensity at 3.0 and 2.6 ppm.2 At natural abundance, the
sensitivity of the 13C CP/MAS experiment is too low to adjust
properly the experimental parameters of the CP and to obtain a
signal within a reasonable experimental time. However, we have
shown that it is possible to characterize the carbon spectrum
by labeling 2a to the extent of 10% on all carbons in the
R-position to the metal center (2b). Signals in the carbon CP/
MAS spectrum at 29, 46, and 246 ppm were observed and
t
(dCHtBu)(CH2 Bu)] prepared by the reaction of triphenylsilanol
and 1a (vide infra Scheme 4, Table 2, and 4-5). Nevertheless,
it would be desirable to obtain more direct evidence that would
confirm these assignments. Recently, we have shown that 2D
HETCOR (heteronuclear correlation) solid-state NMR using
magic angle spinning (MAS) can be used in the same way as
solution-state experiments to clearly ascertain chemical shift
assignments.5 The 2D 1H-13C HETCOR MAS NMR spectrum
(3) Chemical titration and numerous data obtained for the grafting of other
perhydrocarbyl complexes of Ti (1.3%wt, 0.27 mmol/g), Zr (2.3%wt, 0.25
mmol/g), Hf (4.2% wt, 0.24 mmol/g), Ta (4.2% wt, 0.24 mmol/g), and Mo
(2.1%wt, 0.22 mmol/g) onto SiO2-(700) show the presence of around 17-
(5) (a) Petroff Saint-Arroman, R.; Chabanas, M.; Cope´ret, C.; Basset, J.-M.;
Lesage, A.; Emsley, L. J. Am. Chem. Soc. 2001, 123, 3082. (b) Chabanas,
M.; Quadrelli, A.; Cope´ret, C.; Thivolle-Cazat, J.; Basset, B.; Fenet, J.-
M.; Lesage, A.; Emsley, L. Angew. Chem., Int. Ed. 2001, 43, 4493.
0.27 mmol of accessible SiOH groups per gram of SiO2-(700)
.
(4) By “neopentyl-like”, we mean either neopentyl-, neopentylidene-, or
neopentylidyne-type ligands.
9
J. AM. CHEM. SOC. VOL. 125, NO. 2, 2003 493

A R T I C L E S
Chabanas et al.
Scheme 3. Proposed Structure Combining Solid-State NMR and
EXAFS Data
correlations (b and c) with protons at about 2.6 and 3.2 ppm,
t
which correspond to the two diastereotopic protons (CHAHB -
1
Bu), previously reported in the 1D H spectrum at 2.6 and 3.1
ppm. Finally, a correlation peak (d) can be observed between
the proton at 11.1 ppm and the carbon at 246 ppm, confirming
the assignment of these signals to the carbenic proton and
carbon, respectively (dCHtBu). Note that, in this case, the use
of a contact time of 0.5 ms allows the selective observation of
C-H nuclei which are spatially very close (i.e., directly bonded).
Under these conditions of relatively short contact time, a solid-
state HETCOR experiment can be interpreted much like
analogous HETCOR experiments in solution NMR (also called
H-C COSY or HSQC), although the pulse sequence and
magnetization transfer mechanism involved in these two se-
quences are completely different (the transfer is mediated by
through-space dipolar interactions in solids versus through-bond
scalar couplings in liquids).6 Using longer contact times (>1
ms), we found that it is possible to observe extra correlation
peaks, which arise from longer range dipolar through-space
interactions. These long-range spectra yield further information
1
about the structure of 2a. The H-13C HETCOR spectrum of
2b was therefore recorded with a contact time of 5 ms (Figure
2a,b). The expanded view of the aliphatic region in Figure 2b
shows that the correlation peak called “a” appeared in this
spectrum as two well-separated components a1 (1.4 ppm; 30
ppm) and a2 (1.6 ppm; 28 ppm). In fact, the correlation peak a1
corresponds to two correlations, but the resolution and the
chemical shift difference are not sufficient to clearly identify
each of them. The correlation peak g (11.1 ppm; 44 ppm)
corresponds to the interaction of the carbenic proton with a
carbon at 44 ppm, which must be relatively close, probably the
quaternary carbon of the neopentylidene ligand (dCHC(CH3)3).
This carbon at 44 ppm also correlates with a proton at around
1.4 ppm (h), which could therefore be assigned to the methyl
resonance of the neopentylidene ligand (dCHC(CH3)3). The
signal in F1 around 1.4 ppm also corresponds to the methyl
resonance of the neopentyl ligand (CH2C(CH3)3) because it
Figure 2. 1H-13C HETCOR solid-state NMR spectra of the Re surface
species 2b using a contact time for the cross-polarization of 5 ms: (a) Full
spectrum. (b) Expanded region.
t
correlates (peak i) with the carbon at 46 ppm (CH2 Bu).
Therefore, the correlation peak a1 probably corresponds to the
two methyl groups of both the neopentyl and the neopentylidene
ligands. As a consequence, a2 (1.6 ppm, 28 ppm) could be
assigned to the methyl group of the neopentylidyne ligand
(tCC(CH3)3). The corresponding proton resonance at 1.6 ppm
((tCC(CH3)3) clearly correlates with a sharp signal at 53 ppm
(peak j), probably the quaternary carbon of the neopentylidyne
ligand (tCC(CH3)3). Noteworthy is also the correlation (peak
k) of only one of the diastereotopic protons at 3.1 ppm with a
carbon at 30 ppm, which can be assigned to either a carbon of
a methyl group (CH2C(CH3)3) or, more likely, the quaternary
carbon R to the methylene (CH2C(CH3)3), as observed on the
neopentylidene and neopentylidyne ligands. The other two
Figure 3. Re LIII-edge k3-weighted EXAFS (left) and Fourier transform
(right) of 2a: gray lines, experimental; black lines, spherical wave theory.
of 2b recorded with a contact time of 0.5 ms (Figure 1) shows
a correlation (a) between protons around 1.4 ppm and carbons
around 29 ppm. Moreover, the 13C resonance of the methylene
(6) Derome, A. E. Modern NMR Techniques for Chemistry Research; Organic
t
group of the neopentyl ligand (CH2 Bu, 45-46 ppm) gives two
Chemistry Series Vol. 6; Pergamon Press: Oxford, 1987.
9
494 J. AM. CHEM. SOC. VOL. 125, NO. 2, 2003

Perhydrocarbyl ReVII Complexes
A R T I C L E S
Table 2. Discernible 1H and 13C NMR Resonances for the
Surface Compound syn-2b
[OCMe(CF3)2](THF).9 Therefore, these data are most consistent
with the presence of an extra siloxane bridge as a two-electron
donor ligand.10 The large Debye-Waller factor suggests that
either only some of the surface complexes possess this bond
or, more likely, a large variation of (Re-O) bond lengths is
present (heterogeneity of the silica surface). The final component
is due to scattering from the tertiary carbon atoms of the ligands.
Additional scattering shells due either to multiple scattering or
to scattering from Si in the support do not significantly improve
the fit of the model to the data. Overall, the EXAFS data are in
excellent agreement with the structure of 2 with an additional
dative bond from a siloxane bridge in the silica support as shown
in Scheme 3.
a
1H NMR, δ/ppm
13C NMR, δ/ppm
resonances
assignments
resonances
assignments
1.1
1.3
C(CH3)3)
C(CH3)3)
28
30
tCC(CH3)3)
C(CH3)3)
1.5
2.6
3.1
11.1
tCC(CH3)3)
30
44
46
53
246
292
CH2C(CH3)3)
dCHC(CH3)3)
CH2 Bu
tCC(CH3)3)
dCHtBu
tCtBu
t
CHAHB Bu
t
t
CHAHB Bu
dCHtBu
a Some data are assigned via 2D HETCOR solid-state NMR.
Table 3. EXAFS Parameters for 2aa
Syn-Anti Isomerism, A Comparison with Molecular
Complexes. Rotational isomers (rotamers) are common in the
organometallic chemistry of alkylidene complexes. The alky-
lidene substituent can indeed be pointed either toward (syn)
another multiply bonded ancillary ligand (imido, alkylidyne, etc.)
or away (anti) from it. Some alkylidene organometallic syntheses
yield only one rotamer (usually the syn one), which can be
generally converted into a mixture of the syn and the anti
rotamers either thermally or photochemically. An interesting
example is the molecular complex syn-[Re(tCtBu)(dCHtBu)-
(ORF)2],1b whose chemical structure is quite similar to that of
2a. The reaction of 1a with 1 equiv of Ph3SiOH in benzene at
room temperature quantitatively gives neopentane and 2m [Ph3-
distance
(Å)b
Debye−Waller
factor (Å )
2 b
element
# of atoms
assignment
C
2
1.789(4)
0.0125(4)
RetCC(Me)3
RedCHC(Me)3
Re-OSit
Re-CH2C(Me)3
Rer(OSi2)
RetCC(Me)3
RedCHC(Me)3
Re-CH2C(Me)3
O
C
O
1
1
1
2.015(3)
2.015c
2.420(8)
0.0056(2)
0.0056c
0.012(1)
C
3
3.286(5)
0.0044(4)
a E0 for all shells is -6.9(5) eV. b Number in parentheses is the standard
deviation of the parameter in the fit. c Parameter linked to the preceding
shell.
t
Si-O-Re(tCtBu)(dCHtBu)(CH2 Bu)] (see Supporting Infor-
t
methyl groups from the Bu fragments of the neopentyl and
neopentylidene ligands cannot be fully assigned under these
conditions. Nevertheless, combining the information given by
1D and 2D solid-state NMR spectroscopy provides a relatively
complete NMR data set for the surface compound 2a. All of
these data are fully consistent and indicate that the reaction of
1a with SiO2-(700) leads to the formation of [(tSiO)-Re(tC-
mation). Very similar NMR data were obtained for another
molecular model 2n [(c-C5H9)7Si7O12Si-O-Re(tCtBu)(dCHt-
t
Bu)(CH2 Bu)], prepared by the reaction of 1a with the poly-
oligomeric silsesquioxane 3 [(c-C5H9)7Si7O12Si-OH]. These
complexes are rather unstable as compared to their surface
complex analogue, which has so far prevented us from obtaining
suitable crystals for X-ray analysis (slow decomposition at room
temperature in our hands). Additionally, they are obtained as a
10/1 mixture of syn and anti rotamers along with about 1 equiv
of neopentane (Scheme 4). The surface complex 2a has chemical
shift values very close to those obtained for syn isomers, which
further points out that 2a is obtained as the syn isomer, as the
sole surface species.
tBu)(dCHtBu)(CH2 Bu)], 2a, as the sole surface species (Scheme
t
1 and Table 2).
EXAFS Data. EXAFS data collected from the Re surface
complex 2a (Figure 3) are consistent with the NMR data with
one difference: a dative bond from a siloxane bridge can also
be observed. The parameters derived from fitting the EXAFS
data (Table 3) describe the local environment of the rhenium
center. The distances to the two nearest carbon neighbors,
corresponding to the (dCHtBu) and (tCtBu) ligands, cannot
be resolved with the data range available. The large Debye-
Waller factor for this shell is consistent with this assignment.
When the surface complex syn-2b was heated at 120 °C under
Ar during 30 min, typical signals for syn-2b were still present
1
in both H and 13C spectra, but a new signal at 12.6 ppm (in
the carbenic proton range) was observed in the MAS 1H NMR
(Figure 4c), and extra peaks at 68, 257, and 302 ppm were also
detected by direct excitation solid-state 13C NMR spectroscopy
(Figure 4d). The extra signal at 12.6 ppm in the MAS 1H NMR
spectrum matches the carbenic proton signal of the anti isomer
of the molecular models (12.65 ppm for anti-2m and 12.63 ppm
for anti-2n, Figure 5 (and Supporting Information) for com-
parison). Moreover, the extra signals at 257 and 302 ppm in
13C solid-state NMR also match, respectively, the carbenic and
the carbynic signals of the anti isomers of the molecular models,
providing good evidence for the partial isomerization of the
surface complex syn-2b into its anti rotamer, anti-2b.11,12
t
Similarly, the (-OSit) and -CH2 Bu cannot be resolved and
were fit using two shells with the same parameters. The 1.79 Å
distance for the first shell is consistent with an average of Re-C
distances in alkylidene and alkylidyne complexes of ReVII
:
1.85-1.89 C and 1.74-1.76 Å, respectively,7 and the 2.01 Å
distance for the second shell is consistent with an average of
Re-O and Re-C distances in ReVII complexes: 1.90-2.00 Å7
and 2.11 Å,8 respectively. Moreover, an extra atom, oxygen, is
also present in the coordination sphere of Re, at a longer Re-O
distance (2.42 Å), which is most likely due to a dative bond
from an adjacent siloxane bridge because it is similar to that of
the Re-OC4H8 (THF, 2.398 Å) in syn-Re(tCtBu)(dCHtBu)-
(9) Schofield, M. H.; Schrock, R. R.; Park, L. Y. Organometallics 1991, 10,
1844. See also refs 1a and 7d.
(10) Similar observations have been made on other grafted organometallic
complexes, see: Corker, J.; Lefebvre, F.; Le´cuyer, C.; Dufaud, V.; Quignard,
F.; Choplin, A.; Evans, J.; Basset, J.-M. Science 1996, 271, 966. Vidal,
V.; The´olier, A.; Thivolle-Cazat, J.; Basset, J.-M.; Corker, J. J. Am. Chem.
Soc. 1996, 118, 4595.
(7) For Re-O, RedC, and RetC bond distances, see: (a) ref 1. (b) Cai, S.;
Hoffman, D. M.; Wierda, D. A. J. Chem. Soc., Chem. Commun. 1988,
1489. (c) Toreki, R.; Schrock, R. R.; Vale, M. G. J. Am. Chem. Soc. 1991,
113, 3610. (d) Toreki, R.; Vaughan, G. A.; Schrock, R. R.; Davis, W. M.
J. Am. Chem. Soc. 1993, 115, 127.
(8) For a Re-C bond distance, see ref 7b.
9
J. AM. CHEM. SOC. VOL. 125, NO. 2, 2003 495

A R T I C L E S
Chabanas et al.
1
Figure 4. MAS H and HPDEC 13C solid-state NMR spectra of 2b before ((a) and (b)) and after ((c) and (d)) heating at 120 °C under argon for 30 min.
Spectra recorded at 25 °C.
Scheme 4. Reaction of 1a with Triphenylsilanol and 3 To Give syn- and anti-2n and -2m, Respectively
These assignments were further confirmed by 2D HETCOR
solid-state NMR (Figure 6). Integration of the carbenic proton
peaks over several experiments (120 °C, 30 min) gave a syn/
anti ratio of approximately 2.0 ( 0.5. Almost the same NMR
spectrum with a measured syn/anti ratio of 2.0 was obtained
for a sample treated for a longer time (1 h 25 min) at 120 °C.
9
496 J. AM. CHEM. SOC. VOL. 125, NO. 2, 2003

Perhydrocarbyl ReVII Complexes
A R T I C L E S
t
Figure 5. Direct comparison of NMR data for syn and anti rotamers of both the surface complex 2b [(tSiO)-Re(tCtBu)(dCHtBu)(CH2 Bu)] (i) and the
t
1
molecular model 2m [Ph3SiO)-Re(tCtBu)(dCHtBu)(CH2 Bu)] (ii). (a) H NMR (carbenic region). (b) 13C NMR (carbenic and carbynic region).
Scheme 5. Equilibrium between the Syn and the Anti Rotamers of
the Surface Compound 2a
25 °C over a 12-48 h period according to 1H NMR. Similarly,
the syn/anti ratio measured at 25 °C for the molecular model
2m (reaction of 1a with Ph3SiOH at 25 °C) did not vary, even
after prolonged storage in the dark at 25 °C (syn/anti ) 10).
All of these data are consistent with a very slow transformation
of the syn to the anti rotamer of 2a at 25 °C. The isomerization
becomes observable at higher temperatures (Scheme 5). There-
fore, the syn/anti ratio of 2.0 ( 0.5 measured at 25 °C after
treatment at 120 °C (67 ( 7% syn, 33 ( 7% anti) corresponds
probably to the thermodynamic equilibrium at 120 °C. ∆G° of
this transformation can be measured to be +2 ( 1 kJ/mol at
Figure 6. 1H-13C HETCOR solid-state NMR spectra of the Re surface
120 °C.
species syn- and anti-2b using a contact time for the cross-polarization of
Interestingly, storage of 2a for 24 h at 25 °C in a Pyrex
Schlenk tube exposed to daylight also induced syn-anti
0.5 ms. * notes the correlation peaks with spinning side bands.
isomerization. After this treatment, approximately 40% of syn-
2a was converted into anti-2a (syn/anti ) 1.52). In conclusion,
2a can be indeed present as two different rotational isomers, as
for analogous molecular complexes, but is generated selectively
as the syn rotamer upon grafting. It is probably the kinetic
product even though our data do not allow us to certify it.
Investigation of the Grafting Mechanism. Because three
types of perhydrocarbyl ligands (neopentyl, neopentylidene, and
neopentylidyne) are present in the molecular precursor 1a,
several mechanisms for the grafting could in principle occur
(Scheme 6): (i) direct cleavage of the Re-C bond of the
neopentyl ligand by a silanol O-H bond (electrophilic cleav-
age), (ii) addition of a silanol O-H bond onto the neopentyl-
idene moiety followed by R-H abstraction, and/or (iii) addition
of a silanol O-H bond onto the neopentylidyne moiety followed
by R-H abstraction.
This shows that the syn/anti ratio of 2.0 corresponds to a
stabilized composition at this temperature.
In situ solid-state NMR experiments provided further infor-
mation about this isomerization. A rotor, filled with pure syn-
2a, was heated in the NMR probe at 60 °C13 for 30 min, and
MAS 1H NMR spectra were recorded at this temperature, giving
a new carbenic signal at 12.5 ppm. Upon cooling, the relative
intensities of the peaks at 11 and 12.5 ppm did not change.
Finally, no syn-anti isomerization of 2a was observed at
(11) In four-coordinate d0 species, syn alkylidenes exhibit a lower 1JCH coupling
constant and upfield 1H and 13C NMR resonances relative to the anti
rotamer. ReVII: see ref 1b. MoVI : Oskam, J. H.; Schrock, R. R. J. Am.
Chem. Soc. 1993, 115, 11831. See also: LaPointe, A. M.; Schrock, R. R.;
Davis, W. M. J. Am. Chem. Soc. 1995, 117, 4802 and references therein.
t
(12) The signal for CH2 Bu of anti-2b, which should appear around 47 ppm
(according to the model compound), is probably buried under the broad
t
CH2 Bu signal of syn-2b (44 ppm). The extra peak at 68 ppm is probably
One should note at this point that the preparation of 1a
due to a partial decomposition of the product because it does not appear
reproducibly.
t
involves two distinct alkylation steps with [Mg(CH2 Bu)Cl],
(13) The actual temperature in the probe is probably much higher that the one
fixed (>100 °C).
allowing in principle for the introduction of first the alkylidyne
9
J. AM. CHEM. SOC. VOL. 125, NO. 2, 2003 497

A R T I C L E S
Chabanas et al.
Scheme 6. Possible Mechanisms for the Grafting of 1a onto Silica
Scheme 7. Preparation of Isotopomers of 1
tion of 13C labeling in the methylene (CH2 Bu), the methine
(dCHtBu), and the alkylidyne (tCtBu) carbons, respectively.
This shows that the second step of alkylation is probably not
very selective.
t
and the alkylidene, and then the two alkyl moieties (Scheme
7). This strategy allows in principle the selective synthesis of
t
four distinct isotopomers depending on the use of [Mg(CH2 -
Bu)Cl] and/or [Mg(13CH2 Bu)Cl] for each alkylating step.
t
Therefore, it is possible to generate the unlabeled or fully 13C-
labeled dimers, 4a and 4b, respectively, which can in turn be
transformed via a second alkylation step into 1a (unlabeled),
1c (dilabeled, selectively on the two alkyl ligands), 1d (dilabeled,
selectively on the alkylidene and alkylidyne ligands), and 1e
(tetralabeled), depending on the use of a labeled or unlabeled
Grignard reagent at each alkylation step. While the synthesis
of 1a and 1e provided the expected isotopomers, the selectivity
in the formation of 1c and 1d was only partial. Therefore, 1c
was obtained with 92 and 15% incorporation of 13C labeling in
The alkylidene and alkyl ligands can interexchange readily
under the reaction conditions, probably via H-transfer,14 while
the alkylidyne seems reluctant to participate in such a process;
hence no incorporation of 13C (via an exchange process) in this
ligand is observed.15 To have an estimation of the stability of
the isotopomer composition, a 0.1 M solution in C6D6 of 1c
(14) Schrock, R. R.; Clark, D. N.; Sancho, J.; Wengrovius, J. H.; Rocklage, S.;
Pedersen, S. F. Organometallics 1982, 1, 4802. Caulton, K. G.; Chilsom,
M. H.; Streib, W. E.; Xue, Z. J. Am. Chem. Soc. 1991, 113, 6082. Lapointe,
A. M.; Schrock, R. R.; Davis, W. M. J. Am. Chem. Soc. 1995, 117, 4802.
Li, L.; Hung, M.; Xue, Z. J. Am. Chem. Soc. 1995, 117, 12746. Chen, T.;
Wu, Z.; Li, L.; Sorasaenee, K. R.; Diminie, J. B.; Pan, H.; Guzei, I. A.;
Rheingold, A. L.; Xue, Z. J. Am. Chem. Soc. 1998, 120, 13519.
(15) No 13C-enrichment of the alkylidyne carbon was detected by NMR
spectroscopy (<2% 13C).
the methylene (CH2 Bu) and the methine (dCHtBu) carbons,
t
respectively, with no labeling in the alkylidyne carbon (tC-
tBu), while 1d was obtained with 27, 65, and >85% incorpora-
9
498 J. AM. CHEM. SOC. VOL. 125, NO. 2, 2003

Perhydrocarbyl ReVII Complexes
A R T I C L E S
1
Figure 7. (a) MAS H and (b) HPDEC 13C solid-state NMR of 2c.
Scheme 8. Grafting of 1c onto SiO2-(700) To Give 2c
was stored at 25 °C in the dark, and the isotopic composition
was measured regularly over a 2-month period by H NMR
1
spectroscopy. The percentage of 13C-label on the methylene
carbon slowly decreased, whereas the percentage of 13C-label
in the carbenic carbon increased.16 Interestingly, the singlet for
the methyl group of the neopentylidyne (1.38 ppm) did not
change, showing that no 13C-enrichment occurs on the carbynic
ligand. Integration of the carbenic signals gives the percentage
of 13C incorporated into the carbenic moiety. The isotopic
composition stabilized after 1000 h (about 1.5 months!) when
the carbenic carbon and the methylene carbons were 13C-
enriched to 64 and 68%, respectively. The alkylidyne ligand
was still not enriched (<2%). This corresponds to a quasi
statistical redistribution of the 13C label over alkyl and alkylidene
ligands (that is, about 66% 13C enrichment on each position).
R-H transfer reactions can readily account for these observa-
tions.13 According to our data, only the transfer of a methylene
proton (of the neopentyl) to the alkylidene moiety takes place,
while the alkylidyne is a spectator ligand and does not participate
in R-H transfer reactions.17 It is worth noting that the isotopomer
distribution can be considered stable on the reaction time scale
used for impregnation. Therefore, the grafting of 1c onto
SiO2-(700) gave the surface product 2c. CP/MAS and especially
a direct excitation 13C solid-state NMR spectrum show abso-
lutely no carbynic resonance (Figure 7), even though the pulse
sequence and the parameters used were those known to favor
the observation of carbynic resonances.
ligand during the grafting (Scheme 8). The molecular precursor
1c is partially 13C-enriched on the alkyl and alkylidene ligands
to 92 and 15%, respectively. The isotopic distribution of the
neopentane released during its grafting onto SiO2-(700) should
provide information about the grafting mechanism involved
(eq 1).
The relative importance of pathways (i) and (ii) can be in
principle estimated (eqs 1 and 2) using the percentage of 13C
incorporated in the neopentane liberated during grafting (%Np*)
and the percentage of 13C labeling on each of the ligands in 1c
(Px*). For pathway (i), every alkyl ligand has an equiprobability
to be transformed into neopentane; hence the probability to
observe labeled neopentane (P1) is equal to the percentage of
labeling of the alkyl moiety (Palkyl*). For pathway (ii), the
probability of observing labeled neopentane (P2; eq 3) depends
for 2/3 on the percentage of labeling of the alkyl moiety (Palkyl*)
and for 1/3 on the percentage of labeling of the alkylidene moiety
(Palkylidene*).
This rules out the participation of mechanism (iii) in the
grafting of 1a because it implies first the formation of an
intermediate containing two alkylidene ligands, of which one
would be partially 13C-labeled thus leading to scrambling in
the R-H abstraction step (Scheme 8). No strong 12C/13C isotopic
effect is expected for the R-H abstraction step, and therefore if
mechanism (iii) were operating, we would have observed at least
a partial 13C enrichment of the carbynic ligand during the
grafting. The absence of labeling (<5%) in the alkylidyne ligand
in the surface species shows that the carbyne in 1c is a spectator
(%Np*/100) - P2
fraction of pathway (i) )
(1)
P1 - P2
fraction of pathway (ii) ) 1 - (fraction of pathway (i)) (2)
with
2
1
3
P1 ) Palkyl
*
and P2 ) × Palkyl* + × Palkylidene
* (3)
3
The grafting of 1c onto SiO2-(700) yielded neopentane 81 (
5% 13C-labeled according to mass spectroscopy. Therefore,
applying this model to the experimental results suggests that
57 ( 20% of grafting would occur via mechanism (i) and the
remaining amount (43 ( 20%) occurs via mechanism (ii).
t
(16) A partially deuterated analogue, Re(tCtBu)(dCHtBu)(CD2 Bu)2 has shown
no evidence for H/D scrambling among the neopentyl and neopentylidene
ligands at 80 °C in toluene-d8, see: Lapointe, A. M.; Schrock, R. R.
Organometallics 1995, 14, 1875.
(17) In Re(tCtBu)(dCHtBu)(OR)2, the alkylidyne ligand was found to behave
in most cases as an ancillary ligand, see ref 7c.
9
J. AM. CHEM. SOC. VOL. 125, NO. 2, 2003 499

A R T I C L E S
Chabanas et al.
Scheme 9. Isotopic Distribution of Neopentane Released During the Grafting of 1c onto SiO2-(700), Following Either Pathway (i) or Pathway
(ii)
Moreover, these reaction pathways are usually estimated from
the reaction of the organometallic complex and deuterated silica
readily obtained from H/D exchange with D2O. Therefore, the
grafting of 1a was performed on a 92% deuterated silica.
Because the addition of (tSiO-H) onto the carbynic moiety
has been ruled out, the relative contribution of pathways (i) and
(ii) can be determined by quantifying monodeuterated and
nondeuterated neopentane released during the grafting of 1a onto
partially deuterated SiO2-(700) (eqs 4 and 5).
Moreover, because the deuteration of SiO2-(700) is usually
only partial (typically 90-95%), the model must integrate the
percentage of nondeuterated silanol groups (%SiOH) present
on the partially deuterated SiO2-(700). The reaction of 1a with
tSiOH produces indeed only nondeuterated neopentane what-
ever the pathway is. With all of these facts in hand, it is possible
to obtain the fraction of the pathways (i) and (ii) depending on
the percentage of deuteration in neopentane liberated in the gas
phase (%NpD, eqs 4-6). The probability for the elimination
of monodeuterated neopentane (P1) in pathway (i) is equal to 1
in any case, while the probability for the elimination of
monodeuterated neopentane (P2) in pathway (ii) is 1/2 if isotope
effects in the R-H abstraction step are considered negligible.
If isotope effects (kH/kD) are taken into account, P2 can easily
be estimated (eq 6 and vide infra for comments).
the sole isotopomer, that is, a mixture 92/8 of mono- and
nondeuterated neopentane on 92% deuterated silica. The forma-
tion of 26% of nondeuterated neopentane shows that the grafting
has to also involve the addition of the silanol on the alkylidene
moieties of 1a followed by an R-abstraction (vide supra
mechanism (ii)). Applying this model to our experimental results
(%SiOH ) 8%, %NpD ) 74 ( 5%, and %NpH ) 26 ( 5%,
kH/kD ) 1) shows that grafting occurs 61 ( 10% via mechanism
(i) and 39 ( 10% via mechanism (ii) (Scheme 9), which is to
be compared with the results obtained from the experiment using
13C-labeled 1c and which shows that grafting occurs 57 ( 20%
via mechanism (i) and 43 ( 20% via mechanism (ii). Note-
worthy is the good agreement between these two independent
methods to determine the relative contribution of mechanisms
(i) and (ii). Moreover, looking into the influence of isotope
effects on the relative distribution of mono- and nondeuterated
neopentane shows that they have little effect on the determi-
nation of each pathway (it falls within experimental errors,
which preclude, in turn, their evaluation).18 In conclusion, about
60 ( 10% of 1a reacts with SiO2-(700) via mechanism (i), while
40 ( 10% reacts via mechanism (ii) (Scheme 10). Because 1a
contains two alkyl ligands for one alkylidene ligand, the
reactivity ratio between alkylidene and alkyl ligands is 1.3 (
0.6, showing that, in this case, alkyl and alkylidene moieties
have in fact a similar reactivity toward silanols.
%NpD - P2 × (100 - %SiOH)
fraction of pathway (i) )
These data are in sharp contrast to what has been observed
in molecular chemistry because the reaction of 1a with Bro¨nsted
acids (HX, X ) Cl, OC6F5, BF4, OTf, etc.) gives the products
resulting exclusively from the protonation onto the alkylidene
(100 - %SiOH) × (P1 - P2)
(4)
fraction of pathway (ii) ) 1 - (fraction of pathway (i)) (5)
with
t
moiety, [Re(tCtBu)(CH2 Bu)3X], which can be isolated in fairly
good yields (60-80%).1b These compounds do not undergo
spontaneously R-H abstraction of neopentane to give alkylidene
complexes. This R-H abstraction step has to be induced by
adding donor ligands such as pyridine, acetonitrile, or methanol.
By comparison, the reaction of 1a with silica would proceed
via both addition of tSiO-H to the carbenic moiety, giving
(kH/kD) + 1
P1 ) 1 and P2 )
3 (kH/kD) + 1
(P2 ) 0.5 for kH/kD ) 1) (6)
[(tSiO)-Re(tCtBu)(CH2 Bu)3] as a transient state, and direct
t
t
The grafting of 1a onto 92% deuterated SiO2-(700) yielded
neopentane on average 74 ( 5% monodeuterated and 26 ( 5%
nondeuterated according to mass spectroscopy. A simple elec-
trophilic cleavage of a Re-alkyl bond by a surface SiO-D bond
(mechanism (i)) would produce monodeuterated neopentane as
electrophilic cleavage of the Re-CH2 Bu bond, both leading
to syn-2a without addition of an external molecular ligand. The
(18) For example, using a kH/kD of 2 and 7 would lead to our case of F(i) equal
to 66 and 69%, respectively, to be compared with 61% assuming no isotope
effect.
9
500 J. AM. CHEM. SOC. VOL. 125, NO. 2, 2003

Perhydrocarbyl ReVII Complexes
A R T I C L E S
Scheme 10. Percentage of Non- and Monodeuterated Neopentane Produced During the Reaction of tSiOD Groups with 1a Following
Either Pathway (i) or Pathway (ii)
Scheme 11. Understanding of the Origin of the Stereoselectivity
in the Preparation of syn-2a
the surface via one SiO-Re linkage and containing a neopentyl,
a neopentylidene, and a neopentylidyne ligand along with a
siloxane bridge, ([(tSiO)-Re(tCtBu)(dCHtBu)(CH2 Bu)-
t
(tSiOSit)]). This surface complex was found to be present
only as its syn rotamer (syn-2a). The grafting probably proceeds
via two pathways: direct protolytic cleavage of a Re-neopentyl
bond of 1a by SiO-H (60%) and addition of SiO-H bond onto
the alkylidene moiety of 1a followed by R-H abstraction (40%).
The neopentylidyne ligand of 1a has been found unreactive
under these conditions and acts as a spectator ligand during the
grafting. Interestingly, syn-2a can be partially isomerized into
its anti rotamer (anti-2) either by heating or by exposing it to
light. This study further demonstrates that surface organometallic
chemistry can allow the preparation of well-defined surface
organometallic species, which can be characterized at a mo-
lecular level. Finally, like Schrock et al. stated in 1983,1a
EXAFS data point out the presence of a siloxane bridge (donor
ligand) in the coordination of 2a, which could have played the
role of the external ligand to induce the spontaneous extrusion
t
of the neopentyl ligand in [(tSiO)-Re(tCtBu)(CH2 Bu)3] to
form directly 2a. Another possibility would conciliate the
observation in molecular (sole reactivity of the alkylidene ligand)
and surface organometallic chemistry, that is, if the addition of
the silanol onto the carbene would be reversible. Indeed, in this
case, successive addition and its reverse reaction would lead to
scrambling of the labels (2H or 13C), which would therefore
preclude the possibility to evaluate the proportion of pathway
(i) because pathway (ii) would have a similar effect on the
isotopomer distribution. Therefore, the surface reaction could
well-occur completely via pathway (ii) if the addition of the
silanol onto the carbene is reversible. It is, however, difficult
to probe this mechanism.
“[Re(tCtBu)(dCHtBu)(CH2 Bu)2] is an especially interesting
t
species in the sense it fills out the series of four-coordinate
compounds...,” we would like to make the same comment
concerning 2a, [(tSiO)Re(tCtBu)(dCHtBu)(CH2 Bu)2], in
t
view of the recently characterized corresponding surface
complexes of Ta,5b Mo,5a and W20 as [(tSiO)Ta(dCHtBu)-
(CH2 Bu)2] and [(tSiO)M(tCtBu)(CH2 Bu)2] for M ) Mo and
t
t
W.
Finally, as a molecular chemist, one might ask why 2a is
obtained as a single isomer. The molecular complex 1a is present
as a single isomer, probably the syn rotamer according to NMR
data. While the direct electrophilic cleavage should provide syn-
2a in any case, the process involving the addition to the carbenic
moiety has to be stereoselective. The addition of the silanol
involves first its coordination to the Re center to form probably
a distorted trigonal bipyramidal surface complex19 (Scheme 11),
followed by the addition of the proton to the carbenic moiety,
Experimental Section
All experiments were carried out in the strict absence of oxygen
and water. Standard Schlenk techniques under argon were used for
organometallic syntheses. Surface compounds were dried under high
vacuum (10-5 Torr). Pentane, dichloromethane, diethyl ether, and THF
were purified according to the published procedures.21 C6D6 (SDS) was
distilled over Na/benzophenone and stored over 3 Å molecular sieves.
[Re(dNtBu)2Cl3] was prepared in a one-pot reaction involving Re2O7
(99.9%, Strem Chemicals), tBuNH2 (99.5%, Aldrich), (CH3)3SiCl (99%,
Aldrich), and anhydrous gaseous HCl (Air Liquide), used as received,
following a literature procedure.1 2,4-Lutidine hydrochloride was
prepared by bubbling an excess of anhydrous gaseous HCl into a
solution of 2,4-lutidine (98%, Acros) in diethyl ether, followed by
filtration and washing steps. Ph3SiOH (Aldrich) and the polyoligomeric
silsesquioxane 3 (Aldrich) were dried under high vacuum prior use, at
25 °C (4 h) and at 40 °C (12 h), respectively. SiO2 (Aerosil Degussa,
200 m2/g) was calcined at 400 °C in air for 2 h, partially dehydroxylated
t
while the Bu group points away from the surface. The
subsequent R-H abstraction step thus generates syn-2a.
Conclusion
The reaction of 1a with silica dehydroxylated at 700 °C has
been investigated by IR, elemental analysis, chemical reactivity,
EXAFS, 1D and 2D solid-state NMR spectroscopy, and
synthesis of molecular models. The data are fully consistent
with the formation of only one surface species 2a, grafted to
(20) Chabanas, M. Ph.D. Thesis, Universite´ Claude Bernard Lyon I, 2001.
(21) Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Purification of Laboratory
Chemicals; Pergamon Press: New York, 1980.
(19) Schrock, R. R.; Crowe, W. E.; Bazan, G. C.; DiMare, M.; O’Regan, M.
B.; Schofield, M. H. Organometallics 1991, 10, 1832. See also ref 1b.
9
J. AM. CHEM. SOC. VOL. 125, NO. 2, 2003 501

A R T I C L E S
Chabanas et al.
at 500 °C under high vacuum (10-5 mmHg) for 12 h, and then at 700
°C for 4 h (support referred to as SiO2-(700)). In experiments requiring
deuterated silica, a similar procedure was used, but the first partial
dehydroxylation at 500 °C was followed by D2O treatment and partial
dehydroxylation at 500 °C for 5 h (five cycles). After partial
dehydroxylation at 700 °C for 4 h, silica was 92-93% deuterated
according to IR spectroscopy. Solution NMR spectroscopy was recorded
Synthesis of 1a. The rhenium complex 1a was prepared in four steps
according to a literature procedure1a starting from Re2O7 and involving
the following intermediates: [Re(dNtBu)2Cl3],1b [Re(dNtBu)2(dCH-
t
t
tBu)(CH2 Bu)],1a and [Re(tCtBu)(dCHtBu)-(NH2 Bu)Cl2]2‚(THF).1a
Overall yields were in the range 30-40% (lit. 45%). 1H NMR (C6D6):
δ 1.10 (s, 18H, CH3), 1.27 (s, 9H, CH3), 1.38 (s, 9H, CH3), 1.55 (d,
2JHaHb ) 12.3 Hz, 2H, ReCHaHb Bu), 1.89 (d, JHaHb ) 12.3 Hz, 2H,
t
2
1
t
on a Bruker ACX-300 and DRX-500 spectrometer. H NMR spectra
ReCHaHb Bu), 7.65 (s, 1H, RedCHtBu) ppm. 13C NMR (CDCl3): δ
were referenced to C6D5H at 7.15 ppm. 13C NMR spectra were
referenced to C6D6 at 128.0 ppm. GC analysis of neopentane was
performed on a gas chromatograph HP 5890, equipped with a flame
ionization detector (FID) and a KCl/Al2O3 on fused silica column (50
m × 0.32 mm). 1D MAS 1H and 13C CP/MAS solid-state NMR spectra
were recorded on a Bruker DSX-300 spectrometer operating at 300
and 75 MHz for 1H and 13C, respectively. The samples were introduced
under Ar in a zirconia rotor, which was then tightly closed. In all
experiments, the sample rotation frequency was set to 10 kHz. Chemical
shifts are given with respect to TMS as an external standard, with a
29.7 (CC(CH3)3), 32.4 (CC(CH3)3), 34.6 (CC(CH3)3), 35.1 (ReCH2-
C(CH3)3), 43.1 (RedCHC(CH3)3), 52.6 (RetCC(CH3)3), 78.6 (ReCH2-
tBu), 224.4 (RedCHtBu), 295.1 (RetCtBu) ppm.
Preparation of Labeled 1b-e. (1-13C, 99%) Monolabeled neopen-
tylmagnesium chloride was prepared in four steps starting from 13C-
labeled (carbonyl, 1-13C, 99%) dimethylformamide (H13CONMe2,
t
DMF*), using the following sequence: reaction of BuLi with DMF*
to form (1-13C, 99%) 2,2-dimethylpropanal (not isolated), followed by
a reduction with LiAlH4 to yield the corresponding alcohol (step 1), a
tosylation (step 2), and a nucleophilic substitution with LiCl to give
13C-monolabeled (1-13C, 99%) 2,2-dimethylchloropropane (step 3). It
was then transformed into the corresponding Grignard reagent (step
4).
1
precision of 0.2 and 1 ppm for H and 13C NMR, respectively. For
CP/MAS 13C NMR, the following sequence was used: 90° pulse on
the protons (pulse length 3.8 µs), then a cross-polarization step with a
contact time typically set to 5 or 10 ms, and finally acquisition of the
13C signal under high power proton decoupling. The delay between
t
Step 1: Synthesis of Bu13CH2OH. To a mixture of DMF* (6.9 g,
t
93.2 mmol) and diethyl ether (200 mL) at -78 °C was added BuLi
1
each scan was set to 1 s, to allow the complete relaxation of the H
(1.7 M solution in pentane, 73 mL, 124 mmol) in 30 min under vigorous
stirring. The light yellow resulting reaction mixture was allowed to
warm to 25 °C, stirred for 30 min, and transferred via a cannula into
200 mL of a 3 M aqueous solution of HCl. The aqueous phase was
extracted with diethyl ether. The combined organic layers were washed
with a saturated aqueous solution of NaHCO3 and dried 30 min over
MgSO4. The resulting solution was treated by LiAlH4 (2.7 g, 71.1
mmol), stirred for 15 min, and treated with a saturated aqueous solution
of Na2SO4. The aqueous phase was extracted with diethyl ether, and
the combined organic phases were dried over MgSO4. After evaporation
of diethyl ether under atmospheric pressure, a colorless oil (7.43 g,
89.5%) was obtained. 1H NMR (C6D6): δ 0.81 (d, 3JCH ) 4.8 Hz, 9H,
nuclei. For direct excitation 13C NMR, the following sequence was used:
pulse on the carbons (impulsion length 3.0 µs) and recording of the
13C signal under high power proton decoupling. The delay between
each scan was set to 1 s. For both CP/MAS and direct excitation 13C
NMR, an apodization function (exponential) corresponding to a line-
broadening of 50 Hz was applied. The 2D solid-state NMR spectroscopy
experiments were conducted on a Bruker DSX 500 spectrometer using
a 4-mm MAS probe. For the cross-polarization step, a ramped radio
frequency (RF) field centered at 77 kHz was applied to protons, while
the carbon RF field was matched to obtain optimal signal. During
acquisition, the proton decoupling field strength was also set to 77 kHz.
A total of 64 t1 increments with 256 scans each were collected. The
sample spinning frequency was 10 kHz, and the contact time for the
cross-polarization step was set to a value as defined in the figures.
Quadrature detection in ω1 was achieved using the TPPI method.22 For
EXAFS experiments, the samples were packaged in aluminum holders
with Kapton tape and sealed inside two Mylar pouches in an Ar filled
drybox. The samples were then put into glass jars and sealed with Teflon
tape inside the drybox. Everything except for the Teflon tape was baked
out at 110 °C overnight. X-ray absorption spectra were acquired at the
Stanford Synchrotron Radiation Laboratory (SSRL) at beam-line 4-1
using a Si(220) double crystal monochromator detuned 50% to reduce
the higher order harmonic content of the beam. X-ray absorption spectra
were obtained in the transmission mode at room temperature using argon
filled ionization chambers. The data analysis was performed by standard
procedures using the EXAFSPAK suite of programs developed by G.
George of SSRL.23 Fitting of the spectrum was done on the k3-weighted
data using the following EXAFS equation where S02 is the scale factor,
fixed at 0.9; Ni is the coordination number of shell i; Si is the central
atom loss factor for atom i; Fi is the EXAFS scattering function for
atom i; Ri is the distance to atom i from the absorbing atom; λi is the
photoelectron mean free path; σi is the Debye-Waller factor; φi is the
EXAFS phase function for atom i; and φc is the EXAFS phase function
for the absorbing atom.
t
1
(CH3)3C13CH2OH), 0.96 (br, 1H, Bu13CH2OH), 3.03 (d, JCH ) 140
t
Hz, 2H, Bu13CH2OH) ppm.
Step 2: Synthesis of tBu13CH2OTs. p-Toluenesulfonyl chloride (16.7
t
g, 87.7 mmol) was slowly added to a solution containing BuCH2OH
(7.43 g, 83.5 mmol), pyridine (13.4 mL, 166 mmol), and CH2Cl2 (85
mL). The reaction mixture was stirred at room temperature for 2 days
and then treated by 130 mL of a 1.5 M HCl aqueous solution. The
aqueous phase was extracted four times with CH2Cl2 (20 mL). The
combined organic phases were washed with a saturated aqueous solution
of NaHCO3 (40 mL), dried 30 min over MgSO4, and evaporated to
give a colorless oil (19.0 g, 93.5%), which crystallized after 24 h at
1
3
room temperature. H NMR (CDCl3): δ 0.88 (d, JCH ) 4.7 Hz, 9H,
(CH3)3C13CH2OTs), 2.44 (s, 3H, CH3-Ar), 3.64 (d, JCH ) 148 Hz,
1
2H, Bu13CH2OTs), 7.33 (d, JHH ) 8.3 Hz, 2H, Ar), 7.77 (d, JHH
)
t
3
3
8.3 Hz, 2H, Ar) ppm. 13C NMR (CDCl3): δ 21.6 (CH3-Ar), 26.0
((CH3)3C13CH2OTs), 31.6 ((CH3)3C13CH2OTs), 79.5 (tBu13CH2OTs),
127.9 (Ar), 129.7 (Ar), 133.0 (Ar), 144.6 (Ar) ppm.
Step 3: Synthesis of tBu13CH2Cl. To a 100 mL round-bottom flask
equipped with a short path distillation apparatus and a receiver Schlenk
t
were added under argon Bu13CH2OTs (19.0 g, 78.2 mmol), DMPU
(1,3-dimethyl-3,4,5,6-tetrahydro-2(1H)-pyrimidone, 35 mL), and an-
hydrous LiCl (9.2 g, 217 mmol). The mixture was heated to 130 °C
t
under vigorous stirring, and Bu13CH2Cl (7.4 g, 88%) slowly distilled
n
off in 16 h. 1H NMR (C6D6): δ 0.77 (d, 3JCH ) 5.3 Hz, 9H, (CH3)3C13-
NiSi(k,Ri)Fi(k,Ri)
-2Ri
ø(k) = S20
exp
exp(-2σ2i k2i ) sin
1
t
CH2Cl), 2.95 (d, JCH ) 148.3 Hz, 2H, Bu13CH2Cl) ppm. 13C NMR
∑
(
)
kR2i
λ(k,Ri)
i)1
[2kRi + φi(k,Ri) + φc(k)]
(22) Marion, D.; Wu¨trich, K. Biochem. Biophys. Res. Commun. 1983, 113, 967.
(23) Koningsberger, D. C.; Prins, R. X-ray Absorption: Principles, Applications,
Techniques of EXAFS, SEXAFS, and XANES; John Wiley & Sons: New
York, 1988.
The program FEFF824 was used to calculate theoretical values for Si,
Fi, λi, φi, and φc on the basis of atomic positions taken from the crystal
structure of the most similar complex.
(24) Ankudinov, A.; Ravel, B.; Rehr, J. J.; Conradson, S. Phys. ReV. 1998, B58,
7565.
9
502 J. AM. CHEM. SOC. VOL. 125, NO. 2, 2003

Perhydrocarbyl ReVII Complexes
A R T I C L E S
(CDCl3): δ 27.1 ((CH3)3C13CH2Cl), 33.1 ((CH3)3C13CH2Cl), 57.3
(tBu13CH2Cl) ppm.
) 4.8 Hz, 9H, CH3), 1.38 (d, 3JCH ) 4.7 Hz, 9H, CH3), 1.89 (dd, 1JCH
t
) 116 Hz, 2JHaHb ) 13 Hz, 2H, Re13CHaHb Bu), 7.65 (dd, 1JCH ) 115
3
Hz, JHCReC ) 5.3 Hz, 1H, Red13CHtBu) ppm. 13C NMR (CDCl3)
t
Step 4: Synthesis of Bu13CH2MgCl. Turnings of Mg (1.6 g, 65.8
t
labeled carbons: δ 78.6 (ReCH2 Bu), 224.4 (RedCHtBu), 295.1 (Ret
mmol) were activated at 400 °C under vacuum. A solution of
tBu13CH2Cl (6.8 mL; 54.8 mmol) in diethyl ether (10 mL) and THF
(10 mL) was then added under argon, heated to reflux, treated with
0.1 mL of 1,2-dibromoethane, and then heated at 110 °C overnight.
The resulting mixture was then filtered, and the filtrate was diluted
CtBu) ppm.
Synthesis of 1d. A solution of [Re(tCtBu)(dCHtBu)(NH2 Bu)Cl2]2‚
t
(THF) (0.32 g, 0.31 mmol) in THF (10 mL) was cooled to -40 °C,
and a 0.87 M solution of tBu13CH2MgCl in diethyl ether/THF (1.6 mL,
1.39 mmol) was added dropwise. The mixture was stirred for 45 min
at -40 °C, and the solvent was then evaporated. The residue was
extracted with pentane (15 mL). The extract was filtered through Celite,
and the solvent was removed from the filtrate in vacuo, leaving a
yellow-orange oil which can be purified by sublimation at room
temperature under vacuum (10-5 Torr) onto a coldfinger (liquid N2).
t
with diethyl ether (30 mL). A 0.87 M solution of Bu13CH2MgCl in
diethyl ether/THF was obtained (42 mL, 66.7%).
Synthesis of 1b. This complex, 10% randomly 13C-labeled on the
R-carbons, can be prepared in a manner similar to that of 1a, by using
a 90/10 mixture of nonlabeled and 100% 13C-labeled neopentylmag-
nesium chloride as alkylating agent in the two alkylation steps.
t
3
3
Synthesis of [Re(t13CtBu)(d13CHtBu)(13CH2 Bu)2], 2b. This com-
1H NMR (C6D6): δ 1.11 (d, JCH ) 4.8 Hz, 18H, CH3), 1.27 (d, JCH
) 4.8 Hz, 9H, CH3), 1.38 (d, 3JCH ) 4.7 Hz, 9H, CH3), 7.65 (dd, 1JCH
) 115 Hz, 3JHCReC ) 5.3 Hz, 1H, Red13CHtBu) ppm. 13C NMR (CDCl3)
pound was prepared in a three-step procedure starting from [Re(dN-
tBu)2Cl3], via the intermediates [Re(dNtBu)2(d13CHtBu)(13CH2tBu)] and
t
t
[Re(t13CtBu)(d13CHtBu)(NH2 Bu)Cl2]2‚(THF).
labeled carbons: δ 78.6 (ReCH2 Bu), 224.4 (RedCHtBu), 295.1 (Ret
CtBu) ppm.
Step 1: Synthesis of [Re(dNtBu)2(d13CHtBu)(13CH2 Bu)]. To a
t
solution of [Re(dNtBu)2Cl3] (1.3 g, 2.99 mmol) in diethyl ether (50
mL) was added dropwise at -78 °C a 0.87 M solution of Bu13CH2-
Synthesis of 1e. A solution of 4b [Re(t13CtBu)(d13CHtBu)(NH2-
tBu)Cl2]2‚(THF) (0.30 g, 0.30 mmol) in THF (10 mL) was cooled to
-40 °C, and a 0.87 M solution of tBu13CH2MgCl in diethyl ether/THF
(1.32 mmol) was added dropwise. The mixture was stirred for 45 min
at -40 °C, and the solvent was then evaporated. The residue was
extracted with pentane (15 mL). The extract was filtered through Celite,
and the solvent was removed from the filtrate in vacuo, leaving a
yellow-orange oil which can be purified by sublimation at room
temperature under vacuum (10-5 Torr) onto a coldfinger (liquid N2).
t
MgCl in diethyl ether/THF (10.3 mL, 8.97 mmol). The resulting deep
purple solution was allowed to warm to room temperature, and the
color turned to brown-green. The reaction mixture was stirred at room
temperature for 2 h, filtered through Celite, and the filtrate was
evaporated to dryness to give a dark residue. Sublimation under vacuum
(10-5 Torr) at 80 °C gave a yellow oil (0.83 g, 59%). 1H NMR (C6D6):
δ 1.19 (d, 3JCH ) 4.8 Hz, 9H, CH3), 1.29 (d, 3JCH ) 4.8 Hz, 9H, CH3),
2
1
1.37 (s, 18H, dNtBu), 2.69 (dd, JHaHb ) 13.3 Hz, JCHa ) 126 Hz,
3
3
1H NMR (C6D6): δ 1.11 (d, JCH ) 4.8 Hz, 18H, CH3), 1.27 (d, JCH
t
2
1
) 4.8 Hz, 9H, CH3), 1.38 (d, 3JCH ) 4.7 Hz, 9H, CH3), 7.65 (dd, 1JCH
2H, ReCHaHb Bu), 2.90 (dd, JHaHb ) 13.3 Hz, JCHb ) 126 Hz, 2H,
ReCHaHb Bu), 12.02 (d, JCH ) 133 Hz, 1H, RedCHtBu) ppm. 13C
t
1
) 115 Hz, JHCReC ) 5.3 Hz, 1H) ppm. 13C NMR (CDCl3) labeled
3
t
t
carbons: δ 78.6 (ReCH2 Bu), 224.4 (RedCHtBu), 295.1 (RetCtBu)
NMR (C6D6) labeled carbons: δ 34.5 (ReCH2 Bu), 262.4 (RedCH-
tBu) ppm.
ppm.
Step 2: Synthesis of [Re(t13CtBu)(d13CHtBu)(NH2 Bu)Cl2]2‚
t
Synthesis of 2a by Impregnation of 1a onto SiO2-(700). A mixture
of 1a (200 mg, 0.43 mmol) and SiO2-(700) (1.00 g) in pentane (10 mL)
was stirred at 20 °C for 2 h. After being filtered, the solid was washed
three times with pentane, and all volatile compounds were condensed
into another reactor of known volume to quantify neopentane evolved
during the grafting. The resulting yellow powder was dried under
vacuum (10-5 Torr) to yield 1.1 g of 2a. Analysis by gas chromatrog-
raphy indicated the formation of 0.24 mmol of neopentane during
grafting. Elemental analysis of 1a: C, 4.55%w and Re, 4.75%w (found,
C/Re ) 14.86; calcd for 1a, C/Re ) 15.00). Solid-state MAS 1H NMR
(300 MHz): δ 1.2, 2.6, 3.0, and 11.1 ppm. CP/MAS 13C NMR: δ
30.3, 44.4, and 245.8 (weak) ppm. IR: 2960, 2932, 2905, 2873, 2742,
1474, 1461, 1390, 1363 cm-1. ESR: no signal.
(THF).4b 2,4-Lutidine hydrochloride (0.62 g, 4.31 mmol) was added
t
to a solution of [Re(dNtBu)2(d13CHtBu)(13CH2 Bu)] (0.67 g, 1.42
mmol) in CH2Cl2 cooled at -40 °C. The color changed from yellow
to orange. The mixture was stirred at -40 °C for 1.5 h and then at 0
°C for 1 h. The white powder of tBuNH3Cl was removed by filtration
through Celite and carefully extracted with 3 × 2 mL of CH2Cl2. The
combined fractions were evaporated in vacuo to give a light orange
powder (0.63 g, 87%) which can be crystallized from a 1:1 mixture of
THF and pentane at -40 °C. 1H NMR (CDCl3): major isomer δ 1.18
3
(s, 18 H, (CH3)3CNH2), 1.30 (d, JCH ) 4.4 Hz, 18H, CH3), 1.40 (d,
3JCH ) 4.4 Hz, 18H, CH3), 1.84 (m, THF), 3.72 (m, THF), 4.25 (d,
t
2
2JHaHb ) 12.5 Hz, 2H, BuNHaHb), 5.34 (d, JHaHb ) 13.0 Hz, 2H,
1
3
tBuNHaHb), 13.64 (dd, JCH ) 123.3 Hz, JHCReC ) 3.4 Hz, 2H, Red
CHtBu); minor isomer δ 1.19 (s, 18 H, (CH3)3CNH2), 1.29 (d, 3JCH
5.5 Hz, 18H, CH3), 1.41 (d, 3JCH ) 4.0 Hz, 18H, CH3), 1.84 (m, THF),
Synthesis of 2a by Impregnation of 1a onto Deuterated SiO2-(700)
.
)
A similar procedure was used in which SiO2-700 was replaced by
deuterated SiO2-(700). Analysis of the evolved neopentane by GC/MS
gave reproducibly the following isotopomeric composition: d0-neopen-
tane (26 ( 5%), d1-neopentane (74 ( 5%), d2 and other isotopomers
(traces < 1%).
Synthesis of 2b by Impregnation of 1b onto SiO2-(700). The surface
compound 2b was prepared in a manner similar to that of 2a, using 1b
in place of 1a. Solid-state MAS 1H NMR (500 MHz): δ 1.1, 1.3, 1.5,
2.6, 3.1, and 11.1 ppm. CP/MAS 13C NMR: δ 29, 46, and 247 ppm.
Direct excitation 13C NMR: δ 30, 44, 246, and 292 ppm.
2
t
3.72 (m, THF), 4.39 (d, JHaHb ) 12.5 Hz, 2H, BuNHaHb), 5.13 (d,
2JHaHb ) 13.0 Hz, 2H, tBuNHaHb), 13.59 (dd, 1JCH ) 123.8 Hz, 3JHCReC
) 3.6 Hz, 2H, RedCHtBu) ppm.13C NMR (CDCl3) labeled carbons:
major isomer δ 294.6 (d, 2JCReC ) 6.6 Hz, RetCtBu), 298.7 (d, 2JCReC
) 6.6 Hz, 1JCH ) 123 Hz, RedCHtBu); minor isomer δ 294.0 (d, 2JCReC
) 6.6 Hz, RetCtBu), 299.5 (d, 2JCReC ) 6.6 Hz, 1JCH ) 124 Hz, Red
CHtBu) ppm.
t
Step 3: Conversion of 4b [Re(t13CtBu)(d13CHtBu)(NH2 Bu)Cl2]2‚
(THF) into 1c. A solution of 4b (0.15 g, 0.15 mmol) in THF (5 mL)
t
was cooled to -40 °C, and a 0.87 M solution of Bu13CH2MgCl in
Synthesis of 2c by Impregnation of 1c onto SiO2-(700). Compound
2c was prepared in a manner similar to that of 2a, using 1c in place of
diethyl ether/THF (0.7 mL, 0.61 mmol) was added dropwise. The
mixture was stirred for 45 min at -40 °C, and the solvent was then
evaporated. The residue was extracted with pentane (15 mL). The
extract was filtered through Celite and evaporated in vacuo to give a
yellow-orange oil, which can be purified by sublimation at room
temperature under vacuum (10-5 Torr) onto a coldfinger (liquid N2).
1
1a. Solid-state MAS H NMR (300 MHz): δ 1.1 and 10.8 ppm. CP/
MAS 13C NMR: δ 31, 46, and 246 ppm. Direct excitation 13C NMR:
δ 31, 45, and 247 ppm. No carbynic signal observed.
Synthesis of 2d by Impregnation of 1d onto SiO2-(700). Compound
2d was prepared in a manner similar to that of 2a, using 1d in place
3
3
1
1H NMR (C6D6): δ 1.11 (d, JCH ) 4.8 Hz, 18H, CH3), 1.27 (d, JCH
of 1a. Solid-state MAS H NMR (300 MHz): δ 1.1 and 10.9 ppm.
9
J. AM. CHEM. SOC. VOL. 125, NO. 2, 2003 503

A R T I C L E S
Chabanas et al.
CP/MAS 13C NMR: δ 30, 45, 246, and 292 ppm. Direct excitation
13C NMR: δ 30, 47, 247, and 293 ppm.
Reaction of 1a with the Polyoligomeric Silsesquioxane 3. In an
NMR tube equipped with a Teflon screw cap (provided by Young
Scientific Ltd.), the polyoligomeric silsesquioxane 3 (12.0 mg, 43.5
µmol) was added at room temperature to a yellow-orange solution of
1a (20.0 mg, 42.8 µmol) in 0.5 mL of C6D6. Compound 5 immediately
dissolved, and the color turned brown-red. Monitoring the reaction by
1H NMR spectroscopy showed the almost complete (>95%) disap-
pearance of 1a after 12 h at room temperature with the concomitant
formation of neopentane and 2n as a 10 to 1 mixture of its syn and
anti isomers. After evaporation of the reaction mixture, an orange-red
oil was obtained as an inseparable mixture of 2n of its syn and anti
Synthesis of 2e by Impregnation of 1e onto SiO2-(700). Compound
2e was prepared in a manner similar to that of 2a, using 1e in place of
1
1a. Solid-state MAS H NMR (300 MHz): δ 1.0 and 11.0 ppm. CP/
MAS 13C NMR: δ 31, 47, 247, and 294 ppm. Direct excitation 13C
NMR: δ 32, 47, 248, and 294 ppm.
Hydrogenolysis of the Surface Complex 2a. First 20-40 mg of
silica was pressed into a 18 mm self-supporting disk, put into a sealed
glass high-vacuum reactor equipped with CaF2 windows, and partially
dehydroxylated under vacuum (500 °C, 12 h and 700 °C, 4 h).
Compound 1a was then sublimed under dynamic vacuum at 50 °C on
the silica disk, which turned yellow-orange. After 1 h of reaction at 25
°C, the excess of 1a was removed by reverse sublimation in a liquid
nitrogen cooled tube, which was then sealed off using a torch. For
each step, IR spectra were recorded. IR (2a): 2955, 2927, 2898, 2864,
2744, 1476, 1459, 1390, 1362 cm-1. Anhydrous H2 (70 000 Pa, 6.6
mmol) and 80 mg of 1a (4.75% Re, 20.4 µmol Re) were heated at 250
°C for 86 h in a Schlenk tube (234 mL) to give 0.288 mmol of methane
as the sole gaseous product according to GC analysis. This corresponds
to a ratio of CH4 evolved per Re of 14.1 or, in other words, 2.83
neopentyl-like ligand per Re. The reaction can be monitored by in situ
IR spectroscopy using 2a prepared in an IR-cell.
1
rotamers in a 10/1 ratio. H NMR (C6D6): major isomer (syn) δ 1.09
(s, 9H, ReCH2C(CH3)3), 1.09-1.25 (br m, c-CH(CH2)4), 1.25 (s, Red
CHC(CH3)3), 1.37 (s, 9H, RetCC(CH3)3), 1.4-2.1 (br m, c-CH(CH2)4),
2
t
2
2.56 (d, JHaHb ) 12.8 Hz, 1H, ReCHaHb Bu), 3.00 (d, JHaHb ) 12.8
Hz, 1H, ReCHaHb Bu), 11.00 (s, 1H, RedCHtBu) ppm. 13C NMR: δ
t
22.7, 22.9, 27.3, 27.5, 27.8, 28.0 (c-CH(CH2)4)), 29.7 (RetCC(CH3)3),
31.6 (RedCHC(CH3)3), 32.5 (ReCH2C(CH3)3), 32.6 (ReCH2C(CH3)3),
t
45.1 (RedCHC(CH3)3), 49.3 (ReCH2 Bu), 53.4 (RetCC(CH3)3), 247.2
(RedCHtBu, 1JCH ) 118.0 Hz), 291.4 ppm (RetCtBu). The following
signals were discernible for the anti rotamer, 1H NMR (C6D6): δ 1.16
(s, 9H, ReCH2C(CH3)3), 1.27 (s, 9H, RedCHC(CH3)3), 1.39 (s, 9H,
2
t
RetCC(CH3)3), 2.20 (d, JHaHb ) 12.6 Hz, 1H, ReCHcHd Bu), 3.14
2
t
Reaction of 1a with Ph3SiOH. In an NMR tube equipped with a
Teflon screw cap (provided by Young Scientific Ltd.), Ph3SiOH (12.0
mg, 43.5 µmol) was added at room temperature to a yellow-orange
solution of 1a (20.0 mg, 42.8 µmol) in 0.5 mL of C6D6. Ph3SiOH slowly
dissolved, and the color turned brown-red. Monitoring the reaction by
1H NMR spectroscopy showed the complete disappearance of 1a after
12 h at room temperature with the concomitant formation of neopentane
and 2m as a 10 to 1 mixture of its syn and anti isomers. After
evaporation of the reaction mixture, a red oil was obtained as an
inseparable mixture of 2m of its syn and anti rotamers in a 10/1 ratio.
1H NMR (C6D6): major isomer (syn) δ 1.10 (s, 9H, ReCH2C(CH3)3),
1.12 (s, 9H, RetCC(CH3)3), 1.24 (s, 9H, RedCHC(CH3)3), 2.62 (d,
(d, JHaHb ) 12.6 Hz, 1H, ReCHcHd Bu), 12.63 (s, 1H, RedCHtBu).
13C NMR: δ 28.4 (RetCC(CH3)3), 29.3 (RedCHC(CH3)3), 33.8
t
(ReCH2C(CH3)3), 42.1 (RedCHC(CH3)3), 47.3 (ReCH2 Bu), 54.1 (Ret
CC(CH3)3), 254.9 (RedCHtBu, JCH ) 156.8 Hz), 300.7 (RetCtBu).
1
The H and C assignments were established by HSQC and HMBC 2D
NMR experiments. 29Si NMR (C6D6): -99.9 (1Si), -65.7 (4Si), -65.2
(3Si).
Acknowledgment. We are grateful to Prof. R. R. Schrock
and Dr. E. A. Quadrelli for helpful discussions. We wish to
thank the C.N.R.S., the ESCPE-Lyon, and ENS-Lyon for
financial support. M.C. is also grateful to the French Ministry
for Education and Research for a fellowship (MENRT). The
EXAFS work was supported by the Director, Office of Science,
Office of Basic Energy Sciences, of the U.S. Department of
Energy under Contract No. DE-AC03-76SF00098. The EXAFS
work was performed at the Ernest O. Lawrence Berkeley
National Laboratory, which is operated by the U. S. DOE under
Contract No. DE-AC03-76SF00098, and at the Stanford Syn-
chrotron Radiation Laboratory, which is operated by the
Director, U. S. DOE, Office of Science, Office of Basic Energy
Sciences, Division of Chemical Sciences.
2JHaHb ) 13.1 Hz, 1H, ReCHaHb Bu), 3.07 (d, JHaHb ) 13.1 Hz, 1H,
t
2
t
ReCHaHb Bu), 7.0-7.5 (m, 9H, Ph), 7.6-8.0 (m, 6H, Ph), 10.86 (s,
1H, RedCHtBu) ppm. 13C NMR: δ 29.5 (RetCC(CH3)3), 31.6 (Red
CHC(CH3)3), 32.6 (ReCH2C(CH3)3), 32.7 (ReCH2C(CH3)3), 45.1 (Red
CHC(CH3)3), 48.0 (ReCH2tBu), 53.3 (RetCC(CH3)3), 128.0 (Ph), 130.0
(Ph), 135.5 (Ph), 137.4 (Ph), 245.9 (RedCHtBu, 1JCH ) 116 Hz), 292.4
ppm (RetCtBu). The following signals were discernible for the anti
rotamer of 2m. 1H NMR (C6D6): δ 1.14 (s, 9H, RetCC(CH3)3), 1.16
(s, 9H, ReCH2C(CH3)3), 1.21 (s, 9H, RedCHC(CH3)3), 2.26 (d, 2JHaHb
t
) 12.7 Hz, 1H, ReCHcHdtBu), 3.22 (d, 2JHaHb ) 12.7 Hz, 1H, ReCHcHd -
Bu), 12.65 (s, 1H, RedCHtBu). 13C NMR: δ 28.4 (RetCC(CH3)3),
29.4 (RedCHC(CH3)3), 33.9 (ReCH2C(CH3)3), 42.1 (RedCHC(CH3)3),
46.8 (ReCH2 Bu), 53.9 (RetCC(CH3)3), 253.8 (RedCHtBu, JCH
)
t
1
Supporting Information Available: NMR spectra and evolu-
tion of the 13C label on the carbenic carbon in 1c as a function
of time (PDF). This material is available free of charge via the
Internet at http://pubs.acs.org.
159 Hz), 301.2 (RetCtBu). The H and C assignments were established
by DEPT 135, HSQC, and HMBC 2D NMR experiments. The syn
assignment for the major rotamer is based on the difference in 1H
chemical shift and 1JCH coupling for the carbenic protons between the
syn and anti rotamers.1b
JA020136S
9
504 J. AM. CHEM. SOC. VOL. 125, NO. 2, 2003