Reactions of the rhenium allyl hydrido complex Cp*Re(η3-C3H5)(CO)(H) with the electrophiles CF3CO2H, CF3SO3H, NOBF4, [p-N2C6H4OMe][BF4] and [Ph3C][BF4]

Article abstract of DOI:10.1016/S0022-328X(98)00959-0

The rhenium hydrido allyl complex Cp*Re(η³-C₃H₅)(CO)(H) (1) reacted with NOBF₄ or [p-N₂C₆H₄OMe][BF₄] at -78°C to give [Cp*Re(η²-C₃H₆)(CO)(NO)][BF ₄] (2) or [Cp*Re(η²-C₃H₆)(CO)(N ₂C₆H₄OMe)][BF₄] (3), respectively. In these reactions addition of the electrophile is accompanied by transfer of the hydride ligand to the allyl group to form the propene ligand. Complex 1 reacted with [Ph₃C][BF₄] in CH₃CN at -78°C to abstract the hydride ligand and form the allyl acetonitrile complex [Cp*Re(η³-C₃H₅)(CO)(CH ₃CN)][BF₄] (4). Complex 1 reacted with CF₃CO₂H or CF₃SO₃H to form the hydrido propene complexes Cp*Re(η²-C₃H₆)(H)(CO)(CF ₃CO₂) (5) or Cp*Re(η²-C₃H₆)(H)(CO)(CF ₃SO₃) (6). Studies with CF₃CO₂D demonstrated that both Cp*Re(η²-C₃H₆)(D)(CO)(CF ₃CO₂) and Cp*Re(η²-C₃H₅D)(H)(CO)(CF ₃CO₂) are formed. This is consistent with a mechanism in which D⁺ attacks the metal followed by transfer of either D or H to the allyl group. Complex 1 was found not to react with C₂H₄, hexene, cyclohexene, MeCCMe, PhCCMe, CH₃CN or PMe₃.

Full text of DOI:10.1016/S0022-328X(98)00959-0

Journal of Organometallic Chemistry 572 (1999) 213–223
Reactions of the rhenium allyl hydrido complex
Cp*Re(p3-C3H5)(CO)(H) with the electrophiles CF3CO2H, CF3SO3H,
NOBF4, [p-N2C6H4OMe][BF4] and [Ph3C][BF4]
Ying-Xia He, Derek Sutton *
Department of Chemistry, Simon Fraser Uni6ersity, Burnaby, BC V5A 1S6, Canada
Received 28 May 1998
Abstract
The rhenium hydrido allyl complex Cp*Re(p3-C3H5)(CO)(H) (1) reacted with NOBF4 or [p-N2C6H4OMe][BF4] at −78°C to
give [Cp*Re(p2-C3H6)(CO)(NO)][BF4] (2) or [Cp*Re(p2-C3H6)(CO)(N2C6H4OMe)][BF4] (3), respectively. In these reactions
addition of the electrophile is accompanied by transfer of the hydride ligand to the allyl group to form the propene ligand.
Complex 1 reacted with [Ph3C][BF4] in CH3CN at −78°C to abstract the hydride ligand and form the allyl acetonitrile complex
[Cp*Re(p3-C3H5)(CO)(CH3CN)][BF4] (4). Complex 1 reacted with CF3CO2H or CF3SO3H to form the hydrido propene
complexes Cp*Re(p2-C3H6)(H)(CO)(CF3CO2) (5) or Cp*Re(p2-C3H6)(H)(CO)(CF3SO3) (6). Studies with CF3CO2D demonstrated
that both Cp*Re(p2-C3H6)(D)(CO)(CF3CO2) and Cp*Re(p2-C3H5D)(H)(CO)(CF3CO2) are formed. This is consistent with a
mechanism in which D+ attacks the metal followed by transfer of either D or H to the allyl group. Complex 1 was found not
to react with C2H4, hexene, cyclohexene, MeCꢀCMe, PhCꢀCMe, CH3CN or PMe3. © 1999 Elsevier Science S.A. All rights
reserved.
Keywords: Rhenium allyl hydrido complex; Electrophiles; Ligand
1. Introduction
be better explored [16]. Here, we report some of the
reactions with some electrophiles. Some attempted reac-
tions are also described. These include treatment with
some alkenes, alkynes and nitriles, with which 1 was
found to be unreactive.
Previous work from our group has reported the
formation of the half-sandwich rhenium p3-allyl hy-
drido complex Cp*Re(p3-C3H5)(CO)(H) (1), which was
obtained by photochemical C–H activation of the co-
ordinated propene ligand in Cp*Re(p2-C3H6)(CO)2
[1,2]. By comparison with most other transition metal
p3-allyl hydrido complexes reported in the literature
[1,3–15], complex 1 is relatively inert, and the endo and
exo isomers interconvert only slowly, which has allowed
the X-ray structures of both isomers to be obtained.
Complex 1 has recently been made non-photochemi-
cally in improved yield from the chloro compound
Cp*Re(p3-C3H5)(CO)Cl so that now its chemistry may
2. Results
2.1. Reactions of 1 with [NO]+, [N2C6H4OMe]+, and
[Ph3C]+
These reactions are shown in Scheme 1(a). Complex
1 reacted with [NO][BF4] in acetone at −78°C to
produce [Cp*Re(p2-C3H6)(CO)(NO)][BF4] (2). In this
reaction the attack of NO+ at the metal occurs with
migration of the hydride ligand to the p3-allyl group to
give a propene ligand. The IR spectrum of 2 exhibited
* Corresponding author. Tel.: +1-604-291-4885; fax: +1-604-291-
3765; e-mail: dsutton@sfu.ca.
0022-328X/99/$ - see front matter © 1999 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(98)00959-0

214
Y.-X. He, D. Sutton / Journal of Organometallic Chemistry 572 (1999) 213–223
Scheme 1. (a) Reactions of Cp*Re(p3-C3H5)(CO)(H) (1) with electrophiles: (i) NOBF4; (ii) [p-N2C6H4OMe][BF4]; (iii) [Ph3C][BF4]/MeCN; (iv)
CF3CO2H; (v) CF3SO3H; (vi) HBF4. (b). Propene position number scheme.
w(CO) at 2047, and w(NO) at 1765 cm−1. These are in
the typical regions for coordinated CO and NO absorp-
tions in similar cationic rhenium compounds [17–20].
acetonitrile complex 4, which was first synthesized pre-
viously from [Cp*Re(p3-C3H5)(CO)2][BF4] by using
PhIO in CH3CN to oxidatively remove one of the CO
ligands [23].
1
The H-NMR spectrum of 2 showed resonances indica-
tive of the presence of two diastereomers 2a and 2b in
a ratio of 1:1.1 by integration for the propene protons;
the two Cp* resonances were overlapped. There was no
change in this ratio when the sample in CDCl3 was kept
at room temperature (r.t) for 1 week. The two
diastereomers may be accounted for by considering the
position of attack of NO+ at the rhenium center as
shown in Scheme 2 and discussed below.
2.2. Protonation of Cp*Re(p3-C3H5)(CO)(H) (1)
When complex 1 was reacted with CF3CO2H at
−78°C in acetone-d6, the IR spectrum of the reaction
solution after 5 min showed a new w(CO) band at 1962
cm−1. The reaction was completed at about −60°C
and the product was isolated, purified, and identified by
spectroscopy as the new hydrido propene complex
Cp*Re(p2-C3H6)(H)(CO)(CF3CO2) (5) (Scheme 1). The
CI-MS of 5 gave a peak for the parent ion at m/z 506,
but a stronger peak at m/z 505 for [M+ –H], and a base
peak at m/z 391 which is consistent with loss of
CF3CO2H and one H atom from the parent ion. The
1H-NMR spectrum of isolated 5 in C6D6 showed all the
expected resonances for the propene protons, which
were assigned as a multiplet at l 3.22 for H3, two at l
2.60 and l 2.20 for the terminal methylene protons, and
a doublet at l 2.36 integrating for three methyl protons.
The hydride ligand gave a singlet at l −9.28, which
was in the expected region for a neutral rhenium hy-
dride complex by comparison with 1.
When
complex
1
was
treated
with
[N2C6H4OMe][BF4] in acetone at −78°C, the
[N2C6H4OMe]+ ion attacked the rhenium instead of
inserting into the Re–H bond [21], and the hydride
transferred to the p3-allyl group to produce the propene
complex [Cp*Re(p2-C3H6)(CO)(N2C6H4OMe)][BF4] (3).
The IR of 3 showed w(CO) at 1982, and w(NN) at 1711
cm−1, indicative of a singly-bent aryldiazenide ligand
1
[22]. The H-NMR spectrum of 3 showed a multiplet at
l 7.04 (integration 4H) for the C6H4 protons and a
singlet at l 3.87 for OCH3. The coordinated propene
proton resonances were assigned as multiplets at 4.10,
3.47, and l 3.10 for H3, H1 and H2, respectively, and a
doublet at l 2.16 for the methyl group (see Scheme
1(b)). The Cp* resonance was a singlet at l 2.17.
Complex 1 reacted with [Ph3C][BF4] in CH3CN at
−78°C to remove the hydride ligand and form the
The protonation of 1 was conducted again by using
CF3SO3H in either acetone-d6 at −78°C, or C6D6 at
r.t., and the hydrido propene complex Cp*Re(p2-
C3H6)(H)(CO)(CF3SO3) (6) was obtained (Scheme 1).

Y.-X. He, D. Sutton / Journal of Organometallic Chemistry 572 (1999) 213–223
215
Scheme 2. Scheme to show the formation of diastereomers 2a and 2b from the reaction of exo and endo 1 with NO+
.
1
The IR spectrum of 6 in C6D6 showed w(CO) at 1975
cm−1. The CI-MS of 6 gave a parent peak at m/z 542.
The H-NMR spectrum of 6 in C6D6 gave almost the
same propene resonances as did 5. The Cp* signal was
a singlet at l 1.58, and the major Re–H resonance was
at l −9.68. In addition, there were two minor hydride
resonances at l −9.51 and l −9.33.
2.3. Low temperature H-NMR study of the
protonation of 1
1
2.3.1. Protonation of Cp*Re(p3-C3H5)(CO)(H) (1) with
CF3CO2H in CD2Cl2
The above protonation of 1 with CF3CO2D gave
results that could be accommodated by attack of D+
at the metal or the metal–hydride bond with forma-
tion of a hydrido deuterio intermediate, followed by
transfer of either H or D to the allyl group. The
purpose of a low temperature study was therefore to
try to observe any such dihydride intermediate pro-
duced in the protonation reaction. It was also hoped
to observe whether different stereoisomers of a dihy-
dride intermediate might be produced, and, if so,
whether these might lead to different isomers of the
hydrido propene complex product observable at the
lower temperatures.
The CF3CO2H was added to a solution of 1 (mix-
ture of exo and endo isomers, ratio endo: exo ca. 3:1)
in CD2Cl2 at 173 K in an NMR tube, and the H-
NMR spectrum was recorded 5 min later at 183 K.
The hydride region of the spectrum showed two broad
signals at l −9.32 and l −9.13 in a ratio of 3.9:1.0.
The spectrum was obtained again at 183 K after a
further 10 min. It showed that the resonance intensity
at l −9.13 had decreased slightly relative to that at l
The protonation of 1 with CF3CO2D was conducted
in C6D6 in an attempt to investigate the reaction path-
way. A mixture of Cp*Re(p2-C3H5D)(H)(CO)(CF3CO2)
(7) and Cp*Re(p2-C3H6)(D)(CO)(CF3CO2) (8) was ob-
1
tained with the ratio 7:8=3.3:1 (see below). The H-
NMR spectrum of the mixture gave almost the same
resonances as for complex 5, except that the signal at l
2.33 (assigned to –CH2D and –CH3) is broad because
of coupling with the single D atom, and the central
proton H3 also showed a broad multiplet at l 3.20. The
2H{1H}-NMR of this mixture gave a broad signal at l
2.24, which corresponds to the deuterium in a CH2D
group, and is at a chemical shift similar to the proton
resonance of this group, confirming the presence of a
CH2D group in complex 7. A resonance at l −9.31,
corresponding to Re–D, indicated the presence of com-
1
2
plex 8. The integration of the H resonances indicated a
ratio of 7:8=3.3:1.
The protonation of 1 was also carried out with HBF4
in an attempt to observe an intermediate (possibly a
dihydrido complex) from proton addition prior to coor-
dination of the anion, since BF4− may be a poorer
ligand compared with CF3CO2− or CF3SO3−. However,
1
−9.32. The H-NMR spectrum was recorded at 203,
213 and 253 K over 3 h (Fig. 1), and the intensity
ratio of these resonances continued to change in fa-
vour of the l −9.32 resonance. When the tempera-
ture was increased to 294 K, the signal at −9.32
1
both the IR and the H-NMR indicated that a hydrido
propene complex was again produced which is formu-
lated as Cp*Re(p2-C3H6)(H)(CO)(FBF3) (9).

216
Y.-X. He, D. Sutton / Journal of Organometallic Chemistry 572 (1999) 213–223
Fig. 1. Variable temperature 1H-NMR spectra (Re–H resonances) following the addition of CF3CO2H to 1 in CD2Cl2 at 173 K. Spectrum f was
recorded in C6D6, all others in CD2Cl2).
2.3.2. Protonation of Cp*Re(p3-C3H5)(CO)(H) (1) with
shifted to l −9.20, and the one at −9.13 shifted to l
CF3SO3H in acetone-d6
−9.08, and was now of only small intensity (Fig. 1e).
At all temperatures, only one Cp* resonance was ob-
served, and was a singlet at l 2.03 at 294 K. The
resonances for the propene protons were very broad
(possibly indicating exchange) and were difficult to
assign. The sample was then pumped to dryness and
The reaction of 1 with CF3SO3H was conducted in
acetone-d6 by using the pure exo isomer of 1. The
1H-NMR spectrum of 1 in acetone-d6 was first recorded
at 213 K, and showed only the hydride resonance for
the exo isomer at l −9.68. Then, CF3SO3H was added
to a solution of 1 in acetone-d6 in an NMR tube at 183
K. A spectrum taken at 213 K 5 min after CF3SO3H
was added showed three hydride resonances at l
−9.19, l −9.30 and l −9.42 assigned to unknown
hydride species i–iii (Table 1). A sequence of spectra
with increasing temperature in the range 213–273 K
was obtained, and showed changes in the hydride reso-
nances as indicated in Table 1. The l −9.19 and l
−9.42 resonances generally decayed away as the tem-
perature was increased. At 273 K a further spectrum
was obtained 0.8 h later, and showed that a further
change had occurred with time while the temperature
was held constant. The temperature was lowered again
to 253 K, but the observed spectrum was not the same
as previously obtained at this temperature, indicating
that the changes were not reversible with temperature.
Instead, it appeared that at 273 K two species were
present (with hydride resonances at l −9.24 and l
1
redissolved in C6D6. The H-NMR spectrum in C6D6
gave a spectrum essentially identical to the one ob-
served for Cp*Re(p2-C3H6)(CO)(H)(CF3CO2) (5) in the
r.t. synthesis above. Only a single hydride resonance
occurred at l −9.33 (Fig. 1f), and the propene and
Cp* resonances were well-resolved.
It is conceivable that the two hydride resonances
initially observed result from inequivalent products
formed from the endo and exo isomers of 1, respectively
as the initial ratio of the resonances is somewhat similar
to the relative amounts of these isomers in the sample
of 1 employed. However, there is no direct evidence for
this. This, of course, could be tested by studying a pure
isomer of 1. This was done, as described next (but using
CF3SO3H in acetone for the protonation), and two
isomers of the propene complex again resulted, which
argues against this idea.

Y.-X. He, D. Sutton / Journal of Organometallic Chemistry 572 (1999) 213–223
217
Table 1
1H-NMR data for the temperature and time dependence of the species produced from the protonation of the exo isomer of 1 in acetone-d6 at 213
K
T (K)
Time (h)
Speciesa
i
ii
iii
6
l (ReH) (%)b
l (ReH) (%)b
l (ReH) (%)b
l (ReH) (%)b
213
223
233
243
253
263
273
273
253
253
294
0.00
0.50
1.18
1.68
2.65
2.95
3.32
4.12
4.47
4.63
72.0
−9.19; 42.1
−9.18; 36.7
−9.18; 35.2
−9.18; 15.4
−9.16; 13.3
−9.16; 7.5
−9.16; 0.0
−9.30; 32.6
−9.30; 30.1
−9.30; 36.7
−9.27; 62.0
−9.24; 43.8
−9.24; 37.3
−9.24; 65.3
−9.24; 43.5
−9.24; 40.4
−9.24; 35.5
−9.42; 25.3
−9.42; 33.2
−9.39; 28.0
−9.38; 22.5
−9.35; 23.7
−9.34; 22.3
−9.34; 0.0
−9.45; 19.2
−9.44; 33.0
−9.44; 34.7
−9.44; 56.5
−9.44; 59.7
−9.44; 64.6
−9.44; 100
a Species i–iii are unknown intermediates or isomers of 6, see text.
b The relative percentages of the species were obtained from the intensity of the hydride resonances.
−9.44), and on decreasing the temperature to 253 K,
and then increasing it from 253 to 294 K, the ratio
changed such that the resonance at l −9.24 eventually
disappeared, and only the one at l −9.44 remained.
The solvent was removed, and the spectrum was re-
recorded in C6D6. The H-NMR spectrum of this sam-
ple was identical to that previously recorded for
complex 6.
3. Discussion
In the reaction of 1 with NOBF4 the 1H-NMR
spectrum of product 2 gave two sets of propene reso-
nances. We assign these to the two possible
diastereomers 2a and 2b illustrated in Scheme 2. These
should ideally have different w(CO) and w(NO) IR
bands, and we presume that these are not resolved in
the polar solvents necessary to dissolve this ionic com-
plex. Each diastereomer has a corresponding enan-
tiomer 2a% and 2b%. The sample of 1 used for this
reaction was a mixture of both the endo and the exo
isomers. Scheme 2 shows that each of these can gener-
ate both diastereomers of 2. Theoretically, there are
three possible interligand positions for NO+ to attack
the rhenium in 1. These are numbered 1–3. We assume
that a four-legged piano stool intermediate is formed
after NO+ attack, as shown in square brackets in
Scheme 2. The NO ligand must be a bent 1e-donor at
this stage, and we assume the p3-allyl still adopts the
same orientation as it had in 1. (The formation of a
bent NO ligand from electrophilic NO+ attack at a
basic metal site is well established) [24]. If the hydride
transfers only to the nearest end of the allyl to produce
the propene ligand together with conversion of the NO
ligand from a 1e- to a 3e-donor to give 2, only the
reaction of 1 with NO+ in positions 1 and 2 will be
productive. This is because (assuming there is no fast
stereochemical reorganisation) the hydride will be in a
position trans to the allyl if NO+ attacks 1 at position
3, which will preclude transfer of the hydride to the
allyl. For a given enantiomer of exo-1, the two
diastereomers 2a and 2b can be envisaged to result from
NO+ attacking at the different positions 1 and 2,
respectively (Scheme 2). The other enantiomer of exo-1
1
2.4. Reaction of Cp*Re(p3-C3H5)(CO)(H) (1) with
alkenes and alkynes
We anticipated that 1 might react with alkenes to
give the corresponding alkyl complexes Cp*Re(p3-
C3H5)(CO)(R). Some of these are known complexes
that have been synthesized in this laboratory from the
chloro complex Cp*Re(p3-C3H5)(CO)Cl [16]. Complex
1 was treated with ethylene, hexene or cyclohexene at
r.t. in hexane. The IR spectrum showed no change in
the w(CO) absorption of 1 over 24 h. Similarly, 1
showed no reaction with PhCꢀCMe or MeCꢀCMe in
hexane at r.t. for 24 h.
2.5. Reaction of Cp*Re(p3-C3H5)(CO)(H) (1) with
CH3CN
A solution of 1 in pure CH3CN was stirred at r.t. for
24 h and was monitored by IR. No reaction occurred.
2.6. Reaction of Cp*Re(p3-C3H5)(CO)(H) (1) with
PMe3
Complex 1 was treated with PMe3 overnight in an
attempt to replace the CO to give Cp*Re(p3-
C3H5)(PMe3)(H). No reaction occurred.

218
Y.-X. He, D. Sutton / Journal of Organometallic Chemistry 572 (1999) 213–223
will then give the enantiomers 2a% and 2b%. These are all
also formed in a corresponding way from the endo
isomer of 1 as also shown in Scheme 2. It should
perhaps be pointed out that additional isomers of 2 are
possible if propene rotamers are considered. Chiral
half-sandwich rhenium nitrosyl alkene complexes simi-
lar to 2, but with PPh3 replacing CO have been exten-
sively studied by Gladysz and co-workers. The
orientation of the alkene CꢁC vector is consistent with
rhenium–alkene frontier orbital overlap considerations,
and steric preferences with regard to the other co-lig-
ands have been used to define the most likely alkene
rotamer in the observed diastereomers [25]. Here, we
have considered that the preferred rotamer has the
methyl group avoiding the bulky Cp* ligand, as shown.
It is also conceivable that alkene rotation scrambles any
possible isomers based on alkene rotamers. An entirely
different mechanism for the formation of 2 in which a
hydride-to-allyl migration occurs first, followed by co-
ordination of NO+ as a 2e-donor is considered to be
less likely, in view of the lack of reactivity of 1 with
other 2e-donors.
Because of the presence of the Re–H bond, there was
the possibility that 1 might react with an arenediazo-
nium ion to give an aryldiazene or an arylhydrazide
complex, formally the products of an insertion of the
diazonium ion into the metal–hydride bond [26–28].
However, 1 reacted with [N2C6H4OMe][BF4] in the way
that it did with NOBF4. The N2Ar ligand is coordi-
nated as a 3-electron donor and the H ligand has
migrated to the allyl to give a propene ligand. Com-
plexes containing an aryldiazenide ligand and an alkene
are uncommon [29]. Complex 3 is an addition to a
range of cationic rhenium half sandwich carbonyl aryl-
diazenido complexes of the type [Cp*ReLL%(N2Ar)]+
synthesized in this laboratory [30]. The hydride in 1
exhibits hydridic character in the reaction of 1 with
Ph3C+ and is simply abstracted to leave a vacant site
that is filled by the solvent MeCN yielding 4.
The protonation of 1 with CF3CO2H and CF3SO3H
resulted in the formation of the new rhenium hydrido
propene complexes Cp*Re(p2-C3H6)(CO)(H)(CF3CO2)
(5) and Cp*Re(p2-C3H6)(CO)(H)(CF3SO3) (6). This is
the first time, to our knowledge, that protonation of a
hydrido allyl complex has yielded a characterizable
propene complex. In the two previous cases that we
know of protonation has resulted in the elimination of
the propene. Henderson and co-workers observed the
production of propene in the protonation of Mo(H)(p3-
C3H5)(dppe)2 with HCl at low concentration [31]. The
protonation of Os(CO)(H)(p3-C3H5)(PR3)2 (R=PiPr3,
PMetBu2) with CH3CO2H, HBF4 and HCl was studied
by Schlu¨nken and Werner [5]. The elimination of
propene resulted in all cases, no matter which acid was
used for the protonation, and the formation of the
hydrido complex Os(CO)(H)(PR3)2(CH3CO2) was iden-
tified in the case of CH3CO2H. No intermediate
propene complex was observed. In our case, the spec-
troscopic data clearly indicated the coordination of the
propene ligand in complexes 5 and 6.
Because of the expected four-legged piano-stool ge-
ometry for complexes 5 and 6 there are three possible
positional geometric isomers in which the hydride lig-
and is trans to either CO, propene, or the oxyanion
ligand. In addition, each of these stereoisomers has a
diastereomeric partner where the propene is bound by
the opposite enantioface. For each of these six chiral
isomers there is a corresponding enantiomer. We do not
presently have enough evidence to assign the specific
structure of the isomer ultimately obtained for 5 and 6.
It may be a single one of these six isomers or the
1H-NMR could result from exchange between more
than one isomer. As shown in Scheme 3 and Scheme 4,
using exo-1 and protonation with CF3SO3H for illus-
tration, protonation at rhenium (or at the Re–H bond)
can give two possible dihydride intermediates, in which
the hydrides are either cis (Scheme 3) or trans (Scheme
4) in a four-legged piano-stool structure. The two
Schemes depict the possible product isomers that can
result in each case. In Scheme 3, two enantiomers I-˜
and I-™ for the cis dihydride I arise from the enan-
tiomers of exo-1. It is reasonable to assume that forma-
tion of the propene ligand results only from migration
of a H atom cis to an allyl CH2 group, and that
migration of a H atom trans to the allyl ligand does not
occur. Following H migration, attack of the oxyanion
may then occur at a site opposite any one of the H,
propene, or CO ligands in the unsaturated three-legged
piano-stool cationic intermediate III-˜ or its enan-
tiomer III-™. For III-˜ this generates isomers 6a-˜–6c-
˜, respectively. The enantiomer III-™ produces the
enantiomers 6a-™–6c-™. In 6a-˜–6c-˜ the propene is
bound to Re by the si face, whereas in 6a-™–6c-™ it is
bound by the re face. In a similar fashion, as shown in
Scheme 4, the trans dihydride II generates six corre-
sponding diastereomers 6a%-˜–6c%-˜ and 6a%-™–6c%-™ of
the above products, depending upon which of the two
hydrogens migrates to the allyl group, followed by
attack of the oxyanion as above. Thus, all six possible
stereoisomers and their enantiomers of 6 can in princi-
ple result, depending upon which of sites 1–3 opposite
the different basal ligands in exo-1 is thought to be the
site of initial proton attack. Note further that the allyl
complex 1 exists in exo and endo isomeric forms, and
each of these forms is capable of yielding all of the
above stereoisomers of 5 and 6, whether or not inter-
conversion of these forms occurs.
The study done using CF3CO2D in C6D6 shows the
formation of two singly D-labelled analogues of 6 in
unequal amounts. In the major product (7) the D has
been incorporated into the propene methyl group, while
the other product (8) has D bound to Re. The first of

Y.-X. He, D. Sutton / Journal of Organometallic Chemistry 572 (1999) 213–223
219
Scheme 3. Scheme illustarting the formation of isomers 6a-˜–6c-˜ from exo-1-˜ and the corresponding enantiomers 6a-™–6c-™ from exo-1-™
following protonation of these enantiomers of exo-1 at positions 1 or 2. Symbols h and i signify enantiomers.
these products, 7, (but clearly not 8) could be accounted
for by direct attack of D+ on the allyl ligand. On the
other hand, both products 7 and 8 are compatible with
the mechanism discussed above, and shown in Scheme
3 and Scheme 4. In this case, attack of D+ at the
metal–hydrogen bond or at the metal is envisaged to give
an intermediate cationic Re(H)(D) complex that can then
transfer either H or D to the allyl group. If the different
amounts of 7 and 8 observed to be produced are kinetic
in origin, this may indicate that both mechanisms are
operating. This possibility is consistent with Henderson’s
results on the protonation of Mo(H)(p3-C3H5)(dppe)2,
with HCl [31]. In that case, protonation was proposed
to occur at the metal center to give a dihydrido p3-allyl
cationic intermediate, or at the allyl ligand to give a
hydrido propene cationic intermediate, and these inter-
mediates lead to the elimination of the observed products
propyne and propene, respectively. However, if D+
attack is exclusively at the metal, or if 7 and 8 are in
equilibrium, then the equilibrium deuterium isotope
effect would also favor 7 (C–D bond) over 8 (Re–D
bond) Eq. ((1)) A rough estimate of KH/KD for 7:8,
ignoring statistical effects, can be made by using ballpark
wavenumber values w(Re–H) ca. 2000 cm−1 and
w(C–H) ca. 3000 cm−1 and the formula KH/KD=exp
{(mH−mD)/KBT}, where mH−mD=hc/2×(1−1/1.40)
(wCH−wReH) [32–36]. This gave a value of KH/KD=2.
((1))

220
Y.-X. He, D. Sutton / Journal of Organometallic Chemistry 572 (1999) 213–223
Scheme 4. Scheme illustrating the formation of isomers 6a%-˜-6c%-˜ and the corresponding enantiomers 6a%-™-6c%-™ following protonation of exo-1
at position 3 and subsequent choice of migration of a H atom to the allyl group. Symbols h and i signify enantiomers. Primed labels (e.g. 6a%)
signify a diastereomer of the unprimed label (e.g. 6a) arising from coordination of propene by the opposite enantioface.
unresolved mutual coupling and coupling to inequiva-
lent allyl terminal protons consistent with assignment
to the cis dihydride structure I in Scheme 3. The l
−9.42 resonance might therefore tentatively be as-
signed to the trans dihydride intermediate II in Scheme
4. As the temperature was changed, both of these
decayed away and the hydride resonance at l −9.30
underwent a complicated change in intensity. It is pos-
sible that this arises from one or more isomers of the
propene complex Cp*Re(p2-C3H6)(CO)(H)(CF3SO3)
(6). Overall, this resonance eventually decreases as the
resonance of l −9.44 increases, and presumably rep-
resents an isomer that is unstable with respect to the
final product.
Although it was hoped that the low temperature
protonation studies would allow observation of inter-
mediates such as the proposed cis or trans dihydride
complexes I or II, these experiments did not yield
unambiguously interpretable results, except to show
that the final products were the same as those formed
in the r.t. syntheses. While at the low temperatures
two (in the case of protonation with CF3CO2H in
CD2Cl2) or three (in the case of protonation with
CF3SO3H in acetone) hydride resonances were initially
observed, and converted to those of the final products
on increasing the temperature, we cannot be sure of
their assignment. There is no evidence specifically
pointing to assignment as one or other of the dihy-
dride intermediates I or II, though this is a possibility.
In the low temperature protonation with CF3SO3H the
resonance at l −9.19 is very broad and could be
composed of two inequivalent hydride resonances with
As expected in view of the previous reports in which
protonation of an allyl complex led to evolution of
propene [5,31], the hydrido propene complexes 5 and 6
are not very stable, but were sufficiently so for full

Y.-X. He, D. Sutton / Journal of Organometallic Chemistry 572 (1999) 213–223
221
spectroscopic characterization. It is likely that their
decomposition also results from the elimination of the
equipped with a Phrasor Inc. fast atom bombardment
accessory. Masses are given for the 187Re isotope. Cor-
rect isotopic distribution patterns were observed for all
parent peaks. Microanalyses were performed by M.K.
Yang of the SFU Microanalytical Laboratory.
propene. Therefore, the protonation of
1
with
CF3CO2H in acetone-d6 at −78°C was also conducted
in the presence of PMe3, in an attempt to obtain a
PMe3 complex Cp*Re(CO)(H)(PMe3)(CF3CO2) to sta-
bilize the unsaturated species produced after the elimi-
nation. The product in this reaction was complex 5,
instead of the expected compound. Possibly the added
PMe3 was largely protonated by CF3CO2H to form the
phosphonium ion [HPMe3]+.
The failure of 1 to react with alkenes, alkynes,
MeCN or PMe3 under the conditions utilized here
contrasts with the reactivity shown towards the elec-
trophilic unsaturated species NO+ and N2Ar+. Pre-
sumably these are prevented from binding to the metal
or inserting into the Re–H bond by the reluctance of 1
to provide a vacant site, either by an p3-p1 transition of
the allyl group or ring slippage of the Cp*.
5.2. [Cp*Re(p2-C3H6)(CO)(NO)][BF4] (2)
Complex 1 (mixture of endo/exo ca. 4/1, 13 mg, 0.033
mmol) was dissolved in acetone (2 ml). At −78°C,
NOBF4 (10 mg, 0.086 M in acetone) was added, and
the mixture was stirred for 10 min. By this time, the
solution changed from colorless to yellow, and the
temperature increased to −60°C. The solvent was
pumped off, the residue was extracted with CH2Cl2,
then recrystallized from ether/hexane. The pure product
was obtained as yellow crystals (16.1 mg, 0.032 mmol,
74%), m.p.: decomposed at 149°C. IR (CH2Cl2, cm−1):
wCO=2047, wNO=1765. FAB-MS (m/z): 422 (M+,
base), 380 (M+ –C3H6), 350 (M+ –C3H6–CO–2H).
1H-NMR (CDCl3, l), isomer 2a (or 2b): 4.27, (1H, m,
H3), 3.59 (1H, dd, J12=2.5 Hz, J13=15.0 Hz, H1), 3.29
(1H, d, J23=13.5 Hz, H2), 2.52 (3H, d, J=6.5 Hz,
CH3), 2.23 (15H, s, Cp*). Isomer 2b (or 2a): 3.93, (1H,
m, H3), 3.00 (1H, dd, J12=2.5 Hz, J13=9.0 Hz, H1),
2.78 (1H, d, J23=9.0 Hz, H2), 2.22 (15H, s, Cp*), 2.12
(3H, d, J=6.5 Hz, CH3). Anal. Calc. for
C14H21BF4NO2Re: C, 33.08; H, 4.16; N, 2.76. Found:
C, 32.85; H, 3.95; N, 3.06.
4. Conclusions
The most notable feature to emerge from this study is
the electrophile-induced migration of the hydride ligand
to the allyl group to give observable propene com-
plexes. In the reactions with NO+ and [N2C6H4OMe]+
attack occurs at the metal, and the nitrosyl and aryldi-
azenido ligands act as 3-electron donors, fulfilling the
required electron count in the resulting propene com-
plexes 2 and 3. In the corresponding reactions with the
strong acids CF3CO2H, CF3SO3H and HBF4, the re-
sults are compatible with protonation at the metal
center to give a dihydride intermediate, though proto-
nation at the p3-allyl or the Re–H bond must also be
considered. Again, migration of hydride to the allyl
group occurs and coordination of the anion is required
to fulfil the required electron count of the resulting
propene complex.
5.3. [Cp*Re(p2-C3H6)(CO)(N2C6H4OCH3)][BF4] (3)
Complex 1 (mixture of endo/exo ca. 8/1, 15 mg, 0.038
mmol) was dissolved in 2 ml acetone. At −78°C,
[p-N2C6H4OCH3][BF4] (10 mg, 0.09 M in 0.5 ml ace-
tone) was added to this solution. The solution immedi-
ately changed from colorless to orange and the IR
showed the disappearance of the w(CO) for 1, and a
new CO absorption at 1995 cm−1 for 3. The solvent
was pumped off, and the residue was recrystallized
from CH2Cl2/ether to give the pure product as brown
solid (12.3 mg, 0.021 mmol, 63%). IR (CH2Cl2, cm−1):
wCO=1982, wNN=1711. 1H-NMR (CDCl3, l): 7.04
(4H, m, C6H4), 4.10 (1H, m, H3), 3.87 (3H, s, OCH3),
3.47 (1H, m, H2), 3.10 (1H, dd, J13=13.0 Hz, J21=2.0
Hz, H1), 2.17 (15H, s, Cp*), 2.16 (3H, d, J=3.6 Hz,
CH3).
5. Experimental section
5.1. General procedures
All reactions were carried out under dry nitrogen in
Schlenk apparatus. Solvents were purified by standard
methods and were freshly distilled under dry nitrogen.
All reagents were obtained from Aldrich except where
mentioned. FT-IR spectra were recorded on a Bomem
Michelson-120 instrument in hexane, ether or C6D6
5.4. [Cp*Re(p3-C3H5)(CO)(NCCH3)][BF4] (4)
Complex 1 (mixture of endo/exo ca. 7/1, 12 mg, 0.031
mmol) was dissolved in 3 ml of freshly distilled
CH3CN. At −35°C, [Ph3C][BF4] (10 mg, 0.11 M in 0.5
ml CH3CN) was added to this solution, then stirred for
10 min. The IR showed disappearance of the w(CO)
band from 1, and a new w(CO) at 1971 cm−1 appeared.
1
solutions. H- and 13C-NMR spectra were recorded by
M.M. Tracey of the SFU NMR Service using a Bruker
WM-400 instrument operating at 400.13 and at 100.6
MHz. MS spectra were obtained by G. Owen on a
Hewlett-Packard Model 5985 GC-MS instrument

222
Y.-X. He, D. Sutton / Journal of Organometallic Chemistry 572 (1999) 213–223
the solvent was pumped off, and the residue was recrys-
tallized from CH2Cl2/ether to give 4 as a yellowish
solid. IR (CH3CN, cm−1): wCO=1971. 1H-NMR
(CDCl3, l): 4.83, (1H, m, Hc), 3.26 (1H, dd, Jsc=6.0
Hz, Jss=3.5 Hz, Hs), 3.11 (1H, dd, Jsc=6.0 Hz, Jss=
3.5 Hz, Hs), 2.80, (3H, s, CH3CN), 2.03 (1H, d, Jac=
11.6 Hz, Ha), 1.96 (15H, s, Cp*), 1.32 (1H, d, Jac=9.0
Hz, Ha).
min. The IR showed only one new CO band at 1971
cm−1. The solvent was pumped off, and the residue
was dried for 5 h. The pure product was obtained as a
yellowish solid (mixture of 7 and 8, ratio 7:8=3:1; 18.9
mg, 0.037 mmol, 81%). IR (C6D6, cm−1): wCO=1971.
CI-MS (m/z): 465 (M+ –C3H6), 463 (M+ –C3H6–D,
base), 394 (M+ –CF3CO2), 393 (M+ –CF3CO2H), 392
1
(M+ –CF3CO2D). H-NMR (C6D6, l): 3.20, (1H, m,
H3), 2.59 (1H, d, J13=12.0 Hz, H1), 2.33 (2H, br,
CH2D), 2.18 (1H, d, J23=8.8 Hz, H2), 1.55 (15H, s,
5.5. Cp*Re(p2-C3H6)(H)(CO)(CF3CO2) (5)
2
Cp*), −9.32 (1H, s, Re–H). H{1H}-NMR (C6D6, l):
Complex 1 (mixture of endo/exo ca. 5/1, 20 mg, 0.051
mmol) was dissolved in acetone-d6. At −78°C, five
drops of CF3CO2H were added to the solution, which
was then stirred for 1 h. The IR showed that the
reaction was not finished at this time. The reaction was
continued for another hour while the temperature was
raised to −60°C. The IR showed the disappearance of
the CO absorption from 1, and appearance of a new
CO band at 1962 cm−1 for 5. The solvent was pumped
off, and the residue was extracted with hexane. This
hexane solution was pumped dry for 12 h, and the pure
product was obtained as a yellowish solid. IR (hexane,
cm−1): wCO=1981. CI-MS (m/z): 506 (M+), 505
(M+ –H), 465 (M+ –C3H5), 463 (M+ –C3H6–H), 421
2.24 (s, CH2D), −9.31 (s, Re–D).
5.8. Cp*Re(p2-C3H6)(H)(CO)(FBF3) (9)
Complex 1 (mixture of endo/exo ca. 4/1, 15 mg, 0.038
mmol) was dissolved in C6D6 (2 ml). At r.t., HBF4
(three drops, 48% ether solution) was added and the
solution was stirred for 5 min. The IR of the solution
showed a new w(CO) band at 1971 cm−1 for the
product, and the w(CO) of 1 disappeared. The solvent
was pumped off, and the residue was dried under
vacuum for 8 h. The product was obtained as a brown-
ish solid. IR (C6D6, cm−1): wCO=1971. CI-MS (m/z):
1
393 (M+ –FBF3, base). H-NMR (C6D6, l): 2.73, (1H,
1
(M+ –C3H5–CO2), 391 (M+ –CF3CO2H–H, base). H-
m, H3), 2.69 (1H, d, J13=13.0 Hz, H1), 2.47 (1H, d,
J23=5.8 Hz, H2), 2.33 (3H, d, J=6.0 Hz, CH3), 1.60
(15H, s, Cp*), −9.45 (1H, s, Re–H). There are two
minor hydride resonances at −9.30 and l −9.26.
NMR (C6D6, l): 3.22, (1H, m, H3), 2.60 (1H, d,
J13=12.0 Hz, H1), 2.36 (3H, d, J=6.7 Hz, CH3), 2.20
(1H, d, J23=8.5 Hz, H2), 1.59 (15H, s, Cp*), −9.28
(1H, s, Re–H).
5.9. Protonation of 1 with CF3COOH in CD2Cl2 at
low temperature
5.6. Cp*Re(p2-C3H6)(H)(CO)(CF3SO3) (6)
Complex 1 (mixture, endo/exo ca. 5/1, 18 mg, 0.046
mmol) was dissolved in C6D6 (2 ml). At r.t., CF3SO3H
(three drops) was added to this solution, which was
then stirred for 10 min. The IR of the solution showed
a new w(CO) band at 1975 cm−1, and the w(CO) of 1
disappeared. The solvent was pumped off, and the
residue was dried under vacuum for 3 h. The product
was obtained as a brownish solid (18.5 mg, 0.034 mmol,
74%). IR (C6D6, cm−1): wCO=1975. CI-MS (m/z): 542
(M+), 541 (M+ –H), 407 (M+ –CF3SO2–2H), 391
Complex 1 (mixture, endo/exo ca. 3/1, 15 mg, 0.038
mmol) was dissolved in 1 ml CD2Cl2 in a H-NMR
1
tube. This was then placed in a Schlenk tube, degassed
three times, cooled to 173 K, and then placed in the
1
probe. The H-NMR spectrum was recorded at 183 K,
which showed a mixture of endo and exo isomers of 1 in
a ratio of endo/exo=3/1. The sample was removed
from the probe, and five drops of CF3CO2H were
immediately added to this solution at 173 K. The NMR
1
tube was again placed in the probe, and the H-NMR
1
(M+ –CF3SO3H–H, base). H-NMR (C6D6, l): 4.00,
spectrum was recorded at 183 K after 5 min and
showed two hydride resonances at l −9.32 and l
−9.13. The spectrum was obtained again 10 min later,
and showed only a slight decrease of the signal at l
−9.13. The temperature was then raised to 203, 213,
253, and 294 K, and the 1H-NMR spectra at these
temperatures were recorded. As the temperature in-
creased, the resonance at l −9.13 decreased, and the
one at l −9.32 increased. At 294 K, the signal at l
−9.13 disappeared, and the one at l −9.32 was the
only remaining hydride resonance. The resonances of
the propene and the allyl protons of the possible iso-
mers were broad and overlapped, which made the
(1H, m, H3), 2.63 (1H, d, J13=12.2 Hz, H1), 2.39 (3H,
d, J=6.0 Hz, CH3), 2.19 (1H, dd, J23=8.8 Hz, J21=
1.0 Hz, H2), 1.58 (15H, s, Cp*), −9.68 (1H, s, Re–H).
There were two minor hydride resonances at −9.51
and l −9.33.
5.7. Cp*Re(p2-C3H5D)(H)(CO)(CF3CO2) (7) and
Cp*Re(p2-C3H6)(D)(CO)(CF3CO2) (8)
Complex 1 (mixture of endo/exo ca. 3/1, 18 mg, 0.046
mmol) was dissolved in 2 ml C6D6 at r.t., CF3CO2D
(five drops) was added, and the mixture stirred for 10

Y.-X. He, D. Sutton / Journal of Organometallic Chemistry 572 (1999) 213–223
223
assignments of these signals impossible. Only one Cp*
References
resonance was observed in all the temperature range, it
was a singlet at l 2.03. The product resulting from the
low temperature protonation was pumped to dryness,
and redissolved in C6D6. The 1H-NMR spectrum of it in
C6D6 was identical to that of obtained for complex 5.
[1] R.J. Batchelor, F.W.B. Einstein, R.H. Jones, J.M. Zhuang, D.
Sutton, J. Am. Chem. Soc. 111 (1989) 3468.
[2] J.M. Zhuang, D. Sutton, Organometallics 10 (1991) 1516.
[3] R.C. Schnabel, D.M. Roddick, Organometallics 15 (1996) 3550.
[4] M. Schulz, H. Werner, J. Organomet. Chem. 470 (1994) 243.
[5] C. Schlunken, H. Werner, J. Organomet. Chem. 454 (1993) 243.
[6] J.E. Bercaw, J.R. Moss, Organometallics 11 (1992) 639.
[7] W.D. McGhee, R.G. Bergman, J. Am. Chem. Soc. 110 (1988)
4246.
5.10. Protonation of 1 with CF3SO3H in acetone-d6 at
low temperature
[8] W.D. McGhee, R.G. Bergman, J. Am. Chem. Soc. 107 (1985)
3388.
Complex 1 (pure exo isomer, 15 mg, 0.038 mmol) was
dissolved in acetone-d6 in an NMR tube, degassed as
above, and the 1H-NMR spectrum of it was recorded at
213 K, which showed only one resonance in the hydride
region as expected for the pure exo isomer of 1. The
sample was then removed from the probe, and five drops
of CF3SO3H were added at 183 K. This gave a suspen-
sion because of the insolubility of CF3SO3H. The sample
was replaced in the probe set at 213 K, and the spectrum
was recorded after 5 min at 213 K, which showed three
hydride resonances at l −9.19, l −9.30 and l −9.42.
The temperature was then raised in 10°C steps to 273 K,
and the spectrum was recorded. As the temperature
increased, the resonances at l −9.19 and l −9.42
decreased, and the one at l −9.30 showed a compli-
cated change. At 253 K, there were four hydride reso-
nances at l −9.16, l −9.24, l −9.35 and l −9.45,
respectively. At 273 K, the 1H-NMR spectrum was
recorded twice at different times, and the result indicated
that the resonance at l −9.30 decreased as the time
increased, and the one at l −9.44 increased. At 294 K,
the other three resonances disappeared, and the one at
l −9.44 was the only signal in the hydride region (Table
1). The resonances of the propene and the allyl protons
of the species were broad and overlapped, which made
the assignments of these signals impossible. Several Cp*
resonances were observed in all the temperature range,
[9] R.H. Heyn, K.G. Caulton, J. Am. Chem. Soc. 115 (1993) 3354.
[10] T.H. Tulip, J.A. Ibers, J. Am. Chem. Soc. 101 (1979) 4201.
[11] E.O. Sherman, M. Olson, J. Organomet. Chem. 172 (1979) C13.
[12] G. Carturan, A. Scrivanti, F. Morandini, Angew. Chem. Int. Ed.
Engl. 20 (1981) 112.
[13] R. Bertani, G. Carturan, A. Scrivanti, Angew. Chem. Int. Ed.
Engl. 22 (1983) 246.
[14] J. Schwartz, J. Mol. Catal. 72 (1992) 373.
[15] M.D. Fryzuk, X. Gao, S.J. Rettig, J. Am. Chem. Soc. 117 (1995)
3106.
[16] Y. He, D. Sutton, J. Organomet. Chem. 538 (1997) 49.
[17] C. Roger, T.S. Peng, J.A. Gladysz, J. Organomet. Chem. 439
(1992) 163.
[18] J.R. Sweet, W.A.G. Graham, Organometallics 1 (1982) 982.
[19] J.R. Sweet, W.A.G. Graham, Organometallics 2 (1983) 135.
[20] J.R. Sweet, W.A.G. Graham, J. Organomet. Chem. 217 (1981)
C37.
[21] B.L. Haymore, J. Organomet. Chem. 137 (1977) C11.
[22] G.C-Y. Kim, R.J. Batchelor, X. Yan, F.W.B. Einstein, D.
Sutton, Inorg. Chem. 34 (1995) 6163.
[23] R.J. Batchelor, F.W.B. Einstein, Y. He, D. Sutton, J.
Organomet. Chem. 468 (1994) 183.
[24] (a) D.J. Hodgson, N.C. Payne, J.A. McGinnety, R.G. Pearson,
J.A. Ibers, J. Am. Chem. Soc. 90 (1968) 4486. (b) D.J. Hodgson,
J.A. Ibers, Inorg. Chem. 7 (1968) 2345. (c) B.F.G. Johnson, B.L.
Haymore, J.R. Dilworth in: G. Wilkinson, R.D. Gillard, J.A.
McCleverty, (Eds.), Comprehensive Coordination Chemistry,
vol. 2, Pergamon Press, Oxford, England, 1987, p. 99.
[25] (a) J.A. Gladysz, B.J. Boone, Angew. Chem. Int. Ed. Engl. 36
(1997) 550. (b) T.-S. Peng, J.A. Gladysz, J. Am. Chem. Soc. 114
(1992) 4174. (c) G.S. Bodner, T.-S. Peng, A.M. Arif, J.A.
Gladysz, Organometallics 9 (1990) 1191.
1
and they were at l 2.0–d 2.1. Finally, the H-NMR
spectrum of the ultimate product from this low temper-
ature protonation was measured in C6D6 at r.t., and was
found to be identical to that of obtained for complex 6.
[26] G.W. Parshall, J. Am. Chem. Soc. 89 (1967) 1822.
[27] J.A. Carroll, D. Sutton, Inorg. Chem. 19 (1980) 3137.
[28] F.W.B. Einstein, T. Jones, A.J.L. Hanlan, D. Sutton, Inorg.
Chem. 21 (1982) 2585.
[29] X. Yan, F.W.B. Einstein, D. Sutton, Can. J. Chem. 73 (1995)
939.
[30] A. Cusanelli, R.J. Batchelor, F.W.B. Einstein, D. Sutton,
Organometallics 13 (1994) 5096.
Acknowledgements
This work was supported by the Natural Sciences and
Engineering Research Council of Canada through a
research grant to D.S., and by the Simon Fraser Univer-
sity through a Graduate Research Fellowship to Y.H.
We are indebted to Professors J.M. Mayer and A.
Bennet for advice on the calculation of equilibrium
isotope effects.
[31] K.E. Oglieve, R.A. Henderson, J. Chem. Soc. Chem. Commun.
(1993) 474.
[32] E. Buncel, C.C. Lee (Eds.), Isotopes in Organic Chemistry, vol.
6, Elsevier, Amsterdam, 1984, p. 161.
[33] H.C. Urey, D. Rittenberg, J. Chem. Phys. 1 (1933) 137.
[34] J. Bigeleisen, M.G. Mayer, J. Chem. Phys. 15 (1947) 261.
[35] J. Bigeleisen, J. Chem. Phys. 17 (1949) 675.
[36] B.R. Bender, J. Am. Chem. Soc. 117 (1995) 11239.
.