Indenyl rhenium alkyne complexes: CO substitution via alkyne-assisted ring slippage and CO-catalyzed phosphine substitution

Article abstract of DOI:10.1021/om020884q

CO substitution of the indenyl alkyne complex (η⁵-C₉H₇)(CO)₂ Re(η²-MeC≡CMe) (3) to form (η⁵-C₉H₇)Re(CO)₃ (8) is much faster than that of either the indenyl alkene complex (η⁵-C₉H₇)(CO)₂ Re(η²-cis-MeHC=CHMe) (2) or the cyclopentadienyl alkyne complex (η⁵-C₅H₅)(CO)₂ Re(η²-MeC≡CMe) (5). The acceleration of the reaction of 3 with CO is proposed to involve cooperativity between slippage of the indenyl ring from an η⁵ 6e-donor to an η³ 4e-donor and shifting of the alkyne ligand from a 2e- to a 4e-donor as CO associates with the rhenium complex. The rate of reaction of 3 with CO showed a linear dependence on CO pressure between 1 and 10 atm, consistent with CO-promoted indenyl ring slippage. At very high CO pressure, 3 is rapidly converted to (η¹-C₉H₇)(CO)₄ Re(η²-MeC≡CMe) (12). Upon release of CO pressure, 12 reverts to a 12:1 mixture of starting material 3 and CO substitution product 8 at a rate that is inversely dependent on CO pressure. When the reaction of ¹³CO with 3 was carried to about 50% conversion, extensive labeling of starting material 3 and product 8 was observed. The distribution of label in 3 and 8 requires largely stereospecific ¹³CO addition to form mainly one isomer of an η³-indenyl intermediate (η³-C₉H₇)(CO)₂(¹³-CO) Re(η²-MeC≡CMe) (B-(¹³CO)₁) along with a minor amount of a second stereoisomer of B-(¹³CO)₁ and the involvement of double-labeled η¹-indenyl complex 12-(¹³CO)₂. The reaction of 3 with PPh₃ to form the phosphine complex (η⁵-C₉H₇)(CO)₂ Re(PPh₃) (9) is catalyzed by ¹³CO. The observation of ¹³CO in 9 is consistent with the proposal that 9 is formed by trapping a reactive intermediate in the CO substitution reaction of 3.

Full text of DOI:10.1021/om020884q

Organometallics 2003, 22, 1183-1195
1183
Articles
In d en yl Rh en iu m Alk yn e Com p lexes: CO Su bstitu tion
via Alk yn e-Assisted Rin g Slip p a ge a n d CO-Ca ta lyzed
P h osp h in e Su bstitu tion
Charles P. Casey,* Thomas E. Vos, J ohn T. Brady, and Randy K. Hayashi
Department of Chemistry, University of Wisconsin-Madison, Madison, Wisconsin 53706
Received October 23, 2002
CO substitution of the indenyl alkyne complex (η5-C9H7)(CO)2Re(η2-MeCtCMe) (3) to form
(η5-C9H7)Re(CO)3 (8) is much faster than that of either the indenyl alkene complex
(η5-C9H7)(CO)2Re(η2-cis-MeHCdCHMe) (2) or the cyclopentadienyl alkyne complex (η5-
C5H5)(CO)2Re(η2-MeCtCMe) (5). The acceleration of the reaction of 3 with CO is proposed
to involve cooperativity between slippage of the indenyl ring from an η5 6e-donor to an η3
4e-donor and shifting of the alkyne ligand from a 2e- to a 4e-donor as CO associates with
the rhenium complex. The rate of reaction of 3 with CO showed a linear dependence on CO
pressure between 1 and 10 atm, consistent with CO-promoted indenyl ring slippage. At very
high CO pressure, 3 is rapidly converted to (η1-C9H7)(CO)4Re(η2-MeCtCMe) (12). Upon
release of CO pressure, 12 reverts to a 12:1 mixture of starting material 3 and CO substitution
product 8 at a rate that is inversely dependent on CO pressure. When the reaction of 13CO
with 3 was carried to about 50% conversion, extensive labeling of starting material 3 and
product 8 was observed. The distribution of label in 3 and 8 requires largely stereospecific
13CO addition to form mainly one isomer of an η3-indenyl intermediate (η3-C9H7)(CO)2(13-
CO)Re(η2-MeCtCMe) (B-(13CO)1) along with a minor amount of a second stereoisomer of
B-(13CO)1 and the involvement of double-labeled η1-indenyl complex 12-(13CO)2. The reaction
of 3 with PPh3 to form the phosphine complex (η5-C9H7)(CO)2Re(PPh3) (9) is catalyzed by
13CO. The observation of 13CO in 9 is consistent with the proposal that 9 is formed by trapping
a reactive intermediate in the CO substitution reaction of 3.
Sch em e 1
In connection with our search for heterobimetallic
dihydrides that might serve as powerful new reducing
agents, we synthesized the heterobimetallic dihydride
Cp(CO)2Re(µ-H)Pt(H)(PPh3)2 (1) by reaction of Cp-
(CO)2ReH2 with (CH2dCH2)Pt(PPh3)2.1,2 We were de-
lighted to find that 1 reduced alkynes to produce high
yields of rhenium alkene complexes and platinum
alkyne complexes (Scheme 1). We also found that Pt(0)
complexes catalyzed the reaction of trans-Cp(CO)2ReH2
with alkynes to give rhenium alkene complexes. Unfor-
tunately, because of the high stability of rhenium alkene
complexes, these reactions were stoichiometric in rhe-
nium.
Since the stability of these Cp(CO)2Re(alkene) com-
plexes toward substitution interfered with the develop-
ment of a catalytic process, we began investigations of
related (η5-indenyl)(CO)2Re(alkene) complexes which we
anticipated would undergo alkene exchange more rap-
idly. Substitution of the indenyl ligand for the cyclo-
pentadienyl ligand in metal complexes often leads to
increased rates of substitution, usually referred to as
the “indenyl effect”. Basolo found that the rate of PPh3
substitution on (η5-C9H7)Rh(CO)2 was 108 times faster
than the analogous Cp system (Scheme 2).3 The rate
dependence on phosphine concentration indicated that
the reaction proceeded via an associative process. A
mechanism involving simultaneous addition of PPh3 to
rhodium and indenyl ring slippage from η5 to η3
coordination was proposed. Mawby showed that the
indenyl ligand can also increase the rates of substitution
in dissociative reactions. The reaction of (η5-C9H7)Fe-
(CO)2I with P(OEt)3 proceeded 575 times faster than the
analogous Cp system.4 Initially the indenyl effect was
explained in terms of enhanced stability of the η3-
indenyl due to increased aromaticity of the six-mem-
(1) Casey, C. P.; Rutter, E. W., J r.; Haller, K. J . J . Am. Chem. Soc.
1987, 109, 6886.
(2) (a) Casey, C. P.; Rutter, E. W., J r. J . Am. Chem. Soc. 1989, 111,
8917. (b) Casey, C. P.; Wang, Y.; Tanke, R. S.; Hazin, P. N.; Rutter, E.
W., J r. New J . Chem. 1994, 18, 43.
(3) Rerek, M. E.; Basolo, F. J . Am. Chem. Soc. 1984, 106, 5908.
(4) J ones, D. J .; Mawby, R. J . Inorg. Chim. Acta 1972, 6, 157.
10.1021/om020884q CCC: $25.00 © 2003 American Chemical Society
Publication on Web 02/08/2003

1184 Organometallics, Vol. 22, No. 6, 2003
Casey et al.
Sch em e 2
bered ring upon slippage. More recently, it has also been
noted that the η5-indenyl ligand is bound to the metal
less strongly than an η5-Cp ligand due to weaker
bonding to the carbons shared with the aryl ring, and
it has been suggested that η5-indenyl binding might be
better described as “η3 + η2” binding.5
Here we report the synthesis of (η5-C9H7)(CO)2Re(η2-
cis-MeHCdCHMe) (2) and (η5-C9H7)(CO)2Re(η2-Me-
CtCMe) (3) and their substitution reactions. In the
course of studying these substitution reactions, we
uncovered a unique cooperative interaction between an
indenyl ligand and an alkyne ligand that leads to
unusually fast processes. The ability of the alkyne to
function as either a 2e- or a 4e-donor apparently
stabilizes the transition state for CO-promoted η5- to
η3-indenyl ring slippage. Further evidence for ring
slippage was obtained by the observation of the
η1-intermediate (η1-C9H7)(CO)4Re(η2-MeCtCMe) (12) at
very high CO pressure. We have also found that CO
catalyzes the substitution of PPh3 for butyne in 3. This
is the first example of a CO-catalyzed phosphine sub-
stitution reaction. While it is now clear that the indenyl
system will not provide sufficient acceleration of alkene
dissociation to be useful in heterobimetallic catalytic
systems, the results reported here provide new insight
into the mechanism of substitution reactions that are
often of crucial importance in catalytic processes.
F igu r e 1. X-ray crystal structure of (η5-C9H7)(CO)2Re(η2-
cis-MeHCdCHMe) (2).
Sch em e 3
room temperature over several hours to form a yellow-
orange solution of (η5-C9H7)(CO)2Re(η2-cis-MeHCd
CHMe) (2), which was isolated as a yellow-orange
crystalline solid (43%). The structure of 2 was estab-
lished spectroscopically and confirmed by X-ray crystal-
lography (Figure 1). The indenyl ligand is bound with
three shorter distances to carbon [Re-C(8), 2.267(7) Å;
Re-C(9), 2.283(8) Å; Re-C(10), 2.302(8) Å] and two
longer distances to the carbons shared with the aryl ring
[Re-C(11), 2.382(7) Å; Re-C(7), 2.393(8) Å]; the folding
angle Ω of the indenyl ligand is 7(1)°. The angle between
the CdC bond of coordinated 2-butene and the indenyl
ring is 22°.
The indenyl rhenium THF complex 7 reacted with
excess 2-butyne in THF over 1 h to produce the alkyne
complex (η5-C9H7)(CO)2Re(η2-MeCtCMe) (3), which was
isolated as an orange solid (35%) (Scheme 3). The
structure of 3 was established spectroscopically and
confirmed by X-ray crystallography (Figure 2). As in the
case of alkene complex 2, the indenyl ligand of 3 is
bound with three shorter distances to carbon [Re-C(8b),
2.229(13) Å; Re-C(9b), 2.286(13) Å; Re-C(10b), 2.320-
(13) Å] and two longer distances to the carbons shared
with the aryl ring [Re-C(11b), 2.409(13) Å; Re-C(7b),
2.424(13) Å]; the folding angle Ω of the indenyl ligand
is 2(2)°. The angle between the CtC bond of coordinated
2-butyne and the indenyl ring is 6°. The methyl groups
of 2-butyne are bent away from Re by 27°.
Resu lts
One important obstacle to developing alkyne hydro-
genation catalysts based on the chemistry of Cp(CO)2Re-
(µ-H)Pt(H)(PPh3)2 is the high thermal stability of the
resulting rhenium alkene products. The Cp(CO)2Re-
(alkene) complexes do not release alkene until heated
above 140 °C. For example, treatment of a C6D6 solution
of Cp(CO)2Re(η2-cis-MeHCdCHMe) (4) with excess 2-
butyne at 140 °C for 33 h led to only 31% conversion of
4 to a mixture of Cp(CO)2Re(η2-MeCtCMe) (5) (16%)
and unidentified products (15%). In search of more labile
alkene complexes, we began investigations of the
indenyl complex (η5-C9H7)(CO)2Re(η2-cis-MeHCdCHMe)
(2) as a possibly more reactive alternative.
Syn t h esis of (η5-C9H 7)(CO)2R e(η2-cis-MeHCdC-
HMe) (2) a n d (η5-C9H7)(CO)2Re(η2-MeCtCMe) (3).
The indenyl rhenium THF complex (η5-C9H7)Re(CO)2-
(THF) (7) obtained by photolysis of (η5-C9H7)Re(CO)3 (8)
proved to be a valuable intermediate for the synthesis
of alkene, alkyne, and phosphine complexes. A yellow
THF solution of 7 reacted with excess cis-2-butene at
(5) (a) Calhorda, M. J .; Veiros, L. F. Coord. Chem. Rev. 1999, 186,
37. (b) Veiros, L. F. Organometallics 2000, 19, 3127. (c) Calhorda, M.
J .; Roma˜o, C. C.; Veiros, L. F. Chem. Eur. J . 2002, 8, 868. (d) Kubas,
G. J .; Kiss, G.; Hoff, C. D. Organometallics 1991, 10, 2870.

Indenyl Rhenium Alkyne Complexes
Organometallics, Vol. 22, No. 6, 2003 1185
Sch em e 4
MeCHdCdCH2))6 at 80 °C under 2.3 atm 13CO for 7 h
led to complete disappearance of alkyne complex 5, 16%
conversion to C5H5Re(CO)3 (6), and decomposition to
uncharacterized black solids. Thus, the reactions of both
4 and 5 with CO proved to be so slow that quantitative
rate measurements were not readily determined.
Rea ction of CO w ith (η5-C9H7)(CO)2Re(η2-cis-
MeHCdCHMe) (2) a n d (η5-C9H7)(CO)2Re(η2-MeCt
CMe) (3). The reactions of indenyl alkene and alkyne
complexes 2 and 3 with CO were much faster than those
of the corresponding Cp complexes. The conversion of 2
to (η5-C9H7)Re(CO)3 (8) occurred over several days in
C6H6 solution at 25 °C under 4.1 atm CO. The reaction
of 2 with CO at 25 °C was monitored by IR analysis of
0.3 mL aliquots removed from a pressure reaction
vessel. The rate of disappearance of 2 was first order in
F igu r e 2. X-ray crystal structure of (η5-C9H7)(CO)2Re(η2-
MeCtCMe) (3).
2-Bu tyn e Exch a n ge w ith (η5-C9H7)(CO)2Re(η2-
cis-MeHCdCHMe) (2). The indenyl alkene complex
(η5-C9H7)(CO)2Re(η2-cis-MeHCdCHMe) (2) proved to
have high thermal stability like its Cp analogue 4.
Reaction of excess 2-butyne with alkene complex 2 in
C6D6 at 140 °C led to the formation of the alkyne
complex (η5-C9H7)(CO)2Re(η2-MeCtCMe) (3) (maximum
29% yield) along with decomposition products. The
disappearance of alkene complex 2 at 140 °C was
rhenium through three half-lives (kobs of 6.4 × 10-6 s-1
,
t1/2 ) 30 h, ∆Gq at 25 °C ≈ 24.5 kcal mol-1). The
carbonylation of indenyl alkene complex 2 was much
faster than the carbonylation of Cp alkene complex 4.
The indenyl ligand accelerated the rate of CO substitu-
tion but was ineffective in promoting reaction of 2 with
2-butyne. CO reacted with 2 under very mild conditions
(4.1 atm CO, 25 °C), but reaction of 2 with 2-butyne
required heating to 140 °C for days.
1
monitored by H NMR spectroscopy at 25 °C. The rate
of disappearance of 2 was first order in rhenium to over
3 half-lives (kobs ) 4.1 × 10-6 s-1, t1/2 ) 2 days, ∆Gq at
140 °C ) 34.6 kcal mol-1). The reverse reaction where
alkyne complex 3 reacted with excess cis-2-butene
proceeded at lower temperature (85 °C) (kobs ) 5.0 ×
10-7 s-1, t1/2 ) 16 days, ∆Gq at 85 °C ) 31.4 kcal mol-1
)
The reaction of indenyl alkyne complex 3 with CO was
27 times faster than the reaction of indenyl alkene
complex 2 with CO. When a C6H6 solution of 3 under
4.1 atm CO at 25 °C was rapidly stirred, formation of
(η5-C9H7)Re(CO)3 (8) occurred over several hours (Scheme
4). The rate of disappearance of the carbonyl bands of
3 at 1943 and 1869 cm-1 was first order in rhenium
through 3 half-lives (kobs of 1.8 × 10-4 s-1, t1/2 ) 1.1 h).
Examination of the rates of reaction of CO with the
indenyl and Cp complexes of alkenes and alkynes
indicates that the special combination of indenyl and
alkyne ligands leads to an unusually fast reaction. This
apparent cooperativity led us to investigate the reaction
of 3 with CO in greater detail in an attempt to
understand how the indenyl and alkyne ligands work
together to accelerate substitution reactions.
P r essu r e Dep en d en ce of th e CO Su bstitu tion of
3. The rate of formation of 8 from 3 in the presence of
CO was measured over a wide range of CO pressures
(1-120 atm) (Table 1, Figure 3). The rate of CO
substitution in the pressure range of 1-10 atm CO was
measured by following the appearance of the IR bands
of 8 and the disappearance of the IR bands of 3. To avoid
depleting the concentration of CO in solution, the
solutions were rapidly stirred and the concentration of
rhenium complexes (4-8 mM) was kept low compared
with the concentration of CO in solution (8-30 mM).7
The rate of CO substitution of 3 was also followed by
NMR in the pressure range 2-5 atm. While NMR
and cleanly produced 2-butene complex 2 (only ∼5%
uncharacterized decomposition products). The activation
barrier for substitution of alkyne complex 3 was 3.2 kcal
mol-1 lower than that of alkene complex 2. The slow
rate of alkene-alkyne exchange of these indenyl com-
plexes indicates that use of an indenyl analogue of
alkyne hydrogenation catalyst Cp(CO)2Re(µ-H)Pt(H)-
(PPh3)2 (1) would not provide the improvement that we
had hoped for.
Attem p ted CO Su bstitu tion Rea ction s of C5H5-
(CO)2Re(η2-cis-MeHCdCHMe) (4) an d C5H5(CO)2Re-
(η2-MeCdCMe) (5). To establish a baseline of reactivity
for comparison with reactions of indenyl complexes, we
first investigated the reactions of C5H5(CO)2Re(η2-cis-
MeHCdCHMe) (4) and C5H5(CO)2Re(η2-MeCtCMe) (5)
with CO. When a C6H6 solution of Cp alkene complex 4
was heated at 80 °C under 4 atm of CO, no reaction
was observed after 3 days (∆Gq at 80 °C > 31 kcal
mol-1). This is consistent with the high thermal stability
of 4 observed in its slow reaction with 2-butyne at 140
°C.
The reaction of CO with C5H5(CO)2Re(η2-MeCtCMe)
(5) was also very slow. Treatment of a C6D6 solution of
5 with 4 atm of 13CO at 25 °C led to only 3.3% conversion
of 5 to C5H5Re(CO)3 (6) after 24 days (∆Gq at 25 °C >
28 kcal mol-1). Treatment of a C6D6 solution of
C5H5(CO)2Re(η2-MeCtCMe) (5) (containing 0.35 equiv
of 2,6-dimethylpyridine to prevent the acid-catalyzed
isomerization to the allene complex C5H5(CO)2Re(η2-2,3-
(6) Casey, C. P.; Brady, J . T. Organometallics 1998, 17, 4620-4629.

1186 Organometallics, Vol. 22, No. 6, 2003
Casey et al.
Ta ble 1. CO P r essu r e Dep en d en ce on th e Ra te of
F or m a tion of (η5-C9H7)Re(CO)3 (8) a t 25 °C fr om
(η5-C9H7)(CO)2Re(η2-MeCtCMe) (3)
Sch em e 5
CO (atm) [CO] (mM) [Re] (mM) kobs × 104 (s-1
)
t1/2 (min)
1.1a
2.1a
8
15
8
10
5
0.53
1.1
1.8
1.8
0.60
1.3
1.5
1.7
2.4
2.6
4.8
7.4
8.2
8.1
8.3
8.0
4.2
8.0
9.7
9.5
8.7
6.9
5.6
217
103
63
66
193
90
80
66
49
45
24
16
14
14
14
14
28
14
12
12
13
16
19
3.0a
21
29
15
20
26
28
28
36
4.1a
8
2.2b
58
55
46
66
57
41
2
3
2
2
5
2
5
5
5
6
4
4
4
3.0b
3.9b
4.3b
4.2b
5.3b
10.9c
17.6c
27.6c
37.9c
40.8c
76
123
193
265
285
699
71
143
238
475
535
713
839
100c
of appearance of 8 by two different methods. In the first
method, we determined the rate of appearance of 8 by
two-point kinetic analysis. We added a C6H6 solution
of 3 into a 30 mL stainless steel bomb equipped with a
stir bar to ensure rapid mixing of gas with the solution.8
The bomb was placed in a 25 °C water bath and
pressurized with 38 atm of CO. After 10 min, CO and
solvent were rapidly evaporated and the residue was
analyzed by 1H NMR spectroscopy. Integration showed
a 1.0:0.6 ratio of 3 to 8. Assuming a first-order depen-
dence on rhenium, the rate of appearance of 8 was
determined (kobs ) 8.3 × 10-4 s-1, t1/2 ) 14 min). When
the reaction of 3 with CO was carried out under 18 atm
of CO, a similar rate was observed (kobs ) 7.4 × 10-4
s-1, t1/2 ) 16 min). Plots of kobs versus CO pressure
leveled off at high pressure (Figure 3). This analysis
rests on the assumptions that the reaction is first order
in 3 (as observed at low pressures) and that the major
species in solution is still 3. Several mechanisms that
might account for the rate leveling at higher CO
pressure (>18 atm) will be discussed before a second
and more accurate method of monitoring the rate of
formation of 8 at CO pressures between 10 and 120 atm
is presented.
10.2d
20.4d
34.0d
68.0d
76.5d
102.0d
120.0d
a
Rate measured by IR spectroscopy after release of CO pres-
sure. Rate measured by in situ 1H NMR spectroscopy. c Reaction
b
run in high-pressure reaction vessel and mixture analyzed by 1H
d
NMR spectroscopy after release of pressure. Rate measured by
in situ IR spectroscopy.
The near linear increase in the rate of appearance of
8 versus CO pressure from 1 to 5 atm is consistent with
an associative mechanism involving initial addition of
CO to 3 followed by subsequent loss of alkyne to
generate 8. The near zero-order rate dependence on CO
at higher pressure requires a change in the rate-
determining step to one in which the major species
present in solution reacts in a first-order process to
generate an intermediate that is converted to product
8. The first mechanism that we considered was an
unprecedented non-nucleophile-assisted ring slippage to
the η3-indenyl intermediate A followed by trapping of
the intermediate by CO to form B (Scheme 5). The η3-
indenyl intermediate A is suggested to be stabilized by
the alkyne ligand which can switch from a 2e- to a 4e-
donor as ring slippage occurs. At high CO pressure,
intermediate A is efficiently trapped by CO to form B
and the formation of A becomes the rate-determining
step. A prediction of this mechanism is that at very high
F igu r e 3. CO pressure dependence of the rate of formation
of (η5-C9H7)Re(CO)3 (8) from (η5-C9H7)(CO)2Re(η2-MeCt
CMe) (3). b ) kobs from IR spectroscopy after release of
CO pressure. [ ) kobs from in situ 1H NMR spectroscopy.
9 ) kobs from reaction run in high-pressure reaction vessel
and product mixture analyzed by 1H NMR spectroscopy
after release of pressure. 2 ) kobs from in situ IR spectros-
copy.
allowed in situ monitoring, the limited gas/liquid surface
area in NMR tubes limits the rate of gas dissolution and
can result in mass transfer limited rates. Despite this
problem, the rates measured by the two methods are
in fair agreement with one another. The rate of appear-
ance of 8 measured by NMR spectroscopy and IR
spectroscopy showed a linear dependence on CO pres-
sure over the range from 1 to 10 atm. At higher
pressures of CO (>10-120 atm), we measured the rate
(7) The concentration of CO in solution was calculated using the
Ostwald coefficient: (a) Solubility Data Series: Carbon Monoxide;
Cargill, R. W., Ed.; Pergamon Press: Oxford, U.K., 1990; Vol. 43, p
113. (b) J ust, G. Z. Phys. Chem., Stoechiom. Verwandtschaftsl. 1901,
37, 342. (c) Park, J .; Yi, X.; Gasem, K. A. M.; Robinson, R. L., J r. J .
Chem. Eng. Data 1995, 40, 245.
(8) As a test for maintenance of the CO concentration in solution,
the rates of reaction of 3 at 40 atm CO were run at 1.7 and 4.9 mM
concentrations of 3. If the reactions were CO starved, then the effect
of CO starvation should have been more pronounced at the higher
concentrations of 3 where CO depletion would be accentuated. How-
ever, similar rates were observed.

Indenyl Rhenium Alkyne Complexes
Organometallics, Vol. 22, No. 6, 2003 1187
13CO pressure no incorporation of 13CO into recovered
starting material will be observable. A second possible
mechanism involves the buildup of an η3- (B) or η1-
indenyl complex (12) at high CO pressure. Once the new
species is the major species present, the pressure
dependence on CO should disappear. This type of
mechanism requires that the new species be detectable
by direct monitoring of the reaction mixtures at high
pressure.
for other η1-indenyl rhenium complexes. Resonances at
δ 55.4 and 11.1 for 2-butyne bound to 12, resonances at
δ 67.6 and 10.9 for 2-butyne bound to 3, and resonances
at δ 72.6 and -0.8 for free 2-butyne were also seen. We
also observed one 13C resonance in the rhenium carbonyl
chemical shift region for 12 at δ 190.4 along with 13C
resonances for 3 (δ 204.1) and 8 (δ 194.2). While this
method was useful for observing and identifying 12 in
situ, the highest ratio of 12:8:3 observed was 30:20:50.
The reaction of 3 with 10-120 atm of CO was
monitored in a stainless steel pressure reactor equipped
with a React-IR probe.9 For example, when a solution
of 3 in C6H6 in a stainless steel pressure reactor was
placed under 68 atm of CO, IR carbonyl bands of 3
rapidly disappeared in 2-3 min and two new peaks not
Higher enrichment of 12 was needed for measurement
of the rate of loss of CO and alkyne from 12. We
succeeding in synthesizing a 92:4:4 mixture of 12:8:3
by cooling a C6D5CD3 solution of 3 in a stainless steel
pressure vessel at -10 °C, adding 100 atm of CO,
stirring for 1 h, cooling to -78 °C, and then releasing
pressure. The resulting solution of highly enriched 12
was transferred to NMR tubes cooled to -78 °C.
associated with 8 or 3 appear at 1946 and 1995 cm-1
.
The new peaks slowly disappeared and peaks for 8 at
1927 and 2024 cm-1 appeared. A plot of ln(([8]inf - [8])/
[8]inf) versus time was linear through 3 half-lives,
indicating the appearance of 8 over time followed first-
order kinetics (kobs ) 9.5 × 10-4 s-1, t1/2 ) 12 min). The
rate of appearance of 8 was measured for a number of
CO pressures between 10 and 120 atm of CO (Table 1,
Figure 3). The rate of appearance of 8 showed an inverse
dependence on CO pressure at these high pressures. The
previous apparent leveling of the rate of formation of 8
at high CO pressure when the reactions were monitored
by release of pressure and NMR analysis is now at-
tributed to the inadequacy of this analysis procedure.
(η1-C9H7)Re(CO)4(η2-MeCtCMe) (12). While the
ReactIR allowed us to observe a new major species (12)
in the reaction of (η5-C9H7)(CO)2Re(η2-MeCtCMe) (3)
with 68 atm of CO, we could not positively identify this
species from an IR spectrum of a mixture of 8, 3, and
12. However, 1H NMR spectroscopy provides a very
powerful method for differentiating between η5-, η3-, and
η1-indenyl binding modes to rhenium.10
Ster eoch em istr y of (η1-C9H7)Re(CO)4(η2-MeCt
CMe) (12). 13C NMR and IR spectroscopy established
the trans relationship of the alkyne and η1-indenyl
ligands of 12. In the 13C NMR spectrum, only one 13CO
resonance at δ 190.4 was observed, consistent with the
trans ligand assignment; the cis isomer has four differ-
ent CO ligands. The IR spectrum had two CO bands at
1946 (br, m) and 1995 (s) cm-1. Since the near C4v
symmetry of the trans isomer is broken by the alkyne
ligand to approximate C2v symmetry, two IR bands are
expected. We suggest that the alkyne is aligned parallel
to one pair of trans CO’s and that the alkyne and this
pair of CO’s compete for back-bonding from the same
metal d-orbital. The IR band for this pair of parallel
CO’s therefore appears at substantially higher fre-
quency (1995 cm-1) than the IR band (1946 cm-1) for
the pair of CO’s perpendicular to the alkyne ligand. The
alternative cis isomer would have near Cs symmetry,
and three strong CO bands and a weak CO band would
have been expected.
CO Loss fr om (η1-C9H 7)R e(CO)4(η2-MeCtCMe)
(12). Upon warming a C6D5CD3 solution of 12 to -9 °C,
conversion to a 12:1 mixture of alkyne complex 3 (δ 5.36)
In considering optimum conditions for NMR observa-
tion of the new species, we noted that the species was
observable by IR spectroscopy only above ∼20 atm CO.
Our hypothesis that the new species was either an η3-
or η1-indenyl complex formed by addition of one or two
CO ligands predicted that ∆S for the formation of the
addition product would be strongly negative and adduct
formation would be more favorable at low temperature.
A solution of 3 in C6D5CD3 was placed under 10 atm
CO in a thick walled NMR tube and maintained at -15
1
and tricarbonyl complex 8 (δ 5.70) was observed by H
NMR spectroscopy during the first half-life. The ratio
of 3:8 decreased at longer times due to an increase in
the amount of CO in solution released from 12.
The rate of disappearance of the δ 4.5 1H NMR
resonance of (η1-C9H7)Re(CO)4(η2-MeCtCMe) (12) was
measured between -24 and 2 °C. The headspace of the
NMR tube was left under vacuum to facilitate the
removal of released CO. The rate was first order in 12
at -9 °C through 3 half-lives (kobs ) 1.3 × 10-4 s-1, t1/2
) 87 min). Between -4 and 2 °C, the rate of disappear-
ance of 12 was determined during the first half-life to
avoid inhibition by released CO. An Eyring plot yielded
activation parameters of ∆Hq ) 17.9 ( 1.0 kcal mol-1
and ∆Sq ) -8.3 ( 3 eu for the loss of CO from 12. These
activation parameters were used to estimate the rate
of CO loss of 12 at 25 °C as 7.7 × 10-3 s-1 (t1/2 ) 1.5
min).
1
°C for 6 h. The H NMR spectrum at -65 °C showed
that the solution contained (η1-C9H7)Re(CO)4(η2-MeCt
CMe) (12) along with 8 and 3. The structure of 12 is
supported by the observation of seven new and distinct
indenyl proton resonances as expected for an η1-indenyl
complex. The resonance for coordinated butyne in 12
appeared at δ 1.78; resonances for bound butyne of 3
(δ 2.10) and for free butyne (δ 1.62) were also seen. In
the 13C NMR spectrum, there were nine distinct reso-
nances for indenyl carbons with the C1 resonance
shifted to much lower frequency (δ 19.4) as observed
F or m a tion of (η1-C9H7)Re(CO)5 (13). At 100 atm
of CO pressure, 8 reacts slowly with CO to form (η1-
C9H7)Re(CO)5 (13). The rate of disappearance of the
1929 cm-1 band of 8 followed pseudo-first-order kinetics
for over 3 half-lives (kobs ) 4.76 × 10-5 s-1, t1/2 ) 4 h,
∆Gq ) 23.4 kcal mol-1). 13 was stable at room temper-
(9) A 30 mL PARR stainless steel pressure vessel equipped with a
SiComp probe of a ASI ReactIR infrared spectrometer. SiComp probe
is an attenuated total reflectance (ATR) device with silicon as the ATR
surface.
(10) Chemical shift and the proton resonance pattern are distinctive
for the three different modes of coordination. O’Connor, J . M.; Casey,
C. P. Chem. Rev. 1987, 87, 307.

1188 Organometallics, Vol. 22, No. 6, 2003
Casey et al.
Ta ble 2. P r essu r e Dep en d en ce of th e Ra te of
Rea ction of CO w ith
Ta ble 3. 13CO La bel in (η5-C9H7)Re(CO)3 (8) a fter
Com p lete Rea ction of 3 w ith 13CO
(η5-C9H7)(CO)2Re(η2-cis-MeHCdCHMe) (2) a t 25 °C
pressure (atm)
% 8-(13CO)1
% 8-(13CO)2
% 8-(13CO)3
CO (atm) [CO] (mM) [Re] (mM) kobs × 105 (s-1
)
t1/2 (h)
1
2
3
4
5.3
15
67 ( 6
72 ( 6
54 ( 6
45 ( 6
45 ( 6
32 ( 6
23 ( 4
20 ( 4
30 ( 4
31 ( 4
30 ( 4
26 ( 4
10 ( 6
8 ( 6
16 ( 6
24 ( 6
25 ( 6
43 ( 6
2.0a
2.7a
14
19
28
190
285
606
11
7
13
5
0.15
0.23
0.64
1.9
128
84
30
10
5
4.1a
27.2b
40.8b
87c
5
4.0
5
7.4
2.6
Ta ble 4. 13CO In cor p or a tion in to
(η5-C9H7)(CO)2Re(η2-MeCtCMe) (3) a fter ∼50%
Con ver sion to (η5-C9H7)Re(CO)3 (8)
a
Rate measured by IR spectroscopy after release of CO pres-
b
sure. Reaction run in high-pressure reaction vessel and mixture
analyzed by 1H NMR spectroscopy after release of pressure. c Rate
measured by in situ IR spectroscopy.
13CO pressure
(atm)
% 3-
% 3-
% 3-
%
13CO in
3:% 8
8
(
13CO)0
(
13CO)1
(
13CO)2
1
2
3
4
12
46% 77 ( 4
49% 65 ( 4
50% 62 ( 4
46% 64 ( 4
29% 64 ( 4
18 ( 4
24 ( 4
24 ( 4
23 ( 4
22 ( 4
4 ( 4
0.26
0.37
0.37
0.43
0.88
ature for days and lost CO to re-form 8 upon heating at
90 °C. The rate of disappearance of 13 at 90 °C (t1/2 10
min) was determined by monitoring the disappearance
of its proton resonance (H1) at δ 3.9.
12 ( 4
13 ( 4
14 ( 4
14 ( 4
(η1-C9H7)Re(CO)5 (13) is kinetically much more robust
than η1-indenyl alkyne complex 12, which loses CO
rapidly at room temperature (rate extrapolated to 25
°C ) 7.7 × 10-3 s-1, t1/2 ) 1.5 min). Similarly, addition
of CO to (η5-C9H7)Re(CO)3 (8) to produce 13 is much
slower than addition of CO to alkyne complex (η5-C9H7)-
Re(CO)2(MeCtCMe) (3). Thus, an alkyne complex not
only accelerates CO addition to η5-indenyl complexes but
also accelerates CO dissociation from η1-indenyl com-
plexes. In Os(CO)4(η2-CF3CtCCF3), the alkyne in-
creases the labilization of CO ligands cis to the alkyne
by a factor of 106 compared to the labilization of CO from
the parent Os(CO)5 complex.11 Similar cis labilization
of CO by the alkyne ligand in complex 12 explains the
rapid loss of CO from 12.
P r essu r e Dep en d en ce of th e CO Su bstitu tion of
Alk en e Com p lex 2. The rate of reaction of CO with
alkene complex 2 showed a linear dependence on CO
pressure over the entire range from 1 to 87 atm (Table
2), in contrast to observations for alkyne complex 3.
When the reaction of 2 under 87 atm CO was monitored
in situ by IR spectroscopy, no η1-indenyl complex was
observed and the rate of disappearance of 2 was the
same as the rate of appearance of 8 (kobs ) 7.4 × 10-5
s-1, t1/2 ) 2.6 h).
Ta ble 5. 13CO La bel in (η5-C9H7)Re(CO)3 (8) a fter
50% Con ver sion of 3
13CO pressure
(atm)
conversion
% 8-
% 8-
% 8-
to 8
(
13CO)1
(
13CO)2
(
13CO)3
1
2
3
3.6
4
12
13
46%
49%
50%
42%
46%
29%
50%
91 ( 6
82 ( 6
83 ( 6
89 ( 6
82 ( 6
82 ( 6
65 ( 6
8 ( 4
1 ( 6
6 ( 8
7 ( 6
3 ( 6
6 ( 6
6 ( 6
12 ( 4
12 ( 4
8 ( 4
12 ( 4
12 ( 4
23 ( 4
12 ( 6
carbon monoxide from Cambridge Isotope Laboratories
contained approximately 10% of 13C18O along with 90%
13C16O. A significant increase in multiply labeled 8 was
seen as the 13CO pressure was raised.
The appearance of multiple labels in tricarbonyl
product 8 could have resulted from exchange into
starting material 3 during the course of reaction, from
exchange into a reactive intermediate during the con-
version of 3 to 8, or from exchange into product 8. A
control experiment excluded exchange into product 8
under the reaction conditions. Treatment of a C6D6
solution of 8 with 5 atm of 13CO for 2 days at room
temperature showed no change in the 13C NMR spec-
trum. This requires that exchange of 13CO into 8 be at
least 500 times slower than the conversion of 3 to 8.
The reaction of 3 with 13CO was followed by 13C NMR
spectroscopy to determine if the label was incorporated
into 3 during the course of reaction. When the reaction
of 3 with 4.2 atm of 13CO at 25 °C was followed by 13C
NMR spectroscopy, the 13CO resonance of 3 at δ 204.1
initially increased at a rate similar to that of the 13CO
resonance for 8 at δ 194.3. This indicated roughly
comparable rates of 13CO exchange into 3 and of
conversion of 3 to 8. In another set of experiments using
various 13CO pressures, the reactions of 3 with 13CO
were stopped after about 50% conversion, and both 8
and 3 were isolated by thin-layer chromatography and
analyzed by mass spectrometry (Tables 4 and 5).
Significant 13CO incorporation into recovered 3 (23-
50%, depending on pressure) was seen after about 50%
conversion to 8 and 13CO incorporation into 3 increased
somewhat as the 13CO pressure was raised. The relative
rate of 13CO incorporation into 3 compared with conver-
sion of 3 to 8 is given by {[3-(13CO)1] + [3-(13CO)2]}/[8]
and increased with CO pressure (Table 4). Figure 4 is
The linear dependence on CO pressure is consistent
with an associative process involving initial addition of
CO to 2 followed by loss of alkene. The overall rate of
CO addition and alkene loss is much slower than that
of CO addition and alkyne loss.
Reaction of 13CO with (η5-C9H7)(CO)2Re(η2-MeCt
CMe) (3). The reaction of 13CO with 3 was studied over
the pressure range 1-15 atm. Reactions were run to
completion, and the tricarbonyl complex (η5-C9H7)Re-
(CO)3 (8) was isolated and its isotopic composition
analyzed by electron impact mass spectrometry. The
cluster of peaks at m/e 384-389 due to the isotopomers
of 8 (37% 185Re, 63% 187Re) was used to determine the
13C labeling pattern. The ratio of 8-(13CO)1:8-(13CO)2:
8-(13CO)3 was determined by fitting the observed ratio
of peaks in the 384-389 range as the weighted sum of
the patterns calculated for one, two, or three 13CO labels
(Table 3). The expected isotopic patterns of 8-(13CO)1:
8-(13CO)2:8-(13CO)3 took into account that the labeled
(11) Pearson, J .; Cooke, J .; Takats, J .; J ordan, R. B. J . Am. Chem.
Soc. 1998, 120, 1434.

Indenyl Rhenium Alkyne Complexes
Organometallics, Vol. 22, No. 6, 2003 1189
Sch em e 7
also seen for phosphine complex 9: 35 ( 4% 9-(13CO)0:
44 ( 4% 9-(13CO)1:21 ( 4% 9-(13CO)2.
The cooperativity between CO and PPh3 in accelerat-
ing the rate of disappearance of 3 is consistent with
PPh3 trapping of a reactive intermediate reversibly
formed during the reaction of 3 with CO. The high
concentration of phosphine in solution leads to more
efficient trapping of this reactive intermediate than
when 13CO is the only trapping agent present. The
sequential addition of 13CO and PPh3 generates an
intermediate that then loses alkyne and either CO or
13CO. This explains the high incorporation of 13CO into
9.
F igu r e 4. CO pressure dependence of the relative rates
of 13CO incorporation into 3 compared with conversion of
3 to 8. The relative rate is given by the ratio of {[3-(13CO)1]-
+ [3-(13CO)2]}:[8] at partial conversion of 3 to 8.
Sch em e 6
For m ation of (η1-C9H7)Re(CO)2(CNt-Bu )2(η2-MeCt
CMe) (11). t-BuNC is isoelectronic with CO and often
is used as a model for CO; unlike CO, it is easy to obtain
high concentrations of t-BuNC in solution. The reaction
of 3 with t-BuNC was investigated for comparison with
the reaction of 3 with CO. Treatment of a CD2Cl2
solution of alkyne complex 3 with slightly less than 1
equiv of t-BuNC (0.2 M) at room temperature for several
hours led to the formation of the t-BuNC substitution
product (η5-C9H7)(CO)2Re(CNt-Bu) (10) (Scheme 7). The
1H NMR spectrum of 10 showed typical η5-indenyl shifts
at δ 7.43 (m), 7.02 (m), 5.64 (d), and 5.53 (t) for the ring
hydrogens as well as a singlet at δ 1.35 for the t-BuNC
ligand.
consistent with a combination of pressure-independent
and pressure-dependent routes for this rate ratio.
P P h 3 Tr a p p in g of Rea ctive In ter m ed ia tes in th e
CO Su b st it u t ion of (η5-C9H 7)(CO)2R e(η2-MeCt
CMe) (3). In an effort to trap intermediates in the
reaction of (η5-C9H7)(CO)2Re(η2-MeCtCMe) (3) with CO,
we studied carbonylation in the presence of PPh3. We
first explored the reaction of PPh3 with 3 without added
CO. The reaction of 3 with PPh3 in C6D6 at 25 °C slowly
produced the known phosphine complex (η5-C9H7)(CO)2-
Re(PPh3).12 The rate of disappearance of 3 was moni-
1
tored by H NMR spectroscopy following the disappear-
When the reaction of 3 with less than 1 equiv of
t-BuNC was carried out in CD2Cl2 at -78 °C, rapid
formation of the η1-indenyl complex (η1-C9H7)Re(CO)2-
(CNt-Bu)2(η2-MeCtCMe) (11) was observed. The 1H
NMR spectrum at -70 °C showed only starting alkyne
complex 3 and η1-indenyl complex 11 in solution. The
1H NMR spectrum of 11 showed two resonances at δ
1.45 and 1.42 for the diastereotopic t-BuNC ligands, an
alkyne methyl resonance at δ 2.41, and seven distinct
indenyl resonances for the η1-indenyl hydrogens. Sub-
sequent warming of the solution to room temperature
for several hours resulted in the formation of t-BuNC
complex 10.14
ance of the resonance at δ 2.05 of the 2-butyne ligand.
The rate was first order in rhenium and first order in
PPh3 (kobs ) 8.6 × 10-7 s-1, t1/2 ) 225 h, 0.21 M PPh3;
kobs ) 17 × 10-7 s-1, t1/2 ) 113 h, 0.41 M PPh3). Basolo
previously showed that (η5-C9H7)Re(CO)3 reacts with
phosphines through an associative mechanism to give
(η5-C9H7)(CO)2Re(phosphine) and suggested an indenyl
ring slip mechanism for the reaction.13 Significantly, the
reaction of 3 with PPh3 was more than 100 times slower
than the reaction of 3 with CO.
Under 4 atm 13CO, the reaction of 3 with 0.42 M PPh3
at 25 °C was greatly accelerated and a 1.0:5.0 mixture
of (η5-C9H7)Re(CO)3 (8):(η5-C9H7)Re(CO)2Re(PPh3) (9)
was formed (Scheme 6). Surprisingly, the disappearance
of 3 was faster for the system with both 13CO and PPh3
(kobs ) 5.1 × 10-4 s-1, t1/2 ) 22 min, 0.42 M PPh3, 4 atm
CO) than the reaction with only PPh3 (kobs ) 1.7 × 10-6
s-1, t1/2 ) 113 h, 0.41 M PPh3) or only 13CO present
(kobs ) 1.5 × 10-4 s-1, t1/2 ) 66 min, 4.1 atm CO).
Both 8 and 9 were isolated by TLC and analyzed by
mass spectrometry to determine the amount of 13CO
incorporation into each. (η5-C9H7)Re(CO)3 (8) showed
significant multiple labeling: 43 ( 6% 8-(13CO)1:38 (
6% 8-(13CO)2 :19 ( 6% 8-(13CO)3. Multiple labeling was
1
The H NMR spectrum of η1-indenyl complex 11 at
room temperature showed evidence for a fluxional
process involving a 1,3 shift of the η1-indenyl ligand.
This process interchanges the environments of the
indenyl hydrogens and results in the coalescence of the
seven resonances seen at low temperature to four broad
resonances at δ 7.41, 6.94, 6.33, and 4.35. The fluxional
process also interchanges the environments of the
t-BuNC ligands seen at δ 1.43. The stereochemistry
shown for 11 is consistent with the observation of two
t-BuNC resonances and a single 2-butyne methyl reso-
nance at low temperature and with the observation of
a single t-BuNC resonance at room temperature.
(12) Kolobova, N. E.; Lobanova, I. A.; Zdanovich, V. I.; Petrovskii,
P. V. Izv. Akad. Nauk SSSR, Ser. Khim. 1981, 5, 935.
(13) Bang, H.; Lynch, T. J .; Basolo, F. Organometallics 1992, 11,
40.
(14) Addition of excess CNt-Bu to 3 in an effort to obtain complete
conversion led to formation of uncharacterized decomposition products.

1190 Organometallics, Vol. 22, No. 6, 2003
Casey et al.
Sch em e 8
alkene complex 4 did not react under 4 atm CO up to
80 °C, the indenyl alkene complex 2 reacted under 4
atm CO at 25 °C (t1/2 ) 30 h). This corresponds to a
difference in activation barriers of more than 7.5 kcal
mol-1. While the Cp alkyne complex 5 barely reacted
under 4 atm CO at 25 °C (only 3% reaction after 24
days), the indenyl alkyne complex 3 reacted quickly
under 4 atm CO at 25 °C (t1/2 ) 66 min). This corre-
sponds to a difference in activation barriers of more than
5 kcal mol-1. These large rate increases are a signature
of an indenyl effect occurring in the CO substitution
reactions of 2 and 3 and support formation of a ring-
slipped intermediate in these associative substitutions.
In d en yl a n d Alk yn e Liga n d Coop er a tivity. The
combination of the indenyl ligand along with the alkyne
ligand of 3 led to uniquely fast rates of CO substitution.
Both indenyl and alkyne ligands are variable electron
donors: the indenyl ligand can slip from an η5 6e-donor
to an η3 4e-donor,4,16 while the alkyne ligand can switch
from a 2e- to a 4e-donor ligand.17 Ring slippage of Cp
and indenyl ligands allows associative addition of a new
ligand while maintaining an 18e count at the metal.
Indenyl ligands slip more readily than Cp ligands
because of their inherently weaker binding as η5 ligands
and because of their greater stability as η3 ligands due
to the increased aromaticity of the six-membered ring
in the slipped indenyl intermediate.3-5
Acceleration of ligand dissociation by alkyne ligands
has been attributed to the alkyne ligand’s ability to
switch to a 4e-donor mode and stabilize a developing
“vacant” coordination site. Wulff suggested that coupling
of a carbene ligand with CO to give a vinyl ketene might
be promoted by a coordinated alkyne changing from a
2e- to a 4e-donor.18 Takats suggested that the alkyne
osmium complex Os(CO)4(η2-MeCtCMe) loses CO prior
to reacting with a second alkyne and the intermediate
formed from loss of CO is stabilized by the alkyne ligand
changing from a 2e- to a 4e-donor.19 Takats and J ordan
found that Os(CO)4(η2-CF3CtCCF3) reacts with phos-
phines by a dissociative mechanism at a rate 106 times
faster than Os(CO)5.11
F or m a tion of (η1-C9H7)Re(CO)2(P Me3)2(η2-MeCt
CMe) (14). Addition of PMe3 to alkyne complex 3
occurred at -40 °C to give (η1-C9H7)Re(CO)2(PMe3)2(η2-
MeCtCMe) (14), which was isolated by low-tempera-
ture evaporation of solvent (Scheme 8). 14 was unstable
at room temperature but was shown to be formed in 90%
yield by low-temperature NMR of the redissolved solid.
The 13C{1H} NMR spectrum had two CO resonances,
each with one large coupling to a trans phosphine and
one small coupling to a cis phosphine that established
the stereochemistry of 14. Upon warming to room
temperature, 14 lost 2-butyne and formed a 1:2 mixture
of (η5-C9H7)Re(CO)2(PMe3) (15) and (η1-C9H7)Re(CO)2-
(PMe3)3 (16).15
Pure 16 was obtained from reaction of 3 with excess
PMe3 at room temperature. The η1-indenyl ligand of 16
shows separate resonances for H1 and H3 at low
temperature, but a fluxional process causes coalescence
of these peaks at 0 °C (∆Gq ) 11.8 kcal mol-1). In the
low-temperature 31P NMR spectrum of 16, resonances
are seen for nonequivalent diastereotopic trans phos-
phines at δ -17.0 (dd, J ) 156, 26 Hz) and -20.5 (dd,
J ) 156, 26 Hz) and for a third phosphine at δ -28.2 (t,
J ) 26 Hz). The 13C NMR resonance at δ 200.9 (dt, J P-C
) 60, 8 Hz, CO) establishes that one CO is trans to a
phosphine ligand, while the resonance at δ 195.7 (q,
J P-C ) 8 Hz CO) is assigned to CO trans to the
η1-indenyl ligand and cis to three phosphines.
The CO substitution of indenyl alkyne complex 3
takes advantage of both the indenyl and alkyne ligand
rate-enhancing effects to give even faster rates than
those attributed to having just one of the ligands. This
ligand cooperativity might be due to simultaneous
indenyl ring slippage and a switch of the alkyne ligand
from a 2e- to a 4e-donor; in this extreme, an intermedi-
ate A would be reversibly formed and trapped by CO
at a rate that never approaches the saturation limit
(Scheme 5). An alternative explanation of the cooper-
ativity involves concerted CO attack and ring slippage,
with the electron-deficient transition state being stabi-
lized by the additional electron donation from the alkyne
ligand. These alternatives are not kinetically distin-
guishable.
Discu ssion
We will focus on three significant aspects of this
work: (1) the unique cooperativity between alkyne and
indenyl ligands in promoting substitution reactions, (2)
insight into the details of alkyne substitution gained
from kinetic and 13CO labeling studies, and (3) CO-
catalyzed phosphine substitution.
In d en yl Effect on CO Su bstitu tion of Alk yn e a n d
Alk en e Com p lexes. The rate of the CO substitution
reactions of indenyl alkene and alkyne complexes
displayed a linear dependence on CO pressure up to 10
atm that provided evidence for an associative mecha-
nism. The rates of the indenyl substitutions were much
faster than those of their Cp analogues. While the Cp
Kin etic Evid en ce for th e Br oa d Ou tlin es of th e
Mech a n ism of CO Su bstitu tion of Alk yn e Com p lex
3. The rate of conversion of alkyne complex 3 to CO
(16) J i, L.-N.; Rerek, M. E.; Basolo, F. Organometallics 1984, 3, 740.
(17) Templeton, J . L. Adv. Organomet. Chem. 1989, 29, 1.
(18) Bos, M. E.; Wulff, W. D.; Miller, R. A.; Chamberlin, S.;
Brandvold, T. A. J . Am. Chem. Soc. 1991, 113, 9293.
(19) Washington, J .; McDonald, R.; Takats, J .; Menashe, N.; Reshef,
D.; Shvo, Y. Organometallics 1995, 14, 3996.
(15) The ratio of (η5-C9H7)Re(CO)2(PMe3) (15) and (η1-C9H7)Re(CO)2-
(PMe3)3 (16) was not 1:1 as expected from stoichiometry because there
was still approximately 0.8 equiv of free PMe3 left in solution after
removing the solvent from (η1-C9H7)Re(CO)2(PMe3)2(η2-MeCtCMe) (14)
and redissolving in CD2Cl2.

Indenyl Rhenium Alkyne Complexes
Organometallics, Vol. 22, No. 6, 2003 1191
substitution product 8 depends linearly on CO pressure.
At low CO pressure (1-10 atm), rate ) k1obs[3][CO]; k1obs
) 4.45 × 10-5 atm-1 s-1 at 25 °C. At higher CO pressure
the rate of disappearance of starting material 3 contin-
ued to increase, but the formation of significant quanti-
ties of η1-indenyl complex 12 complicated analysis of the
kinetics at higher CO pressure.
The η1-indenyl complex 12 was prepared in 90% yield
under high CO pressure at low temperature. In the
absence of added CO pressure, 12 reacted to generate
a 12:1 ratio of starting alkyne complex 3:CO substitution
product 8. This partitioning requires that k1obs
)
k1[k3/(k3 + k-1)] ) k1[1/13] and gives the rate constant
k1 ) 5.8 × 10-4 atm-1 s-1 at 25 °C for the conversion of
3 to η3-indenyl intermediate B (Scheme 9).
F igu r e 5. Plot of 1/kobs for conversion of 12 to (η5-C9H7)-
Re(CO)3 (8) vs pressure of CO above 68 atm.
As detailed earlier, we independently measured the
rate of CO loss from 12 in the absence of added CO
between -24 and 2 °C. Extrapolation of the Eyring plot
to 25 °C gave an estimate of the first-order rate constant
for loss of CO from 12 of k-2 ) 7.7 × 10-3 s-1 at 25 °C.
The rate of CO loss from η1-indenyl alkyne complex 12
was approximately 106 times faster than from the all-
carbonyl η1-indenyl complex (13). This remarkable rate
acceleration is attributed to the ability of the alkyne
ligand to facilitate CO loss by switching from 2e- to
4e-donor in the transition state for CO loss.11,17,19
In principle, η1-indenyl complex 12 could lose alkyne
first to form intermediate C, which could then lose CO
to form 8. Apparently this is not a significant pathway
since ∼92% of 12 loses 2 CO’s to re-form alkyne complex
3. Detailed kinetic analysis also demonstrates that
alkyne loss occurs from η3-indenyl complex B and not
from η1-indenyl complex 12. The rate law shown in eq
1 describes the reactions in Scheme 9. If 2-butyne were
lost only from 12 (k4 . k3 ≈ 0), eq 1 simplifies to eq 2
and the rate of appearance of 8 would depend on the
square of the CO pressure at low CO pressure. However,
a linear CO dependence was observed at pressures
below 10 atm of CO. Therefore, loss of alkyne from 12
is not an important process at low CO pressure.
that loses butyne (k4[12] . k3[B]), the rate of appear-
ance of 8 should be independent of CO pressure (eq 4).
If intermediate B is the only complex that loses butyne,
the rate expression for the appearance of 8 simplifies
to eq 5, which has an inverse dependence on CO
pressure. Experimentally, the rate of appearance of 8
was inversely dependent on CO pressure at pressures
> 50 atm of CO where the major species in solution was
12. Therefore, we conclude that butyne loss occurs from
B and not from 12, consistent with the low-pressure
limit of losing butyne from B.
The intercept of the plot of 1/kobs versus CO pressure
for conversion of 12 to 8 at pressures above 50 atm gives
k-2 ) 8.8 × 10-3 s-1 at 25 °C, which is the rate constant
for CO dissociation from 12 (Figure 5 and eq 6). As
detailed earlier, we independently measured the rate
of CO loss from 12 in the absence of added CO between
-24 and 2 °C. Extrapolation of the Eyring plot to 25 °C
gave an estimate of k-2 as 7.7 × 10-3 s-1, in close
agreement with the measurement from Figure 5. The
slope of the plot in Figure 5 gives k2/k-2k3 ) 13.6 atm-1
s-1. This allows calculation of k2/k3 ) 0.12 atm-1 with
k-2 ) 8.8 × 10-3 s-1
.
The observation that (η1-indenyl)Re)(CO)5 (13) was
not detected by IR during the conversion of 12 to 8
under high CO pressure is also consistent with the
failure of 12 to lose butyne. If 12 lost butyne to generate
intermediate C, trapping of C by CO would have given
13, which is kinetically stable at 25 °C.
k1k3[CO][3]
k-1 + k3 + k2[CO] k-1 + k3 + k2[CO]
k4[12] (1)
k-2k3[12]
d[8]
dt
)
+
+
if k3 ) 0 and [3] . [12]
d[8]
Refin em en t of th e CO Su bstitu tion Mech a n ism
Usin g 13CO La belin g Da ta . The observation of exten-
sive 13CO incorporation into 3 at ∼50% conversion of 3
to 8 requires reversibility of the CO addition to 3 to give
an η3-indenyl intermediate B (Table 4). Loss of CO from
B was independently established by the observation
that the major reaction of B (generated by CO loss from
η1-indenyl complex 12) was loss of a second CO to
generate 3.
k1k2k4[CO]2[3]
(k-1 + k2[CO])(k-2 + k4)
)
(2)
(3)
dt
k1k3[CO][3]
k-1 + k3
d[8]
dt
if k4 ) 0 and [3] . [12]
)
d[8]
dt
if k3 ) 0 and [12] . [3]
if k4 ) 0 and [12] . [3]
) k4[12] (4)
The quantitative extent of 13CO incorporation into 3
provides insight into the stereochemistry of B. Since we
know that B loses CO to form 3 about 12 times more
rapidly than it loses alkyne to form 8, if 13CO entered
B randomly or if label scrambled in B, then at 50%
conversion to 8, the rhenium complexes would have had
to cycle through B-(13CO)1 six times and two-thirds of
it would have lost CO and led to 3-(13CO)1. This would
have resulted in incorporation of multiple 13CO ligands
into 3, but 3 recovered from reaction with 1 atm 13CO
k-2k3[12]
d[8]
dt
)
(5)
k3 + k2[CO]
k2[CO]
1
kobs
1
k-2
if k4 ) 0 and [12] . [3]
)
+
(6)
k-2k3
We will consider two pathways for 2-butyne loss
under high CO pressure where η1-indenyl complex 12
is the major species. If complex 12 is the major species

1192 Organometallics, Vol. 22, No. 6, 2003
Casey et al.
Sch em e 9
give rise to 13CO exchange and one of the two isomers
with symmetrically related 13CO and CO ligands as the
minor species which is solely responsible for exchange.
In this case, 13CO enters and leaves only one site to
generate each of the two isomers, and the two isomers
are not required to have the same relationship between
the alkyne and η3-indenyl ligands (for example, major
species B-(13CO)1-3 and minor species B-(13CO)1-2).
Sch em e 10
Additional information about the likely stereochem-
istry of B comes from the conversion of B to the
η1-indenyl complex 12, in which the indenyl and alkyne
ligands are trans. The only isomers of B that can lead
to 12 by CO replacement of the η3-indenyl with retention
of stereochemistry are the three isomers B-(13CO)1-5,
-6, and -7. The two incoming 13CO ligands in 12-(13CO)2
are likely to be cis to one another, since this is the
stereochemistry seen in the η1-indenyl PMe3 complex
14 and the η1-indenyl isonitrile complex 11. It is
interesting to note that the stereochemistry of these η1-
indenyl complexes results from CO substitution with
retention of stereochemistry as the indenyl ring slips.
So far, we have discussed only routes to labeled 3 via
η3-indenyl intermediate B, but routes involving η1-
indenyl complex 12 also need to be considered. We will
first consider whether routes through 12 are the only
pathway to labeled 3. If 12 with trans 13CO ligands were
formed selectively, microscopic reversibility would re-
quire loss of either two trans CO or two trans 13CO
ligands to give only 3-(13CO)2 and 3. This possibility is
clearly excluded since much more monolabeled 3-(13CO)1
was observed than doubly labeled 3-(13CO)2. If 12 were
formed with cis 13CO ligands, microscopic reversibilty
would require loss of cis CO ligands and would have
given a 2:1 ratio of 3-(13CO)1: 3-(13CO)2. Under 1 atm
13CO, the observed 4.5:1 ratio of 3-(13CO)1: 3-(13CO)2
at 50% conversion to 8 is too low for 13CO exchange into
3 to involve only routes through 12.
contained only 18% 3-(13CO)1 and 4% 3-(13CO)2. Clearly,
13CO does not enter B randomly. There are three
different isomers of unlabeled B with different relation-
ships between the alkyne and η3-indenyl ligand and
seven total isomers of B-(13CO)1 shown in Scheme 10.
If one of the two isomers with symmetry-related CO and
13CO ligands (B-(13CO)1-1 or B-(13CO)1-2) were the only
isomer formed, microscopic reversibility would lead to
13CO exchange into 3 one out of two times that B is
generated; this would have resulted in incorporation of
multiple 13CO ligands into 3, which was not seen.
Therefore, B-(13CO)1-1 and B-(13CO)1-2 can be excluded
as the major isomer of B. If one of the other five isomers
(B-(13CO)1-3, -4, -5, -6, or -7) were the only isomer
formed, microscopic reversibility would lead to no
exchange.
We next considered whether the amount of 3-(13CO)2
can be explained solely via sequential exchange passing
through monolabeled 3-(13CO)1. Under 1 atm 13CO, at
46% conversion of 3 to 8, 9.7% 3-(13CO)1 had formed;
sequential labeling predicts that only 0.5% 3-(13CO)2
should have been seen instead of the 2.1% observed. At
higher 13CO pressures, there are larger discrepancies:
under 12 atm, at 29% conversion of 3 to 8, 15.6%
3-(13CO)1 and 9.9% 3-(13CO)2 had formed, but sequen-
tial labeling predicts formation of only 0.8% 3-(13CO)2.
Our data can be accommodated in two different ways.
One possibility is that one of these five isomers is the
major species generated by 13CO addition to 3 and that
it has a second minor mode of dissociating CO (and by
microscopic reversibility, a second minor mode of adding
CO). In this scenario, the two isomers have the same
relationship between the alkyne and η3-indenyl ligands
(for example, major species B-(13CO)1-3 and minor
species B-(13CO)1-1). A second possibility involves one
of the five isomers as the major species which does not

Indenyl Rhenium Alkyne Complexes
Organometallics, Vol. 22, No. 6, 2003 1193
Sch em e 11
Clearly, (1) sequential reactions cannot alone account
for the amount of double label seen and (2) a second
direct route to double labeling becomes more important
as 13CO pressure is increased. This suggests that there
are two routes to 13CO exchange, one involving at least
two different isomeric η3-indenyl intermediates B and
another involving cis 12-(13CO)2, which becomes in-
creasingly important at higher pressure.
compound 3, but the simulation systematically overes-
timated the amount of double labeled 8-(13CO)2. For 1
atm, the labeling patterns (observed, simulated) were
3-(13CO)0 (77%, 75%), 3-(13CO)1 (18%, 20%), 3-(13CO)2
(4%, 5%); and 8-(13CO)1 (91%, 80%), 8-(13CO)2 (8%,
19%), 8-(13CO)3 (1%, 2%). For 4 atm, the labeling
patterns (observed, simulated) were 3-(13CO)0 (64%,
69%), 3-(13CO)1 (22%, 19%), 3-(13CO)2 (14%, 13%); and
8-(13CO)1 (82%, 72%), 8-(13CO)2 (12%, 20%), 8-(13CO)3
(6%, 6%). For 12 atm, the labeling patterns (observed,
simulated) were 3-(13CO)0 (64%, 69%), 3-(13CO)1 (22%,
19%), 3-(13CO)2 (14%, 13%); and 8-(13CO)1 (82%, 79%),
8-(13CO)2 (12%, 17%), 8-(13CO)3 (6%, 1%).
CO-Ca ta lyzed P h osp h in e Su bstitu tion . The phos-
phine substitution of the indenyl alkyne complex 3 to
form phosphine complex 9 was much faster in the
presence of 13CO, indicating that CO catalyzed the
phosphine substitution of 3.20 The rate of reaction of 3
with PPh3 in the absence of added CO depended on
[PPh3], which is consistent with an associative mecha-
nism involving indenyl ring slippage. When 13CO was
added to the system, the rate of formation of phosphine
complex 9 occurred several hundred times faster. The
disappearance of 3 was about 3 times faster (t1/2 ) 22
min vs t1/2 ) 66 min) in the presence of 13CO and PPh3
than in the presence of 13CO alone.
A plot of the ratio of the sum of 3-(13CO)1
+
3-(13CO)2:8 versus CO pressure provides semiquanti-
tative evidence for two pathways for 13CO incorporation
into 3 (Figure 4). The nonzero intercept is due to
pressure-independent 13CO incorporation into 3 via
isomers of B, while the positive slope indicates a
pressure-dependent pathway via η1-indenyl intermedi-
ate 12. The intercept indicates the rate of CO loss from
intermediate B is about 0.25 as fast as alkyne loss.
Assuming cis 13CO in 12, 13CO incorporation into 3 will
occur 75% of the time that 12 is formed and the slope
of Figure 4 is 0.75 × k2/k3 ) 0.055 atm-1. This allows
calculation of k2/k3 ) 0.075 atm-1. This ratio is similar
to the value of 0.12 atm-1 calculated from the graph of
1/kobs versus CO pressure for the observed rate of
formation of 8 from 12 at CO pressures above 50 atm
(Figure 5).
A mechanism that explains all of the facets described
above and that gives a reasonable quantitative agree-
ment with our data is shown in Scheme 11 with the
indicated relative rate constants. (There are, of course,
other combinations of isomers of B-(13CO)1 that would
also explain the data.) In this mechanism, 13CO addition
to 3 produces mainly intermediate B-(13CO)1-5 by
partial displacement of the indenyl ligand with retention
of stereochemistry and a smaller amount of B-(13CO)1-
6, which can lose CO to form 3-(13CO)1. Addition of a
second 13CO to B-(13CO)1-5 leads to cis 12-(13CO)2 and
can account for excess double label 3-(13CO)2 at early
reaction time and at higher pressure.
These results are explained in terms of the mecha-
nism shown in Scheme 12. In the absence of added CO,
PPh3 slowly attacks 3 to form the phosphine-substituted
η3-indenyl intermediate E, which can lose alkyne to
form the phosphine substitution product 9. Interest-
ingly, neither an η5-indenyl phosphine alkyne complex
nor an η5-indenyl diphosphine complex was observed,
indicating that η3-indenyl intermediate E loses alkyne
or phosphine faster than CO.
With CO alone, a more rapid attack of CO on 3
produces η3-indenyl intermediate B, which we have seen
reverts to 3 about 12 times faster than it loses alkyne
This reaction scheme and relative rate constants were
used to simulate the time course of the formation of
labeled 3 and 8 under 1, 4, and 12 atm 13CO. Reasonable
agreement was obtained for the labeling pattern in
(20) In contrast, we previously observed PMe3 catalysis of CO
insertion into the methyl rhenium complex C5H5Re(NO)(PMe3)(CH3)
to give C5H5Re(NO)(PMe3)(COCH3). Casey, C. P.; Widenhoefer, R. A.;
O’Connor, J . M. J . Organomet. Chem. 1992, 428, 99.

1194 Organometallics, Vol. 22, No. 6, 2003
Casey et al.
Sch em e 12
carbonyl band of 3 at 1869 cm-1. A plot of the ln[3] versus
time was linear (kobs ) 1.1 × 10-4 s-1, t1/2 ) 103 min).
to form 8. An explanation for the rate enhancement
when both CO and PPh3 are present is that the rapidly
formed intermediate B is intercepted by PPh3 to give
the intermediate η1-C9H7(CO)2(13CO)(PPh3)Re(η2-MeCt
CMe) (D), which then loses CO and the alkyne ligand
to form 9. This scenario explains the observations of
13CO incorporation in phosphine complex 9 and of
enhanced rates when both CO and PPh3 are present.
(b) A C6D6 (0.32 mL) solution of 3 (7.3 mg, 18 µmol) was
placed under 2.9 atm 13CO (205 µmol, 20 mM) in a 1.98 mL
resealable NMR tube. The disappearance of the alkyne methyl
resonance of 3 at δ 2.05 was followed by 1H NMR spectroscopy
at 25 °C. A plot of ln[3] versus time was linear (kobs ) 1.3 ×
10-4 s-1, t1/2 ) 90 min).
(c) A C6H6 (10 mL) solution of 3 (10.2 mg, 25 µmol) was
placed in a 30 mL stainless steel bomb containing a star
stirring bar. The bomb was placed in a 25 °C water bath and
pressurized with 18 atm CO (260 mM). After 10 min rapid
stirring, CO and solvent were removed by rapid evaporation.
The remaining solid was dissolved in C6D6, and 1H NMR
Exp er im en ta l Section
spectroscopy showed a 2.0:1.1 ratio of 3:8, indicating kobs
)
7.4 × 10-4 s-1, t1/2 ) 16 min.
(d) A C6H6 (10 mL) solution of 3 (20 mg, 49 µmol) was placed
in a 25 mL stainless steel Parr pressure bomb equipped with
a high-pressure SiComp probe attached to an ASI ReactIR
1000 spectrometer. At 25 °C, the bomb was pressurized with
68 atm CO. The appearance of peaks at 1927 and 2024 cm-1
was monitored to determine the rate of appearance of 8. A plot
of ln[([8]inf - [8])/[8]inf] versus time was linear (kobs ) 9.5 ×
10-4 s-1, t1/2 ) 12 min). In the case of the reaction of 3 with 10
atm of CO, only 3 and 8 were observed, and both the rate of
appearance of 8 (kobs ) 4.16 × 10-4 s-1, t1/2 ) 28 min) and the
rate of disappearance of 3 (kobs ) 4.13 × 10-4 s-1, t1/2 ) 28
min) were measured.
In d en yl Nu m ber in g Used in NMR Assign m en ts. (η5-
C9H 7)(CO)2R e(η2-MeCtCMe) (3). Addition of excess 2-
butyne (6.5 mmol) to a THF solution of 7 (250 mg, 0.58 mmol)
resulted in the formation of 3 after an hour at room temper-
ature. Flash chromatography (silica gel, 8:1 pentane-CH2Cl2)
gave 3 (84 mg, 0.20 mmol, 35% yield) as an orange crystalline
1
solid. H NMR (500 MHz, C6D6): δ 6.76 (AA′BB′, H5,6), 6.60
(AA′BB′, H4,7), 5.36 (t, J ) 3 Hz, H2), 5.23 (d, J ) 3 Hz, H1,3),
2.05 (s, Me). 13C{1H} NMR (126 MHz, C6D6): δ 204.1 (CO),
125.8 (C5,6), 123.2 (C4,7), 111.5 (C8,9), 90.5(C2), 73.7 (C1,3),
67.6 (Ct), 10.9 (Me). IR (THF): 1943 (s), 1869 (vs) cm-1
.
At pressures between 10 and 50 atm of CO, mixtures of 3,
8, and 12 were seen by IR and the rate of appearance of 8 was
calculated (Table 1). The IR spectra were deconvoluted into a
sum of the spectra of 3, 8, and 12. The molar absorptivities of
the three species were used to obtain their concentrations as
a function of time (Figure 3S in Supporting Information).
Deter m in a tion of P er cen t 13CO in (η5-C9H7)(CO)2Re-
(η2-MeCtCMe) (3) a n d (η5-C9H7)Re(CO)3 (8). The isotopic
composition of 8 obtained from reaction of 3 with 13CO was
analyzed by electron impact mass spectrometry. The cluster
of peaks in the range m/e 384-389 due to the isotopomers of
the molecular ion of 8 (37% 185Re, 63% 187Re, and taking into
HRMS (EI) calcd (found) for
(412.0467). Anal. Calcd for C15H13O2Re: C, 43.79; H, 3.18.
Found: C, 43.57; H, 2.95.
C15H13O2Re: m/z 412.0474
Rea ction of (η5-C9H7)(CO)2Re(η2-MeCtCMe) (3) w ith
CO. (a) A C6H6 (10 mL) solution of 3 (39 mg, 95 µmol) was
placed in a 6 oz. pressure reaction vessel. A star stirring bar
and high stirring rates were employed to ensure equilibration
of gaseous and dissolved CO. The system was pressurized with
2.1 atm of CO (15 mM). The pressure vessel was placed in a
25 °C constant-temperature bath, and 0.3 mL aliquots were
periodically removed and analyzed by IR spectroscopy. The
concentration of 3 was determined by integration of the

Indenyl Rhenium Alkyne Complexes
Organometallics, Vol. 22, No. 6, 2003 1195
account the 10% 13C18O concentration present in labeled
13CO) was used to determine the 13C labeling pattern of 8. The
observed ratio of peaks in the 384-389 range was fit as the
weighted sum of the patterns calculated for 8-(13CO)1:8-
(C2), 140.7 (C8 or C9), 123.7, 123.4, 123.1, 121.8, 118.7, 55.4
(MeCt), 19.4 (C1), 11.1 (MeCt).
Ra te of Con ver sion of (η1-C9H7)Re(CO)4(η2-MeCtCMe)
(12) to 8 a n d 3. A C6D5CD3 solution (0.5 mL) of 12 (8.6 mg,
0.021 mmol) was synthesized by method b described above.
The NMR tube was sealed under vacuum and placed in an
NMR probe cooled to -9 °C. With a headspace of approxi-
mately 1 mL, the ending CO pressure was calculated to be
between 0.9 and 1.0 atm of CO after all the CO is released
from 12. Disappearance of 12 was monitored by 1H NMR
spectroscopy by integrating the proton resonance at δ 4.5 for
12 compared to the internal standard 1,4-dioxane. The rate
of disappearance of 12 was first order in 12 to over 3 half-
lives (kobs ) 1.3 × 10-4 s-1, t1/2 ) 87 min). The rate of
disappearance of 12 was measured between -24 and 2 °C, but
the reaction was monitored for only 1 half-life at 4 and -2 °C
since there was significant deviation from first-order decay at
longer times due to inefficient removal of CO. A 1:12 mixture
of 8:3 was formed during the first half-life as determined by
integrating the proton resonances at δ 6.58 for 8 and at δ 6.60
for 3. The ratio of 1:12 of 8:3 is the average over all temper-
atures.
(
13CO)2:8-(13CO)3 by minimizing the sum of the squares of the
difference between calculated and observed peak intensities
(Table 3, 5). Similarly the isotopic composition of 3 recovered
from partial reaction of 3 with 13CO was analyzed by electron
impact mass spectrometry. The cluster of peaks in the range
m/e 410-415 due to the isotopomers of the molecular ion of 3
(37% 185Re, 63% 187Re, and taking into account the 10% 13C18
O
concentration present in labeled 13CO) was used to determine
the 13C labeling pattern of 3. The observed ratio of peaks in
the 410-415 range was fit as the weighted sum of the patterns
calculated for 3-(13CO)0:3-(13CO)1:3-(13CO)2 by minimizing the
sum of the squares of the difference between calculated and
observed peak intensities (Table 4).
(η1-C9H7)Re(CO)4(η2-MeCtCMe) (12). (a) At 77 K, a
measured amount of CO was condensed into a thick walled
NMR tube containing 3 (8 mg, 19 µmol) in C6D5CD3 (0.35 mL)
in order to obtain a pressure of 10 atm of CO at -15 °C. After
6 h at -15 °C, 1H NMR spectroscopy at -50 °C showed a 6:87:6
mixture of 8 (δ 1.62 for free butyne):3 (δ 2.10):12 (δ 1.73). After
30 h at -15 °C, the ratio of 8:3:12 was 30:20:50.
Ack n ow led gm en t. Financial support from the De-
partment of Energy, Office of Basic Energy Sciences, is
gratefully acknowledged.
(b) A 4:4:92 ratio of 8:3:12 was obtained as follows. A C6D5-
CD3 (1.5 mL) solution of 3 (29.5 mg, 0.072 mmol) was placed
in a 30 mL stainless steel pressure vessel with a stirring bar.
At -10 °C, 100 atm of CO was added and the solution was
stirred for 1 h at -10 °C. At -78 °C, CO was released from
the pressure vessel and the toluene solution was transferred
to a NMR tube cooled to -78 °C. Four freeze-pump-thaw
cycles were performed to remove dissolved CO before measur-
ing the rate of conversion of 12 to 8 and 3. IR (C6H6): 1998
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures for synthesis of 7, 2, 13, 14, and 16; reactions of 2
with CO, 3 with PPh3, 13CO, CNt-Bu; plot of CO pressure
dependence of the rate of formation of 8 from 3; Eyring plot
for disappearance of 12; simulations of 13CO incorporation into
3 and 8 at 1, 4, and 12 atm 13CO; and X-ray crystal data for 2
and 3. This material is available free of charge via the Internet
at http://pubs.acs.org.
1
(s), 1946 (m) cm-1. H NMR (500 MHz, -65 °C, C6D5CD3): δ
7.74 (d, J ) 7 Hz, H7), 7.68 (d, J ) 7 Hz, H4), 7.35 (t, J ) 7
Hz, H6), 7.28 (t, J ) 7 Hz, H5), 7.20 (d, J ) 5 Hz, H2), 6.73 (d,
J ) 5 Hz, H3), 4.5 (s, H1), 1.73 (s, MeCt). 13C{1H} NMR (125
MHz, -65 °C, C6D5CD3): δ 190.5 (CO), 160.6 (C8 or C9), 152.2
OM020884Q