Mechanism of the formation of carbyne complexes of rhenium upon protonation of vinylidene precursors

Article abstract of DOI:10.1021/om970398g

The reactions of trans-[ReX(=C=CHR)(dppe)₂] with [NHEt₃][BPh₄] to form the carbyne complexes trans-[ReX(sCCH₂R)(dppe)₂]⁺ (X = Cl; R = Ph, C₆H₄Me-4, Bu^t, CO₂Me, CO₂Et; X = F; R = CO₂Et; dppe = Ph₂PCH₂CH₂PPh₂) have been studied by stopped-flow spectrophotometry and shown to proceed via three pathways whose relative contribution depends on the nature of R and X. The most direct pathway involves regiospecific protonation at the β-carbon of the vinylidene. However, under some conditions initial protonation at the metal to form [Re(H)X(=C=CHR)(dppe)₂]⁺ is more rapid, and this hydride subsquently rearranges to form the carbyne by an intramolecular pathway or by protonation of [Re(H)X-(=C=CHR)(dppe)₂]⁺ at the β-carbon of the vinylidene ligand to give [Re(H)X(S=CCH₂R)-(dppe)₂]²⁺, which then undergoes deprotonation to form [ReX(=CCH₂R)(dppe)₂]⁺. For the R = C₆H₅ or C₆H₄Me-4 complexes, kinetic analysis indicates that all three pathways occur, whereas for the bulky R = Bu^t analogue, the pathways that involve direct addition to the vinylidene ligand do not operate. For [ReX(=C=CHCO₂R)(dppe)₂] (R = Et, Me) the strong electron-withdrawing effect of the ester group results in slow proton transfer from [NHEt₃]⁺ to the vinylidene ligand in [Re(H)Cl(=C=CHCO₂R)(dppe)₂]⁺. The formation of an adduct is evident from the kinetic studies with these complexes and is proposed to be the species in which [NHEt₃]⁺ is hydrogen-bonded to the β-carbon of the vinylidene ligand. Rate-limiting proton transfer within this adduct completes the reaction and is associated with a large primary isotope effect. The way in which the trans-halide influences this reactivity has also been investigated and fluoride shown to highly promote the rate of protonation.

Full text of DOI:10.1021/om970398g

Organometallics 1997, 16, 5441-5448
5441
Mech a n ism of th e F or m a tion of Ca r byn e Com p lexes of
Rh en iu m u p on P r oton a tion of Vin ylid en e P r ecu r sor s
Maria Fernanda N. N. Carvalho, S´ılvia S. P. R. Almeida, and
Armando J . L. Pombeiro*
Centro de Quı´mica Estrutural, Complexo I, Instituto Superior Te´cnico, Av. Rovisco Pais,
1096 Lisboa codex, Portugal
Richard A. Henderson
Nitrogen Fixation Laboratory, J ohn-Innes Centre, Norwich Research Park, Colney Lane,
Norwich NR4 7UH, U.K.
Received May 13, 1997X
The reactions of trans-[ReX(dCdCHR)(dppe)2] with [NHEt3][BPh4] to form the carbyne
complexes trans-[ReX(tCCH2R)(dppe)2]+ (X ) Cl; R ) Ph, C6H4Me-4, But, CO2Me, CO2Et;
X ) F; R ) CO2Et; dppe ) Ph2PCH2CH2PPh2) have been studied by stopped-flow
spectrophotometry and shown to proceed via three pathways whose relative contribution
depends on the nature of R and X. The most direct pathway involves regiospecific protonation
at the â-carbon of the vinylidene. However, under some conditions initial protonation at
the metal to form [Re(H)X(dCdCHR)(dppe)2]+ is more rapid, and this hydride subsquently
rearranges to form the carbyne by an intramolecular pathway or by protonation of [Re(H)X-
(dCdCHR)(dppe)2]+ at the â-carbon of the vinylidene ligand to give [Re(H)X(tCCH2R)-
(dppe)2]2+, which then undergoes deprotonation to form [ReX(tCCH2R)(dppe)2]+. For the R
) C6H5 or C6H4Me-4 complexes, kinetic analysis indicates that all three pathways occur,
whereas for the bulky R ) But analogue, the pathways that involve direct addition to the
vinylidene ligand do not operate. For [ReX(dCdCHCO2R)(dppe)2] (R ) Et, Me) the strong
electron-withdrawing effect of the ester group results in slow proton transfer from [NHEt3]+
to the vinylidene ligand in [Re(H)Cl(dCdCHCO2R)(dppe)2]+. The formation of an adduct is
evident from the kinetic studies with these complexes and is proposed to be the species in
which [NHEt3]+ is hydrogen-bonded to the â-carbon of the vinylidene ligand. Rate-limiting
proton transfer within this adduct completes the reaction and is associated with a large
primary isotope effect. The way in which the trans-halide influences this reactivity has
also been investigated and fluoride shown to highly promote the rate of protonation.
In tr od u ction
photometry, can define the sites of protonation and the
factors that determine the rate of formation or cleavage
of metal-hydrogen or carbon-hydrogen bonds.
Protonation of small unsaturated molecules, activated
by coordination to a transition metal center, is of current
interest in coordination chemistry, particularly applied
to catalysis1 or to mimic the elementary reactions of
enzymes such as nitrogenases.2 The investigation of the
mechanisms of such reactions with, for example, ni-
triles,3 allenes,4 isocyanides,5 ethylene,6a or alkynyl,6b
using rapid techniques such as stopped-flow spectro-
A particulary simple reaction (at least stoichiomet-
rically) is the protonation of vinylidene ligands to give
the corresponding carbyne (alkylidyne) species,7 in
which an apparently regiospecific proton addition to the
remote â-carbon atom of the vinylidene has occurred,
at a variety of transition metal centers. We have
prepared8 a series of rhenium carbyne complexes of the
type trans-[ReX(tC-CH2R)(dppe)2]+ (X ) Cl, F; R )
alkyl, aryl, ester), upon protonation of the corresponding
vinylidene complexes trans-[ReX(dCdCHR)(dppe)2] (eq
* Corresponding author: pombeiro@alfa.ist.utl.pt.
X Abstract published in Advance ACS Abstracts, November 1, 1997.
(1) See, e.g.: Crabtree, R. H. The Organometallic Chemistry of the
Transition Metals; J ohn Wiley & Sons: New York, 1988; Masters, C.
Homogeneous Transition-Metal Catalysis; Chapman and Hall: London,
1981.
(2) Evans D. J .; Henderson R. A.; Smith B. E. In Bioinorganic
Catalysis; Reedijk J ., Ed.; Marcel Dekker: New York, 1993; Chapter
5, p 89. Leigh, G. J . Acc. Chem. Res. 1992, 25, 177. Pombeiro, A. J . L.
Inorg. Chim. Acta 1992, 198-200, 179. Pombeiro, A. J . L.; Richards,
R. L. Coord. Chem. Rev. 1990, 104, 13.
(3) Frau´sto da Silva, J . J . R.; Guedes da Silva, M. F. C.; Henderson,
R. A.; Pombeiro, A. J . L.; Richards, R. L. J . Organomet. Chem. 1993,
461, 141.
(4) Henderson, R. A.; Pombeiro, A. J . L.; Richards, R. L.; Wang, Y.
J . Organomet. Chem. 1993, 447, C11.
(5) Henderson, R. A.; Pombeiro, A. J . L.; Richards, R. L.; Frau´sto
da Silva, J . J . R.; Wang, Y. J . Chem. Soc., Dalton Trans. 1995, 1193.
(6) (a) Oglieve, K. E.; Henderson, R. A. J . Chem. Soc., Dalton Trans.
1991, 3295. (b) Henderson, R. A.; Oglieve, K. E. J . Chem. Soc., Dalton
Trans. 1996, 3397.
(7) Bruce, M. I. Chem. Rev. 1991, 91, 197 and references therein.
Beevor, R. G.; Green, M.; Orpen, A. G.; Williams, I. D. J . Chem. Soc.,
Dalton Trans. 1987, 1319. Nicklas, P. N., Selegue, J . P.; Young, B. A.
Organometallics 1988, 7, 2248. Birdwhistell, K. R.; Tonker, T. L.;
Templeton, J . L. J . Am. Chem. Soc. 1985, 107, 4474. Kostic, N. M.;
Fenske, R. F. Organometallics 1982, 1, 974.
(8) (a) Almeida, S. S. P. R.; Pombeiro A. J . L. Organometallics 1997,
16, 4469. (b) Pombeiro, A. J . L.; Hills, A.; Hughes, D. L.; Richards, R.
L. J . Organomet. Chem. 1988, 352, C5.
S0276-7333(97)00398-1 CCC: $14.00 © 1997 American Chemical Society

5442 Organometallics, Vol. 16, No. 25, 1997
Carvalho et al.
Sch em e 1
F igu r e 1. Kinetic data for the reaction of trans-[ReCl-
(dCdCHC6H4Me-4)(dppe)2] (1.0 × 10-4 mol dm-3) with
[NHEt3][BPh4] in THF, at 25 ( 0.1 °C, monitored at λ )
420 nm. The line drawn is that defined by eq 2 and the
values given in the text. Inset: kobs vs [NEt3] for a constant
concentration of [NHEt3][BPh4] (1.0 × 10-3 mol dm-3).
1), and in a preliminary kinetic study9 of the reaction
trans-[ReX(dCdCHR)(dppe)2] + H+
f
In a previous study9 with trans-[ReCl(dCdCHPh)-
(dppe)2], we have proposed that initial protonation
occurs at rhenium to form [Re(H)Cl(dCdCHPh)(dppe)2]+,
although spectroscopic confirmation for this proposal
has not yet been obtained. The kinetics of the conver-
sion of this intermediate to give [ReCl(tCCH2Ph)-
(dppe)2]+ can occur by an acid-independent pathway.
This can be rationalized only if protonation of the
complex has already occurred, albeit at the “wrong” site.
trans-[ReX(tCCH2R)(dppe)2]+ (1)
of trans-[ReCl(dCdCHPh)(dppe)2] with [NHEt3][BPh4],
the complexity of the rate law is incompatible with
regiospecific protonation but rather demonstrates that
the reaction occurs by three pathways (Scheme 1; X )
Cl, R ) Ph): (i) direct protonation at the remote carbon;
(ii) protonation of the metal followed by intramolecular
migration, or (iii) protonation of the metal followed by
an acid-catalyzed rearrangement pathway.
In this paper, we report further kinetic studies on
complexes of this type which show that with a variety
of R substituents (R ) But, C6H4Me-4, CO2Me, CO2Et)
on the vinylidene ligand these three pathways are
observed and that the nature of R influences which
pathway dominates. In addition, the effect of changing
the trans-halide from X ) Cl to X ) F was also
investigated. Complexes with carbene or carbyne ligands
stabilized by a trans-fluoro ligand are rare10 and these
systems provide an opportunity to compare the effect
of F- and Cl- ligands
The exponential nature of the absorbance-time curves
for all the complexes trans-[ReCl(dCdCHR)(dppe)2]
indicates that the kinetics exhibit a first-order depen-
dence on the concentration of the complex. However,
the dependence on the acid concentration varies with
the nature of the substituent R and X as will be outlined
in the following discussion.
tr a n s-[ReCl(dCdCHC6H4Me-4)(d p p e) ]. The re-
2
action of trans-[ReCl(dCdCHC6H4Me-4)(dppe)2] with
[NHEt3][BPh4] shows a dependence on the concentration
of acid as described by eq 2 and shown in Figure 1 (here
kobs ) k4 + k3[NHEt3]
(2)
Resu lts a n d Discu ssion
and throughout this paper, kobs is the observed pseudo-
first-order rate constant measured under conditions
where [NHEt3+]/[Re] g 10). The rate constants k3 and
k4 correspond to the elementary reactions shown in
Scheme 1.
The reaction of trans-[ReCl(dCdCHR)(dppe)2] with
anhydrous HCl in THF (THF ) tetrahydrofuran) to
form [ReCl(tCCH2R)(dppe)2]+ is complete within the
dead time of the stopped-flow apparatus (2 ms). How-
ever, with the weaker acid [NHEt3][BPh4], the reaction
is sufficiently slow to be studied by this technique.
For all vinylidene complexes studied in this work, the
absorbance-time traces show a similar behavior.
When reaction 1 is studied on a stopped-flow ap-
paratus, it occurs in two distinct stages: an initial
absorbance change complete within the dead time of the
apparatus followed by an exponential absorbance-time
curve whose final absorbance (A∞) is that of the carbyne
product. This behavior is consistent with an initial
rapid reaction, complete within the dead time of the
apparatus, to form an intermediate that subsquently
converts to the carbyne product.
Analysis of the data in Figure 1 gives k4 ) 0.28 (
0.01 s-1 and k3 ) 74 ( 1 dm3 mol-l s-l.
The addition of the conjugate base, NEt3, while
keeping the concentration of [NHEt3+] constant, per-
turbs the kinetics (Figure 1 inset).
Consideration of the mechanism shown in Scheme 1
leads to the rate law of eq 3 (see Appendix 1 for
derivation of this rate law). This mechanism describes
three pathways for the formation of the carbyne, and
the derivation of the rate law assumes that the rate-
limiting steps for each of these pathways are k2, k3, and
k4 and that initial protonation of the metal by [NHEt3]+
(Kl) is a rapidly established equilibrium. A linear
dependence on the concentration of NEt3 is observed
(Figure 1) and hence we can conclude that, in the
denominator, Kl[Et3NH+]/[NEt3] >> 1. Under these
(9) Carvalho, M. F. N. N.; Henderson, R. A.; Pombeiro, A. J . L.;
Richards, R. L. J . Chem. Soc., Chem. Commun. 1989, 1796.
(10) Doherty, N. M.; Hoffman, N. W. Chem. Rev. 1991, 91, 553 and
references therein.

Carbyne Complexes of Rhenium
Organometallics, Vol. 16, No. 25, 1997 5443
The k2 route corresponds to the direct â-protonation
of the vinylidene ligand, whereas the k3 and k4 ones
involve a prior protonation of the metal followed by an
acid-catalyzed rearrangement (k3) or by an intramo-
lecular 1,3-hydrogen migration (k4) from the metal to
the vinylidene â-carbon. The last route (k4) can be
considered to occur either by hopping of the hydrogen
atom along from the Re to the â-carbon atom or through
a more complex acid-catalyzed pathway such as that
postulated in Scheme 2. This involves an initial slow
insertion step (k4) of the vinylidene ligand into the
Re-H bond, a process that parallels the known11
hydride migration to a carbene ligand in other systems.
The derived vinyl ligand, in Re-C(H)dCHR+, following
a type of process reported12 for other vinyl complexes,
would then undergo proton addition at the â-carbon to
form the carbene species RedC(H)CH2R2+ which, upon
proton loss, would give the final carbyne product.
tr a n s-[ReCl(dCdCHBu t)(d p p e)2]. The reaction of
trans-[ReCl(dCdCHBut)(dppe)2] with [NHEt3][BPh4] in
THF occurs at a rate that shows no systematic variation
with the concentration of acid or base as shown in
Figure 3. This behavior is consistent with the reaction
proceeding exclusively by the hydride migration path-
way (k4). The tert-butyl group is a good electron donor
and hence would be expected to render the â-carbon
atom sufficiently basic to be protonated; however, its
bulky nature does not allow the close approach of the
relatively large [NHEt3]+ to the â-carbon site. Conse-
quently, a pathway involving less unfavorable steric
interactions dominates, in which initial protonation of
the metal (complete within the dead time of the ap-
paratus) is followed by rate-limiting intramolecular
migration (k4 ) 0.16 ( 0.04 s-1). We can establish a
limit for K1 for this reaction, by considering that even
when [NHEt3+] ) 5.0 × 10-3 mol dm-3 and [NEt3] ) 10
× 10-3 mol dm-3 (highest concentration of base) at least
90% of the rhenium is present as [Re(H)Cl(dCdCHBut)-
(dppe)2]+ and hence K1 g 18.
F igu r e 2. (kobs - k4)/[NHEt3+] vs [NEt3]/[NHEt3+], for
[NHEt3+] ) 1.0 × 10-3 mol dm-3, for the reaction of trans-
[ReCl(dCdCHC6H4Me-4)(dppe)2] (1.0 × 10-4 mol dm-3
)
with [NHEt3][BPh4] in THF, at 25 ( 0.1 °C, monitored at
+
λ ) 420 nm. The points corresponding to [NEt3]/[NHEt3
e 10, for which kobs approaches k4, have not been considered
in this analysis because of the great errors associated to
them. The line drawn is that defined by eq 4 and the values
given in the text.
]
conditions eq 3 simplifies to eq 4, from which eq 5 is
d[Re]
-
) (k4 K1[NHEt3+]/[NEt3] + (k3K1[NHEt3+]/
dt
[NEt3] + k2)[NHEt3+])/(1 + K1[NHEt3+]/[NEt3])[Re]
(3)
kobs ) k4 + (k3 + k2[NEt3]/K1[NHEt3+])[NHEt3+] (4)
kobs - k4
k2 [NEt3]
) k3 +
(5)
+
+
K
1 [NHEt3 ]
[NHEt3
]
obtained upon simple rearrangement. Since K1[Et3-
NH+]/[NEt3] >> 1, the equilibrium concentration of
[Re(H)Cl(dCdCHC6H4Me-4)(dppe)2]+ is much higher
than that of [ReCl(dCdCHC6H4Me-4)(dppe)2]. Addition
of NEt3 perturbs the K1 protolytic equilibrium back
towards the vinylidene complex. With appreciable
amounts of [ReCl(dCdCHC6H4Me-4)(dppe)2] present,
the slow but direct protonation of the vinylidene ligand
(k2 route) makes a significant contribution to the reac-
tion rate.
Analysis of the data requires a knowledge of the value
of k4 ) 0.28 s-l established from the studies in the
absence of NEt3 (eq 2). The graph of (kobs - k4)/
[NHEt3+] vs [NEt3]/[NHEt3+] gives k2/K1 ) 2.6 ( 0.3
dm3 mol-1 s-1 and k3 ) 60 ( 10 dm3 mol-1 s-1, which is
in good agreement with the value obtained for the
studies in the absence of NEt3 (Figure 2).
We can also estimate that k3 e 1.6 dm3 mol-l s-l for
the reaction of trans[ReCl(dCdCHBut)(dppe)2]. This
estimate is reached by considering that perhaps a small
(less than 10%) contribution from the acid-dependent
step may have gone undetected. This would give k3-
[NHEt3+] e 0.1k4, i.e., k3 < 1.6 dm3 mol-1 s-1
.
tr a n s-[ReCl(dCdCHCO2Me)(d p p e)2]. If the sub-
stituent on the vinylidene ligand is an electron-
withdrawing ester group (CO2Me or CO2Et), then the
kinetics of formation of the carbyne complex are quite
different. Although a first-order dependence on the
concentration of the complex is still observed, the
dependence on the concentration of [NHEt3]+ is more
complicated.
The important chemical feature about this kinetic
treatment is that the data demonstrate that at least
three pathways will convert vinylidene species to car-
byne complexes and that the values of the elementary
rate and equilibrium constants are similar to those
obtained for the analogous trans-[ReCl(dCdCHPh)-
(dppe)2] as would be expected. This kinetic approach
is employed with other vinylidene complexes described
in this paper. Which of the three pathways dominates
with these vinylidene complexes depends on both the
electronic and steric effects of R, as well as on the
electronic effect of the halide ligand (chloride or fluoride)
trans to the metal-binding vinylidene (Table 1).
The reaction of trans-[ReCl(dCdCHCO2Me)(dppe)2]
with [NHEt3][BPh4] in THF exhibits a nonlinear de-
pendence on the acid concentration as shown in Figure
(11) (a) Davey, C. E.; Osborn, V. A.; Winter, M. J .; Woodward, S. In
Advances in Metal Carbene Chemistry; Schubert U., Ed.; NATO ASI
Series C 269; Kluwer Academic Press: Dordrecht, The Netherlands,
1989; p 159. (b) Le Bozec, H.; Devanne, D.; Dixneuf, P. H., Reference
11a, p 107. (c) Burrell, A. K.; Clark, G. R.; J effrey, J . G.; Richard, C.
E. F.; Roper, W. R. J . Organomet. Chem. 1990, 388, 391. (d) Michelin,
R. A.; Bertani, R.; Mozzon, M.; Bombieri, G.; Benetollo: F.; Guedes da
Silva, M. F. C.; Pombeiro, A. J . L. Organometallics 1993, 12, 2372.
(12) (a) Casey, C. P.; Miles, W. H.; Tukada, H.; O’Connor, J . M. J .
Am. Chem. Soc. 1982, 104, 3761. (b) Grime, R. W.; Whiteley, M. W. J .
Chem. Soc. Dalton Trans. 1994, 1671.

5444 Organometallics, Vol. 16, No. 25, 1997
Ta ble 1. Ra te Con sta n ts for P r oton a tion Rea ction s of Vin ylid en e Com p lexes
Carvalho et al.
mechanism k2/K1 (or K1)b
compound
typea
(dm3 mol-1 s-1
)
k4 (s-1
)
k3 (dm3 mol-1 s-1
)
K5 (dm3 mol-1
)
k6 (s-1
)
trans-[ReCl(dCdCHPh)(dppe)2]
trans-[ReCl(dCdCHC6H4Me-4)(dppe)2]
trans-[ReCl(dCdCHBut)(dppe)2]
trans-[ReCl(dCdCHCO2Me)(dppe)2]
trans-[ReCl(dCdCHCO2Et)(dppe)2]
trans-[ReF(dCdCHCO2Et)(dppe)2]
1
1
1
2
2
1
17.6 ( 0.5
2.6 ( 0.3
(K1 g 18)c
0.073 ( 0.002 9.4 ( 0.5
0.28 ( 0.01
0.16 ( 0.04
74 ( 1
e1.6d
e3.8e (K1 g 36)c 0.084 ( 0.005
213 ( 1
24 ( 1
0.54 ( 0.05
0.7 (0.3
7.3 ( 0.5
0
0.023 ( 0.004
10 ( 6
9.2 × 103 ( 0.8 × 103
Types 1 and 2 correspond to Schemes 1 and 2, respectively. b K1 values (dimentionless) given in parentheses. c Assuming K1 equilibrium
a
d
lying at least 90% to the right-hand side. Assuming that the k3 pathway contributes less than 10% for kobs
.
e Assuming the k2 pathway
contributes less than 10% for kobs
.
Sch em e 2
addition, the curve does not go through the origin but
has an intercept of kobs ) 0.084 ( 0.005 s-l. Analysis
of these data gives the empirical rate law shown in eq
6. This rate law is consistent (see below) with the
d[ReCl(dCdCHCO2Me)(dppe)2]
-
)
dt
+
0.084 + 114[NHEt3
]
+
1 + 213[NHEt3
]
[ReCl(dCdCHCO2Me)(dppe)2] (6)
mechanism shown in Scheme 3, which is an elaboration
of the mechanism in Scheme 1. The only distinction
between the two schemes is the timing of proton transfer
from [NHEt3]+ to the vinylidene ligand. In Scheme 1,
protonation of the vinylidene ligand is a single, ther-
modynamically favorable reaction. In Scheme 3 proton
transfer involves two steps: formation of a hydrogen-
bonded adduct (which accumulates to detectable con-
centrations) followed by rate-limiting proton transfer.
This nonlinear dependence on the concentration of
[NHEt3]+ is due to the electron-withdrawing ester
substituent, which decreases electron density at the
â-carbon, and thus proton transfer from [NHEt3]+
becomes slow. Consequently, [NHEt3]+ hydrogen bonds
to the vinylidene carbon, but transfer of the proton from
N to C occurs slowly within the hydrogen-bonded
complex.
F igu r e 3. Dependence of kobs vs acid concentration for the
reaction of trans-[ReCl(dCdCHBut)(dppe)2] (1.0 × 10-4 mol
dm-3) with [NHEt3][BPh4] in THF, at 25 ( 1 °C, monitored
at λ ) 420 nm. Inset: Dependence of kobs vs base concen-
tration at constant [NHEt3][BPh4] ) 5.0 × 10-3 mol dm-3
.
The addition of the conjugate base NEt3 at a constant
acid concentration did not affect the rate of the reactions
(Figure 4 inset).
The general rate law associated with the mechanism
in Scheme 3 is given by eq 7. In deriving this rate law
d[ReCl(dCdCHCO2R)(dppe)2]
-
) k2[NHEt3+] +
(
dt
+
[NHEt3
[NEt3]
]
(k4 + K5k6[NHEt3+])K1
/
)
+
[NHEt3
]
1 + (1 + K5[NHEt3+])K1
(
)
[NEt3]
F igu r e 4. kobs vs acid concentration for the reaction of
[ReCl(dCdCHCO2R)(dppe)2] (7)
trans-[ReCl(dCdCHCO2Me) (dppe)2] (1.0 × 10-4 mol dm-3
)
with [NHEt3][BPh4] (a) or [NDEt3][BPh4] (b) in THF, at
25 ( 1 °C, monitored at λ ) 350 nm. Inset: kobs vs base
(NEt3) concentration, at [NHEt3][BPh4] ) 2.5 (O) or 5.0
mmol dm-3 (4). The curves drawn are those defined by eq
7 and the rate and equilibrium constants given in the text.
d[ReCl(dCdCHCO2R)(dppe)2]
-
)
dt
+
k4 + K5k6[NHEt3
]
[ReCl(dCdCHCO2R)(dppe)2]
+
1 + k5[NHEt3
]
(8)
4a. Thus, at low concentrations of acid, the rate of the
reaction exhibits a first order dependence on the con-
centration of [NHEt3+], but at high concentrations, the
rate is independent of the concentration of acid. In
(see Apendix 2) it is assumed that (i) protonation of the
metal is rapid and complete within the dead time of the
stopped-flow apparatus, and (ii) k4 and k6 are rate-

Carbyne Complexes of Rhenium
Organometallics, Vol. 16, No. 25, 1997 5445
Sch em e 3
limiting, with K5 being a rapidly established equilibri-
um. Moreover, when K1[NHEt3+]/[NEt3] >> 1, essen-
of the hydrogen-bonded adduct {RedCdCHCO2Me‚‚‚
H+NEt3}, from which the molar extinction coefficient ꢀ
) (7.0 ( 0.1) × 103 dm3 mol-1 cm-1 can be calculated.
At low concentrations of acid, the absorbance is that of
[Re(H)Cl(dCdCHCO2Me)(dppe)2]+ with ꢀ ) (5.90 (
0.05) × 103 dm3 mol cm-1. At intermediate concentra-
tions of [NHEt3]+, a mixture of these two species is
formed. Analysis13 of the effect of acid concentration
tially
stoichiometric
concentrations
of
[Re(H)X(dCdCHR)(dppe)2]+ are formed within the dead
time of the stopped-flow apparatus and hence the k2
pathway does not operate. Under these conditions eq
7 simplifies to eq 8, which is of the same form as that
observed experimentally (eq 6). Comparison of eqs 6
and 8 gives k4 ) 0.084( 0.005 s-1 (intercept of the
curve), K5 ) 213 ( 1 dm3 mol-l and k6 ) 0.54 ( 0.05
s-l.
Parenthetically it is worth emphasizing the similarity
between the mechanisms in Schemes 1 and 3. Equation
8 (Scheme 3) is similar to eq 2 (Scheme 1), except that
the term k3 has now been replaced by K5k6/(l +
K5[NHEt3+]); this replacement merely reflects the slow
proton transfer to the â-carbon atom in [ReH-
(dCdCHCO2R)(dppe)2]+ and hence the accumulation of
the hydrogen-bonded intermediate.
In this system, the limit K1 g 36 can be estimated by
considering that at least 90% of the rhenium complexes
are present as [Re(H)Cl(dCdCHCO2Me)(dppe)2]+, even
at the highest base concentration ([NEt3] ) 10 mmol
dm-3), and [NHEt3+] ) 2.5 mmol dm-3. We can also
estimate a limit for k2/K1 if in the studies with NEt3 a
10% contribution from the k2 pathway, to the total rate,
has been overlooked. The general rate law (7) assumes
the simplified form (9) if we employ the usual (see above)
on the initial absorbance gives K5 ) 210 ( 60 dm3 mol-1
,
in good agreement with the value determined kineti-
cally. The relatively large error associated with K5
estimated spectrophotometrically is a consequence of the
relatively small difference in molar extinction coef-
ficients of [Re(H)Cl(dCdCHCO2Me)(dppe)2]+ and its
hydrogen-bonded adduct with [HNEt3]+ (see above) and
of the consequent error of the estimated molar extintion
coefficient of the latter complex.
Corroboration that both k4 and k6 steps involve rate-
limiting proton transfer is the observation that both
steps show a primary kinetic isotope effect. The kinetics
of the reaction between trans-[ReCl(dCdCHCO2Me)-
(dppe)2] and [NDEt3]+ show the same kinetic behavior
as for the reaction with [NHEt3+] (Figure 4b). Analysis
D
of the data with [NDEt3+] gives k4 ) 0.025 ( 0.003
s-1, k6D ) 0.15 ( 0.02 s-1, and K5D ) 127 ( 8 dm3 mol-1
,
/
H
with the corresponding isotope effects k4H/k4D ) 3.4, k6
D
D
k6 ) 3.0, and K5H/K5 ) l.6. The marked primary
isotope effect for k6 confirms that slow proton transfer
is occurring in this step. Significantly, the k4 step is
also associated with a primary isotope effect. This is
consistent with our proposal that the k4 step corre-
sponds to intramolecular migration of the hydrido ligand
from rhenium to the â-carbon site. The origin of this
hydride ligand is the acid, and thus, in the reactions
with [NDEt3]+, [Re(D)Cl(dCdCHCO2Me)(dppe)2]+ is
formed and the migration of the deuteride is slower than
the analogous hydride. The smaller difference between
+
(k2/K1)[NEt3] + k4 + K5k6[NHEt3
]
kobs
)
(9)
+
1 + K5[NHEt3
k2[NEt3]
K1{1 + K5[NHEt3+]}
]
e 0.1kobs
(10)
simplification Kl[NHEt3+]/[NEt3] >> 1, and eq 10 fol-
lows if the above contribution from k2 is included.
Since K5 ) 213 dm3 mol-l when [NHEt3+] ) 2.5 mmol
dm-3 and kobs ) 0.25 s-1 (see Figure 4), and the highest
(13) Assuming that protonation of the metal is rapid and complete
within the dead time of the stopped-flow apparatus and that K5 is a
rapidly established equilibrium, the initial absorption (Ai) is given by
the following expression:
concentration of NEt3 used is [NEt3] ) 10.0 mmol dm-3
,
we calculate k2/K1 e 3.8 dm3 mol-l s-1
.
{ꢀ(ReH) + ꢀ(ReH‚NHEt3)K5[NHEt3+]}[Re]
Ai )
Spectroscopic evidence for the formation of the hy-
drogen-bonded adduct is obtained by careful inspection
of the absorbance-time curves. The initial absorbance
of these depends on the concentration of [NHEt3+]. At
high concentration of acid, the initial absorbance is that
+
1 + K5[NHEt3
]
In this equation [Re] is the total concentration of rhenium, ꢀ(ReH) )
(7.0 ( 0. l × 103 dm3 mol-1 cm-1 and ꢀ(ReH‚NHEt3) ) (5.90 ( 0.05 ×
103 dm3 mol-1 cm-1. From the data with [Re] ) l.0 × l0-4 mol dm-3
,
K5 ) 210 ( 60 dm3 mol-1
.

5446 Organometallics, Vol. 16, No. 25, 1997
Carvalho et al.
k2 [NEt3]
kobs - k4
) K5k6 +
(11)
+
+
K
1[NHEt3 ]
[NHEt3
]
Using k4 ) 0.023 s-l, graphical analysis of the data
(Figure 6) gives k2/Kl ) 7.3 ( 0.5 dm3 mol-1 s-1 and
K5k6 ) 17.4 ( 0.8 dm3 mol-1 s-1. The latter value is in
excellent agreement with the product of the K5 and k6
values estimated from the data in Figure 5 (K5 k6 ) 17
( 2 dm3 mol-1 s-1).
tr a n s-[ReF (dCdCHCO2Et)(d p p e)2]. The reaction
of trans-[ReF(dCdCHCO2Et)(dppe)2] with [NHEt3]-
[BPh4] exhibits a first-order dependence on the concen-
tration of both acid and complex (Figure 7). NEt3 has
no effect on the rate of the reaction (Figure 7 inset). This
behavior is consistent with the mechanism in Scheme
1, involving initial protonation at rhenium. Direct
protonation of the vinylidene (k2 route) is not detected.
Carbyne formation occurs by proton migration to the
vinylidene â-carbon (k4 route) or further proton attack
at this atom (k3 route).
F igu r e 5. kobs vs acid concentration for the reaction of
trans-[ReCl (dCdCHCO2Et)(dppe)2] (1.0 × 10-4 mol dm-3
)
with [NHEt3][BPh4].
Analysis of the data as outlined for the other systems
in this paper, according to Scheme 1 (eq 2), gives k3 )
(9.2 ( 0.8) × 103 dm3 mol-1 s-1 and k4 ) 10 ( 6 s-1
,
respectively.
An intriguing feature of this study is that protonation
of trans[ReF(dCdCHCO2Et)(dppe)2] is faster than that
of the chloro analogue. A similar effect of the trans-
halide group on the proton-affinity of ligands has been
observed15 in the deprotonation of trans-[M(NH)X-
(dppe)2] (M ) Mo or W; X ) F, Cl, Br, I) and trans-
[M(NNH2)X(dppe)2]+. ln all cases, the behavior can be
rationalized if the fluoro ligand is a better overall
electron releaser to the metal than the chloro ligand,
thus facilitating protonation of the vinylidene group and
retarding deprotonation of imide or hydrazide ligands.
The stronger electron donation of fluoride compared to
chloride opposes the electron-withdrawing effect of the
ester group, and consequently the hydrogen-bonded
adduct does not accumulate in detectable concentra-
tions.
F igu r e 6. (kobs-k4)/[NHEt3+] vs [NHEt3+] at [NHEt3+] )
2.5 (∆) and 5 mM (9) for the reaction of trans-[ReCl-
(dCdCHCO2Et)(dppe)2] (1.0 × 10-4 mol dm-3) with [NHEt3]-
[BPh4] (s, linear regression).
H
D
the K5 and K5 values is in agreement with the
expected14 isotope effects on equilibrium constants.
tr a n s-[ReCl(dCdCHCO2Et)(d p p e)2]. The reaction
of trans-[ReCl(dCdCHCO2Et)(dppe)2] with [NHEt3]-
[BPh4] shows a behavior similar to that exhibited by
the methyl ester. However, the nonlinear dependence
of the reaction rate on the concentration of [NHEt3]+
(Figure 5) is less marked with the ethyl ester. Analysis
of the data in the manner described above gives k4 )
0.023 ( 0.004 s-l, k6 ) 0.7 ( 0.3 s-l, and K5 ) 24 ( 1
dm3 mol-l.
Con clu sion s
Although analysis of the product of protonation of the
vinylidene ligand apparently indicates direct regiospe-
cific protonation of the â-carbon, it is only by detailed
mechanistic studies, such as those described herein, that
a variety of pathways leading to the carbyne product is
revealed. In particular, for electron-rich complexes
initial protonation can occur at either the metal or the
vinylidene ligand. In the systems discussed in this
paper, the kinetically controlled product is the hydrido
species [ReX(H)(dCdCHR)(dppe)2]+; however, subse-
quent rearrangement pathways ultimately result in the
thermodynamic carbyne product [ReX(tCCH2R)(dppe)2]+
in which the proton binds the carbon ligand.
It is important to emphasize that only three pathways
are necessary to describe the mechanisms by which all
the vinylidene complexes discussed herein are converted
to carbyne. Direct, regiospecific protonation of the
â-carbon (k2 route), regiospecific protonation at the
metal (K1 route) followed by intramolecular transfer of
It appears that the bulky ethyl substituent hinders
the approach of the [NHEt3]+ ion to the â-carbon site
more than the methyl group.
In contrast to the reaction of trans-[ReCl(dCdCHCO2-
Me)(dppe)2] with [Et3NH][BPh4] (Figure 4 inset), addi-
tion
of
NEt3
in
the
reaction
between
trans[ReCl(dCdCHCO2Et)(dppe)2l and [NHEt3]+ per-
turbs the kinetics as shown in Figure 6. Addition of
NEt3 disturbs the protolytic equilibrium K1, resulting
in accumulation of the parent vinylidene complex, thus
promoting the k2 route relatively to the K1 pathway.
This can be rationalized since at low concentrations of
[NHEt3]+, where the concentration of the hydrogen-
bonded adduct does not accumulate, K5 [NHEt3+] <<
1, eq 9 simplifies to eq 11.
(14) Bell, R. P. The Proton in Chemistry; 2nd ed.;Chapman and
Hall: London, 1973; Chapters 11 and 12 and references therein.
(15) Lane, J . D.; Henderson, R. A. J . Chem. Soc., Dalton Trans. 1986,
2155.

Carbyne Complexes of Rhenium
Organometallics, Vol. 16, No. 25, 1997 5447
PCH2CH2PEt2), initial protonation is at the ligand,
although protonation to the metal can compete to form
the hydrido product (as in the dinitrogen complexes).
Such a metal/ligand competition for the proton has also
recently been recognized6b for the alkynyl complex
[MoH3(CtCBut)(dppe)2].
Exp er im en ta l Section
All manipulations in the synthetic and kinetic studies were
routinely performed under an atmosphere of dinitrogen using
standard Schlenk or syringe techniques as appropriate. The
solvent, THF, was freshly distilled immediatly prior to use.
The chloro-vinylidene complexes were prepared by treat-
ment of a THF or toluene solution of trans-[ReCl(N2)(dppe)2]
with the appropriate l-alkynel7 in refluxing conditions or under
tungsten filament light or sunlight.8a The fluoro-vinylidine
complex trans-[ReF(dCdCHCO2Et)(dppe)2] was obtained by
treatment of a CH2Cl2 solution of the carbyne complex trans-
[ReF(tCCH2CO2Et)(dppe)2][BF4] with [NBu4]0H.8a,18 The pu-
rity of these complexes was established by elemental analyses
and IR, 31P{1H}, and 1H and/or 13C NMR spectroscopies.
Standard solutions of [NHEt3][BPh4], the complexes, and
NEt3 were prepared in THF under an atmosphere of dinitro-
gen. The acid or (base + acid) working solutions were
prepared, in volumetric flasks, by the appropriate dilution of
the corresponding standard solutions. All the solutions were
transferred to the stopped-flow apparatus using gas-tight, all-
glass syringes. All kinetic studies were complete within 1 h
of preparing the stock acid and base solutions in order to
minimize any complications associated with the acid-catalyzed
ring opening of THF.
The kinetics were studied on a Canterbury SF-40 stopped-
flow spectrophotometer with a spectrophotometer unit SU-40,
from Hi-Tech Scientifics, at 25 °C, monitoring the absorbance
changes associated with the rhenium complexes at the wave-
lengths shown in the figures.
The absorbance-time traces were single exponentials in all
cases and the rate constants, together with the initial and final
absorbances were computed by use of the Rapid Kinetics
Software Suite (version 1.0) program on a City Desk 386-SX
computer interfaced to the stopped-flow spectrophotometer.
Curves were exponential for at least three half-lives.
F igu r e 7. kobs vs acid concentration for the protonation
of trans-[ReF (dCdCHCO2Et)(dppe)2] (1.0 × 10-4 mol
dm-3) by [NHEt3][BPh4] in THF, at 25 ( 1 °C, monitored
at λ ) 420 nm. Inset: (kobs - k4)/[NHEt3+] vs [NEt3]/
[NHEt3+] at [NHEt3+] ) 2.5 × 10-3 mol dm-3
.
the hydrogen to the â-carbon (k4 route), or further
protonation of the hydride species at the â-carbon (k3
route).
With trans-[ReCl(dCdCHR)(dppe)2] (R ) C6H5, C6H4-
Me-4), all three pathways are observed, while with the
R ) But complex, the bulky nature of this substituent
enforces the exclusive use of the intramolecular path-
way. The effect of the more electron-withdrawing ester
substituents in trans-[ReCl(dCdCHCO2R)(dppe)2] (R )
Me, Et) decreases the electron density at the â-carbon.
This results in proton transfer to the â-carbon site
becoming rate-limiting and detection of the precursor
hydrogen-bonded adduct.
Finally, the mechanisms of protonation of the vi-
nylidene ligand can be modulated by changing the trans
halide. Thus, although trans-[ReCl(dCdCHCO2Et)-
(dppe)2] reacts by all three pathways, the analogous
trans-[ReF(dCdCHCO2Et)(dppe)2] reacts via the hy-
drido species only. In addition, the fluoro ligand ap-
parently renders the â-carbon sufficiently basic that the
hydrogen-bonded adduct is too short-lived to be de-
tected.
Ack n ow led gm en t. This work was partially sup-
ported by the J NICT (Portugal)/British Council (U.K.)
Protocol, by J NICT and the PRAXIS XXI Programme
(Portugal), and by the EC Network ERBCHRXCT
940501. We also thank Dr. R. L. Richards for stimulat-
ing discussions.
In general, for an unsaturated hydrocarbon residue
bound to an electron-rich site such as {M(dppe)2} (M )
Mo, W, Re), it appears that there is a fine balance as to
which site (metal or carbon) the proton attacks most
rapidly, and a priori it is as yet not possible to predict
which of these two is protonated more rapidly. For
Ap p en d ix 1. Der iva tion of th e Ra te La w for th e
Mech a n ism of Sch em e 1
Assuming that Kl is a rapidly established equilibrium
and that the rate-limiting steps are k2, k3, and k4, the
rate law is given by expression A.1.
instance, with the allene complex trans-[ReCl(η2-CH2-
d[Re]
dt
CdCHPh)(dppe)2], initial protonation is at the metal (as
observed in the vinylidene complexes), and only subse-
quently is the thermodynamic η2-vinyl product trans-
-
) k2[NHEt3+][Re]e + k3[NHEt3+][ReH]e +
k4[ReH]e (A.1)
[ReCl{C(CH2)CH2Ph}(dppe)2]+ formed4. In contrast, for
trans-[ReCl(NCR)(dppe)2] (R ) aryl), initial protonation
is at the nitrile ligand to form a methyleneamido ligand
in [ReCl(NCHR)(dppe)2]+. Moreover, with trans-[ML2-
(diphos)2] (M ) Mo, W; L ) CNR (R ) Me, But),5 CH2 )
CH2;6a diphos ) dppe; L ) N2;16 diphos ) dppe, Et2-
where [Re]e and [ReH]e stand for the equilibrium
concentrations of trans-[ReX()CdCHR)(dppe)2] and
[Re(H)X()CdCHR)(dppe)2]+, respectively, and [Re] is
(17) Pombeiro, A. J . L.; Almeida, S. S. P. R.; Silva, M. F. C. G.;
J effrey, J . C.; Richards, RL. J . Chem. Soc., Dalton Trans. 1989, 2381.
(18) Almeida, S. S. P. R.; Frausto da Silva, J . J . R.; Pombeiro, A. J .
L. J . Organomet. Chem. 1993, 450, C7.
(16) Henderson, R. A. J . Chem. Soc., Dalton Trans. 1984, 2259.

5448 Organometallics, Vol. 16, No. 25, 1997
Carvalho et al.
the total concentration of rhenium in solution, i.e.
d[Re]
dt
-
) k2[NHEt3+][Re]e + k4[ReH]e +
k6[ReH‚NHEt3]e (A.7)
[Re] ) [Re]e + [ReH]e
(A.2)
where [Re]e, [ReH]e, and [ReH‚NHEt3+]e are the equi-
librium concentrations of trans-[ReX(dCdCHR)(dppe)2],
[Re(H)X(dCdCHR)(dppe)2]+, and [Re(H)(dCdCHR)
(dppe)2]+‚‚‚HNEt3+, respectively, and [Re] is the total
concentration of rhenium in solution i.e.
K1 is given by
[ReH]e[NEt3]
K1 )
+
[Re]e[NHEt3
]
(A.3)
+
[Re] ) [Re]e + [ReH]e + [ReH‚NHEt3
]
(A.8)
e
From eqs A.2 and A.3, expressions relating [Re]e and
[ReH]e in terms of [Re] (eqs A.4 and A.5) are derived
which, upon substitution into eq A.1, lead to eq A.6
which is eq 3 of the text.
K1 and K5 are given by eqs A.3 and A.9, respectively.
+
[ReH‚NHEt3
]
K5 )
(A.9)
+
[ReH]e[NHEt3
]
[Re]
From eqs A.8, A.3, and A.9, expressions relating [Re]e,
[ReH]e and [ReH‚NHEt3]e in terms of [Re] (eqs A.10-
A.12) are derived which, upon substitution into eq A.7,
lead to eq A.13 which corresponds to eq 7 of the text.
[Re]e )
(A.4)
+
[NHEt3
]
1 + K1
[NEt3]
[Re]
+
[Re]e )
(A.10)
[NHEt3
]
+
[NHEt3
]
K1
[Re]
+
[NEt3]
1 + K1
(1 + K5[NHEt3 ]
[NEt3]
[ReH]e )
(A.5)
+
[NHEt3
[NEt3]
]
+
1 + K1
K1[NHEt3
]
[ReH]e )
[ReH‚NHEt3+]e )
d[Re]
[Re]e
(A.11)
[NEt3]
+ 2
d[Re]
dt
K1K6[NHEt3
]
-
)
[Re]e (A.12)
[NEt3]
+
+
[NHEt3
[NEt3]
]
[NHEt3
[NEt3]
]
+
k4K1
+ k3K1
+ k2 [NHEt3
]
(
)
-
)
[Re]
dt
+
[NHEt3
[NEt3]
]
+
[NHEt3
[NEt3]
]
1 + K1
k2[NHEt3+] + {k4 + K5k6[NHEt3+]}K1
[Re]
(A.6)
+
[NHEt3
[NEt3]
]
1 + K1
{1 + K5[NHEt3+]}
Ap p en d ix 2. Der iva tion of th e Ra te La w for th e
Mech a n ism of Sch em e 3
(A.13)
Su p p or tin g In for m a tion Ava ila ble: Tables of stopped-
flow kinetic data (6 pages). Ordering information is given on
any current masthed page.
Assuming that K1 and K5 are rapidly established
equilibria, and that steps k4 and k6 are rate-limiting,
the rate law is given by expression A.7.
OM970398G