Insertions of electrophiles into metal hydride bonds. Reactions of (C5Me5)Rh(PMe3)H2 with activated alkynes to produce η2-alkene complexes

Article abstract of DOI:10.1016/S0020-1693(99)00457-0

The dihydride complex (C₅Me₅)Rh(PMe₃)H₂ (1) reacts with the electron deficient alkynes phenylacetylene (PhC≡CH), diphenylacetylene (PhC≡CPh), methyl propiolate (HC≡CE, E = CO₂Me), dimethylacetylenedicarboxylate (EC≡CE, DMAD) and hexafluoro-2-butyne (F₃CC≡CCF₃) to produce in all cases η²-olefin complexes of the hydrogenated alkyne (2a-j). The complexes (C₅Me₅)Rh(PMe₃)(η²-methacrylate) (2a, 2b), (C₅Me₅)Rh(PMe₃)(η²-styrene) (2c, 2d) and (C₅Me₅)Rh(PMe₃)(η²-cis-stilbene) (2e, 2f) exist as pairs of rotamers that do not interconvert. The reaction of 1 with PhC≡CPh also produces (C₅Me₅)Rh(PMe₃)(η²-trans-stilbene) (2g) at higher temperatures. Insertion of F₃CC≡CCF₃ into 1 occurs stereoselectively to yield (C₅Me₅)Rh(PMe₃)(η²-trans-hexafluoro-2-butene) (2h). A mixture of cis- and trans- insertion products is observed in the reaction of 1 with EC≡CE to yield (C₅Me₅)Rh(PMe₃)(η²-fumarate) (2i) and (C₅Me₅)Rh(PMe₃)(η²-maleate) (2j) complexes. A crossover experiment using 1 and 1-d₂ with DMAD shows only d₀ and d₂ η²-olefin products, demonstrating that both hydrogens come from the same molecule of 1. A second crossover experiment using (C₅Me₅)Rh(PMe₃)(η²-fumarate)-d₂ and free dimethylfumarate shows no evidence for incorporation of the free olefin. The vinyl hydride intermediates (C₅Me₅)Rh(PMe₃)(CR≡CHR)H are observed below - 20°C for R = E and at room temperature for R = CF₃. Cyclic voltammetry of the metal complex 1 and of the substituted alkynes shows irreversible behavior and half-wave redox potentials E 1/2 are reported. CHCl₃ quenching of the vinyl hydride products from the reaction of 1 with F₃CC≡CCF₃ yields crystals of the complex (C₅Me₅)Rh(PMe₃)[trans-C(CF₃)=CH(CF₃)]Cl (5a) which was established as the trans complex by X-ray crystallography. (C) 2000 Elsevier Science S.A.

Full text of DOI:10.1016/S0020-1693(99)00457-0

www.elsevier.nl/locate/ica
Inorganica Chimica Acta 300–302 (2000) 138–150
Insertions of electrophiles into metal hydride bonds. Reactions of
(C5Me5)Rh(PMe3)H2 with activated alkynes to produce h2-alkene
complexes
Anthony D. Selmeczy, William D. Jones *
Department of Chemistry, Uni6ersity of Rochester, Rochester, NY 14627, USA
Received 5 August 1999; accepted 8 October 1999
Abstract
The dihydride complex (C5Me5)Rh(PMe3)H2 (1) reacts with the electron deficient alkynes phenylacetylene (PhCꢀCH),
diphenylacetylene (PhCꢀCPh), methyl propiolate (HCꢀCE, E=CO2Me), dimethylacetylenedicarboxylate (ECꢀCE, DMAD) and
hexafluoro-2-butyne (F3CCꢀCCF3) to produce in all cases h2-olefin complexes of the hydrogenated alkyne (2a–j). The complexes
(C5Me5)Rh(PMe3)(h2-methacrylate) (2a, 2b), (C5Me5)Rh(PMe3)(h2-styrene) (2c, 2d) and (C5Me5)Rh(PMe3)(h2-cis-stilbene) (2e,
2f) exist as pairs of rotamers that do not interconvert. The reaction of 1 with PhCꢀCPh also produces (C5Me5)Rh(PMe3)(h2-trans-
stilbene) (2g) at higher temperatures. Insertion of F3CCꢀCCF3 into 1 occurs stereoselectively to yield (C5Me5)Rh(PMe3)(h2-trans-
hexafluoro-2-butene) (2h). A mixture of cis- and trans- insertion products is observed in the reaction of 1 with ECꢀCE to yield
(C5Me5)Rh(PMe3)(h2-fumarate) (2i) and (C5Me5)Rh(PMe3)(h2-maleate) (2j) complexes. A crossover experiment using 1 and 1-d2
with DMAD shows only d0 and d2 h2-olefin products, demonstrating that both hydrogens come from the same molecule of 1. A
second crossover experiment using (C5Me5)Rh(PMe3)(h2-fumarate)-d2 and free dimethylfumarate shows no evidence for incorpo-
ration of the free olefin. The vinyl hydride intermediates (C5Me5)Rh(PMe3)(CRꢁCHR)H are observed below −20°C for R=E
and at room temperature for R=CF3. Cyclic voltammetry of the metal complex 1 and of the substituted alkynes shows
irreversible behavior and half-wave redox potentials E1/2 are reported. CHCl3 quenching of the vinyl hydride products from the
reaction of 1 with F3CCꢀCCF3 yields crystals of the complex (C5Me5)Rh(PMe3)[trans-C(CF3)ꢁCH(CF3)]Cl (5a) which was
established as the trans complex by X-ray crystallography. © 2000 Elsevier Science S.A. All rights reserved.
Keywords: Rhodium complexes; Hydride complexes; Alkyne complexes; Insertion
mechanisms have been proposed to account for these
experimental results.
1. Introduction
Many examples exist of alkyne insertions into metal
hydride complexes to give cis-products only [3]. Most
authors agree that formation of cis-products is likely to
arise from the four-centered transition state depicted in
Eq. (1).
The insertion of unsaturated organic substrates into
metal hydride bonds is one of the most fundamental
steps in both stoichiometric and catalytic organometal-
lic reactions such as hydrogenation, oligomerization,
and polymerization [1]. Olefin insertion into metal hy-
drides has been well investigated in terms of its relation-
ship to Ziegler–Natta polymerization [2]. The reactions
of alkynes with metal hydride complexes have also been
well studied, though no general trend has been pro-
posed to explain the assortment of regiochemical and
stereochemical outcomes for different systems. Several
(1)
Stereoselective formation of trans-insertion products is
less common and several different mechanisms have
been proposed to account for this phenomenon [4].
Other studies have shown that both cis- and trans-
products can be formed depending on the substrate
* Corresponding author. Tel.: +1-716-275 5493; fax: +1-716-473
6889.
E-mail address: jones@chem.rochester.edu (W.D. Jones)
0020-1693/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved.
PII: S 0 0 2 0 - 1 6 9 3 ( 9 9 ) 0 0 4 5 7 - 0

A.D. Selmeczy, W.D. Jones / Inorganica Chimica Acta 300–302 (2000) 138–150
139
used and the reaction conditions [5]. A few reports have
observed the interconversion of the vinyl hydride interme-
diates that ostensibly determine the final stereochemical
outcome of the reaction [6], and one has noted isomer-
ization of an h2-olefin complex [7]. An additional path-
way of terminal alkyne CꢂH bond activation has also been
seen [8].
showed two new methoxy resonances at l 3.624 (s, 3H)
and 3.483 (s, 3H) in a 2.2:1 ratio along with corresponding
C5Me5 and PMe3 resonances. In addition, a new group
of resonances appeared in the region l 2.6–1.9 with
complex coupling patterns consistent with a p-bound
olefin complex. No new hydride resonances were ob-
served. 31P{1H} NMR spectroscopy confirmed the pres-
ence of two new products with large JRh–P couplings
(ꢀ190 Hz) consistent with rhodium in the +1 formal
oxidation state [11]. These two products were assigned
as rotamers of the h2-methacrylate complex
(C5Me5)Rh(PMe3)[h2-CH2ꢁCH(E)] 2a and 2b.
A similar result was obtained by reacting 1 with
phenylacetylene (Eq. (2)). Addition of 1 equiv. of
PhCꢀCH to a benzene solution of 1 resulted in no color
change, and NMR spectroscopy confirmed that no
reaction had occurred at room temperature. Heating the
mixture at 80°C for 2 h caused the solution to become
bright red. 1H NMR spectroscopy revealed two new
ortho-phenyl resonances at l 7.501 (d, J=8.6 Hz, 2H)
and 7.444 (d, J=8.6 Hz, 2H) accompanied by new pairs
of C5Me5 and PMe3 resonances in a 2.4:1 ratio. The
characteristic h2-olefin resonances were observed in the
region l 2.1–3.8. Homonuclear decoupling experiments
in this region, however, showed that only two h2-olefin
resonances for each product were present within this
region. Irradiation at l 1.67 resulted in the simplification
of two of these resonances, confirming the existence of
a third peak masked by the C5Me5 resonances. Again no
new hydrides were seen, and the 31P{1H} NMR spectrum
showed two new products with coupling constants indica-
tive of Rh in the +1 formal oxidation state. These two
products were assigned as rotamers of the h2-styrene
complex (C5Me5)Rh(PMe3)[h2-CH2ꢁCH(Ph)] 2c and 2d.
These products were also observed previously in the
thermal reaction of (C5Me5)Rh(PMe3)(Ph)H (3) with
styrene [12].
Disparate mechanistic proposals have been offered to
account for competitive formation and/or isomerization
of cis- and trans-vinyl complexes, some of which include
electron transfer [5c] phosphine-promoted interconver-
sion [6c], and concerted trans-insertion [5a]. Few reason-
able mechanisms have been proposed to account for the
E/Z interconversion of h2-olefin complexes, apart from
photochemically induced isomerization. In addition, the
reactions of terminal and unsymmetrical alkynes with
metal hydride complexes show no consistent trends
concerningfactorsthatdeterminetheultimateregiochem-
istry of the insertion. In the case of terminal alkynes,
insertion can take place to yield either a- or b- substituted
vinyl complexes, and both regiochemical outcomes have
been observed [9]. A few papers have also reported
subsequent insertions of alkynes into the initially formed
vinyl complexes to yield butadienyl or even hexatrienyl
fragments [10].
In this section are reported the reactions of the
dihydride complex (C5Me5)Rh(PMe3)H2 (1) with the five
electron-deficient alkynes phenylacetylene (PhCꢀCH),
diphenylacetylene (PhCꢀCPh), methyl propiolate (HCꢀ
CE, E=CO2Me), dimethylacetylenedicarboxylate (EC
ꢀCE, DMAD) and hexafluoro-2-butyne (F3CCꢀCCF3).
In all cases the final products were h2-alkene complexes
of the hydrogenated alkyne. Intermediate vinyl hydride
complexes could be observed in the reactions of 1 with
dimethylacetylenedicarboxylate and hexafluoro-2-bu-
tyne. Mechanistic implications based on kinetic and
crossover experiments and electrochemical measurements
are discussed.
The reaction of 1 with diphenylacetylene was somewhat
more complicated (Eq. (3)).
2. Results
2.1. Reactions of 1 with alkynes
The reactions of 1 with phenylacetylene and with
methyl propiolate gave similar results Eq. (2).
(3)
Addition of excess PhCꢀCPh to a benzene solution of 1
led to no reaction at room temperature and a very slow
reaction at 70°C. After 140 h at 70°C two major products
could be identified in a 3:1 ratio with new ortho-phenyl
aromatic resonances at l 7.786 (d, J=7.4 Hz, 4H)
and 7.567 (d, J=7.2 Hz, 4H) and corresponding
(2)
Addition of excess HCꢀCE (E=CO2Me) to a benzene
solution of 1 at 25°C resulted in a rapid color change from
light tan to red over the course of a few minutes.
1
Inspection of the solution by H NMR spectroscopy

140
A.D. Selmeczy, W.D. Jones / Inorganica Chimica Acta 300–302 (2000) 138–150
h2-olefin peaks at 3.730 (d, J=2 Hz, 2H) and 3.391 (dd,
J=7.3, 2.4 Hz, 2H). New C5Me5 and PMe3 resonances
were also visible. The 31P{1H} NMR spectrum exhibited
large Rh–P couplings (JRh–P=196, 208 Hz) for the two
doublet resonances of the products. These two complexes
were assigned as rotamers of the h2-cis-stilbene complex
(C5Me5)Rh(PMe3)[h2-cis-CH(Ph)ꢁCH(Ph)] 2e and 2f.
After heating the mixture at 105°C for 2 days, however,
a major new product could be detected with inequivalent
h2-olefin resonances at l 4.325 (dd, J=10.6, 2 Hz, 1H)
and 4.013 (td, J=10.6, 2 Hz, 1H), a C5Me5 resonance
at l 1.521 (d, J=1.7 Hz, 15H) and a PMe3 resonance
at l 0.673 (d, J=8.4 Hz, 9H). The 31P{1H} NMR
spectrum showed a new doublet at l −4.046 (d, J=210
Hz). Continued heating for 6 days at 105°C resulted in
formation of this species as the primary product along
with a decrease in 2e and 2f. Based on the inequivalent
h2-olefin resonances, this complex was assigned as the
trans-stilbene complex (C5Me5)Rh(PMe3)[h2-trans-
CH(Ph)=CH(Ph)] 2g.
In order to confirm the identity of the products in the
reaction with diphenylacetylene, (C5Me5)Rh(PMe3)(Ph)-
H was treated with cis- and trans-stilbene. Heating a
hexane solution of (C5Me5)Rh(PMe3)(Ph)H at 65°C with
excess trans-stilbene led to loss of benzene and coordina-
tion of the free alkene. 1H and 31P{1H} NMR spectra were
identical to those for 2g above and allowed recording of
the 13C{1H} NMR spectrum in the absence of the
additional unknown products. When the same reaction
was carried out with cis-stilbene, however, 1H and
31P{1H} NMR spectra showed exclusive formation of the
trans-stilbene complex 2g along with the observation of
a white crystalline material precipitating from the hexanes
solution. 1H NMR identified the white material as
trans-stilbene, with no spectral evidence for any cis-stil-
bene remaining in solution. A control experiment of
thermolysis of a hexanes solution of cis-stilbene at 65°C
overnight resulted in no detectable isomerization. The
origin of this E/Z interconversion of the free olefin is
unknown, but is apparently catalyzed by the rhodium
complex.
When gaseous F3CCꢀCCF3 was condensed into a frozen
benzene solution of 1 on the high-vacuum line, an
immediatecolorchangetodarkyellowwasobservedupon
thawing the sample, accompanied by the thickening of
the solution to a viscous fluid1. Careful removal of the
volatiles left a dark orange–brown solid which was
1
dissolved in C6D6 and filtered. Examination of the H
NMR spectrum revealed two new products in a 2.5:1 ratio
with hydride resonances at l −13.120 (dd, J=46, 27 Hz,
1H) and −13.950 (dd, J=49, 28 Hz, 1H) and a complex
multiplet at l 7.11, due to overlapping vinylic hydrogens.
A
31P{1H} NMR spectrum showed two doublets at l
3.523 and 2.358 with small JRhꢂP couplings (146 and 140
Hz) indicating the presence of rhodium in the +3
oxidation state. These two products were assigned as vinyl
hydride complexes (C5Me5)Rh(PMe3)[C(CF3)ꢁCH-
(CF3)]H 3a and 3b. In order to determine the stereochem-
istry of the trifluoromethyl groups, 19F NMR spectra were
recorded. Previous work by Clark [5c] and others [14]
demonstrated that vicinal through-space 19Fꢂ19F cou-
pling is much larger for cis-trifluoromethyl groups
(5JFꢂF\10 Hz) than for trans-trifluoromethyl groups
(5JFꢂFB3 Hz). Inspection of the 19F NMR spectrum
showed a pair of resonances for each isomer: l 20.765
(br s) and 19.260 (d, J=8.6 Hz) for 3a; and l 17.134 (t,
J=8.7 Hz) and 15.278 (br s) for 3b. The absence of a
large quartet coupling for each resonance suggested that
the geometry about the double bond was trans, and
consequently 3a and 3b were assigned as rotamers. After
1 day at 25°C, a new product could be observed growing
in. A 1H NMR spectrum exhibited new C5Me5 and PMe3
resonances along with a complex pattern at l 2.439 (m,
2H). 31P{1H} NMR spectra showed the new product to
be a RhI complex (JRhꢂP=187 Hz) and 19F NMR
spectroscopy showed resonances at l 20.156 (dt, J=9.6,
3 Hz) and 17.513 (d, J=9.6 Hz), again indicating a
trans-arrangement of the CF3 substituents. This product
was assigned to be the h2-hexafluoro-2-butene complex
(C5Me5)Rh(PMe3)[h2-trans-CH(CF3)ꢁCH(CF3)] (2h).
Heating the sample at 100°C for 5 h resulted in quanti-
tative formation of 2h. A crystal structure of the chloride
complex (C5Me5)Rh(PMe3)[C(CF3)ꢁC(CF3)H]Cl (4a)
was obtained by quenching the initial mixture of vinyl
hydride complexes 3a and 3b with CHCl3. Fig. 1 shows
an ORTEP drawing of the molecule, confirming the trans
geometry of the vinyl group.
The reaction of 1 with hexafluoro-2-butyne at room
temperature was instantaneous (Eq. (4)).
(4)
The reaction of 1 with dimethyl acetylenedicarboxylate
was also rapid (Eq. (5)).
1 This may be due to oligomerization or polymerization of the
F3CC2CF3, catalyzed by some metal species. For references on poly-
merization of this alkyne, see: Ref. [13].

A.D. Selmeczy, W.D. Jones / Inorganica Chimica Acta 300–302 (2000) 138–150
141
matography and isolated (see Section 5). No evidence
for interconversion of isomers was detected after several
days in solution at room temperature.
2.2. Intermediates in alkyne insertion reactions
In an attempt to observe the intermediate vinyl hy-
dride complexes, the reaction was repeated at low tem-
perature. Addition of 5 equiv. of ECꢀCE to a frozen
d8-toluene solution of 1 was performed under a nitro-
gen purge and the solution warmed to −78°C and
gently swirled to mix the components. The NMR tube
was immediately transferred to the pre-cooled NMR
probe (−40°C) and 1H and 31P{1H} NMR spectra
recorded. Two new products could be detected by
31P{1H} NMR spectroscopy at l 11.103 (d, J=148.3)
and 8.336 (d, J=145.4 Hz) with small JRhꢂP values,
indicating RhIII complexes. Although incomplete 1H
NMR data were obtained due to DMAD impurities,
new hydride resonances could be seen at l −11.467
(dd, J=47.6, 23.2 Hz, 1H) and −13.744 (dd, J=48.2,
29.3 Hz, 1H), confirming that the new materials
were the expected vinyl hydride complexes (C5Me5)Rh-
(PMe3)[trans-C(E)ꢁCH(E)]H (3c) and (C5Me5)Rh-
(PMe3)[cis-C(E)ꢁCH(E)]H (3d). Smaller 1H and
31P{1H} NMR resonances were also detected for com-
plexes 2i and 2j. As the sample was slowly allowed to
warm, the resonances for 3c and 3d decreased with a
(5)
Addition of excess ECꢀCE (E=CO2Me) to a benzene
solution of 1 resulted in an immediate color change to
dark red. Inspection of the H NMR spectrum showed
1
formation of
a major product with inequivalent
methoxy resonances at l 3.509 (s, 3H) and 3.413 (s, 3H)
and inequivalent vinylic hydrogens at l 3.492 (ddd,
J=11.3, 9.5, 2 Hz, 1H) and 3.383 (dd, J=11.3, 2 Hz,
1H). A minor product was also detected with a new
methoxy peak at l 3.631(s, 6H) which integrated as two
OMe groups relative to its corresponding C5Me5 and
PMe3 resonances. A resonance at l 2.404 (dd, J=8.8,
2 Hz, 2H) showed that the vinylic hydrogens for this
material were also equivalent. The 31P{1H} NMR spec-
trum indicated that both of the new materials were RhI
complexes by their large coupling constants (JRhꢂP
=
191, 182 Hz). These two products were assigned as the
h2-dimethyl fumarate complex (C5Me5)Rh(PMe3)[h2-
trans-CH(E)ꢁCH(E)] 2i and the h2-dimethyl maleate
complex (C5Me5)Rh(PMe3)[h2-cis-CH(E)ꢁCH(E)] 2j.
These materials could be separated by thin-layer chro-
Fig. 1. ORTEP drawing of (C5Me5)Rh(PMe3)(trans-CCF3ꢁCHCF3)Cl.
Ellipsoids are shown at the 50% level, and hydrogen atoms (except on
,
vinyl ligand) have been omitted for clarity. Selected distances (A) and
angles (°): RhꢂCl(1), 2.414(3); RhꢂC(15), 2.05(1); C(14)ꢂC(15),
1.81(3); C(15)ꢂC(16), 1.11(3); C(16)ꢂC(17), 1.60(4); RhꢂC(15)ꢂC(16),
155(3), C(14)ꢂC(15)ꢂC(16), 90(2); C(15)ꢂC(16)ꢂC(17), 109(3).
Fig. 2. 31P{1H} NMR spectra of (a) 1, (b) 1-d2, and (c) mixture of 1
and 1-d2 showing that 1 and 1-d2 do not undergo H/D exchange.

142
A.D. Selmeczy, W.D. Jones / Inorganica Chimica Acta 300–302 (2000) 138–150
concomitant increase in 2i and 2j. The stereochemical
assignments for the two vinyl hydride complexes 3c and
3d is somewhat tentative, and is primarily based on the
observation that the total amount of trans-material (i.e.
[3c]+[2i]) and the total amount of cis-material (i.e.
[3d]+[2j]) each remained constant throughout the
reaction.
The origin of the vinylic hydrogens in the products 2i
and 2j was probed with a deuterium crossover experi-
ment. A mixture of 1-d0 and 1-d2 showed distinct
resonances in the 31P{1H} NMR spectrum due to an
intrinsic isotopic shift [15] and demonstrated that H/D
exchange does not occur between the two complexes
(Fig. 2(a)–(c)). Addition of DMAD to this solution
exhibited 31P{1H} NMR resonances for only 2i-d0,
2i-d2, 2j-d0, and 2j-d2 (Fig. 3(a)–(d)) with no d1 prod-
ucts detectable. Consequently, both hydrogens in prod-
ucts 2i or 2j came from the same molecule of dihydride
complex 1.
In order to determine if free olefin was generated
prior to formation of 2i and 2j or if intermolecular
olefin exchange was occurring, the following deuterium
labeling study was carried out. 1-d2 was allowed to
react with DMAD in the presence of free dimethyl
fumarate. Inspection of the 31P{1H} NMR spectrum
showed formation of 2i-d2 and 2j-d2, but no 2i-d0 or
2j-d0. This observation demonstrated that olefin ex-
change in this complex is slow. Also, no free dimethyl
fumarate or dimethyl maleate can be involved, since
none of the d0-olefin was incorporated into the h2-
olefin products.
Fig. 3. 31P{1H} NMR spectra of (a) 1 + DMAD, (b) 1-d2+DMAD,
(c) mixture of 2i, 2i-d2, 2j, and 2j-d2, (d) reaction of 1+1-d2 from
Fig. 2(c) with DMAD.
butyl peroxide) or in the presence of a radical spin
trapping agent (PBN, a-phenyl-N-tert-butyl nitrone)
showed comparable rates (kTBP=4.3(7)×10−4 M−1
The electrochemistry of complex 1 and the substi-
tuted acetylenes was investigated. Cyclic voltam-
mograms for PhCꢀCH, PhCꢀCPh, HCꢀCE, ECꢀCE
and F3CCꢀCCF3 and (C5Me5)Rh(PMe3)H2 were
s
−1, kPBN=5.8(7)×10−4 M−1 −1, see Section 5),
s
demonstrating that free radical species are not involved.
recorded in THF solvent (TBA+PF6 supporting elec-
−
trolyte) using a glassy-carbon working electrode against
an internal Fe(C5H5)2 reference (E° for ferrocene=
0.400 V). Reversible couples were not obtained even at
sweep rates up to 5 V s−1. Typical sweep rates were
200–500 mV s−1. In order to approximate the mid-
point between the measured reduction wave and the
unobserved oxidation couple the half-wave potential
E1/2 was defined as the midpoint of the measured
reduction peak current and the baseline current (see
Fig. 4). Complex 1 also showed an irreversible oxida-
tion wave, and its E1/2 was determined as for the
acetylenes. Reduction potentials for the substituted
acetylenes along with the oxidation potential for 1 are
reported in Table 1.
3. Discussion
Although the stereochemical outcome of alkyne in-
sertions into metal–hydride bonds has been well stud-
ied, many conflicting explanations have been invoked
to account for the disparate results. In cases where
cis-insertion occurs exclusively, a four-membered con-
certed transition state has been frequently invoked.
When trans-insertion is observed, the proposed mecha-
nisms to account for E/Z isomerization are diverse.
Clark et al. showed that an electron transfer mechanism
was the most likely for their trans-Pt(PR3)H2 system
based on EPR results and electrochemical measure-
ments [5c]. Otsuka and Nakamura proposed a con-
certed four-membered transition state that results in
direct formation of a trans-vinyl hydride complex [5a].
Bergman and Huggins examined the insertion of alky-
nes into the NiꢂCH3 bond of a series of Ni(acac)-
(PR3)CH3 complexes and concluded that phosphine
The reaction of 1 with HCꢀCE was found to be
bimolecular. Reaction of 1 (0.027 M) with HCꢀCE
(0.054 M) was found to fit a second order plot of
ln{([M]0−[M])/2[M]} versus time and was linear over 5
half-lives (kobs=5.9(5)×10−4 M−1 −1). Similar reac-
s
tions in the presence of a radical initiator (TBP, di-tert-

A.D. Selmeczy, W.D. Jones / Inorganica Chimica Acta 300–302 (2000) 138–150
143
electron-deficient alkyne were to occur, the ion-paired
intermediate A could then undergo a proton transfer
from the metal hydride cation radical to the alkynyl
anion radical, leading to the neutral cage pair B which
contains a vinyl radical. Rapid E/Z isomerization of
this radical species prior to cage collapse is consistent
with the mixture of cis- and trans-vinyl hydride materi-
als observed in this reaction and also with the predom-
inance of the trans-stereoisomer. In this scheme, the
initial ratio of vinyl hydride intermediates determines
the final ratio of h2 materials that are formed, as is
observed experimentally. Alternatively, H atom transfer
from 1 to the acetylene could occur, giving the same
vinyl radical intermediate.
Fig. 4. Sample cyclic voltammograms of HCꢀCCO2Me showing
estimation of E1/2 referenced to ferrocene.
Table 1
Irreversible oxidation and reduction potentials for (C5Me5)Rh-
(PMe3)H2 (1) and for substituted alkynes R1CꢀCR2 in THF/TBAPF6
To examine the feasibility of this mechanistic pro-
posal, the electrochemistry of the substituted alkynes
and the metal complex 1 was investigated by cyclic
voltammetry. Reversible couples were not obtained
even at the fastest sweep rates (5 V s−1), and irre-
versible values are reported as half-wave potentials E1/2
(see Table 1). The oxidation potential for complex 1
was measured to be +0.25 V versus NHE and was also
irreversible. Reduction potentials for the series of alky-
nes ranged from −1.03 V (F3CCꢀCCF3) to −2.65 V
(PhCꢀCH). From these redox values, DG for each
reaction was calculated from the relationship DG= −
R1ꢁ
R2ꢁ
E1/2 (V) a
DG° (kcal mol−1
)
Ph
MeO2C
Ph
CO2Me
CF3
1
H
H
Ph
CO2Me
CF3
–
−2.65
−2.45
−2.39
−1.49
−1.03
+0.25
51.9
50.7
49.4
28.6
18.0
–
a All values are referenced to reversible Fe(C5H5)2 couple at Eo=
0.400 V. E1/2 are one electron reductions in all cases except for 1
which is a one electron oxidation.
ox
1/2
red
1/2
nFErxn, where Erxn=E (1)+E (alkyne). As these
electron transfers are all uphill, they represent mini-
mum kinetic barriers for the rate of reaction. It is clear
that electron transfer rates calculated from these free
energies are far too slow to be responsible for the
reactions studied, and based on these E1/2 values outer
sphere electron transfer is not deemed a plausible mech-
anism in these cases2.
In the series of reactions studied here, all possible
stereochemical outcomes are observed depending upon
the acetylene substrate. Diphenylacetylene reacts with 1
to give cis-products only, with the observation of the
trans-stilbene adduct 2g only under severe conditions.
In contrast, hexafluoro-2-butyne gives stereospecifically
the trans-vinyl hydride and h2-olefin complex, while
dimethyl acetylenedicarboxylate gives a mixture of E
and Z products both at the vinyl hydride and the
h2-olefin stages. The rational for this range of behavior
remains unclear. However, it seems apparent from these
experiments that the stereochemical outcome of the
reaction is fixed by the stereoselectivity of the initial
insertion to produce the E and/or Z vinyl hydride
complex(es). Since vinyl hydride intermediates are not
loss followed by alkyne coordination and methyl migra-
tion to the least hindered alkyne carbon atom produced
vinyl derivatives that could rapidly interconvert stereo-
chemistry, leading to the observation of both kinetic
and thermodynamic products [6c]. In this case, the E/Z
isomerization was proposed to occur via a zwitterionic
vinyl intermediate that could be stabilized by | bases
such as free phosphine.
The observation of a mixture of E/Z isomers in the
reaction of 1 with DMAD could be accommodated by
an electron-transfer mechanism (Scheme 1). If single
electron transfer from the metal complex 1 to the
2 Due to the fact that irreversible couples were obtained, the
measured potentials are only taken as upper limits for the actual
reversible values. As a result, rates that are calculated based on these
irreversible electrochemical potentials cannot be slower, but may
indeed be faster. It is unlikely, however, that the E1/2 values differ
substantially from the reversible values, and therefore initial electron
transfer is kinetically too slow to explain the observed rates of
reaction.
Scheme 1.

144
A.D. Selmeczy, W.D. Jones / Inorganica Chimica Acta 300–302 (2000) 138–150
cis-vinyl hydride intermediate (C5Me5)Rh(PMe3)(cis-
CRꢁCHR)H as the kinetic product via acetylene inser-
tion. A separate equilibration of the vinyl hydride
would then yield the thermodynamically more stable
(C5Me5)Rh(PMe3)(trans-CRꢁCHR)H (and (C5Me5)Rh-
(PMe3)-(trans-h2-CHRꢁCHR)) materials in competi-
tion with (C5Me5)Rh(PMe3)(cis-h2-CHRꢁCHR) from
CpRh(PMe3)(cis-CRꢁCHR)H. Potential mechanisms
for isomerization of vinyl hydride intermediates are
shown in Scheme 2. Scheme 2a illustrates an
organometallic equivalent to an allylic rearrangement,
effectively a [1,3] sigmatropic H-shift. The rhodium
carbene intermediate C can freely rotate around CaꢂCb
producing the trans-vinyl hydride intermediates and
ultimately the trans-h2-olefin materials. A similar mech-
anism in Scheme 2b forms a rhodium carbene hydride
intermediate D that places a negative charge on Cb
which is stabilized by the attached substituent. For the
cases R=Ph and CO2Me, a resonance stabilized anion
is formed (a benzylic anion for R=Ph, and an enolate
anion for R=CO2Me) while in the R=CF3 case the
anion is stabilized by the inductive effect of an electron
withdrawing group. This latter pathway seems most
reasonable and consistent with the observations.
Assuming that the cis-vinyl hydrides are the kinetic
products formed in these reactions as shown in Scheme
3, several specific relationships between equilibrium
constants and relative rates must be true. If the rate of
equilibration of vinyl hydride complexes is fast relative
to k1 and k2, then Keqꢁ1 for R=Ph, Keqꢀ1 for
R=E, and Keqꢂ1 for R=CF3. On the other hand, it
is possible that k1 is faster than equilibration when
R=Ph (leading to cis-olefin product), but equilibration
is faster than k1 for R=CF3 (leading to trans-olefin
Scheme 2.
observed in the reaction with diphenylacetylene, the
cis-insertion adduct must form exclusively.
Although dihydride 1 is a coordinatively saturated 18
electron complex, it is known to react bimolecularly
with D2, PMe3 and CS2 [15,16]. In order to account for
the associative rate dependence on these substrates, the
possibility of h5“h3 ring slippage and hydride to ring
migration pre-equilibria were proposed. Of these reac-
tions, the insertion of CS2 into the metal hydride bond
of 1 is the most closely related to the insertion chem-
istry of the alkynes. Based on detailed kinetic studies in
these systems it was proposed that hydride to ring
migration to produce the intermediate (h4-C5Me5H)-
Rh(PMe3)H occurred, producing a vacant coordination
site and allowing pre-coordination of free substrate.
This mechanism accounts for the associative nature of
these reactions, which would otherwise require loss of a
ligand (i.e. PMe3 or H2) prior to reaction with the
substrate. Rates of H2 loss and PMe3 exchange are
slow, and alternative pathways are required to explain
the rapid reaction of 1 with alkynes and CS2.
In order to accommodate the observed stereochemi-
cal outcomes in the reactions of 1 with PhCꢀCPh,
ECꢀCE and F3CCꢀCCF3, the possibility of hydrogen
atom transfer from 1 to the alkyne to form a vinyl
radical species was considered. The observation of only
cis-materials for the reaction of 1 with PhCꢀCPh, how-
ever, is inconsistent with this mechanism as the
diphenyl vinyl radical formed would be expected to
isomerize quickly to give trans-products. Also, the lack
of crossover in the reaction of 1-d2 with DMAD argues
against free vinyl radical intermediates, as does the
inability to trap the vinyl radical with a spin trapping
agent.
A second proposal that allows for the observation of
cis- and trans-materials would be formation of only the
Scheme 3.

A.D. Selmeczy, W.D. Jones / Inorganica Chimica Acta 300–302 (2000) 138–150
145
4. Conclusions
product). For R=CO2Me, equilibration appears to be
more rapid than reductive elimination since the same
ratio of cis/trans vinyl hydride intermediates is seen
throughout the reaction.
The complex (C5Me5)Rh(PMe3)H2 reacts with the
electron poor acetylenes PhCꢀCH, ECꢀCH, PhCꢀCPh,
ECꢀCE and F3CCꢀCCF3 to give h2-olefin complexes of
the hydrogenated alkynes. PhCꢀCH and ECꢀCH react
similarly to each yield two species assigned as rotamers.
Stereoselective cis-insertion (two rotamers) is observed
for PhCꢀCPh at 70°C, with isomerization to the trans-
complex only occurring above 100°C. A mixture of cis-
and trans-products is seen in the reaction with ECꢀCE
and stereoselective trans-insertion occurs in the reaction
with F3CCꢀCCF3. Vinyl hydride intermediates are de-
tected at −40°C for the ECꢀCE reaction and at room
temperature for the reaction with F3CCꢀCCF3. The
stereochemistry of the h2-olefin complexes is likely to
be determined by the initial stereochemistry of the vinyl
hydride intermediates.
Cyclic voltammetry measurements give irreversible
reduction potentials for the series of acetylenes and the
irreversible oxidation potential for 1. Based on these
E1/2 values, electron transfer is tentatively ruled out as
the first step in the insertion mechanism for all of the
alkynes except hexafluoro-2-butyne (and perhaps
dimethyl acetylenedicarboxylate).
The least satisfying explanation is to simply propose
different mechanistic pathways for each substrate. This
may be possible, however, given the extremely different
nature of the acetylenes used. Perhaps stereospecific
cis-insertion occurs for diphenylacetylene, while
stereospecific trans-insertion occurs for hexafluoro-2-
butyne. Both pathways would have to be competitive in
the reaction of 1 with DMAD, leading to both cis- and
trans-materials. There is ample precedent supporting a
four centered transition state leading to cis-stereochem-
istry for the vinyl hydride materials. In order to accom-
modate trans-stereochemistry, a perpendicular insertion
into the metal hydride shown in E below has been
invoked to explain the concerted formation of trans-
vinyl hydride materials without isomerization of an
initially formed cis-vinyl hydride complex.
5. Experimental
This mechanism is attractive in that it involves the same
intermediate h2-alkyne complex and differs only in the
orientation of the CꢀC bond in relation to the metal
hydride. Smaller substrates such as F3CCꢀCCF3 may
be able to achieve this transition state geometry while
bulkier alkynes such as PhCꢀCPh cannot. If distinct
paths are responsible for the observation of different
stereochemical outcomes in this series of reactions,
these two pathways that differ only in the trajectory of
the bound alkyne seem to be the most reasonable. The
direct insertion pathway also accounts for the lack of
crossover in the reaction of 1-d2 with DMAD.
Previous work studying the insertion of DMAD into
the metal hydride bond of (C5Me5)Rh(PMe3)(Ph)H
supports this proposal. Both cis- and trans-isomers of
the complex (C5Me5)Rh(PMe3)[C(E)ꢁCH(E)](Ph) were
isolated and structurally characterized [17]. Since this
vinyl complex has no metal hydride, the allylic rear-
rangement discussed above is not possible, and propos-
ing a zwitterionic mechanism to account for the
observed stereochemistry (leading to an intermediate
similar to D) seems unlikely as these materials are
stereochemically stable and do not interconvert. This
closely related system lends support to the notion of a
‘direct’ insertion of the acetylene into the metal hydride
bond that determines the E/Z stereochemistry about
the double bond.
5.1. General considerations
All operations and routine manipulations were per-
formed under an atmosphere of nitrogen, either on a
high-vacuum line using modified Schlenk techniques or
in a Vacuum Atmospheres dri-lab. Benzene, THF, and
hexanes were distilled from dark purple solutions of
benzophenone ketyl and stored in glass ampules fitted
with Teflon sealed vacuum line adapters. Alkane sol-
vents were made olefin-free by stirring over H2SO4,
washing with aqueous KMnO4 and water, and distilling
from dark purple solutions of tetraglyme/benzophenone
ketyl. Phenylacetylene, methyl propiolate, dimethyl
acetylenedicarboxylate, dimethyl fumarate, cis-stilbene,
trans-stilbene, Red–Al (sodium bis(2-methoxyethoxy)
aluminum hydride), sodium borohydride and sodium
borodeuteride were purchased from Aldrich and dis-
tilled before use. Bis-triphenylphosphino iminium chlo-
ride (PPN+Cl−) and diphenylacetylene were purchased
from Strem Chemicals and used as received. Hex-
afluoro-2-butyne was purchased from PCR gases and
used as received.
1H, 13C{1H}, and 31P{1H} spectra were recorded on
Bruker AMX400 or Bruker WP200 NMR spectrome-
ters. 19F NMR (470 MHZ) were obtained on a Varian
VXR500S spectrometer. All H NMR chemical shifts
are reported in ppm (l) relative to TMS (tetramethylsi-
1

146
A.D. Selmeczy, W.D. Jones / Inorganica Chimica Acta 300–302 (2000) 138–150
1
lane) and referenced to the residual solvent resonances
(benzene, l 7.15; chloroform, l 7.24; acetone, l 2.04;
THF, l 3.58, 1.73; toluene, l 2.10). 13C{1H} NMR
chemical shifts are referenced to residual C6D6 (l
128.00). 31P{1H} NMR chemical shifts are referenced to
external 10% H3PO4 (l 0.00). 19F NMR chemical shifts
are referenced to external F3CCO2H (l 0.00). All tem-
peratures for variable temperature NMR spectroscopy
were calibrated relative to the chemical shift differences
in the NMR spectra of known standards (4% methanol
in methanol-d4). Analyses were obtained from Desert
Rh(PMe3)(p2-cis-stilbene) Rotamer 1. H NMR (C6D6):
l 7.567 (d, J=7.2 Hz, 4H), 6.9–7.3 (obsc, 4H), 6.815
(t, J=7.8 Hz, 2H), 3.391 (dd, J=7.3, 2.4 Hz, 2Hvinylic),
1.467 (d, J=2.6 Hz, 15H), 0.821 (d, J=8.6 Hz, 9H).
31P{1H} NMR (C6D6): l 1.710 (d, J=196 Hz). Ro-
1
tamer 2. H NMR (C6D6): l 7.786 (d, J=7.4 Hz, 4H),
6.9–7.3 (obsc, 4H), 6.718 (t, J=7.2 Hz, 2H), 3.730 (d,
J=2 Hz, 2Hvinylic), 1.732 (d, J=1.6 Hz, 15H), 0.542 (d,
J=8.3 Hz, 9H). 31P{1H} NMR (C6D6): l −9.083 (d,
J=208 Hz). After heating at 105°C for 2 days a new
major product was formed. (C5Me5)Rh(PMe3)(p2-trans-
Analytics. PPN+BH4 and PPN+BD4 were prepared
by the method of Bau [18]. The complexes (C5Me5)-
Rh(PMe3)Cl2, (C5Me5)Rh(PMe3)Br2 [19], (C5Me5)-
Rh(PMe3)H2, (C5Me5)Rh(PMe3)(PMe3)(C2H4) [20], and
(C5Me5)Rh(PMe3)(Ph)H [21], have all been prepared
previously.
stilbene) H NMR (C6D6): l 7.995 (d J=7.4 Hz, 4H),
−
−
1
6.9–7.3 (obsc, 4H), 6.862 (t, J=7.4 Hz, 2H), 4.325 (dd,
J=10.6, 2 Hz, 1Hvinylic), 4.013 (td, J=10.6, 2 Hz,
1Hvinylic), 1.521 (d, J=1.7 Hz, 15H), 0.673 (d, J=8.4
Hz, 9H). 31P{1H} NMR (C6D6): l −4.046 (d, J=210
Hz). The product could be isolated as a red oil by
chromatography on silica gel, but the oil decomposes
upon standing at room temperature under nitrogen.
5.2. Reaction of 1 with phenylacetylene
20 mg (C5Me5)Rh(PMe3)Cl2 (0.052 mmol) was con-
5.4. Reaction of (C5Me5)Rh(PMe3)(Ph)H with
trans-stilbene
verted to the dihydride 1 using PPN+BH4 and dis-
−
solved in C6D6 in a resealable NMR tube. 6 ml (0.056
mmol) PhCꢀCH was injected under a nitrogen atmo-
sphere and the solution heated to 80°C for 2 h. The
solution turned red. The volatile components were re-
moved under vacuum leaving a dark red oil. NMR
spectroscopy showed two products. (C5Me5)Rh-
(PMe3)(p2-styrene) Rotamer 1. 1H NMR (C6D6): l
7.444 (d, J=8.6 Hz, 2H), 7.1–7.25 (obsc, 2H), 7.003 (t,
J=8.6 Hz, 1H), 3.427 (dtd, J=10.8, 10.1, 2.2 Hz,
1Hvinyl), 2.145 (d, J=10 Hz, 1Hvinyl), 1.67 (obsc,
1Hvinyl), 1.565 (d, J=2 Hz, 15H), 0.840 (d, J=8.2 Hz,
9H). 31P{1H} NMR (C6D6): l −6.318 (d, J=208 Hz).
13C{1H} NMR (C6D6): l 147.79 (s), 130.26 (s), 123.01
(s), 95.87 (t, J=3.4 Hz), 46.80 (dd, J=14.4, 2.5 Hz),
24.37 (dd, J=16.3, 3.4 Hz), 16.55 (d, J=25.4 Hz),
10.30 (s). Rotamer 2. 1H NMR (C6D6): l 7.501 (d,
J=8.6 Hz, 2H), 7.1–7.25 (obsc, 2H), 7.055 (t, J=8.6
Hz, 1H), 3.779 (ddd, J=11.6, 7.6, 2.8 Hz, 1Hvinyl),
2.198 (d, J=11.6 Hz, 1Hvinyl), 1.810 (d, J=2 Hz, 15H),
1.67 (obsc, 1 Hvinyl), 0.624 (d, J=8.9 Hz, 9H). 31P{1H}
NMR (C6D6): l −0.494 (d, J=199 Hz). 13C{1H}
NMR (C6D6): l 150.15 (s), 132.32 (s), 123.82 (s), 95.63
(t, J=3.4 Hz), 49.67 (dd, J=14.3, 1.7 Hz), 28.48 (dd,
J=16, 3.4 Hz), 18.00 (d, J=24.7 Hz), 10.413 (s).
20 mg (C5Me5)Rh(PMe3)(C2H4) (0.058 mmol) was
dissolved in C6H6 and photolyzed using a u\345 nm
filter for 2 days. 31P{1H} NMR spectroscopy showed
95%
conversion
to
the
phenyl
hydride
(C5Me5)Rh(PMe3)(Ph)H along with some formation of
(C5Me5)Rh(PMe3)2. The benzene was removed under
vacuum and the yellow solid dissolved in hexanes. 71
mg trans-stilbene (0.394 mmol) was added under nitro-
gen and the solution was heated to 65°C for 10 h.
Examination of the 31P{1H} NMR spectrum showed
formation of (C5Me5)Rh(PMe3)(h2-trans-stilbene) as
the major product. The solution was taken to dryness
1
and dissolved in C6D6. H and 31P{1H} NMR spectra
were identical to those of the final product obtained
from the reaction of 1 with PhCꢀCPh (6ide supra).
13C{1H} NMR (C6D6): l 150.00 (t, J=2 Hz), 147.37
(d, J=2.5 Hz), 130.58 (s), 129.24 (s), 128.51 (s),
127.62(s), 123.88 (s), 123.07 (s), 96.43 (t, J=3.8 Hz),
45.98 (dd, J=14, 3 Hz), 44.03 (dd, J=16, 2 Hz), 18.09
(d, J=26 Hz), 10.23 (s).
5.5. Reaction of (C5Me5)Rh(PMe3)(Ph)H with
cis-stilbene
5.3. Reaction of 1 with diphenylacetylene
20 mg (C5Me5)Rh(PMe3)(C2H4) (0.058 mmol) was
dissolved in C6H6 and photolyzed using a u\345 nm
filter for 2 days. 31P{1H} NMR spectroscopy showed
95% conversion to phenyl hydride along with some
formation of (C5Me5)Rh(PMe3)2. The solution was
taken to dryness under vacuum and the yellow solid
dissolved in hexanes. 60 ml cis-stilbene (0.337 mmol)
was added via a syringe under nitrogen and the solution
was heated to 65°C for 4 h. A white crystalline precipi-
10 mg (C5Me5)Rh(PMe3)Cl2 (0.026 mmol) was con-
verted to the dihydride 1 using Red–Al and dissolved
in C6D6 in a resealable NMR tube. 47 mg PhCꢀCPh
(0.26 mmol) was added under nitrogen. The solution
was heated at 70°C for 6 days. The solution turned red.
1
Examination of the solution by H and 31P{1H} NMR
spectroscopy revealed two major products. (C5Me5)-

A.D. Selmeczy, W.D. Jones / Inorganica Chimica Acta 300–302 (2000) 138–150
147
tate was observed. Examination of the 31P{1H} NMR
spectrum showed formation of (C5Me5)Rh(PMe3)(h2-
trans-stilbene) as the major product. The solution was
taken to dryness and dissolved in C6D6. 1H NMR
spectroscopy revealed that all of the cis-stilbene had
been converted to trans-stilbene.
the tube while under a stream of nitrogen. The tube was
warmed to −78°C in a dry ice/acetone bath and
swirled gently to mix the components. The sample was
quickly transferred to the pre-cooled (−40°C) NMR
probe and monitored by 31P{1H} NMR spectroscopy as
the probe was warmed in 10° (C) increments. Four new
products were observed by 1H and 31P{1H} NMR
spectroscopy. (C5Me5)Rh(PMe3)[trans-C(R)CH(R)]H
1H NMR (d8-toluene, 235 K): l 3.464 (s, 3H), 3.368 (s,
3H), 1.767 (s, 15H), 1.040 (d, J=10.4 Hz, 9H), −
11.467 (dd, J=47.6, 23.2 Hz, 1H). 31P{1H} NMR
(d8-toluene, 235 K): l 11.103 (d, J=148.3 Hz).
(C5Me5)Rh(PMe3)[cis-C(R)CH(R)]H 1H NMR (d8-
toluene, 235 K): l 1.623 (s, 15H), 1.167 (d, J=10.3 Hz,
9H), −13.744 (dd, J=48.2, 29.3 Hz, 1H). 31P{1H}
NMR (d8-toluene, 235 K): l 8.336 (d, J=145.4 Hz).
(C5Me5)Rh(PMe3)(h2-maleate) 1H NMR (d8-toluene,
235 K): l 1.875 (s, 15H), 0.528 (d, J=8.3 Hz, 9H).
31P{1H} NMR (d8-toluene, 235 K): l 3.267 (d, J=
180.9 Hz). (C5Me5)Rh(PMe3)(h2-fumarate) 1H NMR
(d8-toluene, 235 K): l 1.633 (s, 15H), 1.091 (d, J=10.4
Hz, 9H). 31P{1H} NMR (d8-toluene, 235 K): l 1.440 (d,
J=189.7 Hz).
5.6. Reaction of 1 with methyl propiolate
10 mg (C5Me5)Rh(PMe3)Cl2 (0.026 mmol) was con-
verted to the dihydride 1 using Red–Al. The tan oil
was dissolved in C6D6 and placed in a resealable NMR
tube. 11.5 ml HCꢀCE (0.129 mmol) was added under
nitrogen atmosphere. The solution became red after
several minutes. Examination of the solution by
NMR spectroscopy revealed two products. (C5Me5)-
Rh(PMe3)(p2-methylacrylate) Rotamer 1. 1H NMR
(C6D6): l 3.624 (s, 3H), 2.58 (m, 1Hvinylic), 2.420 (dd,
J=9.9, 1.7 Hz, 1Hvinylic), 1.925 (d, J=8 Hz, 1Hvinylic),
1.806 (d, J=2.3 Hz, 15H), 0.683 (d, J=8.6 Hz, 9H).
31P{1H} NMR (C6D6): l −0.924 (d, J=192 Hz).
13C{1H} NMR (C6D6): l 176.35 (s), 96.13 (t, J=4 Hz),
50.03 (s), 35.59 (d, J=16 Hz), 28.79 (dd, J=16, 3 Hz),
1
16.16 (d, J=26 Hz), 10.10 (s). Rotamer 2. H NMR
(C6D6): l 3.483 (s, 3H), 2.60 (m, 1Hvinylic), 2.49 (m,
1Hvinylic), 1.92 (m, 1Hvinylic), 1.699 (d, J=1.5 Hz, 15H),
1.125 (d, J=9 Hz, 9H). 31P{1H} NMR (C6D6): l
−5.510 (d, J=199 Hz). 13C{1H} NMR (C6D6): l
177.12 (s), 95.92 (t, J=3 Hz), 50.30 (s), 36.93 (d. J=16
Hz), 32.87 (d, J=16 Hz), 17.86 (d, J=25 Hz), 10.10
(s).
5.9. Chromatographic separation of (C5Me5)Rh-
(PMe3)(p2-maleate) and (C5Me5)Rh(PMe3)(p2-fumarate)
complexes
100 mg (C5Me5)Rh(PMe3)Cl2 (0.260 mmol) was con-
verted to the dihydride 1 using Red–Al. The tan solid
was dissolved in benzene and 100 ml DMAD added
slowly under a nitrogen atmosphere. The mixture was
allowed to stir for 15 min. The benzene was removed
under vacuum leaving a dark red oil. The oil was
dissolved in THF and chromatographed on thick layer
silica plates using a 1:1 benzene/THF mixture. Two
bands could be observed. The leading yellow–orange
band was determined to be the fumarate complex
(C5Me5)Rh(PMe3)[trans-CH(R)CH(R)], and the second
yellow band was determined to be the maleate complex
(C5Me5)Rh(PMe3)[cis-CH(R)CH(R)] by NMR spec-
troscopy. (C5Me5)Rh(PMe3)(p2-fumarate) 1H NMR
(C6D6): l 3.509 (s, 3H), 3.413 (s, 3H), 3.478 (ddd,
J=11.3, 9.5, 2 Hz, 1Hvinylic), 3.356 (dd, J=11.3, 2 Hz,
1Hvinylic), 1.696 (d, J=2.5 Hz, 15H), 1.061 (d, J=9.2
Hz, 9H). 31P{1H} NMR (C6D6): l −4.087 (d, J=191
Hz). 13C{1H} NMR (C6D6): l 176.51 (s), 175.11 (s),
96.49 (t, J=4 Hz), 49.98 (s), 49.77 (s), 39.20 (dd,
J=16, 4 Hz), 36.50 (d, J=16 Hz), 16.94 (d, J=26
5.7. Kinetic studies of 1 with methyl propiolate
Three solutions of 1 in 0.84 ml C6D6 (0.020 M in 1)
were freeze–pump–thaw degassed three times. Three
stock solutions containing the following were prepared:
(i) 0.0875 g ECꢀCH in 5 ml C6D6 ([ECꢀCH]=0.208
M), (ii) 0.0991 g a-phenyl-N-tert-butyl nitrone (PBN)
in 2 ml of solution (i) above ([PBN]=0.280 M), (iii)
0.0876 g ECꢀCH, 0.0168 g tert-butyl peroxide (TBP) in
5 ml C6D6 ([TBP]=0.023 M). All three stock solutions
were thoroughly degassed. For each kinetic experiment,
250 ml of the appropriate stock solution was injected
into a C6D6 solution of 1 under a nitrogen atmosphere
1
and room temperature H NMR spectra immediately
recorded at 5 min intervals for at least 5 half-lives.
5.8. Reaction of 1 with dimethyl acetylenedicarboxylate
1
20 mg (C5Me5)Rh(PMe3)Cl2 (0.052 mmol) was con-
verted to the dihydride 1 using PPN+BH4−. The tan
oil was dissolved in d8-toluene in a resealable NMR
tube and freeze–pump–thaw degassed three times. The
tube was frozen in liquid nitrogen and 32 ml dimethyl
acetylenedicarboxylate (0.26 mmol) was injected into
Hz), 9.25 (s). (C5Me5)Rh(PMe3)(p2-maleate) H NMR
(C6D6): l 3.631 (s, 6H), 2.404 (dd, J=8.8, 2 Hz,
2Hvinylic), 1.766 (d, J=2.6 Hz, 15H), 0.600 (d, J=8.6
Hz, 9H). 31P{1H} NMR (C6D6): l −1.959 (d, J=182
Hz). 13C{1H} NMR (C6D6): l 49.87 (s), 34.92 (dd,
J=16, 2 Hz), 14.86 (d, J=25 Hz), 13.94 (s).

148
A.D. Selmeczy, W.D. Jones / Inorganica Chimica Acta 300–302 (2000) 138–150
5.10. Crosso6er experiment between d2-(C5Me5)-
Rh(PMe3)(p2-fumarate) and free dimethyl fumarate
9H), −13.120 (dd, J=46, 27 Hz, 1H). 31P{1H} NMR
(C6D6): l 3.523 (d, J=146 Hz). 19F NMR (C6D6): l
20.765 (br s), 19.260 (d, J=8.7 Hz). Rotamer 2. H
1
A sample of (C5Me5)Rh(PMe3)D2 was prepared by
reacting 9 mg (C5Me5)Rh(PMe3)Br2 (0.019 mmol) with
PPN+BD4−. The dideuteride 1-d2 was dissolved in
d8-toluene and 11.2 mg dimethyl fumarate (0.0778
mmol) added under a nitrogen atmosphere. The solu-
tion was placed in an NMR tube capped with a rubber
septum and cooled to −20°C in the NMR probe.
31P{1H} NMR spectra showed no reaction had oc-
curred. The sample was removed from the probe and
4.7 ml DMAD (0.038 mmol) injected through the sep-
tum. The tube was inverted three times and quickly
returned to the NMR probe. 31P{1H} NMR spectra
revealed that no crossover between the d2-(C5Me5)
Rh(PMe3)(h2-fumarate) complex and free dimethyl fu-
marate had taken place.
NMR (C6D6): l 7.11 (m, 1Hvinylic), 1.578 (d, J=2 Hz,
15H), 0.807 (d, J=9 Hz, 9H), −13.950 (dd, J=49, 28
Hz, 1H). 31P{1H} NMR (C6D6): l 2.358 (d, J=140
Hz). 19F NMR (C6D6): l 17.134 (t, J=8.7 Hz), 15.278
(br s). After heating the above mixture at 100°C for 5 h
a new product is formed quantitatively. (C5Me5)-
Rh(PMe3)(p2-trans-hexafluoro-2-butene). 1H NMR
(C6D6): l 2.439 (m, 2Hvinylic), 1.614 (d, J=2 Hz, 15H),
0.816 (d, J=9 Hz, 9H). 31P{1H} NMR (C6D6): l
−5.480 (d, J=187 Hz). 19F NMR (C6D6): l 20.156
(dt, J=9.6, 3 Hz), 17.513 (d, J=9.6 Hz).
5.13. Electrochemistry of (C5Me5)Rh(PMe3)H2
and substituted alkynes
Cyclic voltammetric measurements were carried out
on a Princeton Applied Research 173 potentiostat and
PAR 175 sweep generator and a Houston 2000 XY
recorder. A single-compartment, three-electrode cell
containing a glassy-carbon (3.2 mm diameter) working
electrode, a platinum wire auxiliary electrode, and a
silver wire reference electrode was used for all measure-
ments. In a typical experiment, several mg of the alkyne
5.11. Crosso6er experiment between (C5Me5)-
Rh(PMe3)H2/(C5Me5)Rh(PMe3)D2 and DMAD
A sample of (C5Me5)Rh(PMe3)D2 was prepared by
reacting 9 mg (C5Me5)Rh(PMe3)Br2 (0.019 mmol) with
PPN+BD4−. A similar sample of (C5Me5)Rh(PMe3)H2
was prepared by reacting 9 mg (C5Me5)Rh(PMe3)Br2
(0.019 mmol) with PPN+BH4−. The two materials
were dissolved in d8-toluene and mixed. 31P{1H} NMR
spectra showed no scrambling had occurred. The solu-
tion was placed in an NMR tube capped with a rubber
septum and cooled to −20°C in the NMR probe. The
sample was removed from the probe and 4.7 ml DMAD
(0.038 mmol) injected through the septum. The tube
was inverted three times and quickly returned to the
NMR probe. 31P{1H} NMR spectra revealed formation
of 2i, 2i-d2, 2j, and 2j-d2 with no evidence for d1
materials.
or of 1 with 200 mg TBA+PF6 as the supporting
−
electrolyte (0.11 M) was placed in the cell and purged
with nitrogen for 5 min. Dry, degassed THF (5 ml) was
introduced and cyclic voltammograms recorded imme-
diately. For the measurement of E1/2 for F3CCꢀCCF3,
the gas cylinder was attached to a syringe needle and
the alkyne slowly bubbled into the degassed THF/
TBA+PF6 solution for 30 s.
−
5.14. X-ray structural determination of (C5Me5)-
Rh(PMe3)[trans-CH(CF3)CH(CF3)]Cl
5.12. Reaction of 1 with hexafluoro-2-butyne
A small yellow crystal of the complex was mounted
on a glass fiber with epoxy and placed in the cold
stream (−20°C) of the diffractometer. 25 reflections
with values of x between 5 and 70° were centered and
used for cell determination. Data were collected in a
primitive monoclinic crystal system, and data reduction
showed absences consistent with space group P21/c.
Solution of the Patterson map allowed placement of the
rhodium atom, and use of the program DIRDIF allowed
location of all remaining atoms. An absorption correc-
tion was applied following isotropic refinement with the
program DIFABS. In the final model, hydrogens were
placed in idealized positions and all non-hydrogen
atoms were refined anisotropically with the exception of
C16, which tended towards non-positive definite refine-
ment. Data collection and refinement parameters are
given in Table 2.
20 mg (C5Me5)Rh(PMe3)Cl2 (0.052 mmol) was con-
verted to the dihydride 1 using Red–Al and dissolved
in THF in a resealable NMR tube. The solution was
subjected to three freeze–pump–thaw degas cycles.
ꢀ100 torr F3CCꢀCCF3 (3.21 mmol) was introduced
into the high-vacuum line (approx. volume 600 ml) and
condensed into the frozen resealable NMR tube. Upon
warming to room temperature, the solution turned
yellow and thickened to a viscous fluid. Removal of the
volatile components under high vacuum left a brown
solid which was extracted with C6D6 and filtered
through a cotton-plugged pipette, yielding a bright
yellow solution. NMR spectroscopy showed formation
of two products. (C5Me5)Rh(PMe3)[trans-C(CF3)CH-
(CF3)]H. Rotamer 1. 1H NMR (C6D6): l 7.11 (m,
1Hvinylic), 1.662 (d, J=2 Hz, 15H), 0.892 (d, J=10 Hz,

A.D. Selmeczy, W.D. Jones / Inorganica Chimica Acta 300–302 (2000) 138–150
149
Table 2
[2] (a) W. Keim, A. Behr, M. Ro¨per, Comprehensive
Organometallic Chemistry, Vol. 8, Pergamon, Oxford, 1982, p.
371. (b) J.A. Ewen, J. Am. Chem. Soc. 106 (1984) 6355. (c)
P.L. Watson, J. Am. Chem. Soc. 104 (1982) 337. (d) H.W.
Turner, R.R. Schrock, J. Am. Chem. Soc. 105 (1983) 4942. (e)
L. Clawson, J. Soto, S.L. Buchwald, M.L. Steigerwald, R.H.
Grubbs, J. Am. Chem. Soc. 107 (1985) 3377. (f) P. Cossee,
Tetrahedron Lett. 17 (1960) 12. (g) K.J. Ivin, J.J. Rooney,
C.D. Stewart, M.L.H. Green, J.R. Mahtab, J. Chem. Soc.
Chem. Commun. (1978) 604.
[3] (a) H.C. Clark, C.S. Wong, J. Organomet. Chem. 92 (1975)
C31. (b) M.R. Torres, A. Vegas, A. Santos, J.J. Ros, J.
Organomet. Chem. 309 (1986) 169. (c) J. Lopez, A. Romero,
A. Santos, A. Vegas, A.M. Echavarren, P.J. Noheda, J.
Organomet. Chem. 373 (1989) 249. (d) A.M. Stolzenberg, E.L.
Mutterties, Organometallics 4 (1985) 1739. (e) B. Longato, S.
Bresadola, Inorg. Chem. 21 (1982) 168. (f) R.H. Crabtree, J.
Chem. Soc. Chem. Commun. (1975) 648. (g) M. Dubeck, R.A.
Schell, Inorg. Chem. 3 (1964) 1757.
[4] (a) J. Amaudrut, J. LeBlanc, C. Moise, J. Sala-Pala, J.
Organomet. Chem. 295 (1985) 167. (b) G.E. Herberich, I.
Savvopoulos, J. Organomet. Chem. 362 (1989) 345. (c) M.M.
Kubicki, R. Kergoat, M. Cariou, J.E. Guerchais, P. L’Hari-
don, J. Organomet. Chem. 322 (1987) 357.
[5] (a) A. Nakamura, S. Otsuka, J. Am. Chem. Soc. 94 (1972)
1886. (b) B.L. Booth, A.D. Lloyd, J. Organomet. Chem. 35
(1972) 195. (c) H.C. Clark, G. Ferguson, A.B. Goel, E.G.
Janzen, H. Ruegger, P.Y. Siew, C.S. Wong, J. Am. Chem.
Soc. 108 (1986) 6961.
[6] (a) G.E. Herberich, H. Mayer, J. Organomet. Chem. 347
(1988) 93. (b) D.L. Reger, S.A. Klaeren, J.E. Babin, R.D.
Adams, Organometallics 7 (1988) 181. (c) R.G. Bergman, J.M.
Huggins, J. Am. Chem. Soc. 103 (1981) 3002.
[7] G.E. Herberich, W. Barlage, Organometallics 6 (1987) 1924.
[8] (a) M. Hackett, G.M. Whitesides, J. Am. Chem. Soc. 110
(1988) 1449. (b) H. Werner, M.A. Esteruelas, H. Otto,
Organometallics 5 (1986) 2295.
[9] (a) M. Cariou, M. Etienne, J.E. Guerchais, R. Kergoat, M.M.
Kubicki, J. Organomet. Chem. 327 (1987) 393. (b) C. Bian-
chini, P. Innocenti, D. Masi, A. Meli, M. Sabat, Organometal-
lics 5 (1986) 72. (c) M.K. Whittlesey, Ph.D. Thesis, University
of York, Heslington, UK, 1991. (d) H. Scordia, R. Kergoat,
M.M. Kubicki, J.E. Guerchais, P. L’Haridon, Organometallics
2 (1983) 1681.
Summary of crystallographic data for (C5Me5)Rh(PMe3)[trans-
CCF3ꢁCHCF3]Cl
Crystal parameters
Chemical formula
Formula weight
Crystal system
Space group (no.)
Z
RhPC17H25F6Cl
512.7072
P21/c
14
4
9.834(6)
12.527(7)
16.737(5)
91.53(3)
2061(3)
1.652
,
a (A)
,
b (A)
,
c (A)
i (°)
3
,
V (A )
zcalc (g cm−3
)
Crystal dimensions (mm)
Temperature (°C)
1.1×2.2×3.4
−20
Measurement of intensity data
Diffractometer
Radiation (monochrom.)
Scan type
Enraf–Nonius CAD4
,
Mo, 0.71073 A (graphite)
2q/ꢀ
Takeoff angle (°)
Total bkgd time
2.6
(scan time)/2
2–16.5
0.7+0.35 tan q
4–40
+h,+k,9l
3607
1850
235
10.722
h0l, l odd; 0k0, k odd
differential
0.78–1.18
Scan rate (° min−1
Scan range (°)
2q Range (°)
)
Data collected
Number of data collected
Number of unique data F2\3|(F2)
Number of parameters varied
v (cm−1
)
Systematic absences
Absorption correction
Range of transmission factors
Equiv. data
Agreement between equiv. data (Fo)
R(Fo)
(
0kl=0k1
0.0467
0.0568
0.0590
1.779
Rw(Fo)
Goodness-of-fit
6. Supplementary material
[10] H. Scordia, R. Kergoat, M.M. Kubicki, J.E. Guerchais, J.
Organomet. Chem. 249 (1983) 371. (b) A. Romero, A. Santos,
A. Vegas, Organometallics 7 (1988) 1988. (c) M.R. Torres, A.
Santos, J. Ros, X. Solans, Organometallics 6 (1986) 1091. (d)
E. Negishi, T. Takahashi, S. Baba, D.E. Van Horn, N.
Okukada, J. Am. Chem. Soc. 109 (1987) 2393.
Supplementary material available includes tables of
crystallographic data, atomic coordinates, bond dis-
tances and angles, and anisotropic thermal parameters
for 4a (8 pages).
[11] B. Klingert, H. Werner, Chem. Ber. 116 (1983) 1450.
[12] L. Dong, Ph.D. Thesis, University of Rochester, Rochester,
NY, 1991.
Acknowledgements
[13] (a) J.F. Harris, R.J. Harder, G.N. Sausen, J. Org. Chem. 25
(1960) 633. (b) H.C. Brown, H.L. Gewanter, J. Org. Chem. 25
(1960) 2071.
Acknowledgment is given to the US Department of
Energy (FG02-86ER13569) for its support of this work.
[14] P.M. Treichel, E. Pitcher, F.G.A. Stone, Inorg. Chem.
1
(1962) 511; F.Y. Petillon, F. Le Floch-Perennou, J.E. Guer-
chais, D.W.A. Sharp, L. Manojlovic-Muir, K.W. Muir, J.
Organomet. Chem. 202 (1980) 23.
References
[15] W.D. Jones, V.L. Kuykendall, A.D. Selmeczy, Organometallics
10 (1991) 1577.
[16] W.D. Jones, A.D. Selmeczy, Organometallics 11 (1992) 889.
[1] (a) J.P. Collman, L.S. Hegedus, J.R. Norton, R.G. Finke,
Principles and Applications of Organotransition Metal Chem-
istry, University Science Books, Mill Valley, CA, 1987. (b) S.
Otsuka, A. Nakamura, Adv. Organomet. Chem. 14 (1976) 245.
[17] W.D. Jones, V.L. Chandler, F.J. Feher, Organometallics
(1990) 164.
9

150
A.D. Selmeczy, W.D. Jones / Inorganica Chimica Acta 300–302 (2000) 138–150
[18] S.W. Kirtley, M.A. Andrews, R. Bau, G.W. Grynkewich, T.J.
Marks, D.L. Tipton, B.L. Whittlesey, J. Am. Chem. Soc. 99
(1977) 7154.
[19] J.W. Kang, P.M. Maitlis, J. Organomet. Chem. 26 (1971) 383.
[20] H. Werner, B.J. Klingert, J. Organomet. Chem. 218 (1981) 395.
[21] (a) W.D. Jones, F.J. Feher, J. Am. Chem. Soc. 104 (1982) 4240.
(b) W.D. Jones, F.J. Feher, J. Am. Chem. Soc. 106 (1984) 1650.
(c) W.D. Jones, F.J. Feher, J. Am. Chem. Soc. 107 (1985) 620.
(d) W.D. Jones, F.J. Feher, Organometallics 2 (1983) 686. (e)
W.D. Jones, F.J. Feher, Inorg. Chem. 23 (1984) 2376.
.