Reaction of alkynes with Rh4(CO)12. A mid-infrared vibrational and kinetic study of (μ4-η2-alkyne) Rh4(CO)8(μ-CO)2

Article abstract of DOI:10.1021/om0510019

The pure component mid-infrared spectra of the butterfly clusters (μ₄-η²-alkyne)Rh₄(CO) ₈(μ-CO)₂, alkyne = 3-hexyne, 1-heptyne, 1-octyne, 4-octyne, 1-phenyl-1-hexyne, and 1-phenyl-1-butyne, were obtained from multicomponent solutions using band-target entropy minimization (BTEM). DFT was used to carry out full geometric optimization and mid-infrared vibrational prediction of (μ₄-η²-butyne)Rh₄-(CO) ₈(M-CO)₂ and (μ₄-η²-propyne) Rh₄(CO)₈(μ-CO)₂ as simple models for the terminal and symmetric alkyne clusters, respectively. The predicted spectra were in good agreement with the experimentally obtained deconvoluted pure component spectra of this class of complexes. The kinetics for the formation of the butterfly cluster (μ₄-η²-3-hexyne)Rh ₄(CO)₈(μ-CO)₂ from the reaction of Rh ₄(CO)₁₂ with 3-hexyne in n-hexane, i.e., Rh ₄(CO)₁₂ + 3-hexyne → (μ₄- η²-3-hexyne)Rh₄(CO)₈(μ-CO)₂ + 2CO, was also investigated. The rate of formation of the butterfly cluster was found to follow the rate expression, rate = k_obs[Rh ₄(CO)₁₂]¹[CO]^-1, with the apparent activation parameters ΔH^? = 123.6 ±11.0 kJ/ mol·K and ΔS^? = (7 ± 4) × 10 J·mol^-1·K^-1. A mechanism is proposed consistent with the observed kinetic rate expression. This involves a dissociation of one of the carbonyl ligands from Rh₄(CO)₁₂ prior to the alkyne coordination.

Full text of DOI:10.1021/om0510019

2182
Organometallics 2006, 25, 2182-2188
Reaction of Alkynes with Rh4(CO)12. A Mid-Infrared Vibrational
and Kinetic Study of (µ4-η2-alkyne)Rh4(CO)8(µ-CO)2
Ayman D. Allian, Martin Tjahjono, and Marc Garland*
Department of Chemical and Biomolecular Engineering, 4 Engineering DriVe 4,
National UniVersity of Singapore, Singapore 117576
ReceiVed NoVember 21, 2005
The pure component mid-infrared spectra of the butterfly clusters (µ4-η2-alkyne)Rh4(CO)8(µ-CO)2,
alkyne ) 3-hexyne, 1-heptyne, 1-octyne, 4-octyne, 1-phenyl-1-hexyne, and 1-phenyl-1-butyne, were
obtained from multicomponent solutions using band-target entropy minimization (BTEM). DFT was used
to carry out full geometric optimization and mid-infrared vibrational prediction of (µ4-η2-2-butyne)Rh4-
(CO)8(µ-CO)2 and (µ4-η2-propyne)Rh4(CO)8(µ-CO)2 as simple models for the terminal and symmetric
alkyne clusters, respectively. The predicted spectra were in good agreement with the experimentally
obtained deconvoluted pure component spectra of this class of complexes. The kinetics for the formation
of the butterfly cluster (µ4-η2-3-hexyne)Rh4(CO)8(µ-CO)2 from the reaction of Rh4(CO)12 with 3-hexyne
in n-hexane, i.e., Rh4(CO)12 + 3-hexyne f (µ4-η2-3-hexyne)Rh4(CO)8(µ-CO)2 + 2CO, was also
investigated. The rate of formation of the butterfly cluster was found to follow the rate expression, rate
) kobs[Rh4(CO)12]1[3-hexyne]1[CO]-1, with the apparent activation parameters ∆Hq ) 123.6 ( 11.0 kJ/
mol‚K and ∆Sq ) (7 ( 4) × 10 J‚mol-1‚K-1. A mechanism is proposed consistent with the observed
kinetic rate expression. This involves a dissociation of one of the carbonyl ligands from Rh4(CO)12 prior
to the alkyne coordination.
reaction has been shown to occur as both a stoichiometric metal-
mediated and catalytic reaction.9 Both intramolecular and
intermolecular Pauson-Khand reactions are used as the main
routes to generate high-value-added and biologically active
substituted cyclopentenones. Substituted cyclopentenones are
of considerable interest to the pharmaceutical industry, as they
often display high potency as antitumor and antibacterial
agents.10
Introduction
The reaction of alkynes with organometallic clusters has been
a subject of considerable experimental as well as theoretical
interest due to the wide variety of coordination modes possible.1-3
Thus the coordination of alkynes to organometallic clusters of
various nuclearities has attracted considerable attention. In
particular, the decrease in bond order of the coordinated alkynes,
and hence lengthening of the carbon-carbon bond, due to the
back-bonding from multiple metal centers has been an issue of
keen interest. Such studies have also provided an excellent
model for the understanding of the adsorption/coordination of
alkynes to metal surfaces.2-4
In addition to the interest in alkynes and metal clusters due
to their coordination chemistry alone, the synthetic utility of
stoichiometric metal-mediated and catalytic reactions involving
alkynes has been the focus of numerous other studies. For
example, in the absence of other reactants, alkyne trimerization
has been observed using a wide variety of metals such as CpCo-
(CO)2.5 Alkynes are also known to react with a second reagent,
i.e., alkenes, to give conjugated dienes.6
The rhodium carbonyl cluster Rh4(CO)9(µ-CO)3, first pre-
pared and characterized by Chini,11-13 has often been the group
9 cluster of choice for mechanistic and synthetic studies
involving tetranuclear clusters.14-17 Recent reports indicate that
18
Rh4(CO)12 and other rhodium-based organometallics19 can
mediate the Pauson-Khand reaction, even catalytically. How-
ever, understanding of the coordination chemistry of Rh4(CO)12
with alkynes is minimal. Diphenylacetylene was reported to react
with Rh4(CO)12, and the new species was tentatively assigned
to be (µ4-η2-C6H5CtCC6H5)Rh4(CO)8(µ-CO)2 based on its
20
mid-infrared spectral similarity to the butterfly cluster (µ4-η2-
(9) See for example: (a) Gibson, S. E.; Stevenazzi, A. Angew. Chem.,
Int. Ed. 2003, 42, 1800. (b) Blanco-Urgoiti, J.; Perez-Castells, J. Chem.
Soc. ReV. 2004, 33, 32. (c) Brummond, K. M.; Joseph, J. L. Tetrahedron
2000, 56, 3263. (d) Gibson, S. E.; Mainolfi, N. Angew. Chem., Int. Ed.
2005, 44, 3022.
(10) Gibson, S. E.; Lewis, S. E.; Mainolfi, N. J. Organomet. Chem. 2004,
689, 3873.
(11) Chini, P.; Martinengo, S. Chem. Commun. 1968, 5, 251.
(12) Chini, P.; Martinengo, S. Inorg. Chim. Acta 1969, 3, 315.
(13) Chini, P.; Heaton, B. T. Top. Curr. Chem. 1977, 71, 1.
(14) Cotton, F. A.; Kruczynski, L.; Lewis, J.; Shapiro, B. L.; Johnson,
L. F. J. Am. Chem. Soc. 1972, 94, 6191.
(15) Evans, J.; Johnson, B. F. G.; Lewis, J.; Norton, J. R.; Cotton, F. A.
J. Chem. Soc., Chem. Commun. 1973, 21, 807.
Perhaps most notably, alkynes react with both alkenes and
CO to give substituted cyclopentenones. The Pauson-Khand7,8
* To whom correspondence should be addressed. E-mail:
chemvg@nus.edu.sg. Phone: +65-6-874-6617. Fax: +65-6-779-1936.
(1) Jeannin, Y. Transition Met. Chem. 1993, 18, 122.
(2) Muetterties, E. L. Science 1977, 196, 839.
(3) Stanghellini, P. L.; Rossetti, R. Inorg. Chem. 1990, 29, 2047.
(4) Kesmodel, L. L.; Baetzold, R. C.; Somorjai, G. A. Surf. Sci. 1977,
66, 299.
(5) Su¨nkel, K. J. Organomet. Chem. 1990, 391, 247.
(6) See for example: (a) Trost, B. M.; Machacek, M.; Schnaderbeck,
M. J. Org. Lett. 2000, 2, 1761. (b) Herrmann, W. A.; Fischer, R. A.;
Herdtweck, E. Organometallics 1989, 8, 2821.
(7) Khand, I. U.; Knox, G. R.; Pauson, P. L.; Watts, W. E. J. Chem.
Soc., Chem. Commun. 1971, 36.
(16) Band, E.; Muetterties, E. L. Chem. ReV. 1978, 78, 8, 636.
(17) Roberts, Y. V.; Johnson, B. F.; Benfield, R. E. Inorg. Chim. Acta
1995, 229, 221.
(8) Khand I. U.; Knox, G. R.; Pauson, P. L.; Watts, W. E. J. Chem.
Soc., Perkin Trans. 1 1973, 975.
(18) Kobayashi, T.; Koga, Y.; Naraska, K. J. Organomet. Chem. 2001,
624, 73.
10.1021/om0510019 CCC: $33.50 © 2006 American Chemical Society
Publication on Web 03/21/2006

Reaction of Alkynes with Rh4(CO)12
Organometallics, Vol. 25, No. 9, 2006 2183
CHtCH)Co4(CO)8(µ-CO)2.21 Stanghellini22 later carried out a
detailed study of (µ4-η2-CHtCH)Co4(CO)8(µ-CO)2 where both
the mid-infrared assignment and the X-ray structure were
reported. Shortly thereafter, Horvath23 reported the X-ray single-
crystal structure of the mixed metal butterfly cluster (µ4-η2-
RC2R)Co2Rh2(CO)8(µ-CO)2, alkyne ) C6F5, Ph. In addition,
the X-ray single-crystal structure24 of (µ4-η2-PhC2Ph)Co3Rh-
(CO)8(µ-CO)2 and recently the structure of (µ-dmad)3CoRh3-
(CO)9, dmad ) (CH3O)COCCCO(OCH3), were reported.25
In the present contribution, understanding of the reactivity
and coordination chemistry of alkynes with Rh4(CO)12 is
considerably extended. The pure component mid-infrared spectra
of various butterfly clusters formed by reacting Rh4(CO)12 with
symmetric, terminal, asymmetric alkynes were obtained. DFT
calculations were performed on (µ4-η2-2-butyne)Rh4(CO)8(µ-
CO)2 and (µ4-η2-propyne)Rh4(CO)8(µ-CO)2 as simple models
for the symmetric and terminal alkyne clusters. The DFT-
optimized geometries yielded predicted infrared spectra that
were very consistent with the experimentally observed spectra.
The DFT-optimized geometries for the butterfly clusters are
reported. The kinetics of the reaction of Rh4(CO)12 with
3-hexyne in n-hexane solutions was also investigated in detail
using in-situ infrared spectroscopy.
Figure 1. DFT-predicted spectrum of (µ4-η2-2-butyne)Rh4(CO)8-
(µ-CO)2 using (a) GaussView03 and (b) Molden43 and the (c) high-
resolution experimental spectrum of (µ4-η2-3-hexyne)Rh4(CO)8(µ-
CO)2.
overlap of two vibrational modes in the experimental spectrum
at 2027 and 2036 cm-1 is clearly predicted by DFT at 1998
and 2006 cm-1
.
Table 1 indicates that the current density functional method
overestimated the wavenumber of the bridging carbonyls in (µ4-
η2-symmetric alkyne)Rh4(CO)8(µ-CO)2 by ca. 0.3% while
underestimating the wavenumber of the terminal metal carbonyl
by ca. 1.3%. The DFT and experimental wavenumbers for (µ4-
η2-symmetric alkyne)Rh4(CO)8(µ-CO)2 are compared to the
complex (µ4-η2-PhC2Ph)Rh4(CO)8(µ-CO)2 reported by Iwashita
and Tamura20 and the complex (µ4-η2-HC2H)Co4(CO)8(µ-CO)2
reported by Stangellini.22
The optimized structure of (µ4-η2-2-butyne) Rh4(CO)8(µ-CO)2
is shown in Figure 2. The rhodium skeleton appears as a
butterfly with Rh(3) and Rh(4) occupying the wingtip positions
and Rh(1) and Rh (2) forming the body. There are two
inequivalent bridging carbonyls and eight terminal carbonyls.
The alkyne group, C(6)-C(7), is not exactly parallel to the Rh-
(1)-Rh(2) bond, in agreement with the reported X-ray of (µ4-
η2-HC2H)Co4(CO)8(µ-CO)222 and (µ4-η2-RC2H)Co2Rh2(CO)6(µ-
CO)4, R ) FeCp2.28 The structure has low symmetry (approxi-
mate C2 symmetry). The Rh(2)-Rh(3) predicted bond length
is 2.81 Å, while the Rh(2)-Rh(4) is 2.71 Å. The predicted
dihedral angle between the M(1)-M(2)-M(3) and M(1)-
M(4)-M(3) planes is 120.8°, consistent with the reported
experimental values of 116° for (µ4-η2-HC2H)Co4(CO)8(µ-
CO)222 and 116.9° for (µ4-η2-RC2R)Co2Rh2(CO)8(µ-CO)2, R )
C6F5.23 It should be mentioned that in the two reported mixed
Co2Rh2 alkyne clusters, two CO ligands are semibridged,
whereas in neither the X-ray structure of (µ4-η2-HC2H)Co4-
(CO)8(µ-CO)2 nor the DFT-predicted geometry of (µ4-η2-
symmetric alkyne)Rh4(CO)8(µ-CO)2 are there indications for
semibridged behavior.
Results and Discussion
High-Resolution Vibrational Study of (µ4-η2-3-hexyne)-
Rh4(CO)8(µ-CO)2. The reaction of 40 mg of Rh4(CO)12 with
0.95 equiv of 3-hexyne under atmospheric argon, in 100 mL of
n-hexane as a solvent, was monitored via in-situ FTIR spec-
troscopy using 1 cm-1 resolution. The collected reaction spectra
were deconvoluted using band target entropy minimization
(BTEM).26,27 Besides the pure component spectra of the starting
materials, Rh4(CO)9(µ-CO)3 and 3-hexyne, a new mid-infrared
spectrum was obtained, Figure 1c. The band positions and
intensity are strikingly similar to the infrared spectrum of the
species reported to be (µ4-η2-PhC2Ph)Rh4(CO)8(µ-CO)2 and
20
species (µ4-η2-HC2H)Co4(CO)8(µ-CO)2, which was confirmed
by X-ray.22
Density functional theory (DFT) was used to carry out full
geometric optimization and mid-infrared vibrational prediction
of the slightly simpler species (µ4-η2-2-butyne) Rh4(CO)8(µ-
CO)2 (see Experimental Section for further details). Figure 1
shows that the experimental spectrum and the DFT-predicted
spectrum are in good agreement. The vibrational patterns are
similar, and the relative intensities of the bands are similar. The
(19) See for example: (a) Schmid, T. M.; Consiglio, G. Chem. Commun.
2004, 20, 2318. (b) Jeong, N.; Sung, B. K.; Choi, Y. K. J. Am. Chem. Soc.
2000, 122, 6771. (c) Shibata, T.; Toshida, N.; Takagi, K. Org. Lett. 2002,
4, 1619. (d) Suh, W. H.; Choi, M.; Lee, S. I.; Chung, Y. K. Synthesis 2003,
2169. (e) Jeong, N.; Lee, S.; Byung, B. K. Organometallics 1998, 17, 3642.
(f) Koga, Y.; Kobayashi, T.; Narasaka, K. Chem. Lett. 1998, 249. (g)
Kobayashi, P. T.; Koga, Y.; Naraska, K. J. Organomet. Chem. 2001, 624,
73. (h) Jeong, N.; Sung, B. K.; Kim, J. S.; Park, S. B.; Seo, S. D.; Shin, J.
Y.; In, K. Y.; Choi, Y. K. Pure Appl. Chem. 2002, 74, 85.
(20) Iwashita, Y.; Tamura, F. Bull. Chem. Soc. Jpn. 1970, 43, 1517.
(21) Kruerk, U.; Hubel, W. Chem. Ber. 1961, 95, 2829.
(22) Gervasio, G.; Rossetti, R.; Stanghellini, P. L. Organometallics 1985,
4, 1612.
(23) Horvath, I. T.; Zsolnai, L.; Huttner, G. Organometallics 1986, 5,
180.
(24) Tunik, S. P.; Krym, V. R.; Starova, G. L.; Nikol’skii, A. B.;
Podkorytov, I. S.; Ooi, S.; Yamasaki, M.; Shiro, M. J. Organomet. Chem.
1997, 481, 83.
(25) Watson, W. H.; Poola, B.; Richmond, M. G. Organometallics 2005,
24, 4687.
The coordinated alkyne in Figure 2 is predicted to have a
bond length of ca. 1.42 Å, and the bond angle C17-C6-C7 is
124°. The reported bond lengths for the alkyne moieties in refs
22, 23, 24, and 28 were 1.399, 1.369, 1.42, and 1.419 Å,
respectively. These values can be compared to a bond length
of 1.22 Å predicted for free 2-butyne using the same DFT
method, PBE/DGDZVP. Overall, the coordinated alkyne in
Figure 2 has a ca. 1.5 bond order.2
(26) Chew, W.; Widjaja, E.; Garland, M. Organometallics 2002, 21,
1882.
(27) Widjaja, E.; Li, C.; Garland, M. Organometallics 2002, 21, 1991.
(28) Zhu, B.; Zhang, W.; Zhao, Q.; Bian, Z.; Hu, B.; Zhang, Y.; Yu-
Hua, Y.; Sun, J. J. Organomet. Chem. 2002, 650, 181.

2184 Organometallics, Vol. 25, No. 9, 2006
Allian et al.
Table 1. Mid-infrared Carbonyl Vibrational Spectra (cm-1) of (µ4-η2-symmetric alkyne)M4(CO)8(µ-CO)2 Butterfly Clusters
(M ) Rh, Co)
bridging
terminal
DFTa
1893
1895
1990
1991
1998
2027
2006
2036
2021
2047
2023
2050
2033
2061
2070
2053
2061
2091
2095
2093
3-hexynea
PhC2Phb
acetylenec
1889
1885
1879
2012.6
2020
2000
2035
2023
2055
2040
2043
a This study M ) Rh. b Ref 20 M ) Rh. c Ref 22 M ) Co.
Figure 2. DFT-optimized geometry of (µ4-η2-2-butyne)Rh4(CO)8-
(µ-CO)2 using PBE/DGDZVP.
Low-Resolution Vibrational Study of (µ4-η2-terminal
alkyne)Rh4(CO)8(µ-CO)2]. Using a similar approach, the
reaction of 40 mg of Rh4(CO)12 with 0.95 equiv of terminal
alkynes, namely, 1-heptyne or phenylacetylene, was monitored
via in-situ mid-infrared FTIR spectroscopy using 4 cm-1
resolution. The collected reaction spectra were deconvoluted
using BTEM. Again, new pure component spectra were
obtained. These are shown in Figure 3c,d along with the
vibrational predictions, Figure 3a,b, obtained for full geo-
metrically optimization of (µ4-η2-propyne)Rh4(CO)8(µ-CO)2.
Figure 3 shows that the DFT-predicted spectrum of (µ4-η2-
propyne)Rh4(CO)8(µ-CO)2 is in very good agreement with
experimentally observed deconvoluted spectra of (µ4-η2-1-
heptyne)Rh4(CO)8(µ-CO)2 and (µ4-η2-phenylacetylene)Rh4-
(CO)8(µ-CO)2. A more detailed comparison of the vibrational
spectra is provided in Table 2. The mid-infrared spectra of the
class (µ4-η2-terminal alkyne)Rh4(CO)8(µ-CO)2 are slightly but
noticeably distinct from those of the class (µ4-η2-symmetric
alkyne) Rh4(CO)8(µ-CO)2.
Figure 3. DFT-predicted spectrum of (µ4-η2-propyne)Rh4(CO)8-
(µ-CO)2 using (a) GaussView03 and (b) Molden43 and the low-
resolution experimental spectra of (c) (1-heptyne)Rh4(CO)8(µ-CO)2
and (d) (µ4-η2-phenylacetylene)Rh4(CO)8(µ-CO)2.
were on the order of hours. These reactions were much slower
than those for terminal alkynes with Rh4(CO)12. Therefore,
3-hexyne was selected for kinetic studies of the reaction shown
in eq 1. This reaction was studied on the interval of 0-0.95
equiv of alkyne to cluster.
The DFT-optimized structure of (µ4-η2-propyne)Rh4(CO)8-
(µ-CO)2 is shown in Figure 4. Overall, the structural parameters
are very close to those observed for the optimized structure of
(µ4-η2-2-butyne)Rh4(CO)8(µ-CO)2.
Rh4(CO)12 + 3-hexyne f
(µ4-η2-3-hexyne)Rh4(CO)8(µ-CO)2 + 2CO (1)
Low-Resolution Vibrational Study of (µ4-η2-asymmetric
alkyne)Rh4(CO)8(µ-CO)2. For completeness, the reactions of
40 mg of Rh4(CO)9(µ-CO)3 with 0.95 equiv each of 1-phenyl-
1-butyne and 1-pheny-1-hexyne were monitored using in-situ
mid-infrared FTIR spectroscopy using 4 cm-1 resolution. The
collected data were deconvoluted with BTEM, and again a
spectrum with a pattern similar to that observed earlier for
terminal and symmetric alkynes was observed. In addition, the
length of the aliphatic chain had little or no influence on the
observed band positions at 2092, 2064, 2048, 2028, 2013, and
It is rather important to note that the range of this reaction
was limited to only 0.95 equiv of alkyne to parent cluster in
the present study. Indeed, new very weak signals were observed
in the SVD of the present data sets, but these signals were not
deconvolutable by BTEM. The presence of these new weak
signals suggests other organometallics at very low concentra-
tions. These signals increase significantly in the vicinity of 1
equiv of alkyne to parent cluster, and thus these observations
suggest the further fragmentation of the butterfly cluster to
binuclear species. A similar observation was made by Horvath23
for (µ4-η2-PhC2Ph)Co2Rh2(CO)8(µ-CO)2 only after adding
1891 cm-1
.
Kinetics. Reaction of the Rh4(CO)9(µ-CO)3 with 3-Hexyne.
The typical reaction times for symmetric alkynes with Rh4(CO)12

Reaction of Alkynes with Rh4(CO)12
Organometallics, Vol. 25, No. 9, 2006 2185
Table 2. Mid-infrared Carbonyl Vibrational Spectra (cm-1) of (µ4-η2-terminal alkyne)Rh4(CO)8(µ-CO)2 Butterfly Clusters
bridging
terminal
DFT
1-heptyne
(Ph)C2H
1894
1897
1994
1996
2002
2030
2031
2012
2042
2044
2024
2027
2036
2063
2066
2065
2094
2095
1895
1897
2015
2018
2051
2054
excess alkynes. The nature and the structure of the species in
homometallic rhodium experiments will be the subject of further
study.
high-resolution spectra shown in Figure 1. Furthermore, the
spectrum of Rh6(CO)16 was observed, with bands at 2075 and
1819.6 cm-1. As Rh6(CO)16 is present at trace levels, its spectra
possessed few artifacts and had lower signal-to-noise ratio. The
spectrum of the newly identified all-terminal Rh4(CO)12, which
is present at only a 1% level, was not deconvolutable.29 No
other organometallic pure component spectra could be recon-
structed.
Relative Concentrations. Upon obtaining the pure compo-
nent spectra, which are not yet scaled to the correct extinction
coefficient, their relative concentration profiles (contributions
to overall signal) can be readily obtained as shown in Figure 8.
These relative concentration profiles are simply the contribution
of each pure component signal to the total experimentally
measured spectral signal. The relative concentration profiles of
the species in batch 3 are shown in Figure 8. The organometallic
species contribute ca. 20% of the experimental signal, while
the rest of the signal is coming from the background, moisture,
and the solvent. The maximum contribution from Rh6(CO)16,
which exists at trace levels, was 0.5% of the total signal. Figure
8 shows that a characteristic reaction time is ca. 6.5 h.
Real Concentrations. Using the relative concentrations and
knowing the initial moles of Rh4(CO)9(µ-CO) in each batch,
Spectroscopic Aspects. Twelve isothermal batch reactions
were conducted under 1 atm of pressure of Ar and/or CO using
hexane as a solvent. The concentrations and the operating
condition were design to allow mechanistic investigation (see
Experimental Section for details). All experiments were moni-
tored via on-line mid-infrared FTIR spectroscopy with a
resolution of 4 cm-1
.
As an example, the raw spectra of the carbonyl vibration
region ν(CO) at 1800-2200 cm-1 for batch number 3 are shown
in Figure 6. Upon adding 0.95 equiv of 3-hexyne (6 µL), the
decline of the Rh4(CO)9(µ-CO)3 band and the rise of new bands
was observed.
Pure Component Spectra. BTEM was used to deconvolute
the pure component spectra present. The rhodium organome-
tallics observed are shown in Figure 7. The pure component
spectrum of the starting material Rh4(CO)9(µ-CO)3 at 2073.8,
2069.6, 2043.2, and 1886.4 cm-1 was consistent with previous
studies and references therein.29 The vibrational spectrum of
(µ4-η2-C2H5CCC2H5)Rh4(CO)8(µ-CO)2 was reconstructed with
excellent high signal-to-noise and is in excellent agreement with
Figure 6. Time series of raw reaction spectra for the reaction of
3-hexyne with Rh4(CO)9(µ-CO)3.
Figure 4. DFT-optimized geometry of µ4-η2-(1-propyne)Rh4(CO)8-
(µ-CO)2 using PBE/DGDZVP.
Figure 5. Low-resolution experimental spectra of (a) (µ4-η2-phenyl-
1-hexyne)Rh4(CO)8(µ-CO)2 and (b) (µ4-η2-1-phenyl-1-butyne)Rh4-
(CO)8(µ-CO)2.
Figure 7. Pure component spectra obtained from BTEM analysis.

2186 Organometallics, Vol. 25, No. 9, 2006
Allian et al.
Figure 8. Relative metal carbonyl concentrations provided by
BTEM analysis.
Table 3. Estimated Values of kobs as a Function of
Temperature
temp (K)
288.15
293.15
298.15
k
obs (106 s-1)
1.4 ( 0.1
3.5 ( 0.2
8.1 ( 0.5
Figure 9. Mole fraction of (µ4-η2-C2H5CCC2H5)Rh4(CO)8(µ-CO)2
as a function of reaction time: (a) Rh4(CO)12 variation, (b) 3-hexyne
variation, and (c) CO partial pressure variation.
Table 4. Experimental Design for Kinetic Experiments
Rh4(CO)12
,
CO, Ar 3-hexyne, temperature,
bar bar
batch
experiment
mg
µL
K
Scheme 1. Proposed Mechanism A for Eq 1 Which Involves
the Dissociation of One of the Carbonyl Ligands from
Rh4(CO)9(µ-CO)3 Prior to the Alkyne Attack
1
2
3
4
5
6
7
8
9
3-hexyne variation
40.60
40.60
39.60
39.60
49.54
60.60
40.15
39.60
39.56
40.76
39.39
40.42
0.03 0.988
0.03 0.988
0.03 0.988
0.03 0.988
0.03 0.988
0.03 0.988
0.01 1.008
0.03 0.988
0.05 0.968
3
4.5
6
6
6
6
6
6
6
298.15
298.15
298.15
298.15
298.15
298.15
298.15
298.15
298.15
288.15
293.15
298.15
Rh4(CO)12 variation
CO variation
10 temperature variation
11
12
0
0
0
1.018
1.018
1.018
6
6
6
be classified as a dissociative mechanism31 where one of
carbonyl ligands dissociates prior to the coordination of the
ligand. DFT was performed, and the results suggest that apical
CO dissociation may be preferred over basal (axial and radial)
or bridging CO dissociation.
In Scheme 1, steps 1 and 2 are considered elementary. As
step 3 is assumed to be the rate-limiting step, the rate of product
formation can be presented as shown in eq 2.
the pure component can be scaled to the correct extension
coefficient, and thus the real concentrations of all organome-
tallics can be obtained. The arithmetic details are discussed in
detail elsewhere.30 Since quantitative information on 3-hexyne
is not available in the 1800-2200 cm-1 spectral region, the
time-dependent concentration of 3-hexyne was calculated using
the mass balance shown below.
d[(RC2R)Rh4(CO)10]
nhexyne,t ) nhexyne,t)0 - nbutterfly,t
rate )
) k3[(RC2R)Rh4(CO)11] (2)
dt
Kinetics and Mechanism of the Reaction. Nine kinetic
experiments were performed to determine the rate expression
for the formation of the butterfly cluster at 298 K. These
experiments and can be divided into three sets, Rh4(CO)12
variation, 3-hexyne variation, and CO variation, as shown in
Table 4 (see Experimental Section). In each set, one experi-
mental parameter was systematically varied while the remaining
variables were held essentially constant.
As the intermediates (RC2R)Rh4(CO)11 and Rh4(CO)11 were
not observed, their accumulation can be assumed negligible,
i.e., d[intermediate]/dt = 0. Using this steady-state approxima-
tion,32 eq 2 can be rewritten as shown in eq 3.
d[(RC2R)Rh4(CO)10]
k1[RC2R][Rh4(CO)12]
)
dt
(k-1/k2)(1 + k-2/k3)[CO] + [RC2R]
Figure 9 shows that the rate of formation of (µ4-η2-C2H5-
CCC2H5)Rh4(CO)8(µ-CO)2 increased with increasing Rh4(CO)12
and 3-hexyne. However, a drastic drop in the rate of its
formation was observed when the CO mole fraction was
increased, indicating that the coordination of the alkyne is
poisoned/retarded by the presence of CO. The proposed mech-
anism, see Scheme 1, for the alkyne substitution reaction can
(3)
With the assumption that k3 is the rate-limiting step, the term
k-2/k3 in the denominator is expected to be large, i.e., (k-1/k2)-
(1 + k-2/k3)[CO] . [RC2R]. Therefore, eq 3 can be further
simplified as shown in eq 4, where kobs equals k1/(k-1/k2)(1 +
k-2/k3).
(29) Allian, A. D.; Garland, M. J. Chem. Soc., Dalton Trans. 2005, 11,
1957.
(30) Widjaja, E.; Li, C.; Garland, M. J. Catal. 2004, 223, 278.
(31) Richens, D. T. Chem. ReV. 2005, 105, 1961.
(32) Masel, R. I. Chemical Kinetics and Catalysis; John Wiley & Sons:
New York, 2001.

Reaction of Alkynes with Rh4(CO)12
Organometallics, Vol. 25, No. 9, 2006 2187
d[(RC2R)Rh4(CO)10]
[RC2R][Rh4(CO)12]
[CO]
) kobs
(4)
dt
To test eq 4, the orders of Rh4(CO)9(µ-CO)3, 3-hexyne, and
CO were computed from the experimental data; that is, the
values of R, â, and γ in eq 5 were obtained by fitting
experimental concentration profiles using linear regression. The
numerical values of R, â, and γ were 1.0362 ( 0.0199, 1.1061
( 0.0232, and -1.0187 ( 0.0525, respectively.
Figure 10. Experimental setup for in situ FTIR measurements.
Legend: (1) argon tank; (2) argon purification column; (3) pressure
transducer; (4) jacketed continuous stirrer tank reactor (CSTR); (5)
hermetically sealed Teflon pump; (6) FTIR with high-pressure flow
through cell; (7) data acquisition.
rate ) kobs[Rh4(CO)12]R[alkyne]â[CO]γ
(5)
Although the results are in a good agreement with the
suggested mechanism, it should be mentioned that the first step
might instead involve transformation of the C3V Rh4(CO)9(µ-
CO)3 to another new structure, i.e., an open polyhedron, which
results from the cleavage of one Rh-Rh metal bond. Indeed,
there exist kinetic and mechanistic studies where open polyhe-
dron M4 clusters have been invoked in order to rationalize the
observations.33
Apparent Eyring Activation Parameters. To determine the
apparent activation parameters, four isothermal batch experi-
ments were carried out at temperatures of 288.15, 293.15, and
298.15 K. The rate of the formation of (µ4-η2-C2H5CCC2H5)-
Rh4(CO)8(µ-CO)2 increased with increasing temperature. The
observed rate constants, kobs, are shown in the Table 3.
The values of kobs were used to regress the values of the
enthalpy of activation ∆Hq and the entropy of activation ∆Sq
using eq 7. The obtained values were 123.6 ( 11.0 kJ‚mol-1
and 72.4 ( 37.3 J‚mol-1‚K-1, respectively.
hexyne, 1-phenyl-1-butyne (Aldrich, 99%), and the 1,6-enyne
(GFS Chemicals, 99%) were used directly without further purifica-
tion.
Equipment. All syntheses and kinetic runs were performed in a
closed recycle system. The experimental apparatus consisted of a
jacketed glass reactor (Aceglass) equipped with a magnetic stirrer,
a Teflon membrane pump (Cole-Parmer), and a Perkin-Elmer
System 2000 mid-infrared FTIR spectrometer. The fluid was
pumped under isobaric (argon/CO, atmospheric pressure) and
isothermal conditions (temp control Polyscience 9105, with tem-
perature stability ( 0.05 K) from the reactor through the pump,
then a high-pressure infrared cell with recycle back to reactor.
Connections for vacuum and argon were provided. A 2.000 piezo-
transducer (PAA-27W, Keller AG, Switzerland) was used through-
out for pressure measurements. The entire experimental setup was
gastight. To ensure quantitative working conditions, and hence
minimal decomposition of the moisture-, oxygen-, and light-
sensitive metal complexes, the system was thoroughly rinsed with
anhydrous solvent under argon and then evacuated to dryness prior
to each experiment. In addition, the reactor was fully covered with
aluminum foil, thus providing dark reaction conditions.34
Hamilton gastight 10 µL and 1.0 mL syringes were used for
introducing 3-hexyne solutions into the reactor through a rubber
septum. The high-pressure infrared cell was constructed at the ETH
Zu¨rich of SS316 steel and could be heated and cooled. The KBr
single-crystal windows used (Korth Monokristalle, Kiel, Germany)
had dimensions of diameter 40 mm by thickness 15 mm. Two sets
of Viton and silicone gaskets provided sealing, and Teflon spacers
were used between the windows. The construction of the flow-
through cell is a variation on the design by Noack (1968)35 and
differs in some respects from other high-pressure infrared cells.36
The high-pressure cell was situated in the Perkin-Elmer 2000 FTIR
infrared spectrometer. The cell chamber was purged with purified
nitrogen (Soxal, Singapore, 99.999%). The instrumental resolutions
used were 1 and 4 cm-1, and recorded spectra were obtained from
10 co-additions. The reported spectra from 1800-2200 cm-1 have
0.2 cm-1 data intervals. A schematic diagram of the experimental
setup is shown in Figure 10.
κT
h
-∆Hq ∆Sq
kobs
)
exp
+
(7)
(
)
RT
R
Conclusion
The addition of alkyne to Rh4(CO)12 results in the formation
of the butterfly clusters (µ4-η2-alkyne)Rh4(CO)8(µ-CO)2. The
mid-infrared spectra of various butterfly clusters (µ4-η2-alkyne)-
Rh4(CO)8(µ-CO)2 were obtained. Geometries for these new
nonisolatable complexes were obtained by matching the ex-
perimentally obtained vibrational pattern in the mid-infrared with
the predicted DFT vibrational patterns. The kinetics of the
formation of the butterfly cluster (µ4-η2-3-hexyne)Rh4(CO)8(µ-
CO)2 were also studied. The kinetics were well behaved and
provided quite clear reaction orders. The experimental kinetics
together with mechanistic considerations suggest that two
different mechanisms are consistent with the observations. These
involve, as a first step, either a CO dissociative mechanism or
the cleavage of a Rh-Rh bond to form a coordinately
unsaturated open polyhedron.
Kinetic Experiments. The kinetic study consisted of 12 iso-
thermal batch reactions. All of these batches were conducted in a
similar manner. Thus, only a description of the first isothermal
batch, conducted at 298.15 K and under 0.03 mbar CO, is provided
in detail.
Experimental Section
General Information. All solution preparations were carried out
under argon (Soxal, Singapore, 99.999%) using standard Schlenk
techniques.16 The solvents n-hexane (Fluka, puriss, 99.6%+) and
d-benzene (Cambridge-Isotope, MA, 99.5%) were refluxed over
sodium potassium alloy under argon. The argon was further purified
prior to use by passage through a column containing 100 g of
reduced BTS-catalyst (Fluka AG Buchs, Switzerland) and 100 g
of 4 Å molecular sieves to adsorb traces of oxygen and water,
respectively. The metal complex Rh4(CO)12 with stated purity of
at least 98% was obtained from Strem Chemicals (Newport, MA).
All alkynes, 3-hexyne, 1-heptyne,1-octyne, 4-octyne, 1-phenyl-1-
(33) Bor, G.; Dietler, U. K.; Pino, P. J. Organomet. Chem. 1978, 154,
301.
(34) Garland, M. In Mechanisms in Homogeneous Catalysis: A Spec-
troscopic Approach; Heaton, B., Ed.; Wiley-VCH: Weinheim, 2005;
Chapter 4, p 157.
(35) Noack, K. Spectrochim. Acta 1968, 24, 1917.
(36) Whyman, R. In Laboratory Methods in Vibrational Spectroscopy,
3rd ed.; Willis, H. A., van der Maas, J. H., Miller, R. G., Eds.; Wiley:
New York, 1987; Chapter 12.

2188 Organometallics, Vol. 25, No. 9, 2006
Allian et al.
Background spectra of the empty IR chamber were recorded,
and only then was the high-pressure cell loaded into the IR
compartment. Then, 40.60 mg of Rh4(CO)12 was added to 100 mL
of distilled hexane in a Schlenk tube and stirred for 20 min under
atmospheric argon. The reddish solution was then transferred to
the evacuated reactor, the stirrer was started, and argon was
introduced to a total pressure of 0.750 bar and allowed to equilibrate.
Subsequently, 30 mbar of CO was carefully added to the reactor,
and this was topped up with argon to 1.018 bar. The Teflon pump
was started to allow fluid circulation through the high-pressure IR
cell and back to the reactor. After equilibration of the entire system,
infrared spectra of the Rh4(CO)12/n-hexane solution in the high-
pressure cell were recorded. Reaction was initiated by injection of
6 µL of 3-hexyne dissolved in 1 mL of hexane through the rubber
septum injection port. Directly after the injection, spectra were
recorded every 5 min for ca. 14 h. The experimental design for all
12 experiments is shown in Table 4.
The solubility of CO in n-hexane was calculated using the
Henry’s constant reported in ref 37. Furthermore, due to the low
CO partial pressures used, and the fact that CO is evolved during
the reaction in eq 1, the partial pressure and mole fraction of
dissolved CO had to be continuously updated as a function of extent
of reaction. In this manner, proper kinetic analysis could be
performed.
Then, an equimolar amount of alkyne was added to each batch,
namely, 3-hexyne, 1-heptyne, 1-octyne, and 4-octyne. Upon inject-
ing the respective alkyne to each Rh4(CO)12 solution, it was quickly
transferred to a standard IR cell with a path length of 0.5 cm and
was placed in the IR compartment to monitor the formation of the
respective butterfly cluster. Undoubtedly, the inhomogeneity of the
solution due to the lack of stirring contributed to the nonlinearity
and the high noise level observed in the collected spectra.
Nevertheless, BTEM was successfully able to reconstruct the pure
component of the each formed butterfly cluster with decent signal-
to-noise ratio and minor artifacts.
BTEM was applied to the kinetic batch experiments in order to
get the following pure component spectra in n-hexane: (µ4-η2-3-
hexyne)Rh4(CO)8(µ-CO)2, Rh4(CO)12, and Rh6(CO)16.
DFT Calculations. Density functional theory was used to carry
out full geometric optimization and vibrational assignments of (µ4-
η2-2-butyne)Rh4(CO)8(µ-CO)2 and (µ4-η2-1-propyne)Rh4(CO)8(µ-
CO)2. The PBEPBE density functional with the DGDZVP basis
set was used, as this method successfully predicted the geometry
and the vibrational spectra of a variety of rhodium carbonyl
clusters.38 DFT-predicted spectra in Figures 1 and 3 were presented
in two formats. First is the direct output of the calculated vibrational
spectra, while the second is plotted using Molden43, where
Lorenzian distribution is used to represent the vibrational modes,
which can be related easily to the experimental data.
To confirm the validity of the approach taken (functional and
basis set used), the optimized geometry for (µ4-η2-HCCH)Co4(CO)8-
(µ-CO)2 and the mid-infrared spectra were predicted and compared
to the available literature experimental results.22 Very good agree-
ment is obtained. These results are archived in the Supporting
Information.
Attempts at Crystallization. Attempts to obtain crystals for
X-ray single-crystal analysis have thus far been unsuccessful.
Spectral Processing. Band-target entropy minimization was used
to obtain all the pure component experimental spectra reported in
this contribution, by deconvolution of the corresponding reactive
multicomponent solutions. Details of the mathematical constructs
of BTEM can be found elsewhere.26,27
Single-batch experiments were performed to obtain the BTEM
deconvolutions of the following pure component spectra in n-
hexane: (µ4-η2-3-hexyne)Rh4(CO)8(µ-CO)2 at 1 cm-1 resolution,
(µ4-η2-1-heptyne)Rh4(CO)8(µ-CO)2, (µ4-η2-phenylacetylene)Rh4-
(CO)8(µ-CO)2, (µ4-η2-1-phenyl-1-butyne)Rh4(CO)8(µ-CO)2, and (µ4-
η2-1-phenyl-1-hexyne)Rh4(CO)8(µ-CO)2.
Acknowledgment. The authors wish to thank Prof. Mark
Saeys for helpful discussions.
Note Added after ASAP Publication. In the version of
this paper published on the Web March 21, 2006, two different
equations were both labeled as eq 2. The version that now
appears is correct.
The pure component spectra of (µ4-η2-1-octyne)Rh4(CO)8(µ-
CO)2, (µ4-η2-4-octyne)Rh4(CO)8(µ-CO)2, (µ4-η2-1-heptyne)Rh4-
(CO)8(µ-CO)2, and (µ4-η2-3-hexyne)Rh4(CO)8(µ-CO)2 in d-benzene
were obtained using the following procedure. Four batches of ca.
3 mg of Rh4(CO)12 dissolved in 0.5 mL of d-benzene were prepared.
Supporting Information Available: This material is available
free of charge via the Internet at http://pubs.acs.org.
OM0510019
(37) Garland, M.; Horvath, I.; Bor, G.; Pino, P. Organometallics 1991,
10, 559.
(38) Allian, A. D.; Wang, Y.; Saeys, M.; Kuramshina, G. M.; Garland,
M. Vib. Spectrosc., in press.