Dimerization of alkynes promoted by a pincer-ligated iridium complex. C-C reductive elimination inhibited by steric crowding

Article abstract of DOI:10.1021/ja0647194

The pincer-ligated species (PCP)Ir (PCP = κ³-C ₆H₃-2,6-(CH₂P^tBu₂) ₂) is found to promote dimerization of phenylacetylene to give the enyne complex (PCP)Ir(trans-1,4-phenyl-but-3-ene-1-yne). The mechanism of this reaction is found to proceed through three steps: (i) addition of the alkynyl C-H bond to iridium, (ii) insertion of a second phenylacetylene molecule into the resulting Ir-H bond, and (iii) vinylacetylide reductive elimination. Each of these steps has been investigated, by both experimental and computational (DFT) methods, to yield unexpected conclusions of general interest. (i) The product of alkynyl C-H addition, (PCP)Ir(CCPh)(H) (3), has been isolated and, in accord with experimental observations, is calculated to be 29 kcal/mol more stable than the analogous product of benzene C-H addition. (ii) Insertion of a second PhCCH molecule into the Ir-H bond of 3 proceeds rapidly, but with a 1,2-orientation. This orientation gives (PCP)Ir(CCPh)(CPh=CH₂) (4) which would yield the 1,3-diphenyl-enyne if it were to undergo C-C elimination; however, the insertion is reversible, which represents the first example, to our knowledge, of simple β-H elimination from a vinyl group to give a terminal hydride. The 2,1-insertion product (PCP)Ir(CCPh)(CH=CHPh) (6) forms more slowly, but unlike the 1,2 insertion product it undergoes C-C elimination to give the observed enyne. (iii) The failure of 4 to undergo C-C elimination is found to be general for (PCP)Ir(CCPh)(vinyl) complexes in which the vinyl group has an α-substituent. Thus, although C-C elimination relieves crowding, the reaction is inhibited by increased crowding. Density-functional theory (DFT) calculations support this surprising conclusion and offer a clear explanation. Alkynyl-vinyl bond formation in the C-C elimination transition state involves the vinyl group π-system; this requires that the vinyl group must rotate (around the Ir-C bond) by ca. 90° to achieve an appropriate orientation. This rotation is severely inhibited by steric crowding, particularly when the vinyl group bears an α-substituent.

Full text of DOI:10.1021/ja0647194

Published on Web 01/04/2007
Dimerization of Alkynes Promoted by a Pincer-Ligated Iridium
Complex. C-C Reductive Elimination Inhibited by Steric
Crowding
Rajshekhar Ghosh, Xiawei Zhang, Patrick Achord, Thomas J. Emge,
Karsten Krogh-Jespersen,* and Alan S. Goldman*
Contribution from the Department of Chemistry and Chemical Biology, Rutgers,
the State UniVersity of New Jersey, Piscataway, New Jersey 08854
Received July 16, 2006; E-mail: alan.goldman@rutgers.edu; krogh@rutchem.rutgers.edu
Abstract: The pincer-ligated species (PCP)Ir (PCP ) κ3-C6H3-2,6-(CH2PtBu2)2) is found to promote
dimerization of phenylacetylene to give the enyne complex (PCP)Ir(trans-1,4-phenyl-but-3-ene-1-yne). The
mechanism of this reaction is found to proceed through three steps: (i) addition of the alkynyl C-H bond
to iridium, (ii) insertion of a second phenylacetylene molecule into the resulting Ir-H bond, and (iii) vinyl-
acetylide reductive elimination. Each of these steps has been investigated, by both experimental and
computational (DFT) methods, to yield unexpected conclusions of general interest. (i) The product of alkynyl
C-H addition, (PCP)Ir(CCPh)(H) (3), has been isolated and, in accord with experimental observations, is
calculated to be 29 kcal/mol more stable than the analogous product of benzene C-H addition. (ii) Insertion
of a second PhCCH molecule into the Ir-H bond of 3 proceeds rapidly, but with a 1,2-orientation. This
orientation gives (PCP)Ir(CCPh)(CPhdCH2) (4) which would yield the 1,3-diphenyl-enyne if it were to
undergo C-C elimination; however, the insertion is reversible, which represents the first example, to our
knowledge, of simple â-H elimination from a vinyl group to give a terminal hydride. The 2,1-insertion product
(PCP)Ir(CCPh)(CHdCHPh) (6) forms more slowly, but unlike the 1,2 insertion product it undergoes C-C
elimination to give the observed enyne. (iii) The failure of 4 to undergo C-C elimination is found to be
general for (PCP)Ir(CCPh)(vinyl) complexes in which the vinyl group has an R-substituent. Thus, although
C-C elimination relieves crowding, the reaction is inhibited by increased crowding. Density-functional theory
(DFT) calculations support this surprising conclusion and offer a clear explanation. Alkynyl-vinyl bond
formation in the C-C elimination transition state involves the vinyl group π-system; this requires that the
vinyl group must rotate (around the Ir-C bond) by ca. 90° to achieve an appropriate orientation. This
rotation is severely inhibited by steric crowding, particularly when the vinyl group bears an R-substituent.
Introduction
a metal-carbon or metal-hydrogen bond, and then elimination
of either C-H or C-C bonds, respectively (see Scheme 1).
The formation of carbon-carbon bonds is the foundation of
organic chemical synthesis. Regioselective activation of a C-H
bond, followed by carbon-carbon bond formation, is accord-
ingly an extremely attractive challenge that has become the
subject of intense study in recent years.1 Particularly since the
report by Murai of ruthenium-catalyzed regioselective addition
of aryl C-H bonds across C-C double bonds,2 considerable
advances have been made by several groups3-6 in the use of
transition-metal complexes to promote such reactions. Perhaps
the most conceptually straightforward pathways for such
transformations generally involve C-H addition by a metal
center, followed by insertion of an unsaturated species into either
The Murai systems, for example, have been shown to proceed
via the latter type of pathway, that is, olefin insertion into a
Ru-H bond followed by C-C elimination.2
One example of catalytic addition of a C-H bond across a
multiple bond, that has been fairly well developed, is the
dimerization of acetylenes, eq 1.7-38
(4) (a) Matsumoto, T.; Taube, D. J.; Periana, R. A.; Taube, H.; Yoshida, H. J.
Am. Chem. Soc. 2000, 122, 7414-7415. (b) Matsumoto, T.; Periana, R.
A.; Taube, D. J.; Yoshida, H. J. Mol. Catal. A: Chem. 2002, 180, 1-18.
(c) Periana, R. A.; Liu, X. Y.; Bhalla, G. Chem. Commun. 2002, 3000-
3001. (d) Bhalla, G.; Oxgaard, J.; Goddard, W. A., II; Periana, R. A.
Organometallics 2005, 24, 5499-5502. (e) Bhalla, G.; Oxgaard, J.;
Goddard, W. A., III; Periana, R. A. Organometallics 2005, 24, 3229-
3232. (f) Oxgaard, J.; Bhalla, G.; Periana, R. A.; Goddard, W. A., III.
Organometallics 2006, 25, 1618-1625.
(1) (a) Guari, Y.; Sabo-Etiennne, S.; Chaudret, B. Eur. J. Inorg. Chem. 1999,
1047-1055. (b) Ritleng, V.; Sirlin, C.; Pfeffer, M. Chem. ReV. 2002, 102,
1731-1769. (c) Jun, C.-H.; Lee, J. H. Pure Appl. Chem. 2004, 76, 577-
587.
(2) (a) Murai, S.; Kakiuchi, F.; Sekine, S.; Tanaka, Y.; Kamatani, A.; Sonoda,
M.; Chatani, N. Nature 1993, 366, 529-531. (b) Kakiuchi, F.; Murai, S.
Acc. Chem. Res. 2002, 35, 826-834.
(3) Weissman, H.; Song, X.; Milstein, D. J. Am. Chem. Soc. 2001, 123, 337-
338.
(5) (a) Lail, M.; Arrowood, B. N.; Gunnoe, T. B. J. Am. Chem. Soc. 2003,
125, 7506-7507. (b) Goj, L. A.; Gunnoe, T. B. Curr. Org. Chem. 2005,
9, 671-685.
(6) (a) Thalji, R. K.; Ellman, J. A.; Bergman, R. G. J. Am. Chem. Soc. 2004,
126, 7192-7193. (b) Thalji, R. K.; Ahrendt, K. A.; Bergman, R. G.; Ellman,
J. A. J. Am. Chem. Soc. 2001, 123, 9692-9693.
(7) Hoehn, A.; Werner, H. J. Organomet. Chem. 1990, 382, 255-272.
(8) Werner, H.; Brekau, U. Z. Naturforsch., B: Chem Sci. 1989, 44, 1438-
1446.
9
10.1021/ja0647194 CCC: $37.00 © 2007 American Chemical Society
J. AM. CHEM. SOC. 2007, 129, 853-866
853

A R T I C L E S
Ghosh et al.
Although this may be considered a somewhat “easy” case,
since the acetylenic C-H bond is generally much more reactive
than sp3 or even sp2 C-H bonds (and C-C triple bonds are
generally more reactive than double bonds39), we felt that a
mechanistic study of acetylene dimerization could yield insights
relevant to the broader class of C-H addition across multiple
bonds. Herein we report the results of such a study, which has
revealed unanticipated fundamental aspects of the insertion of
alkyne bonds into M-H bonds, and, in particular, C-C bond
elimination.
Figure 1. X-ray structure of (PCP)Ir(trans-1,4-phenyl-but-3-ene-1-yne) (1).
Scheme 1
Results and Discussion
1. Phenylacetylene Dimerization. The reaction of (PCP)-
t
IrH2 (PCP ) κ3-C6H3-2,6-(CH2PR2)2; R ) Bu for all experi-
mental work, and for computational work as well unless noted
otherwise) with 2 equiv norbornene (NBE) yields norbornane
and a compound of the composition (PCP)Ir(NBE). Although
not well characterized, “(PCP)Ir(NBE)” readily acts as a source
of (PCP)Ir. Addition of 2 equiv phenylacetylene (PhCCH) to a
xylene solution of this precursor results, after ca. 8 h at room
temperature, in the clean formation of a single complex (1) as
1
indicated by 31P NMR. The H NMR spectrum is consistent
(9) Schaefer, M.; Mahr, N.; Wolf, J.; Werner, H. Angew. Chem. 1993, 105,
1377-1379.
with characterization of 1 as an adduct of (PCP)Ir and the
PhCCH dimerization product, trans-1,4-phenyl-but-3-ene-1-yne
(2), as indicated in eq 2.
(10) Schaefer, M.; Wolf, J.; Werner, H. Organometallics 2004, 23, 5713-5728.
(11) Werner, H. Coord. Chem. ReV. 2004, 248, 1693-1702.
(12) Krueger, P.; Werner, H. Eur. J. Inorg. Chem. 2004, 481-491.
(13) Schaefer, M.; Wolf, J.; Werner, H. J. Chem. Soc., Dalton Trans. 2005,
1468-1481.
(14) Weng, W.; Guo, C.; Celenligil-Cetin, R.; Foxman, B. M.; Ozerov, O. V.
Chem. Commun. 2006, 197-199.
(15) Ogoshi, S.; Ueta, M.; Oka, M.-A.; Kurosawa, H. Chem. Commun. 2004,
2732-2733.
(16) Lee, C.-C.; Lin, Y.-C.; Liu, Y.-H.; Wang, Y. Organometallics 2005, 24,
136-143.
(17) Nishiura, M.; Hou, Z. J. Mol. Catal. A: Chem. 2004, 213, 101-106.
(18) Yang, C.; Nolan, S. P. J. Org. Chem. 2002, 67, 591-593.
(19) Boese, W. T.; Goldman, A. S. Organometallics 1991, 10, 782-786.
(20) Li, X.; Vogel, T.; Incarvito, C. D.; Crabtree, R. H. Organometallics 2005,
24, 62-76.
Support for this characterization was obtained by the addition
of CO to the resulting solution, yielding (PCP)Ir(CO)40 and 2
(eq 3). Enyne 2 was characterized (1H and 13C NMR) by
comparison with a known sample16 and by single-crystal X-ray
diffraction of cocrystallized (PCP)Ir(CO) and 2 (Supporting
Information).
(21) Esteruelas, M. A.; Herrero, J.; Lopez, A. M.; Olivan, M. Organometallics
2001, 20, 3202-3205.
(22) Dash, A. K.; Eisen, M. S. Org. Lett. 2000, 2, 737-740.
(23) Trost, B. M.; Sorum, M. T.; Chan, C.; Ruehter, G. J. Am. Chem. Soc. 1997,
119, 698-708.
(24) Trost, B. M.; Chan, C.; Ruhter, G. J. Am. Chem. Soc. 1987, 109, 3486-
3487.
(25) Derien, S.; Monnier, F.; Dixneuf, P. H. Top. Organomet. Chem. 2004, 11,
1-44.
(26) Musaev, D. G.; Morokuma, K. Top. Organomet. Chem. 2005, 12, 1-30.
(27) Bianchini, C.; Peruzzini, M.; Vacca, A.; Zanobini, F. Organometallics 1991,
10, 3697-707.
(28) Bruce, M. I. Chem. ReV. 1991, 91, 197-257.
(29) Perez-Carreno, E.; Paoli, P.; Ienco, A.; Mealli, C. Eur. J. Inorg. Chem.
1999, 1315-1324.
(30) Bruneau, C.; Dixneuf, P. H. Acc. Chem. Res. 1999, 32, 311-323.
(31) Grotjahn, D. B.; Zeng, X.; Cooksy Andrew, L. J. Am. Chem. Soc. 2006,
128, 2798-2799.
(32) Herberich, G. E.; Barlage, W. Organometallics 1987, 6, 1924-1930.
(33) Bassetti, M.; Casellato, P.; Gamasa, M. P.; Gimeno, J.; Gonzalez-Bernardo,
C.; Martin-Vaca, B. Organometallics 1997, 16, 5470-5477.
(34) Frohnapfel, D. S.; White, P. S.; Templeton, J. L. Organometallics 2000,
19, 1497-1506.
(35) Bassetti, M.; Marini, S.; Diaz, J.; Gamasa, M. P.; Gimeno, J.; Rodriguez-
Alvarez, Y.; Garcia-Granda, S. Organometallics 2002, 21, 4815-4822.
(36) Gao, Y.; Jennings, M. C.; Puddephatt, R. J. J. Chem. Soc., Dalton Trans.
2003, 261-265.
Crystals of 1 were obtained by slow evaporation of a pentane
solution; the characterization of 1 as the (PCP)Ir adduct of 2
was confirmed by X-ray diffraction (Figure 1). The triple bond
of the enyne is π-bound to the iridium center; the C-C triple
bond distance is 1.270(4) Å, significantly greater than the crystal-
(37) Navarro, J.; Sola, E.; Martin, M.; Dobrinovitch, I. T.; Lahoz, F. J.; Oro, L.
A. Organometallics 2004, 23, 1908-1917.
(38) Hirabayashi, T.; Sakaguchi, S.; Ishii, Y. AdV. Synth. Catal. 2005, 347, 872-
876.
(39) (a) Marchenko, A. V.; Gerard, H.; Eisenstein, O.; Caulton, K. G. New J.
Chem. 2001, 25, 1244-1255. (b) Marchenko, A. V.; Gerard, H.; Eisenstein,
O.; Caulton, K. G. New J. Chem. 2001, 25, 1382-1388.
(40) Morales-Morales, D.; Redon, R.; Wang, Z.; Lee, D. W.; Yung, C.;
Magnuson, K.; Jensen, C. M. Can. J. Chem. 2001, 79, 823-829.
9
854 J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007

Dimerization of Alkynes
A R T I C L E S
related aryl (and vinyl) hydride complexes have been found to
undergo rapid (NMR time scale) exchange with free arene, at
temperatures as low as -40 °C, via a dissociative pathway (as
determined by dynamic NMR kinetics experiments). Consistent
with a dissociative exchange pathway, no H/D exchange
between arenes is observed to be effected by (PCP)Ir(aryl)(H)
complexes. For example, the rapid equilibrium shown in eq 5
is observed in a solution of C6D6 and toluene-d0.42 Thus, less
than 1 of every ca. 106 arene/arene exchange events results in
H/D exchange.
(PCP)Ir(C6D5)(D) + toluene-d0 h
(PCP)Ir(tolyl)(H) (isomers) + C6D6 (5)
Figure 2. X-ray structure of (PCP)Ir(H)(CCPh) (3) with thermal ellipsoids
shown at 50% thermal probability. Only one of two inequivalent molecules
per asymmetric unit is shown.
In contrast to the aryl hydride analogues, the NMR spectrum
of 3 is that of a nonfluxional complex even at 90 °C, thus
alkynyl C-H elimination is orders of magnitude slower than
elimination of aryl or vinyl C-H bonds. This suggests that
addition of the alkynyl C-H bond is much more favorable than
addition of a phenyl C-H bond. Accordingly, the addition of
phenylacetylene (1.0 equiv) to a benzene solution of (PCP)-
IrPhH results in complete conversion to complex 3.
lographic value for the free enyne (1.21 Å) and in the range of
other acetylenes coordinated to neutral iridium(I) centers.41
Although the dimerization does not proceed catalytically (for
reasons that will be discussed below), eq 2 is clearly relevant
to several late-transition-metal systems which do catalyze
phenylacetylene dimerization in some cases to give, predomi-
nantly or exclusively, enynes of the same regio- and stereo-
chemistry as 2.10,13-17 Prompted by our interest in the formation
of C-C bonds via C-H activation, we undertook a mechanistic
study of reaction 2.
2. Addition of the Alkynyl C-H Bond. Addition of a single
equiv of phenylacetylene to (PCP)Ir(NBE) results in the
immediate clean formation of a single product (3) as indicated
by 31P NMR spectroscopy (δ 72.3 ppm).
(PCP)Ir(H)(Ph) + HCCPh u
(PCP)Ir(H)(CCPh) (3) + PhH (6)
As the equilibrium of eq 6 lies too far to the right to be
measured, even in benzene solution, we obtain a lower limit,
K6 > 104, corresponding to ∆G < -5.4 kcal/mol. If we
approximate ∆S ) 0, then ∆H < -5.4 kcal/mol. Since the
alkynyl C-H bond is ca. 18 kcal/mol stronger than the phenyl-H
bond,43,44 it follows that the Ir-CCPh bond of 3 is at least 23
kcal/mol stronger than the Ir-Ph bond of (PCP)IrPhH. Density-
functional theory (DFT) calculations (discussed in section 8)
indicate that the difference is even greater; reaction 6 is
calculated to be exoergic by 28.6 kcal/mol. Direct calculation
of bond dissociation energies (BDEs) from consideration of the
proper homolytic bond-cleavage reactions leads to the following
Ir-C and H-C values (kcal/mol): Ir-CCPh ) 114.0, Ir-C(Ph)
) 72.1, H-CCPh ) 124.5, and H-C(Ph) ) 108.2. Thus, the
computed Ir-CCPh BDE is substantial and about 42 kcal/mol
greater than that of the Ir-C(Ph) bond.
Although 3 is nonfluxional on the NMR time scale, it does
undergo alkyne/alkyne exchange within minutes of mixing at
room temperature. When PhCCD is added to a solution of 13
C
The 1H NMR spectrum, in particular the strongly upfield signal
at δ -43.48 ppm, is indicative of a five-coordinate Ir(III)
complex with H trans to a vacant coordination site. The spectral
features attributable to the t-butyl and methylene protons indicate
that the molecule has only Cs symmetry. These observations
indicated that 3 is the product of addition of the phenylacetylene
alkynyl C-H bond to (PCP)Ir. Crystals of 3 were obtained by
slow evaporation of a hexane solution of the compound. X-ray
diffraction confirmed its spectroscopic characterization as the
phenylacetylene C-H addition product; an ORTEP diagram is
shown in Figure 2.
labeled 3, (PCP)Ir(13CCPh)(H), full isotopic scrambling is
observed according to the equilibrium of eq 7.
(PCP)Ir(13CCPh)(H) + DCCPh h (PCP)Ir(CCPh)(D) +
H13CCPh h (PCP)Ir(13CCPh)(D) + HCCPh h
(PCP)Ir(CCPh)(H) + D13CCPh (7)
Thus, in spite of a much slower rate of alkyne/alkyne
exchange as compared with the arene/arene exchange reactions
of (PCP)Ir(H)(aryl), H/D exchange is much faster for the alkynyl
hydride 3. The isotopic exchange is suggestive of an associative
mechanism for alkyne/alkyne exchange. Such a mechanism
Five-coordinate products of aryl and vinyl C-H bond addition
to (PCP)Ir have been previously reported.42 (PCP)Ir(H)(Ph) and
(41) (a) Calabrese, J. C.; Roe, D. C.; Thorn, D. L.; Tulip, T. H. Organometallics
1984, 3, 1223-1230. (b) Grotjahn, D. B.; Hoerter, J. M.; Hubbard, J. L. J.
Am. Chem. Soc. 2004, 126, 8866-8867.
(42) Kanzelberger, M.; Singh, B.; Czerw, M.; Krogh-Jespersen, K.; Goldman,
A. S. J. Am. Chem. Soc. 2000, 122, 11017-11018.
(43) Barckholtz, C.; Barckholtz, T. A.; Hadad, C. M. J. Am. Chem. Soc. 1999,
121, 491-500.
(44) Ervin, K. M.; Gronert, S.; Barlow, S. E.; Gilles, M. K.; Harrison, A. G.;
Bierbaum, V. M.; DePuy, C. H.; Lineberger, W. C.; Ellison, G. B. J. Am.
Chem. Soc. 1990, 112, 5750-5759.
9
J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007 855

A R T I C L E S
Ghosh et al.
could proceed, inter alia, through an Ir(V) intermediate,45 an
Ir(V) transition state (or oxidative hydrogen migration),46 or a
concerted σ-bond metathesis pathway.
3. Insertion of Phenylacetylene into Ir-H Bonds. Addition
of 1 equiv phenylacetylene to a benzene solution of acetylide
complex 3 at room-temperature results in the rapid formation
of two new species, 4a and 4b, with 31P NMR chemical shifts
of δ 36.8 and 36.5 ppm, respectively. No hydride peaks are
1
present in the H NMR spectrum, which does reveal two sets
of triplets possibly attributable to the protons of two (PCP)Ir-
bound vinyl groups; one set appears at 4.35 and 4.39 ppm, while
the second set is observed at 4.79 and 6.65 ppm. 31P-1H hetero-
COSY revealed that the first set correlates to the 31P NMR peak
at 36.8 ppm, and the second set correlates to the peak at 36.5
ppm. No cross correlation is observed between these two sets
of peaks in the 1H COSY spectrum. The spectral data are
consistent with the assignment of the two species as products
of insertion of the PhCCH triple bond into the Ir-H bond of 3,
to give two rotamers, as indicated in eq 8.
Figure 3. X-ray structure of (PCP)Ir(PhCdCH2)(CCPh)(CO) (4-CO) with
thermal ellipsoids shown at 50% thermal probability.
Addition of CO (1 atm) to a 4a/4b mixture at room-
temperature results in a single CO addition product, 4-CO.
Complex 4-CO was crystallized by slow evaporation of a
hexane solution. X-ray diffraction reveals that 4-CO has the
rotational conformation (CH2 group syn to PCP aryl) assigned
to 4a (Figure 3). When CO (1.7 atm) was added to a 4a/4b
mixture at -35 °C, all 4a was immediately converted to 4-CO,
but 4b remained unreacted. Slowly warming the solution to
20 °C eventually resulted in the disappearance of 4b and
complete conversion to 4-CO. These results clearly offer further
support to the assignment of 4a and 4b as the iridium(1-
phenylvinyl)(2-phenylacetylide) complexes, and in particular as
the respective rotamers indicated in eqs 8 and 9.
The large difference in the chemical shift of the vinylic
protons of 4b is consistent with the proximity of one proton to
the acetylide group, a well-precedented effect attributed to the
magnetic susceptibility of the triple bond.47
When complexes 4a and 4b are prepared from the reaction
of (PCP)Ir(NBE) with 2 equiv PhC13CH, their 13C NMR
spectroscopic parameters are found to be quite similar but with
one notable difference: a 13C-13C coupling constant of 7.6 Hz
is observed for the isotopically enriched positions of 4b, whereas
no such coupling is observable for 4a. The assignment of 4a
and 4b as indicated in eq 8 was strongly supported by DFT
calculations (discussed in detail in section 8); the spin-spin
couplings between the acetylide-R-C and vinyl-â-C nuclei were
calculated as 9.5 and 0.5 Hz for 4b and 4a, respectively. Thus,
this surprisingly large difference (for which we have no
qualitative explanation) was very well captured by the calcula-
tions.
The formation of complexes 4a and 4b was an unexpected
result, given that the ultimate product of (PCP)Ir with 2 equiv
of PhCCH is the 1,4-diphenyl-butyne-ene adduct, 1. As indicated
in Scheme 1, PhCCH insertion into the Ir-H bond is, a priori,
a plausible first step toward that product; however, the regio-
chemistry of the observed insertion is inconsistent with the
mechanism of Scheme 1, since C-C elimination would give
1,3-diphenyl-butyne-ene.
The observed “wrong” regiochemistry of the PhCCH insertion
(a 1,2-addition), to give 4a and 4b, is offset by an even more
surprising result. The acetylene insertion reaction of eq 8 is
found to be reVersible, yielding easily detected equilibrium
concentrations of 3, 4a, 4b, and phenylacetylene. Thus, the
observed 1,2-addition, to give 4a and 4b, is presumably a dead-
end on the pathway to complex 1.
(45) (a) Klei, S. R.; Tilley, T. D.; Bergman, R. G. J. Am. Chem. Soc. 2000,
122, 1816-1817. (b) Kawamura, K.; Hartwig, J. F. J. Am. Chem. Soc.
2001, 123, 8422-8423. (c) Mohammad, H. A. Y.; Grimm, J. C.; Eichele,
K.; Mack, H.-G.; Speiser, B.; Novak, F.; Quintanilla, M. G.; Kaska, W.
C.; Mayer, H. A. Organometallics 2002, 21, 5775-5784. (d) Krogh-
Jespersen, K.; Czerw, M.; Summa, N.; Renkema, K. B.; Achord, P. D.;
Goldman, A. S. J. Am. Chem. Soc. 2002, 124, 11404-11416. (e) Webster,
C. E.; Hall, M. B. Coord. Chem. ReV. 2003, 238-239; 315-331. (f) Lam,
W. H.; Lam, K. C.; Lin, Z.; Shimada, S.; Perutz, R. N.; Marder, T. B. J.
Chem. Soc., Dalton Trans. 2004, 1556-1562. (g) Bernskoetter, W. H.;
Lobkovsky, E.; Chirik, P. J. Organometallics 2005, 24, 4367-4373.
(46) (a) Oxgaard, J.; Muller, R. P.; Goddard, W. A.; Periana, R. A. J. Am. Chem.
Soc. 2004, 126, 352-363. (b) Oxgaard, J.; Periana, R. A.; Goddard, W.
A., III. J. Am. Chem. Soc. 2004, 126, 11658-11665.
To our knowledge there are no reported precedents of a well-
characterized reversible insertion of an alkyne into a metal-H
bond, or, even more generally, â-H elimination from a metal-
vinyl complex to give a simple metal hydride. Perhaps the only
(47) (a) Agarwal, A.; McGlinchey, M. J. Can. J. Chem. 1978, 56, 959-961.
(b) Alkorta, I.; Elguero, J. New J. Chem. 1998, 22, 381-385. (c) Abraham,
R. J.; Reid, M. J. Chem. Soc., Perkin Trans. 2001, 2, 1195-1204. (d)
Martin, N. H.; Nance, K. H. J. Mol. Graphics Modell. 2002, 21, 51-56.
9
856 J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007

Dimerization of Alkynes
A R T I C L E S
Complex 5 was characterized by 1H and 31P NMR; particularly
diagnostic are the vinyl proton signals, which are broad peaks
at δ 8.60 and δ 4.65 ppm, with JHH ) 13.5 Hz. A closely related
vinyl complex, in which an analogous R-vinyl proton resonates
at δ 8.22 ppm, has been reported by Crabtree.20
well characterized example of alkyne deinsertion from a metal-
vinyl group was reported by Puddephatt.36 In that case, an
equilibrium was observed and measured; however, the product
was a bridging hydride and the formation of such a bridge must
obviously affect the thermodynamics to some degree.
Complex 5 was difficult to obtain in pure form but addition
of CO atmosphere to a solution of complex 5 gave the CO
adduct, 5-CO, which was successfully purified by column
chromatography.
The 1H and 31P NMR parameters of 5-CO were similar to that
of 5 (δ 8.73 ppm, JHH ) 18.2 Hz, JHP ) 2.2 Hz; δ 5.83 ppm,
JHH ) 18.2 Hz, JHP ) 2.2 Hz). The complex was recrystallized
from 25% hexane in benzene, and its structure was crystallo-
graphically determined (Figure 6).
We further exploited the ability to obtain pure samples of
the CO adduct. Complex 5-CO was synthesized from the
reaction of (impure) 5 with CO. Following column chroma-
tography, it was reacted with trimethylamine oxide to give back
complex 5, which was then recrystallized from hexane solution.
The structure of 5 was confirmed by crystallography (Figure
6).
To further investigate the novel equilibrium of eq 10, the
temperature dependence was determined from 25 up to 45 °C
at intervals of 5 °C (and then back down to 25 °C at the same
intervals to confirm that the observed concentrations were in
fact equilibrium values). Plots of R ln K versus 1/T (Figure 4)
yielded the following values of ∆H and ∆S: for formation of
4a, -15.4(5) kcal mol-1 and -38.9(16) cal deg-1 mol-1, and
for formation of 4b, -13.9(4) kcal mol-1 and -37.0(14) cal
deg-1 mol-1. These very negative entropy values are consistent
with the formation of a highly crowded species in which
numerous low-frequency vibrational modes are restricted, rela-
tive to complex 3 and free phenylacetylene.
At -35 °C, the reaction of complex 3 (37 mM) with 1.5 equiv
of phenylacetylene gave complex 4a exclusively; the conversion
of 3 to 4a is observed to be essentially complete in ap-
proximately 45 min. This result is consistent with DFT
calculations (discussed in detail in section 8), which show that
the most favorable approach of the incoming acetylene is a
vacant coordination site trans to the PCP iridium-bound carbon.
Such an approach would yield the rotameric form 4a, as
indicated in Figure 5. Warming the resulting solution of 4a to
room-temperature yields a mixture of 4a and 4b (in the same
ratio, 1:0.6, as obtained when PhCCH is added to 3 at room
temperature).
The kinetic data was fit using the program Gepasi48 to give
the rate constants 1.08(9) × 10-1 M-1 s-1 for k11(4a) (insertion)
and 1.3(3) × 10-4 s-1 for k-11(4a) (â-H elimination). These rates
correspond to activation free energies (-35 °C) of 14.9 kcal/
mol and 18.1 kcal/mol, respectively.
The kinetic barrier to the insertion reaction of eq 11 is
significantly greater than that of reaction 8 (insertion of PhCCH
into the Ir-H bond of 3). Reaction 11 proceeds slowly at
90 °C (approximately 2 × 10-4 M-1 s-1), as compared with
reaction 8 which is relatively fast at -35 °C (1.08(9) × 10-1
M-1 s-1). While reaction 11 does not proceed cleanly, giving
only about a 40% NMR yield of 5 (the remaining products were
unstable to chromatography and not characterized), the 1H NMR
spectrum of the product solution (vinyl H region) reveals the
absence of any significant quantity of the 1,2-insertion product.
Since we see no reason why the kinetic regioselectivity of
addition would be completely reversed for (PCP)IrHCl versus
(PCP)IrH(CCPh), we suspect that 1,2-addition is also kinetically
favorable for (PCP)IrHCl, but reversible, and that the 2,1-
addition product is the thermodynamic sink (probably more
favorable owing to steric effects).
Further investigation of the reactivity of PhCCH with iridium
hydride complexes was conducted with (PCP)IrHCl. As in the
case of the reaction with hydride 3, insertion occurred, and with
complete regioselectivity. Quite surprisingly, however, in the
case of (PCP)IrHCl, the only insertion product observed was
the result of the reverse regioselectivity, that is, 2,1-insertion.
To summarize this section, PhCCH inserts readily into the
Ir-H bond of 3, as required by the Ir-H addition pathway of
Scheme 1. However, the 1,2 regiochemistry of the insertion is
inconsistent with this pathway. But the insertion is easily
reversible (which represents an unprecedented example of simple
(48) (a) Mendes, P. Comput. Appl. Biosci. 1993, 9, 563-571. (b) Mendes, P.
Trends Biochem. Sci. 1997, 22, 361-363. (c) Mendes, P.; Kell, D.
Bioinformatics (Oxford) 1998, 14, 869-883.
9
J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007 857

A R T I C L E S
Ghosh et al.
Figure 4. Equilibrium plots for the reversible insertion of phenylacetylene (eq 10): formation of complex 4a and formation of complex 4b.
Addition of CO to a benzene solution of 6 yields the carbonyl
adduct 6-CO. Crystals of 6-CO were obtained from pentane,
and its structure was determined by crystallography (Figure 7).
Figure 5. Two possible approaches of a phenylacetylene molecule for 1,2-
insertion into the Ir-H bond of 3. Without rotation around the resulting
Ir-C bond, each approach can yield one of the two isomer/rotamers of 4.
Complex 6 (Ar ) Ph) is isomeric with complexes 4a and
4b, with a phenyl on the vinylic â-carbon (trans to Ir) rather
than the R-carbon. Complex 6 is therefore presumably less
crowded than 4a and 4b which do not undergo C-C elimination.
Thus, a priori, it would be expected that 6 would also not
undergo C-C elimination. In fact, in striking contrast with the
behaVior of 4, complex 6 does undergo facile, and clean, C-C
bond elimination to yield complex 1.
â-H elimination from a transition-metal vinyl complex), and it
presumably does not play a role in the formation of the 1,4-
diphenylbutene-yne product. Interestingly, PhCCH inserts into
the Ir-H bond of (PCP)IrHCl to give exclusively the product
of 2,1-insertion.
4. C-C Elimination To Give Complex 1. In the context of
the mechanism of eq 2, the reversible 1,2-insertion of PhCCH
into the Ir-H bond of 3 presumably represents an out-of-cycle
dead end. In that respect the two pathways of Scheme 1, Ir-H
addition and Ir-C addition, are equally consistent with this
insertion reaction. The very low barrier to the 1,2-insertion (∆Gq
≈ 15 kcal/mol), however, suggested to us that the 2,1-insertion
(i.e., the regioselectivity consistent with the observed product)
was likely to be at least facile enough to be consistent with the
overall kinetics of reaction 2. This suspicion was supported by
the observation that the reaction of PhCCH with (PCP)IrHCl
gave exclusively the 2,1-insertion product; thus the barrier to a
2,1-insertion into a (PCP)IrH bond is not necessarily prohibi-
tively high.
With this hypothesis in mind, we attempted an independent
synthesis of the product that would result from the 2,1-insertion
of PhCCH into the Ir-H bond of 3. Complex 5 was reacted
with 2.5 equiv lithium arylacetylide (1 M solution in THF) to
yield the desired iridium(acetylide)vinyl complex, 6, in 92%
yield (eq 14; Ar ) tolyl). Lithium p-tolylacetylide (rather than
phenylacetylide) was used to aid in NMR characterization, and
to confirm the origin of the acetylide group in product 6 and in
any subsequent elimination reactions.
At 40 °C, the rate constant of reaction 16 (Ar ) Ph) is
3.1(5) × 10-4 s-1 (corresponding to ∆Gq ) 23.4 kcal/mol;
Figure 8). (It should be noted that it was possible to remove
from solution most, but not all, excess THF from the solution
in which 6 was generated; continued efforts to remove all THF
were accompanied by loss of 6. Residual THF concentrations
varied, which did not affect the rate of eq 16. However, we
cannot rigorously exclude the possibility that coordination of
THF could be inhibiting C-C elimination;49 in that case the
intrinsic difference between the C-C elimination rates of
complex 6 and complexes 4 would be even greater than these
results indicate.)
(49) (a) Crumpton, D. M.; Goldberg, K. I. J. Am. Chem. Soc. 2000, 122, 962-
963. (b) Crumpton-Bregel, D. M.; Goldberg, K. I. J. Am. Chem. Soc. 2003,
125, 9442-9456 and references therein.
9
858 J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007

Dimerization of Alkynes
A R T I C L E S
Figure 6. X-ray structures of (PCP)Ir(CHdCHPh)(Cl) (5) and (PCP)Ir(CHdCHPh)(Cl)(CO) (5-CO) with thermal ellipsoids shown at 50% thermal probability.
complex 3 (eq 4) is observed to occur rapidly upon addition of
PhCCH to (PCP)Ir(NBE); (ii) insertion of PhCCH into the Ir-H
bond of 3 occurs rapidly, with the wrong regioselectivity (1,2
insertion) to lead to complex 2; however, this reaction is readily
reversible; (iii) insertion of PhCCH into the Ir-H bond of
(PCP)IrHCl occurs (slowly) to give exclusively the 2,1-insertion
product; thus, in a species closely related to 3, the barrier to
2,1-insertion is high, but not prohibitively so; (iv) most
importantly, complex 6, the putative product of a 2,1-insertion
of PhCCH into the Ir-H bond of 3, undergoes C-C elimination
to give 1, with a rate faster than that of the overall reaction of
eq 2; that is, 6 is a kinetically competent intermediate.
5. Kinetics of Enyne Formation. Kinetics of the reaction
of acetylide hydride 3 (85 mM) with 2 equiv phenylacetylene,
to give enyne complex 1, were monitored at 40 °C by 31P NMR.
Figure 7. X-ray structure of (PCP)Ir(CO)(CHdCHPh)[CC(p-tol)] (6-CO;
Ar ) o-Tol) with thermal ellipsoids shown at 50% thermal probability.
The data obtained was fit using Gepasi48 (Figure 9) using a
scheme (Scheme 2) which involves reversible formation of
complex 4a and 4b (in equilibrium, K) and the relatively slow
formation of complex 6 which subsequently transforms to
complex 1. The rate constant for the formation of complex 6
(k3) was thereby determined to be 1.04(6) × 10-2 M-1 s-1. The
quality of the fit was less sensitive to the values of rate constants
other than k3, which are therefore less reliable: k1 ) 2.7 M-1
s-1, k-1 ) 3.7 × 10-2 s-1, k2 ) 5.5 × 10-1 M-1 s-1, k-2 ) 1.3
× 10-2 s-1, k4 ) 1.4 × 10-3 s-1, k5 ) 1.4 × 10-4 M-1 s-1 (k5
refers to a secondary reaction of 1 with PhCCH; see below).51
The failure to observe any 1,3-diphenyl-but-3-ene-1-yne
permits us to extrapolate an upper limit for the rate of C-C
bond elimination from 4a or 4b. By estimating a 2% detection
limit, we can obtain an upper limit of 3 × 10-6 s-1 for
Figure 8. Kinetics of eq 16, C-C reductive elimination from 6 (Ar )
elimination from 4a, or 5 × 10-6 s-1 for elimination from 4b.
Ph), at 40 °C. Lines shown are theoretical first-order fits.
These values correspond to lower limits, of the activation free
energies of elimination, of 26.3 kcal/mol and 26.0 kcal/mol,
respectively.
Consistent with an intramolecular mechanism for C-C
elimination from complex 6, the kinetics of eq 16 are first order
(Figure 8) while the coordinated enyne of complex 1-tol has a
6. C-C Elimination: Further Studies. Although, from the
tolyl group at the acetylide position and a phenyl group bound
outset of this investigation, we had considered the pathway
to the vinylic carbon (as shown in eq 16). The 1H NMR
spectrum of the enyne has a single signal attributable to a
p-methyl group, at δ 1.91 ppm, which corresponds to the methyl
on the phenylacetylenyl unit; the p-methyl on the styrenyl unit
(50) The trend of the p-methyl on the styrenyl being deshielded as compared to
the p-methyl on the phenylacetylene is seen in trans-1-(4-tolyl)-prop-1-
ene: Yu, J.; Gaunt, M.J.; Spencer, J. B. J. Org. Chem. 2002, 67, 4627-
4629. Seen in 1-(4-methylphenyl)-prop-1-yne: Weiss, H. M.; Touchette,
appears at δ 1.95 ppm (see Supporting Information for full
K. M.; Angell, S.; Khan, J. Org. Biomol. Chem. 2003, 1, 2152-2156.
(51) The value chosen for k1 was extrapolated to 40 °C from the value determined
experimentally at -35 °C, by assuming an entropy of -39 eu (the
thermodynamic entropy of insertion). The value of k-1 then gave a best fit
consistent with the observed equilibrium, K1. Since the reactions corre-
sponding to k1, k-1, k2, and k-2 are fast relative to k3, only the ratios (i.e.
K1 and K2) were constrained by the data.
characterization).50
The results discussed above and in preceding sections strongly
implicate the Ir-H insertion pathway (Scheme 1) as the
mechanism of eq 2. In particular (i) C-H addition to give
9
J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007 859

A R T I C L E S
Ghosh et al.
rather than by the Ir-H addition step. Ir-H addition occurs
preferentially in the “wrong” (1,2) direction; however, only the
product of the much less favorable (2,1) addition reaction
undergoes C-C elimination. This completely unanticipated
conclusion is all the more surprising because the more easily
formed insertion isomer is the more crowded one (4), but this
one does not undergo reductive elimination. In view of the
profound importance of C-C elimination in organometallic and
organic chemistry, this conclusion clearly merited further study.
The reaction of 1-phenylpropyne (PhCCMe) with complex
3 (50 °C, ca. 30 min) yields complexes 7a and 7b, in a ratio of
1.1:1. Complex 7a has been crystallographically characterized
as the product of 2,1 addition of Ir-H across the acetylene triple
bond (i.e., the product with phenyl bound to the â-vinyl carbon;
1
Figure 10 and eq 17). On the basis of its H and 31P NMR
spectral data, complex 7b is proposed to be an Ir-C rotamer
of 7a, in analogy with the phenylacetylene insertion products
4a and 4b. It is not clear if formation of 7a and 7b (the 2,1
addition products) are kinetically favored over 1,2-addition
products; in view of the results obtained with phenylacetylene,
it seems entirely plausible that the 1,2 addition is kinetically
favored, but reversible, while complexes 7a and 7b constitute
the thermodynamic sink.
In contrast with the closely related â-phenylvinyl complex
6, complex 7 shows no indication of undergoing C-C elimina-
tion. Complex 7 is stable up to ca. 60 °C (in contrast with 6,
which undergoes rapid C-C elimination at 40 °C, eq 16); at
70 °C, it undergoes conversion to give products resulting from
a complex rearrangement. This rearrangement is beyond the
scope of this paper; the important point is that the presence of
the methyl substituent on the vinyl group precludes vinyl-alkynyl
C-C elimination from 7, even at 70 °C.
In the course of this work, one other complex of the type
(PCP)Ir(CCPh)(CRdCHR′) was isolated. As noted above, the
formation of enyne 2 is not catalytic. Apparently, this is not
because the enyne is bound too strongly to be displaced. Instead,
if more than 2 equiv phenylacetylene are added to a solution of
(PCP)Ir(NBE), the formation of secondary products are observed
including one complex, 8, with a 31P NMR signal at 41.6 ppm.
This complex could be synthesized in almost quantitative yield
by treatment of complex 3 with independently synthesized enyne
Figure 9. Kinetic fits for the formation of enyne complex 1 from the
reaction of acetylide hydride 3 with phenylacetylene. (Note that most of 3
is rapidly converted to an equilibrium concentration of 4a and 4b prior to
the first data point.)
leading to product as shown in Scheme 2 to be very plausible,
the above results present a very surprising conclusion: the
regioselectivity of eq 2 is determined by the C-C elimination
Scheme 2
9
860 J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007

Dimerization of Alkynes
A R T I C L E S
Figure 10. X-ray structures of (PCP)Ir[C(Me)dCHPh](CCPh) (7a) and (PCP)Ir[C(CHCHPh)dCHPh](CCPh)(CO) (8-CO) with thermal ellipsoids shown
at 50% thermal probability.
Figure 12. trans-1,4-Phenyl-but-3-ene-1-yne (2), formed from the reaction
of (PCP)Ir(NBE) with PhCCH and PhCCD with isotope effects indicated.
conclusion would appear to have important potential implica-
tions for other systems in which C-C bonds (particularly with
an sp2 carbon) are formed. Insight into the origin of this effect
Figure 11. Characterized complexes (PCP)Ir(CCPh)(CRdCHR′)
was obtained from computational studies described below
(section 8).
2. The CO adduct of 8, which appears as a broad peak at 42.6
ppm in the 31P NMR spectrum, was crystallized from hexane,
and its structure was determined. The structure (Figure 10)
reveals that 8 is (or at least can be viewed as) the product of
2,1-addition of the Ir-H bond across the triple bond of 2 (eq
18).
7. Isotope Effects and Consideration of Vinylidene Path-
ways. Because of the rapid reaction of 3 with excess PhCCH,
it was not possible to directly measure the equilibrium isotope
effect for eq 20 with PhCCH/PhCCD. However, the sterically
hindered alkyne 2-ethynyltoluene reacts extremely slowly with
the tolyl analogue of 3, (PCP)IrH[CC(o-Tol)] (9) (formation of
the o-Tol analogue of enyne complex 1 has a half-life of ca. 5
days at 90 °C). The reaction of (PCP)Ir(NBE) with 1 equiv
DCC(o-Tol) (2-ethynyl-d1-toluene) immediately gave 9-d1; to
this solution was added 6.6 equiv of a 1:1 mixture of DCC(o-
Tol) and HCC(o-Tol). 1H NMR was used to determine the ratio
of 9-d0/9-d1 as well as the ratio of HCC(o-Tol)/DCC(o-Tol),
yielding a value of EIE20 ) 1.87(5) (Ar ) o-Tol) which we
assume to be very nearly equal to that of EIE20 for Ar ) Ph.
No indication of C-C elimination from complex 8 has been
observed. Heating a benzene solution of 8 at 55 °C results only
in slow decomposition over ca. 2 days.
Thus, we have observed and fully characterized at least four
species of the formula (PCP)Ir(CCPh)(CRdCHR′) (Figure 11).
Notably, only one of these complexes undergoes C-C elimina-
tion, namely 6, which is, surprisingly, the least sterically
hindered of the four complexes. The presence on the R-carbon
of either an electron-donating substituent (methyl; complex 7),
or electron-withdrawing substituents (phenyl and vinyl; com-
plexes 4 and 8, respectively), inhibits C-C elimination.
Therefore we must conclude that steric crowding, specifically
due to the presence of substituents at the vinyl R-carbon, inhibits
C-C elimination in the species shown in Figure 2. This
To determine the overall isotope effect for formation of the
enyne complex, (PCP)Ir(NBE) was reacted with 50 equiv each
of PhCCH and PhCCD, at 50 °C for 24 h. The coordinated
enyne was displaced from the resulting product 1 by addition
of CO to the reaction mixture. The displaced enyne 2 was
isolated by column chromatography and analyzed by 1H NMR
and H NMR in CDCl3 and CHCl3, respectively; the vinylic
H/D ratios were found to be 1.0(1):1 and 2.2(1):1 at the C3
and C4 positions, respectively (Figure 12).
2
The H/D ratio of unity for the C3 vinylic position is expected
(if secondary isotope effects are assumed to be small) for either
9
J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007 861

A R T I C L E S
Ghosh et al.
Figure 13. An a priori plausible mechanism for alkyne dimerization proceeding through vinylidene intermediates.
Table 1. Energy, Enthalpy, Entropy, and Free Energy (25 °C) Calculated (using PBE and B3LYP Functionals) for Selected Energy Minima
and Transition States and Relevant Experimental Valuesa
PBE
B3LYP
S
species
E
H
S
Gb
experimental/commentsc
H
G
(PCP)Ir + HCCH
+ HCCPh
3′ + HCCPh
3′(π-PhCCH)
TS-ins-6′
TS-ins-4a′
TS-ins-4b′
4a′
42.0
42.3
32
32.8
not observed
37.1
31
27.7
(0)
-3.9
9.7
5.7
15.4
-20.7
-19.5
-34.4
4.8
(0)
-1.9
9.8
6.1
16.4
(0)
-46
(0)
11.9
23.6
20.2
30.9
-3.2
-0.8
-19.7
22.6
-1.8
-33.2
-25.6
-7.9
0
(0)
9.3
22.6
18.5
(0)
-44
-46
-47
(0)
not observed
22.5
36.3
32.6
-46
-47
-48
-45
-48
-37
-50
-43
-37
-42
3
G: 21.2 (40°); 20.6 (25°)d
G: 14.9 (-35°); 17.2 (25°)e
no data
-16.6
H, S, G: -15.4; -39; -3.8 (25°)f
H, S, G: -13.9; -37; -2.9 (25°)f
no thermodynamic data
G > 22.5g
-8.1
-42
4.6
4b′
6′
-15
-30.8
7.7
-23.2
22.4
-37
-47
-12.0
36.6
TS-4a′-elim
TS-6′-elim
1′
-18
-14.6
-44.1
-38.3
-7.1
∆Gq ) 23.4h
-48.6
-43.1
-10.5
no thermodynamic data
not observed
-34.0
-28.8
-6.8
-34
-35
3
-23.8
-18.4
-7.8
(PCP)Ir(3-Ph-enyne)i
PCP)Ir + 2′ (free)
not observed
a All values in kcal/mol except entropy (cal/(deg•mol)). b Free energy values; 25 °C, 1.0 M standard state. c Experimentally based data for corresponding
observed species and transition states. d Obtained by fitting kinetic data (40 °C), to give ∆Gq ) 21.2 kcal/mol and then extrapolated to 25 °C, assuming ∆Sq
) -38 eu. e Obtained from direct kinetic measurement of eq 8 at -35 °C, to give ∆Gq ) 14.9 kcal/mol; then extrapolated to 25 °C, assuming ∆Sq
)
-38 eu. f ∆H; ∆S; ∆G; obtained from equilibrium measurements (eq 10) over the range 25-45 °C. g Obtained from fitting kinetic data (40 °C) to give a
lower limit for ∆Gq for C-C bond elimination from 4a (26.3 kcal/mol); then adding free energy of formation of 4a from 3 and PhCCH (-3.8 kcal/mol).
h Obtained from direct kinetic measurement (40 °C) of the reaction of independently synthesized 6 (eq 16). i Isomer of complex 1′ with phenyl group at C3
rather than C4 (as would result from C-C elimination from 6′ instead of 4′).
of the two insertion-type dimerization mechanisms indicated
in Scheme 1. The H/D ratio of 2.2 found for the C4 vinylic
position is larger than the value of 1.87 found for the C-H
addition EIE20. If secondary isotope effects are neglected, the
M-C insertion pathway (assuming M-C insertion is rate-
determining) would predict an H/D ratio equal to EIE20. The
M-H insertion pathway, however, would predict a KIE equal
to the product of EIE20 and the KIE for insertion into the Ir-
H(D) bond.
At least three distinct reaction pathways have been proposed
for the catalytic hydroalkynylation of triple bonds. Two such
pathways, involving C-H addition, insertion, then elimination,
are indicated in Scheme 1.19,32-38 Additionally, vinylidene
and B3LYP54 exchange and correlation functionals were applied
in the calculations; only results obtained with PBE were
generally in very good agreement with experiment, and only
these results will be discussed in detail here. The B3LYP
functionals generally gave substantially higher energies for the
more crowded species, in poor agreement with experimental
data and observations (see Table 1). Most calculations were
performed using the full nontruncated tBuPCP ligand, but changes
to the dimerization energy profile resulting from the use of a
truncated model pincer ligand, MePCP, in which t-butyl groups
are replaced with methyls, were also investigated. Assuming
that the phenyl group of the phenylacetylide ligand was too far
removed from the metal center to significantly impact the
dimerization reaction energies, we modeled it as a simple
acetylide (CCH) ligand, while the incoming phenylacetylene
molecule was not truncated. Computational models of experi-
mental tBuPCP complexes (n) described in the preceding
experimental sections, are labeled n′; calculated MePCP ana-
logues are labeled n′′.
8a. Insertion of Acetylene into the Ir-H Bond. Addition
of the Ir-H bond of 3′ to the triple bond of phenylacetylene
(i.e., insertion of phenylacetylene) is calculated to proceed with
a preference for the 1,2 orientation (TS-ins-4a′), in accord with
experiment. This preference is also calculated for 3′′, the
MePCP complex (Figure 16); however, the difference between
1,2 and 2,1 orientations (0.7 kcal/mol) is much smaller than in
the case of the full tBuPCP ligand (3.4 kcal/mol). This suggests
mechanisms have been proposed by Werner, Crabtree, and
10-13,20,27-31,39,52
others.7,
Unlike the insertion mechanism of
Scheme 1, any such vinylidene mechanism involves the cleavage
of the acetylenic C-H bonds of both alkyne components (see
Figure 13). Accordingly the observation of a KIE of unity for
C3 argues against such a mechanism. We also note that the
insertion mechanisms, but not the vinylidene mechanism, readily
explain the exclusive selectivity for formation of the trans double
bond geometry.
8. Computational Studies of the Full Dimerization Reac-
tion. The full mechanism of the acetylene dimerization,
proposed on the basis of experimental results discussed in the
preceding sections, along with related aspects such as acetylene/
acetylene exchange, was investigated computationally with the
use of DFT-based electronic structure calculations. Both PBE53
(53) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys ReV. Lett. 1996, 77, 3865.
(54) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652. (b) Lee, C.; Yang,
W.; Parr, R. G. Phys. ReV. B: Condens. Matter Mater. Phys. 1988, 37,
785.
(52) Li, X.; Incarvito, C. D.; Crabtree, R. H. J. Am. Chem. Soc. 2003, 125,
3698-3699.
9
862 J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007

Dimerization of Alkynes
A R T I C L E S
to the â-carbon, while also playing a dominant role in disfaVor-
ing the kinetics of formation of the same complex, as discussed
above.
In particular, it should be noted that the 1,2-insertion is 19.8
kcal/mol exoergonic for the MePCP complex at 25 °C (see Figure
16); that is, it is calculated to be effectively irreversible. In
contrast, the calculations predict that the free energy of 1,2-
insertion into the Ir-H bond of the tBuPCP complex (3′, to give
4a′) is -3.2 kcal/mol, and therefore effectively reVersible. This
result is in excellent agreement with the observation of reversible
insertion (eq 10) and, quantitatively, with the experimental free
energy value of -3.8 kcal/mol (25 °C) for insertion to give 4a.
Thus the calculated energies strongly indicate that the highly
unusual â-vinyl-hydrogen elimination from 4 (eq 10) is ther-
modynamically possible because of the unusual degree of
crowding. Considering the presence of the bulky t-butyl groups
on phosphorus and the sterically demanding phenyl group at
the R position it seems reasonable that complex 4a might be
unusually unfavorable thermodynamically. Inspection of the
calculated structure of 4a′ indeed reveals significant crowding,
and the contrast with the less crowded 2,1-insertion product 6′
is evident (Figure 17).
Figure 14. Calculated structures for TS-ins-4a′ (R-phenylvinyl) and TS-
ins-6′ (â-phenylvinyl). The perspective is approximately as viewed down
the acetylide C-C axis (atoms shown at 75% van der Waals radii; vinyl
carbons in red, alkynyl carbons in blue). TS-ins-4a′ is calculated to be 3.4
kcal/mol energetically below TS-ins-6′.
that the observed difference between the two orientations results
largely from steric factors. This conclusion may seem counter-
intuitive; steric crowding might be expected to favor the TS
with the phenyl group further from the iridium center. But the
phosphino alkyl groups are positioned away from the site trans
to the PCP carbon, and hence 1,2 insertion (TS-ins-4a′) may
be less sterically demanding than 2,1 insertion (TS-ins-6′). No
difference in the degree of crowding is obvious from inspection
of the calculated structures, in accord with the relatively small
energetic difference of 3.4 kcal/mol (see Figures 5 and 14). Note,
however, when the insertion reaction is complete, the phenyl
group is found in a much more sterically crowded position in
the case of the 1,2 orientation (see below).
A 1,2-addition in which the PhCCH approaches trans to the
acetylide group (TS-ins-4b′) was calculated to be 10.7 kcal/
mol higher in activation free energy than an approach trans to
the PCP aryl (TS-ins-4a′), in accord with the observation that
4a is the kinetically preferred rotamer resulting from the 1,2-
insertion reaction of PhCCH with 3 (see Figure 5). A 2,1
addition in which PhCCH approaches trans to the acetylide
group is likewise calculated to be higher in activation free energy
than a 2,1 addition approaching trans to the PCP aryl (TS-ins-
6′); in this case, however, the difference is only 1.9 kcal/mol.
The calculated value of 17.3 kcal/mol at -35 °C (20.2 kcal/
mol at 25 °C) for the 1,2-addition (to give 4a), compares well
with the experimentally determined activation free energy of
14.9 kcal/mol at -35 °C. The experimental value for the 2,1-
addition, extrapolated from the overall kinetics of eq 2, is
21.2(1) kcal/mol at 40 °C. If we estimate an activation entropy
of +38 eu for both addition reactions (the value determined as
the thermodynamic entropy of the 1,2-addition), then the
experimental free energy barriers at 25 °C may be estimated as
17.2 kcal/mol (1,2-addition) and 20.6 kcal/mol (2,1-addition).
The difference of 3.4 kcal/mol favoring the 1,2-addition is
exactly equal to the calculated difference (see Figure 15). While
the level of this agreement is certainly fortuitous, the experi-
mental and calculated differences between the two directions
of addition are clearly of the same order of magnitude, while
the absolute calculated values are 3 kcal/mol too high.
In both structures 4a′ and 6′, the PCP ligand is “twisted”
such that the t-butyl groups immediately above and below the
vinyl group maximize their distance from that group. In the
case of 4a′, however, numerous short interatomic distances
between the phenyl group and the PCP and acetylide ligands
are present in spite of this twisting (Figure 17). Furthermore,
the greater twisting of the PCP group in 4a′ versus 6′ results in
short contacts between those t-butyl groups that are on the side
opposite the phenylvinyl group (including the H-H distance
of 1.93 Å indicated in Figure 17).
8b. C-C elimination. The most surprising conclusion of our
experimental studies is that C-C bond elimination from
acetylide-vinyl species is disfavored by the presence of either
electron-donating or -withdrawing substituents on the R-carbon
of the vinyl group. Thus, increased steric bulk apparently
disfaVors elimination. This counterintuitive conclusion is well
supported by calculations. The barrier to elimination from the
R-phenylvinyl complex 4a′ is calculated to be 25.8 kcal/mol.
Although the less crowded â-phenylvinyl complex 6′ is
thermodynamically much more stable (by 16.5 kcal/mol), it has
a much lower calculated barrier to C-C elimination, just 17.9
kcal/mol, and the TS is 1.8 kcal/mol below reactants in free
energy (Figure 15). Also in accord with C-C elimination being
inhibited by crowding, the calculated values for the barrier to
elimination are substantially lower for the less crowded MePCP
analogues: 17.3 kcal/mol for 4′′ and 12.4 kcal/mol for 6′′.
Thus, the R-phenyl species (4′ and 4′′) are higher in energy
than the â-phenyl isomers, as expected on the basis of steric
considerations but, unexpectedly, they also have higher barriers
to elimination. Together these two factors result in very large
differences for the total TS free energies of elimination: 24.4
kcal/mol for the tBuPCP complexes (22.6 kcal/mol (TS-4a′-elim)
- (-1.8) kcal/mol (TS-6′-elim), Figure 15). This overall
difference strongly supports the idea that the observed regiose-
lectivity may be viewed as being controlled by the kinetics of
C-C elimination. For the MePCP complexes, the difference is
considerably smaller, 9.8 kcal/mol, in support of the conclusion
that the difference in activation energies reflects steric effects.
The kinetically more favorable orientation of insertion (1,2)
is calculated to give a product that is thermodynamically less
favorable. The thermodynamic difference between the 2,1- and
1,2-insertion products is calculated to be 4.9 kcal/mol for the
MePCP species and 16.5 kcal/mol for the tBuPCP complex. This
suggests that steric effects play a dominant role in favoring the
thermodynamics of the product with the phenyl group bound
9
J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007 863

A R T I C L E S
Ghosh et al.
Figure 15. Calculated free energy profile (25 °C) for the reaction of (tBuPCP)Ir(H)(CCH) with phenylacetylene. The experimentally observed 2,1-regiochemistry
is in blue and the 1,2-regiochemistry (not observed) is in red. Experimental values (extrapolated to 25 °C) are shown in parentheses. The acetylide group
(but not the incoming phenylacetylene) is truncated and modeled as a simple acetylide.
Figure 16. Calculated free energies for addition of the C-H bond of
acetylene across the triple bond of phenyl acetylene with 2,1- and 1,2-
regiochemistry for a model MePCP ligand. The acetylide group (but not the
incoming phenylacetylene) is truncated and modeled as a simple acetylide.
Figure 18. Calculated transition states (atoms shown at 75% van der Waals
radii) for acetylide/phenylvinyl C-C elimination from (RPCP)Ir(CCH)-
(trans-CHdCHPh) and (RPCP)Ir(CCH)(CPhdCH2) (RPCP ) MePCP and
Figure 17. Calculated structures (atoms shown at 75% van der Waals radii)
tBuPCP). The perspective is approximately as viewed down the acetylide
C-C axis. Schematic structures shown with Ir-C and C-C distances (Å).
for (tBuPCP)Ir(acetylide)(phenylvinyl) isomers 4a′ (R-phenylvinyl) and 6′
(â-phenylvinyl). The perspective is approximately as viewed down the
acetylide C-C axis, and atoms are shown at 75% van der Waals radii.
Particularly short interatomic distances (Å) are indicated. Complex 4a′ is
calculated to be 16.5 kcal/mol above 6′ in free energy.
adducts 4-CO, 5-CO, 6-CO, and 8-CO. The calculated struc-
tures of 4a′ and 6′ (Figure 17) also reveal this structural motif.
In all the calculated TSs for C-C bond elimination (TS-X-
elim), the vinyl groups are rotated approximately perpendicular
to the plane bisecting the P-Ir-P axis, in contrast to the
structures of the energy minima. This rotation presumably allows
the use of the vinyl group π-system for incipient C-C bond
formation. Figure 18 shows the structures of the calculated TSs,
approximately as viewed down the acetylide C-C axis. Clearly,
rotation of the vinyl group engenders substantial steric repulsion,
The DFT calculations help to elucidate a very simple and
compelling explanation for the slower elimination rates from
the R-substituted vinyl complexes. The vinyl groups of all
(
tBuPCP)Ir vinyl complexes discussed in this work are ap-
proximately located in the plane bisecting the (tBuPCP)Ir moiety
(or the plane bisecting the P-Ir-P axis.) This can be seen in
the X-ray structure of complexes 5 and 7a, as well as the CO
9
864 J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007

Dimerization of Alkynes
A R T I C L E S
Scheme 3
neighboring meta-proton (the signal due to one of the meta-
protons overlaps with that of C6D5H and was not used for
fitting). Warming the sample results in line broadening, reaching
coalescence at about 60 °C. With the use of the program
WinDNMR,58 the spectra can be modeled nearly perfectly over
the range 25-75 °C, assuming exchange between the two
protons, which presumably results from rotation about the Ir-C
bond. An Eyring plot of ln(k/T) versus 1/T gives a straight line
(R2 ) 0.995) with slope and intercept of -8343 deg and 23.136,
respectively, which yields ∆Hq ) 16.6 kcal/mol and ∆Sq )
-1.2 eu. The activation free energy barrier to rotation is thus
∆Gq ≈ 17 kcal/mol. Such an energetic contribution (or even a
substantial fraction thereof) to the barrier to C-C elimination
would reduce the overall reaction rate by many orders of
magnitude.
particularly in the R-phenylvinyl species and particularly with
the tBuPCP ligand. For example, a vinyl-H/t-butyl-H distance
of 1.80 Å and a phenyl-C/t-butyl-H distance of 2.45 Å are
calculated for TS-4a′-elim (typical values for van der Waals
radii of H and C are 1.2 and 1.7 Å, respectively55); no such
close contacts are calculated for TS-6′-elim.
Although the vinyl acetylide C-C eliminations ultimately
result in C-C σ-bond formation with no formal change in the
vinyl π-bond, the C-C bond formation in the TS apparently
uses the vinyl group π-system. To our knowledge this is not a
well precedented model for C-C elimination. In the most
comprehensive theoretical study of the reductive elimination
of unsaturated hydrocarbyl groups to date, Musaev and Moro-
kuma invoke the lesser directionality of sp2 versus sp3 carbon
orbitals to explain the greater facility of vinyl versus methyl
elimination (from Pd(II) and Pt(II)).56 Our results indicate that
the presence of a π-system in vinyl ligands, and perhaps the
greater directionality of the carbon p-orbitals, is a major factor
in the present system. Efforts to develop a detailed MO-based
model for the involvement of π-systems in this and other C-C
elimination reactions are currently underway.
Conclusions
A combined experimental/computational study of arylacetyl-
ene dimerization promoted by (PCP)Ir has revealed fundamental
and surprising aspects of (i) the addition of metal-hydrogen
bonds across C-C triple bonds and, in particular, (ii) carbon-
carbon reductive elimination.
(i) Insertion of the triple bond of phenylacetylene into the
Ir-H bond of (tBuPCP)Ir(CCPh)H proceeds readily, and revers-
ibly, even at -35 °C. To our knowledge, this is the first example
of reversible addition of a metal-hydride bond across a C-C
triple bond and, indeed, the first example of â-H elimination
from a metal-vinyl complex to give a nonbridging metal
hydride. This unusual behavior is attributable in large part to
crowding in the vinyl acetylide complex, which favors the
thermodynamics of acetylene deinsertion.
The facile insertion proceeds with the wrong orientation (1,2)
to afford the observed enyne product; the reversibility is
therefore key to the enyne formation, in that the alkyne
ultimately undergoes an insertion with the “correct” (2,1)
orientation.
(ii) The failure of the 1,2-insertion product (i.e., the iridium-
(R-phenylvinyl)(acetylide) complex, 4) to undergo C-C elimi-
nation to give enyne is attributable to steric effects exerted by
the phenyl group at the R position. In support of this conclusion
vinyl-acetylide elimination also fails to occur for analogous
R-methylvinyl (7) and R-vinylvinyl (8) acetylide complexes.
The 2,1-insertion product (i.e., the iridium(â-phenylvinyl)-
(acetylide) complex, 6) does undergo elimination to give the
observed product; this is confirmed by independent synthesis
of 6. Thus, even though elimination alleviates crowding, there
is a greater barrier for elimination of the sterically more hindered
vinyl groups.
We are aware of one example of C-C elimination in which
the ligand π-system appears to play a key role: Morokuma and
co-workers,57 in modeling C-C reductive elimination from
ruthenium in the Murai system, proposed an attack by an alkyl
group on an aryl group. In that case, the aryl group is constrained
by chelation to face the attacking alkyl group. In the present
system, steric effects significantly hinder the ability of the vinyl
group to rotate so as to face the acetylide group. The energetic
cost is prohibitive, when the vinyl group has an R-substituent,
and the feasibility of vinyl-group rotation ultimately appears to
determine the directionality of the overall dimerization reaction.
9. An Experimentally Determined Barrier to Rotation. A
key conclusion from the above computational studies is that
the phosphinoalkyl groups impose a significant barrier to a ca.
90° rotation around the Ir-C(vinyl) bond, which in turn
substantially increases the barrier to C-C bond elimination. As
a means to determine the magnitude of such barriers to rotation
we considered dynamic NMR studies of the complex (tBuPCP)-
Ir(CO)PhH, in which rotation around the Ir-C(phenyl) bond is
a degenerate event. (Note that the five-coordinate species (tBu
-
PCP)IrPhH undergoes facile elimination of benzene, which
precludes its use for this purpose.) In steric terms, the phenyl
group is probably about equivalent to an R-methylvinyl group
(and probably not very different from an R-phenylvinyl group),
as indicated in Scheme 3.
In the low-temperature limit (ca. ambient temperature or
below) the ortho protons of the (tBuPCP)Ir(CO)PhH phenyl group
appear as two signals, each one a doublet split by the respective
DFT calculations allow the elucidation of a clear explanation
for the inhibition of elimination owing to the R-vinyl substit-
uents. In the energy minima, the vinyl groups rest approximately
in the plane perpendicular to the P-P axis. In the elimination
TS’s however, the vinyl groups are rotated by 90° to allow an
attack of the acetylide group on the vinyl π-system. This rotation
is substantially inhibited by the presence of R-vinyl substituents.
We believe that the observation of steric inhibition of C-C
bond elimination, and the conclusion that vinyl groups must
orient in a specific direction to undergo C-C bond formation,
(55) Bondi, A. J. Phys. Chem. 1964, 68, 441-451.
(56) Ananikov, V. P.; Musaev, D. G.; Morokuma, K. Organometallics 2005,
24, 715-723.
(57) (a) Matsubara, T.; Koga, N.; Musaev, D. G.; Morokuma, K. J. Am. Chem.
Soc. 1998, 120, 12692-12693. (b) Matsubara, T.; Koga, N.; Musaev, D.
G.; Morokuma, K. Organometallics 2000, 19, 2318-2329.
(58) Reich, H. J. J. Chem. Educ.: Software, Ser. D 1996, 3D, 17-19.; 27-33.;
56-66.
9
J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007 865

A R T I C L E S
Ghosh et al.
normal-mode analysis. The (unscaled) vibrational frequencies formed
the basis for the calculation of vibrational zero-point energy (ZPE)
corrections. Standard thermodynamic corrections (based on the har-
monic oscillator/rigid rotor approximations and ideal gas behavior) were
made to convert from purely electronic reaction or activation energies
(∆E, ∆Eq; no ∆ZPE) to (standard) enthalpies (∆H, ∆Hq; ∆ZPE
included) and free energies (∆G, ∆Gq; T ) 298 K, P ) 1 atm).67 All
calculations were executed using the Gaussian 03 series of computer
programs.68
may have significant implications for the understanding and
design of other C-C bond formation systems.
General Experimental Methods and Computational
Details
General Experimental Methods. All reactions, recrystallizations,
and routine manipulations were performed at ambient temperature in
an argon-filled glove box, or under argon using standard Schlenk
techniques. Benzene and p-xylene were distilled from sodium/potassium
alloy and vacuum transferred under argon. Deuterated solvents for use
in NMR experiments were dried as their protiated analogues. (PCP)-
IrH2 was synthesized according to published methods.59 All other
We also carried out a large number of calculations on the acetylene
dimerization reaction profile with a variety of two-layer ONIOM-type
approaches (DFT/MM; DFT/HF; DFT/DFT) in which the four t-Bu
groups were treated as the “low level” layer.69 The results of these
calculations will not be discussed in detail here, but we did make some
comparisons to the experimental kinetic and thermodynamic observa-
tions described above and found that the ONIOM-type approaches
consistently overestimated the steric repulsions exerted by the t-Bu
groups. In particular, the largest disagreements with the experimental
data were shown by the computationally most efficient ONIOM
approach in which the t-Bu groups were represented using a molecular
mechanics force field (UFF).70 As the sophistication in the treatment
of the “low level” layer increased, for example from replacing the UFF
treatment by a small basis set ab initio Hartree-Fock or DFT treatment,
the strong overestimation of steric effects attenuates slowly. As the
computational treatment of the t-Bu groups approaches the high level
applied to the remainder of the (PCP)Ir-acetylene species, the agreement
between computed and experimentally derived data gradually improves;
regular all-electron DFT-type calculations on the entire molecular
species clearly provide the best description of the chemistry observed
here. We have actually computed most species with both B3LYP54 and
PBE functionals, but results presented in the discussion, unless noted
otherwise, were obtained using the PBE functionals which uniformly
gave results in much better agreement with experiment.
1
chemicals were used as received from commercial suppliers. H and
31P{1H} NMR spectra were obtained on a 400-MHz, Varian Inova-
400 spectrometer or on a 300-MHz, Varian Mercury-300 spectrometer.
1H chemical shifts are reported in ppm downfield from tetramethylsilane
and were referenced to residual protiated (1H) or deuterated solvent.
31P NMR chemical shifts were referenced to 85% H3PO4. Mesitylene
was employed as an internal standard when yields were determined by
NMR spectroscopy. (E)-1,4 diphenyl-1-buten-3-yne was independently
synthesized by a literature procedure.60 Experimental details and spec-
troscopic characterizations are present in the Supporting Information.
Computational Details. The electronic structure calculations em-
ployed the DFT method61 with the PBE53 exchange and correlation
functionals. The relativistic, small-core ECP and corresponding basis
sets (6s5p3d) of Dolg et al. were used for the Ir atom (SDD model).62
We used all-electron, full double-ú plus polarization function D95(d)
basis sets63 for the C atoms in the acetylene adducts. Carbon atoms in
the PCP unit were also covered with the D95(d) basis set in calculations
where the representation of the PCP species was truncated and placed
methyl groups on the P atoms (i.e., PR2 ) PMe2). Most calculations
were made modeling the PCP species with the t-Bu groups actually
used in the experiments, PR2 ) P(t-Bu)2. The C atoms (and their
attached H atoms) in the t-Bu groups were then described by the
minimal STO-3G basis set.64 Regular hydrogen atoms carried a double-ú
quality 21G basis set,65 whereas acetylenic hydrogen atoms, which
formally become hydrides in the metal complexes, were described by
the triple-ú plus polarization 311G(p) basis set.66
Acknowledgment. Financial support by the National Science
Foundation (Grant CHE-0316575, and Grant CHE-0091872 for
purchase of the Bruker SMART APEX diffractometer) is
gratefully acknowledged.
Reactant, transition state, and product geometries were fully
optimized, and the stationary points were characterized further by
Supporting Information Available: Crystallographic data for
complexes 1, 3, 4-CO, 5-CO, 5, 6-Ar-CO, 7a, 8-CO, and
(PCP)Ir(CO) cocrystallized with 2; experimental details and
spectroscopic characterization of all complexes; complete ref
68. This material is available free of charge via the Internet at
http://pubs.acs.org.
(59) (a) Gupta, M.; Hagen, C.; Flesher, R. J.; Kaska, W. C.; Jensen, C. M. Chem.
Commun. 1996, 2083-2084. (b) Zhu, K.; Achord, P. D.; Zhang, X.; Krogh-
Jespersen, K.; Goldman, A. S. J. Am. Chem. Soc. 2004, 126, 13044-13053.
(c) Goldman, A. S.; Ghosh, R. In Handbook of C-H Transformations -
Applications in Organic Synthesis; Dyker, G., Ed.; Wiley-VCH: New York,
2005; pp 616-621.
(60) Moskalev, N. V.; Timoshchenko, L. V.; Filimonov, V. D. Zh. Org. Khim.
1991, 27, 2233.
(61) Parr, R. G.; Yang, W. Density-Functional Theory of Atoms and Molecules;
University Press: Oxford, 1989.
JA0647194
(62) Dolg, M.; Wedig, U.; Stoll, H.; Preuss, H. J. Chem. Phys. 1987, 86, 866.
(63) Dunning, T. H.; Hay, P. J. Modern Theoretical Chemistry. In Modern
Theoretical Chemistry; Schaefer, H. F., III, Ed.; Plenum: New York, 1976;
pp 1-28.
(67) McQuarrie, D. A. Statistical Thermodynamics; Harper and Row: New York,
1973.
(68) Frisch, M. J.; et al. Gaussian 03, revision B.02; Gaussian, Inc.: Wallingford,
CT, 2004.
(69) Wu¨llen, C. J. v. J. Comput. Chem. 1997, 18, 1985-1992.
(70) Rappe, A. K.; Casewit, C. J.; Colwell, K. S.; Goddard, W. A., III; Skiff,
W. M. J. Am. Chem. Soc. 1992, 114, 10024-10035.
(64) Hehre, W. J.; Stewart, R. F.; Pople, J. A. J. Chem. Phys. 1969, 51, 2657.
(65) Binkley, J. S.; Pople, J. A.; Hehre, W. J. J. Am. Chem. Soc. 1980, 102,
939.
(66) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem. Phys. 1980,
72, 650.
9
866 J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007