Mechanistic studies of the transfer dehydrogenation of cyclooctane catalyzed by iridium bis(phosphinite) p-XPCP pincer complexes

Article abstract of DOI:10.1021/ja048393f

Reaction of bis(phosphinite) PCP iridium pincer complexes (p-XPCP)IrHCl (5a-f) [X = MeO (5a), Me (5b), H (5c), F (5d), C₆F₅ (5e), Ar^F(= 3,5-bis(trifluoromethyl)phenyl) (5f)] with NaOtBu in neat cyclooctane (COA) generates

Full text of DOI:10.1021/ja048393f

Published on Web 07/13/2004
Mechanistic Studies of the Transfer Dehydrogenation of
Cyclooctane Catalyzed by Iridium Bis(phosphinite) p-XPCP
Pincer Complexes
Inigo Go¨ttker-Schnetmann and Maurice Brookhart*
Contribution from the Department of Chemistry, UniVersity of North Carolina at Chapel Hill,
Chapel Hill, North Carolina 27599-3290
Received March 21, 2004; E-mail: mbrokkhart@unc.edu
Abstract: Reaction of bis(phosphinite) PCP iridium pincer complexes (p-XPCP)IrHCl (5a-f) [X ) MeO
(5a), Me (5b), H (5c), F (5d), C6F5 (5e), ArF() 3,5-bis(trifluoromethyl)phenyl) (5f)] with NaOtBu in neat
cyclooctane (COA) generates 1:1 mixtures of the respective (p-XPCP)IrH2 complexes 4a-f and the
cyclooctene (COE) olefin complexes (p-XPCP)Ir(COE) (6a-f) at 23 °C. At higher temperatures, complexes
4 and 6 are equilibrated because of the degenerate transfer dehydrogenation of COA with free COE (6 +
COA h 4 + 2COE), as was shown by temperature-dependent equilibrium constants and spin saturation
transfer experiments at 80 °C. At this temperature, the COE complexes 6 exchange with free COE on the
NMR time scale with the more electron-deficient complexes 6 exchanging COE faster. The exchange is
dissociative and zero order in [COE]. Further analysis reveals that the stoichiometric hydrogenation of
COE by complex 4f, and thus the separated back reaction 4f + 2COE f 6f + COA proceeds at temperatures
as low as -100 °C with the intermediacy of two isomeric complexes (p-ArFPCP)Ir(H)2(COE) (8f, 8f′). COE
deuteration with the perdeuterated complex 4f-d38 at -100 °C results in hydrogen incorporation into the
hydridic sites of complexes 8f,8f′-d38 but not in the hydridic sites of complex 4f-d38, thus rendering COE
migratory insertion in complexes 8f,8f′ reversible and COE coordination by complex 4f rate-determining
for the overall COE deuteration.
thermocatalytic4a,b,5 dehydrogenation of alkanes, though indus-
trial applications are far from realized. As the reverse of olefin
Introduction
Invention of catalytic processes for conversion of abundant
alkane feedstocks to higher value products based on transition-
metal-mediated C-H bond activations remains a challenging
task.1 Heterogeneous alkane dehydrogenation processes such
as (hydro)cracking or catalytic reforming are the most important
industrial source of olefins and arenes.2 Development of
analogous homogeneous processes, due to drastically decreased
reaction temperatures (typically heterogeneous processes operate
at 400-600 °C), is expected to provide higher selectivities and
higher energy efficiency. In addition, such studies might lead
to a better understanding of the requirements for efficient alkane
C-H functionalization.
hydrogenation, homogeneous hydrogenation catalysts in prin-
ciple should accomplish the alkane dehydrogenation.3a,6 The
endothermicity of this reaction however,7 requires very high-
temperature conditions (>150 °C), which are often not compat-
ible with the thermal stability of potential catalysts. Partly
because of their improved thermal stability, bis(phosphine) and
bis(phosphinite) PCP iridium pincer complexes 1-4 (Chart 1)
have been shown to be the most active homogeneous catalysts
(3) (a) Crabtree, R. H.; Mihelcic, J. M.; Quirk, J. M. J. Am. Chem. Soc. 1979,
101, 7738-7740. (b) Baudry, M. J.; Ephritikine, M.; Felkin, H.; Holmes-
Smith, R. J. Chem. Soc., Chem. Commun. 1983, 788-789. (c) Baudry, M.
J.; Crabtree, R. H.; Parnell, C. P.; Uriarte, R. J. Organometallics 1984, 3,
816-817.
(4) (a) Burk, M. J.; Crabtree, R. H.; McGrath, D. V. J. Chem. Soc., Chem.
Commun. 1985, 1829-1830. (b) Burk, M. J.; Crabtree, R. H. J. Am. Chem.
Soc. 1987, 109, 8025-8032. (c) Nomura, K.; Saito, Y. J. Chem. Soc., Chem.
Commun. 1988, 161-162. (d) Sakakura, T.; Sodeyama, T.; Tokunaga, M.;
Tanaka, M. Chem. Lett. 1988, 263-264. (e) Maguire, J. A.; Boese, W. T.;
Goldman, A. S. J. Am. Chem. Soc. 1989, 111, 7088-7093. (f) Sakakura,
T.; Sodeyama, T.; Tanaka, M. New J. Chem. 1989, 13, 737-745. (g)
Sakakura, T.; Abe, F.; Tanaka, M. Chem. Lett. 1991, 297-298. (h) Choi,
J.-C.; Kobayashi, Y.; Sakakura, T. J. Org. Chem. 2001, 66, 5262-5263.
(5) (a) Maguire, J. A.; Goldman, A. S. J. Am. Chem. Soc. 1991, 113, 6706-
6708. (b) Maguire, J. A.; Petrillo, A.; Goldman, A. S. J. Am. Chem. Soc.
1992, 114, 9492-9498. (c) Belli, J.; Jensen, C. M. Organometallics 1996,
15, 1532-1534.
Since the first examples of homogeneous alkane dehydroge-
nation stoichiometric and substoichiometric in transition metal,3
much progress has been made providing effective photo-4 and
(1) For recent reviews on this topic, see: (a) Kakiuchi, F.; Chatani, N. AdV.
Synth. Catal. 2003, 345, 1077-1101. (b) Labinger, J. A.; Bercaw, J. E.
Nature 2002, 417, 507-514. (c) Crabtree, R. H. J. Chem. Soc., Dalton
Trans. 2001, 2437-2450. (d) Jia, C.; Kitamura, T.; Fujiwara, Y. Acc. Chem.
Res. 2001, 34, 633-639. (e) Jensen, C. M. Chem. Commun. 1999, 2443-
2449. (f) Sen, A. Acc. Chem. Res. 1998, 31, 550-557. (g) Shilov, A. E.;
Shul’pin, G. B. Chem. ReV. 1997, 97, 2879-2932. For leading reviews on
the functionalization of sp- and sp2 C-H bonds, see: (h) Ritleng, V.; Sirlin,
C.; Pfeffer, M. Chem. ReV. 2002, 102, 1731-1769. (i) Guari, Y.; Sabo-
Etienne, S. Eur. J. Inorg. Chem. 1999, 1047-1055. (j) Dyker, G. Angew.
Chem., Int. Ed. 1999, 38, 1699-1712.
(6) (a) Crabtree, R. H. Chem. ReV. 1985, 85, 245-269. (b) Crabtree, R. H.
Chem. ReV. 1995, 95, 987-1007.
(7) The dehydrogenation enthalpy of COA (23.3 kcal/mol) is among the lowest
known for alkanes, which makes it an even more suitable substrate for
alkane dehydrogenation. NIST Chemistry WebBook, NIST Standard
Reference Database Number 69. http://webbook.nist.gov/chemistry/ (re-
leased March, 2003).
(2) (a) Wieseman, P. Petrochemicals; Ellis Horwood: Chichester, England
1986; pp 90-91. (b) Weissermel, K.; Arpel, H.-J. Industrial Organic
Chemistry, Wiley-VCH: Weinheim, Germany, 2003; pp 59-89.
9
9330
J. AM. CHEM. SOC. 2004, 126, 9330-9338
10.1021/ja048393f CCC: $27.50 © 2004 American Chemical Society

Transfer Dehydrogenation of Cyclooctane
A R T I C L E S
Chart 1. Active PCP Iridium Pincer Catalysts for the Transfer Dehydrogenation of COA with TBE
for the thermochemical transfer dehydrogenation of, for ex-
ample, cyclooctane (COA) in the presence of tert-butylethylene
(TBE), as shown in eq 1.1e,8-11
concentrations. Furthermore, catalytically active species can be
easily generated in situ by addition of NaOtBu to a solution of
the respective air stable hydridochloro precursor (p-XPCP)IrHCl
(5a-f) in 1:1 mixtures of COA and TBE.12 We now wish to
report the nature of the catalytically active species generated
from the hydridochloro precursors 5a-f as well as mechanistic
results with relevance to the catalytic transfer dehydrogenation
of COA.
Results and Discussion
Oxidative Addition/C-H Activation of COA by the
Fragment (p-XPCP)Ir (A). In recent studies, we have shown
that C-H oxidative addition of toluene and other arenes can
be easily accomplished by reaction of hydridochloro complexes
5a-f with NaOtBu in the arene solvent.13 Furthermore, a
dissociative exchange of arene from the formed (p-XPCP)Ir-
(H)(aryl) complexes comprising a reductive elimination/oxida-
tive addition sequence through the (presumed) intermediacy of
a 14e- species (p-XPCP)Ir (Aa-f) was established by line-
broadening techniques at temperatures above -30 °C.13 Since
alkane C-H activation is one prerequisite step in the transfer
dehydrogenation of COA with tert-butylethylene (TBE) cata-
lyzed by complexes 4a-f, or by 5a-f in the presence of
NaOtBu, we have chosen to examine the stoichiometric reaction
of 5a-f, NaOtBu, and COA. Thus, reaction of complexes 5a-f
with 1.1 equiv of NaOtBu in neat COA forms within ca. 30
min at 23 °C the respective (p-XPCP)IrH2 complexes 4a-f as
was proven by matching the 31P{1H} NMR signals and the
hydridic IrH2 resonances in neat COA to independently prepared
complexes.13 A second Ir complex was formed in each case
with a characteristic 31P{1H} NMR resonance in the range
between 173 and 180 ppm. Additionally, for these species a
second set of signals for the ligand backbone of each p-XPCP
ligand as well as a broad multiplet in the narrow range of 4.2-
4.3 ppm could be detected in the 1H NMR spectra. The ratio of
the (p-XPCP)IrH2 (4a-f) and the new complexes is 1:1, as
determined by integration of the corresponding signals in the
1H NMR spectrum as well as integration of the 31P{1H} NMR
signals. No other complexes are formed in these reactions. We
have assigned the new Ir complexes as (p-XPCP)Ir(COE) (6a-
Initial turnover frequencies (TOF) of 0.2 s-1 and turnover
numbers (TON) up to 1000 have been observed with catalyst 1
at 200 °C. Inhibition of catalysis using 1 by TBE and the formed
cyclooctene (COE) was shown to restrict higher TON and
required the portionwise addition of TBE to obtain the reported
TON.8 Mechanistic studies by Goldman et al. reveal a first-
order dependence of this reaction in TBE at low [TBE], which
becomes inverse first-order at high [TBE] because of the pre-
equilibrium formation of the Ir(III) complex (PCP)Ir(H)(tert-
butylvinyl), the catalyst resting state under these conditions.11a
Furthermore, under conditions where [TBE] < 0.3 M, hydro-
genation of TBE, and more specifically reductive elimination
of TBA from the (PCP)Ir(H)(neo-hexyl) complex, was found
to be the rate-determining step for the overall reaction in the
same study. Recently, we have shown that complexes 4a-f
exhibit a catalytic activity for the reaction of eq 1 up to 1 order
of magnitude higher than that of complex 1 under standard
conditions without the need of maintaining relatively low TBE
(8) Gupta, M.; Hagen, C.; Flesher, R. J.; Kaska, W. C.; Jensen, C. M. Chem.
Commun. 1996, 2083-2084.
(9) (a) Xu, W.-W.; Rosini, G. P.; Gupta, M.; Jensen, C. M.; Kaska, W. C.;
Krogh-Jespersen, K.; Goldman, A. S. Chem. Commun. 1997, 2273-2274.
(b) Liu, F.; Goldman, A. S. Chem. Commun. 1999, 655-656.
(10) Haenel, M. W.; Oevers, S.; Angermund, K.; Kaska, W. C.; Fan, H.-J.; Hall,
M. B. Angew. Chem., Int. Ed. 2001, 40, 3596-3600.
(11) (a) Renkema, K. B.; Kissin, Y. V.; Goldman, A. S. J. Am. Chem. Soc.
2003, 125, 7770-7771. (b) Krogh-Jespersen, K.; Czerw, M.; Summa, N.;
Renkema, K. B.; Achord, P. D.; Goldman, A. S. J. Am. Chem. Soc. 2002,
124, 11404-11416. (c) Krogh-Jespersen, K.; Czerw, M.; Zhu, K.; Singh,
B.; Kanzelberger, M.; Darji, N.; Achord, P. D.; Renkema, K. B.; Goldman,
A. S. J. Am. Chem. Soc. 2002, 124, 10797-10809. (d) Li, S.; Hall, M. B.
Organometallics 2001, 20, 2153-2160. (e) Morales-Morales, D.; Lee, D.
W.; Wang, Z.; Jensen, C. M. Organometallics 2001, 20, 1144-1147. (f)
Krogh-Jespersen, K.; Czerw, M.; Kanzelberger, M.; Goldman, A. S. J.
Chem. Inf. Comput. Sci. 2001, 41, 56-63. (g) Kanzelberger, M.; Singh,
B.; Czerw, M.; Krogh-Jespersen, K.; Goldman, A. S. J. Am. Chem. Soc.
2000, 122, 11017-11018. (h) Liu, F.; Pak, E. B.; Singh, B.; Jensen, C.
M.; Goldman, A. S. J. Am. Chem. Soc. 1999, 121, 4086-4087. (i) Niu, S.;
Hall, M. B. J. Am. Chem. Soc. 1999, 121, 3992-3999. (j) Liu, F.; Goldman,
A. S. Chem. Commun. 1999, 655-656. (k) Lee, D. W.; Kaska, W. C.;
Jensen, C. M. Organometallics 1998, 17, 1-3. (l) Gupta, M.; Hagen, C.;
Kaska, C. W.; Cramer, R. E.; Jensen, C. M. J. Am. Chem. Soc. 1997, 119,
840-841.
1
f) on the basis of matching 31P{1H} and H NMR resonances
to those species formed by reacting (p-HPCP)IrH2 (4c) or (p-
ArFPCP)IrH2 (4f) with 2.5 equiv of COE in neat COA. We have
been unable to isolate complexes 6c,f in a pure form. Generation
of complexes 4a-f and 6a-f from the hydridochloro precursors
5a-f and NaOtBu in COA can be rationalized by dehydro-
(12) Go¨ttker-Schnetmann, I.; White, P. S.; Brookhart, M. J. Am. Chem. Soc.
2004, 126, 1804-1811.
(13) Go¨ttker-Schnetmann, I.; White, P. S.; Brookhart, M. Organometallics 2004,
23, 1766-1776.
9
J. AM. CHEM. SOC. VOL. 126, NO. 30, 2004 9331

A R T I C L E S
Go¨ttker-Schnetmann and Brookhart
Scheme 1. Generation of (p-XPCP)IrH2 (4) and (p-XPCP)Ir(COE) (6) by Dehydrochlorination of Complexes 5 with NaOtBu in Neat COA
Scheme 2. Equilibrium Reaction of Complexes 4 and 6 Generated from Complexes 5
Table 1. 31P NMR Spin Saturation Transfer Data for the Equilibrium 6 + COA h 4 + 2COE at 80 °C
d
e
I
35 dB (4)a
(%)
I
35 dB (6)b
(%)
T (4)c
(s)
T (6)c
(s)
R
R
[4/6] (from K
eq
1
1
obs1
(s-1
obs-1
4,6
)
(s-1
)
R
obs1/R
measurement)
obs-1
a
b
c
f
84.5
64.6
47.8
51.5
74.5
52.5
39.2
42.8
2.1
2.2
2.7
1.6
1.8
2.0
2.3
1.5
0.19
0.45
0.67
0.89
0.09
0.25
0.40
0.59
2.18
1.82
1.67
1.51
2.23
1.92
1.78
1.62
a
b
Relative integration of the 31P NMR signal of 4 obtained for complete versus no saturation of 6. Relative integration of the 31P NMR signal of 6
c
obtained from complete versus no saturation of 4. Obtained by the inversion recovery method. Based on the observed T1 values, the relaxation delay in
each SST experiment was uniformly set to 15 s. Rate for the conversion of 6 to 4. Rate for the conversion of 4 to 6.
d
e
chlorination of 5a-f to give an undetected 14e- fragment (p-
XPCP)Ir (Aa-f). Oxidative addition of COA to Aa-f followed
by â-hydride elimination generates compounds 4a-f and 1
equiv of COE, which traps another 14e- fragment (p-XPCP)Ir
to yield 6a-f (Scheme 1).
free COE and the iridium(III) complexes 4 from their iridium-
(I) precursors 6 and COA.
The 31P NMR resonances of 6 and 4 are sharp in the range
of 25-80 °C and give no indication of the interchange of 6
and 4 as shown in Scheme 2. However, 31P NMR spin saturation
transfer (SST) experiments at 80 °C provide clear evidence for
the equilibrium as well as quantitative rates for the interchange
of the 6a-c,f and 4a-c,f pairs. In a typical set of experiments,
a 1:1 mixture of complexes 4f and 6f, obtained by reaction of
20 µmol complex 5f and 1.1 equiv of NaOtBu in 300 mg of
COA in a J. Young NMR tube at 23 °C, was heated to 80 °C,
while equilibration resulted in generation of free COE and
increase of the 4f/6f ratio. SST experiments on the well-
separated (by ca. 30 ppm) 31P NMR resonances were then
carried out. Irradiation and saturation of the resonance for 4f
resulted in a decrease of the net magnetization of the 31P signal
for 6f, while saturation of the signal for 6f resulted in decrease
of the integral for 4f. These data, together with T1 measurements
for both nuclei, are summarized in Table 1 and can be used to
calculate the rate of formation (not rate constants) of 4f from
6f (Robs1) as well as the reverse rate, formation of 6f from 4f
(Robs-1). Similar data for the 4a-c/6a-c pairs are also sum-
marized in Table 1.
Significantly, we find here that the methylene-bridged pincer
complex {C6H3-2,6-[CH2P(tBu)2]2}IrHCl14 undergoes the same
reaction with NaOtBu in neat COA, but the ratio of the
generated complexes {C6H3-2,6-[CH2P(tBu)2]2}IrH2 (1):{C6H3-
2,6-[CH2P(tBu)2]2}Ir(COE) is ca. 8-9:1 at 23 °C, and free COE
is detected by 1H NMR experiments. Furthermore, we find that
the ratio of each 4a-c,f/6a-c,f pair gradually increases as
temperature is raised with concomitant liberation of free COE.15
More electron-rich ligand backbones p-XPCP result in larger
4/6 ratios (and larger amounts of free COE) at a given
temperature, for example: 4a/6a ) 2.23, 4b/6b ) 1.92, 4c/6c
) 1.78, and 4f/6f ) 1.62 at 80 °C. The ratios 4a-c,f/6a-c,f
return to 1:1 again when the temperature is lowered back to 23
°C. These observations strongly suggest that an equilibrium is
established between complexes 4 and 6 by the reaction
(p-XPCP)Ir(COE) (6) + COA h (p-XPCP)IrH2 (4) + 2COE
(Scheme 2). It is noteworthy that more electron-rich ligand
backbones p-XPCP thermodynamically favor the formation of
As a check on the accuracy of the SST experiments, the ratio
of Robs/Robs-1 should correspond to the independently measured
equilibrium ratio of the 4/6 pairs. These ratios and the Keq values
(14) Moulton, C. J.; Shaw, B. L. J. Chem. Soc., Dalton Trans. 1976, 1020-
1024.
(15) The olefinic signal at 5.6 ppm (calibrated to the COA signal at 1.57 ppm)
is dynamically broadened.
9
9332 J. AM. CHEM. SOC. VOL. 126, NO. 30, 2004

Transfer Dehydrogenation of Cyclooctane
A R T I C L E S
oxidative addition) in the respective (p-XPCP)Ir(H)(aryl) com-
plexes,13 exchange of COE in complexes 6, as judged by the
signal width “∆ν1/2*” of the olefinic resonances, qualitatively
proceeds faster for the more electron-deficient (p-XPCP)Ir
fragments (Aa-f). This is also in accordance with the quantita-
tive SST experiments that revealed that electron-deficient
p-XPCP ligand backbones result in enhanced rates for the
transformation of 6 to 4 (vide supra). A faster dissociative ligand
exchange with increasing electron deficiency at the metal center
is not unprecedented.17
The equilibria of Scheme 2 constitute a full catalytic cycle
of a putative transfer dehydrogenation, since the COE/COA
couple can be considered as a degenerate hydrogen acceptor/
donor system. At “low” temperature (23 °C) and “low” [COE],
both complexes 4 and 6 are resting states of the catalytic cycle,
while at “high” [COE] (i.e., in the presence of an additional
1.2 equiv of COE with respect to the catalyst) the sole resting
state is COE complex 6 (Figure 1, 23 °C spectrum). Increasing
temperature, however, entropically favors the formation of three
species (4 and two COE molecules) from two species (6 and
COA) and shifts the equilibrium of Scheme 2 to the right (Figure
1, 70 °C spectrum). Pre-equilibrium COE dissociation from the
resting state 6 presumably generates the 14e- species A, which
is capable of COA C-H oxidative addition and subsequent
â-hydride elimination to generate dihydride 4 and COE.
Stoichiometric COE Hydrogenation by Complex 4f. To
gain a better understanding of the equilibrium described in
Scheme 2, we have studied the stoichiometric reaction of
dihydride 4f with COE to generate COA and 6f (and thus the
separated back reaction of Scheme 2). While similar studies
have been conducted for the hydrogenation of TBE with
methylene-bridged complex 1 at 55 °C to obtain kinetic data,11a
reaction of complex 4f with COE (11-37 equiv) in neat COA
solvent is instantaneous at 23 °C and produces complex 6f as
the sole product. Low-temperature hydrogenation of COE was
therefore conducted in toluene-d8 and, surprisingly, proceeds
too fast for NMR monitoring even at -40 °C. At -90 °C,
however, it was possible to follow the consumption of 4f in
the presence of 11 equiv of COE over the course of ca. 3 h (ca.
4 to 5 half-lives) by 1H and 31P{1H} NMR experiments.
Significantly, the final product 6f is not formed at -90 °C, but
the buildup of two new species with characteristic resonances
at δ ) 181.5 and 167.7 in the 31P{1H} NMR and at -9.86,
Figure 1. Temperature-dependent equilibrium 6c + COA h 4c + 2COE
in the aromatic and olefinic region of the 1H NMR spectrum. Sample
obtained from reaction of 5c and ca. 2.2 equiv of COE in neat COA.
are summarized in the last two columns and are clearly in close
agreement. It is significant to note that the rates of interchange
of species 4 and 6 increase as the electron-withdrawing ability
of para-substituent increases. For example, the rate of conversion
of 6f to 4f is 0.89 s-1 compared to 0.19 s-1 for conversion of
6a to 4a. The reverse rate is similarly accelerated, with the rate
of 4f to 6f equal to 0.59 s-1 compared to 0.09 s-1 for 4a to 6a.
Additional details concerning the 4/6 equilibration were
revealed by reacting (p-HPCP)IrHCl (5c) with 1.1 equiv of
NaOtBu in neat COA in the presence of ca. 2.2 equiv of COE.
Quantitative conversion of 5c into 6c is complete within ca. 40
min at 23 °C, and a ca. 1:1.2 mixture of 6c/COE (sharp multiplet
at 5.61 ppm) results. Heating this reaction mixture to 70 °C
causes the appearance of the iridium dihydride complex 4c (ca
5% with respect to 6c). In the same experiment, increasing line-
broadening of the signals for free COE at 5.61 ppm and
coordinated COE (slightly shifted downfield by ca. 0.05 ppm)
can be observed upon heating the sample gradually to 70 °C
(Figure 1).
Upon addition of COE (12, 41 equiv), the disappearance of
4c at 70 °C is observed. Furthermore, we find that the olefinic
signal for free COE becomes increasingly sharp with COE
addition, while the width at half-height (∆ν*1/2) of the olefinic
signal of complex 6c remains at ca. 25.0 Hz. Similar results
have been obtained for the olefinic resonance of complexes 6a,f.
At 75 °C, ∆ν*1/2(6a) ) ca. 19 ( 1 Hz, ∆ν*1/2(6c) ) ca. 34 (
1 Hz, and ∆ν*1/2(6f) ) ca. 48( 1 Hz in the presence of 2-40
equiv of free COE. Upon heating a solution of 5 equiv of free
COE and 6f in neat COA from 70 to 100 °C, the olefinic COE
signal of 6f gradually broadens into the baseline while the signal
for free COE reaches ∆ν*1/2 ) ca. 72 Hz at δ ) 5.55 ppm
(instead of 5.61 ppm at 23 °C). Further heating to 120 °C results
in a diminished signal width ∆ν*1/2 ) ca. 64 Hz and δ ) 5.45
ppm (the weighted average position for fast exchange is 5.41
ppm). The lack of dependence of line-broadening of the olefinic
resonances of 6a,c,f on [COE] clearly indicates the exchange
is zero-order in [COE] with COE dissociation from complexes
6 as the rate-limiting step.16 As was quantitatively shown for
the dissociative arene exchange (via reductive elimination/
1
3.61, and 2.95 ppm (in a 1:1:1 ratio) in the H NMR spectra
can be observed. While the 31P NMR resonance at 181.5 ppm
at -90 °C matches that of (p-ArFPCP)Ir(D)(tolyl-d7) (7f),13 the
31P NMR resonance at 167.7 ppm and the 1H NMR resonances
are indicative of a six-coordinate Ir complex containing a
coordinated COE double bond (3.61 ppm), the allylic signals
of coordinated COE (2.95 ppm), and two averaged hydride
ligands (-9.86 ppm). Over time, a minor isomer characterized
by a hydridic resonance at -9.41 ppm can also be observed by
1H NMR experiments. However, disappearance of these isomers
and further buildup of 7f can be monitored by 31P{1H} and 2D
NMR (characteristic resonance at -43.1 ppm) at -90 °C. We
tentatively assign the aforementioned Ir(H2)(COE) complexes
(16) In principle, the dissociation of COE could be induced by association of
COA C-H bonds which are present in ca. 100-10 000-fold excess with
respect to the olefinic CdC bond of free COE in this study.
(17) Zhong, H. A.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 2002, 124,
1378-1399.
9
J. AM. CHEM. SOC. VOL. 126, NO. 30, 2004 9333

A R T I C L E S
Go¨ttker-Schnetmann and Brookhart
Scheme 3. COE Hydrogenation in Toluene-d8 at -90 °C and Formation of COE Complex 6f at -40 °C
as cis- and trans-isomers of (p-ArFPCP)Ir(H)2(COE) 8f,8f′,18
which upon insertion of coordinated COE into one Ir-H bond
generates the (p-ArFPCP)Ir(H)(cyclooctyl) complex Bf. Reduc-
tive elimination of COA from Bf generates the 14e- species
Af, which C-D activates the toluene-d8 solvent and results in
formation of (p-ArFPCP)Ir(D)(tol-d7) (7f) (Scheme 3).
When complex 4f is treated with COE in methylcyclohexane-
d14 (in place of toluene-d8) at -90 °C, formation of both
isomeric complexes 8f,8f′ and the final product 6f is observed
by 1H NMR experiments. By warming samples of 4f and COE
from -110 °C to -70 °C in the NMR spectrometer, there is a
small buildup of complexes 8f,8f′ of up to 10% (10.5 equiv of
COE), 12% (14 equiv of COE), and 15% (20.5 equiv of COE)
during the heating period. At -70 °C, species 8f,8f′ are
consumed faster than they are generated to leave 6f as the sole
product. The half-life of dihydride complex 4f in these reactions
was determined to be ca. 17 min (10.5 equiv of COE) and ca.
9 min (20.5 equiv of COE) at -70 °C. Finally, when the
perdeutero complex (p-ArFPCP)IrD2 (4f-d38)21 in methylcyclo-
hexane-d14 is treated at -100 °C with 35 equiv of COE
(conditions chosen to build up complexes 8f-d38,8f′-d38 in
substantial amounts), an exchange of Ir-deuteride by hydride is
observed in complexes 8f-d38,8f′-d38. On the basis of the
integrals for both the Ir-H and olefinic resonances of 8f,8f′,
H/D exchange is not complete prior to formation of 6f. These
experiments indicate that the insertion of coordinated COE into
one of the Ir-D bonds of 8f-d38,8f′-d38 and â-hydride (â-
deuteride) elimination in Bf are reversible and competitive with
reductive elimination of COA from the unobserved (p-ArFPCP)-
Ir(D)(cyclooctyl-d1) complex (Bf-d38). Remarkably, no protio
label is transferred into the hydridic sites of the starting (p-ArF-
PCP)IrD2 complex 4f-d38 at -100 °C (Scheme 4). Thus, COE
coordination to 4f-d38 is irreversible. This step must be the rate-
limiting step for overall COE hydrogenation at low [COE] where
little 8f,8f′ grow in during reaction. Figure 2 shows a qualitative
free-energy diagram for the overall reaction. To accommodate
the H/D exchange results, transition-state 1 (TS1) must be higher
in energy than transition-states 2 (TS2) and 3 (TS3) and
transition-states 2 and 3 must be comparable in energy.
While aromatic and aliphatic C-H(C-D) activations at very
low temperatures are not unprecedented, the generation of
coordinatively unsaturated species capable of accomplishing this
reaction usually requires photolytic conditions19 rather than
“low-temperature” thermochemical conditions.20 Oxidative ad-
dition of toluene at -90 °C starting from 4f and COE implies
that the activation barriers for the hydrogenation of COE,
including reductive elimination of COA from complex Bf to
form the 14e- species Af, are remarkably low. The observed
Ir(D)(tolyl-d7) complex 7f is kinetically stable at -90 °C. Its
protio analogue (p-ArFPCP)Ir(H)(tolyl) was shown by line-
broadening techniques to exchange toluene by a dissociative
mechanism at temperatures above -30 °C.13 Heating the sample
to -40 °C results in reductive elimination of toluene-d8 and
finally in quantitative generation of the (p-ArFPCP)Ir(COE) (6f)
as the thermodynamic product within minutes (Scheme 3).
Complex 6f exhibits no detectable 31P NMR resonance at -40
°C (400 MHz) because of the exchange of two diastereotopic
31P nuclei, which occurs at a rate that broadens signals into the
baseline. At -90 °C, the 31P NMR resonances appear as two
doublets (2JPP ) 345 Hz) at 184.8 and 169.2 ppm, while the
fast exchange singlet is detected at 177.0 ppm above 0 °C.
(18) We believe that the cis-dihydride gives rise to one hydridic resonance
because of fast exchange presumably through an Ir(η2-H2) species.
(19) (a) Janowicz, A. H.; Bergman, R. G. J. Am. Chem. Soc. 1982, 104, 352-
354. (b) Arndtsen, B. A.; Bergman, R.; Mobley, T. A.; Peterson, T. H.
Acc. Chem. Res. 1995, 28, 154-162 and references cited herein.
(20) A few examples of oxidative C-H addition after thermochemical generation
of coordinatively unsaturated species at -86 to -60 °C are known: (a)
Periana, R.; Bergman R. G. J. Am. Chem. Soc. 1986, 108, 7332-7346. (b)
Golden, J. T.; Andersen, R. A.; Bergman, R. G. J. Am. Chem. Soc. 2001,
123, 5837-5838.
(21) Deuteration of all tert-butyl and hydridic positions of compound 4f-d38 is
achieved by heating 4f in benzene-d6 for 12 h to 150 °C.
9
9334 J. AM. CHEM. SOC. VOL. 126, NO. 30, 2004

Transfer Dehydrogenation of Cyclooctane
A R T I C L E S
Scheme 4. Reversible and Irreversible Steps of COE Deuteration at -100 °C in Methlcyclohexane-d14
significant to note that in the case of 1 the Ir(III) vinyl hydride,
{C6H3-2,6-[CH2P(tBu)2]2}Ir(H)[CHdCHC(CH3)3], is formed
rather than the olefin complex.11a,g,k Similar to the case of COE,
no hydride label is transferred into the deuterio sites of 4f-d38
upon reaction with TBE over the course of 160 min at -80 °C,
thus implying an irreversible step between 4f and the (p-ArF-
PCP)Ir(H)(neo-hexyl) complex (Cf). Given the increased steric
bulk of TBE relative to COE, it seems likely that TBE
coordination to form the (unobserved) cis- and trans-(p-ArF-
PCP)Ir(H)2(TBE) complexes requires a higher activation barrier
than COE coordination. This suggests that coordination of TBE
is the rate-determining step in the hydrogenation.
Figure 2. Free-energy G versus reaction coordinate ê for the hydrogenation
of COE by 4f, 4f-d38 at -100 °C (the intermediacy of Bf and Af is
postulated).
The transfer dehydrogenation of COA by TBE was studied
1
by H and 31P NMR spectroscopy in COA solution at 70 °C.
Treatment of a mixture of COA/6f (330:1) (obtained by reaction
of 4f with 2 equiv of COE in neat COA solvent at 23 °C) with
20 equiv of TBE resulted in one turnover after ca. 20 min, as
shown by generation of 1 equiv of TBA. At this stage, the major
iridium species is the COE complex 6f, but a small amount of
TBE complex 9f (ca. 10%) can be detected. After 5.5 turnovers
(ca. 190 min), 6f is the sole iridium complex present, even
though a large excess of TBE is still present.
Increasing temperature should accelerate the first-order
conversion of 8f,8f′ to 6f relative to the second-order formation
(negative ∆Sq) of 8f,8f′ from 4f. Thus, it is not surprising that
at higher temperatures, i.e., -70 °C, it becomes increasingly
difficult to detect 8f,8f′ in the reaction of COE with 4f.
TBE Hydrogenation. The hydrogenation of TBE by complex
4f requires slightly higher temperatures than COE hydrogenation
but is still several orders of magnitude faster than with the
standard Kaska/Jensen/Goldman pincer complex 1. Thus 4f
disappears with t1/2 ) ca. 70 min in the presence of 12 equiv of
TBE in methylcyclohexane-d14 solution at -80 °C, which
compares to the reported disappearance of complex 1 with t1/2
) ca. 10 min in the presence of 12 equiv of TBE at 55 °C.11a
No complexes other than 4f and the product olefin complex
(p-ArFPCP)Ir(TBE) (9f) are observed in this reaction.22 It is
TBE is therefore much more weakly bound to the fragment
Af than COE. These experiments show that under transfer
dehydrogenation conditions previously reported once [COE]
begins to rise, the catalyst resting state is the COE complex 6f,
and as observed, free COE inhibits turnover to a much higher
degree than TBE.12,13
The higher binding affinity of COE relative to TBE was
independently proved by reaction of precatalyst 5f with NaOtBu
and 2.7 or 12.7 equiv of TBE in benzene-d6 at 23 °C. These
reactions do not give quantitative conversion of 5f to 9f but
give mixtures of (p-ArFPCP)Ir(TBE) (9f) and (p-ArFPCP)Ir-
(D)(C6D5) (11f-d6)13 [2.7 equiv of TBE: 9f/11f-d6 ) 1:4, 12.7
(22) The assignment of complex 9f is based on the absence of any hydridic
resonances down to -100 °C as well as on the appearance of three
multiplets in the 1H NMR spectrum in a 1:1:1 ratio at 4.98, 4.03, and 3.15
ppm.
9
J. AM. CHEM. SOC. VOL. 126, NO. 30, 2004 9335

A R T I C L E S
Go¨ttker-Schnetmann and Brookhart
Scheme 5. Equilibration of Complexes 11f-d6 and 9f by Dissociative Ligand Exchange
Chart 2. Catalytic Cycles for the Transfer Dehydrogenation of COA with TBE by Complexes 1 (left side)a or 4 (right side)
a Resembling the results obtained by Goldman et al. according to ref 11a.
equiv of TBE: 9f/11f-d6 ) 1.4:1]. In contrast, reaction of 5f
with NaOtBu with 2 equiv of COE in benzene-d6 already results
in quantitative generation of COE complex 6f at 23 °C. Since
complexes 9f and 11f-d6 are equilibrated through a dissociative
ligand exchange (see Scheme 5) in a manner similar to that
found for complexes (p-XPCP)Ir(H)(aryl)13 and complexes 6,
it is clear COE has a substantially higher binding affinity than
TBE.
than COE hydrogenation although still several orders of
magnitude faster than with complex {C6H3-2,6-[CH2P(tBu)2]2}-
Ir(H)2 (1).
The results reported permit a detailed comparison of the
catalytic cycles for the Kaska/Jensen/Goldman system based
on a pincer complex of the phosphine {C6H4-1,3-[CH2P(tBu)2]2}
and the pincer catalysts studied here employing the more
electron-deficient phosphinites {5-X-C6H4-1,3-[OP(tBu)2]2}.13
Although very similar at first glance, the catalytic cycles for
the transfer dehydrogenation of COA with TBE by complexes
1 or 4 exhibit some remarkably different characteristics.
Catalytic cycles for the two systems are shown separately in
Chart 2. Both systems C-H activate COA via a 14e- fragment,
{4-X-C6H2-2,6-[OP(tBu)2]2}Ir [) (p-XPCP)Ir] (Aa-f) or {C6H3-
2,6-[CH2P(tBu)2]2}Ir (C), respectively, while generation of the
14e- fragment results from hydrogenation of TBE by the
dihydride complexes 1 or 4. â-Hydride elimination from
unobserved Ir(H)(cyclooctyl) complexes generates COE and the
catalytically active Ir(H)2 complexes 1 or 4. Under typical
catalysis condition, that is, at high [TBE] or high [COE] relative
to [Ir], the catalyst resting states are predominantly controlled
by strong olefin to Ir interactions, while under more stoichio-
metric conditions (low [olefin]:[Ir] ratios) 1 or 4 are competitive
resting states relative to the iridium olefin complexes (Chart
2).
Summary
Reaction of (p-XPCP)IrHCl complexes 5a-f with NaOtBu
in neat cyclooctane at 23 °C generates 1:1 mixtures of
complexes 4a-f, (p-XPCP)IrH2, and complexes 6a-f, (p-
XPCP)Ir(COE). At 70 °C, complexes 4a-f and 6a-f are
equilibrated by the degenerate transfer dehydrogenation of COA
with COE. At higher [COE], the equilibria 6 + COA h 4 +
2COE are shifted to the (p-XPCP)Ir(COE) complexes 6, while
a degenerate COE exchange in complexes 6 was found to be
independent of [COE] and faster for the more electron-deficient
ligand backbones p-XPCP. The separated back reaction 4f +
2COE f 6f + COA already proceeds at -100 °C via the
intermediacy of two observable isomeric (p-ArFPCP)Ir(H)2-
(COE) complexes 8f,8f′. At -70 °C, complexes 8f,8f′ are
consumed faster than they are generated under pseudo-first-
order conditions, and t1/2 ) ca. 9 min (20.5 equiv of COE) and
ca. 17 min (10.5 equiv of COE) for the overall COE hydrogena-
tion. Deuteration of COE (35 equiv) by 4f-d38 at -100 °C
reveals that coordination of COE to 4f-d38 is rate-limiting for
the overall hydrogenation reaction and that migratory insertion
of (p-ArFPCP)Ir(H)2(COE) 8f,8f′ is partially reversible, indicat-
ing that the reductive elimination of (p-ArFPCP)Ir(cyclooctyl)-
(H) is competitive with â-elimination to reform the dihydride
8f,8f′. Hydrogenation of TBE by 4f proceeds somewhat slower
A striking difference between catalysts derived from 1 and 4
is the interaction of their 14e- fragments with olefins. {C6H3-
2,6-[CH2P(tBu)2]2}Ir (C) has a high affinity for TBE because
of formation of the oxidative addition product {C6H3-2,6-[CH2P-
(tBu)2]2}Ir(H)[CHdCHC(CH3)3]. The thermodynamic stability
of this Ir(III)(vinyl)(hydride) complex results in its dominance
at [TBE] > 0.3 M and strong inhibition of catalysis. Goldman’s
9
9336 J. AM. CHEM. SOC. VOL. 126, NO. 30, 2004

Transfer Dehydrogenation of Cyclooctane
A R T I C L E S
General Procedure for in Situ Generation of COE Complexes
6a-d,f. To a mixture of complex 5a and 1.1 equiv of NaOtBu in a
medium-walled J. Young NMR tube was added ca. 330 mg of COA
containing the appropriate amount of COE in an argon atmosphere
glovebox. The tube was occasionally shaken and treated with ultrasound
until 5a was completely dissolved. After 30-60 min at 23 °C, the
starting complex had disappeared, and the solution was deep red
colored. If no COE was added to the reaction mixture, 1:1 mixtures of
4a/6a were obtained at 23 °C. A similar technique was used to generate
1:1 mixtures of 4b,c,f/6b,c,f. Complexes 6c and 6f were alternatively
generated by reaction of complexes 4c and 4f, respectively, with 2 to
kinetic studies established an inverse first-order dependence of
the catalytic turnover frequency on [TBE] because of the
unfavorable equilibrium between C + TBE and the vinyl
hydride complex, {C6H3-2,6-[CH2P(tBu)2]2}Ir(H)[CHdCHC-
(CH3)3].11a In contrast, the affinity of {4-X-C6H2-2,6-[OP-
(tBu)2]2}Ir (A) for TBE is rather low; TBE is not oxidatively
added but simply π-coordinated to the metal. Such a different
binding mode of TBE, despite the very similar geometric
environment at the Ir centers in A and C, is clearly electronic
in nature considering that A is much more electron-deficient
than C and therefore prefers substrate interactions which do not
increase its oxidation state from Ir(I). Furthermore, the steric
requirements for coordination (instead of oxidative addition)
of TBE or COE at A are rather different, radically disfavoring
TBE coordination. Consequently, TBE is only a weak inhibitor
of the catalytically active fragment A, and high TONs and TOFs
can be achieved even at high [TBE] (∼5.5 M). However, the
much higher affinity of A for COE compared to TBE causes
the catalytic inhibition after ca. 2200 TON for the best
performing fragment Af due to pre-equilibrium formation of
6f.
The second, and truly striking, difference between catalysts
1 and 4 is that 4f hydrogenates TBE or COE at much lower
temperatures (-70 °C) than 1 (55 °C) with comparable rate
constants. Significantly, the rate-determining step for the
hydrogenation of COE, and probably of TBE, by 4f is the
coordination of the olefin to 4f rather than reductive elimination
of neo-hexane, for example, from the intermediate (pincer)Ir-
(H)(neo-hexyl) complex, as was demonstrated by Goldman for
the Kaska/Jensen/Goldman system.11a Once again, the higher
stability of the Ir(III) oxidation state in complexes derived from
{C6H3-2,6-[CH2P(tBu)2]2}Ir (C) might account for this behavior
when compared to a higher stability of Ir(I) oxidation states in
complexes derived from (p-XPCP)Ir (A). Furthermore, the
increased electrophilicity in these bis(phosphinite) complexes
suggests a higher rate of insertion of the olefin dihydride
complex.
1
2.5 equiv of COE in neat COA at 23 °C. H NMR spectra in neat
COA were referenced to the COA signal ()1.57 ppm). NMR data for
6a-d,f are reported in neat COA. The 31P resonances are temperature-
dependent. Samples for 31P {1H}NMR spin saturation transfer experi-
ments were prepared by an identical procedure in the absence of COE.
Capillaries containing a toluene-d8 solution of the p-HPCP ligand of
complex 4c {)C6H4-1,3-[OP(tBu)2]2} were used as internal integration
standards and gave identical results when compared to the software
calibration routine provided by the Bruker XWinNMR package. T1
relaxation times for the 31P NMR resonances were determined by the
inversion recovery method. The relaxation delay was uniformly set to
15 s (>5 × T1) for each SST experiment. Five independent SST
experiments at 80 °C were conducted with 64 transients each, to obtain
accurate rate data for the equilibrium of Scheme 2. Irradiation of the
31P{1H} NMR resonance of complexes 4 or 6 with dB ) 120 gave no
changes of the integrals of 4 and 6. Irradiation of the 31P{1H} NMR
resonance of complexes 4 with dB ) 35 gave complete saturation of
complex 4 and signal depletion of the respective complex 6. Irradiation
of the 31P{1H} NMR resonance of complexes 6 with dB ) 35 gave
complete saturation of complex 6 and signal depletion of the respective
complex 4. Finally, irradiation in the middle of both resonances with
dB ) 35 gave no signal depletion of 4 or 6. The T1 relaxation times,
in combination with the degree of signal depletion (and the thermo-
dynamic ratio of 4/6 as a crosscheck), allowed for a quantitative analysis
of the rates of the separated forward and back reactions of the
equilibrium 6 + COA h 4 + 2COE. Rates Robs1 for the half reaction
6 + COA f 4 + 2COE were obtained by the algorithm Robs1 ) T1(6)-1
× [(1/I(6)) - 1] after irradiation of the 31P NMR signal of 4, where
I(6) is the relative integral of 6 obtained by complete versus no
saturation of 4. Rates Robs-1 for the back reaction were obtained by the
respective algorithm Robs-1 ) T1(4)-1 × [(1/I(4)) - 1] after irradiation
of the 31P NMR signal of 6, where I(4) is the relative integral of 4
obtained by complete versus no saturation of 6.
Experimental Section
General Considerations. All manipulations were carried out using
standard Schlenk, high vacuum, and glovebox techniques. Argon was
purified by passage through columns of BASF R3-11 (Chemalog) and
4 Å molecular sieves. Toluene-d8, benzene-d6, and methylcyclohexane-
d14 were degassed, vacuum-transferred, and stored over mol sieves in
an argon atmosphere glovebox. COA was stirred with concentrated
H2SO4 for several hours until it was olefin- and arene-free (GC and
NMR analysis), and then it was distilled under vacuum and stored in
an argon atmosphere glovebox. Cyclooctene was degassed and distilled
under vacuum and stored in an argon atmosphere glovebox prior to
use. TBE as received from Aldrich was degassed, vacuum-transferred,
and stored in an argon atmosphere glovebox. Complexes 4 and 5 were
synthesized by known procedures,12,13 and complex 4f-d38 (deuterated
in all tert-butyl and hydridic positions) was obtained by heating a
benzene-d6 solution of 4f for 12 h to 150 °C followed by high vacuum
sublimation of the frozen solvent at 0 °C.23 NMR spectra were recorded
on a Bruker DRX 400 MHz instrument and are referenced either to
residual protio solvent or to TMS as internal standard. Samples in neat
protio solvent were shimmed by optimizing the intensity of the FID
and referenced to the solvent chemical shift in CDCl3. 31P chemical
shifts are referenced to an external H3PO4 standard.
1
6a: H NMR (400.1 MHz, 80 °C, neat COA): δ 6.11 (s, 2 H, 3-
and 5-H), 4.18 [m br, 2 H, Ir(COE)], 3.67 (s, 3 H, OCH3), remaining
signals not observed because of COA. 31P{1H} NMR (162 MHz, 80
°C, neat COA): δ 171.4. 6b: 1H NMR (400.1 MHz, 80 °C, neat
COA): δ 6.33 (s, 2 H, 3- and 5-H), 4.22 [m br, 2 H, Ir(COE)],
remaining signals not observed because of COA. 31P{1H} NMR (162
1
MHz, 80 °C, neat COA): δ 171.2. 6c: H NMR (400.1 MHz, 80 °C,
2
neat COA): δ 6.78 (t br, 1 H, 4-H), 6.47 (d, JH-H ) 8.0 Hz, 2 H, 3-
and 5-H), 4.28 [m br, 2 H, Ir(COE)], remaining signals not observed
because of COA. 31P{1H} NMR (162 MHz, 80 °C, neat COA): δ 170.5.
6d: 1H NMR (400.1 MHz, 80 °C, neat COA): δ 6.2 (d, 2JF-H ) 10.3
Hz, 2 H, 3- and 5-H), 4.25 [m br, 2 H, Ir(COE)], remaining signals
not observed because of COA. 31P{1H} NMR (162 MHz, 80 °C, neat
COA): δ 174.5. 6f: 1H NMR (400.1 MHz, 80 °C, neat COA): δ 7.79
(s, 2 H, 2′- and 6′-H), 7.52 (s, 1 H, 4′-H), 6.59 (s, 2 H, 3- and 5-H),
(23) In analogy to complex 1, complexes 4 exchange their protio or deuterio
label in the hydridic positions with the tert-butyl groups. See also ref 11k.
9
J. AM. CHEM. SOC. VOL. 126, NO. 30, 2004 9337

A R T I C L E S
Go¨ttker-Schnetmann and Brookhart
4.21 [m br, 2 H, Ir(COE)], remaining signals not observed because of
COA. 31P{1H} NMR (162 MHz, 80 °C, neat COA): δ 179.0.
Low-Temperature Hydrogenation of COE by Complex 4f in
Toluene-d8 and Methylcyclohexane-d14 and by Complex 4f-d38 in
Methylcyclohexane-d14. A solution of 4f (8.1 mg, 10 µmol) and 300
µL toluene-d8 in a septum-capped NMR tube was cooled to ca. -110
°C in a toluene/liquid nitrogen slush. A solution of COE (12.1 mg,
110 µmol, 11 equiv; 20.6 mg, 187 µmol, 18.7 equiv; 40.8 mg, 370
µmol, 37 equiv) in 100 µL toluene-d8 was slowly added via syringe,
and the content was carefully mixed while the tube remained in the
toluene/liquid nitrogen slush. The tube was then inserted into the
precooled (-105 °C) probe of the NMR spectrometer and shimmed
before being heated to -90 °C and reshimmed. Formation of complexes
8f,8f′ and traces of 7f-d8 were already evidenced by the first acquired
experiment at -90 °C (after ca. 5 min). The buildup of complexes
8f,8f′ and 7f-d8 was monitored over the course of 3 h (11 equiv of
COE), while >95% (4 to 5 half-lives) of 4f and 8f,8f′ was consumed
to leave 7f-d8. Warming of the sample to -40 °C (within ca. 2 min)
resulted in quantitative generation of the COE complex 6f. Increasing
amounts of COE added to complex 4f resulted in higher buildup of
complexes 8f,8f′.
A solution of 4f (8.1 mg, 10µmol) and 280 µL methylcyclohexane-
d14 in a septum-capped NMR tube was cooled to ca. -125 °C in an
n-hexane/liquid nitrogen slush. A solution of COE (11.6 mg, 105 µmol,
10.5 equiv; 15.4 mg, 140 µmol, 14 equiv; 22.6 mg, 205 µmol, 20.5
equiv) in 100 µL methylcyclohexane-d14 was slowly added via syringe,
and the content was carefully mixed while the tube remained in the
n-hexane/liquid nitrogen slush. The tube was then inserted into the
precooled (-110 °C) probe of the NMR spectrometer, and the sample
was shimmed. An 1H NMR experiment at -110 °C indicated no
consumption of complex 4f. The sample was then heated to -70 °C in
the probe and reshimmed. Substantial amounts of complexes 8f,8f′ (10-
15%, depending on the equivalents of COE added) and 6f (10-16%)
were already formed by the time the first experiment was acquired
(ca. 4 min). The consumption of 4f and 8f,8f′ as well as the buildup of
complex 6f was monitored by 1H NMR experiments. The well-separated
signals of 3- and 5-H of complexes 4f (6.83 ppm), 8f,8f′ (6.58 ppm),
and 6f (6.73 ppm) were suitable for accurate integration and allowed
for the determination of t1/2 of complex 4f ) ca. 17 min (10.5 equiv of
COE) and ca. 9 min (20.5 equiv of COE). Similar experiments were
conducted with 4f-d38 and 35 equiv of COE in methylcylohexane-d14
at -100 °C. In these experiments, the buildup of hydridic resonances
for complexes 8f,8f′-d38 was observed, while no hydridic label in
complex 4f-d38 was detectable. 4f 1H NMR (400.1 MHz, -70 °C,
methylcyclohexane-d14): δ 7.94 (s, 2 H, 2′- and 6′-H), 7.69 (s, 1 H,
4′-H), 6.83 (s, 2 H, 3- and 5-H), tBu not observed because of excess
COE, -15.80 (t br, 2 H, IrH2). 31P{1H} NMR (162 MHz, -70 °C,
methylcyclohexane-d14): δ 205.4.
6f: 1H NMR (400.1 MHz, -70 °C, methylcyclohexane-d14): δ 7.92
(s, 2 H, 2′- and 6′-H), 7.65 (s, 1 H, 4′-H), 6.73 (s, 2 H, 3- and 5-H),
4.23 (m), 2.48 (m), tBu not observed because of excess COE. 31P{1H}
NMR (162 MHz, -70 °C, methylcyclohexane-d14): δ. 184.2 and 168.6
2
(d each, JPP ) 348 Hz).
Low-Temperature Hydrogenation of TBE by Complexes 4f and
4f-d38. A solution of 4f (8.1 mg, 10 µmol) and 280 µL methylcyclo-
hexane-d14 in a septum-capped NMR tube was cooled to ca. -125 °C
in an n-hexane/liquid nitrogen slush. A solution of TBE (10.1 mg, 120
µmol, 12 equiv; 29.5 mg, 350 µmol, 35 equiv) in 100 µL methylcy-
clohexane-d14 was slowly added via syringe, and the content was
carefully mixed while the tube remained in the n-hexane/liquid nitrogen
slush. The tube was then inserted into the precooled (-110 °C) probe
of the NMR spectrometer and shimmed before heating to -80 °C and
reshimming. With 12 equiv of TBE, slow formation of the TBE
complex 9f was observed with a t1/2 ) ca. 70 min at -80 °C.
Deuteration of TBE (35 equiv) with 4f-d38 at -80 °C resulted in
1
observation only of complexes 4f-d38 and 9f-d36. 9f: H NMR (400.1
MHz, -80 °C, methylcyclohexane-d14): δ 7.93 (s, 2 H, 2′- and 6′-H),
7.66 (s, 1 H, 4′-H), 6.72 (s, 2 H, 3- and 5-H), 4.83 (m, 1 H,), 3.81 (m,
1 H,), 3.22 (m, 1 H), tBu not observed because of residual protio-
methylcyclohexane/excess TBE. 31P{1H} NMR (162 MHz, -80 °C,
2
methylcyclohexane-d14): δ 180.5 and 170.3 (d each, JPP ) 342 Hz
each, 1:1 P).
In Situ Generation of the TBE Complex (ArFPCP)Ir(TBE) (9f)
from 5f. Mixtures of complexes 9f and 11f-d6 were obtained by reacting
complex 5f and NaOtBu with variable amounts of TBE in benzene-d6
1
solution (see Scheme 5). 9f: H NMR (400.1 MHz, 23 °C, benzene-
d6): δ 7.87 (s, 2 H, 2′- and 6′-H), 7.67 (s, 1 H, 4′-H), 7.02 (s, 2 H, 3-
and 5-H); 4.99, 4.07, and 3.15 (m each, 1 H each, TBE), 1.28 (s, 9 H,
TBE), 1.13 (vt, JPH ) 7.0 Hz, 36 H, 4 × tBu). 31P{1H} NMR (162
MHz, 23 °C, benzene-d6): δ 176.9. 11f-d6: 1H NMR (400.1 MHz, 23
°C, benzene-d6): δ 7.87 (s, 2 H, 2′- and 6′-H), 7.67 (s, 1 H, 4′-H),
6.97 (s, 2 H, 3- and 5-H), 1.34 (vt, JPH ) 7.0 Hz, 36 H, 4 × tBu).
31P{1H} NMR (162 MHz, 23 °C, benzene-d6): δ 182.0.
Acknowledgment. We gratefully acknowledge funding by
the National Institutes of Health (GM No. GM 28938) and the
Deutsche Akademie der Naturforscher Leopoldina (Grant No.
BMBF-LPD 9901/8-60 to I.G.-S.). We express our thanks to
Alan Goldman for fruitful discussions.
8f,8f′: 1H NMR (400.1 MHz, -70 °C, methylcyclohexane-d14,
second isomer in {}): δ 7.85 {7.85} (s, 2 H, 2′- and 6′-H), 7.60 {7.60}
(s, 1 H, 4′-H), 6.58 {6.58} (s, 2 H, 3- and 5-H), 3.86 {3.42} (m, COE),
2.69 {2.69} (m, COE), tBu not observed because of excess COE, -9.80
{-10.23} (t br, 2JPH ) 15.5 Hz 2 H, IrH2). 31P{1H} NMR (162.1 MHz,
-70 °C, methylcyclohexane-d14): δ 167.2 {167.1} (br).
JA048393F
9
9338 J. AM. CHEM. SOC. VOL. 126, NO. 30, 2004