Mechanism of alkane transfer-dehydrogenation catalyzed by a pincer-ligated iridium complex

Article abstract of DOI:10.1021/ja0289200

The mechanism of (PCP)Ir-catalyzed transfer-dehydrogenation has been elucidated for the prototypical substrate/acceptor couple, COA/TBE, at 55 °C (COA = cyclooctane; TBE = tert-butylethylene). The catalytic cycle may be viewed as the sum of two reactions: (i) hydrogenation of TBE by (PCP)IrH₂ and C-H addition of a second mole of TBE to give (PCP)IrH(tert-butylvinyl), and (ii) dehydrogenation of COA by (PCP)IrH(tert-butylvinyl) to give (PCP)IrH₂, COE, and TBE. These two stoichiometric reactions have been observed independently and their kinetics determined. The overall catalysis has also been monitored in situ, and (PCP)IrH₂ and (PCP)IrH(tert-butylvinyl) have been observed as the resting states; the ratio of these two complexes is found to be proportional to [TBE]². Based upon the proportionality constant thus obtained and the catalytic rate as a function of [TBE] (which reaches a maximum at ca. 0.3 M), the respective rate constants for the hydrogenation and dehydrogenation segments can be obtained. Good agreement is found between the rates independently obtained from stoichiometric and catalytic runs. Within the overall TBE-hydrogenation reaction, labeling experiments indicate that the rate-determining step is the reductive elimination of TBA (2,2-dimethylbutane) from (PCP)IrH(tert-butylethyl) (which is formed via insertion of TBE into an Ir-H bond of (PCP)IrH₂). Based upon considerations of microscopic reversibility, it can be further inferred that the rate-determining step for the alkane dehydrogenations is C-H addition (and not β-H elimination). Copyright

Full text of DOI:10.1021/ja0289200

Published on Web 06/06/2003
Mechanism of Alkane Transfer-Dehydrogenation Catalyzed by a
Pincer-Ligated Iridium Complex
Kenton B. Renkema, Yury V. Kissin, and Alan S. Goldman*
Department of Chemistry, Rutgers, The State UniVersity of New Jersey, New Brunswick, New Jersey 08903
Received October 12, 2002; E-mail: goldman@rutchem.rutgers.edu
The selective dehydrogenation of alkanes and alkyl groups is a
reaction with a wide range of applications of great potential value.1
Presently, the most effective systems for selective alkyl-group
dehydrogenation are those based upon pincer-ligated iridium.2,3
Herein we report elucidation of the reaction mechanism of the first
and prototypical example of such reactions: transfer-dehydroge-
nation of cyclooctane/tert-butylethene (COA/TBE) catalyzed by
(PCP)IrH2 (1; PCP ) 2,6-C6H3(CH2PtBu2)2), to give cyclooctene
(COE) and 2,2-dimethylbutane (tert-butylethane; TBA).
and then study its kinetics, the suspected second half of the cycle
(eq 4) is too rapid to measure at 55 °C at low [TBE] (<ca. 0.2 M).
Higher [TBE] inhibited reaction 4, but under such conditions
reaction 2 then proceeds too rapidly to allow the independent
measurement of the rate of reaction 4. Nevertheless, the observation
of inhibition by added TBE suggests that the reaction proceeds via
free “(PCP)Ir”.
We therefore used an alternative source of “(PCP)Ir” which
permitted study of COA dehydrogenation (as in eq 4) without the
complications introduced by the presence of free TBE. (PCP)IrH-
(Ph) undergoes dissociative hydrocarbon exchange in analogy with
the behavior of 2 (eq 3). Accordingly, it reacts with COA (eq 5),
in analogy with the reaction of 2 (eq 4).
(PCP)IrH(Ph) + COA f 1 + C6H6 + COE
(5)
The kinetics of eq 5 (55 °C, p-xylene solvent) are found to be
consistent with the rate law of eq 6, with kdh(IrPh) ) 0.0023
Our approach to the mechanism involves independent determi-
nation of the kinetics of both the catalytic system and its
stoichiometric component reactions.
In the absence of alkane (in mesitylene solvent), 1 reacts with
TBE: 1 equiv of TBE is hydrogenated, and a second equivalent
undergoes vinylic C-H addition to (PCP)Ir (eq 2) to give (PCP)-
IrH(tBuVi) (2; tBuVi ) trans-2-[tert-butyl]vinyl).4
min-1
.
-d[(PCP)IrH(Ph)]/dt ) d[1]/dt )
k
dh(IrPh)[(PCP)IrH(Ph)][COA][C6H6]-1 (6)
To extrapolate the analogous rate constant for the actual complex
of interest, 2, it is sufficient to determine the equilibrium constant
for eq 7;5 KPh/Vi was found to equal 0.33.
KPh/Vi
(PCP)IrH(Ph) + TBE y z (PCP)IrH(tBuVi) + PhH (7)
Applying this value to kdh(IrPh), we obtain a value of 0.0070 min-1
for kdh, the analogous rate constant for the reaction of 2 with COA
(eq 4). This extrapolation is valid assuming that reactions 4 and 5
both proceed through a pre-equilibrium involving (PCP)Ir or
through any other common intermediate.
Based on the mechanism of eqs 2 and 4 and the corresponding
rate laws, the ratio of [2] to [1] under catalytic conditions can be
predicted to be that of eq 8.5
We have previously reported 2 and found that it undergoes rapid
exchange with free TBE (on the NMR time scale) at ambient
temperature via reversible loss of TBE (eq 3).4
(PCP)IrH(tBuVi) h “(PCP)Ir” + TBE
(2)
(3)
The kinetics of reaction 2 were studied by monitoring the
disappearance of 1 and appearance of 2 at 55 °C in mesitylene.
The reaction was found to proceed with a rate constant kh ) 0.57
M-1 min-1, where -d[1]/dt ) d[2]/dt ) kh[1][TBE].
When COA is added to 2 (in p-xylene solvent at 55 °C), a rapid
reaction ensues (eq 4).
[2]/[1] ) kh[TBE]2/kdh[COA]
(8)
1
The catalytic reaction (eq 1) was monitored in situ (31P and H
NMR) at 55 °C. A plot of [2]/[1] vs [TBE]2 ([TBE] ) 0.12-0.72
M) yields a straight line, in excellent agreement with eq 8.5 The
slope, kh/kdh[COA], yields kh/kdh ) 56 M-1
.
2 + COA f 1 + TBE + COE
(4)
The catalytic turnover rate predicted by the mechanism of eqs 2
The sum of reactions 2 and 4 is the catalytic reaction of eq 1.
Thus, the two reactions constitute a possible mechanism for the
catalytic transfer-dehydrogenation.
Unfortunately, whereas it was possible to isolate the suspected
half of the catalytic cycle leading to TBE hydrogenation (eq 2)
and 4 is expressed in the overall rate law of eq 9.5
d[COE]/dt ) khkdh[TBE][COA][Irtot]/(kdh[COA] +
kh[TBE]2) (9)
9
7770
J. AM. CHEM. SOC. 2003, 125, 7770-7771
10.1021/ja0289200 CCC: $25.00 © 2003 American Chemical Society

C O M M U N I C A T I O N S
Figure 3. Reaction of 1 with TBE: competition between â-H (or â-D)
elimination and C-H bond elimination (H* ) H or D).
The labeling experiment of Figure 3 demonstrates that the rate-
determining step of reaction 2, and thus the rate-determining step
of the catalytic cycle under typical conditions, is C-H elimination
of TBA (2b, Figure 2). Based on considerations of microscopic
reversibility, C-H addition must be the rate-determining step of
the reverse reaction, the (hypothetical) dehydrogenation of TBA.
This conclusion can be further extrapolated to the terminal
dehydrogenation of n-alkanes (the lack of a bulky t-Bu group should
facilitate â-H elimination more than C-H addition) and probably
to cycloalkanes. This finding is an important contrast with previ-
ously reported alkane dehydrogenation systems, in which â-H
elimination was found to be rate-determining;1 it is clearly relevant,
for example, to the high regioselectivity shown by 1 in the terminal
dehydrogenation of n-alkanes.3
Figure 1. Rate of transfer-dehydrogenation (eq 1) vs [TBE].
In summary, the mechanism of (PCP)Ir-catalyzed COA/TBE
transfer-dehydrogenation has been elucidated. The two segments
of the cycle have been observed independently, and the nature of
the rate-determining step within each segment has been determined.
Noteworthy features of the mechanism include the following:
(i) Although olefin hydrogenation is one of the most widely and
readily catalyzed reactions of organometallic complexes, the rate-
determining (slow) segment of the present cycle under typical
conditions is hydrogenation of TBE.
(ii) Although C-H addition is often assumed to be the “difficult”
step in alkane functionalization, in this case C-H elimination of
hydrogenated alkane product (TBA; eq 2b) is apparently rate-
determining.
Figure 2. Mechanism of eq 1 (presumed intermediates in brackets). Outer
(dashed) arrows refer to eqs 2 and 4 (overall observed reactions, with rate
constants kh and kdh). 2a, 2b, 4a, 4b are inferred reaction steps.
According to eq 9, in the limit of low [TBE], the catalytic rate
is equal to kh[TBE][Irtot] (first order in [TBE]), while in the limit
of high [TBE], the rate should be kdh[TBE]-1[COA][Irtot] (inverse
first-order in [TBE]). Accordingly, the curve of rate vs [TBE]
reveals a maximum (at ca. 0.34 M TBE; Figure 1).
The curve shown in Figure 1 is calculated according to eq 9 to
a best fit of the data, with kh/kdh held fixed at 56 M-1 (the value
obtained from eq 8 and a plot of [2]/[1] vs [TBE]2). This therefore
represents a one-parameter fit yielding the two rate constants, each
obtained solely from in situ observation of the catalytic system:
(iii) In the regime where dehydrogenation is the rate-determining
segment (very high [TBE]), C-H addition is inferred to be rate-
determining. Nevertheless, a substantial contribution to the overall
barrier derives from the thermodynamic cost of eliminating the vinyl
C-H bond of TBE prior to addition of the alkane C-H bond.
kh ) 0.53 M-1 min-1, kdh ) 0.0094 min-1
.
We now arrive at the critical question in the assessment of the
proposed mechanism of eqs 2 and 4: Are the rate constants obtained
from the stoichiometric experiments consistent with those obtained
independently from the catalytic experiments? Considering experi-
mental error and possible effects of different solvents (stoichiometric
runs were necessarily conducted in non-alkane solvents), the
agreement between the sets of rate constants is excellent (catalytic
kh ) 0.53 M-1 min-1, kdh ) 0.0094 min-1; stoichiometric kh )
0.57 M-1 min-1, kdh ) 0.0070 min-1). The overall set of kinetic
equations is thus strongly overdetermined, and we consider these
kinetics to be, effectively, proof of the mechanism of eqs 2 and 4
(Figure 2).
Acknowledgment. We thank the Division of Chemical Sciences,
Office of Basic Energy Sciences, Office of Energy Research, U.S.
Department of Energy, for support of this research.
Supporting Information Available: General experimental infor-
mation, derivations of eqs 8 and 9, graphical results and discussion
concerning eqs 5 and 8, discussion of isotope effects, and kinetics-
modeling procedures (PDF). This material is available free of charge
via the Internet at http://pubs.acs.org.
References
Somewhat surprisingly, under typical conditions ([TBE] < ca.
0.3 M), the TBE-hydrogenation part of the cycle is rate-determining.
To determine the rate-determining reaction step (within the overall
reaction 2), a labeling experiment was designed to reveal if insertion
of TBE into the Ir-H bond is irreversible (i.e., if k2b . k-2a, Figure
3). (PCP)IrD2 was reacted with TBE. The initial rate of formation
of the H/D-exchanged product, CH2dCDtBu, was found to be 4.8
times greater than that of the hydrogenated product, CH2DsCHD-
(1) Crabtree, R. H. J. Chem. Soc., Dalton Trans. 2001, 17, 2437-2450 and
references therein.
(2) (a) Jensen, C. M. Chem. Commun. 1999, 2443-2449. (b) Gupta, M.;
Hagen, C.; Flesher, R. J.; Kaska, W. C.; Jensen, C. M. Chem. Commun.
1996, 2083-2084. (c) Liu, F.; Pak, E. B.; Singh, B.; Jensen, C. M.;
Goldman, A. S. J. Am. Chem. Soc. 1999, 121, 4086.
(3) Previous reports of (PCP)Ir-catalyzed transfer-dehydrogenation (ref 2)
involved temperatures g150 °C; however, for the purposes of mechanistic
study, 55 °C was found to give much more convenient rates.
(4) Kanzelberger, M.; Singh, B.; Czerw, M.; Krogh-Jespersen, K.; Goldman,
(tBu). Neglecting isotope effects,5 we therefore obtain k-2a/k2b
)
A. S. J. Am. Chem. Soc. 2000, 122, 11017.
(5) See Supporting Information.
9.6. (Statistically, formation of CH2dCDtBu reflects half the rate
of â-H/D elimination).
JA0289200
9
J. AM. CHEM. SOC. VOL. 125, NO. 26, 2003 7771