A mechanistic description of unexpected room temperature C-H and P-C bond activation by wilkinson's catalyst

Article abstract of DOI:10.1016/j.inoche.2012.04.006

The Wilkinson's catalyst, (PPh3)3RhCl dissolved in benzene under nitrogen or under high vacuum undergoes a series of unreported reactions at room temperature. The preliminary mechanistic description involves a series of consecutive intramolecular and intermolecular phosphine exchanges, benzene oxidative addition, and biphenyl reductive elimination.

Full text of DOI:10.1016/j.inoche.2012.04.006

Inorganic Chemistry Communications 21 (2012) 43–46
Contents lists available at SciVerse ScienceDirect
Inorganic Chemistry Communications
journal homepage: www.elsevier.com/locate/inoche
A Mechanistic Description of Unexpected Room Temperature C-H and P-C Bond
Activation by Wilkinson's Catalyst
Elvin Igartúa-Nieves, Antonio J. Rivera-Brown, José E. Cortés-Figueroa ⁎
Organometallic Chemistry Research Laboratory, Department of Chemistry, University of Puerto Rico, Mayagüez, PR 00681-9019
a r t i c l e i n f o
a b s t r a c t
Article history:
The Wilkinson's catalyst, (PPh3)3RhCl dissolved in benzene under nitrogen or under high vacuum undergoes
a series of unreported reactions at room temperature. The preliminary mechanistic description involves a
series of consecutive intramolecular and intermolecular phosphine exchanges, benzene oxidative addition,
and biphenyl reductive elimination.
Received 25 January 2012
Accepted 15 April 2012
Available online 24 April 2012
©
2012 Elsevier B.V. All rights reserved.
Keywords:
Wilkinson's catalyst
Consecutive reactions
C-H bond activation
oxidative addition
reductive elimination
metal hydride
Introduction
3 2 2
Ar-Cl (Ar=Ph, p-tolyl) produce trans-(Ph P) Rh(Ph PF)(Cl) (4) and
Ar-Ph (Eq. (2)).
3 3
The Wilkinson's catalyst, (PPh ) RhCl (1Cl), is ubiquitous in inor-
ganic synthesis and in homogeneous catalysis [1]. However, key
mechanistic information concerning structure-reactivity relationship
of its reactions in aromatic solvents remains unravel. The fluoro
analogue of 1Cl, (Ph P) Rh(X) (1 F, X=F) [2], was recently prepared
3 3
and characterized along with the assessment of its reactivity promot-
ed by reversible intramolecular Rh-F/P-Ph exchange [3] producing
3 2 2
cis-(Ph P) Rh(Ph)(Ph PX) (2 F, X=F). The chemical species 1Cl and
In C
Rh(η -(C
6
D
6
1 F produces trans-(Ph
3 2 2 3 2
P) Rh(Ph PF)(F) (5) and (Ph P)
its fluoro analogue 1 F are somewhat similar in terms of their struc-
ture and reactivity [2,3]. For example, they exhibit similar geometry
parameters in the solid state, and formation of the bridged dimers
2
6
H
4
PPh )) (6) (Eq. (3)) [2].
2
It was proposed that reactions 3 and 4 proceed via rate determin-
ing reversible intramolecular P-Ph/Rh-F exchange (Eq. (2)) [2,3].
(
3 4 2 2 3
Ph P) Rh (μ-X) (X=F, Cl) (3) upon PPh dissociation (Eq. (1)).
It was reported that unlike 1Cl that was thought to produce
Ph Rh (μ-Cl) (3Cl) and traces of (Ph P) (Ph)Rh(Cl) as the
sole products when dissolved and heated in aromatic solvents, 1 F
exhibits a series of unexpected reactions [2]. Reactions of 1 F with
In the present communication we are reporting that 1Cl exhibits a
similar activity under milder conditions than reported for 1 F. In addi-
tion to formation of the chloro-bridged dimer (3Cl), 1Cl undergoes a
(
3
P
2
)
4
2
2
3
2
2
series of reactions when it is dissolved in benzene (or in C
room temperature under nitrogen or under high vacuum. However,
unlike 1 F that produces trans-(Ph P) Rh(Ph PF)(F) (5) and (Ph P)
)) (6) when it is dissolved in C , 1Cl produces
))(Ph)(H) (8) and Ph-
6 6
D ) at
3
2
2
3
2
2
Rh(η -(C
Ph P) Rh(D) (7), (Ph
6 6
when it is dissolved in C D
6
H
4
PPh
2
6 6
D
2
(
C
3
3
P)
3 2
Rh(η -(C
6 4 2
H PPh
⁎
Corresponding author.
E-mail address: jose.cortes.figueroa@usa.net (J.E. Cortés-Figueroa).
6
D
5
(Eq. (4)).
1387-7003/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.inoche.2012.04.006

44
E. Igartúa-Nieves et al. / Inorganic Chemistry Communications 21 (2012) 43–46
Experimental
The nature of the chemical species was established by 1 H NMR
spectroscopy. The reactions’ progresses were monitored by observing
the decrease of absorbance values at 370 nm. Plots of absorbance vs.
time were biexponential consisting of two consecutive absorbance
decays: a relatively rapid absorbance decay (referred as the fast
segment of the plot) followed by a relatively slow absorbance decay
The set of consecutive reactions in Eqs. (5) and (6) can be
expressed as a general set of consecutive first order reactions
Eq. (7)) where R is the reactant, I is a non-steady state intermediate,
P is the product(s), and kobsd and k’obsd are pseudo-first-order rate
constants [4]. Eq. (8) is the corresponding integrated and evaluated
form of the rate law when absorbance is used to monitor the
(
(
referred as the slow segment of the plot) (Fig. 1).
Consecutive pseudo-first order rate constant values were deter-
mined in triplicate by fitting the absorbance vs. time plots using a
non-linear curve program (OriginPlus 7.5™). Error limits of the aver-
age rate constant values and of the activation parameters, reported
in Table 1 of supporting information are given in parentheses as the
uncertainties of the last digit(s). Following the reactions’ progresses
t ∞
reaction's progress. In Eq. (8), A and A are absorbance at time t
and at time infinity, respectively, and the coefficients α and β are
constants whose values depend on kobsd, k’obsd, and on the extinction
coefficients of the chemical species involved in the consecutive reac-
tions [4].
1
by H NMR spectroscopy permitted correlation of the segments of
the biexponential plots to specific chemical processes.
′
kobsd k obsd
R → I → P
ð7Þ
ð8Þ
Results and Discussion
t
′t
−k
−k obsd
obsd
At ¼ αe
þ βe
þ A∞
The biexponential plots of absorbance vs. time of solutions of 1Cl
in benzene are shown in Fig. 1. The nature of the chemical processes
involved in the fast and slow segments of the plots depends on the
experimental conditions (vide infra). For example, for reactions of
The observed biexponential decay of absorbance with time is
mathematically described by Eq. (8). A detailed explanation about
why correlation of kobsd and k’obsd to specific chemical reactions
depends on experimental conditions and how the actual assignment
was achieved will be deferred until the next two paragraphs. Results
1
Cl with benzene, the fast and slow segments of the plots were corre-
lated to consecutive reconversion of 1Cl from 3Cl and intramolecular
P-Ph/Rh-Cl exchange [2] on 1Cl producing 2Cl, respectively (Eq. (5)).
from experiments where the reactions of 1Cl with C
6
D
6
were monitored
by H NMR spectrometry evidenced formation of uncoordinated PPh
(Fig. 2). Observation of free PPh is consistent with formation of 3Cl. In-
terestingly, formation of C -Ph and a Rh-H hydride was observed
only under flooding conditions where [PPh added >>[1Cl] ([1Cl] =-
initial 1Cl concentration). The interpretation of these observables is
that oxidative addition of C to 2Cl followed by reductive elimination
accounts for the formation of C -Ph. In turn, the hydrido species is
formed from 9. The basis for this interpretation rests on the observation
that 1 H NMR spectra of solutions containing 1Cl and added PPh
in
displays signals corresponding to C -Ph and a Rh-H hydride.
The absence of these H NMR signals obtained four hours after 1Cl
was dissolved in C without added PPh suggests that C -Ph and
the hydrido species are formed exclusively from 9 during the time
scale of this study. Oxidative addition of C to 9 followed by reductive
elimination forming (Ph P) Rh(D) (7) accounts for formation of C
1
It is being proposed that the hydrido and Ph
duced by a parallel intramolecular oxidative addition (cyclometalation)
and intermolecular oxidative addition of benzene to (Ph P) Rh(Ph) (9),
that in turn is produced by intermolecular PPh /PPh Cl exchange on 2Cl
Eq. (6)). The basis for this interpretation rests on aspects of the reported
2
species are respectively pro-
3
3
3
3
6 5
D
3
2
3
]
0
0
(
mechanism for the reactions of 1 F with Ar-X (X=H, Cl) [2] and on results
from kinetics experiments being reported in this communication.
6 6
D
6 5
D
3
C
6
D
6
6 5
D
1
D
6 6
3
6 5
D
0
0
0
0
0
0
.40
.36
.32
.28
.24
.20
6 6
D
3
3
6 5
D -
Ph but not for the formation of Rh-H hydride (Eq. (6)). However, intra-
molecular oxidative addition on 9 accounts for the observation of a
Rh-H hydride (Eq. (6)). It is expected that 9, like its methyl congener
Ph P) Rh(Me), will undergo facile cyclometalation [5]. These observa-
3 3
tions and interpretation are summarized on the mechanistic description
in Scheme 1.
This mechanism's rate law and the rate law from Eq. (7) would be
mathematically equivalent when 2Cl (or 1Cl, depending on experimen-
tal conditions) is a non-steady state intermediate. However, one must
exercise caution on assigning kobsd and k’obsd to specific mechanistic
steps. For example, instead of the previous correlation where kobsd
and k’obsd were associate to the set of consecutive reactions in Eq. (5),
(
f)
(
(e)
(
(
(
d)
c)
b)
(
a)
0
1000
2000
3000
4000
5000
6000
Time (s)
3 0
for reactions under flooding conditions ([PPh ]added >>[1Cl] ), kobsd
and k’obsd are related to reactions in Scheme 1 by Eqs. (9) and (10),
respectively.[4] It was observed that under flooding conditions where
[PPh ]added >>[1Cl] , kobsd and k’obsd are independent and dependent,
3 0
Fig. 1. Plots of absorbance (370 nm) vs. time for reactions of 1Cl with benzene. (a), at 62.1 °C;
added =2.29×10- M;
4
(
b), at 42.1 °C; (c), at 56.1 °C; (d), at 32.1 °C in presence added [PPh ]
3
-
3
(
e), at 32.1 °C; (f), at 32.1 °C in presence added [PPh
3
]
added =7.37×10 M.

E. Igartúa-Nieves et al. / Inorganic Chemistry Communications 21 (2012) 43–46
45
Rh-H
Rh-PPh3/
PPh3(added)
*
(
c)
PPh
3
added
-
1.35 -1.40 -1.45 -1.50 -1.55 -1.60
ppm
Rh-PPh3
Ph-C D
6
5
Ph-C D
6
5
(
b)
Rh-PPh3
~
*
Rh-PPh3
PPh3
(
a)
Fig. 3. Plot of k’obsd vs. [PPh
at 32.1 °C (■) (k’obsd =0.21(1) [PPh
3
] for reactions of 1Cl with benzene containing added [PPh
added, 42.1 °C (k’obsd =0.44(3) [PPh added) and
3
(k’obsd =2.5(4) [PPh ]added).
3
]
3
]
3
]
52.1 °C
7.8
7.6
7.4
7.2
7.0
ppm
Fig. 2. 1H NMR spectra of (a) 1Cl in C
solution [PPh added =0.0314 M.
6 6 6 6 6 6 3
D /C H *(0.04%), (b) and (c) 1Cl in C D /PPh
‡
3
]
ΔS =65(83) J/K mol). The activation parameters estimated in our
study are in the ballpark of the corresponding values reported for
intermolecular PPh
kJ/mol, ΔS =+18(1) J/K mol) [10]. These values suggest that the
intermolecular exchange may proceed via an interchange mechanism
involving solvent(s) molecule(s). Moreover, the observation that
‡
respectively of [PPh
obsd to (kexch+kdim). Because [PPH
formation, kobsd ≈kexch. Likewise, the observation that k’obsd values
are dependent on [PPh added, permits a preliminary assignment of
k’obsd to parallel intermolecular PPh /Ph Cl exchange on 2Cl and oxida-
3
]added (Fig. 3). This observation permits associate
3
/PPh
3
exchange on 1Cl in THF (ΔH =77.2(4)
‡
k
3
]added is expected to inhibits 3Cl
3
]
k
PPh /PPh3values for intermolecular PPh
solvent-dependent (at 30 °C; kPPh /PPh3(in THF)=2.9 s ; kPPh /PPh3(in
3
/PPh
3
exchange on 1Cl are
3
2
3
-
1
tive addition of benzene and cyclometalation on 2Cl (Eq. (10)).
3
3
toluene)=0.31 s-1) [10] supports an associative-interchange process.
As mentioned above, in kinetics runs where [PPh3]added =0, kobsd
was correlated to k-dim that governs reconversion of 1Cl from 3Cl,
whereas k’obsd was associated to intramolecular Cl/Ph kexch. Rate
constant values obtained using these correlations for reactions at
various temperatures and corresponding estimated activation pa-
rameters are being reported in this paper for formation of 1Cl from
3Cl (k-dim) and intramolecular exchange on 1Cl (kexch). Activation
parameters (ΔH exch =66(2) kJ/mol, ΔS exch =-100(8) J/K mol) for
intramolecular P-Ph/Rh-Cl exchange were obtained from the Eyring
equation by plotting ln(kobsd =(kexch))/T vs. 1/T (Fig. 4).
Density functional theory has been used to model a series of plausible
P-Ph/Rh-F exchange pathways [2b,3]. Results from these calculations on
kobsd¼ kexch
ð9Þ
0
k
obsd¼ k
½PPh ꢀþ k
½C H ꢀþ k
ð10Þ
PPh 3=PPh Cl
3
oxad
6
6
cm
2
k
0obsd¼ k
½PPh3ꢀ
ð11Þ
PPh 3=PPh Cl
2
Eq. (10) predicts linear plots of k’obsd vs. [PPh
and intercept are related to kPPh /PPh2Cland (koxad[C
3
]
added which slope
]+kcm) values,
respectively. The observation that intercept values of these plots are
zero within experimental error indicates that oxidative addition of
benzene to 2Cl and cyclometalation on 2Cl are not competig process-
es. This kinetics result is in line with the observation that Rh-H and
‡
‡
6
H
6
3
C
6
H
5
-C
added >>[ICl]
PPh /PPh2Clat 32.1 °C, and 42.1 °C and 52.1 °C obtained from k’obsd vs.
6
D
6
species are formed under flooding conditions where
3
(PhP) Rh(X) indicate that the most likely pathway involves formation of
[PPh
3
]
0
. Thus, Eq. (10) reduces to Eq. (11). Values of
a metallophosphorane [2b,3a] intermediate where the involved phos-
phorus is five-coordinated. Formation of this five-coordinated phos-
phorus promotes P-Ph activation, X/Ph exchange and formation of
k
3
[PPh
3
]added plots in Fig. 3 were used to estimate the activation param-
‡
eters for PPh
3 2
/PPh Cl exchange on 2Cl. (ΔH =99(26) kJ/mol,
3 2 2
cis-RhPh(PPh ) (PXPh ) [3]. The highly negative experimental entropy
6 6
Scheme 1. Proposed mechanistic description of reactions of 1Cl in benzene (or in C D ).

46
E. Igartúa-Nieves et al. / Inorganic Chemistry Communications 21 (2012) 43–46
‡
of activation (ΔS exch =-100(8) J/K mol) in our study indicates substan-
tial loss of degrees of freedom in the transition state structure leading to
formation of 2Cl. The corresponding enthalpy of activation for the forma-
structural properties and chemical and physical behaviour in solution
of the above complexes is important not only because it is necessary
information to design catalysts for specific applications, but to identi-
fy and to mitigate chemical processes that may reduce their catalytic
activity. Experimental and theoretical investigations are underway in
our laboratory to address this concern.
‡
tion of 2Cl (ΔH exch =66(2) kJ/mol) is considerably smaller than exper-
‡
‡
imental (ΔH exch =92(2) kJ/mol, ΔS exch =-42(1) J/K mol) [2b] and
‡
computed (ΔH exch =92.3 kJ/mol) [2a] values reported for formation of
2
F from 1 F. Although P-R/Rh-X exchange must be investigated in a va-
3 2
riety of (R P) (Ph)Rh(X) complexes (R=alky, aryl groups, X=Cl, F, I…),
Acknowledgment
this set of activations parameter for the formation of 2Cl and 2 F from 1Cl
and 1 F, respectively, suggests the existence of a compensation effect be-
Acknowledgment is made to the Donors of The Petroleum Re-
search Fund, administered by the American Chemical Society, (ACS-
PRF # 36623-B3). Valuable comments from Professor Deborah A.
Moore-Russo (SUNY-Buffalo) are gratefully acknowledged. The ex-
perimental assistance of the participants of the UPRM undergraduate
research program (CHEM 4998-4999) is also acknowledged.
‡
‡
‡
tween ΔH exch and ΔS exch, so that relatively large ΔH exch are accompa-
‡
nied by relative large ΔS exch [6]. The preliminary interpretation
proposed here is that P-Ph/Rh-X (X=Cl, F) exchange proceeds via a
common mechanism where the transition states exhibit degree of P-X
bond formation and Rh-X bond dissociation [7–9].
Appendix A. Supplementary material
Conclusion
Supplementary data to this article can be found online at http://
dx.doi.org/10.1016/j.inoche.2012.04.006.
The set of consecutive reactions ascribable to the two consecutive
exponential decays of absorbance (370 nm) with time and correlation
of corresponding kobsd and k’obsd parameters in Eq. (8) depends on
References
3
experimental conditions: i) when [PPh ]added =0 the consecutive re-
actions are
[1] (a) M.A. Bennett, P.A. Longstaff, Chem. Ind. (London) (1965) 846;
(b) J.A. Osborn, F.H. Jardine, J.F. Young, G. Wilkinson, J. Chem. Soc. A (1966)
1711;
(
(
c) F.H. Jardine, Prog. Inorg. Chem. 28 (1981) 63;
d) A. Grünberg, X. Yeping, H. Breitzke, G. Buntkowsky, Chem. Eur. J. 16 (2010)
6993.
[
2] (a) V.V. Grushin, W.J. Marshall, J. Am. Chem. Soc. 126 (2004) 3068;
(b) S.A. Macgregor, D.C. Roe, W.J. Marshall, K.M. Bloch, V.I. Bakhmutov, V.V.
Grushin, J. Am. Chem. Soc. 127 (2005) 15304.
[3] S.A. Macgregor, T. Wondimagegn, Organometallics 26 (2007) 1143;
b) S.A. Macgregor, Chem. Soc. Rev. 36 (2007) 67.
(
k
obsd =k-dim and k’obsd =kexch); ii) whereas under flooding where
(
[PPh added >>[1Cl] , the consecutive reactions are
3
]
0
[
4] (a) J.H. Espenson, Chemical Kinetics and Reaction Mechanisms, 2nd ed.,
McGraw-Hill, New York, 1995, pp. 74–75;
(b) J.I. Steinfeld, J.S. Francisco, W.L. Hase, Chemical Kinetics and Dynamics, 2nd
ed., Prentice Hall, Upper Saddle River, NJ, 1998, pp. 25–32;
(
(
c) T. Félix, J.E. Cortés-Figueroa, Inorg. Chem. Commun. 12 (2009) 347;
d) T. Félix, J.E. Cortés-Figueroa, J. Chem. Educ. 87 (2010) 426.
[
[
5] W. Keim, J. Organomet. Chem. 14 (1968) 179.
6] B. Giese, Acc. Chem. Res. 17 (1984) 438;
C.D. Johnson, Chem. Rev. 75 (6) (1975) 755;
J. Rooney, J. Catal. Lett. 50 (1998) 15;
(
k
obsd =kexch and k'obsd=kPPh /PPh2Cl [PPh
3
]). Moreover, observation
of unexpected reactivity due to seemingly small structural and elec-
tronic differences in (Ph P) (L)Rh(X) (X=CF , CH , H, Ph, F, Cl, I;
L=PPh , PPh Cl, PPh F) [11] species warrants further studies involv-
3
N. Koga, J. Šesták, J. Therm. Anal. 37 (1991) 1103;
L. Liu, Q.X. Guo, Chem. Rev. 101 (2001) 673.
[7] Although the validity of a compensation effect has been questioned in the past by
3
2
3
3
3
2
2
8
some investigators , its validity is now firmly established empirically by rigorous
ing a well designed investigation on the structure/reactivity rela-
tionship on these species. Understanding the relationship between
9
statistical methods .
[
[
8] O. Exner, Nature 227 (1970) 366;
G.C. McBane, J. Chem. Educ. 75 (7) (1998) 919;
R.V. Muraleedharan, J. Therm. Anal. 41 (1994) 53.
9] A.L. Rukhin, Chemom. Intell. Lab. Syst. 90 (2008) 116;
C. Ouvrard, M. Berthelot, T. Lamer, O. Exner, J. Chem. Inf. Comput. Sci. 41 (2001)
-
-
-
4
6
8
1141;
K.A. Bunten, A.J. Poe, New J. Chem. 30 (2006) 1638;
R. Larsson, J. Mascetti, React. Kinet. Catal. Lett. 85 (1) (2005) 107.
[
[
10] J. Goodman, V.V. Grushin, R.B. Larichev, S.A. Macgregor, W.J. Marshall, D.C. Roe,
J. Am. Chem. Soc. 132 (2010) 12013.
11] Meaningful activation parameter values require knowledge of rate constant
values without unresolved concentration dependences or summation of rate
constants. J.H. Espenson, Chemical Kinetics and Reaction Mechanisms, 2nd ed.,
McGraw-Hill, New York, 1995, pp. 155–157.
-
-
-
-
-
10
12
14
16
18
0.0029
0.0030
0.0031
0.0032
0.0033
-
1
1
/T (K )
Fig. 4. Plots of ln(k/Temperature) vs. 1/Temperature. k=kPPh2Cl/PPh3 on 2Cl ; k=k-dim
on 3Cl (■);k=kexch on 1Cl ; k=kexch on 1 F
.