C-H bond activation by rhodium(I) phenoxide and acetate complexes: Mechanism of H-D exchange between arenes and water

Article abstract of DOI:10.1021/om7012259

New rhodium(I) complexes (PNP)Rh(X) (PNP = 2,6-bis(di-tert- butylphosphinomethyl)pyridine) (X = OTf(1), OAc (3), OH (8), OCH ₂CF₃ (9), OC₆H₅ (10), OC ₆H₄NO₂ (11)) have been prepared. Hydroxide complex 8 and trifluoroethoxide complex 9 undergo stoichiometric activation of benzene-d₆ to form the phenyl complex (PNP)Rh(C₆D ₅). Acetate and aryloxide complexes 3, 10, and 11 are active catalysts for H-D exchange between arenes and water. Control experiments indicate that the rhodium complexes are the active catalysts and that the observed exchange is not catalyzed by adventitious acid. Mechanistic studies of the H-D exchange reaction support a pathway involving dissociation of aryloxide or acetate ligand. The reaction is accelerated by added alcohol and, for the acetate complex, inhibited by added sodium acetate.

Full text of DOI:10.1021/om7012259

1454
Organometallics 2008, 27, 1454–1463
C-H Bond Activation by Rhodium(I) Phenoxide and Acetate
Complexes: Mechanism of H-D Exchange between Arenes and
Water
Susan Kloek Hanson, D. Michael Heinekey,* and Karen I. Goldberg*
Department of Chemistry, Box 351700, UniVersity of Washington, Seattle, Washington 98195-1700
ReceiVed December 18, 2007
New rhodium(I) complexes (PNP)Rh(X) (PNP ) 2,6-bis(di-tert-butylphosphinomethyl)pyridine) (X
) OTf (1), OAc (3), OH (8), OCH2CF3 (9), OC6H5 (10), OC6H4NO2 (11)) have been prepared. Hydroxide
complex 8 and trifluoroethoxide complex 9 undergo stoichiometric activation of benzene-d6 to form the
phenyl complex (PNP)Rh(C6D5). Acetate and aryloxide complexes 3, 10, and 11 are active catalysts for
H-D exchange between arenes and water. Control experiments indicate that the rhodium complexes are
the active catalysts and that the observed exchange is not catalyzed by adventitious acid. Mechanistic
studies of the H-D exchange reaction support a pathway involving dissociation of aryloxide or acetate
ligand. The reaction is accelerated by added alcohol and, for the acetate complex, inhibited by added
sodium acetate.
does not directly activate C-H bonds.4 Thus water serves to
Introduction
inhibit C-H bond activation, and rigorously water-free condi-
tions result in higher activity for C-H bond activation.5
Labinger and Bercaw have also shown that alcohols can bind
to the metal center and be subject to C-H bond activation
themselves, at rates comparable to that of alkanes.5 This
incompatibility with alcohols places limits on the potential
catalytic functionalization of hydrocarbons by many metal
systems.
The discovery of efficient catalysts for the activation and
functionalization of C-H bonds could have tremendous envi-
ronmental and economic impact. In particular, catalysts for the
selective oxidation of organic compounds could streamline
synthesis in the chemical and pharmaceutical industry, reducing
energy use and waste production by allowing for direct
utilization of cheaper and more available feedstocks.1
However, the development of alkane functionalization cata-
lysts has proved to be a major challenge. The catalyst must be
able to selectively react with alkane C-H bonds, even in the
presence of functionalized products such as alcohols. Although
many late transition metal complexes have been shown to
activate C-H bonds, the large majority of these reactions are
inhibited by or incompatible with water and alcohols.2 This is
due in part to the fact that C-H bond activation generally
requires an available site at the metal center to bind alkane.
Water and alcohols often bind to transition metals more
effectively than arenes or alkanes, blocking the open site and
suppressing the desired reactivity.2
For example, Labinger, Bercaw, and Tilset have studied C-H
bond activation by platinum(II) diimine complexes and have
shown that C-H bond activation occurs at a platinum(II) solvent
complex.3,4 In the presence of water, the platinum(II) solvent
complex is in equilibrium with a water-bound complex, which
In this context, an ideal reaction would be C-H bond
activation by a metal alkoxide or hydroxide complex, breaking
a C-H bond, and liberating an alcohol product in one step.
The first examples of this reaction were reported in 2005 by
Gunnoe and by Periana.6,7 Gunnoe and co-workers described
arene activation by (Tp)Ru(PMe3)2OH (Tp ) hydridotris(pyra-
zoyl)borate), which exchanged deuterium into the hydroxide
ligand when heated in benzene-d6 and was demonstrated to be
an active catalyst for H-D exchange between arenes and water.6
Periana and co-workers documented the reactivity of
(acac)2Ir(OCH3)py, which upon heating underwent stoichio-
metric benzene activation to convert to the phenyl complex
(acac)2Ir(C6H5)py and was also shown to be an active catalyst
for H-D exchange between arenes and water.7 Subsequent
mechanistic studies carried out by both groups suggest similar
mechanisms involving dissociation of a neutral ancillary ligand,8
followed by a concerted C-H bond activation across the metal
hydroxide or alkoxide ligand. This four-center interaction
* Corresponding author. E-mail: goldberg@chem.washington.edu;
heinekey@chem.washington.edu.
(1) (a) Shilov, A. E.; Shul’pin, G. B. Chem. ReV. 1997, 97, 2879–2932.
(b) Stahl, S. S.; Labinger, J. A.; Bercaw, J. E. Angew. Chem., Int. Ed. 1998,
37, 2180–2192. (c) Jones, W. D. Science 2000, 287, 1942–1943. (d)
Crabtree, R. H. J. Chem. Soc., Dalton Trans. 2001, 2437–2450. (e) Labinger,
J. A.; Bercaw, J. E. Nature 2002, 417, 507–514. (f) ActiVation and
Functionalization of C–H Bonds; Goldberg, K. I., Goldman, A. S., Eds.;
ACS Symposium Series 885; American Chemical Society: Washington, DC,
2004.
(2) (a) Periana, R. A.; Bhalla, G.; Tenn, W. J., III; Young, K. J. H.;
Liu, X. Y.; Mironov, O.; Jones, C. J.; Ziatdinov, V. R. J. Mol. Catal. A
2004, 220, 7–25. (b) Conley, B. L.; Tenn, W. J., III; Young, K. J. H.;
Ganesh, S. K.; Meier, S. K.; Ziatdinov, V. R.; Mironov, O.; Oxgaard, J.;
Gonzales, J.; Goddard, W. A., III; Periana, R. A. J. Mol. Catal. A 2006,
251, 8–23.
(3) Johansson, L.; Tilset, M.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem.
Soc. 2000, 122, 10846–10855.
(4) Zhong, H. A.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc.
2002, 124, 1378–1399.
(5) Owen, J. S.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 2006,
128, 2005–2016.
(6) Feng, Y.; Lail, M.; Barakat, K. A.; Cundari, T. R.; Gunnoe, T. B.;
Petersen, J. L. J. Am. Chem. Soc. 2005, 127, 14174–14175.
(7) Tenn, W. J., III; Young, K. J. H.; Bhalla, G.; Oxgaard, J.; Goddard,
W. A., III; Periana, R. A. J. Am. Chem. Soc. 2005, 127, 14172–14173.
(8) (a) Feng, Y.; Lail, M.; Foley, N. A.; Gunnoe, T. B.; Barakat, K. A.;
Cundari, T. R.; Petersen, J. L. J. Am. Chem. Soc. 2006, 128, 7982–7994.
(b) Tenn, W. J., III; Young, K. J. H.; Oxgaard, J.; Nielsen, R. J.; Goddard,
W. A., III; Periana, R. A. Organometallics 2006, 25, 5173–5175.
10.1021/om7012259 CCC: $40.75
2008 American Chemical Society
Publication on Web 03/04/2008

Rh Catalyzed H-D Exchange Between Arenes and Water
Organometallics, Vol. 27, No. 7, 2008 1455
pathway involving the lone pair of the alkoxide or hydroxide
group is somewhat similar to σ-bond metathesis and has been
termed “internal electrophilic substitution”.9–12
We recently reported arene activation by rhodium(I) hydrox-
ide, trifluoroethoxide, and phenoxide complexes bearing the PNP
ligand (PNP ) 2,6-bis(di-tert-butylphosphinomethyl)pyridine).13
The hydroxide and trifluoroethoxide complexes undergo sto-
ichiometric benzene activation to form the rhodium(I) phenyl
complex, and the phenoxide complex is an active catalyst for
H-D exchange between arenes and water.13 Rhodium(I) acetate
and 4-nitrophenoxide complexes are also found to catalyze the
exchange. Not only is the C-H bond activation compatible with
water, but the presence of added phenol actually accelerates
the reaction. While early studies suggested similarities to the
reactions studied by Gunnoe and Periana,13 new experimental
evidence suggests that the H-D exchange reaction proceeds
by a different mechanism than that of the ruthenium(II) and
iridium(III) systems.
Figure 1. ORTEP drawing of the molecular cation of 1-N2 (thermal
ellipsoids at 50% probability, H atoms omitted for clarity). Select
bond distances (Å) and angles (deg): N1-N2 ) 1.116(4), Rh1-N1
) 1.898(3), Rh1-N3 ) 2.074(3), Rh1-P1 ) 2.304(1), Rh1-P2
) 2.297(1), P1-Rh1-P2 ) 168.07(3), N1-Rh1-N3 ) 178.9(1),
P1-Rh1-N1 ) 95.93(8), P2-Rh1-N1 ) 94.71(8).
Results
A. Synthesis and Characterization of PNP Rhodium(I)
Complexes. Rhodium(I) triflate, trifluoroacetate, and acetate
complexes bearing the PNP ligand (PNP ) 2,6-bis(di-tert-
butylphosphinomethyl)pyridine) were prepared by reaction of
0.5 equiv of [Rh(COE)2Cl]2 (COE ) cyclooctene) with silver
triflate, silver trifluoroacetate, or silver acetate in THF, followed
by filtration and addition of PNP ligand (eq 1). Complexes
(PNP)Rh(X) (X ) OTf (1), O2CCF3 (2), OAc (3)) were isolated
in high yields (83–95%) and fully characterized. The 31P NMR
spectra of these compounds exhibit a doublet due to coupling
ligand is only weakly activated. No reaction with nitrogen was
observed for trifluoroacetate complex 2 or acetate complex 3.
The PNP rhodium(I) methyl and hydroxide complexes could
not be cleanly prepared by salt metathesis. The methylene groups
on complexes 1 and 1-N2 are susceptible to deprotonation;
reaction of 1 or 1-N2 with strong bases (MeLi, KOtBu, or
LiN(SiMe3)2) resulted in formation of the neutral methylene-
deprotonated dinitrogen complex 4 (eq 2). Neutral dinitrogen
complex 4 was characterized by 1H and 31P NMR spectroscopy,
IR spectroscopy, and elemental analysis. The 31P NMR spectrum
of 4 shows a characteristic A-B pattern of doublets at 67.74
ppm (2JP-P ) 269 Hz, 1JRh-P ) 131 Hz) and 63.98 ppm (2JP-P
) 269 Hz, 1JRh-P ) 131 Hz), indicating inequivalent phosphorus
atoms. The IR spectrum of 4 shows the N-N stretch at 2122
cm-1, slightly red-shifted from the N-N stretch of 1-N2 (2153
cm-1).
1
1
to rhodium, with JRh-P ) 145 Hz for 1, JRh-P ) 151 Hz for
1
2, and JRh-P ) 156 Hz for 3.
(1)
The triflate ligand of complex 1 is readily displaced; in
benzene or THF solution 1 was observed to be in equilibrium
with the monomeric dinitrogen complex [(PNP)Rh(N2)]OTf (1-
N2). When a benzene-d6 solution of 1 at room temperature was
exposed to an atmosphere of N2 and protected from light,
quantitative conversion to 1-N2 occurred over the course of 6
days. Heating or subjecting a solution of 1-N2 to vacuum
resulted in slow re-formation of 1. The monomeric dinitrogen
complex 1-N2 was isolated in 86% yield and fully characterized.
The X-ray structure of 1-N2 (Figure 1) reveals that the N-N
bond distance (1.116(4) Å) is only slightly elongated from that
in free dinitrogen (1.09 Å).14 The IR spectrum of 1-N2 shows
νN-N ) 2153 cm-1, an additional indication that the dinitrogen
(2)
The rhodium(I) methyl complex (PNP)Rh(CH3) (5) was
prepared by transmetalation with dimethylzinc. Reaction of 1
or 1-N2 with dimethylzinc initially resulted in formation of a
rhodium-zinc adduct, formulated as (PNP)Rh(CH3)Zn-
(CH3)O2SOCF3 (6), on the basis of NMR spectroscopy and
1
elemental analysis. The H NMR spectrum of 6 shows two
virtual triplets corresponding to the tert-butyl groups on the
ligand, indicating a low-symmetry structure. The Rh-CH3 group
exhibits a doublet of triplets at 0.71 ppm, with 3JP-H ) 5.7 Hz
2
and JRh-H ) 1.8 Hz, while the Zn-CH3 group appears as a
singlet at 0.45 ppm, integrating to 3H.
(9) Hutschka, F.; Dedieu, A.; Eichberger, M.; Fornika, R.; Leitner, W.
J. Am. Chem. Soc. 1997, 119, 4432–4443.
Hydrolysis of the rhodium-zinc adduct 6 resulted in re-
formation of triflate complex 1. However, reaction of 6 with 1
equiv of bipy (2,2′-bipyridyl) released rhodium(I) methyl
complex 5. The zinc products of this reaction were not
characterized. Complex 5 was isolated in 67% yield by
extraction into pentane and characterized by NMR spectroscopy,
elemental analysis, and X-ray crystallography.13 The 1H NMR
spectrum of complex 5 shows a doublet of triplets for the Rh-
(10) Milet, A.; Dedieu, A.; Kapteijn, G.; van Koten, G. Inorg. Chem.
1997, 36, 3223–3231.
(11) Oxgaard, J.; Tenn, W. J., III; Nielsen, R. J.; Periana, R. A.; Goddard,
W. A., III. Organometallics 2007, 26, 1565–1567.
(12) Cundari, T. R.; Grimes, T. V.; Gunnoe, T. B. J. Am. Chem. Soc.
2007, 129, 13172–13182.
(13) Kloek, S. M.; Heinekey, D. M.; Goldberg, K. I. Angew. Chem.,
Int. Ed. 2007, 46, 4736–4738.
(14) MacKay, B. A.; Fryzuk, M. D. Chem. ReV. 2004, 104, 385–402.

1456 Organometallics, Vol. 27, No. 7, 2008
Hanson et al.
CH3 group at 0.75 ppm, with 3JP-H ) 6.0 Hz and 2JRh-H ) 2.1
Hz. Following an analogous procedure with 1, treatment with
diphenylzinc and then bipy allowed for isolation of the
rhodium(I) phenyl complex (PNP)Rh(C6H5) (7) in 61% yield.
Complex 7 was characterized by NMR spectroscopy and
elemental analysis.
Rhodium(I) hydroxide, trifluoroethoxide, phenoxide, and
4-nitrophenoxide complexes (PNP)Rh(X) (X ) OH (8),
OCH2CF3 (9), OC6H5 (10), OC6H4NO2 (11)) were prepared by
reaction of a solution of 5 in THF with water, trifluoroethanol,
phenol, or 4-nitrophenol (eq 3). However, phenoxide complex
10 and 4-nitrophenoxide complex 11 were prepared more
conveniently by the reaction of 0.5 equiv of [Rh(COD)OH]2
(COD ) 1,5-cyclooctadiene) with 1 equiv of PNP ligand and 1
equiv of the appropriate phenol in toluene and were recrystal-
lized from toluene or ether. Complexes 8–11 were characterized
by 1H and 31P NMR spectroscopy and elemental analysis. The
31P NMR spectra of complexes 8–11 exhibit a doublet with
Figure 2. ORTEP drawing of 10-HOC6H5 (thermal ellipsoids at
50% probability, H atoms and a second molecule of cocrystallized
phenol omitted for clarity). Select bond lengths (Å) and angles
(deg): Rh1-O1 ) 2.1096(18), Rh1-N1 ) 2.037(2), Rh1-P1 )
2.2951(7), Rh1-P2 ) 2.3140(7), O1-O2 ) 2.573(2), P1-Rh1-P2
) 163.68(3), N1-Rh1-O1 ) 178.90(8), P1-Rh1-N1 ) 82.61(6),
P2-Rh1-N1 ) 84.05(6).
1
1
coupling to rhodium of JRh-P ) 157 Hz for 8, JRh-P ) 161
with an authentic sample of complex 7 prepared from 1 and
diphenylzinc. Consistent with this finding, no reaction was
observed when a benzene-d6 solution of complex 7 was treated
with water (1.6 equiv) at room temperature. Similarly, heating
a benzene-d6 solution of trifluoroethoxide complex 9 at 100 °C
for 158 h resulted in partial conversion to 7-d5 (40%), along
with unreacted starting material (40%). Some yellow solid
precipitate was also observed in this reaction. When the reaction
time was extended, increased decomposition was observed, with
no increase in the yield of 7-d5.
Hz for 9, 1JRh-P ) 157 Hz for 10, and 1JRh-P ) 153 Hz for 11.
(3)
The synthesis of phenoxide complex 10 and 4-nitrophenoxide
complex 11 required use of exactly 1 equiv of phenol. Additional
phenol interacted with 10 by hydrogen bonding, which could
be detected by 1H NMR spectroscopy and X-ray crystallography.
When 1 equiv of phenol was added to a benzene-d6 solution of
10, the 1H NMR spectrum showed a set of signals corresponding
to the aryl protons on the phenoxide ligand, as well as signals
arising from the hydrogen-bonded phenol at 7.16, 6.86, and 6.84
ppm, which appeared significantly downfield from the chemical
shifts of free phenol in benzene-d6 (7.04, 6.77, and 6.57 ppm).
The chemical shifts of the phenoxide ligand were found to be
dependent on the concentration of phenol; further increasing
the concentration of phenol caused the multiplet arising from
the four phenoxide aryl protons (7.40–7.29 ppm) to resolve into
distinct doublet and triplet signals and shift downfield. In
contrast, addition of water to a benzene-d6 solution of 10 showed
no changes in the chemical shifts of 10 nor any conversion to
hydroxide complex 8.
(4)
Thermolysis of complexes 8 and 9 in benzene-d6 led to partial
deuterium incorporation into the ligand methylene groups of
the product 7 (40–60%). This deuterium incorporation may
result from exchange with the deuterated water or trifluoro-
ethanol product, respectively, of the reaction. In a separate
experiment, approximately 75% deuterium was exchanged into
the methylene positions when a benzene solution of 7 was
treated with D2O (10 equiv) for 24 h at room temperature (as
1
2
determined by H and H NMR spectroscopy) (eq 5).13
When crystals of 10 were grown in the presence of excess
phenol, the phenoxide complex cocrystallized with 2 equiv of
phenol; one molecule of phenol is oriented directly toward the
phenoxide ligand in a hydrogen-bonding interaction, and the
other is oriented away from the rhodium, adjacent to the ligand
tert-butyl groups. The X-ray structure of 10-HOC6H5 is shown
in Figure 2. The O1-O2 distance of 2.573(2) Å is consistent
with a hydrogen bond between the phenoxide ligand and the
associated phenol.15
B. Stoichiometric Activation of Benzene by (PNP)Rh-
(X) Complexes. As described previously,13 heating a benzene-
d6 solution of hydroxide complex 8 at 60 °C for 95 h resulted
in formation of the deuterated phenyl complex 7-d5 in 60% yield
(eq 4), along with formation of an intractable yellow solid
precipitate. The identity of 7-d5 was confirmed by comparison
(5)
In contrast to the reactivity observed for the hydroxide and
trifluoroethoxide complexes, heating a solution of phenoxide
complex 10 in benzene-d6 resulted in no reaction. Even after
heating for 1 week at 100 °C, no conversion to phenyl complex
7 was observed.16 Consistent with this result, reaction of 7 with
2.5 equiv of phenol resulted in quantitative formation of 10 at
(16) A small amount (ca. 15%) of deuterium incorporation into the
methylene positions of the phenoxide complex 10 was observed when a
solution of 10 was heated in dry benzene-d6 at 100 °C for 118 h.
(15) Fulton, J. R.; Holland, A. W.; Fox, D. J.; Bergman, R. G. Acc.
Chem. Res. 2002, 35, 44–56.

Rh Catalyzed H-D Exchange Between Arenes and Water
Organometallics, Vol. 27, No. 7, 2008 1457
Figure 3. Plot of turnovers of the H-D exchange between benzene-
d6 and H2O for phenoxide complex 10 vs time (h) at 100 °C.
Figure 4. Plot of turnover frequency of the exchange between
benzene-d6 and H2O for phenoxide complex 10 vs [phenol] at 100
°C.
room temperature. Like phenoxide complex 10, rhodium(I)
triflate, trifluoroacetate, acetate, and 4-nitrophenoxide complexes
1, 2, 3, and 11 showed no reaction when heated in benzene at
100 °C for at least one week.
(THF) revealed that approximately 47 turnovers had occurred,
with a selectivity (statistically corrected) of 1 para:1.2 meta:0
ortho:0 CH3.13 Finally, when the exchange reaction between
H2O and benzene-d6 was conducted with 4-nitrophenoxide
complex 11 in the presence of the base Na2CO3 (18 equiv), the
turnover frequency was 0.11(2) turnovers/h, the same as in the
absence of added base (0.10(2) turnovers/h).
Although control experiments indicated that phenol itself does
not catalyze the exchange, added phenol was found to promote
the rhodium-catalyzed reaction. For the H-D exchange between
benzene-d6 and H2O, the turnover frequency of phenoxide
complex 10 was measured at concentrations of phenol ranging
from 0 to 0.8 M. A plot of TOF (turnovers/h) vs [phenol] (Figure
4) reveals that the turnover frequency depends linearly on the
concentration of phenol. Trifluoroethanol also serves to acceler-
ate this reaction. The turnover frequency for catalyst 10
measured in the presence of 5 equiv of trifluoroethanol, 0.48(3)
turnovers/h, is 5 times greater than the turnover frequency
without added alcohol (0.09(2) turnovers/h). In contrast, the
turnover frequency for catalyst 10 with added tert-butanol (5
equiv) (0.085(20) turnover/h) is no different from that in the
absence of added alcohol.
C. Catalytic H-D Exchange between Arenes and
Water. As reported in a preliminary communication, phenoxide
complex 10 catalyzes H-D exchange between arenes and
water.13 Acetate complex 3 and 4-nitrophenoxide complex 11
are also active catalysts for this reaction. The H-D exchange
between benzene-d6 and H2O was conveniently monitored by
1H NMR (eq 6), where the extent of the reaction was determined
by integration of the residual benzene signal against an internal
standard (THF or hexamethylbenzene). No intermediates were
observed in the catalytic reactions; at all times the only rhodium
1
species evident by H NMR was 3, 10, or 11.
(6)
A plot of turnovers versus time was linear for 3, 10, and 11
(Figure 3). The turnover frequency for each catalyst was
determined by averaging three individual runs, with 0.09(2)
turnovers/h for 10, 0.10(2) turnovers/h for 11, and 0.09(2)
turnovers/h for 3. In contrast, trifluoroacetate complex 2 and
triflate complex 1 are poor catalysts for the exchange. Triflate
complex 1 showed only 1.5 turnovers after 150 h at 100 °C,
and trifluoroacetate complex 2 showed no exchange after 150 h
at 100 °C.
Since strong acids are known to catalyze H-D exchange
between arenes and water,17 a number of control experiments
were carried out to ensure that the observed H-D exchange
was not catalyzed by trace acid. Control experiments with
phenol, H2O, and benzene-d6 or toluene-d8 (no rhodium) showed
no reaction, even after 300 h at 100 °C. In addition, no exchange
was observed when acetic acid (0.027 mM) was heated at 100
°C in benzene-d6 with H2O (no rhodium).
When the H-D exchange reaction was conducted with
phenoxide complex 10 in the presence of a high concentration
of phenol (0.9 M), the turnover frequency was sufficiently high
that after 100 h an equilibrium distribution of hydrogen and
deuterium was achieved (approximately 425 turnovers), and no
further increase in the residual benzene solvent signal was
1
observed (Figure 5). Integration of the H NMR spectrum of
the reaction mixture confirmed that no catalyst decomposition
had occurred, and thus the lack of increase in the residual
benzene signal was due to unproductive exchanges.
Protic additives also promote the H-D exchange reaction
between benzene-d6 and H2O when the acetate complex 3 is
the catalyst. With 2 equiv of added acetic acid, the turnover
frequency is 0.30(2) turnovers/h, about 3 times greater than with
no additives (0.09(2) turnovers/h). Phenol also increases the
catalytic activity; the turnover frequency with 3 and 2 equiv of
phenol is 0.23(2) turnovers/h.
In toluene-d8 with H2O, the selectivity of the exchange
catalyzed by phenoxide complex 10 could be determined. A
toluene-d8 solution of 10 was heated with H2O at 100 °C for
360 h, and the residual toluene-h signals were monitored by
1H NMR spectroscopy. Integration against an internal standard
The effect of added sodium acetate on the exchange reaction
with acetate complex 3 was studied. The H-D exchange
between benzene-d6 and H2O with 3 was monitored in the
presence 8, 21, and 32 equiv of NaOAc. Although determination
(17) (a) Lauer, W. M.; Matson, G. W.; Stedman, G. J. Am. Chem. Soc.
1958, 80, 6433–6437. (b) Stock, L. M.; Brown, H. C. J. Am. Chem. Soc.
1959, 81, 3323–3329.

1458 Organometallics, Vol. 27, No. 7, 2008
Hanson et al.
1.116(4) is similar to the value of 1.108(3) determined for the
related neutral dimeric dinitrogen complex [(iPrPCP)Rh]2(N2)
(iPrPCP ) 1,3-bis(diisopropylphosphinomethyl)benzene).19
Neutral methylene-deprotonated dinitrogen complex 4 was
formed from the reaction of 1 with base under a nitrogen
atmosphere (eq 2). The deprotonation of ligand methylene
groups has been observed before for Ru, Ir, Pd, and Pt
complexes bearing similar pyridine-phosphine-based ligands.20,21
This reactivity of the methylene groups with base prevented
the synthesis of methyl complex 5 using alkylating agents such
as MeLi and MeMgBr. Instead, complex 5 was prepared from
the milder reagent ZnMe2, which initially led to formation of
the rhodium-zinc adduct 6. Rhodium adducts of zinc,22
mercury,23 and aluminum24 have been described previously. The
addition of bipy to 6 allowed for the convenient separation of
rhodium methyl complex 5 from the zinc byproducts, presum-
ably due to the coordination of bipy to zinc.
Figure 5. Equilibrium isotopic distribution reached in the exchange
between benzene-d6 and H2O for complex 10 with 0.9 M phenol
at 100 °C.
Phenoxide complex 10 interacts with phenol by hydrogen
bonding, which was evident in both the 1H NMR spectrum and
the X-ray structure of 10-HOC6H5. A strong hydrogen-bonding
interaction between phenol and the related rhodium(I) phenoxide
complex (PMe3)3Rh(OC6H4CH3) has previously been studied
by Bergman and co-workers.25 The hydrogen bonding was
1
detected by H NMR spectroscopy, where the chemical shifts
of both coordinated and hydrogen-bonded phenol were found
to be concentration and temperature dependent. Calorimetry
measurements allowed for the determination of ∆H ) -11.4(5)
kcal/mol for formation of the hydrogen bond in benzene.25 The
X-ray structure of (PMe3)3Rh(OC6H4CH3)-HOC6H4CH3 shows
an O-O distance of 2.62 Å,25 slightly longer than the O1-O2
distance in 10-HOC6H5 (2.573(2) Å). A similar interaction has
also been reported by Vrieze and co-workers, who determined
by 1H NMR spectroscopy that the rhodium(I) phenoxide
complex (NNN)Rh(OC6H5) (NNN ) 2,6-(CHdNiPr)2C5H3N)
forms a hydrogen-bonded adduct of the form (NNN)Rh(OC6H5)-
HOC6H5 with excess phenol.26
Figure 6. Plot of turnovers of the H-D exchange between benzene-
d6 and H2O vs time (h) for acetate complex 3 in the presence of 0
(9), 8 (2), 21 (b), and 32 (*) equiv of sodium acetate at 100 °C.
of the exact concentration of NaOAc in the benzene layer is
not possible due to the biphasic nature of the reaction mixture,
it was qualitatively observed that added NaOAc inhibited the
exchange reaction. With 8, 21, and 32 equiv of added NaOAc,
the turnover frequencies were found to be 0.05(1), 0.04(1), and
0.02(1) turnovers/h, respectively (Figure 6). An analogous
experiment with phenoxide complex 10 and added potassium
phenoxide could not be conducted; an equilibrium between the
potassium phenoxide and water generated phenol, which could
Mechanism of Rhodium-Catalyzed H-D Exchange be-
tween Arenes and Water. Deuterium oxide is an inexpensive
and convenient source of deuterium, and there is considerable
interest in methods for labeling organic molecules with D2O.
Efficient transition metal catalysts for H-D exchange between
1
be detected in the benzene-d6 layer by H NMR spectroscopy
and has an accelerating effect on the reaction (vide supra).
Discussion
(19) Cohen, R.; Rybtchinski, B.; Gandelman, M.; Rozenberg, H.; Martin,
J. M. L.; Milstein, D. J. Am. Chem. Soc. 2003, 125, 6532–6546.
(20) (a) Sacco, A.; Vasapollo, G.; Nobile, C. F.; Piergiovanni, A.;
Pellinghelli, M. A.; Lanfranchi, M. J. Organomet. Chem. 1988, 356, 397–
409. (b) Zhang, J.; Leitus, G.; Ben-David, Y.; Milstein, D. J. Am. Chem.
Soc. 2005, 127, 10840–10841. (c) Zhang, J.; Leitus, G.; Ben-David, Y.;
Milstein, D. Angew. Chem., Int. Ed. 2006, 45, 1113–1115. (d) Gunanathan,
C.; Ben-David, Y.; Milstein, D. Science 2007, 317, 790–792.
(21) Ben-Ari, E.; Leitus, G.; Shimon, L. J. W.; Milstein, D. J. Am. Chem.
Soc. 2006, 128, 15390–15391.
Synthesis of Complexes 1–11. Reaction of [Rh(COE)2Cl]2
with the appropriate silver salt, followed by filtration and the
addition of the PNP ligand, allowed for efficient preparation of
complexes 1-3. In solution, the triflate complex 1 was observed
to be in equilibrium with 1-N2. Although there are multiple
examples of neutral rhodium(I) dinitrogen complexes,18 1-N2
is a rare example of a cationic rhodium(I) dinitrogen complex.
The slight elongation of the N-N bond distance of 1-N2
(1.116(4) Å) compared to free dinitrogen (1.09 Å) is consistent
with relatively weak binding.18 The N-N bond length of
(22) (a) Geerts, R. L.; Huffman, J. C.; Caulton, K. G. Inorg. Chem.
1986, 25, 590–591. (b) Fryzuk, M. D.; McConville, D. H.; Rettig, S. J.
Organometallics 1990, 9, 1359–1360. (c) Fryzuk, M. D.; McConville, D. H.;
Rettig, S. J. Organometallics 1993, 12, 2152–2161.
(23) Cano, M.; Heras, J. V.; Lobo, M. A.; Pinilla, E.; Gutierrez, E.;
Monge, M. A. Polyhedron 1989, 8, 2727–2729.
(18) (a) For examples, see Van Gaal, H. L. M.; Moers, F. G.; Steggerda,
J. J. J. Organomet. Chem. 1974, 65, C43–C45. (b) Urtel, H.; Meier, C.;
Eisentrager, F.; Rominger, F.; Joschek, J. P.; Hofmann, P. Angew. Chem.,
Int. Ed. 2001, 40, 781–784. (c) van der Boom, M. E.; Milstein, D. Chem.
ReV. 2003, 103, 1759–1792. (d) Masuda, J. D.; Stephan, D. W. Can.
J. Chem. 2005, 83, 324–327.
(24) Mayer, J. M.; Calabrese, J. C. Organometallics 1984, 3, 1292–
1298.
(25) Kegley, S. E.; Schaverien, C. J.; Freudenberger, J. H.; Bergman,
R. G.; Nolan, S. P.; Hoff, C. D. J. Am. Chem. Soc. 1987, 109, 6563–6565.
(26) Haarman, H. F.; Kaagman, J. F.; Smeets, W. J. J.; Spek, A. L.;
Vrieze, K. Inorg. Chim. Acta 1998, 270, 34–45.

Rh Catalyzed H-D Exchange Between Arenes and Water
Organometallics, Vol. 27, No. 7, 2008 1459
Scheme 1. Possible Mechanisms for the H-D Exchange Reaction
Table 1. Crystallographic Data Collection Parameters for 1-N2 and
10-HOC6H5
arenes and water are relatively rare,27 with one of the most active
examples being the iridium(III) complex Cp*Ir(PMe3)Cl2
developed by Tilley and Bergman.28As described above, PNP
rhodium(I) aryloxide and acetate complexes are active
catalysts for the H-D exchange reaction. No intermediates
are detected in this process. The reaction is accelerated by
protic additives, such as phenol, trifluoroethanol, and acetic
acid. For phenoxide complex 10, the turnover frequency
depends linearly on the concentration of phenol. The catalytic
activity of acetate complex 3 is inhibited by added NaOAc.
Complexes with more weakly basic ligands, such as triflate
and trifluoroacetate, are inactive for the exchange.
Control experiments indicate that the exchange is not
catalyzed by adventitious acid. The turnover frequency of
catalyst 11 (0.10(2) turnovers/h) is unaffected by added base
(Na2CO3, 0.11(2) turnovers/h). The selectivity of the reaction
of toluene-d8 with H2O catalyzed by 10 (1 para:1.2 meta:0
ortho:0 CH3, statistically corrected) is inconsistent with an acid-
catalyzed electrophilic aromatic substitution mechanism, where
activation would occur preferentially at the ortho and para
positions.17
1-N2
10-HOC6H5
empirical formula
fw
C24H43F3N3O3P2RhS
675.52
C41H60NO3P2Rh
779.75
T (K)
130(2)
130(2)
wavelength (Å)
cryst descrip
space group
unit cell dimens (Å,
deg)
0.71069
prism
P21/n
a ) 8.4223(2)
0.71073
red prism
j
P1
a ) 10.3991(2)
b ) 15.6644(4)
c ) 22.9973(7)
R ) 90
ꢀ ) 93.5385(11)
γ ) 90
b ) 11.5735(2)
c ) 18.0621(4)
R ) 101.4002(8)
ꢀ ) 91.9941(9)
γ ) 110.4954(9)
1983.11(7)
2, 1.306
V (Å3)
3028.26(14)
4, 1.482
Z, F (Mg/m3)
µ (mm-1
F(000)
)
0.787
1400
0.548
824
cryst size
0.24 × 0.24 × 0.18
0.43 × 0.29 × 0.22
mm
mm
no. of reflns for
indexing
502
9361
θ range (deg)
2.96 to 28.28
-11 e h e 11
-20 e k e 20
-30 e l e 30
12 566, 7169
2.10 to 28.50
-13 e h e 13
-15 e k e 15
-21 e l e 24
15 112, 9364
index ranges
Four possible mechanisms for the H-D exchange reaction
are depicted in Scheme 1. Path A is a concerted pathway, where
reaction of the C-H bond occurs in one step. In this path the
oxygen lone pair is involved in forming the new O-H bond
while the R-H bond is breaking.9–12 The reverse reaction with
no. of reflns collected,
unique
completeness to θ
98.8%
99.3%
absorp corr
semiempirical from
equivalents
full-matrix
least-squares on F2
0.935
semiempirical from
equivalents
full-matrix
least-squares on F2
0.971
refinement method
(27) (a) Garnett, J. L.; Hodges, R. J. J. Am. Chem. Soc. 1967, 89, 4546–
4547. (b) Garnett, J. L.; Long, M. A.; Peterson, K. B. Aust. J. Chem. 1974,
27, 1823–1825. (c) Blake, M. R.; Garnett, J. L.; Gregor, I. K.; Hannan,
W.; Hoa, K.; Long, M. A. J. Chem. Soc., Chem. Commun. 1975, 930–932.
(d) Shilov, A. E.; Shteinman, A. A. Coord. Chem. ReV. 1977, 24, 97–143.
(e) Prechtl, M. H. G.; Hoelscher, M.; Ben-David, Y.; Theyssen, N.; Loschen,
R.; Milstein, D.; Leitner, W. Angew. Chem., Int. Ed. 2007, 46, 2269–2272.
(f) Leung, C. W.; Zheng, W.; Wang, D.; Ng, S. M.; Yeung, C. H.; Zhou,
Z.; Lin, Z.; Lau, C. P. Organometallics 2007, 26, 1924–1933.
goodness of fit on F2
R1
0.0450
0.0455
wR (I > 2σI)
0.1132
0.1056
ROD in place of ROH allows for deuterium exchange into the
arene. Supporting this pathway, calculations by Oxgaard and
Periana suggest that this mechanism, resembling σ-bond me-
(28) Klei, S. R.; Golden, J. T.; Tilley, T. D.; Bergman, R. G. J. Am.
Chem. Soc. 2002, 124, 2092–2093.

1460 Organometallics, Vol. 27, No. 7, 2008
Hanson et al.
tathesis, should be accessible to metals with 6 or more
d-electrons, such as the d8 rhodium(I) center.11 However, it is
difficult to reconcile the dependence of the H-D exchange
reaction on protic additives, which might be expected to inhibit
the C-H bond addition by interacting with one of the lone pairs
on the OR ligand.
expected to promote anion dissociation. Finally, it is hypoth-
esized that triflate and trifluoroacetate complexes are inactive
for the exchange due to the low basicity of the triflate and
trifluoroacetate anions. Attempts to test this hypothesis by
conducting the exchange reaction with triflate complex 1 and
base were hampered by the reaction of 1 with base, likely
through the deprotonation of the methylene groups noted above.
In path D, dissociation of the aryloxide or acetate ligand is
followed by deprotonation of one of the methylene groups,
resulting in generation of a neutral methylene-deprotonated
rhodium(I) intermediate. Activation of benzene results in
formation of a neutral methylene-deprotonated rhodium(III)
phenyl hydride complex, which subsequently rearranges to
produce a rhodium(I) phenyl complex. Exchange of ROH for
ROD allows for deuterium incorporation into the arene. A
similar involvement of the methylene protons has been reported
by Milstein and co-workers for the conversion of the methylene-
deprotonated complex Ir(I) COE complex, 12, to the phenyl
complex (PNP)Ir(C6D5) (13) upon heating in benzene-d6 (eq
7).21
Path B involves a direct oxidative addition of the arene C–H
bond to the 16-electron rhodium(I) center. Reductive elimination
of ROH, followed by oxidative addition of ROD, exchanges
deuterium into the arene. The reductive elimination of ROH
and oxidative addition of ROD may be concerted or stepwise
eliminations and additions. Here the argument could be made
that hydrogen bonding of excess phenol to the phenoxide ligand
should reduce the electron density at the rhodium(I) center and
increase steric crowding, making a direct oxidative addition less
favorable in the presence of phenol. However, recent compu-
tational work by Goldman and Krogh-Jespersen indicates that
reducing the π-electron density at the metal center may actually
make C-H oxidative addition to a d8 four-coordinate square-
planar iridium(I) center more favorable.29 Overall, though, very
few examples of direct C-H bond addition to a four-coordinate
square-planar iridium(I) or rhodium(I) center have been
reported.30,31 Also, a direct oxidative addition pathway cannot
account for the inhibition of the acetate catalyst by sodium
acetate.
In path C, the aryloxide or acetate ligand is replaced by
solvent (H2O or arene, H2O is shown in Scheme 1). This could
proceed by preliminary dissociation of -OR followed by solvent
coordination, or by associative substitution of -OR by solvent.
Once the arene has entered the coordination sphere, deproto-
nation of either the resulting σ-complex or phenyl hydride
complex ensues. Exchange of ROH for ROD allows for
incorporation of deuterium into the arene. In order to be
consistent with the experimentally observed inhibition by sodium
acetate, this mechanism must involve a pre-equilibrium loss of
-OR, followed by a rate-determining step of either displacement
of H2O by arene or oxidative addition of benzene.32 Notably,
detailed kinetic studies of the mechanism of benzene activation
by the related square-planar d8 platinum(II) diimine complexes
indicate that both solvent displacement by arene and C-H
oxidative addition may be rate-limiting steps, depending on the
ligand substituents.4 In addition, Tilset and co-workers have
shown that for dimethyl platinum(II) diimine complexes the
metal is the kinetic site of protonation, with the important
implication that the microscopic reverse reaction of oxidative
addition proceeds by deprotonation of a platinum(IV) alkyl
hydride complex.33
(7)
Supporting the possibility that deprotonation of the methylene
backbone may be involved in the PNP rhodium chemistry, as
described above, partial deuteration of the methylene positions
was observed upon stoichiometric benzene activation by hy-
droxide complex 8 and trifluoroethoxide complex 9.16 However,
the methylene groups exchange efficiently with protic solvents
at room temperature (vide supra), making differentiation
between path C and path D on the basis of labeling experiments
impractical. It would be difficult to distinguish the possibility
that the deprotonation of the methylene groups is crucial to the
mechanism from the case where it is an unrelated side reaction.
Nevertheless, path D cannot account for the observed
inhibition by sodium acetate. Although this pathway also
involves a pre-equilibrium loss of aryloxide or acetate ligand,
the participation of the anion in the immediately following
deprotonation step would be expected to offset this in the rate
law and eliminate the overall rate dependence on acetate.34,35
In order for the reaction to exhibit an inhibition by acetate, the
deprotonation by acetate must occur after the rate-limiting step,
as in path C.
Thus, the experimental evidence best supports path C. Most
important in this analysis is that this pathway involves loss of
-OR in the first step, by either associative or dissociative
substitution. A cationic rhodium(I) solvent complex is then
formed, with the rate-limiting step being either displacement
of H2O by arene or C-H oxidative addition. This mechanism
is distinctly different from the ruthenium(II) and iridium(III)
systems studied by Gunnoe and Periana,8 where the OR group
is attached to the metal at the time of C-H bond activation,
and the lone pair electrons of the OR group play a role in the
actual C-H bond cleavage. Notably, pathway C bears remark-
able resemblance to the mechanism believed to be operative in
Consistent with the experimental observations, the pre-
equilibrium loss of aryloxide or acetate ligand should be favored
by protic additives, as hydrogen bonding between the phenoxide
ligand and added phenol could facilitate dissociation of phe-
noxide.34 In addition, increasing the polarity of the reaction
mixture by the addition of protic molecules would also be
(29) 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.
(30) Nuckel, S.; Burger, P. Angew. Chem., Int. Ed. 2003, 42, 1632–
1636.
(31) Martin, M.; Torres, O.; Onate, E.; Sola, E.; Oro, L. A. J. Am. Chem.
Soc. 2005, 127, 18074–18084.
(32) Although determination of a kinetic isotope effect can implicate
C-H oxidative addition as a rate-determining step, the large error associated
with measuring the rate of the exchange reaction between C6H6 and D2O
by 2H NMR prevented investigation of kinetic isotope effects in this system.
(33) Wik, B. J.; Lersch, M.; Tilset, M. J. Am. Chem. Soc. 2002, 124,
12116–12117.
(34) Williams, B. S.; Goldberg, K. I. J. Am. Chem. Soc. 2001, 123, 2576–
2587.
(35) Goldberg, K. I.; Yan, J.; Breitung, E. M. J. Am. Chem. Soc. 1995,
117, 6889–6896.

Rh Catalyzed H-D Exchange Between Arenes and Water
Organometallics, Vol. 27, No. 7, 2008 1461
procedures. Elemental analyses were performed by Atlantic Mi-
crolab Inc. of Norcross, GA.
platinum(II) Shilov chemistry, where the first step is arene or
alkane coordination to a platinum(II) center, followed by
deprotonation, either of a platinum(II) σ-complex or of a
platinum(IV) hydrocarbyl hydride formed by oxidative addition
of the C-H bond.1b
(PNP)Rh(OTf) (1). Under argon, [Rh(COE)2Cl]2 (163.0 mg,
0.227 mmol), and silver triflate (117.3 mg, 0.456 mmol) were
suspended in THF (5 mL). The reaction was stirred for 10 min,
resulting in the precipitation of a light gray solid (AgCl). The
reaction mixture was filtered through a Teflon syringe filter, and
the filter washed with THF (5 mL). Solid PNP ligand (179.3 mg,
0.454 mmol) was immediately added to the filtrate. The reaction
was stirred for 10–15 min, then the solvent was removed under
vacuum. The orange solid was washed with pentane (5 × 1.5 mL)
Conclusions
The ability to activate C-H bonds in the presence of water
or alcohol will be an important characteristic of an effective
catalyst for hydrocarbon oxidation. However, many of the
transition metal complexes known to activate C-H bonds are
incompatible with water or alcohols, which can coordinate to
the metal center and suppress its reactivity toward C-H bonds.
The recent discovery by Gunnoe and Periana of iridium(III) and
ruthenium(II) alkoxide and hydroxide complexes capable of
C-H bond activation represents a potential solution to this
problem.6,7 The addition of a C-H bond across the metal
alkoxide or hydroxide ligand forms a new metal–carbon bond
and releases alcohol or water in one step. The iridium(III) and
ruthenium(II) complexes clearly demonstrate the ability to
activate C-H bonds in the presence of water, catalyzing H-D
exchange between arenes and water.
As described above, rhodium(I) hydroxide, aryloxide, and
acetate complexes are also capable of C-H bond activation in
the presence of water and alcohols. For the H-D exchange
between benzene-d6 and H2O, the turnover frequency of the
aryloxide and acetate complexes is actually increased by added
phenol. On the surface, the reactivity observed for the rhodium
complexes bears similarity to the reactions reported by Gunnoe
and Periana and appears to extend the generality of the reaction.
However, mechanistic studies suggest that unlike the iridium(III)
and ruthenium(II) complexes, where the C-H bond adds across
a coordinated OR group, the rhodium complexes react by
dissociation of the aryloxide or acetate ligand and coordination
of the arene to the metal. This discovery of a different pathway
than previously found in related systems dramatically extends
the range of C-H activation systems capable of operating in
the presence of water and alcohols.
1
and dried under vacuum. Yield: 287 mg (98%). H NMR (C6D6,
300 MHz): δ 6.74 (t, 1H, J ) 7.8 Hz, Py), 6.17 (d, 2H, J ) 7.8
Hz, Py), 2.47 (vt, 4H, J ) 3.3 Hz, CH2), 1.37 (vt, 36H, J ) 6.6
1
Hz, C(CH3)3). 31P{1H} NMR (C6D6): δ 61.78 (d, 2P, JRh-P
)
145 Hz). 19F NMR (C6D6): δ -78.05 (s). Anal. Calcd for
C24H43F3NO3P2RhS: C, 44.52; H, 6.69; N, 2.16. Found: C, 44.88;
H, 6.88; N, 2.19.
(PNP)Rh(N2)OTf (1-N2). Orange solid (PNP)Rh(OTf) (282 mg,
0.435 mmol) was dissolved in THF (30 mL) under an atmosphere
of nitrogen. The dark orange solution was allowed to stand at room
temperature in the dark for 7 days, during which time the color
changed to a pale yellow-orange. The solvent was removed under
vacuum, leaving a pale yellow-orange solid, which was dried under
vacuum 30 min. Yield: 261 mg (89%). 1H NMR (C6D6, 300 MHz):
δ 7.95 (d, 2H, J ) 7.8 Hz, Py), 7.35 (t, 1H, J ) 7.8 Hz, Py), 3.67
(vt, 4H, J ) 4.2 Hz, CH2), 1.11 (vt, 36H, J ) 6.9 Hz, C(CH3)3).
31P{1H} NMR (C6D6): δ 71.71 (d, 2P, 1JRh-P ) 125 Hz). 19F NMR
(C6D6): δ -77.86 (br s). IR (thin film): νN-N ) 2153 cm-1. Anal.
Calcd for C24H43F3N3O3P2RhS: C, 42.67; H, 6.42; N, 6.22. Found:
C, 42.91; H, 6.45; N, 6.20.
(PNP)Rh(O2CCF3) (2). The preparation of complex 2 from the
chloride complex (PNP)Rh(Cl) and AgO2CCF3 has been recently
reported by Milstein and co-workers.39 Under nitrogen,
[Rh(COE)2Cl]2 (139.7 mg, 0.195 mmol) and silver trifluoroacetate
(86.5 mg, 0.391 mmol) were combined in THF (5 mL). The reaction
mixture was stirred 10 min and then filtered through a Teflon
syringe filter to remove solid AgCl. The filter was washed with
THF (3 mL). To the combined filtrate was added solid PNP ligand
(153.8 mg, 0.389 mmol). The resulting red solution was allowed
to stand at room temperature for 30 min, and then the solvent was
removed under vacuum, yielding a bright red solid. The solid was
washed with pentane (3 × 2 mL) and dried under vacuum. Yield:
Experimental Section
1
211.7 mg (89%). H NMR (C6D6, 300 MHz): δ 6.79 (t, 1H, J )
General Considerations. Unless specified otherwise, all reac-
tions were carried out under a dry nitrogen atmosphere using
standard glovebox and Schlenk techniques. Deuterated solvents
were purchased from Cambridge Isotope Laboratories, and benzene-
d6 (99.5% D), toluene-d8, cyclohexane-d12, and THF-d8 were dried
over sodium/benzophenone. Benzene-d6 (99.96% D) was used as
received and stored under N2 at -35 °C. Dimethylzinc (10 wt%
solution in hexanes), diphenylzinc (99%), and di-tert-butylphosphine
were purchased from Strem. All other chemicals were purchased
from Aldrich and used as received. THF, pentane, toluene, and
benzene were dried by passage through activated alumina and
molecular sieve columns. NMR spectra were obtained at room
temperature on Bruker AV300, AV500 MHz, or dpx200 spectrom-
eters, with chemical shifts (δ) reported in ppm downfield of
tetramethylsilane. 31P{1H} and 19F NMR spectra were referenced
externally to 85% H3PO4 and neat CF3COOH (δ -78.5 ppm).
[Rh(COE)2Cl]2,36 [Rh(COD)OH]2,37 and 2,6-bis(di-tert-butylphos-
phinomethyl)pyridine (PNP)38 were prepared according to published
7.8 Hz, Py), 6.19 (d, 2H, J ) 7.8 Hz, Py), 2.52 (vt, 4H, J ) 3.6
Hz, CH2), 1.34 (vt, 36H, J ) 6.3 Hz, C(CH3)3). 31P{1H} NMR
1
(C6D6): δ 61.07 (d, 2P, JRh-P ) 151 Hz). 19F NMR (C6D6): δ
-74.87 (s). Anal. Calcd for C25H43F3NO2P2Rh: C, 49.11; H, 7.09;
N, 2.29. Found: C, 49.31; H, 7.23; N, 2.35.
(PNP)Rh(OAc) (3). Under nitrogen, [Rh(COE)2Cl]2 (107.3 mg,
0.150 mmol) and silver acetate (50.4 mg, 0.302 mmol) were
combined in THF (5 mL). The reaction mixture was stirred 10 min
and then filtered through a Teflon syringe filter. The filter was
washed with THF (3 mL). To the combined filtrate was added solid
PNP ligand (118.1 mg, 0.299 mmol), forming a red-brown solution
upon mixing. The solvent was removed under vacuum, and the
resulting red-brown solid washed with pentane (2 × 3 mL). Yield:
1
138.5 mg (83%). H NMR (C6D6, 500 MHz): δ 6.87 (t, 1H, J )
7.5 Hz, Py), 6.23 (d, 2H, J ) 7.5 Hz, Py), 2.56 (vt, 4H, J ) 3.5
Hz, CH2), 2.28 (s, 3H, OAc), 1.44 (vt, 36H, J ) 6.5 Hz, C(CH3)3).
31P{1H} NMR (C6D6) δ 59.75 (d, 2P, 1JRh-P ) 156 Hz). 13C{1H}
NMR (THF-d8, 50 MHz): δ 174.91 (s, OAc), 165.64 (vt, J ) 6.5
Hz, Py), 130.73 (s, Py), 120.64 (vt, J ) 5.5 Hz, Py), 36.40 (vt, J
) 5.2 Hz, CH2), 35.06 (vt, J ) 5.7 Hz, C(CH3)3), 29.81 (vt, J )
(36) van der Ent, A.; Onderdelinden, A. L. Inorg. Synth. 1990, 28, 90–
92.
(37) Uson, R.; Oro, L. A.; Cabeza, J. A. Inorg. Synth. 1985, 23, 126–
130.
(39) Feller, M.; Ben-Ari, E.; Gupta, T.; Shimon, L. J. W.; Leitus, G.;
Diskin-Posner, Y.; Weiner, L.; Milstein, D. Inorg. Chem. 2007, 10479–
10490.
(38) Hermann, D.; Gandelman, M.; Rozenberg, H.; Shimon, L. J. W.;
Milstein, D. Organometallics 2002, 21, 812–818.

1462 Organometallics, Vol. 27, No. 7, 2008
Hanson et al.
4.4 Hz, C(CH3)3), 24.87 (s, OAc). Anal. Calcd for C25H46NO2P2Rh:
C, 53.86; H, 8.32; N, 2.51. Found: C, 53.80; H, 8.19; N, 2.48.
(PN-P)Rh(N2) (4). Under nitrogen, (PNP)Rh(OTf) (82.5 mg,
0.127 mmol) was dissolved in THF (4 mL), and the solution was
cooled to -35 °C. Methyllithium (95 µL of a 1.4 M solution in
diethyl ether, 0.133 mmol) was added and the solution allowed to
warm to room temperature, turning a bright red color. The solvent
was removed under vacuum. The remaining red residue was
extracted into pentane (20 mL) and filtered through a Teflon syringe
filter. The pentane was then removed under vacuum to give a bright
(d, 2P, 1JRh-P ) 124 Hz). Anal. Calcd for C26H49F3NO3P2RhSZn:
C, 42.03; H, 6.65; N, 1.89. Found: C, 42.00; H, 6.78; N, 1.91.
(PNP)Rh(C6H5) (7). In a thick-walled glass vessel fitted with a
Teflon stopcock, (PNP)Rh(OTf) (43.9 mg, 0.0677 mmol) was
suspended in benzene (4 mL). A suspension of solid diphenylzinc
(44.6 mg, 0.203 mmol) in benzene (2 mL) was added, and the
reaction mixture was heated at 60 °C for 17 h. After cooling to
room temperature, 2,2′-dipyridyl (bipy) (10.7 mg, 0.0686 mmol)
was added to the yellow solution, resulting in a color change to
brown. The solvent was removed under vacuum, leaving a brown
residue. Pentane (15 mL) was added, and the mixture was sonicated
for 1 h, giving a dark brown solution and light brown-yellow solid
material. The pentane solution was filtered through a Teflon syringe
filter, concentrated to approximately 1 mL, and cooled to -35 °C
1
red solid. Yield: 41.3 mg (61%). H NMR (cyclohexane-d12, 300
MHz): δ 6.26 (m, 1H, Py), 6.09 (d, 1H, J ) 8.7 Hz, Py), 5.36 (d,
1H, J ) 6.6 Hz, Py), 3.46 (d, 1H, JH-P ) 3.9 Hz, P-CH-Py), 3.01
(d, 2H, JH-P ) 8.1 Hz, P-CH2-Py), 1.56 (m, 36H, C(CH3)3).
31P{1H} NMR (cyclohexane-d12): δ 67.74 (A-B pattern of
1
to yield brown, solid (PNP)Rh(C6H5). Yield: 24.0 mg (61%). H
2
1
NMR (C6D6, 300 MHz): δ 8.17 (d, 2H, J ) 7.5 Hz, Rh-Ph), 7.26
(t, 2H, J ) 7.5 Hz, Rh-Ph), 6.98 (t, 2H, J ) 7.5 Hz, Rh-Ph (1H),
Py (1H)), 6.46 (d, 2H, J ) 7.5 Hz, Py), 2.86 (vt, 4H, J ) 3.0 Hz,
CH2), 1.26 (vt, 36H, J ) 6.0 Hz, C(CH3)3). 31P{1H} NMR (C6D6)
δ 59.40 (d, 2P, 1JRh-P ) 173 Hz). 13C{1H} NMR (THF, 50 MHz):
doublets, 1P, JP-P ) 269 Hz, JRh-P ) 131 Hz), 63.98 (A-B
2
1
pattern of doublets, 1P, JP-P ) 269 Hz, JRh-P ) 131 Hz). IR
(thin film): νN-N ) 2122 cm-1. Anal. Calcd for C23H42N3P2Rh: C,
52.57; H, 8.06; N, 8.00. Found: C, 52.38; H, 7.93; N, 7.82.
(PNP)Rh(CH3) (5). In a thick-walled glass vessel equipped with
a Teflon stopcock, a solution of (PNP)Rh(OTf) (151.0 mg, 0.233
mmol) in THF (5 mL) was combined with ZnMe2 (0.370 mL of a
10 wt % solution in hexanes, 0.256 mmol). The glass vessel was
sealed and heated at 70 °C for 19 h. The reaction was cooled to
room temperature, and the solvent was removed under vacuum,
leaving a brown residue. The brown residue was dissolved in THF
(5 mL), and 2,2′-bipyridyl (bipy) (36.3 mg; 0.233 mmol) was added,
resulting in an immediate color change to bright pink. The solvent
was removed under vacuum, and the purple solid was extracted
into pentane (50 mL), leaving behind a yellow solid residue. The
purple pentane solution was filtered, and the solvent was removed
under vacuum to yield a dark purple solid. Yield: 80.1 mg (67%).
1H NMR (C6D6, 300 MHz): δ 7.07 (t, 1H, J ) 7.5 Hz, Py), 6.47
(d, 2H, J ) 7.5 Hz, Py), 2.77 (vt, 4H, J ) 3.3 Hz, CH2), 1.39 (vt,
1
2
δ 168.70 (dt, JRh-C ) 32 Hz, JP-C ) 14 Hz, Rh-C), 161.68 (br
s, Py), 142.55 (s, Ph), 131.21 (s, Py), 123.30 (s, Ph), 118.97 (s,
Py), 116.83 (s, Ph), 37.38 (br s, CH2), 35.12 (br s, C(CH3)3), 29.11
(br s, C(CH3)3). Anal. Calcd for C29H48NP2Rh: C, 60.52; H, 8.41;
N, 2.43. Found: C, 60.51; H, 8.31; N, 2.44.
(PNP)Rh(OH) (8). In a glass vessel equipped with a Teflon
stopcock, (PNP)Rh(CH3) (58.0 mg, 0.113 mmol) was dissolved in
THF (1.5 mL). Approximately 10 drops of degassed, deionized
water were added, and the reaction mixture was stirred for 15 min.
The THF was subsequently removed under vacuum, yielding a dark
pink residue. The residue was dissolved in a minimum amount of
THF and crystallized by slow diffusion of pentane at -35 °C. Yield:
45.6 mg (78%). 1H NMR (C6D6, 300 MHz): δ 6.97 (t, 1H, J ) 7.5
Hz, Py), 6.22 (d, 2H, J ) 7.5 Hz, Py), 2.49 (vt, 4H, J ) 3.3 Hz,
CH2), 1.45 (vt, 36H, J ) 6.3 Hz, C(CH3)3), -2.03 (br s, 1H, Rh-
3
2
36H, J ) 6.3 Hz, C(CH3)3), 0.75 (dt, 3H, JP-H ) 6.0 Hz, JRh-H
) 2.1 Hz, Rh-CH3). 31P{1H} NMR (C6D6): δ 60.07 (d, 2P, 1JRh-P
) 169 Hz). 13C{1H} NMR (C6D6, 50 MHz): δ 161.47 (br s, Py),
128.63 (s, Py), 119.72 (br s, Py), 38.87 (br s, CH2), 35.12 (br s,
C(CH3)3), 30.16 (br s, C(CH3)3), -22.70 (dt, 1JRh-C ) 25 Hz, 2JP-C
) 12 Hz, Rh-CH3). Anal. Calcd for C24H46NP2Rh: C, 56.14; H,
9.03; N, 2.73. Found: C, 55.85; H, 9.00; N, 2.73.
1
OH). 31P{1H} NMR (C6D6): δ 57.60 (d, 2P, JRh-P ) 157 Hz).
Anal. Calcd for C23H44NOP2Rh: C, 53.59; H, 8.60; N, 2.72. Found:
C, 53.63; H, 8.59; N, 2.60.
(PNP)Rh(OCH2CF3) (9). (PNP)Rh(CH3) (22.1 mg, 0.0431
mmol) was added to a glass vessel equipped with a Teflon stopcock
containing a solution of trifluoroethanol (0.3 mL) in THF (1 mL).
The solvent was removed under vacuum. The remaining red residue
was dissolved in toluene (1 mL), and the toluene was removed
under vacuum. The red solid product was then recrystallized from
X-ray structure determination of 5:13 Crystallographic data for
this compound is available from the Cambridge Database CCDC
#640089. Crystals were grown from a concentrated pentane solution
at -35 °C. The structure was refined by full-matrix least-squares
on F2. Crystal data for (PNP)Rh(CH3): C24H46NP2Rh, MW )
513.47, orthorhombic, space group P212121, T ) 131(2) K, a )
13.0520(4) Å, b ) 13.8410(5) Å, c ) 14.5260(6) Å, ꢀ ) 90°, Z )
4, R1 ) 0.0531, wR2 ) 0.1277, GOF (F2) ) 0.961, absolute
structure parameter 0.01(5).
1
pentane at –35 °C. Yield: 12.8 mg (50%). H NMR (C6D6, 300
MHz): δ 6.85 (t, 1H, J ) 7.5 Hz, Py), 6.19 (d, 2H, J ) 7.5 Hz,
Py), 4.48 (q, 2H, 3JH-F ) 10.5 Hz, -OCH2CF3), 2.48 (vt, 4H, J )
3.3 Hz, CH2), 1.38 (vt, 36H, J ) 6.6 Hz, C(CH3)3). 31P{1H} NMR
1
(C6D6): δ 57.57 (d, 2P, JRh-P ) 161 Hz). 19F NMR (C6D6): δ
3
-76.64 (t, 3F, JH-F ) 10.2 Hz). 13C{1H} NMR (THF-d8, 50
MHz): δ 165.31 (vt, J ) 5.6 Hz, Py), 129.86 (s, Py), 127.92 (q,
1JC-F ) 284.9 Hz, -OCH2CF3), 120.45 (vt, J ) 5.1 Hz, Py), 76.68
(PNP)Rh(CH3)Zn(CH3)OTf (6). In a thick-walled glass vessel
equipped with a Teflon stopcock, (PNP)Rh(N2)OTf (81.3 mg, 0.120
mmol) was dissolved in THF (5 mL). Dimethylzinc (0.260 mL of
a 10 wt % solution in hexanes, 0.180 mmol) was added under
nitrogen. The glass vessel was sealed and heated at 70 °C for 19 h,
during which time the color changed from yellow to brown. The
reaction was cooled to room temperature and the solution concen-
trated to 0.75 mL under vacuum. Slow diffusion of pentane into
the concentrated THF solution at -35 °C yielded bright yellow
crystals, which were washed with pentane (3 × 1 mL). Yield: 54.8
mg (61%). 1H NMR (C6D6, 300 MHz): δ 7.13 (t, 1H, J ) 7.5 Hz,
Py), 6.81 (d, 2H, J ) 7.5 Hz, Py), 3.68 (dvt, 2H, JH-H ) 17.7 Hz,
JH-P ) 4.2 Hz, CH2), 2.97 (dvt, 2H, JH-H ) 17.7 Hz, JH-P ) 4.5
Hz, CH2), 1.23 (vt, 18 H, J ) 6.9 Hz, C(CH3)3), 0.98 (vt, 18H, J
2
(q vt, JC-F ) 29.9 Hz, J ) 2.6 Hz, -OCH2CF3), 36.73 (vt, J )
6.8 Hz, CH2), 35.14 (vt, J ) 5.1 Hz, C(CH3)3), 29.71 (vt, J ) 4.9
Hz, C(CH3)3). Anal. Calcd for C25H45NF3OP2Rh: C, 50.26; H, 7.59;
N, 2.34. Found: C, 50.41; H, 7.84; N, 2.43.
(PNP)Rh(OC6H5) (10). A solution of phenol (7.6 mg, 0.0809
mmol) in THF (1 mL) was added to a glass vial containing
(PNP)Rh(CH3) (41.5 mg, 0.0809 mmol), forming a deep red
solution. The solvent was removed under vacuum, and the remain-
ing red solid washed with cold pentane (1 mL). Yield: 37.7 mg
(79%).
Alternate Preparation of 10. In a glass vessel equipped with a
Teflon stopcock, [Rh(COD)OH]2 (65.4 mg, 0.1434 mmol), PNP
ligand (112.7 mg, 0.2853 mmol), and phenol (26.5 mg, 0.2819
mmol) were dissolved in toluene (6 mL). The reaction was heated
at 100 °C for 24 h, then cooled to room temperature. The solvent
3
2
) 6.0 Hz, C(CH3)3), 0.71 (dt, 3H, JP-H ) 5.7 Hz, JRh-H ) 1.8
Hz, Rh-CH3), 0.45 (s, 3H, Zn-CH3). 31P{1H} NMR (C6D6): δ 62.34

Rh Catalyzed H-D Exchange Between Arenes and Water
Organometallics, Vol. 27, No. 7, 2008 1463
was removed under vacuum, leaving a red residue. The red solid
was recrystallized from diethyl ether at -35 °C, giving dark purple
crystals of 10, which were washed with pentane (2 × 1 mL). Yield:
at longer times the rate slowed as the reaction approached an
equilibrium distribution of hydrogen and deuterium (approximately
425 turnovers).
1
90.6 mg (54%). H NMR (C6D6, 300 MHz): δ 7.40–7.29 (m, 4H,
H-D Exchange between Toluene-d8 and H2O. A resealable
Teflon-capped NMR tube was charged with deionized H2O (33 µL)
and toluene-d8 (0.30 mL). The mixture was degassed by three
freeze–pump–thaw cycles. (PNP)Rh(OC6H5) (2.9 mg, 0.00491
mmol) and THF (0.00455 mmol) were added, and the tube was
placed under argon. The NMR tube was heated in a stainless steel
jacket in a Neslab Exacal EX-250 HT oil bath at 100 °C and
phenoxide), 6.85 (t, 1H, J ) 7.8 Hz, Py), 6.70 (t, 1H, J ) 6.9 Hz,
phenoxide), 6.22 (d, 2H, J ) 7.8 Hz, Py), 2.57 (vt, 4H, J ) 3.0
Hz, CH2), 1.34 (vt, 36H, J ) 5.7 Hz, C(CH3)3). 31P{1H} NMR
1
(C6D6): δ 57.71 (d, 2P, JRh-P ) 157 Hz). Anal. Calcd for
C29H48NOP2Rh: C, 58.88; H, 8.17; N, 2.37. Found: C, 58.30; H,
8.17; N, 2.39.
1
(PNP)Rh(OC6H4NO2) (11). In a dry glass vial, [Rh(COD)OH]2
(50.0 mg, 0.110 mmol), PNP ligand (86.6 mg, 0.219 mmol), and
4-nitrophenol (30.2 mg, 0.217 mmol) were combined. Toluene (3.5
mL) was added, and the reaction mixture was allowed to stand
undisturbed for 3 days at room temperature, during which time large
red crystals grew in the vial. The supernatant was removed, and
the crystals were washed with pentane (3 × 2 mL). Yield: 123.5
mg (89%). 1H NMR (C6D6, 500 MHz): δ 8.41 (d, 2H, J ) 9.0 Hz,
aryloxide), 6.89 (d, 2H, J ) 9.0 Hz, aryloxide), 6.81 (t, 1H, J )
8.0 Hz, Py), 6.22 (d, 2H, J ) 8.0 Hz, Py), 2.53 (vt, 4H, J ) 3.5
Hz, CH2), 1.20 (vt, 36H, J ) 6.0 Hz, C(CH3)3). 31P{1H} NMR
monitored by H NMR spectroscopy. The residual toluene meta
and para signals increased, but no increase in the ortho and CH3
positions was observed. After 336 h, a total of 47 turnovers had
occurred, with a selectivity of 1.2 meta:1 para:0 ortho:0 CH3
(statistically corrected) determined from the integrals of the residual
toluene signals (versus the THF internal standard). A control
reaction with phenol (3.1 mg, 0.0330 mmol) and no rhodium
showed no detectible increase in the residual toluene signals after
360 h.
Deuterium Exchange into the Methylene Protons of
(PNP)Rh(C6H5) from D2O. (PNP)Rh(C6H5) (3.0 mg; 0.00521
mmol) was dissolved in benzene (0.25 mL). D2O (1 µL; 0.0550
mmol) was added and the mixture allowed to react for 24 h at room
temperature. Examination of the reaction mixture by 2H NMR
showed a broad signal at 2.8 ppm, indicating deuterium had
exchanged into the methylene positions. The volatiles were removed
1
(C6D6): δ 59.19 (d, 2P, JRh-P ) 153 Hz). 13C{1H} NMR (THF-
d8, 50 MHz): δ 180.79 (s, aryloxide), 166.20 (vt, J ) 5.0 Hz, Py),
133.94 (s, aryloxide), 132.47 (s, Py), 126.06 (s, aryloxide), 121.55
(s, aryloxide), 120.89 (vt, J ) 5.7 Hz, Py), 36.21 (vt, J ) 6.4 Hz,
CH2), 35.16 (vt, J ) 7.2 Hz, C(CH3)3), 29.90 (vt, J ) 3.4 Hz,
C(CH3)3). Anal. Calcd for C29H47N2O3P2Rh: C, 54.72; H, 7.44; N,
4.40. Found: C, 55.05; H, 7.51; N, 4.36.
1
under vacuum, and upon examination by H NMR (benzene-d6),
the methylene signal integrated at only 25% of its normal intensity.
General Procedure for the H-D Exchange Reactions. In an
NMR tube fitted to a vacuum adaptor, the rhodium catalyst (ca. 3
mg, 0.005 mmol) was dissolved in benzene-d6 (99.96% D, 350 µL)
containing THF or hexamethylbenzene as an internal standard.
Deionized water (20 µL, 1.11 mmol) was added, and the NMR
tube was flame-sealed under active vacuum. The tube was heated
submerged in a steel jacket in a Neslab Exacal EX-250 HT oil
bath at 100 °C. The reaction was monitored at intervals over the
course of 150 h, with the extent of the exchange being determined
by integration of the residual benzene signal against the internal
standard. 1H NMR spectra were recorded using a 360 s delay (one
scan) to ensure accurate integration of the benzene. For the reactions
conducted with high concentrations of phenol (0.5–0.8 M), the rate
of exchange was determined from the first 20 h of reaction time;
Acknowledgment. This work was supported by the NSF
as part of the Center for Enabling New Technologies through
Catalysis-CENTC (Grant No. 0434568). We thank W. D.
Jones, J. Kovach, W. Borden, D. Hrovat, and A. S. Goldman
for helpful discussions. The crystal structures of 1-N2, 5, and
10-HOC6H5 were solved by A. G. Dipasquale (1-N2), W.
Kaminsky (5), and R. D. Swartz II (10-HOC6H5).
Supporting Information Available: CIF files for 1-N2, 5, and
10-HOC6H5. This material is available free of charge via the Internet
at http://pubs.acs.org.
OM7012259