Synthesis and investigation of [Cp*(PMe3)Rh(H)(H2)]+ and its partially deuterated and tritiated isotopomers: Evidence for a hydride/dihydrogen structure

Article abstract of DOI:10.1021/ja0165990

Hydrogenolysis of [Cp*(PMe₃)Rh(Me)(CH₂Cl₂)]⁺ BAr'₄^- (4, Ar′ = 3,5-C₆H₃(CF₃)₂) in dichloromethane afforded the nonclassical polyhydride complex [Cp*(PMe₃)Rh(H)(H₂)]⁺ BAr′₄^- (1), which exhibits a single hydride resonance at all accessible temperatures in the ¹H NMR spectrum. Exposure of solutions of 1 to D₂ or T₂ gas resulted in partial isotopic substitution in the hydride sites. Formulation of 1 as a hydride/dihydrogen complex was based upon T₁ (T₁(min) = 23 ms at 150 K, 500 MHz), J_H-D (ca. 10 Hz), and J_H-T (ca. 70 Hz) measurements. The barrier (ΔG?) to exchange of hydride with dihydrogen sites was determined to be less than ca. 5 kcal/mol. Protonation of Cp*(PMe₃)Rh(H)₂ (2) using H(OEt₂)₂BAr'₄ resulted in binuclear species [(Cp*(PMe₃)Rh(H))₂(u-H)]⁺ BAr'₄^- (3), which is formed in a reaction involving 1 as an intermediate. Complex 3 contains two terminal hydrides and one bridging hydride ligand which exchange with a barrier of 9.1 kcal/mol as observed by ¹H NMR spectroscopy. Additionally, the structures of 3 and 4, determined by X-ray diffraction, are reported.

Full text of DOI:10.1021/ja0165990

Published on Web 04/10/2002
Synthesis and Investigation of [Cp*(PMe3)Rh(H)(H2)]+ and Its
Partially Deuterated and Tritiated Isotopomers: Evidence for
a Hydride/Dihydrogen Structure
Felicia L. Taw,† Heather Mellows,‡ Peter S. White,† Frederick J. Hollander,§
Robert G. Bergman,*,§ Maurice Brookhart,*,† and D. Michael Heinekey*,‡
Contribution from the Department of Chemistry, UniVersity of North Carolina at Chapel Hill,
CB #3290, Chapel Hill, North Carolina 27599-3290, Department of Chemistry,
UniVersity of Washington, Box 351700, Seattle, Washington 98195-1700, and
Department of Chemistry, UniVersity of California and DiVision of Chemical Sciences,
Lawrence Berkeley National Laboratory, Berkeley, California 94720
Received July 12, 2001
Abstract: Hydrogenolysis of [Cp*(PMe3)Rh(Me)(CH2Cl2)]+BAr′4 (4, Ar′ ) 3,5-C6H3(CF3)2) in dichlo-
-
romethane afforded the nonclassical polyhydride complex [Cp*(PMe3)Rh(H)(H2)]+BAr′4- (1), which exhibits
a single hydride resonance at all accessible temperatures in the 1H NMR spectrum. Exposure of solutions
of 1 to D2 or T2 gas resulted in partial isotopic substitution in the hydride sites. Formulation of 1 as a
hydride/dihydrogen complex was based upon T1 (T1(min) ) 23 ms at 150 K, 500 MHz), JH-D (ca. 10 Hz),
and JH-T (ca. 70 Hz) measurements. The barrier (∆Gq) to exchange of hydride with dihydrogen sites was
determined to be less than ca. 5 kcal/mol. Protonation of Cp*(PMe3)Rh(H)2 (2) using H(OEt2)2BAr′4 resulted
-
in binuclear species [(Cp*(PMe3)Rh(H))2(µ-H)]+BAr′4 (3), which is formed in a reaction involving 1 as an
intermediate. Complex 3 contains two terminal hydrides and one bridging hydride ligand which exchange
1
with a barrier of 9.1 kcal/mol as observed by H NMR spectroscopy. Additionally, the structures of 3 and
4, determined by X-ray diffraction, are reported.
Introduction
of H2 σ electrons to an empty metal dσ orbital balanced by back-
donation of metal dπ electrons to the H2 σ* orbital. Excessive
The study of transition metal polyhydrides comprises a large
portion of organometallic chemistry.1 In particular, these poly-
hydride complexes play a prominent role in many homogeneous
catalytic processes, including hydrogenation of unsaturated
bonds.2 Such complexes have been shown to adopt both classical
structures with terminal hydride ligands and nonclassical
structures containing one or more η2-H2 ligands. The first
classical polyhydride, Fe(H)2(CO)4, was discovered by Hieber
in 1931.3 However, it was not until 1984 that Kubas reported
the first stable nonclassical dihydrogen complex, W(CO)3(Pi-
Pr3)2(H2).4
back-donation into the H2 σ* orbital leads to scission of the
H-H bond and thus formation of terminal hydride species.
The factors determining which structure is adopted in a
particular case are not well understood. Since transition metals
generally form stronger M-H bonds upon descending a column
of the periodic table, classical hydride structures are favored
for 5d metals. Specifically, a comparison of 4d vs 5d metals
with similar co-ligands indicates that the second row metal is
often a dihydrogen complex while the third row metal prefers
the dihydride structure. Numerous examples from the literature
support this idea, such as Tp*Rh(H)2(H2)6 vs Tp*IrH47 (Tp* )
A concise summary of the standard bonding schemes for
metal-dihydrogen and metal-hydride bonds has been offered
by Crabtree.5 Metal-dihydrogen interactions involve donation
+ 10
HB(3,5-Me2pz)3),8 [CpRu(P-P)(H2)+]9 vs [CpOs(P-P)(H)2
]
(P-P ) dppe, dppp), and (PCy3)2Ru(H)2(H2)211 vs (PiPr2Ph)2Os-
(H)6.12 However, exceptions exist in which both second- and
† University of North Carolina at Chapel Hill.
(6) (a) Bucher, U. E.; Lengweiler, T.; Nanz, D.; von Philipsborn, W.; Venanzi,
L. M. Angew. Chem., Int. Ed. Engl. 1990, 29, 548. (b) Nanz, D.; von
Philipsborn, W.; Bucher, U. E.; Venanzi, L. M. Magn. Reson. Chem. 1991,
29, S38.
‡ University of Washington.
§ University of California and Division of Chemical Sciences.
(1) Recent reviews: (a) Maseras, F.; Lledos, A.; Clot, E.; Eisenstein, O. Chem.
ReV. 2000, 100, 601. (b) Sabo-Etienne, S.; Chaudret, B. Chem. ReV. 1998,
98, 2077. (c) Lin, Z.; Hall, M. B. Coord. Chem. ReV. 1994, 135/136, 845.
(d) Heinekey, D. M.; Oldham, W. J., Jr. Chem. ReV. 1993, 93, 913. (e)
Crabtree, R. H. Angew. Chem., Int. Ed. Engl. 1993, 32, 789. (f) Jessop, P.
G.; Morris, R. H. Coord. Chem. ReV. 1992, 121, 155.
(2) Esteruelas, M. A.; Oro, L. A. Chem. ReV. 1998, 98, 577.
(3) Hieber, W.; Leutert, F. Naturwissenschaften 1931, 19, 360.
(4) Kubas, G. J.; Ryan, R. R.; Swanson, B. I.; Vergamini, P. J.; Wasserman,
H. J. J. Am. Chem. Soc. 1984, 106, 451.
(5) Hamilton, D. G.; Crabtree, R. H. J. Am. Chem. Soc. 1988, 110, 4126 and
references therein.
(7) Paneque, M.; Poveda, M. L.; Taboada, S. J. Am. Chem. Soc. 1994, 116,
4519.
(8) Trofimenko, S. Chem. ReV. 1993, 93, 943.
(9) Conroy-Lewis, F. M.; Simpson, S. J. J. Chem. Soc., Chem. Commun. 1987,
1675.
(10) Jia, G.; Ng, W. S.; Yao, J.; Lau, C. P.; Chen, Y. Organometallics 1996,
15, 5039.
(11) Borowski, A. F.; Donnadieu, B.; Daran, J. C.; Sabo-Etienne, S.; Chaudret,
B. Chem. Commun. 2000, 543.
(12) Howard, J. A. K.; Johnson, O.; Koetzle, T. F.; Spencer, J. L. Inorg. Chem.
1987, 26, 2930.
9
5100
J. AM. CHEM. SOC. 2002, 124, 5100-5108
10.1021/ja0165990 CCC: $22.00 © 2002 American Chemical Society

Evidence for a Hydride/Dihydrogen Structure
A R T I C L E S
Scheme 1
third-row metal analogues with similar co-ligands adopt non-
classical structures. Examples include the dihydrogen complexes
M(CO)3(PiPr3)2(H2) (M ) Mo, W),4 [Tp(PMe3)M(H)(H2)]+ (M
) Rh, Ir),13 and [MCl(H2)(depe)2]+ (M ) Ru, Os).14
In these and other complexes, the choice of ligands around
the metal center also plays a key role in defining structure. For
example, Hay has calculated that replacing all the CO ligands
with phosphines in W(CO)3(PiPr3)2(H2) will result in dihydride
formation.15 The CO ligands serve as π-acceptors which stabilize
filled metal d orbitals, thereby inhibiting excessive back-
donation of metal dπ electrons to the H2 σ* orbital. This is
consistent with the observation that W(CO)3(PiPr3)2(H2) contains
16
a dihydrogen ligand but Mo(PMe3)5(H)2 contains terminal
hydrides.
Theoretical studies of the LRh′′H4′′ (L ) Cp, Tp) system
suggest that when L ) Cp a classical tetrahydride structure is
more stable, whereas if L ) Tp a nonclassical isomer is
preferred.17 This difference in behavior can be rationalized by
the observation that Cp is a stronger electron-donor toward Rh
than Tp, thereby contributing more electron density to the H2
antibonding σ* orbital. In addition, the character of the Tp ligand
causes octahedral geometries to be favored (TpRh(H)2(H2) is
octahedral, whereas TpRh(H)4 is not), while Cp complexes can
adopt three or four-legged piano-stool structures.17 For example,
structure would be energetically favored. While the presence
of electron-donating Cp* and PMe3 ligands might induce
hydride formation, rhodium is a second-row metal which prefers
lower oxidation states. Theoretical studies by Hall on the model
complex [Cp(PH3)Rh′′H3′′]+ suggested that the hydride/dihy-
drogen structure and the classical trihydride were very similar
in energy, differing only by 0.5 kcal/mol.25
Kubas has shown that partial substitution of deuterium in
dihydrogen complexes allows direct measurement of JH-D
values, which can be used to verify η2-H2 bonding.4 Similarly,
Crabtree has shown that short T1 times are also indicative of
bound H2 ligands, which display rapid dipole-dipole relaxation
due to the short H-H distances.5,26 A combination of these
techniques is used here, in addition to examination of partially
tritiated complexes, to determine that [Cp*(PMe3)Rh(H)-
while [Cp(PMe3)Ir(H)3 ]
+ 18 and Cp*(PCy3)Ru(H)319 form clas-
sical polyhydrides, the Tp analogues, [Tp(PMe3)Ir(H)(H2)+]20
and Tp*(PCy3)Ru(H)(H2),21 contain η2-H2 ligands. However,
inconsistencies in this pattern exist which cannot be easily
22
explained. For instance, Cp*Ir(H)4 contains only terminal
20
hydride ligands, as expected, but the Tp* analogue Tp*Ir(H)4
also seems to adopt a classical structure based upon chemical
behavior and T1 measurements.
-
(H2)]+BAr′4 adopts a hydride/dihydrogen structure.
The Ir(V) complex [Cp(PMe3)Ir(H)3]+ was shown by Hei-
nekey in 1990 to be a classical trihydride exhibiting quantum
mechanical exchange coupling between the hydrogen nuclei.18
The Cp* analogue [Cp*(PMe3)Ir(H)3]+, which was first char-
acterized by Bergman23 in 1985, was also shown by Heinekey24
in 1996 to be a classical trihydride. This complex undergoes a
rapid hydride permutation process which leads to a single
Additionally, synthesis and characterization of a binuclear
-
rhodium hydride complex, [(Cp*(PMe3)Rh(H))2(µ-H)]+BAr′4
,
is presented. The reaction to form this binuclear species involves
as an intermediate the hydride/dihydrogen [Cp*(PMe3)Rh(H)-
(H2)]+ BAr′4 complex mentioned above. The binuclear com-
-
plex contains two terminal hydrides and one bridging hydride
1
ligand which are exchanging as observed by H NMR spec-
1
hydride resonance in the H NMR spectrum at temperatures
troscopy. The activation barrier for exchange of a terminal
hydride with the bridging hydride has been determined by NMR
line-broadening techniques.
above 220 K. At lower temperatures, two distinct hydride
environments are observed. The classical structure of this
compound is not surprising due to the presence of a Cp* ligand
in addition to being a third-row metal complex.
Herein, we describe the preparation and characterization of
the rhodium analogue, [Cp*(PMe3)Rh′′H3′′]+ (no structure
implied). From the previous discussion, it is not obvious which
Results and Discussion
Generation of [Cp*(PMe3)Rh(H)(H2)]+BAr′4 (1). The
-
rhodium hydride/dihydrogen species, formulated as [Cp*(PMe3)-
-
Rh(H)(H2)]+BAr′4 (1) on the basis of evidence presented
below, has been prepared via two independent routes. Proto-
nation of the previously reported neutral dihydride complex Cp*-
(PMe3)Rh(H)2 (2) using excess H(OEt2)2BAr′4 (1-2 equiv)
(13) Oldham, W. J., Jr.; Hinkle, A. S.; Heinekey, D. M. J. Am. Chem. Soc.
1997, 119, 11028.
27
(14) Cappellani, E. P.; Maltby, P. A.; Morris, R. H.; Schweitzer, C. T.; Steele,
M. R. Inorg. Chem. 1989, 28, 4437.
resulted in the formation of 1. However, this reaction occurred
(15) Hay, P. J. J. Am. Chem. Soc. 1987, 109, 705.
with concomitant formation of a side product, [(Cp*(PMe3)-
(16) Lyons, D.; Wilkinson, G.; Thornton-Pett, M.; Hursthouse, M. B. J. Chem.
Soc., Dalton Trans. 1984, 695.
-
Rh(H))2(µ-H)]+BAr′4 (3), which was present in significant
(17) Gelabert, R.; Moreno, M.; Lluch, J. M.; Lledo´s, A. Organometallics 1997,
16, 3805.
amounts (Scheme 1; see below for characterization of 3). The
use of 2 equiv of H(OEt2)2BAr′4 led to product ratios of approxi-
mately 9:1 for formation of 1 versus 3, whereas protonation
using 1 equiv resulted in product ratios of approximately 1:1.
(18) Heinekey, D. M.; Millar, J. M.; Koetzle, T. F.; Payne, N. G.; Zilm, K. W.
J. Am. Chem. Soc. 1990, 112, 909.
(19) Arliguie, T.; Chaudret, B. J. Chem. Soc., Chem. Commun. 1986, 13, 985.
(20) Heinekey, D. M.; Oldham, W. J., Jr. J. Am. Chem. Soc. 1994, 116, 3137.
(21) Moreno, B.; Sabo-Etienne, S.; Chaudret, B.; Rodriguez, A.; Jalon, F.;
Trofimenko, S. J. Am. Chem. Soc. 1995, 117, 7441.
(22) Gilbert, T. M.; Bergman, R. G. Organometallics 1983, 2, 1458.
(23) Gilbert, T. M.; Bergman, R. G. J. Am. Chem. Soc. 1985, 107, 3502.
(24) Heinekey, D. M.; Hinkle, A. S.; Close, J. D. J. Am. Chem. Soc. 1996, 118,
5353.
(25) Lin, Z.; Hall, M. B. Organometallics 1992, 11, 3801.
(26) Crabtree, R. H.; Lavin, M. J. Chem. Soc., Chem. Commun. 1985, 1661.
(27) Isobe, K.; Bailey, P. M.; Maitlis, P. M. J. Chem. Soc., Dalton Trans. 1981,
2003.
9
J. AM. CHEM. SOC. VOL. 124, NO. 18, 2002 5101

A R T I C L E S
Taw et al.
Scheme 2
Scheme 3
-
Figure 1. ORTEP diagram of [Cp*(PMe3)Rh(Me)(CH2Cl2)]+ (4, BAr′4
counterion omitted for clarity).
Scheme 4
Table 1: Selected Bond Distances and Bond Angles
bond distance
(Å)
bond angle
(deg)
Rh(1)-Cp*(centroid)
Rh(1)-P(1)
1.8620(4)
2.285(1)
2.106(5)
2.488(1)
Cp*-Rh(1)-P(1)
132.03(4)
Hydrogenolysis of the rhodium(III) dichloromethane adduct,
[Cp*(PMe3)Rh(Me)(CH2Cl2)]+BAr′4- (4), produced 1 in good
yields (Scheme 2). In this reaction, small and variable amounts
of a byproduct formulated as the aquo complex [Cp*(PMe3)-
Cp*-Rh(1)-C(11) 123.3(1)
Rh(1)-C(11)
Cp*-Rh(1)-Cl(1)
P(1)-Rh(1)-C(11)
122.23(3)
84.5(1)
Rh(1)-Cl(1)
Rh(H)(OH2)]+BAr′4 (5) were observed. The presence of any
-
-
residual water caused displacement of the labile µ2-H2 ligand
from 1 to form 5. Complex 1 has not been isolated, as rapid
decomposition occurs upon removal of solvent even at low
temperatures. Thus, for all experiments conducted in this report,
1 has been generated in situ by hydrogenolysis of 4. Partial
deuteration and tritiation of 1 were achieved by exposure of
solutions of 1 to D2 or T2 gas.
(10). Exchange of triflate with the noncoordinating BAr′4
ligand to produce 4 was achieved by adding NaBAr′4 to 10 using
CH2Cl2 as solvent (Scheme 3).
We have since discovered that a more efficient synthesis of
4 involved addition of NaBAr′4 to the previously reported
compound Cp*(PMe3)Rh(Me)(Cl) (11).31 A direct exchange of
chloride ligand with BAr′4- occurred to produce 4 in high yields
(Scheme 4). A clear advantage of this method is that it employs
easily prepared, air-stable complex 11 to produce the air-
sensitive compound 4 directly.
Orange-red crystals of 4, which are thermally and air-
sensitive, were isolated in approximately 80% yield and were
fully characterized. An X-ray crystal structure of 4 is shown in
Figure 1. Table 1 lists some selected bond distances and bond
angles for this compound. Complex 4 forms a structure with a
three-legged piano stool geometry. While two dichloromethane
molecules are present per rhodium center in the crystal lattice,
only one is within bonding distance (Rh-Cl distance ) 2.49
Å) to the metal center. The other dichloromethane molecule
and the tetraarylborate anion are clearly noncoordinating.
Synthesis of [Cp*(PMe3)Rh(Me)(CH2Cl2)]+BAr′4- (4). The
cationic rhodium methyl complex 4 can be synthesized using
procedures analogous to those employed for the previously
reported iridium analogue, [Cp*(PMe3)Ir(Me)(CH2Cl2)]+BAr′4
-
(6).28 Excess MeLi was added to the known dichloride Cp*-
(PMe3)Rh(Cl)2 (7) to produce the dimethyl species Cp*-
27
(PMe3)Rh(Me)229 (8). Similarly, excess AgOTf was added to 7
to produce a bis-triflate complex Cp*(PMe3)Rh(OTf)2 (9).30
Species 8 and 9 were then combined in equal amounts in a
conproportionation reaction, resulting in conversion to the
rhodium methyl triflate compound Cp*(PMe3)Rh(Me)(OTf)
(28) Burger, P.; Bergman, R. G. J. Am. Chem. Soc. 1993, 115, 10462.
(29) Diversi, P.; Iacoponi, S.; Ingrosso, G.; Laschi, F.; Lucherini, A.; Pinzino,
C.; Uccello-Barretta, G.; Zanello, P. Organometallics 1995, 14, 3275.
(30) Stang, P. J.; Huang, Y. H.; Arif, A. M. Organometallics 1992, 11, 231.
(31) Jones, W. D.; Feher, F. J. Organometallics 1983, 2, 562.
9
5102 J. AM. CHEM. SOC. VOL. 124, NO. 18, 2002

Evidence for a Hydride/Dihydrogen Structure
A R T I C L E S
Scheme 5
Figure 2. ORTEP diagram of [(Cp*(PMe3)Rh(H))2(µ-H)]+ (3, Hydride
-
ligands were not located; BAr′4 counterion omitted for clarity).
Synthesis and Characterization of [(Cp*(PMe3)Rh(H))2-
(µ-H)]+BAr′4- (3). When excess H(OEt2)2BAr′4 was added to
2 to generate rhodium hydride 1, a second hydride species (3)
was also produced (Scheme 1). After further studies, we found
that the most efficient synthesis of complex 3 could be achieved
by adding exactly 0.5 equiv of H(OEt2)2BAr′4 to rhodium
dihydride 2. The mechanism presumably involves initial pro-
tonation of 2 to form hydride/dihydrogen species 1. Reversible
loss of H2 occurs to produce a rhodium monohydride species
(not observed), which can then combine with 1 equiv of 2 to
produce the binuclear complex (Scheme 5). Observation by 1H
NMR spectroscopy of the initial formation and then disappear-
ance of 1 as the reaction progresses, and the appearance of free
H2 in the product mixture, lend support to this mechanistic
pathway. An X-ray crystal structure of 3 has been obtained;
the ORTEP diagram is shown in Figure 2. The Rh-Rh distance
of 2.94 Å is suggestive of a weak metal-metal interaction.
issue. The iridium analogue of 3, namely [(Cp*(PMe3)Ir(H))2-
(µ-H)]+X- (X ) PF6 or BF4, 14), has been previously reported
but was prepared using a dissimilar route.23 The X-ray crystal
structure of 14 shows that 3 and 14 are isostructural.
Activation parameters for the exchange of a terminal hydride
with a bridging hydride have been reported for the iridium
system, but determination of these parameters for the rhodium
system is nontrivial.23 Solubility of 3 becomes a problem at the
1
low temperatures required to obtain well-resolved H NMR
spectra of the hydride signals. In addition, the numerous coup-
ling interactions present make modeling of the exchange system
difficult. Measurement of change in line widths with temperature
provides an approximate free energy of activation of 9.1 ( 0.2
kcal/mol for exchange of terminal and bridging hydrides.
Characterization of 1 by NMR Spectroscopy. In addition
1
to resonances for Cp* and PMe3, the H NMR spectrum of 1
Although they were not located in the diffraction study, NMR
analysis demonstrates that binuclear rhodium complex 3 contains
two terminal hydride ligands and one bridging hydride moiety.
These hydrides exchange rapidly at room temperature and
exhibit fluxional behavior as observed by variable-temperature
1H NMR spectroscopy. At 313 K, the hydride resonance appears
as a well-resolved triplet of triplets due to averaged coupling
to two rhodium nuclei and two phosphorus nuclei (δ -16.99
ppm; 1JRh-H ) 14.4 Hz, 2JP-H ) 20.0 Hz). Upon lowering the
temperature the resonance broadens significantly. At 223 K, the
hydride resonance is distinguishable only as a structureless lump
above the baseline. Upon further cooling, decoalescence occurs
to reveal two distinct hydride resonances that integrate in a 2:1
ratio. At temperatures below 203 K the 1H NMR signal for the
two equivalent terminal hydrides was easily distinguishable from
the signal for the bridging hydride. Unlike the hydride region,
the alkyl region varied little with temperature.
Werner has reported a similar rhodium complex, [(Cp(PiPr3)-
Rh)2(µ-H)3]+PF6- (12), which was synthesized by protonation
of Cp(PiPr3)Rh(H)2 (13) using CF3CO2H.32 From NMR data
obtained at room temperature, complex 12 was rationalized as
a binuclear species containing three equivalent bridging hydride
ligands. It is likely that the structure of 12 is similar to 3 and
contains only one bridging hydrogen, but to date no low-
temperature NMR studies have been reported to clarify this
exhibits a doublet of doublets in the hydride region correspond-
1
ing to three protons at -7.95 ppm (223 K; JRh-H ) 25 Hz,
1
2JP-H ) 11 Hz). Recording a H{31P} NMR spectrum of 1
results in simplification of the hydride resonance to a doublet
(1JRh-H ) 25 Hz). Additionally, a 31P NMR spectrum obtained
with decoupling of methyl protons reveals a quartet (2JP-H
)
11 Hz). The hydride signal for all three spectra remains
unchanged at all accessible temperatures (down to 123 K using
a 750 MHz NMR spectrometer), suggesting that all three
“hydride” ligands are equivalent on the NMR time scale.
Exposure of a solution of 1 to one atmosphere of D2 gas
results in partial deuteration of the hydride sites, producing a
mixture of 1, 1-d1, and 1-d2. These isotopomers exhibit distinct
resonances in the 1H NMR spectrum, as shown in Figure 3. At
223 K, an upfield isotope shift of 31 ppb with JH-D ) 10.3 Hz
was observed for 1-d1, while a greater upfield shift of 121 ppb
with JH-D ) 9.9 Hz was observed for 1-d2. These isotope shifts
are temperature-dependent, with the shift difference between 1
and 1-d2 increasing to 137 ppb upon lowering the temperature
to 180 K. Figure 4 shows the temperature dependence of the
hydride chemical shifts for all three isotopomers.
Partial tritiation gives similar results, with upfield isotope
shifts for 1-t1 and 1-t2 of 13 and 230 ppb, respectively, as
1
observed by H{31P} NMR spectroscopy. Also observed are
J
H-T values of 72.8 Hz for 1-t1 and 66.8 Hz for 1-t2 (Figure 5).
The isotope shift for 1-t2 increases as the temperature is lowered
(32) Werner, H.; Wolf, J. Angew. Chem., Int. Ed. Engl. 1982, 21, 296.
in a fashion similar to that in the partial deuteration results
9
J. AM. CHEM. SOC. VOL. 124, NO. 18, 2002 5103

A R T I C L E S
Taw et al.
Applying the method of Halpern, the H-H bond length can
be bracketed between 0.82 Å (fast η2-H2 rotation) and 1.03 Å
(slow rotation).33,34 A second method based on measurement
of JH-D in the η2-HD ligand is described below which yields a
more precise value for the H2 bond length.
Partial deuteration of the hydride ligands in complex 1 leads
to the observation of distinct resonances for 1, 1-d1, and 1-d2
in the 1H NMR spectrum, with the deuterated species exhibiting
substantial H-D coupling. The resonances due to 1-d1 and 1-d2
are shifted modestly upfield from that due to 1. The chemical
shift of the resonance due to 1-d2 is temperature-dependent, with
larger upfield shifts observed at lower temperature (Figure 4).
At 223 K, the H-D coupling in 1-d1 is 10.3 Hz, while the H-D
coupling in 1-d2 is 9.9 Hz. These values for JH-D are indicative
of the presence of a dihydrogen ligand. If the deuterium atoms
were distributed statistically in a hydride/dihydrogen structure,
then 1JH-D would be approximately 30 Hz. The observation of
temperature-dependent chemical shift differences and different
values of JH-D in the different isotopomers are characteristic
of isotopic perturbation of an equilibrium and a nonstatistical
distribution of deuterium.
Figure 3. Partial (hydride region) 750 MHz 1H{31P} NMR spectrum of a
mixture of complexes 1, 1-d1, and 1-d2 in CD2Cl2 at 223 K.
above. Table 2 summarizes the NMR data for all the isoto-
pomers discussed above.
Determination of Hydride/Dihydrogen Structure of 1. As
stated previously, the spectroscopic data are consistent with the
presence of three equivalent “hydride” ligands in complex 1.
This observation may result from a fluxional hydride/dihydrogen
structure as was observed for related cationic tris-pyrazolylborate
complexes of Ir and Rh of the form [Tp(PMe3)M(H)(H2)]+ (M
) Rh, Ir).13,20 Alternatively, the ground-state structure of 1 could
be a dynamic trihydride, as reported for the closely related Ir
complexes [Cp(PMe3)Ir(H)3]+ and [Cp*(PMe3)Ir(H)3]+.18,23,24
These two possibilities can be distinguished with confidence
by employing a combination of techniques: measurement of
the relaxation rate of the hydride protons and analysis of the
effect of partial substitution of the hydride ligands with
deuterium or tritium.
As previously reported for the iridium hydride/dihydrogen
complex, the observed chemical shifts and JH-D coupling data
can be analyzed quantitatively to calculate the limiting chemical
shifts of the dihydrogen ligand (δH ) and the terminal hydride
2
1
ligand (δH), JH-D for the η2-HD ligand, and the energy
difference which arises from deuterium substitution at the
dihydrogen site versus the hydride site.13 In the case of complex
1, this analysis is complicated by the relatively small isotope
effects observed upon deuteration, which are similar in mag-
nitude to the intrinsic isotope effects on the hydride chemical
shifts resulting from partial deuteration.
The chemical shift of the H3 isotopomer (δH ) is simply the
3
statistical average of the dihydrogen site and the hydride site,
The measurement of T1 values for the hydride resonance in
complex 1 was performed at temperatures between 240 and 130
K. The T1 value decreases to a minimum of 23 ms at 150 K
(500 MHz). This short T1(min) value observed for the hydride
ligands in complex 1 is qualitatively consistent with the presence
of a bound dihydrogen ligand in a hydride/dihydrogen structure.
The H-H bond length can be estimated from the mutual
thus δH ) (2δH + δH )/3. However, upon substitution of one
3
2
deuterium atom in the hydride positions an equilibrium between
three different species is obtained (eq 1). A similar equilibrium
is obtained if two deuterium atoms are incorporated in the
hydride positions (eq 2).
1
dipole-dipole relaxation rate Rd-d of the H nuclei in the η2-
H2 ligand (Rd-d ) 1/T1(d-d)). The total relaxation rate RH is
2
the sum of Rd-d and other relaxation rates from interaction with
other magnetic dipoles in the complex, Ro (RH ) Rd-d + Ro).13
2
In the case of 1, the Robs (T1(min)) value is the weighted average
1
of the relaxation rates of an H in all sites. Thus, Robs(1) )
(2RH + RH)/3 where RH and RH are relaxation rates of the
2
2
η2-H2 and terminal hydride ligands, respectively. We assume
that RH is modeled by the relaxation rate Robs(5) of the hydride
ligand in the aquo complex 5 (T1(min) of 5 ) 1.3 s at 178 K,
500 MHz), and further, that this rate also approximates Ro.
Therefore, as previously derived,13
In this analysis we assume that there is no energy difference
between structures a and a′. We also assume that the chemical
shift of the exo hydrogen atom in a is the same as the endo
hydrogen atom in a′. Also let a, a′, and b be the mole fractions
of each species, so a + a′ + b ) 2a + b ) 1. The Boltzmann
factor A1, defined as e-∆E/RT, may be introduced to account for
an energy difference between a (or a′) and b. A1 is equal to b/a
Robs(1) ) (2RH + RH)/3 ) [2(Rd-d + Ro) + RH]/3 )
2
[2Rd-d + 3Robs(5)]/3
Rd-d ) 3[Robs(1) - Robs(5)]/2
(33) Desrosiers, P. J.; Cai, L.; Lin, Z.; Richards, R.; Halpern, J. J. Am. Chem.
Soc. 1991, 113, 4173.
(34) For a discussion and early analysis of fast vs slow η2-H2 rotation, see:
Bautista, M. T.; Earl, K. A.; Maltby, P. A.; Morris, R. H.; Schweitzer, C.
T.; Sella, A. J. Am. Chem. Soc. 1988, 110, 7031.
Using the experimental values reported above, Rd-d ) 64
s-1
.
9
5104 J. AM. CHEM. SOC. VOL. 124, NO. 18, 2002

Evidence for a Hydride/Dihydrogen Structure
A R T I C L E S
Figure 4. Chemical shifts for the hydride resonances in a mixture of 1, 1-d1, and 1-d2 as a function of temperature.
intrinsic isotope effects (see below). In a similar fashion the
observed JH-D coupling constant is given by eq 4, if the two-
2
bond JH-D coupling between the dihydrogen ligand and the
terminal hydride ligand is assumed to be zero.
1JH-D
J(H D)
)
(4)
2
2 + A1
Similarly, expressions can be derived which relate the chemical
shift and observed JH-D coupling of the HD2 isotopomer, written
in terms of a second Boltzmann factor, A2, where A2 ) d/c.
2A2δH + δH
2
δHD
)
(5)
(6)
2
2A2 + 1
1JH-DA2
J(HD )
)
2
2A2 + 1
Qualitatively, the observation that the H-D coupling constant
in complex 1-d1 is slightly greater than in 1-d2 is consistent
with some concentration of deuterium in the dihydrogen ligand,
rather than the hydride site. Since the difference in the JH-D
values is only slightly larger than the uncertainty in the
measurements, we sought to verify this situation by measuring
the H-T coupling constants in partially tritiated samples of 1.
By use of the ratio γT/γD ) 6.9465, the JH-T value observed in
Figure 5. Partial (hydride region) 500 MHz 1H{31P} NMR spectrum of a
mixture of complexes 1, 1-t1, and 1-t2 in CD2Cl2 at 223 K.
Table 2. Proton NMR Data for Various Isotopomers of Complex 1
at 223 K
H
3
H D
2
HD
2
H T
2
HT
2
δ (ppm)a
-7.95
24.8 ( 0.2 23.8 ( 0.2 22.1 ( 0.2 25.5 ( 0.2 23.8 ( 0.2
10.3 ( 0.2 9.9 ( 0.2 72.8 ( 0.1 66.8 ( 0.1
-7.96 (31) -8.09 (121) -7.96 (13) -8.10 (230)
J
Rh-H (Hz)
H-D or
1-t1 of 72.8 Hz can be converted to the more familiar JH-D
)
J
-
J
H-T (Hz)
10.5 Hz, in good accord with the value of JH-D ) 10.3 Hz
observed in 1-d1. The agreement of these coupling constants
and the temperature-independent chemical shift of 1-d1 suggest
that the heavy isotope is distributed statistically in 1-d1 and 1-t1
(i.e., a/b and A1 = 1) and that the true value for 1JH-D is ca. 31
Hz. These observations also suggest that the small and tem-
perature invariant isotope shift of ca. 30 ppb observed in 1-d1
is due to an intrinsic isotope effect.
JP-H (Hz)
9.8 ( 0.2 10 ( 0.2 10 ( 0.2 9.8 ( 0.2 9.6 ( 0.2
a The chemical shifts are given in ppm, with the isotope shifts (ppb).
) b/a′. Thus, a and a′ can be combined, and the mole fractions
can be rewritten in terms of the Boltzmann factor, A1. For this
case a ) a′ ) 1/(2 + A1) and b ) A1/(2 + A1), leading to eq
3, as derived previously.13
In contrast, the isotope shift observed for 1-d2 is large and
temperature-dependent, consistent with the isotopic perturbation
of the equilibrium in eq 2 (i.e., a/b and A2 * 1). The value of
A2 can be estimated as 0.88 at 233 K from eq 6, knowing that
1JH-D is ca. 31 Hz and that the observed value for JH-D in 1-d2
δH + δH + A1δH
2
2
δ(H D)
)
(3)
2
2 + A1
This analysis excludes the contribution to the chemical shift of
9
J. AM. CHEM. SOC. VOL. 124, NO. 18, 2002 5105

A R T I C L E S
Taw et al.
1
is 9.9 Hz. It can be reasonably assumed that the intrinsic isotope
effect contributing to the chemical shift in 1-d2 is approximately
60 ppb, or roughly double the value observed in 1-d1. Values
of A2 at various temperatures can be calculated from the
observed values of JH-D in 1-d2 and the chemical shift for the
dihydrogen and hydride ligands in 1 can be deduced using eq
5. The values of A2 obtained vary from 0.77 at 183 K to 0.91
In the analysis of the H NMR spectra of partially deuterated
samples of complex 1, the assumption of negligible H-D
coupling between bound dihydrogen and adjacent hydride is
reasonable.
Summary
The cationic Rh(III) complex [Cp*(PMe3)Rh(Me)(CH2Cl2)]+-
BAr′4 (4) was prepared via chloride abstraction from Cp*-
-
at 243 K. The calculated chemical shifts are δH ) -7.0 ppm
2
and δH ) -9.6 ppm. In light of the numerous approximations
made above, a realistic uncertainty in these chemical shift values
is (0.5 ppm.
(PMe3)Rh(Me)(Cl) (11) using NaBAr′4 in CH2Cl2. Hydro-
genolysis of 4 yielded [Cp*(PMe3)Rh(H)(H2)]+BAr′4 (1) for
-
which the hydride resonance exhibits a single chemical shift
down to 123 K at 750 MHz. The identification of 1 as a fluxional
hydride/dihydrogen complex was made using T1, JH-D, and JH-T
measurements. The interchange of hydride with dihydrogen sites
must have a barrier (∆Gq) less than ca. 5 kcal/mol.
These chemical shifts are not directly available by low
temperature observation, since the rate of hydrogen atom
exchange in 1 is too fast to allow for the observation of a static
spectrum, even at 123 K. Using the chemical shifts derived
above and the fact that a single resonance was observed at 123
K, a maximum activation energy of ca. 5 kcal/mol can be
calculated for the dynamic process, indicating that the strong
H-H bond in the bound dihydrogen ligand is being broken very
rapidly.35
An interesting aspect of these observations is that A1 * A2.
This can be rationalized by noting that the relative energy of
the isotopomers depicted in eq 1 and 2 is proportional to the
reduced mass of the hydrogen (deuterium) atoms bound to the
rhodium center. Substitution of deuterium for hydrogen in the
Protonation of Cp*(PMe3)Rh(H)2 (2) with 1 equiv of H(OEt2)2-
BAr′4 yielded a mixture of 1 and binuclear complex [(Cp*-
(PMe3)Rh(H))2(µ-H)]+BAr′4 (3). Use of 0.5 equiv of acid
-
resulted in clean formation of 3, which was formulated as a
fluxional bridging hydride complex based upon 1H NMR
spectroscopy and an X-ray structure determination. A free
energy of activation of 9.1 kcal/mol was measured for exchange
of terminal and bridging hydrides.
Experimental Section
General Considerations. Unless otherwise noted, all reactions and
manipulations were performed using standard high-vacuum, Schlenk,
or drybox techniques. Argon and nitrogen were purified by passage
through columns of BASF R3-11 catalyst (Chemalog) and 4 Å
molecular sieves. 1H and 13C NMR chemical shifts were referenced to
x
hydride site will change the oscillator energy by 1/2, inde-
pendent of any isotope substitution in the dihydrogen ligand.
In contrast, the reduced mass varies with isotope substitution
in a nonlinear fashion in the bound dihydrogen ligand. This
leads to the observation that replacement of a second hydrogen
atom with deuterium in the bound dihydrogen ligand has a
greater effect than replacing the first one.
1
residual H and 13C NMR signals of the deuterated solvents, respec-
tively. 31P NMR chemical shifts were referenced to an 85% H3PO4
sample used as an external standard. Tritium NMR spectra were
- 43
Using the well-known inverse correlation between the H-H
bond length and the value of 1JH-D, an H-H bond distance of
0.90 Å in the dihydrogen ligand of complex 1 can be
derived.36-39 This suggests that neither of the two distances
calculated above from the relaxation data are correct. The rate
of rotation of the dihydrogen ligand in 1 may be intermediate
between the fast and slow regime.40
The value of 1JH-D derived here (and thus the H-H distance)
is based on the assumption that the two-bond coupling between
the bound dihydrogen and the adjacent hydride is zero. Coupling
between bound dihydrogen and a hydride ligand has been
observed in cases where the dihydrogen and hydride ligands
are trans to each other.41 In one report42 of significant coupling
between bound dihydrogen and a cis hydride ligand, this
coupling is thought to arise from quantum mechanical exchange
coupling, which would be quenched upon partial deuteration.
obtained at 533 or 800 MHz and externally referenced to CHDTCO2 .
The procedures for the safe storage and manipulation of tritium gas
have been described previously.24 Elemental analyses were performed
by the University of California, Berkeley Microanalytical Facility, or
by Atlantic Microlab Inc. of Norcross, GA. Mass spectrometric analyses
were performed by the University of California, Berkeley Mass
Spectrometry Facility.
Materials. All solvents were deoxygenated and dried via passage
over a column of activated alumina.44,45 Deuterated solvents (Cambridge
Isotope Laboratories) were purified by vacuum transfer from CaH2 and
stored over 4 Å molecular sieves. Compressed H2 gas was purchased
from Aldrich, D2 from Matheson, and T2 from American Radiolabeled
Chemical, and all were used without further purification. NaBAr′4 was
purchased from Boulder Scientific and used without further purification.
H(OEt2)2BAr′4 was synthesized as described in the literature.46 A 1.5
M solution of MeLi in Et2O was purchased from Acros and used without
further purification.
Complex 2 was initially reported by Maitlis but was prepared
according to procedures published by Bergman.27,47 Complexes 9 and
11 have been previously reported and were synthesized according to
literature procedures.30,31 Complex 8 is also known but was prepared
using a method different from that reported in the literature (see
below).29 All spectroscopic data for the above compounds are in accord
with published values.
(35) For examples in the determination of H/H2 exchange barriers and H-H
bond distances using inelastic neutron scattering spectroscopy, see: (a)
Eckert, J.; Albinati, A.; Bucher, U. E.; Venanzi, L. M. Inorg. Chem. 1996,
35, 1292. (b) Li, S.; Hall, M. B.; Eckert, J.; Jensen, C. M.; Albinati, A. J.
Am. Chem. Soc. 2000, 122, 2903.
(36) Maltby, P. A.; Schlaf, M.; Steinbeck, M.; Lough, A. J.; Morris, R. H.;
Klooster, W. T.; Koetzle, T. F.; Srivastava, R. C. J. Am. Chem. Soc. 1996,
118, 5396.
(37) Heinekey, D. M.; Luther, T. A Inorg. Chem. 1996, 35, 4396.
(38) Luther, T. A.; Heinekey, D. M. Inorg. Chem. 1998, 37, 127.
(39) Grundemann, S.; Limbach, H. H.; Buntkowsky, G.; Sabo-Etienne, S.;
Chaudret, B. J. Phys. Chem. A 1999, 103, 4752.
(40) Morris, R. H.; Wittebort, R. J. Magn. Reson. Chem. 1997, 35, 243.
(41) Bianchini, C.; Moneti, S.; Peruzzini, M.; Vizza, F. Inorg. Chem. 1997, 36,
5818.
(42) Guari, Y.; Sabo-Etienne, S.; Chaudret, B. J. Am. Chem. Soc. 1998, 120,
4228.
(43) Evans, E. A.; Warrell, D. C.; Elvidge, J. A.; Jones, J. R. Handbook of
Tritium NMR Spectroscopy and Applications; Wiley: New York, 1985.
(44) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.; Timmers,
F. J. Organometallics 1996, 15, 1518.
(45) Alaimo, P. J.; Peters, D. W.; Arnold, J.; Bergman, R. G. J. Chem. Educ.
2001, 78(1), 64.
(46) Brookhart, M.; Grant, B.; Volpe, A. F., Jr. Organometallics 1992, 11, 3920.
(47) Periana, R. A.; Bergman, R. G. J. Am. Chem. Soc. 1986, 108, 7332.
9
5106 J. AM. CHEM. SOC. VOL. 124, NO. 18, 2002

Evidence for a Hydride/Dihydrogen Structure
A R T I C L E S
Spectral Data for BAr′4-. The 1H and 13C NMR resonances of the
BAr′4 (Ar′ ) 3,5-C6H3(CF3)2) counteranion in CD2Cl2 were essentially
invariant for all cationic complexes discussed here, and spectro-
scopic data are not repeated for each compound. 1H NMR (400 MHz,
CD2Cl2) δ 7.73 (s, 8 H, Ho), 7.57 (s, 4 H, Hp). 13C{1H} NMR (101
MHz, CD2Cl2) δ 161.9 (q, 1JC-B ) 49.8, Cipso), 135.0 (s, Co), 129.0 (q,
NaBAr′4 and sealed with a rubber stopper. The flask was placed in an
ice bath, and 5 mL of CH2Cl2 was introduced via syringe. The contents
of the flask were stirred for 10-15 min and then filtered through Celite
supported by a glass frit. The filtrate was concentrated in vacuo and
the flask placed in a freezer (243 K) for 2 d. Orange-red crystals of 4
(223 mg, 0.175 mmol) were isolated in 84% yield. Alternately, this
procedure can be repeated using 75 mg (0.206 mmol) of 11 and 191
mg (0.216 mmol) of NaBAr′4 to produce crystals of 4 (224 mg, 0.175
1
2JC-F ) 31.4, Cm), 124.7 (q, JC-F ) 272.6, CF3), 117.7 (s, Cp).
[Cp*(PMe3)Rh(H)(H2)]+BAr′4 (1). Solutions of complex 1 were
-
1
mmol) in 85% yield. H NMR (400 MHz, CD2Cl2, 273 K) δ 5.33 (s,
generated using the following three procedures. (a) An NMR tube was
charged with 5 mg (0.016 mmol) of 2 and sealed with a septum. The
tube was placed in a 233 K bath, and a solution of H(OEt2)2BAr′4 (32.0
mg, 0.032 mmol) in CD2Cl2 (0.5 mL) was added slowly via syringe.
The contents of the tube were allowed to warm briefly (2-3 min) to
room temperature and then cooled again in the 233 K bath. A 1H NMR
spectrum of the product revealed a mixture of approximately 90% 1
and 10% 3. (b) A J-Young NMR tube was charged with a solution of
4 (5-10 mg of 4 in 0.5 mL CD2Cl2), sealed, and attached to a vacuum
line. The sample was subjected to 4 freeze-pump-thaw cycles and
placed in a 233 K bath. The sample was then exposed to ∼1 atm H2,
the NMR tube was sealed, and complex 1 was formed by vigorous
shaking of the NMR tube. (c) In a third approach, 0.5 mL of CD2Cl2
or CDFCl2 was vacuum-transferred into a 5-mm NMR tube con-
taining 1-3 mg of 4. Upon thawing to 233 K the solution was ex-
posed to 1 atm H2 for several min to afford 1. The NMR tube was
then sealed under ∼500 Torr of H2 gas. As mentioned previously, 1
could not be isolated due to rapid decomposition upon removal of
solvent. However, solutions of 1 can be stored for long periods of time
4
2
CH2Cl2), 1.60 (d, JP-H ) 2.6 Hz, 15 H, Cp*), 1.44 (d, JP-H ) 10.1
Hz, 9 H, PMe3), 0.74 (d, JP-H ) 6.7 Hz, 3 H, Me). 31P{1H}
3
NMR (162 MHz, CD2Cl2, 273 K) δ 3.2 (d, 1JRh-P ) 163.6 Hz, PMe3).
13C{1H} NMR (101 MHz, CD2Cl2, 273 K) δ 101.5 (s, Cp*-Ar), 54.2
1
(s, CH2Cl2), 14.8 (d, JC-P ) 32.1 Hz, PMe3), 9.4 (s, Cp*-Me), 2.0
(dd, 1JC-Rh ) 24.7 Hz, 2JC-P ) 14.4 Hz, Me). Anal. Calcd for C47H41-
BF24Cl2PRh: C, 44.19; H, 3.24. Found: C, 44.21; H, 3.27. FAB
LRMS: m/z 329 ([Cp*(PMe3)Rh(Me)]+).
[Cp*(PMe3)Rh(H)(OH2)]+BAr′4- (5). A trace amount of complex
5 was formed in the reaction to generate 1. Residual H2O present in
the solvent and/or added H2 gas served to displace the labile η2-H2
1
ligand from 1, thereby forming 5. H NMR (400 MHz, CD2Cl2, 223
K) δ 1.72 (s, 15 H, Cp*), 1.45 (d, 2JP-H ) 13.4 Hz, 9 H, PMe3), -9.63
1
2
(dd, JRh-H ) 20.4 Hz, JP-H ) 45.1 Hz, 1 H, hydride).
Cp*(PMe3)Rh(Me)2 (8). In air, a Schlenk flask was charged with
200 mg (0.519 mmol) of 7 and sealed with a rubber stopper. The flask
was evacuated and back-filled with Ar 4×, and 10 mL of dry, degassed
Et2O was added via syringe. While stirring, 1.56 mmol (1.0 mL of a
1.5 M solution of MeLi in Et2O) of MeLi was added slowly via syringe.
The contents of the flask were stirred for 12 h after which ∼0.1 mL of
H2O was added to quench the MeLi. All liquids were cannula-filtered
into another Schlenk flask, and the volatile materials were removed in
vacuo to produce a brown solid. This solid was dissolved in pentane
and filtered through alumina. The light orange filtrate was collected
and the solvent removed in vacuo to produce 8 (114 mg, 0.332 mmol)
in 64% yield. Complex 8 has been previously characterized, but was
prepared using a different route.29
1
in a sealed vessel at 195 K or below. H NMR (400 MHz, CD2Cl2,
4
2
223 K) δ 1.92 (d, JP-H ) 2.7 Hz, 15 H, Cp*), 1.42 (d, JP-H ) 11.2
Hz, 9 H, PMe3), -7.95 (dd, 1JRh-H ) 24.6 Hz, 2JP-H ) 10.7 Hz, 3 H,
averaged H and η2-H2). 31P{1H} NMR (162 MHz, CD2Cl2, 223 K) δ
5.6 (d, JRh-P ) 122.8 Hz, PMe3). 13C{1H} NMR (101 MHz, CD2Cl2,
1
223 K) δ 104.1 (s, Cp*-Ar), 19.2 (d, 1JC-P ) 36.6 Hz, PMe3), 10.4 (s,
Cp*-Me).
Deuterium and Tritium Substitution in 1 (1-d1, 1-d2, 1-t1, and
1-t2). Deuterium was incorporated by freezing an NMR tube containing
a solution of 1, removing the H2 atmosphere, and replacing it with D2
gas. The NMR tube was then sealed under ∼500 Torr D2 gas. Tritiated
samples were obtained by exposing solutions of 1 to 150 Torr T2 at
-40 °C for 20-60 min, yielding 20-60% tritiation or 15-40 mCi of
activity. The apparatus for the safe manipulation of tritium has been
previously described.24 The samples were then frozen, the T2 was
removed, and the tube flame was sealed under vacuum.
Cp*(PMe3)Rh(Me)(OTf) (10). A Schlenk flask was charged with
100 mg (0.290 mmol) of 8 and 178 mg (0.290 mmol) of 9 and sealed
with a rubber stopper. The flask was placed in an ice bath, and 15 mL
of Et2O was introduced via syringe. The contents of the flask were
allowed to warm to rt while stirring. After 12 h of stirring the brown-
black solution was filtered through silanized silica gel and all solvent
removed in vacuo from the filtrate to produce an orange-brown solid.
This solid was recrystallized from Et2O/pentane to produce 10 (181
[(Cp*(PMe3)Rh(H))2(µ-H)]+BAr′4 (3). A Schlenk flask was
-
1
mg, 0.378 mmol) in 65% yield. H NMR (400 MHz, CD2Cl2, rt) δ
charged with 50.0 mg (0.158 mmol) of 2 and placed in a 233 K bath.
A cold (233 K) solution of H(OEt2)2BAr′4 (80.0 mg, 0.079 mmol) in
CH2Cl2 (4 mL) was cannula-transferred to the flask and the mixture
allowed to warm to room temperature. The orange solution was stirred
for 15-20 min and pentane (∼1 mL) was added via syringe. The flask
was placed in a freezer (243 K), and orange crystals of 3 (62.6 mg,
0.042 mmol, 53% yield) were isolated after a few days. 1H NMR (400
MHz, CD2Cl2, 193 K) δ 1.81 (s, 15 H, Cp*), 1.27 (d, 2JP-H ) 10.0 Hz,
9 H, PMe3), -14.91 (br s, 2 H, terminal H’s), -21.33 (br s, 1 H,
4
2
1.60 (d, JP-H ) 2.5 Hz, 15 H, Cp*), 1.45 (d, JP-H ) 10.3 Hz, 9 H,
PMe3), 0.83 (d, JP-H ) 6.9 Hz, 3 H, Me). 31P{1H} NMR (162 MHz,
3
CD2Cl2, RT) δ 6.7 (d, 1JRh-P ) 165.7 Hz, PMe3). 19F NMR (376 MHz,
CD2Cl2, RT) δ -79.1 (s). 13C{1H} NMR (101 MHz, CD2Cl2, RT) δ
1
1
119.7 (q, JC-F ) 320.2 Hz, CF3), 98.5 (s, Cp*-Ar), 14.6 (d, JC-P
30.1 Hz, PMe3), 9.5 (s, Cp*-Me), 4.2 (dd, JC-Rh ) 22.3 Hz, JC-P
)
)
1
2
16.5 Hz, Me).
Measurement of Activation Parameter Using NMR Line Broad-
ening Techniques. The activation barrier for exchange of a terminal
hydride with a bridging hydride in complex 3 was measured using
standard NMR line broadening experiments conducted at temperatures
between 163 and 223 K. Line widths (ν) were measured at half-height
in units of Hz and were corrected for line widths (νo) at the
low-temperature limit. The exchange rates were determined from the
equation k ) π(ν - νo). Free energies of activation were calculated
from the Eyring equation ∆G‡ ) -RT[ln(kh/kBT)].
Crystallographic Studies. Crystallographic studies were performed
by Dr. Peter S. White (complex 3) at the University of North Carolina,
Chapel Hill, Single-Crystal X-ray Facility or by Dr. Frederick J.
Hollander (complex 4) at the University of California, Berkeley, X-ray
Crystallography Facility.
1
4
bridging H). H NMR (400 MHz, CD2Cl2, 253 K) δ 1.90 (d, JP-H
)
2.8 Hz, 15 H, Cp*), 1.37 (d, 2JP-H ) 10.1 Hz, 9 H, PMe3), -16.99 (br
s; averaged signal for exchanging terminal and bridging H’s). 1H NMR
4
(400 MHz, CD2Cl2, 313 K) δ 1.95 (d, JP-H ) 2.8 Hz, 15 H, Cp*),
2
1
1.43 (d, JP-H ) 10.1 Hz, 9 H, PMe3), -16.99 (tt, JRh-H ) 14.4 Hz,
2JP-H ) 20.0 Hz, 3 H, signal for rapidly exchanging terminal and
bridging H’s). 31P{1H} NMR (162 MHz, CD2Cl2, 253 K) δ 4.5 (d,
1JRh-P ) 137.4 Hz, PMe3). 13C{1H} NMR (101 MHz, CD2Cl2, 253
1
K) δ 100.2 (s, Cp*-Ar), 21.1 (d, JC-P ) 34.1 Hz, PMe3), 11.6 (s,
Cp*-Me). Anal. Calcd for C58H63BF24P2Rh2: C, 46.61; H, 4.25.
Found: C, 46.49; H, 4.23.
[Cp*(PMe3)Rh(Me)(CH2Cl2)]+BAr′4 (4). A Schlenk flask was
-
charged with 100 mg (0.209 mmol) of 10 and 94 mg (0.220 mmol) of
9
J. AM. CHEM. SOC. VOL. 124, NO. 18, 2002 5107

A R T I C L E S
Taw et al.
For complex 3, data were collected at 173 K on a Bruker SMART
diffractometer, using the omega scan mode. The structure was solved
by direct methods. Refinement was by full-matrix least squares with
weights based on counter statistics. All non-hydrogen atoms were
refined anisotropically, while H atoms were refined using a riding
model. Crystal data and collection parameters are given in the
Supporting Information. All computations were performed using the
NRCVAX suite of programs.48
For complex 4, all measurements were made at 159 K on a Siemens
SMART diffractometer with graphite monochromated Mo KR radiation.
The structure was solved by direct methods and expanded using Fourier
techniques. Hydrogen atoms were included in calculated positions but
not refined. Data were integrated using the program SAINT and
corrected for Lorentz and polarization effects.49 No decay correction
was applied. The data were analyzed using the program XPREP to
determine space group and cell contents.50 Crystal data and collection
parameters are given in the Supporting Information.
Acknowledgment. Research performed in the M. S. B.
laboratory at UNC was supported by the NSF (CHE-0107810).
Research performed in the D. M. H. laboratory at UW was
supported by the NSF (CHE-9807358). Acquisition of the 750
MHz NMR spectrometer was supported by the Murdoch
Charitable Trust and the NSF. We acknowledge NSF support
(CHE-9710008) of a 500 MHz spectrometer upgrade. The work
done in the R. G. B. laboratory at UCB was supported by the
Director, Office of Energy Research, Office of Basic Energy
Sciences, Chemical Sciences Division, U.S. Department of
Energy, under Contract No. DE-AC03-76SF00098.
Supporting Information Available: Crystallographic data for
complexes 3 and 4 (CIF). This material is available free of
charge via the Internet at http://pubs.acs.org.
JA0165990
(48) Gabe, E. J.; Le Page, Y.; Charland, J.-P.; Lee, F. L.; White, P. S. J. Appl.
Crystallogr. 1989, 22, 384.
(49) SAINT: SAX Area-Detector Integration Program; Version 4.024; Siemens
Industrial Automation, Inc., Madison, WI, 1995.
(50) XPREP: Version 5.03; Part of the SHELXTL Crystal Structure Determi-
nation Package; Siemens Industrial Automation, Inc., Madison, WI, 1995.
9
5108 J. AM. CHEM. SOC. VOL. 124, NO. 18, 2002