Synthesis and characterization of hydrotris(pyrazolyl)borate dihydrogen/hydride complexes of rhodium and iridium

Article abstract of DOI:10.1021/ja971627t

Protonation of TpM(Pr₃)H₂ (M = Rh, Ir) complexes with HBF⁴·Et₂O or [H(Et₂O)₂][B(Ar)₄](Ar = 3,5=(CF₃)₂C₆H₃) affords cationic complexes which exhibit a single hydride resonance at all accessible temperatures in the ¹H NMR spectrum. Formulation as fluxional dihydrogen/hydride complexes is indicated by short T₁(min) values of ca. 22 ms (Ir) and 7 ms (Rh). The relaxation times are consistent with H-H bond lengths of 0.88-1.11 A? in the iridium complexes and 0.73-0.92 A? in the rhodium complexes depending on the relative rate of the dihydrogen rotational motion. In the case of the iridium complexes, partial substitution of the hydrode positions with deuterium or tritium results in large temperature-dependent isotope shifts and resolvable J(H-D) or J(H-T) coupling constants. Analysis of the chemical shift and coupling constant data as a function of temperature is consistent with a preference for the heavy hydrogen isotope to occupy the hydride rather than the dihydrogen site. This analysis also provides the limiting chemical shifts of the dihydrogen and hydride ligands as well as the ¹J(H-D) coupling constant (ca. 25 Hz) in the bound dihydrogen ligand.

Full text of DOI:10.1021/ja971627t

11028
J. Am. Chem. Soc. 1997, 119, 11028-11036
Synthesis and Characterization of Hydrotris(pyrazolyl)borate
Dihydrogen/Hydride Complexes of Rhodium and Iridium
Warren J. Oldham, Jr., Amber S. Hinkle, and D. M. Heinekey*
Contribution from the Department of Chemistry, UniVersity of Washington, Box 351700,
Seattle, Washington 98195-1700
ReceiVed May 19, 1997X
Abstract: Protonation of TpM(PR3)H2 (M ) Rh, Ir) complexes with HBF4‚Et2O or [H(Et2O)2][B(Ar)4] (Ar ) 3,5-
(CF3)2C6H3) affords cationic complexes which exhibit a single hydride resonance at all accessible temperatures in
the 1H NMR spectrum. Formulation as fluxional dihydrogen/hydride complexes is indicated by short T1(min) values
of ca. 22 ms (Ir) and 7 ms (Rh). The relaxation times are consistent with H-H bond lengths of 0.88-1.11 Å in the
iridium complexes and 0.73-0.92 Å in the rhodium complexes depending on the relative rate of the dihydrogen
rotational motion. In the case of the iridium complexes, partial substitution of the hydride positions with deuterium
or tritium results in large temperature-dependent isotope shifts and resolvable JH-D or JH-T coupling constants.
Analysis of the chemical shift and coupling constant data as a function of temperature is consistent with a preference
for the heavy hydrogen isotope to occupy the hydride rather than the dihydrogen site. This analysis also provides
the limiting chemical shifts of the dihydrogen and hydride ligands as well as the 1JH-D coupling constant (ca. 25 Hz)
in the bound dihydrogen ligand.
Since the first report of a stable molecular hydrogen complex
In this paper we investigate the effect of substituting the Cp
ligand of [CpIr(L)H3]BF4 complexes with the hydrotris(1-
pyrazolyl)borate (Tp)9 ligand. The new Tp complexes, [TpIr-
(L)(H2)H]BF4 (L ) PMe3, PPh3), are formulated as dihydrogen/
hydride complexes, although only a single hydride resonance
by Kubas,1 the possibility that a fluxional polyhydride complex
might also contain a dihydrogen ligand has been actively
investigated.2 In general, transition metal polyhydride com-
plexes are characterized by high coordination numbers (CN
7-9) and high formal oxidation states.3 Because several
structures of nearly equivalent energy are available to seven-,
eight-, and nine-coordinate complexes, rapid permutation of the
1
is observed in the H NMR spectrum at all accessible temper-
atures. We have also obtained spectroscopic evidence for the
thermally labile rhodium complex [TpRh(PPh3)(H2)H]+. In-
corporation of deuterium or tritium in the hydride positions of
the iridium complexes results in large temperature-dependent
isotope shifts in the 1H and 3H NMR spectra, which are
attributed to isotopic perturbation of equilibria. A similar
isotope effect is not observed for the rhodium complex. We
show that these isotope effects arise from nonstatistical occupa-
tion of different hydride sites and can be used to obtain structural
information. Some aspects of the iridium chemistry have been
previously communicated.10
1
hydride positions is often observed by H NMR spectroscopy.
As a result, structural characterization in solution depends upon
indirect methods, in which the observed NMR parameters are
a population-weighted average of all the hydride environments.
For example, Crabtree and co-workers have employed T1
measurements to detect short H-H contacts in a range of
polyhydride complexes, including [Ir(PCy3)2H6]+ and Fe-
(PEtPh2)3H4.4 A quantitative treatment of relaxation in poly-
hydride complexes has been developed which allows useful
structural information to be obtained from T1(min) data.4-6
Results
We have previously reported the structure and properties of
cationic iridium complexes of the form [CpIr(L)H3]BF4 (L )
various PR3), which have been shown to adopt iridium(V)
trihydride structures in the solid state.7,8 These complexes
undergo a rapid hydride rearrangement which leads to a single
hydride resonance in the H NMR spectrum above ca. 220 K.
However, at very low temperatures, spectra consistent with the
solid state structure are obtained.
Synthesis. We have prepared complexes of the form TpIr-
(L)H2 (L ) PMe3, PPh3) by two methods. The first approach
was developed following a report by Pedersen and Tilset11 that
[Cp*Ir(PPh3)H3]BF4 reacts in acetonitrile to form [(MeCN)3Ir-
(PPh3)H2]BF4 and free Cp*H. We have found that the related
PMe3 complex [Cp*Ir(PMe3)H3]SO3CF3 also reacts in aceto-
nitrile to form [(MeCN)3Ir(PMe3)H2]SO3CF3 (eq 1). The three
1
X Abstract published in AdVance ACS Abstracts, October 15, 1997.
(1) Kubas, G. J.; Ryan, R. R.; Swanson, B. I.; Vergamini, P. J.;
Wasserman, H. J. J. Am. Chem. Soc. 1984, 106, 451-452.
(2) Heinekey, D. M.; Oldham, W. J., Jr. Chem. ReV. 1993, 93, 913-
926.
(3) Hlatky, G. G.; Crabtree, R. H. Coord. Chem. ReV. 1985, 65, 1-48.
(4) Hamilton, D. G.; Crabtree, R. H. J. Am. Chem. Soc. 1988, 110, 4126-
4133.
(5) Desrosiers, P. J.; Cai, L.; Lin, Z.; Richards, R.; Halpern, J. J. Am.
Chem. Soc. 1991, 113, 4173-4184.
(9) Substitution of the Tp ligand is represented by superscripts as
suggested by Trofimenko. For example, methyl substituents in the 3,
(6) Bautista, M. T.; Earl, K. A.; Maltby, P. A.; Morris, R. H.; Schweitzer,
C. T.; Sella, A. J. Am. Chem. Soc. 1988, 110, 7031-7036.
(7) Heinekey, D. M.; Millar, J. M.; Koetzle, T. F.; Payne, N. G.; Zilm,
K. W. J. Am. Chem. Soc. 1990, 112, 909-919.
5-positions are indicated as TpMe . For a comprehensive review of this class
2
of complexes see: Trofimenko, S. Chem. ReV. 1993, 93, 943-980.
(10) Heinekey, D. M.; Oldham, W. J., Jr. J. Am. Chem. Soc. 1994, 116,
3137-3138.
(8) Heinekey, D. M.; Hinkle, A. S.; Close, J. D. J. Am. Chem. Soc. 1996,
118, 5353-5361.
(11) Pedersen, A.; Tilset, M. Organometallics 1993, 12, 3064-3068.
S0002-7863(97)01627-2 CCC: $14.00 © 1997 American Chemical Society

Dihydrogen/Hydride Complexes of Rhodium and Iridium
J. Am. Chem. Soc., Vol. 119, No. 45, 1997 11029
Table 1. Selected Spectroscopic Data for New Hydride
Complexes
1H NMR (300 MHz, CD2Cl2)
IR (Nujol)
νB-H νIr-H
δM-H
JP-H
(Hz)
T1(min)
(ms)
complex
(ppm)
(cm-1
)
(cm-1
)
1
2
3
4
5
6
7
8
-21.30
-20.47
-22.11
-16.42
-10.40
-9.77
25.0
22.1
26.7
28.4c
11
f
f
542a
359b
2482
2481
2511
2475
2499
2502
2142
2179, 2139
2138
2092, 2069
2199
2197
304d
21e
21e
-11.0
-7.95
f
7g
c
f
a 177 K. b 201 K. JRh-H ) 18.9 Hz. d 195 K. e 182 K. Not resolved.
g 180 K.
MeCN ligands of the new complexes are easily displaced by
the pyrazolyl arms of the Tp ligand under mild conditions to
form TpIr(L)H2 (L ) PMe3 (1), PPh3(2)). These reactions are
conveniently carried out in one pot by generating the tris-
(acetonitrile) intermediate in situ and then adding 1 equiv of
NaTp in a second step. The procedure has also been used to
1
Figure 1. High-field region of the H{31P} NMR spectrum (CD2Cl2,
Me2
prepare TpMe Ir(PMe3)H2 (3) upon addition of KTp
to a
2
solution of [(MeCN)3Ir(PMe3)H2]+ in acetonitrile (eq 2).
500 MHz, 240 K) of a partially deuterated sample of [TpIr(PMe3)-
(H2)]BF4 (5).
decomposes at room temperature over the course of several
hours; however, its lifetime in solution can be extended
considerably if the protonation reaction is carried out with
[H(Et2O)2][B(Ar)4] (Ar ) 3,5-(CF3)2C6H3). Similarly, 4 de-
composes immediately at 210 K to a complex mixture of
products upon addition of HBF4‚Et2O. Reactions carried out
with [H(Et2O)2][B(Ar)4] yield [TpRh(PPh3)(H2)H][B(Ar)4] (8),
which is stable in solution to 250 K (eq 5). Complexes 5 and
6 were found to be thermally stable and can be isolated as
colorless salts upon addition of HBF4‚Et2O to Et2O or Et2O/
CH2Cl2 solutions of the parent dihydride complexes.
Complexes 1-3 are stable to air and water and can be purified
by chromatography on alumina, eluting with toluene. Isolated
yields are typically 55-70%. The dihydride complexes are
colorless, thermally stable, and crystalline and are readily
identified by their appropriate analytical and spectroscopic data,
which are detailed in the Experimental Section. Selected
spectroscopic data are reported in Table 1.
We have developed an alternative approach to these dihydride
complexes using the reaction of phosphines with TpM(C2H4)2
(M ) Rh, Ir). The resulting TpM(PR3)(C2H4) complexes react
with hydrogen to afford the dihydride complexes in very good
yield. For example, TpRh(PPh3)H2 (4) has been prepared using
this method (eq 3). Characterization of the intermediate
Selected data for the new cationic hydride complexes are
presented in Table 1. Complexes 5-8 exhibit a 2:1 pattern of
Tp resonances in the 1H and 13C NMR spectra, which is
characteristic of a structure with Cs symmetry. In all cases,
this pattern is unchanged upon cooling to 180 K. An appropriate
resonance for the respective hydride ligands is observed at high
field near -10.4 ppm in the 1H NMR spectra of 5-7. A similar
resonance at -7.95 ppm is observed for the rhodium complex
8. Generally these resonances are broad with no distinguishing
features. However, in the case of complex 5, the hydride
resonance is a doublet with P-H coupling of 11 Hz. A total
of three hydride ligands (no structure implied) is confirmed by
the observation of a quartet for the PMe3 resonance in the 31P-
1
{Me H} NMR spectrum at -42.7 ppm (JP-H ) 10 Hz). For
complex 5, a sharp IR absorption at 2199 cm-1 is attributed to
an Ir-H stretch and a band at 2499 cm-1 is assigned to the
B-H stretch. Similar observations were made for 6. The
minimum of the longitudinal relaxation time (T1) is found to
be 21-22 ms for 5 and 6 but is 7 ms for 8.
phosphine-ethylene complexes and the mechanism of their
reaction with hydrogen has been separately reported.12
Addition of 1 equiv of HBF4‚Et2O to CD2Cl2 solutions of
1-3 at ambient temperature or below affords new cationic
complexes without elimination of H2 (eq 4). The TpMe complex
NMR Observations for Partially Deuterated Complexes.
Deuteration of the hydride positions is observed over the course
of several hours at room temperature upon exposure of solutions
of 5-7 to an atmosphere of D2. Well-resolved signals for H3,
H2D, and HD2 isotopomers are observed by 1H NMR spectros-
copy. Large temperature-dependent downfield isotope shifts are
noted for the hydride resonances of the deuterated complexes.
A representative spectrum of the hydride region for complex 5
acquired with 31P decoupling at 240 K is shown in Figure 1.
2
(12) Oldham, W. J., Jr.; Heinekey, D. M. Organometallics 1997, 16,
467-474.

11030 J. Am. Chem. Soc., Vol. 119, No. 45, 1997
Oldham et al.
Table 2. Chemical Shifts of Hydride Resonances and JH-D
Coupling Constants for Partially Deuterated Isotopomers of
Complexes 5-8
complex
isotopomer
hydride shift (ppm)
JH-D (Hz)
5a
H3
H2D
HD2
H3
H2D
HD2
H3
H2D
-10.37
-10.18
-10.08
-9.80
7.48
8.78
6a
7a
-9.61
b
b
-9.52
-11.06
-10.84
-10.75
-7.96
7.33
8.90
d
HD2
H3
H3, H2D, HD2
8c
-7.94
e
a
c
1H{31H} NMR (500 MHz, 240 K). b Not resolved. 1H NMR (300
Figure 2. Plot of chemical shift of the hydride resonance in 5, 5-d1,
and 5-d2 as a function of observation temperature.
MHz, 213 K). d Broad; line width ) 75 Hz. e Broad; line width ) 63
Hz.
Figure 3. Plot of H-D coupling in 5-d1 and 5-d2 as a function of
observation temperature.
The resonances of the Tp and phosphine ligands are unaffected
by deuteration of the hydride positions and show no unusual
temperature dependence. The isotope shifts for 5, ∆1 ) δ(H2D)
- δ(H3) and ∆2 ) δ(HD2) - δ(H2D), vary from +228 and
+122 ppb at 215 K to +149 and +72 ppb at 280 K, respectively.
Over a similar temperature range, the observed H-D coupling
of the hydride resonance varies from ca. 6.9 to 7.6 Hz for 5-d1
and from ca. 8.8 to 8.7 Hz for 5-d2. These observations are
represented graphically in Figures 2 and 3. Very similar
temperature-dependent isotope shifts and H-D coupling con-
stants are observed for 6 and 7. The hydride resonances of the
H2D and HD2 isotopomers of 6 are significantly broadened
compared to those of 5 and 7, so that the JH-D coupling constant
could not be accurately measured in this case. However, on
the basis of the line width of the hydride peaks, the JH-D
coupling constants of 6 are similar to those measured for 5 and
7. Data for these complexes acquired at 240 K are summarized
in Table 2.
Figure 4. Top: 1H NMR spectrum (CD2Cl2, 500 MHz) of a partially
tritiated sample of 5. Bottom: 3H NMR spectrum (CD2Cl2, 533 MHz).
this temperature, the hydride resonance is observed at -9.64
ppm. Although the hydride resonance of the H3 isotopomer
cannot be observed at this temperature due to efficient dipolar
relaxation, we estimate its chemical shift to be ca. -10.3 ppm.
The isotope shift, δ(HD2) - δ(H3), at this temperature is then
ca. 660 ppb.
Complex 8 fails to exchange with D2 under the low-
temperature conditions at which it is stable. In this case, partial
deuteration of the hydride positions was accomplished by
reacting TpRh(PPh3)D2 with 1 equiv of [H(Et2O)2][B(Ar)4] at
low temperature. The hydride positions are estimated to contain
ca. 55% deuterium by integration of the hydride resonance
against the Tp resonances in the 1H NMR spectrum (one scan).
A broad featureless hydride resonance at the same chemical shift
as that of the undeuterated complex is observed for the mixture
of 8, 8-d1, and 8-d2. The chemical shift of this resonance shows
no significant temperature dependence.
NMR Spectra of Partially Tritiated Complexes. Incorpo-
ration of tritium into the hydride positions of 5-7 was observed
when CD2Cl2 solutions were exposed to T2 gas (ca. 200 Torr)
for several hours at room temperature. Distinct hydride
resonances for the H3 and H2T isotopomers were observed in
the 1H NMR spectrum. The central peak of the HT2 triplet was
observed in the case of 6 and 7 but was not detected for 5
because the concentration of the HT2 isotopomer was too small.
The sample activity in each case is estimated to be less than 10
mCi. For 5, the hydride resonance of the H2T isotopomer is
shifted downfield of the H3 isotopomer by 280 ppb at 240 K
(Figure 4). The H2T resonance appears as a broad doublet of
doublets due to coupling to phosphorus (JP-H ) 9.9 Hz) and
tritium (JT-H ) 54.5 Hz). The hydride resonance of the H3
isotopomer is a doublet (JP-H ) 11.6 Hz), due to coupling to
phosphorus (see Figure 4).
A CDCl2F solution composed principally of 5-d2 was
1
investigated at low temperature by H NMR spectroscopy. As
the temperature was lowered, the hydride resonance of the HD2
isotopomer continued to shift downfield; however, no evidence
for decoalescence of this resonance was observed to 127 K. At

Dihydrogen/Hydride Complexes of Rhodium and Iridium
J. Am. Chem. Soc., Vol. 119, No. 45, 1997 11031
Table 3. Hydride Chemical Shifts and JP-H, JP-T, and JH-T
Coupling Constants for Partially Tritiated Isotopomers of 5-7
Table 4. Significant Bond Distances and Angles for 5‚CH2Cl2
Distances (Å)
1
3
observed by H and H NMR Spectroscopy at 240 K
Ir-P
2.280(3)
2.105(7)
1.772(12)
1.795(11)
Ir-N(2)
Ir-N(6)
P-C(11)
2.161(8)
2.070(8)
1.805(10)
J
P-H or
Ir-N(3)
P-C(10)
P-C(12)
complex
isotopomer δ (ppm) JP-T (Hz) JH-T (Hz)
5
1H NMR
3H NMR
H3
H2T
T3
T2H
TH2
H3
H2T
HT2
T3
TH2
H3
H2T
HT2
T3
T2H
TH2
-10.35
-10.07
-10.35
-10.50
-10.75
-9.70
11.6
9.9
12.9
11.5
a
9.1
a
a
a
11.6
11.9
a
54.5
Angles (deg)
5
P-Ir-N(2)
96.0(2)
86.3(3)
P-Ir-N(3)
P-Ir-N(6)
N(3)-Ir-N(6)
Ir-P-C(11)
B-Ir-P
177.6(2)
94.(2)
85.5(3)
114.0(4)
128.6(2)
63.3
53.2
N(2)-Ir-N(3)
N(2)-Ir-N(6)
Ir-P-C(10)
Ir-P-C(12)
84.3(3)
6
1H NMR
115.0(4)
114.0(4)
-9.44
52.6
-9.23
-9.70
a
6
7
3H NMR
1H NMR
-10.16
-11.06
-10.75
-10.52
-11.06
-11.22
-11.62
52.2
observed. The closest Ir-F contact is 3.541(15) Å. The closest
contact between iridium and the CH2Cl2 of crystallization is
3.832(15) Å. The bond lengths and bond angles observed for
5‚CH2Cl2 are typical of Ir(III) Tp complexes.13,14 The Ir-N
bond distances to the equatorial pyrazolyl ligands are signifi-
cantly different (2.161(8) vs 2.070(8) Å), which may relate to
the hydride structure. Details of the structure solution and
refinement are included in the Experimental Section.
51.7
a
a
a
a
a
7
3H NMR
65.7
51.0
a Not resolved.
Discussion
Synthesis. In contrast to the well-developed synthetic routes
used to prepare Cp*Ir(PR3)H2 and CpIr(PR3)H2 complexes,
similar approaches for the corresponding Tp complexes are
unsuccessful. The immediate precursor to Cp*Ir(PR3)H2 is the
corresponding dichloride complex. These are easily obtained
upon reaction of [Cp*IrCl2]2 with a range of phosphine donor
ligands.15 Attempts by Powell and co-workers to prepare TpR2-
Ir(L)Cl2 complexes from [TpR2IrCl2]2 were reported to fail
except in the case of L ) AsPhMe2.16 The diiodide complexes
CpIr(PR3)I2 employed in the Cp chemistry are prepared in a
one-pot reaction from CpIr(C2H4)2.7 Both ethylene ligands are
displaced upon addition of 1 equiv of I2 to afford an iodide-
bridged oligomer, [CpIrI2]n, to which 1 equiv of phosphine is
added in a second step to yield CpIr(PR3)I2. We have found
that TpIr(C2H4)2 reacts with I2 to yield [TpIr(C2H4)2I]I, which
reacts with PPh3 to give [TpIr(PPh3)(C2H4)I]I.17 The ethylene
ligand in this complex is not easily displaced.
We find that the TpIr(PR3)H2 complexes can be prepared
from [(MeCN)3Ir(PR3)H2]+, which is readily available by the
reaction of acetonitrile with [Cp*Ir(PR3)H3]+ as reported by
Tilset and Petersen. A more efficient approach involves reacting
TpIr(C2H4)2 sequentially with 1 equiv of a phosphine ligand
and then adding H2 in a second step to give TpIr(PR3)H2. This
reaction is easily extended to include the related rhodium
complexes. The details of this chemistry are reported sepa-
2
Figure 5. ORTEP drawing (50% probability ellipsoids) for [TpIr-
(PMe3)(H2)H]BF4. Hydrogen atoms have been omitted for clarity.
In the 3H NMR spectrum, distinct hydride resonances for the
T3, T2H, and TH2 isotopomers were observed. The T2H and
TH2 isotopomers are shifted significantly upfield of the T3
isotopomer. The isotope shifts at 240 K, ∆3 ) δ(T2H) - δ-
(T3) and ∆4 ) δ(TH2) - δ(T2H) are -150 and -250 ppb,
respectively. The T3 isotopomer was observed at the same
chemical shift as the H3 isotopomer in the 1H NMR spectrum.
The T2H isotopomer appears as a doublet of doublets coupling
to phosphorus (JP-T ) 11.5 Hz) and hydrogen (JT-H ) 63.3
Hz). The TH2 isotopomer is a broad triplet with an observed
T-H coupling constant of 53.2 Hz, which is identical within
experimental error to the H-T coupling constant observed for
the H2T isotopomer in the 1H NMR spectrum. The 1H and 3H
NMR data for 5-7 are summarized in Table 3.
rately.12 A preparation of complex 3 by thermolysis of TpMe
-
IrH4 with PMe3 has been briefly reported, but no details of
characterization were given.18
Characterization of the Cationic Hydride Complexes. The
cationic species resulting from protonation of 1-4 give spec-
troscopic data consistent with new complexes containing a
mirror plane of symmetry. A total of three hydride ligands (no
structure implied) is indicated by the observation of a quartet
in the 31P NMR spectrum of 5 recorded with decoupling of the
(13) Ferna´ndez, M. J.; Rodriguez, M. J.; Oro, L. A.; Lahoz, F. J. J. Chem.
Soc., Dalton Trans. 1989, 2073-2079.
Structural Characterization. Crystals of 5‚CH2Cl2 suitable
for X-ray diffraction were obtained by slow diffusion of Et2O
into a concentrated CH2Cl2 solution of 5. An ORTEP drawing
of the iridium cation and key structural features are given in
(14) Bovens, M.; Gerfin, T.; Gramlich, V.; Petter, W.; Venanzi, L. M.;
Haward, M. T.; Jackson, S. A.; Eisenstein, O. New. J. Chem. 1992, 16,
337-345.
(15) Kang, J. W.; Moseley, K.; Maitlis, P. M. J. Am. Chem. Soc. 1969,
91, 5970-5977.
-
Figure 5 and Table 4, respectively. The hydrogen atoms, BF4
(16) May, S.; Reinsalu, P.; Powell, J. Inorg. Chem. 1980, 19, 1582-
anion, and the CH2Cl2 of crystallization have been omitted for
clarity. The metal-bound dihydrogen ligand and the hydride
ligand were not reliably located. No unusual intermolecular
1589.
(17) Oldham, W. J., Jr.; Heinekey, D. M. Unpublished results.
(18) Paneque, M.; Poveda, M. L.; Toboada, S. J. Am. Chem. Soc. 1994,
116, 4519-4520.
-
interaction between BF4 and the iridium metal center is

11032 J. Am. Chem. Soc., Vol. 119, No. 45, 1997
Oldham et al.
Table 5. H-H Bond Lengths from T1(min) Data in the Limits of
deuterium atoms in the hydride positions of dihydrogen and
polyhydride complexes.7,8,20-24 A few reports of downfield
isotope shifts in both these classes of molecules have also
appeared.5,18,25-35 In some cases, the observed isotope shifts
have been rationalized by invoking isotopic perturbation of
equilibria;10,27,34 however, a detailed study of this phenonenon
has not been reported.
Slow and Fast Rotations
H-H (Å)
complex
slow rotation
fast rotation
[TpIr(PMe3)(H2)H]+ (5)
[TpIr(PPh3)(H2)H]+ (6)
[TpRh(PPh3)(H2)H]+ (8)
1.11
1.12
0.92
0.88
0.89
0.73
Two types of equilibrium situations may be perturbed upon
substitution of the hydride positions of 5-7 with deuterium or
tritium. The observed isotope shifts may result from perturba-
tion of an equilibrium between a dihydrogen/hydride complex
and a trihydride complex (eq 6).27,36 A second possibility is
organic protons. At all temperatures, only a single hydride
resonance is observed in the H NMR spectrum, which may
1
result from a fluxional trihydride or a fluxional dihydrogen/
hydride complex. These possibilites can be distinguished by a
combination of T1(min) measurements and analysis of the effect
of partial substitution of the hydride ligands with deuterium or
tritium.
The minimum of the longitudinal relaxation time (T1(min))
of the dihydrogen ligand has been shown to be a sensitive
indicator of the H-H bond length.4 The short T1(min) values
observed for 5-8 are qualitatively consistent with the presence
of a bound H2 ligand in a dihydrogen/hydride structure. An
accurate calculation of the H-H bond length requires that the
mutual relaxation rate of the hydrogen atoms in the coordinated
dihydrogen ligand (Rd-d) be known explicitly (1/T1 ) relaxation
rate, R). The observed rate of dipolar relaxation for a dihy-
that, within a single dihydrogen/hydride ground state structure,
the heavy hydrogen isotopes may concentrate in a particular
site (eq 7). These two possibilities can be distinguished by
comparing the isotope shifts upon partial deuteration using both
2
1H and H NMR spectroscopy. In our case, very broad lines
drogen ligand (RH ) is actually the sum of the mutual H-H
2
2
dipolar relaxation (Rd-d) and the relaxation resulting from
were observed in the H NMR spectra, so we instead incorpo-
interactions with other dipoles in the molecule (Ro), thus RH
)
rated tritium in the hydride positions and investigated this
problem using 1H and 3H NMR spectroscopy. If the equilibrium
in eq 6 is perturbed upon isotopic substitution, identical
isotopomers will be observed at the same chemical shifts in
the H and H NMR spectra, but the H3 and T3 isotopomers
will not coincide.37 If instead, an equilibrium between different
sites in the same structure is perturbed (eq 7), then the H3 and
T3 isotopomers will be observed at approximately the same
2
Rd-d + Ro.5 Additionally, in the case of fluxional polyhydride
complexes suspected to contain a dihydrogen ligand, the
observed T1(min) value is the population-weighted average of
all the hydride sites. For the dihydrogen/hydride complexes
1
3
under study in this work, the observed relaxation rate (Robs
)
1/T1) is given by Robs ) (2RH + RH)/3 where RH and RH are
2
2
the relaxation rates of the dihydrogen and hydride ligands. The
relaxation rate of the terminal hydride ligand (RH) can be
estimated from the T1(min) value of the parent dihydride
complexes such as 1. Similarly, Ro can also be estimated from
this measurement. Calculation of Rd-d for the dihydrogen ligand
of 5 is shown as follows:
(20) Hamilton, D. G.; Luo, X. L.; Crabtree, R. H. Inorg. Chem. 1989,
28, 3198-3203.
(21) Luo, X. L.; Crabtree, R. H. J. Am. Chem. Soc. 1990, 112, 4813-
4821.
(22) Luo, X.; Michos, D.; Crabtree, R. H. Organometallics 1992, 11,
237-241.
(23) Baird, G. J.; Davies, S. G.; Moon, S. D.; Simpson, S. J.; Jones, R.
H. J. Chem. Soc., Dalton Trans. 1985, 1479-1486.
(24) Casey, C. P.; Tanke, R. S.; Hazin, P. N.; Kemnitz, C. R.; McMahon,
R. J. Inorg. Chem. 1992, 31, 5474-5479.
Robs(5) ) (RH + 2RH )/3 ) (RH + 2(Rd-d + Ro))/3
2
if RH and Ro ≈ Robs(1), then
(25) Heinekey, D. M.; Payne, N. G.; Sofield, C. D. Organometallics 1989,
8, 1824-1826.
(26) Antoniutti, S.; Albertin, G.; Amendola, P.; Bordignon, E. J. Chem.
Soc., Chem. Commun. 1989, 229-230.
R
obs(5) ) (2Rd-d + 3Robs(1))/3
(27) Earl, K. A.; Jia, G.; Maltby, P. A.; Morris, R. H. J. Am. Chem. Soc.
1991, 113, 3027-3039.
and
(28) Nanz, D.; von Philipsborn, W.; Bucher, U. E.; Venanzi, L. M. Magn.
Reson. Chem. 1991, 29, S38-44.
Rd-d ) 3(Robs(5) - Robs(1))/2 ) 69 s-1
(29) Michos, D.; Luo, X.; Crabtree, R. H. Inorg. Chem. 1992, 31, 4245-
4250.
(30) Collman, J. P.; Wagenknecht, P. S.; Hutchison, J. E.; Lewis, N. S.;
Lopez, M. A.; Guilard, R.; L’Her, M.; Bothner-By, A. A.; Mishra, P. K. J.
Am. Chem. Soc. 1992, 114, 5654-5664.
(31) Miller, R. L.; Toreki, R.; LaPointe, R. E.; Wolczanski, P. T.; Van
Duyne, G. D.; Roe, D. C. J. Am. Chem. Soc. 1993, 115, 5570-5588.
(32) Bianchini, C.; Linn, K.; Masi, D.; Peruzzini, M.; Polo, A.; Vacca,
A.; Zanobini, F. Inorg. Chem. 1993, 32, 2366-2376.
(33) Heinekey, D. M.; Liegeois, A.; van Roon, M. J. Am. Chem. Soc.
1994, 116, 8388-8389.
(34) Albinati, A.; Bakhmutov, V. I.; Caulton, K. G.; Clot, E.; Eckert, J.;
Eisenstein, O.; Gusev, D. G.; Grushin, V. V.; Hauger, B. E.; Klooster, W.
T.; Koetzle, T. F.; McMullan, R. K.; O’Loughlin, T. J.; Pe´lissier, M.; Ricci,
J. S.; Sigalas, M. P.; Vymenits, A. B. J. Am. Chem. Soc. 1993, 115, 7300-
7312.
(35) Bullock, R. M.; Song, J.; Szalda, D. J. Organometallics 1996, 15,
2504-2516.
Similar calculations have been carried out for 6 and 8. The
H-H bond length can now be calculated in the limits of slow
and fast hydrogen rotation by the method of Halpern and co-
workers. We find that the H-H bond length of the dihydrogen
ligand of 5 is bracketed by distances of 1.11 and 0.88 Å (Table
5). We can estimate the true H-H bond length after first
considering the results obtained upon partial substitution of the
hydride positions with deuterium or tritium.
Isotopic Perturbation of Equilibrium. We propose that the
unusually large and temperature-dependent isotope shifts ob-
served for the iridium dihydrogen/hydride complexes are a
manifestation of isotopic perturbation of equilibria.19 Generally
only small (0-70 ppb) temperature-independent upfield isotope
shifts are commonly observed upon partial substitution of
(36) Luo, X.; Crabtree, R. H. J. Am. Chem. Soc. 1990, 112, 6912-6918.
(37) An example of an equilibrium between two different structures has
been characterized using this method by Hartwig and De Gala in the case
of an equilibrium between a hydroborate and a boryl complex: Hartwig, J.
F.; De Gala, S. R. J. Am. Chem. Soc. 1994, 116, 3661-3662.
(19) Saunders, M.; Kates, M. R. J. Am. Chem. Soc. 1977, 99, 8070-
8071.

Dihydrogen/Hydride Complexes of Rhodium and Iridium
J. Am. Chem. Soc., Vol. 119, No. 45, 1997 11033
1
chemical shift, but the isotope shifts observed in the H NMR
can be rewritten in terms of the Boltzmann factor, A1. For this
problem, a ) a′ ) 1/(2 + A1) and b ) A1/(2 + A1), leading to
eq 11. In a similar fashion, the observed JH-D coupling constant
is given by eq 12. If the two-bond 2JH-D coupling between the
spectrum will be in the direction opposite to that observed in
the H NMR spectrum.38 The spectra shown in Figure 4 and
3
the data tabulated in Table 3 confirm that the isotope effect
results from a nonstatistical site preference for deuterium (or
tritium) in a single dihydrogen/hydride structure. That is, the
equilibrium which is perturbed upon isotopic substitution is that
described by eq 7. An important assumption which we make
in the quantitative analysis outlined below is to ignore contribu-
tions from intrinsic isotope effects. We assume, for example,
that the isotope shift in the H NMR spectrum resulting from
partial deuteration arises entirely from isotopic perturbation
effects, rather than intrinsic isotope effects.
1JH-D + 2JH-D
1JH-D + 2JH-D
J(H D) ) a
+ a′
+ b(2J
)
H-D
(
)
(
)
2
2
2
(12)
(13)
1JH-D
1
J(H D)
)
2
2 + A1
In principle, the observed chemical shifts and the JH-D
coupling data can be analyzed quantitatively to calculate the
limiting chemical shifts of the dihydrogen ligand (δH2) and the
dihydrogen ligand and the terminal hydride ligand is assumed
to be zero and the a and a′ terms are combined, then the equation
can be rewritten in terms of the Boltzmann factor, A1, and
simplified to give eq 13.
In a similar fashion, 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.
The observed 1H resonance of the HD2 isotopomer is given by
eq 14, where c, d, and d′ are the mole fractions of the species
1
terminal hydride ligand (δH), JH-D for the dihydrogen ligand,
and the energy difference between deuterium substitution of the
dihydrogen site versus the hydride site. This method is based
on the procedure used by Calvert and Shapley39 in the study of
fluxional agostic methyl groups. The general approach is to
1
relate the dynamically averaged observables in the H NMR
spectrum (hydride chemical shifts and coupling constants) to
the limiting values from which they originate.
δ(HD ) ) c(δH) + d(δH ) + d′(δH )
(14)
2
2
2
The hydride chemical shift of the H3 isotopomer (δH ) is
3
simply the statistical average of the dihydrogen site and the
2A2δH + δH
2
hydride site, thus δH ) (2δH + δH )/3. However, upon
3
2
δ(HD )
)
(15)
2
2A2 + 1
substitution of one deuterium atom in the hydride positions an
equilibrium between three different species is obtained (eq 8).
shown in eq 9. Analogous to the previous presentation, c + d
+ d′ ) 1 and the Boltzmann factor A2 is equal to d/c ) d′/c.
Rewriting the mole fractions in terms of A2, c ) 1/(2A2 + 1)
and d ) d′ ) A2/(2A2 + 1). Substituting these back into eq 14
leads to eq 15. The observed H-D coupling of the HD2
isotopomer is given by eq 16. If the 2JH-D term is assumed to
1JH-D + 2JH-D
1JH-D + 2JH-D
J(HD ) ) c(2JH-D) + d
+ d′
(
)
(
)
2
2
2
(16)
(17)
1JH-DA2
J(HD )
)
A similar equilibrium mixture is obtained if two deuterium
atoms are incorporated in the hydride positions (eq 9). In this
analysis we assume that there is no preference between a and
a′. That is, a and a′ are assumed to be degenerate. We also
assume that the chemical shift of the exo hydrogen atom in a is
the same as the endo hydrogen atom in a′. The observed
chemical shift of the H2D isotopomer is given by eq 10, where
2
2A2 + 1
be zero and the substitutions for the mole fractions are made,
then eq 16 reduces to eq 17.
The results of this analysis (calculated using the data presented
in Tables 2 and 3) for complexes 5-7 are presented in Table
6. In the case of 5, the limiting spectroscopic parameters were
also calculated at every temperature indicated in Figures 2 and
3. In this way, the standard deviation (or precision) in each
δH + δH
δH + δH
2
2
δ(H D) ) a
+ a′
+ b(δ ) (10)
(
)
(
)
H2
2
2
2
calculated parameter for complex 5 was obtained: δH ) -8.4
2
1
( 0.1 ppm, δH ) -14.4 ( 0.3 ppm, JH-D ) 24.6 ( 0.3 Hz,
δH + δH + A1δH
2
2
∆E1 ) -130 ( 12 cal/mol, and ∆E2 ) -107 ( 12 cal/mol.
However, comparison of the results calculated from the 1H and
3H NMR data for complexes 6 and 7 indicates that systematic
error introduces larger uncertainties in the calculated parameters
than the precision of the measurement would suggest. The error
in the calculated chemical shifts (estimated from the standard
deviation of the two measurements) is approximately (1 ppm.
The error in the calculated energy differences is (25%, and
that in the calculated H-T coupling constants is (4 Hz. Despite
these uncertainties, the analysis does confirm that the heavy
hydrogen isotope concentrates in the terminal hydride position.
Thus, in relation to eqs 8 and 9, we conclude that the equilibria
lie to the right. Because the chemical shift of the dihydrogen
ligand is downfield of the hydride signal, this leads to the
δ(H D)
)
(11)
(
)
2
2 + A1
a, a′, and b are the mole fractions of each species. Within this
definition, the sum of a + a′ + b ) 1. The Boltzmann factor
A1, defined as e-∆E/RT, may be introduced to account for an
energy difference between a (a′) and b. A1 is equal to b/a )
b/a′. Thus a and a′ can be combined and the mole fractions
(38) An example of an equilibrium between two different proton sites
in the same structure, which has been characterized using both 1H and
2H NMR spectroscopy, is provided by Stone’s investigation of
[Fe2(µ-CH3)(µ-CO)(µ-dmpm)Cp2]PF6: Dawkins, G. M.; Green, M.; Orpen,
A. G.; Stone, F. G. A. J. Chem. Soc., Chem. Commun. 1982, 41-43.
(39) Calvert, R. B.; Shapley, J. R. J. Am. Chem. Soc. 1978, 100, 7726-
7727.

11034 J. Am. Chem. Soc., Vol. 119, No. 45, 1997
Table 6. Calculated Parameters from Isotopic Perturbation Analysis of H and H NMR Spectra of 5-7 Acquired at 240 K
Oldham et al.
1
3
5a
5 (3H)b
6 (1H)c
6 (3H)b
7a
7 (1H)c
7 (3H)b
δH2 (ppm)
-8.3
-14.6
24.7
-8.3
-7.2
-8.0
-9.1
-15.0
24.8
-8.7
-9.2
δH (ppm)
-14.4
-14.6
-13.2
-15.9
-14.8
1JH-D (Hz)
1JH-T (Hz)
∆E1 (cal/mol)
∆E2 (cal/mol)
177
176
180
178
183
-125d
-103e
-136f
-110g
-143f
-143g
-177f
-99g
-153d
-118e
-174f
-171g
-203f
-132g
a
b
c
1H NMR analysis of a partially deuterated sample. 3H NMR analysis of a partially tritiated sample. 1H NMR analysis of a partially tritiated
sample. d Energy difference between Tp(L)Ir(HD)H and Tp(L)Ir(H2)D (see eq 5). e Energy difference between Tp(L)Ir(D2)H and Tp(L)Ir(HD)D
(eq 6). f Energy difference between Tp(L)Ir(HT)H and Tp(L)Ir(H2)T. g Energy difference between Tp(L)Ir(T2)H and Tp(L)Ir(HT)T.
1
1
significant downfield shift in the H NMR spectrum and the
linear inverse correlation between JH-D and the H-H bond
3
corresponding upfield shift in the H NMR spectrum.
length (determined by neutron diffraction or solid state NMR
methods) as reported by Heinekey and Luther.48 A similar
analysis has been used to infer the relatively slow rotational
Structural Considerations. The above analysis establishes
that a dihydrogen/hydride structure is adopted by complexes
5-8. In contrast, the Cp (or Cp*) analogs such as [CpIr(PR3)-
H3]+ are trihydride complexes, with appropriately long T1(min)
values of ca. 300 ms.7 Protonation of CpRh(PiPr3)H2 with HPF6
is reported to yield a hydride-bridged dimer, [{CpRh(PiPr3)}2-
(µ-H3)]PF6, and H2.39 Mononuclear rhodium analogs of [CpIr-
(PR3)H3]+ are not known. A similar change in structure from
trihydride to dihydrogen/hydride upon substitution of Tp for
Cp ligands has been reported by Chaudret and co-workers in
related ruthenium complexes.41,42 The neutral CpRu(PR3)H3
complexes are characterized as classical trihydride complexes;43
49
motion of the dihydrogen ligand in [Cp*Ru(dppe)(H2)]BF4
and the dicationic complex [Os(bpy)(PPh3)2(CO)(H2)](OTf)2.48
For the iridium complexes reported here, the H-D coupling in
the bound dihydrogen ligand is ca. 25 Hz, which is consistent
with an H-H distance of ca. 1 Å. These complexes may belong
to the currently limited group of dihydrogen complexes such
as [Os(H2)(dppe)2X]+ (X ) Cl, Br),50 in which dihydrogen
rotational motion is comparable to the rate of molecular
tumbling.
In the case of the rhodium complex 8, the H-H bond length
is calculated to be between 0.92 Å (slow rotation) and 0.73 Å
(fast rotation). Since the H-H bond length of free hydrogen
is 0.74 Å, the value calculated by assuming fast rotation is
probably not correct. Unfortunately, the rhodium complex gave
only a broad hydride resonance upon partial deuteration of the
hydride positions, so a JH-D value is not available. Assuming
that hydrogen rotation is relatively slow in complex 8, as found
for the iridium analogs, the longer H-H distance of ca. 0.92 Å
results. Although a definitive value for the H-H distance in
complex 8 cannot be obtained, it is clear that the H-H distance
is significantly longer in the iridium complexes than for the
corresponding rhodium complex.
however, substitution of the Cp ligand for TpMe results in stable
2
complexes of the form TpMe Ru(L)(H2)H. In this case, L can
2
be a phosphine, nitrogen, or sulfur donor ligand.
In most cases, the Tp ligand is considered to be a more
efficient donor than the Cp ligand.9 This property would seem
to favor a trihydride structure over the dihydrogen/hydride
structure that is actually observed in this system. However,
for the later transition metals such as iridium, the Tp ligand
transfers less electron density to the metal center. This is
reflected in the higher CO stretching frequency of TpIr(CO)H2
(2020 cm-1, CH2Cl2)44 versus CpIr(CO)H2 (2002 cm-1, CH2-
Cl2)45 and also TpIr(CO)(C2H4) (2000 cm-1, cyclohexane)46
versus CpIr(CO)(C2H4) (1980 cm-1, cyclohexane).47 The
difference in relative donor ability is probably a function of
orbital overlap between iridium and the Cp or Tp ligands. The
soft iridium center is expected to form stronger, more covalent
bonds with the soft Cp ligand. The bonding between iridium
and the hard pyrazolyl nitrogen donors is presumably more ionic
in character. Steric factors will also favor complexes with a
reduced coordination number for Tp (cone angle 184°) versus
Cp (cone angle 136°) complexes.
Hydride Dynamics. Although a static low-temperature-
limiting H NMR spectrum was not obtained for complexes
5-8, partial deuteration (tritiation) and the observation of IPR
as detailed above allow the limiting chemical shifts δH and δH
1
2
to be estimated. From these data and the observation of a single
hydride resonance for 5 at 127 K, the activation energy for
exchange between the dihydrogen ligand and the hydride ligand
can be calculated as ∆Gq e 5 kcal/mol. It should be noted
that these observations were made on a sample of complex 5
which was heavily deuterated in the hydride positions. This
rules out the possibility that exchange coupling could contribute
to the observed spectrum as has been noted for other dynamic
polyhydrides.51
While our data offer no mechanistic insight, it is instructive
to consider possible mechanisms for the rapid exchange of all
three hydride nuclei. Two distinct dynamic processes are
required. Rotation of the dihydrogen ligand around the M-H2
bond axis must be rapid on the chemical shift time scale. This
is quite reasonable, since reported barriers to hydrogen rotation
are very low, except in d2 systems.52,53 A second process
Estimation of H-H Distances. As noted by Morris and
co-workers, H-H bond distances inferred from T1(min) data
depend upon assumptions about the relative rate of dihydrogen
rotation.6 In the case of iridium complexes 5 and 6, the observed
T1(min) values of 21-22 ms are consistent with an H-H
distance of 0.88 Å (fast rotation) or 1.11 Å (slow rotation).
Further insight can be obtained by making use of the roughly
(40) Werner, H.; Wolf, J.; Ho¨hn, A. J. Organomet. Chem. 1985, 287,
395-407.
(41) Moreno, B.; Sabo-Etienne, S.; Chaudret, B.; Rodriguez-Fernandez,
A.; Jalon, F.; Trofimenko, S. J. Am. Chem. Soc. 1994, 116, 2635-2636.
(42) Moreno, B.; Sabo-Etienne, S.; Chaudret, B.; Rodriguez, A.; Jalon,
F.; Trofimenko, S. J. Am. Chem. Soc. 1995, 117, 7441-7451.
(43) Arliguie, T.; Border, C.; Chaudret, B. Organometallics 1989, 8,
1308.
(48) Heinekey, D. M.; Luther, T. A. Inorg. Chem. 1996, 35, 4396-4399.
(49) Klooster, W. T.; Koetzle, T. F.; Jia, G.; Fong, T. P.; Morris, R. H.;
Albinati, A. J. Am. Chem. Soc. 1994, 116, 7677-7681.
(50) 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, 5396-5407.
(51) Heinekey, D. M.; Payne, N. G.; Sofield, C. D. Organometallics 1990,
9, 2643-2645.
(52) Jalo´n, F. A.; Otero, A.; Manzano, B. R.; Villasen˜or, E.; Chaudret,
B. J. Am. Chem. Soc. 1995, 117, 10123-10124.
(44) Fernandez, M. J.; Rodriguez, M. J.; Oro, L. A. J. Organomet. Chem.
1992, 438, 337-342.
(45) Shapley, J. R.; Adair, P. C.; Lawson, J. R.; Pierpont, C. G. Inorg.
Chem. 1982, 21, 1702-1704.
(46) Ciriano, M. A.; Ferna´ndez, M. J.; Modrego, J.; Rodr´ıguez, M. J.;
Oro, L. A. J. Organomet. Chem. 1993, 443, 249-252.
(47) Szajek, L. P.; Lawson, R. J.; Shapley, J. R. Organometallics 1991,
10, 357-361.

Dihydrogen/Hydride Complexes of Rhodium and Iridium
J. Am. Chem. Soc., Vol. 119, No. 45, 1997 11035
benzophenone (except CH2Cl2 from P2O5) under a nitrogen atmosphere.
Deuterated NMR solvents (purchased from Cambridge Isotope Labo-
ratories) were degassed and were stored over CaH2 (CD2Cl2) or Na-
K-benzophenone (C6D6, toluene-d8, THF-d8). Hydrogen (99.999%)
and ethylene (99.7%) were purchased from Airco. Potassium hydrotris-
(1-pyrazolyl)borate (KTp) was prepared by the procedure of Trofi-
menko.55 RhCl3‚3H2O was purchased from Pressure Chemicals.
(NH4)2IrCl6 was recovered from laboratory iridium residues by fol-
lowing published procedures.56 Unless stated otherwise, all other
reagents were obtained from Aldrich and used as received.
Variable-temperature NMR measurements were performed using the
Bruker B-VT1000 temperature control module with a copper-
constantan thermocouple. Temperature calibration was obtained by
measurement of the chemical shift difference between the -CH3 and
-OH peaks of a standard methanol sample using the method of Van
Geet.57 Tritium NMR spectra were obtained on a Bruker WM-500
spectrometer at 533 MHz with a Cryomagnetics tritium probe and
externally referenced to (CHDTCO2)-.58 The procedures employed to
safely store and manipulate tritium gas have been described previously.8
Infrared spectra were recorded as Nujol mulls between NaCl plates
on a Perkin-Elmer model 1600 Fourier transform spectrophotometer
(2.0 cm-1 resolution). Elemental analyses were performed by Canadian
Microanalytical Services, Ltd., Vancouver, BC, Canada.
exchanges hydrogen nuclei between the dihydrogen and the
hydride ligands. Since our complexes are positively charged,
it seems reasonable to speculate that a proton transfer may be
involved, although it has been reported that the hydride ligands
of the closely related ruthenium complexes TpMe Ru(PR3)(H2)H
2
also exchange rapidly on the NMR time scale. In this case,
heterolytic cleavage of the dihydrogen ligand is not facilitated
by an overall positive charge.42
Comparison with Related Complexes. There are several
examples of fluxional cis-dihydrogen/hydride complexes in the
literature. Partial deuteration has been generally employed to
establish the presence of the H2 ligand, but large isotope effects
such as those observed for 5-7 have not been previously
reported. A modest upfield isotope shift has been reported by
Field and co-workers in [Fe{P(CH2CH2CH2PMe2)3}(H2)H]-
BPh4.54 In this iron complex, JH-D(H2D) > JH-D(HD2) (10 and
9 Hz, respectively), consistent with a slight preference for
deuterium to concentrate in the dihydrogen ligand, opposite to
the preference observed in 5-7. Distinct hydride resonances
attributed to the H3 and the H2D isotopomers are observed upon
partial deuteration of [Os{P(CH2CH2PPh2)3}(H2)H]BPh4.32 The
reported isotope shifts are consistent with a modest preference
for deuterium to concentrate in the hydride ligand; however,
the temperature dependence of the isotope shifts was not
reported. The observed H-D couplings for this osmium
complex are nearly identical to the H-D coupling constants
for 5 and 7, indicating that the dihydrogen ligand in each of
these complexes is in a similar electronic environment.
Large and temperature-dependent isotope effects have been
[(C5Me5)Ir(PMe3)H3]SO3CF3. To a 300 mL glass bomb fitted with
a Kontes valve were added (C5Me5)Ir(PMe3)Cl259 (2.01 g, 4.24 mmol)
and AgSO3CF3 (2.37 g, 9.2 mmol). CH2Cl2 (50 mL) was vacuum-
transferred into the flask and the head space backfilled with H2 (1000
Torr). The flask was covered with aluminum foil, and the mixture
was allowed to stir overnight to yield a pale orange solution containing
a white precipitate (AgCl), which was filtered off. The volume of the
filtrate was reduced under vacuum. Layering with Et2O gave colorless
crystals upon standing at -30 °C for 3 days. These were washed with
Et2O (3 × 5 mL) and dried under vacuum. Yield: 2.06 g (88%). The
1H NMR spectrum of the product matches that previously reported.60
[(C5Me5)Ir(PPh3)H3]BF4. This complex was prepared from (C5-
reported in partially deuterated TpMe IrH4, along with substantial
2
H-D coupling, which is suggestive of a dihydrogen/dihydride
structure. However, this complex has quite long hydride
relaxation times, which led Poveda and co-workers to suggest
a tetrahydride structure.18 The corresponding rhodium complex
15
Me5)Ir(PPh3)Cl2 as above but using AgBF4. Yield: 60%. The
1
product was characterized by H NMR spectroscopy.11
TpMe Rh(H2)H2 has been formulated as a dihydrogen/dihydride
TpIr(PMe3)H2 (1). A solution of [(C5Me5)Ir(PMe3)H3]SO3CF3 (146
mg, 0.263 mmol) in CH3CN (15 mL) was stirred under Ar at room
temperature for 2 days. NaTp (66 mg, 0.280 mmol) was added. After
the mixture was stirred for an additional 2 days, the volatiles were
removed under vacuum and the resulting residue was taken up in a
minimum of toluene. Complex 1 was isolated by chromatography on
alumina, eluting with toluene. The toluene was removed by rotary
evaporation and the residue crystallized by adding wet MeOH. The
resulting white powder was dried under vacuum to yield 70 mg (55%)
of analytically pure product. 1H NMR, δ (CD2Cl2): 7.70, 7.63 (d, 2
H each, 3,5-pzeq); 7.67, 7.59 (m, 1 H each, 3,5-pzax); 6.18 (t, 2 H,
4-pzeq); 6.08 (m, 1 H, 4-pzax); 1.63 (d, JP-H ) 10.0 Hz, 9 H, PMe3);
-21.3 (d, JP-H ) 25.0 Hz, 2 H, Ir-H). 13C{1H} NMR, δ (CD2Cl2):
146.2, 134.0 (s, 1 C, 3,5-pzax); 144.0, 138.8 (s, 2 C, 3,5-pzeq); 105.8
(s, 1 C, 4-pzax); 105.6 (s, 2 C, 4-pzeq); 21.7 (d, JP-C ) 39.4 Hz). 31P{Me
1H} NMR, δ (CD2Cl2): -45.6 (t, JP-H ) 25 Hz). 1H NMR (C6D6):
7.93, 7.34 (br s and m, respectively, 1 H each, 3,5-pzax); 7.57, 7.53 (d,
2 H each, 3,5-pzeq); 5.94 (t, 2 H, 4-pzeq); 5.68 (m, 1 H, 4-pzax); 1.32
(d, JP-H ) 9.9 Hz, PMe3); -20.56 (d, JP-H ) 25.5 Hz, 2 H, Ir-H).
13C{1H}, δ NMR, δ (C6D6): 146.4, 133.3 (s, 1 C each, 3,5-pzax); 144.1,
134.4 (s, 2 C, 3,5-pzeq); 105.7 (s, 1 C, 4-pzax); 105.4 (s, 2 C, 4-pzeq);
21.5 (d, JP-C ) 38.8 Hz, PMe3). 31P{Me 1H} NMR, δ (C6D6): -46.55
(t, JP-H ) 24.3 Hz). IR cm-1: 2482 (νB-H); 2143 (νIr-H). MS: m/z
484 (M+). Anal. Calcd for C12H21BIrN6P: C, 29.83; H, 4.38; N, 17.39.
Found: C, 29.51; H, 4.83; N, 17.26.
2
species on the basis of the observation of H-D coupling and
short T1 values for the hydride ligands.28 The apparent structural
difference between these two complexes rests primarily on T1
data. Further structural information on these interesting com-
plexes would be very desirable.
Conclusion
Protonation of TpM(PR3)H2 (M ) Rh, Ir) complexes affords
highly dynamic dihydrogen/hydride complexes. In the case of
the iridium complexes, partial substitution of the hydride
positions with deuterium or tritium results in large temperature-
dependent isotope shifts, which result from a preference for the
heavy isotope to occupy the hydride site. Analysis of these
data gives the limiting chemical shifts of the dihydrogen and
1
hydride ligands as well as the coupling constant JH-D of the
dihydrogen ligand (ca. 25 Hz). These observations, combined
with short T1(min) values of ca. 22 ms for the Ir complexes,
are consistent with an H-H bond length of ca. 1 Å. For the
Rh complex, the T1(min) value of 7 ms is consistent with the
presence of a bound dihydrogen ligand, but H-D coupling was
not resolvable. The T1 data for the Rh complex give an H-H
distance of 0.73-0.92 Å, depending on the rate of dihydrogen
rotation.
TpIr(PPh3)H2 (2). This complex was prepared as above, starting
from [(C5Me5)Ir(PPh3)H3]BF4. Yield: 60%. 1H NMR (C6D6): 8.04,
(55) Trofimenko, S. Inorg. Synth. 1979, 12, 99.
(56) Kauffman, G. B.; Myers, R. D. Inorg. Synth. 1978, 18, 131-133.
(57) Gordon, A. J.; Ford, R. A. The Chemist’s Companion; Wiley: New
York, 1972; p 303.
(58) Evans, E. A.; Warrell, D. C.; Elvidge, J. A.; Jones, J. R. Handbook
of Tritium NMR Spectroscopy and Applications; Wiley: New York, 1985.
(59) Isobe, K.; Bailey, P. M.; Maitlis, P. M. J. Chem. Soc., Dalton Trans.
1981, 2003-2008.
(60) Gilbert, T. M.; Bergman, R. G. J. Am. Chem. Soc. 1985, 107, 3502-
3507.
Experimental Section
General Methods. All manipulations were conducted under a dry
argon or nitrogen atmosphere using standard Schlenk and drybox
techniques. Solvents were purified by distillation from Na-K-
(53) Sabo-Etienne, S.; Chaudret, B.; el Makarim, H. A.; Barthelat, J.;
Daudey, J.; Ulrich, S.; Limbach, H.; Mo¨ıse, C. J. Am. Chem. Soc. 1995,
117, 11602-11603.
(54) Bampos, N.; Field, L. D. Inorg. Chem. 1990, 29, 587-588.

11036 J. Am. Chem. Soc., Vol. 119, No. 45, 1997
Oldham et al.
7.35 (br s and m, respectively, 1 H each, 3,5-pzax); 7.66-7.59 (m, 6
H, PPh3); 7.49, 6.85 (d, 2 H each, 3,5-pzeq); 6.99-6.95 (m, 9 H, PPh3);
4-pzeq); -9.77 (br, 3 H, Ir-H). 31P{1H} NMR, δ (CD2Cl2): 1.1 (s).
IR, cm-1: 2502 (νB-H); 2197 (νIr-H). Anal. Calcd for C27H28B2F4-
IrN6P: C, 42.83; H, 3.73; N, 11.1. Found: C, 40.00; H, 4.25; N, 9.20.
5.75 (m, 1 H, 4-pzax); 5.67 (t, 2 H, 4-pzeq); -19.70 (d, 2 H, JP-H
)
)
1
23.1 Hz, Ir-H). 31P{aromatic H} NMR, δ (C6D6): 18.8 (t, JP-H
[TpMe Ir(PMe3)(H2)H][B(3,5-(CF3)2C5H3)4] (7). A typical sample
2
22.3 Hz). 1H NMR (CD2Cl2): 7.83, 7.63 (br s and m, respectively, 1
H each, 3,5-pzax); 7.66, 6.61 (d, 2 H each, 3,5-pzeq); 7.39-7.24 (m, 15
H, PPh3); 6.13 (m, 1 H, 4-pzax); 5.86 (t, 2 H, 4-pzeq); -20.47 (d, JP-H
) 22.1 Hz, 2 H, Ir-H). 13C{1H} NMR, δ (CD2Cl2): 146.4, 134.8 (s,
1 C, 3,5-pzax); 143.3, 134.7 (s, 2 C, 3,5-pzeq); 135.3 (d, JP-C ) 56 Hz,
i-C6H5); 134.2, 128.2 (d, JP-C ) 10 Hz, o- and m-C6H5); 130.1 (s,
p-C6H5); 106.0 (s, 1 C, 4-pzax); 105.2 (s, 2 C, 4-pzeq). 31P{aromatic
1H} NMR, δ (CD2Cl2): 16.9 (t, JP-H ) 22.0 Hz). IR, cm-1: 2481
(νB-H); 2179, 2139 (νIr-H). Anal. Calcd for C27H27BIrN6P: C, 48.44;
H, 4.07; N, 12.55. Found: C, 48.20; H, 4.01; N, 12.24.
was prepared by vacuum transfer of CD2Cl2 (0.4 mL) into an NMR
tube containing TpMe Ir(PMe3)H2 (12.8 mg, 0.023 mmol) and [H(Et2O)2]-
2
[B(3,5-(CF3)2C6H3)4]61 (23 mg, 0.023 mmol). 1H NMR, δ (CD2Cl2):
7.73 (br, 8 H, o-C6H3(CF3)2); 7.58 (s, 4 H, p-C6H3(CF3)2); 6.08 (s, 2
H, 4-pzeq); 5.75 (s, 1 H, 4-pzax); 2.46, 2.35 (s, 6 H each, 3,5-pzeq);
2.25, 2.06 (s, 3 H each, 3,5-pzax); 1.63 (d, JP-H ) 10.7 Hz, 9 H, PMe3);
-11.0 (vbr, 3 H, Ir-H). 31P{1H} NMR, δ (CD2Cl2): -49.2 (s).
[TpRh(PPh3)(H2)H][B(3,5-(CF3)2C5H3)4] (8). A typical sample was
prepared by vacuum transfer of CD2Cl2 into an NMR tube containing
TpRh(PPh3)H2 (2.6 mg, 0.0045 mmol) and [H(Et2O)2][B(3,5-(CF3)2-
C6H3)4] (5 mg, 0.0049 mmol). The sample was warmed to -78 °C
and stored at this temperature until immediately before being transferred
to a precooled NMR probe (< 250 K). 1H NMR, δ (CD2Cl2): 7.80,
6.57 (d, 2 H each, 3,5-pzeq); 7.72 (br, 8 H, o-C6H3(CF3)2); 7.57 (br,
shoulder on p-C6H3(CF3)2 resonance); 7.54 (br, 3 H, p-C6H3(CF3)2);
7.41, 7.09 (m, 6 H each, o- and m-C6H5PPh2); 6.22 (br m, 1 H, 4-pzax);
6.02 (t, 2 H, 4-pzeq); -7.95 (br, 3 H, Rh-H).
TpMe Ir(PMe3)H2 (3). This complex was prepared on a 1.5 mmol
2
scale by a procedure identical to that described for the Tp analog.
Yield: 74%. 1H NMR, δ (CD2Cl2): 5.84 (s, 2 H, 4-pzeq); 5.70 (s, 1
H, 4-pzax); 2.41, 2.26 (s, 6 H each, 3,5-Me2pzeq); 2.26, 2.09 (s, 3 H
each, 3,5-Me2pzax); 1.63 (d, JP-H ) 9.6 Hz, 9 H, PMe3); -22.11 (d,
JP-H ) 26.7 Hz, 2 H, Ir-H). 13C{1H} NMR, δ (CD2Cl2): 150.8, 144.7
(3,5-pzeq); 150.0, 143.5 (3,5-pzax); 106.0 (4-pzeq); 104.7 (4-pzax); 24.2
(d, JP-C ) 38.2 Hz, PMe3); 17.35, 13.0 (3,5-Me2pzeq); 17.75, 12.7 (3,5-
Me2pzax). 31P{1H} NMR, δ (CD2Cl2): -53.0 (s). 1H NMR, δ
(C6D6): 5.76 (s, 2 H, 4-pzeq); 5.47 (s, 1 H 4-pzax); 2.41, 2.10 (s, 3 H
X-ray Structure of [TpIr(PMe3)(H2)H]BF4‚CH2Cl2. Colorless
crystals suitable for X-ray diffraction were obtained by diffusion of
Et2O into a solution of [TpIr(PMe3)(H2)H]BF4 in CH2Cl2. Diffraction
measurements were made on a crystal fragment of dimensions 0.3 ×
0.3 × 0.35 mm in a nitrogen stream at 183 K on an Enraf-Nonius
CAD4 diffractometer using graphite-monochromated Mo KR radiation
(λ ) 0.710 73 Å). An orientation matrix was determined from 24
centered peaks in the range of 28° e 2θ g 34°. Monoclinic symmetry
(space group P21/n) was indicated on the basis of systematic absences.
The cell parameters are a ) 10.560(2) Å, b ) 13.500(3) Å, c ) 15.880-
(3) Å, â ) 92.54(3)°, and V ) 2261.6(11) Å3 (Z ) 4) with a calculated
density of 1.918 g/cm3. There were 3960 unique reflections collected,
with 2θ e 50°; of those reflections, 3196 with I g 4σ(I) were adjudged
observed. Reduction of the data was carried out with XCAD4, and
further work was carried out using the Siemens version of SHELX.
The structure was solved by direct methods and agreed with the
results of a Patterson function; location of the iridium atom reduced
the R factor to 22%. The rest of the atoms were found from difference
maps. The full refinement proceeded to a final R of 4.5% and an Rw
of 5.9%, with a GOF of 1.9. The weighting scheme required a
correction factor of 0.0015. Tables of data collection, solution, and
refinement details, crystal data, atomic coordinates, and anisotropic
thermal parameters are included in the Supporting Information.
each, 3,5-Me2pzax); 2.30, 2.27 (s, 6 H, 3,5-Me2pzeq); 1.41 (d, JP-H
)
9.6 Hz, 9 H, PMe3); -21.21 (d, JP-H ) 26.1 Hz, 2 H, Ir-H). 13C{1H}
NMR, δ (C6D6): 150.5, 144.1 (3,5-pzeq); 149.8, 142.4 (3,5-pzax); 106.0
(4-pzeq); 104.9 (4-pzax); 23.6 (d, JP-C ) 37.3 Hz, PMe3); 17.5, 12.9
(3,5-Me2pzeq); 17.6, 12.6 (3,5-Me2pzax). 31P{Me 1H} NMR, δ (C6D6):
-52.0 (t, JP-H ) 24.7 Hz). IR, cm-1: 2511 (νB-H); 2138 (νIr-H). Anal.
Calcd for C18H33BIrN6P: C, 38.10; H, 5.86; N, 14.81. Found: C, 38.67;
H, 5.94; N, 14.55.
[TpIr(PMe3)(H2)H]BF4 (5). To a 50 mL Schlenk flask containing
a solution of 1 (24 mg, 0.050 mmol) in Et2O (5 mL) at -78 °C was
added HBF4‚Et2O (85%, 10 mL, 0.07 mmol) dropwise. The mixture
was allowed to gradually warm to room temperature as fine white
microcrystals precipitated from solution. The product was filtered off,
1
washed with Et2O (2 mL), and dried under vacuum. Yield: 87%. H
NMR, δ (CD2Cl2): 7.87, 7.85 (d, 2 H each, 3,5-pzeq); 7.69, 7.67 (br s,
1 H each, 3,5-pzax); 6.43 (t, 2 H, 4-pzax); 6.21 (m, 1 H, 4-pzax); 1.78
(d, JP-H ) 11 Hz, 9 H, PMe3); -10.40 (d, JP-H ) 11 Hz, 3 H, Ir-H).
13C{1H} NMR, δ (CD2Cl2): 146.1, 135.8 (3,5-pzax); 144.9, 136.8 (3,5-
pzeq); 107.8 (4-pzeq); 107.3 (4-pzax); 17.9 (d, JP-C ) 42.9 Hz, PMe3).
1
31P{Me H} NMR (CD2Cl2): -42.7 (q, JP-H ) 9.8 Hz). 1H NMR, δ
(CDCl2F): 7.87, 7.79 (d, 2 H each, 3,5-pzeq); 7.68, 7.59 (br s, 1 H
each, 3,5-pzax); 6.40 (t, 2 H, 4-pzeq); 6.13 (m, 1 H, 4-pzax); 1.76 (d,
JP-H ) 11.6 Hz, 9 H, PMe3); -10.42 (d, JP-H ) 10.5 Hz, 3 H, Ir-H).
IR, cm-1: 2499 (νB-H); 2199 (νIr-H). Anal. Calcd for C12H22B2F4-
IrPN6: C, 25.24; H, 3.88; N, 14.72. Found: C, 25.45; H, 4.03; N,
14.40.
Acknowledgment. We thank the National Science Founda-
tion for support of this research. We are grateful for fellowship
support (W.J.O.) from the Chevron Research and Technology
Co, and we thank Dr. David Barnhart for skilled technical
assistance.
[TpIr(PPh3)(H2)H]BF4 (6). Methylene chloride was added dropwise
to a suspension of TpIr(PPh3)H2 in Et2O until a homogeneous solution
was obtained. To this solution was added 1 equiv of HBF4‚Et2O, and
the mixture was then reduced in volume under vacuum and stored at
-30 °C overnight to afford a colorless powder, which was filtered off,
washed with Et2O, and dried under vacuum. This material did not
give a satisfactory elemental analysis. A color change from white to
pale pink was observed when this complex was exposed to vacuum.
Decomposition following loss of H2 may explain the unsatisfactory
analytical data. 1H NMR, δ (CD2Cl2): 7.83, 6.79 (d, 2 H each, 3,5-
pzeq); 7.76, 7.70 (m, 1 H each, 3,5-pzax); 7.58 (m, 3 H, p-C6H6); 7.44,
7.16 (m, 6 H each, o- and m-C6H6); 6.25 (m, 1 H, 4-pzax); 6.09 (t, 2 H,
Supporting Information Available: An ORTEP drawing
and tables giving a structure determination summary, positional
parameters, thermal parameters, bond distances, and bond angles
for [TpIr(PMe3)(H2)H]BF4 (5) (9 pages). Ordering information
and Internet access instructions are given on any current
masthead page.
JA971627T
(61) Brookhart, M.; Grant, B.; Volpe, A. F., Jr. Organometallics 1992,
11, 3920-3922.