Dihydrogen complexes of rhodium: [RhH2(H2) x(PR3)2]+ (R = Cy, iPr; x = 1, 2)

Article abstract of DOI:10.1021/ic0482739

Addition of H₂ (4 atm at 298 K) to [Rh(nbd)(PR₃) ₂][BAr^F₄] [R = Cy, ⁱPr] affords Rh(III) dihydride/dihydrogen complexes. For R = Cy, complex 1a results, which has been shown by low-temperature NMR experiments to be the bis-dihydrogen/bis- hydride complex [Rh(H)₂(η²-H₂) ₂(PCy₃)₂][BAr^F₄]. An X-ray diffraction study on 1a confirmed the {Rh-(PCy₃)₂} core structure, but due to a poor data set, the hydrogen ligands were not located. DFT calculations at the B3LYP/DZVP level support the formulation as a Rh(III) dihydride/dihydrogen complex with cis hydride ligands. For R = ⁱPr, the equivalent species, [Rh(H)₂(η²- H₂)₂(PⁱPr₃)₂][BAr ^F₄] 2a, is formed, along with another complex that was spectroscopically identified as the mono-dihydrogen, bis-hydride solvent complex [Rh(H)₂(η²-H₂)(CD₂Cl ₂)(PⁱPr₃]₂-[BAr^F ₄] 2b. The analogous complex with PCy₃ ligands, [Rh(H)₂(η₂-H₂)(CD₂Cl ₂)(PCy₃)₂][BAr^F₄] 1b, can be observed by reducing the H₂ pressure to 2 atm (at 298 K). Under vacuum, the dihydrogen ligands are lost in these complexes to form the spectroscopically characterized species, tentatively identified as the bis hydrides [Rh(H)₂(L)₂(PR₃)₂]- [BAr^F₄] (1c R = Cy; 2c R = ⁱPr; L = CD ₂Cl₂ or agostic interaction). Exposure of 1c or 2c to a H₂ atmosphere regenerates the dihydrogen/bis-hydride complexes, while adding acetonitrile affords the bis-hydride MeCN adduct complexes [Rh(H) ₂(NCMe)₂(PR₃)₂][BAr^F ₄]. The dihydrogen complexes lose [HPR₃][BAr ^F₄] at or just above ambient temperature, suggested to be by heterolytic splitting of coordinated H₂, to ultimately afford the dicationic cluster compounds of the type [Rh₆(PR₃)6(μ- H)₁₂][BAr^F₄]₂ in moderate yield.

Full text of DOI:10.1021/ic0482739

Inorg. Chem. 2005, 44, 3162−3171
+
Dihydrogen Complexes of Rhodium: [RhH2(H2)x(PR3)2]
i
(R
)
Cy, Pr; x
)
1, 2)
Michael J. Ingleson, Simon K. Brayshaw, Mary F. Mahon, Giuseppe D. Ruggiero, and
Andrew S. Weller*
Department of Chemistry, UniVersity of Bath, Bath, BA2 7AY, United Kingdom
Received December 9, 2004
Addition of H2 (4 atm at 298 K) to [Rh(nbd)(PR3)2][BArF4] [R
For R Cy, complex 1a results, which has been shown by low-temperature NMR experiments to be the bis-
dihydrogen/bis-hydride complex [Rh(H)2( Rh-
2-H2)2(PCy3)2][BArF4]. An X-ray diffraction study on 1a confirmed the
(PCy3)2 core structure, but due to a poor data set, the hydrogen ligands were not located. DFT calculations at the
B3LYP/DZVP level support the formulation as a Rh(III) dihydride/dihydrogen complex with cis hydride ligands. For
iPr, the equivalent species, [Rh(H)2( 2-H2)2(PiPr3)2][BArF4] 2a, is formed, along with another complex that was
spectroscopically identified as the mono-dihydrogen, bis-hydride solvent complex [Rh(H)2(
2-H2)(CD2Cl2)(PiPr3)2]-
[BArF4] 2b. The analogous complex with PCy3 ligands, [Rh(H)2( 2-H2)(CD2Cl2)(PCy3)2][BArF4] 1b, can be observed
)
Cy, iPr] affords Rh(III) dihydride/dihydrogen complexes.
)
η
{
}
R
)
η
η
η
by reducing the H2 pressure to 2 atm (at 298 K). Under vacuum, the dihydrogen ligands are lost in these complexes
to form the spectroscopically characterized species, tentatively identified as the bis hydrides [Rh(H)2(L)2(PR3)2]-
[BArF4] (1c R
) Cy; 2c R ) ) CD2Cl2 or agostic interaction). Exposure of 1c or 2c to a H2 atmosphere
iPr; L
regenerates the dihydrogen/bis-hydride complexes, while adding acetonitrile affords the bis-hydride MeCN adduct
complexes [Rh(H)2(NCMe)2(PR3)2][BArF4]. The dihydrogen complexes lose [HPR3][BArF4] at or just above ambient
temperature, suggested to be by heterolytic splitting of coordinated H2, to ultimately afford the dicationic cluster
compounds of the type [Rh6(PR3)6(µ ] in moderate yield.
-H)12][BArF
4 2
Introduction
[RhTp(H)(η2-H2)(PPh3)]+,6 RhTp(H)2(η2-H2),7 Rh(η2-H2)-
(PCP),8 [Rh(H)2(η2-H2)(PPP)]+,9 and [RhCp*(η2-H2)-
(dmpm)]2+ 10 [Cp* ) η5-C5Me5, Tp ) HB(pz)3 or a
derivative thereof, PCP ) η3-C6H3-1,3-(PtBu2)2, PPP ) MeC-
(CH2PPh2)3]. No simple rhodium dihydrogen complexes of
the general formula Rh(η2-H2)x(PR3)2 have been reported,
which would be of significant interest on the basis of the
role that they play in homogeneous catalysis, particulary as
several hydrogenation reactions have been postulated to
proceed via dihydrogen intermediates.11 Neutral Rh(H)x(PR3)y
The transition metal chemistry of dihydrogen continues
to be a significant area within coordination chemistry even
though the first recognized example of such a complex,
W(η2-H2)(CO)3(PiPr3)2, was reported by Kubas over 20 years
ago.1 Dihydrogen complexes, and the larger group of
σ-complexes to which they belong, are fascinating in terms
of fundamental structure and bonding as well as being central
to the study of homogeneous catalysis.2 Of the later transition
metals, examples of dihydrogen acting as a ligand are known
across metals of groups 8-10.2-4 Rather surprisingly, given
the important role that rhodium hydrides play in homoge-
neous catalysis, relatively few rhodium dihydrogen com-
plexes have been documented: [RhCp*(H)(η2-H2)(PMe3)]+,5
(5) Taw, F. L.; Mellows, H.; White, P. S.; Hollander, F. J.; Bergman, R.
G.; Brookhart, M.; Heinekey, D. M. J. Am. Chem. Soc. 2002, 124,
5100.
(6) Oldham, W. J.; Hinkle, A. S.; Heinekey, D. M. J. Am. Chem. Soc.
1997, 119, 11028.
* Author to whom correspondence should be addressed. Tel: +44(0)-
1225 383394. Fax: +44(0)1225 386231. E-mail: a.s.weller@bath.ac.uk.
(1) Kubas, G. J.; Ryan, R. R.; Swanson, B. I.; Vergamini, P. J.;
Wasserman, H. J. J. Am. Chem. Soc. 1984, 106, 451.
(7) Bucher, U. E.; Lengweiler, T.; Nanz, D.; Vonphilipsborn, W.; Venanzi,
L. M. Angew. Chem., Int. Ed. Engl. 1990, 29, 548.
(8) Xu, W. W.; Rosini, G. P.; Gupta, M.; Jensen, C. M.; Kaska, W. C.;
Jespersen, K.; Goldman, A. S. Chem. Commun. 1997, 2273.
(9) Bakhmutov, V. I.; Bianchini, C.; Peruzzini, M.; Vizza, F.; Vorontsov,
E. V. Inorg. Chem. 2000, 39, 1655.
(2) Kubas, G. J. Metal dihydrogen and σ-bond complexes; Kluwer: New
York, 2001.
(3) McGrady, G. S.; Guilera, G. Chem. Soc. ReV. 2003, 32, 383.
(4) Heinekey, D. M.; Oldham, W. J. Chem. ReV. 1993, 93, 913.
(10) Gelabert, R.; Moreno, M.; Lluch, J. M.; Lledos, A.; Pons, V.; Heinekey,
D. M. J. Am. Chem. Soc. 2004, 126, 8813.
3162 Inorganic Chemistry, Vol. 44, No. 9, 2005
10.1021/ic0482739 CCC: $30.25
© 2005 American Chemical Society
Published on Web 04/02/2005

Dihydrogen Complexes of Rhodium
a
Scheme 1
a Y ) [B{C6H3-3,5-(CF3)2}4]- or [1-H-closo-CB11Me11]-.
Scheme 2
(y ) 2, x ) 3; y ) 3, x ) 1) complexes were investigated
by Ibers and Otsuka in the 1970s and 1980s,12,13 although
some of these latter compounds were subsequently refor-
mulated as a dimeric species.14 This gap in the chemistry of
dihydrogen ligands is further highlighted by the fact that the
closely related iridium and ruthenium complexes [Ir(H)2(η2-
t
H2)2(PR3)2]+ (R ) Cy3, Bu2Ph)15-17 and Ru(H)2(η2-H2)2-
compounds reported here, the anions [BArF ]- and [1-H-
4
i
(PR3)2 (R ) Cy, Pr)18-20 have been known for almost 20
closo-CB11Me11]- are interchangeable, and consequently,
years and serve as the paradigm for compounds that contain
both hydride and dihydrogen ligands.
only results for [BArF ]- are discussed further. The use of
4
other anions such as [BF4]- and [PF6]- led to decomposition
and intractable mixtures. At room temperature or just above,
under a H2 atmosphere, compound 1a slowly loses [HPCy3]-
We have recently communicated that hydrogenation of
[Rh(nbd)(PiPr3)2][Y] {nbd ) norbornadiene, Y ) [B{C6H3-
3,5-(CF3)2}4]-(BArF ), [1-H-closo-CB11Me11]-} ultimately
4
[BArF ] to afford the cluster species [Rh6(PCy3)6(µ-H)12]-
4
results in the formation of a novel, high hydride content
cluster: [Rh6(PiPr3)6(µ-H)12][Y]2 (Scheme 1), along with
other (as yet unidentified) products.21 This reaction proceeds
via intermediates that we initially identified spectroscopically
as the hydride/bis-dihydrogen complexes [Rh(H)2(η2-H2)x-
(PiPr3)2][Y] (x ) 1 or 2). In this Article, we describe in more
detail the synthesis and spectroscopic characterization of
these intermediate species and their related cyclohexyl
phosphine congeners.
[BArF ] , which is analogous to the Pr3 derivative.21 These
i
4 2
cluster compounds will be reported in more detail in a future
contribution. The formulation of 1 was determined by
solution NMR studies and a preliminary X-ray structure. The
formation of discrete Rh(III) monomers is in contrast to the
analogous complexes with PPh3 ligands, which have been
shown to form dimers such as [Rh(η6-C6H5PPh2)(PPh3)]2-
[Y]2 {Y ) weakly coordinating anion}, in which the
unsaturation at the Rh(I) center is relieved by the coordina-
tion of an aryl group from a PPh3 ligand.22-24
Results and Discussion
1
In the H NMR spectrum of 1a at 298 K under 4 atm of
[Rh(H)2(η2-H2)2(PCy3)2][BArF ]. The addition of H2 (ca.
4
H2, no hydride signals are observed. Free H2 [δ 4.6] is also
not observed, consistent with a fast exchange between free
and bound H2. A single rather broad 31P environment is
observed in the 31P{1H} NMR spectrum [δ - 54.3, J(RhP)
110 Hz]. On progressive cooling to 190 K, the 31P{1H} NMR
spectrum is still a doublet, but at this temperature, it is sharper
[δ 60.1, J(RhP) 92 Hz]. The corresponding 1H NMR
spectrum at 190 K reveals two signals, a broader one at δ
-1.81, which integrates to 4-H relative to both the cyclohexyl
4 atm at 298 K) to a CH2Cl2 solution of the well-separated
ion pair Rh(I) complex [Rh(nbd)(PCy3)2][Y] results in the
reductive hydrogenation of the diene and the formation, in
quantitative yield by NMR spectroscopy, of the new complex
[Rh(H)2(η2-H2)2(PCy3)2][Y] 1a {Y ) [BArF ]-, [1-H-closo-
4
CB11Me11]-} (Scheme 2). For 1a, and all subsequent
(11) Esteruelas, M. A.; Ore, L. A. Chem. ReV. 1998, 98, 577.
(12) Yoshida, T.; Okano, T.; Thorn, D. L.; Tulip, T. H.; Otsuka, S.; Ibers,
J. A. J. Organomet. Chem. 1979, 181, 183.
(13) Yoshida, T.; Thorn, D. L.; Okano, T.; Ibers, J. A.; Otsuka, S. J. Am.
Chem. Soc. 1979, 101, 4212.
(14) Catalytic Aspects of Metal Phosphine Complexes; Thorn, D. L.; Ibers,
J. A., Eds.; American Chemical Society: Washington, DC, 1982; Vol.
196.
(15) Crabtree, R. H.; Lavin, M. J. Chem. Soc., Chem. Commun. 1985, 1661.
(16) Cooper, A. C.; Eisenstein, O.; Caulton, K. G. New J. Chem. 1998,
22, 307.
phosphine and [BArF ]- protons, and a sharper one at δ
4
-14.14, integrating to 2-H (Figure 1). 31P or 103Rh coupling
is not resolved in either of the signals. T1 measurements on
these signals at 190 K show that the integral 4-H peak has
a much shorter relaxation time than the higher field resonance
(14 and 216 ms, respectively). These values unambiguously
place the signals as arising from dihydrogen and hydride
ligands, respectively.17 The relaxation time for the dihydrogen
(17) Crabtree, R. H.; Lavin, M.; Bonneviot, L. J. Am. Chem. Soc. 1986,
108, 4032.
(18) Sabo-Etienne, S.; Chaudret, B. Coord. Chem. ReV. 1998, 180, 381.
(19) Abdur-Rashid, K.; Gusev, D. G.; Lough, A. J.; Morris, R. H.
Organometallics 2000, 19, 1652.
(20) Chaudret, B.; Devillers, J.; Poilblanc, R. Organometallics 1985, 4,
1727.
(21) Ingleson, M. J.; Mahon, M. F.; Raithby, P. R.; Weller, A. S. J. Am.
Chem. Soc. 2004, 126, 4784.
(22) Rifat, A.; Patmore, N. J.; Mahon, M. F.; Weller, A. S. Organometallics
2002, 21, 2856.
(23) Rifat, A. Exo-closo carborane compounds of rhodium and iridium.
Ph.D. Thesis, Univeristy of Bath, United Kingdom, 2003.
(24) Marcazzan, P.; Ezhova, M. B.; Patrick, B. O.; James, B. R. Comptes
Rendus Chimie 2002, 5, 373.
Inorganic Chemistry, Vol. 44, No. 9, 2005 3163

Ingleson et al.
be rotating around the M-H2 axis at low temperatures as
only one η2-H2 environment is observed; this being in accord
with findings for other dihydrogen complexes in which the
barrier to rotation is very small.2,26,27
[Rh(H)2(η2-H2)2(PiPr3)2]+ and [Rh(H)2(η2-H2)(CD2Cl2)-
(PiPr3)2]+. The addition of H2 (ca. 4 atm at room tempera-
ture) to a CH2Cl2 solution of [Rh(nbd)(PiPr3)2][BArF ] results
4
in a mixture of the bis-dihydrogen complex [Rh(H)2(η2-
H2)2(PiPr3)2][BArF ] 2a and the mono-dihydrogen complex
Figure 1. High-field region of the 1H NMR spectrum of 1a showing the
relative integrals at 190 K.
4
[Rh(H)2(η2-H2)(CD2Cl2)(PiPr3)2][BArF ] 2b (Scheme 3). These
4
two complexes have been characterized by low-temperature
Chart 1
NMR spectroscopy.
The 1H NMR and 31P NMR spectra of this mixture under
a H2 atmosphere (4 atm at 298 K) are temperature dependent.
1
In the room temperature H NMR spectrum, as well as
signals due to phosphine and the [BArF ]- anion, a very broad
4
peak is observed in the hydride region centered at δ - 8.62
ppm that integrates to 3.6 protons relative to the [BArF ]-
4
ligand was still decreasing at the lowest accessible temper-
ature (190 K); and thus T1(minimum) was not unambiguously
determined. The chemical shift and line widths of these
signals are similar to those reported at low temperatures by
Crabtree and by Caulton for the dihydrogen/bis-dihydrogen
anion signals. The line-width of this signal suggests that an
exchange process is occurring. No signal due to free H2 is
observed at ca. δ 4.6, consistent with exchange between free
and bound H2. In the 31P{1H} NMR spectrum, a single
doublet is observed at δ 60.4 [J(RhP) 107]. On progressive
t
complexes [Ir(H)2(η2-H2)2(PR3)2]+ (R ) Cy, Bu2Ph).15-17
1
cooling, the broad hydride signal in the H NMR spectrum
The observation of H/D coupling in the H-D isotopomer
of a dihydrogen complex is direct proof of the existence of
a coordinated H2 ligand.2 Exposure of a CD2Cl2 solution of
1a to D2 gas resulted in a decrease in the overall intensity
of the hydride signals, indicating H/D exchange, and a
broadening of the signal at δ -1.81, although unfortunately
no HD coupling was observed. However, as will be shown
later, under similar conditions, the PiPr3 derivative does show
H-D coupling. Furthermore, DFT calculations (vide infra)
on a model complex [Rh(H)6(PMe3)2]+ indicate a dihydride/
bis-dihydrogen complex as a stable isomer. This, along with
chemical shifts, relative integrals, and T1 measurements
allows the confident assignment of 1a as the bis-dihydrogen/
first disappears into the baseline (240 K) eventually to be
replaced at low temperature (200 K) by a more complex set
of signals of varying line widths (Figure 2). At the same
time, in the 31P{1H} NMR spectrum, two major signals are
observed at δ 68.4 and δ 62.1 in a 2:1 ratio, both showing
coupling to rhodium. This ratio changes to 4:1 on cooling
to 190 K, showing that these two complexes are in dynamic
equilibrium. On the basis of chemical shifts, relative integrals,
T1 measurements, and H/D exchange experiments, these two
sets of peaks are assigned to the bis-dihydrogen/bis-hydride
complex [Rh(H)2(η2-H2)2(PiPr3)2][BArF ] 2a and the mono-
4
dihydrogen/bis-hydride solvento complex [Rh(H)2(η2-H2)(CD2-
Cl2)(PiPr3)2][BArF ] 2b.
bis-hydride complex [Rh(H)2(η2-H2)2(PCy3)2][BArF ].
4
4
In the hydride region of the 1H NMR spectrum at 200 K,
two resonances at δ -1.90 and δ -14.23, in the ratio of
4:2, are assigned to complex 2a. The measured T1 values
(at 190 K) for these two signals in 2a provide evidence for
formulation as a bis-dihydrogen/dihydride complex, showing
the characteristic short T1 times for the dihydrogen ligand
as compared with the hydride: δ -1.90 (11 ms) and δ
-14.23 (202 ms). As for 1a, the relaxation time for the
dihydrogen ligand was still decreasing at the lowest acces-
sible temperature (190 K); and thus T1(minimum) was not
unambiguously determined. These chemical shifts are also
similar to those observed for 1a as well as [Ir(H)2(η2-H2)2-
(PCy3)2]+.15,17 Further evidence for the assignment of 2a as
a bis-dihydrogen complex is the observation of a broad 1:1:1
triplet for the higher field signal δ -1.90 when a sample is
degassed and then placed under 4 atm of D2 [J(DH)average
∼30 Hz] (Figure 3). This multiplet is due to a mixture of
The formulation of 1a as the bis-dihydrogen complex [Rh-
(H)2(η2-H2)2(PCy3)2][BArF ] places it as being directly
4
analogous to the complexes first reported by Crabtree, [Ir-
(H)2(η2-H2)2(PCy3)2]+;15,17 Chaudret, Ru(H)2(η2-H2)2(PCy3)2;18,20
(Chart 1) and Shaw [Re(H)8-2x)(η2-H2)x(PCy3)2]+.25 T1 mea-
surements on the dihydrogen and dihydride ligands in the
iridium complex show short relaxation times for both (48
and 73 ms, respectively, at 193 K), unlike in 1a, suggesting
that the exchange is still occurring between these ligands;
while for Ru(H)2(η2-H2)2(PR3)2 this exchange process is still
extremely facile at the lowest attained temperatures, with
only one time averaged signal observed for hydride and
dihydride ligands. For 1a, this exchange must be slower,
leading to the observed order of magnitude difference in the
T1 times. This observation also contrasts with other rhodium
hydride/dihydrogen complexes that are highly fluxional, with
hydride and dihydrogen ligands exchanging at very low
temperatures.5,9 The dihydrogen ligands in 1a are likely to
(26) Li, S.; Hall, M. B.; Eckert, J.; Jensen, C. M.; Albinati, A. J. Am. Chem.
Soc. 2000, 122, 2903.
(27) Rodriguez, V.; Sabo-Etienne, S.; Chaudret, B.; Thoburn, J.; Ulrich,
S.; Limbach, H.-H.; Barthelat, J. C.; Hussein, K.; Marsden, C. J. Inorg.
Chem. 1998, 37, 3475.
(25) Fontaine, X. L. R.; Fowles, E. H.; Shaw, B. L. J. Chem. Soc., Chem.
Commun 1988, 482.
3164 Inorganic Chemistry, Vol. 44, No. 9, 2005

Dihydrogen Complexes of Rhodium
a
Scheme 3
-
a [BArF
]
4
anions not shown.
the partially deuterated isotopomers of 2a: [Rh(H)6-x(D)x-
dihydrogen ligand (9 ms) and the other two as hydrides (149
and 144 ms, respectively). Unfortunately, HD coupling was
not resolved in the partially deuterated sample, although the
signal at δ -0.28 does broaden significantly on the addition
of D2. The two inequivalent hydride resonances observed
for 2b suggest a structure similar to that reported for the
related five-coordinate complex [Ir(H)2(η2-H2)(PtBu2Ph)2]-
(PiPr3)2][BArF ]. We are reluctant to use this value to
4
[BArF ], which shows signals at δ -0.02 (η2-H2), δ -10.0
4
(H), and δ -41.2 (H); the latter signal being assigned to a
hydride trans to an empty coordination site.16 Related cationic
iridium complexes with hydride ligands trans to vacant sites
also show similar, large, upfield chemical shifts (e.g., [Ir-
Figure 2. 1H NMR spectrum (high field region, 200 K) of complexes 2a
and 2b (asterisk marks part of resonances due to iPr3P ligands, and † marks
an unidentified compound).
(H)2(PPh(tBu)2)2][BArF ] δ -37.129 (trans to agostic interac-
estimate28 the H-H distance in the complex as the signal is
so broad. This signal also undergoes a small upfield IPR
shift of 0.044 ppm. Similar upfield shifts have been reported
4
tions) and five-coordinate IrH2I(PtBu2Ph)2 δ -44.4).30 In
contrast, the highest field chemical shift observed for 2b is
δ -22.4, which suggests a different structure, possibly one
in which a ligand occupies the site trans to the hydride.30
On the basis of the chemical shifts of the hydrogen ligands
and observations made on changing the solvent from CD2-
Cl2 (vide infra), we tentatively suggest that the CD2Cl2
solvent occupies this vacant site rather than an agostic
interaction. Attempts to observe the coordinated CH2Cl2
spectroscopically by 13C{1H} NMR at low temperature were
not successful, although bound CH2Cl2 is often difficult to
observe due to rapid exchange with the bulk solvent.31
Examples of dichloromethane complexed with electrophilic
transition metals are now well-established, and both mono-
dentate31,32 and bidentate33,34 binding modes have been
observed. A Rh(III) complex with coordinated CH2Cl2 has
recently been reported and crystallographically characterized,
previously in other dihydrogen complexes RuH2(η2-H2)2-
19
(PiPr3)2 and [RhCp*H(η2-H2)(PMe3)][BArF ].5 For these
4
latter species, and closely related complexes,10 well-resolved
coupling patterns are observed for the various H/D isoto-
pomers. This is in contrast to the broad signal observed for
2a. Under a D2 atmosphere, no H/D exchange was observed
for the protons on the phosphine ligands, similar to that
observed for [Ir(H)2(η2-H2)2(PtBu2Ph)2]+,29 but contrasts with
[Ru(H)2(η2-H2)2(PiPr3)2]+ 19 in which deuterium is incorpo-
rated into the phosphine.
[RhCp*(PMe3)(Me)(CH2Cl2)][BArF ],5 and in this case,
4
coordinated CH2Cl2 is observed in the 13C{1H} NMR
spectrum.
Attempts to use solvents other than CD2Cl2 led to mixed
results. The noncoordinating solvent fluorobenzene freezes
at 231 K, precluding a low-temperature NMR analysis of
any hydride/bis-dihydrogen complexes formed in this solvent.
In a d8-toluene solution, hydrogenation of [Rh(nbd)(PiPr3)2]-
Figure 3. Partial 1H NMR spectrum of the (η2-H2)2 region of 2a (b)
overlaid with that of a mixture of 2a and its partially deuterated isotopomers
(a). Both were measured at 200 K.
The other major species in solution at 200 K is formulated
[BArF ] results in a broad signal at room temperature in the
4
as the solvento complex [Rh(H)2(η2-H2)(CD2Cl2)(PiPr3)2]-
1H NMR spectrum at ∼δ -8, similar to that seen in CD2-
Cl2, but at lower temperatures, the cationic products formed
are not soluble. However, the solvent meta-fluorotoluene
[BArF ] 2b. In addition to the resonance arising from 2a, a
4
doublet is observed in the 31P{1H} NMR spectrum at δ 62.1
that is assigned to 2b. In the 1H NMR spectrum, three high-
field signals observed at δ -0.28, -12.66, and -22.42 in
the ratio 2:1:1 are assigned to 2b. T1 measurements (at 190
K) on these three signals assign the lowest field as a
(30) Hauger, B. E.; Gusev, D.; Caulton, K. G. J. Am. Chem. Soc. 1994,
116, 208.
(31) Butts, M. D.; Scott, B. L.; Kubas, G. J. J. Am. Chem. Soc. 1996, 118,
11831.
(32) Huhmann-Vincent, J.; Scott, B. L.; Kubas, G. J. Inorg. Chem. 1999,
38, 115.
(33) Huang, D. J.; Bollinger, J. C.; Streib, W. E.; Folting, K.; Young, V.;
Eisenstein, O.; Caulton, K. G. Organometallics 2000, 19, 2281.
(34) Seggen, D. M. V.; Hurlburt, P. K.; Anderson, O. P.; Strauss, S. H.
Inorg. Chem. 1995, 34, 3453.
(28) Maltby, P. A.; Schalf, 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.
(29) Cooper, A. C.; Streib, W. E.; Eisenstein, O.; Caulton, K. G. J. Am.
Chem. Soc. 1997, 119, 9069.
Inorganic Chemistry, Vol. 44, No. 9, 2005 3165

Ingleson et al.
a
Scheme 4
Chart 2
-
a [BArF
] anions not shown.
4
(3-F-C6H4Me) proved to be a suitable solvent that dissolved
the cationic species, was noncoordinating and had a workable
temperature range for NMR spectroscopy (melting point -87
°C). Although d7-meta-fluorotoluene is not available, we
successfully used a per-protio solvent for NMR spectroscopy
(by shimming on the FID), while some of the aromatic
at 200 K ) 245, 252, and 7 ms, respectively). These signals
appear in a 1:1:2 ratio, respectively. We assign these signals
to [Rh(H)2(η2-H2)(d8-THF)(PiPr3)2][BArF ], 3b, on the basis
4
of their T1 measurements and the similarity with complex
2b (Chart 2). The 31P{1H} NMR shows two doublets at δ
55.2 [J(RhP) 114 Hz] and δ 64.1 [J(RhP) 103 Hz], in the
ratio of 2:1, respectively. Placing this solution under a partial
vacuum to remove H2 afforded NMR spectra that only
signals due to the [BArF ]- anion are not obscured by the
4
solvent allowing for the relative integration of hydride
signals. Treatment of [Rh(nbd)(PiPr3)2][BArF ] with H2 (4
4
showed the presence of [Rh(H)2(d8-THF)2(PiPr3)2][BArF ].
4
atm at 298 K) in meta-fluorotoluene resulted in an 1H NMR
spectrum at 298 K that showed a broad high field resonance
at δ -5.81 of a relative integral 4.9 and no free H2 signal,
consistent with rapid exchange between bound and free H2.
Cooling to 210 K partially (vide infra) froze this exchange
out, and only two broad high field resonances are observed,
at δ -3.24 and δ -15.08 in the ratio of 4:2, very similar to
that observed for 1a and 2a. The 31P{1H} NMR spectrum at
210 K displayed a single sharp doublet at δ 67.48 [J(RhP)
94 Hz], similar to 2a. T1 measurements at 210 K indicated
that the δ -3.24 resonance can be assigned to bound
dihydrogen (20 ms) and that the δ -15.08 resonance is best
assigned to hydride ligands (76 ms). That the signal due to
the hydride ligand still shows a relatively short T1 time
indicates that there is still some exchange between hydride
and dihydrogen ligands at this temperature in this solvent.
Similar exchange is observed for [Ir(H)2(η2-H2)2(PCy3)2]+ 15,17
and Ru(H)2(η2-H2)2(PCy3)2.18,20 Cooling further to 190 K
results in significant broadening into the baseline of the signal
at δ -3.24 but not the signal at δ -15.08. This is different
behavior to that observed in CD2Cl2 in which relatively sharp
lines are observed at 190 K. Nevertheless, in the absence of
CD2Cl2 solvent at low temperature, that only signals due to
a bis-dihydrogen/bis-hydride complex (i.e., 2a, Scheme 4)
are observed, strongly suggests that 2b is best assigned as a
CD2Cl2 complex.
The formation of both mono- (2a) and bis-dihydrogen (2b)
complexes in CD2Cl2 contrasts to 1a, which is formed as
the sole product. Variation in stability with respect to H2
loss has previously been noted for the neutral Ru(H)2(η2-
H2)x(PR3) systems. With tricyclohexyl phosphine, the stable
ruthenium bis-dihydrogen adduct is formed (cf. 1a), whereas
with isopropyl phosphine, H2 loss is facile, and dimers such
as Ru2H6(PiPr3)4 result.18,19
Pressure Dependence. All the dihydrogen complexes so
far discussed are sensitive to H2 pressure (Scheme 5). For
the cyclohexyl phosphine complexes, reducing the initial H2
pressure to ∼2 atm at 298 K results in the appearance of a
mono dihydrogen complex that is formulated as [Rh(H)2-
(η2-H2)(CD2Cl2)(PCy3)2][BArF ] 1b on the basis of its very
4
similar 1H NMR spectra (at 190 K) in the hydride region as
compared with 2b, namely, a broad signal at δ -0.11 (T1
29 ms, η2-H2) and sharper peaks at δ -12.3 (T1 137 ms)
and δ -22.2 (T1 259 ms) for the now inequivalent hydride
ligands. The 31P{1H} NMR spectrum at this pressure also
indicates that two compounds are present, 1a and 1b, the
latter giving a signal at δ 54.3 [J(RhP) 102 Hz]. The ratio
of 1a/1b at 2 atm H2 is ∼1:1. For the isopropyl derivatives
under the same conditions (2 atm), the previously identified
bis-dihydrogen 2a and mono dihydrogen complexes 2b are
observed but with the latter compound by far the major
species in solution.
Use of d8-THF in place of CD2Cl2 as a solvent in the initial
hydrogenation of [Rh(nbd)(PiPr3)2][BArF ] led to a low
At lower H2 pressures (1 atm at 298 K), one compound
predominates for both 1 and 2 that may be more cleanly
produced by removing H2 in vacuo. These products have
been spectroscopically characterized in CD2Cl2 solution and
4
temperature (200 K) NMR spectrum that showed the
presence of two compounds in a 2:1 ratio, neither of which
was 2a or 2b. The major one showed a broad hydride peak
at δ -24.33 [T1 266 ms at 200 K] in the 1H NMR spectrum,
which we assign as the bis-THF/bis-hydride complex [Rh-
(H)2(d8-THF)2(PiPr3)2][BArf4], 3a, this being similar to [Rh-
(H)2(EtOH)2(PPh3)2][ClO4] that was first described by Os-
borne and Schrock.35 The minor component showed two
hydride resonances at δ -13.08 and δ -24.89, both as
complex multiplets (part of a ABM2X system), and a broad
signal at δ -0.10 characteristic of a dihydrogen ligand (T1
are tentatively identified as [Rh(H)2(L)2(PR3)2][BArF ] (1c
4
i
R ) Cy; 2c R ) Pr; L ) CD2Cl2 or agostic interaction).
For both complexes, a single relatively sharp signal is
observed in the region associated with terminal hydrides:
1c δ -23.0 and 2c δ -24.4. No signal due to coordinated
dihydrogen was observed. 103Rh coupling is resolved in both,
and for 2c coupling to two equivalent phosphines are also
observed. The 31P{1H} NMR spectra for each is a single
doublet showing coupling to rhodium [1c δ 53.9 J(RhP) 111
Hz; 2c δ 57.6 J(RhP) 104 Hz]. We tentatively assign these
(35) Schrock, R. R.; Osborn, J. A. J. Am. Chem. Soc. 1976, 98, 2134.
3166 Inorganic Chemistry, Vol. 44, No. 9, 2005

Dihydrogen Complexes of Rhodium
Scheme 5
Scheme 6
atmosphere and storage at 10 °C. The crystals were poor
diffractors, and the resulting data set was not of high quality,
allowing only the gross structural features of the molecule
to be observed. The asymmetric unit (Figure 4) consists of
one cation and one [1-H-closo-CB11Me11]- anion. The
cationic portion is a linear {Rh(PCy3)2}+ fragment, and no
other significant electron density was located near the Rh
center. Given that the metal fragment would have a nominal
12-electron count, this suggests the presence of hydrogen
ligands or agostic interactions;29 however, given the poor
quality of the data, the location of the hydride ligands around
the metal was precluded, leaving the actual number inde-
terminate. Nevertheless, the structure is consistent with a [Rh-
(H)2(η2-H2)2(PCy3)2]+ fragment, showing a trans disposition
of phosphine ligands [P1-Rh-P1′′ ) 180°]. The structure
is grossly similar to the structures recently reported for Ru-
(H)2(η2-H2)2(PCy3)2 39 and Ru(H)2(η2-H2)2(PiPr3)2,19 both of
which also have a P-M-P bond angle of ∼180° in the solid
state. DFT calculations (vide infra) show that a trans
arrangement of dihydrogen and hydride ligands in [Rh(H)2-
(η2-H2)2(PCy3)2]+ would be expected to give a complex with
a P-Rh-P angle of 180°, while cis arrangements have a
slightly bent P-Rh-P angle. Interestingly, in these gas-phase
calculations, the trans isomer is found to be 27 kcal mol-1
higher in energy as compared to the possible cis isomers,
but given the poor quality of the data, discussion of the
structural metrics is clearly inappropriate apart from confirm-
ing the gross overall structure.
DFT Calculations on [Rh(H)6(PMe3)2]+. We have briefly
investigated the structure of the complex [Rh(H)6(PMe3)2]+,
which is a model for complexes 1a and 2a, by the DFT
method (B3LYP/DZVP). Very similar results were also
obtained using the B3LYP/LANL2DZ level of theory, which
has been successfully used previously to model transition
metal dihydrogen complexes.16,26,27 Optimization of [Rh(H)2-
(η2-H2)2(PMe3)2]+ afforded a stable ground-state structure
with trans phosphines, two hydrides, and two dihydrogen
ligands, confirming the spectroscopic assignment as a Rh-
(III) bis-dihydrogen dihydride. Four possible isomers all
having trans phosphines (A-D) were identified, which are
very similar to those described for RuH2(H2)2(PH3)2.27
Structures A-C have cis hydrides (Figure 5) and are related
to one another by a simple rotation of one or both of the
dihydrogen ligands. There is only a negligible energy
difference between these (less than 1 kcal mol-1). Structure
D has trans hydrides and is 27 kcal mol-1 higher in energy.
The P-Rh-P angles in isomers A-C lie between 164 and
to Rh(III) complexes with either a chelating coordinated
dichloromethane or as bis-agostic complexes with no coor-
dinated dichloromethane. The chemical shift of the hydride
ligands in an agostic species is expected to be nearer δ -40
{viz. [Ir(H)2(PPh(tBu)2)2][BArF ] δ -37.1},29 while the
4
hydride signals in 1c and 2c are more like the dihydride
complexes Rh(H)2{O2S(O)Me}(PiPr3)2
36
[δ -25.3] and
[Ir(H)2(o-Cl2C6H4)(PPh3)2][BF4]37 [δ -20.8], both of which
contain weakly bound ligands trans to the hydrides. This
suggests that the formulation of 1c and 2c as weakly bound
dichloromethane complexes is more appropriate. Whatever
the nature of the weakly bound ligand, repressurization of
1c or 2c with H2 (4 atm) immediately affords the previously
observed mixtures of dihydrogen complexes at 200 K, while
addition of the coordinating solvent acetonitrile to 1c or 2c
resulted in the immediate formation of the bis-acetonitrile
adducts [Rh(H)2(NCMe)2(PR3)2][BArF ] (Scheme 6). Both
these observations are consistent with a weakly bound solvent
or interaction. The acetonitrile adducts are direct congeners
of [Ir(H)2(NCMe)2(PR3)2][PF6]38 first reported by Crabtree
4
1
and are identified by a sharp doublet of triplets in the H
NMR spectrum for the equivalent hydrides [e.g., δ -19.1,
J(RhH) 18, J(PH) 8 Hz] and a single, sharp doublet in the
31P{1H} NMR spectrum [e.g., δ 59.8 J(RhP) 107 Hz].
Attempts to isolate 1c and 2c led to decomposition, leading
to dimeric compounds with bridging chlorides that presum-
ably arise from the coordinated CH2Cl2.
Solid-State Structure of [Rh(H)2(η2-H2)x(PCy3)2]+. At-
tempts to secure the unambiguous characterization of the
dihydrogen complexes 1 and 2 in the solid state by X-ray
crystallography were frustrated by the mixtures of products
formed, the dependence on H2 pressure, and the gradual
formation of [Rh6(PR3)6(µ-H)12][BArF ] and [HPR3][BArF ]
4 2
4
among other products. After repeated attempts, a small
number of crystals were grown from the reaction of [Rh-
(nbd)(PCy3)2][1-H-closo-CB11Me11] with dihydrogen in fluo-
robenzene solution by layering with pentane under a H2
(36) Werner, H.; Bosch, M.; Schneider, M. E.; Hahn, C.; Kukla, F.; Manger,
M.; Windmu¨ller, B.; Weberndo¨rfer, B.; Laubender, M. Dalton Trans.
1998, 3549.
(37) Crabtree, R. H.; Faller, J. W.; Mellea, M. F.; Quirk, J. M. Organo-
metallics 1982, 1, 1361.
(38) Crabtree, R.; Demou, P. C.; Eden, D.; Mihelcic, J. M.; Parnell, C. A.;
Quirk, J. M.; Morris, G. E. J. Am. Chem. Soc. 1982, 104, 6994.
(39) Borowski, A. F.; Donnadieu, B.; Daran, J. C.; Sabo-Etienne, S.;
Chaudret, B. Chem. Commun. 2000, 543.
Inorganic Chemistry, Vol. 44, No. 9, 2005 3167

Ingleson et al.
hydrides. Optimization of a tetrahydride, Rh(V), structure
[Rh(H)4(η2-H2)(PMe3)2]+ resulted in the Rh(III) dihydride/
bis-dihydrogen structure.
Heterolytic Cleavage of Coordinated H2. Under a H2
atmosphere at, or just above, room temperature, the complex
mixtures of 2 lose protonated phosphine, [HPiPr3][BArF ],
4
over a period of days with the concomitant formation of the
cluster dications [Rh6(PiPr3)6(µ-H)12][BArF ] in moderate
4 2
yield (20%).21 A similar cluster and the formation of [HPCy3]-
[BArF ] also ultimately results from mixtures of 1. Other
4
rhodium containing products are also formed in these
reactions that currently have not been fully characterized.
Although the initial organometallic product of this process
is not yet known (we assume it is a neutral low coordinate
{RhP(H)x}n speciessScheme 7), the observation of proto-
nated phosphine means that it could be regarded as arising
from heterolytic activation of a dihydrogen ligand. Confir-
mation that the proton arises from H2 comes from replacing
Figure 4. Solid-state structure of the cationic portion of [Rh(H)2(η2-H2)x-
(PCy3)2][1-H-closo-CB11Me11]. Ellipsoids shown at the 30% probability
level. Only one of the disordered cyclohexyl rings is shown (see
Experimental Procedures for details). Hydrogen atoms are not shown.
Equivalent (primed) atoms were generated by the operations x, y, -z + 3/2
and -x, y, z. Selected bond lengths (Å) and angles (deg): Rh-P1 2.352(3)
and P1-Rh-P1′′ 180.
H2 with D2 and the observation of [DPiPr3][BArF ] in the
4
31P{1H} NMR spectrum [1:1:1 triplet, δ 46.1, J(DP) 68 Hz].
Intramolecular heterolytic cleavage of metal-bound H2 is now
a well-known phenomenon,2,41 for example, the proton
transfer between dihydrogen and thiolate in the complex [Os-
(η2-H2)(CO)(pyridine-2-thiolate)(PPh3)2]BF4.42 As far as we
are aware, the products of heterolytic H2 cleavage and proton
transfer to a phosphine have only been reported once
previously. Protonation at low temperature of Ru(H)(PCy3)2-
(CO)Cl with H[BF4] initially affords the dihydrogen complex
[Ru(η2-H2)2(PCy3)2(CO)Cl][BF4] (Scheme 8), which subse-
quently eliminates [HPCy3][BF4] to leave a transient com-
pound spectroscopically characterized as 14-electron Ru-
(H)(PCy3)(CO)Cl.43 On warming, a tetrametric compound,
[Ru4Cl7(PCy3)4(CO)4][BF4], is isolated. The elimination of
[HPR3]+ from a cationic dihydrogen complex to give an
unsaturated species that subsequently condenses to form a
multimetallic product shows clear parallels with the rhodium
compounds under discussion here.
Figure 5. DFT optimized structures for [Rh(H)2(η2-H2)2(PMe3)2]+ (B3LYP/
DZVP level).
167°, while that for D is 177°, and all are similar to that
calculated for RuH2(H2)2(PH3)2.27 We have not searched for
transition states between the three isomers A-C, but given
that experimentally dihydrogen rotation is faster than the
NMR time scale for 1a and 2a at 190 K, the barrier between
the isomers is expected to be small. Barriers to η2-H2 rotation
have been determined both theoretically and experimentally
on related systems and are indeed small (less than 1 kcal
mol-1).16,26,27 The H-H distance in all three isomers is very
short, at ca. 0.77 Å. This is longer than in free H2 (0.74 Å)
but shorter than in other neutral polyhydride-dihydrogen
complexes studied by DFT methods, such as Ru(H)2(η2-H2)2-
(PH3)2 [∼0.85 Å]27 and IrIH2(η2-H2)(PMe3)2 [0.86 Å].26 Not
unsurprisingly, the H-H distance is more like that calculated
for cationic hydride/dihydrogen systems such as [Ir(H)2(η2-
H2)(PH3)2]+ [0.80 Å]16 or observed by neutron diffraction
in [Fe(H)(η2-H2)(dppe)2][BPh4] [0.816(16) Å].40 The short
H-H distance calculated in [Rh(H)2(η2-H2)2(PMe3)2]+ is
consistent with a dihydrogen ligand showing little back-
bonding with the cationic Rh(III) center, coupled with both
H2 ligands being orientated trans to the high trans influence
The mechanism by which the protonated phosphine results
from mixtures 1 and 2 could either be intramolecular (perhaps
via a σ-bond metathesis-type route) or intermolecular depro-
tonation by traces of free PR3 in solution. Whatever the
mechanism, even though there are no π-acid ligands on 1
or 2, the net positive charge on the rhodium center is
sufficient to render the dihydrogen ligand sufficiently acidic
to protonate PCy3 or PiPr3. Morris has compiled extensive
data on the relative pKa values of coordinated dihydrogen
ligands in nonaqueous solvents.44 Given that, on this scale,
the pKa of [HPCy3][BArF ] is 9.7 and that of [HPiPr3][BArF ]
4
4
is 9.0, this places the relative pKa of the dihydrogen ligands
in 1 and 2 as being comparable to these protonated
phosphines. Dihydrogen complexes that bear a positive
(41) Kubas, G. AdV. Inorg. Chem. 2004, 56, 127.
(42) Schlaf, M.; Lough, A. J.; Morris, R. H. Organometallics 1996, 15,
4423.
(43) Yi, C. S.; Lee, D. W.; He, Z.; Rheingold, A. L.; Lam, K.-C.; Concolino,
T. E. Organometallics 2000, 19, 2909.
(44) Abdur-Rashid, K.; Fong, T. P.; Greaves, B.; Gusev, D. G.; Hinman,
J. G.; Landau, S. E.; Lough, A. J.; Morris, R. H. J. Am. Chem. Soc.
2000, 122, 9155.
(40) Ricci, J. S.; Koetzle, T. F.; Bautista, M. T.; Hofstede, T. M.; Morris,
R. H.; Sawyer, J. F. J. Am. Chem. Soc. 1989, 111, 8823.
3168 Inorganic Chemistry, Vol. 44, No. 9, 2005

Dihydrogen Complexes of Rhodium
Scheme 7
Scheme 8
charge and/or electron-withdrawing ligands are more acidic
than neutral complexes,41 and the estimated pKa for the
dihydrogen complexes 1 and 2 would be thus expected to
be much lower than the neutral isoelectronic complexes
Ru(H)2(η2-H2)2(PR3)2. This is the case, with the R ) PiPr3
complex having a pKa of 39.44 Consistent with this, these
complexes do not eliminate protonated phosphine and
decompose by simple loss of molecular H2 instead to give
bis-phosphine hydride bridged dimers.44,45
ligands, and hydrogenation may proceed by insertion into a
rhodium(III)-hydride bond.
Experimental Procedures
General. All manipulations were performed under an inert
atmosphere of argon, using standard Schlenk-line and glovebox
techniques. Glassware was dried in an oven at 130 °C overnight
and flamed with a blowtorch, under vacuum, three times before
use. CH2Cl2 and pentane were distilled from CaH2, diethyl ether
from sodium/benzophenone ketal, and hexane from sodium. Fluo-
robenzene and m-fluorotoluene were dried over CaH2 and vacuum
distilled. C6D6, d8-THF, and d8-toluene were dried over a potassium
mirror, and CD2Cl2 was distilled under vacuum from CaH2.
Microanalyses were performed by Mr. Alan Carver (University of
Conclusions
The addition of H2 to [Rh(nbd)(PR3)2][BArF ] salts (R )
4
iPr or Cy) in CD2Cl2 results in Rh(III) dihydrogen/hydride
complexes of the general formula [Rh(H)2(η2-H2)2(PR3)2]-
[BArF ] and [Rh(H)2(η2-H2)(CD2Cl2)(PR3)2][BArF ]. These
Bath Microanalytical Service). [Rh(nbd)(PR3)2][BArF ] complexes
4
were prepared by an adaptation of the published route using
4
4
K[BArF ]48 and PR3.49 [Rh(nbd)(PCy3)2][1-H-closo-CB11Me11] was
4
complexes have been characterized at low temperature by
NMR spectroscopy by using a combination of T1 measure-
ments, an observed H-D coupling constant of ∼30 Hz, and
a crystal structure of the cyclohexyl phosphine derivative
that shows a trans disposition of phosphine ligands. Experi-
mental characterization has been supported by DFT calcula-
tions. At room temperature, these complexes slowly lose
prepared by a similar route using Ag[1-H-closo-CB11Me11].50 All
other compounds were used as received from Aldrich or Strem
Chemicals. We were not able to obtain an elemental analysis of
any of the unstable compounds reported in this work.
1
NMR Spectroscopy. H, 11B, 13C, and 31P NMR spectra were
recorded on Bruker Avance 300 and 400 MHz FT-NMR spectrom-
eters. Unless noted, residual protio solvent was used as reference
for 1H NMR spectra (CD2Cl2: δ ) 5.33). 11B and 31P NMR spectra
were referenced against BF3‚OEt2 (external), and 85% H3PO4
(external), respectively. Coupling constants are quoted in Hertz.
Spectral assignments were aided by the use of T1 measurements,
which were made at 190 or 200 K using the software provided
with a Bruker Avance 400 MHz spectrometer using the standard
inversion-recovery-delay method (180°-τ-90°) method. In each
experiment, a waiting period of 5 times longer than the expected
relaxation time and 11 variable delays were used.
[HPR3][BArF ], suggested to result from heterolytic cleavage
4
of coordinated H2, to afford the novel clusters [Rh6(PR3)6-
(µ-H)12][BArF ] .
4 2
The observation of simple complexes of the formula [Rh-
(H)2(η2-H2)2(PR3)2][BArF ] on the hydrogenation of a nor-
4
bornadiene precursor complex further underlines the role that
such dihydride/dihydrogen complexes may play in the
hydrogenation of olefins using cationic rhodium catalysts.11
Intriguingly, the suggestion that heterolytic H2 cleavage
might occur in these complexes suggests that a similar
mechanism could operate for the hydrogenation of olefins
using these Rh(III) {RhP2}+ fragments, as has been suggested
to operate for cationic ruthenium(II) complexes.46,47 Alter-
natively, olefin may simply replace the bound dihydrogen
[Rh(H)2(η2-H2)2(PCy3)2][BArF ] (1a). A solution of [Rh(Cy3P)2-
4
(nbd)][BArF ] (0.020 g) in CD2Cl2 (0.3 mL) in a Youngs NMR
4
tube was degassed and backfilled with 1 atm of H2 at 77 K. At
room temperature, the H2 pressure in the tube is ca. 4 atm (298/77
≈ 4). On thawing, the solution rapid changed color from orange to
pale yellow. Yield: quantitative by NMR spectroscopy (see text).
(45) Arliguie, T.; Chaudret, B.; Morris, R. H.; Sella, A. Inorg. Chem. 1988,
(48) Buschmann, W. E.; Miller, J. S. Inorg. Synth. 2002, 33, 85.
(49) Rifat, A.; Laing, V. E.; Kocoik-Ko¨hn, G.; Mahon, M. F.; Ruggiero,
G. D.; Weller, A. S. J. Organomet. Chem. 2003, 680, 127.
(50) Ingleson, M. J.; Patmore, N. J.; Kociok-Ko¨hn, G.; Mahon, M. F.;
Ruggiero, G. D.; Weller, A. S.; Clarke, A. J.; Rourke, J. P. J. Am.
Chem. Soc. 2004, 126, 1503.
27, 598.
(46) Chan, W.-C.; Lau, C.-P.; Chen, Y.-Z.; Fang, Y.-Q.; Ng, S.-M.; Jia,
G. Organometallics 1997, 16, 34.
(47) Daguenet, C.; Scopelliti, R.; Dyson, P. J. Organometallics 2004, 23,
4849.
Inorganic Chemistry, Vol. 44, No. 9, 2005 3169

Ingleson et al.
Table 1. Crystal Data and Structure Refinement for 1a
[closo-CB11Me11H]
Attempts to isolate solid material in bulk repeatedly failed.
However, small crystals of the [Y ) 1-H-closo-CB11Me11]- salt
(prepared in exactly the same manner using [(Cy3P)2Rh(NBD)]-
[closo-CB11Me11H]) were obtained by slow diffusion of pentane
into a solution of the complex in C6H5F at 10 °C.
empirical formula
formula weight
temperature (K)
wavelength (Å)
crystal system
C48H106B11P2Rh
967.09
150(2)
0.71073
1H NMR (CD2Cl2, 298 K): δ 7.73 (s, 8H, BArF ), 7.57 (s, 4H,
4
orthorhombic
BArF ), 2.24-1.05 (m, 66H, PCy3); no hydride signal visible. 31P-
4
space group
Cmcm
{1H} NMR (CD2Cl2, 298 K): δ 54.3 (d, 1JRhP ) 110 Hz). 11B NMR
(CD2Cl2, 298 K): δ -5.9 (s). Selected 1H NMR (CD2Cl2, 190 K):
unit cell dimensions
a ) 15.1320(4) Å; R ) 90.000
b ) 17.8420(4) Å; â ) 90.000
c ) 21.1990(8) Å; γ ) 90.000
5723.4(3)
δ -1.81 (br, 4H, T1 ) 14 ms, η2-H2), -14.14 (br s, 2H, T1 ) 216
volume (Å3)
1
ms, Rh-H). 31P{1H} NMR (CD2Cl2, 190 K): δ 60.1 (d, JRhP
)
Z
4
92 Hz)
density (calculated) (mg/mm3)
1.122
0.384
2096
absorption coefficient (mm-1
F(000)
)
[Rh(H)2(η2-H2)(CD2Cl2)(PCy3)2][BArF ] (1b). A solution of
4
[Rh(Cy3P)2(nbd)][BArF ] (0.018 g) in CD2Cl2 (0.3 mL) in a Youngs
4
crystal size (mm)
theta range for data collection (deg)
index ranges
0.15 × 0.15 × 0.10
NMR tube was degassed and backfilled with 1 atm of H2 at 170 K
(∼2 atm at 298 K). On thawing, the solution rapidly changed color
from orange to pale yellow to afford a mixture of 1a and 1b in an
approximate 1:1 ratio.
3.53-27.43
-19 r h r 19; -23 r k r 22;
-27 r l r 27
44653
3458 [R(int) ) 0.1009]
1921
reflections collected
independent reflections
reflections observed (>2σ)
data completeness
Selected NMR data for 1b: 1H NMR (CD2Cl2, 190 K): δ 0.11
(2 H, T1 29 ms, η2-H2), -12.3 (br, 1H, T1 137 ms, Rh-H), -22.2
(br, 1H, T1 259 ms, Rh-H). 31P{1H} NMR (CD2Cl2, 190 K): δ
0.993
absorption correction
refinement method
none
full-matrix least-squares on F2
3458/117/227
1.807
1
54.3 (d, JRhP ) 102 Hz).
data/restraints/parameters
goodness-of-fit on F2
final R indices [I > 2σ(I)]
R indices (all data)
[RhH2(L)2(PCy3)2][BArF ] (1c) (L ) Solvent or Agostic
4
Interaction). A sample of [(Cy3P)2Rh(H2)2H2][BArF ] was formed
R1 ) 0.1511; wR2 ) 0.4626
R1 ) 0.2208; wR2 ) 0.4920
3.554 and -1.701
4
in situ by the hydrogenation of [(Cy3P)2Rh(nbd)][BArF ] in C6H5F.
4
largest diff. peak and hole (eÅ-3
)
The solvent was evaporated to dryness, and the residue was dried
in vacuo for 5 h to leave a dark yellow oil. Attempts to obtain
solid material repeatedly failed due to the ready decomposition of
this compound.
obtain analytically pure solid material repeatedly failed due to the
ready decomposition of this compound.
1H NMR (CD2Cl2, 298 K): δ 7.72 (s, 8H, BArF ), 7.56 (s, 4H,
4
1H NMR (CD2Cl2, 298 K): δ 7.74 (s, 8H, BArF ), 7.54 (s, 4H,
3
4
BArF ), 2.33 (m, 6H, CHCH3), 1.25 (doublet of heptets, JPH
)
4
1
BArF ), 2.46-1.27 (m, 66H, PCy3) -24.03 (br d, JRhH ) 42 Hz,
3
1
4
14.4, JHH ) 7.6 Hz, 36H, CH3), -24.36 (dt, JRhH ) 40.4 Hz,
2JPH ) 14.4 Hz, 2H, Rh-H). 31P{1H} NMR (CD2Cl2, 298 K): δ
57.6 (d, 1JRhP ) 104 Hz). 11B NMR (CD2Cl2, 298 K): δ -5.9 (s).
2H, Rh-H). 31P{1H} NMR (CD2Cl2, 298 K): δ 53.9 (d, JRhP
)
1
111 Hz). 11B NMR (CD2Cl2, 298 K): δ -5.9 (s).
[Rh(H)2(η2-H2)2(PiPr3)2][BArF ] (2a) and [Rh(H2)(H)2(CD2Cl2)-
[Rh(THF)2(H)2(PiPr3)2][BArF ] (3a) and [Rh (THF)(η2-H2)-
4
4
(iPr3P)2][BArF ] (2b). A solution of [(iPr3P)2Rh(nbd)][BArF ] (0.015
(H)2(PiPr3)2][BArF ] (3b). A solution of [Rh(iPr3P)2(nbd)][BArF ]
4
4
4
4
g) in CD2Cl2 (0.3 mL) in a Young’s NMR tube was degassed and
backfilled with 1 atm of H2 at 77 K. The tube was sealed, and
upon thawing to room temperature, the solution rapidly changed
color from orange to pale yellow. Yield: quantitative by NMR
spectroscopy (see text).
(0.018 g) in d8-THF (0.3 mL) in a Youngs NMR tube was degassed
and backfilled with 1 atm of H2 at 170 K (∼2 atm at 298 K). On
thawing, the solution rapidly changed color from orange to pale
yellow to afford a mixture of 3a and 3b in an approximate 1:1
ratio.
1H NMR (d8-THF, 200 K), (3a): δ 7.89 (s, 8H, BArF ), 7.76
1H NMR (CD2Cl2, 298 K): δ 7.73 (s, 8H, BArF ), 7.57 (s, 4H,
4
4
(s, 4H, BArF ), 2.27 (m, 6H, CHCH3), 1.23 (m, 36H, CH3), -24.33
BArF ), 2.31 (m, 6H, CHCH3), 1.25 (doublet of heptets, 36H, 3JPH
4
4
(br, 2H, T1 ) 266 ms, Rh-H). 31P{1H} NMR (d8-THF, 200 K): δ
55.17 (d, 1JRhP ) 114 Hz). 1H NMR (d8-THF, 200 K), (3b): δ 7.89
(s, 8H, BArF ), 7.76 (s, 4H, BArF ), 2.27 (m, 6H, CHCH3), 1.23
3
) 14.8 Hz, JHH ) 7.1 Hz, CH3), -8.62 (v br, 3.6H, (H2)/H).
1
31P{1H} NMR (CD2Cl2, 298 K): δ 60.4 (d, JRhP ) 107 Hz). 11B
4
4
NMR (CD2Cl2, 298K): δ -5.9 (s).
At 200 K, the ratio of 2b/2a at 190 K is ∼2:1 and at 190 is
(m, 36H, CH3), 0.30 (br, 2H, T1 ) 7 ms, η2-H2), -13.08 (m, 1H,
T1 ) 245 ms, Rh-H), -24.89 (m, 1H, T1 ) 252 ms, Rh-H). 31P-
∼4:1.
1
{1H} NMR (d8-THF, 200 K): δ 64.14 (d, JRhP ) 103 Hz).
1
[Rh(H)2(η2-H2)2(PiPr3)2][BArF ] (2a). H NMR (CD2Cl2, 190
4
X-ray Crystallography. Single crystals were analyzed using a
Nonius Kappa CCD diffractometer. Details of the data collection,
solutions, and refinements are given in Table 1. The model was
solved and subsequently refined using full-matrix least squares in
SHELXL-97.51 This structure presented problems from the outset
of the X-ray experiment. Although the crystals appeared to be of
adequate as opposed to exceptional quality under the microscope,
early indexation frames on many trial samples revealed less than
desirable quality diffraction spots and a rather large mosaicity
(1.65°). Moreover, there was a fall off in diffraction intensity at
relatively low Bragg angles, necessitating a long data collection
time per frame, in an attempt to ameliorate this difficulty. The main
K): δ 7.61 (s, 8H, BArF ), 7.46 (s, 4H, BArF ), 2.08, (m, 6H,
4
4
CHCH3), 1.02 (m, 36H, CH3), -1.90 (br, 4H, T1 ) 9 ms, η2-H2),
-14.23 (br s, 1H, T1 ) 202 ms, Rh-H). 31P{1H} NMR (CD2Cl2,
1
190 K): δ 68.4 (d, JRhP ) 92 Hz).
[Rh(H)2(η2-H2)(CD2Cl2)(PiPr3)2][BArF ] (2b). 1H NMR (CD2-
4
Cl2, 190 K): δ 7.61 (s, 8H, BArF ), 7.46 (s, 4H, BArF ), 2.08, (m,
4
4
6H, CHCH3), 1.02 (m, 36H, CH3), -0.28 (br, 2H, T1 ) 9 ms, η2-
H2), -12.66 (s br, 1H, T1 ) 149 ms, Rh-H), -22.42 (s br, 1H, T1
) 144 ms, Rh-H). 31P{1H} NMR (CD2Cl2, 190 K): δ 62.1 (d,
1JRhP ) 100 Hz).
[Rh(H)2(L)2(PiPr3)2][BArF ] (2c) (L ) Solvent or Agostic
4
Interaction). A sample of 2a was formed in situ as outlined
previously. The solvent was evaporated to dryness, and the residue
was dried in vacuo for 5 h to leave a dark yellow oil. Attempts to
(51) Sheldrick, G. M. SHELX-97. A computer program for refinement of
crystal structures, University of Go¨ttingen, Germany, 2004.
3170 Inorganic Chemistry, Vol. 44, No. 9, 2005

Dihydrogen Complexes of Rhodium
problems arose at the structure solution stage, the first hurdle being
definitive space group assignment. The two contenders after detailed
scrutiny were Cmcm and C2/c. Integration of the data in either of
the associated Laue symmetries failed to clearly reveal which of
the two options was correct. Thus, parallel solutions and refinements
were conducted in both space groups. Unfortunately, the model in
both cases was seen to suffer from disorder. Given the quality of
the data set, that the R(int) values for either space group differed
by less than 0.6%, and that the shift/esd values were similar in
both refinements, we decided to present our results in the higher
symmetry option. Taking the problems outlined previously into
account, we make no claim herein with regard to the metric data
for this structure. The asymmetric unit in this structure was seen
to consist of a cation, in which a [Rh(PCy3)2]+ fragment was visible,
and a [closo-HCB11Me11] anion. In the cation, the central rhodium
was seen to be located at a special position with local m2m
symmetry, while P1, C1, and C4 are all seated on a mirror plane
perpendicular to the a axis. A 50:50 disorder of the cyclohexyl
ring based on C5 was successfully modeled, with the restraint (in
each of the disordered fragments) that all C-C distances be similar
and that the ADPs of the partial carbons be alike. The anionic
fragment as presented contains four half-occupancy boron atoms,
one full-occupancy boron, four half-occupancy methyl carbon
atoms, and one full-occupancy methyl carbon. All half-occupancy
atoms are located on mirror planes intrinsic in the space group
symmetry. The cage carbon would appear to be disordered over
the two symmetry related B2 sites. However, no attempt was made
to model dual occupancy of this site by both carbon and boron
partials because of the data quality (C2/c did not provide any
additional illumination). Hydrogen atoms were included at calcu-
lated positions throughout. Methyl hydrogens on partial occupancy
carbons in the anion are included such that they are always located
on the same mirror plane as the relevant parent carbon.
program.52 Structures were optimized without any symmetry
constraints using DFT at the B3LYP hybrid method53,54 with the
DZVP basis set.55 Geometry optimization for complexes used the
Berny routine. The nature of each stationary point was verified
through frequency calculations.
Acknowledgment. The Royal Society (A.S.W.) and the
EPSRC (GR/R36824/01 and GR/T10169/01) are thanked for
funding.
Supporting Information Available: CIF file for the structure
of 1a and DFT calculated structures with selected bond lengths
and angles for [RhH6(PMe3)2]+. This material is available free of
charge via the Internet at http://pubs.acs.org.
IC0482739
(52) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. J. A.; Vreven, T.; Kudin,
K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone,
V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G.
A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;
Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai,
H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev,
O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P.
Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.;
Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas,
O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J.
B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen,
W.; Wong, M. W.; Gonzalez, C.; Pople, J. A.; Gaussian 03, Revision
A.1 ed.; Gaussian, Inc.: Pittsburgh, PA, 2003.
(53) Lee, C.; Yang, W.; Parr, G. R. Phys. ReV. B 1988, 37, 785.
(54) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(55) Godbout, N.; Salahub, D. R.; Andzelm, J.; Wimmer, E. Can. J. Chem.
1992, 70, 560.
DFT Calculations. Gas-phase geometry optimization for all
multinuclear solutes were performed using the Gaussian 03
Inorganic Chemistry, Vol. 44, No. 9, 2005 3171