Thermodynamic studies of [H2Rh(diphosphine)2] + and [HRh(diphosphine)2(CH3CN)]2+ complexes in acetonitrile

Article abstract of DOI:10.1021/ic100117y

Thermodynamic studies of a series of [H₂Rh(PP)₂] ⁺ and [HRh(PP)₂(CH₃CN)]²⁺ complexes have been carried out in acetonitrile. Seven different diphosphine (PP) ligands were selected to allow v

Full text of DOI:10.1021/ic100117y

3918 Inorg. Chem. 2010, 49, 3918–3926
DOI: 10.1021/ic100117y
Thermodynamic Studies of [H2Rh(diphosphine)2]þ and
[HRh(diphosphine)2(CH3CN)]2þ Complexes in Acetonitrile
Aaron D. Wilson,†,§ Alexander J. M. Miller,† Daniel L. DuBois,*,‡ Jay A. Labinger,*,† and John E. Bercaw*,†
†Arnold and Mabel Beckman Laboratories of Chemical Synthesis, California Institute of Technology, Pasadena,
California 91125, and ‡Chemical and Materials Sciences Division, Pacific Northwest National Laboratory,
Richland, Washington 99352. § Current address: Interfacial Chemistry, Idaho National Laboratory, P.O. Box
1625, Idaho Falls, Idaho 83415
Received January 19, 2010
Thermodynamic studies of a series of [H2Rh(PP)2]þ and [HRh(PP)2(CH3CN)]2þ complexes have been carried out in
acetonitrile. Seven different diphosphine (PP) ligands were selected to allow variation of the electronic properties of
the ligand substituents, the cone angles, and the natural bite angles (NBAs). Oxidative addition of H2 to [Rh(PP)2]þ
complexes is favored by diphosphine ligands with large NBAs, small cone angles, and electron donating substituents,
with the NBA being the dominant factor. Large pKa values for [HRh(PP)2(CH3CN)]2þ complexes are favored by small
ligand cone angles, small NBAs, and electron donating substituents with the cone angles playing a major role. The
hydride donor abilities of [H2Rh(PP)2]þ complexes increase as the NBAs decrease, the cone angles decrease, and
the electron donor abilities of the substituents increase. These results indicate that if solvent coordination is involved in
hydride transfer or proton transfer reactions, the observed trends can be understood in terms of a combination of two
different steric effects, NBAs and cone angles, and electron-donor effects of the ligand substituents.
Introduction
from H2 gas in the presence of a base will transfer a hydride
ligand to coordinated CO or to BX3 compounds.8-11
A
Transition metal hydride complexes are important in a
broad range of catalytic and stoichiometric reactions. Ex-
amples include biological catalysts such as hydrogenase
enzymes, industrial catalysts such as those involved in hydro-
formylation of olefins, and catalysts used in the synthesis of
L-Dopa for treating Parkinson’s disease.1-6 As a result, more
quantitative models for understanding and predicting the
thermodynamic properties of transition metal hydrides are
important. For example, knowledge of the pKa value and
hydride donor ability of a transition metal hydride allows one
to predict the pH range over which a transition metal hydride
will be stable.7 Similarly, hydricity of transition metal hy-
drides have been used to predict which hydrides generated
quantitative understanding of hydride acceptor abilities is
also useful in designing electrocatalysts for H2 production
and oxidation.12-17
Considerations such as these have inspired efforts to
measure both the kinetic and the thermodynamic hydricity
of transition metal hydrides.7,8,12,13,18-28 Many transition
metal hydrides contain monodentate or bidentate phosphine
ligands, and the steric and electronic properties of these
(9) Ellis, W. W.; Miedaner, A.; Curtis, C. J.; Gibson, D. H.; DuBois, D. L.
J. Am. Chem. Soc. 2002, 124, 1926–1932.
(10) Elowe, P. R.; West, N. M.; Labinger, J. A.; Bercaw, J. E. Organo-
metallics 2009, 28, 6218–6227.
(11) Miller, A. J. M.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc.
2008, 130, 11874–11875.
(12) Berning, D. E.; Noll, B. C.; DuBois, D. L. J. Am. Chem. Soc. 1999,
121, 11432–11447.
(13) Curtis, C. J.; Miedaner, A.; Ellis, W. W.; DuBois, D. L. J. Am. Chem.
Soc. 2002, 124, 1918–1925.
(14) Berning, D. E.; Miedaner, A.; Curtis, C. J.; Noll, B. C.; Rakowski
DuBois, M. C.; DuBois, D. L. Organometallics 2001, 20, 1832–1839.
(15) Curtis, C. J.; Miedaner, A.; Ciancanelli, R.; Ellis, W. W.; Noll, B. C.;
Rakowski DuBois, M.; DuBois, D. L. Inorg. Chem. 2003, 42, 216–227.
(16) Wilson, A. D.; Newell, R. H.; McNevin, M. J.; Muckerman, J. T.;
Rakowski DuBois, M.; DuBois, D. L. J. Am. Chem. Soc. 2006, 128, 358–366.
(17) Wilson, A. D.; Shoemaker, R. K.; Miedaner, A.; Muckerman, J. T.;
DuBois, D. L.; DuBois, M. R. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 6951–
6956.
*To whom correspondence should be addressed. E-mail: daniel.dubois@
pnl.gov (D.L.D.), jal@caltech.edu (J.A.L.), bercaw@caltech.edu (J.E.B.).
(1) Vineyard, B. D.; Knowles, W. S.; Sabacky, M. J.; Bachman, G. L.;
Weinkauff, D. J. J. Am. Chem. Soc. 1977, 99, 5946–5952.
(2) Knowles, W. S. Angew. Chem., Int. Ed. 2002, 41, 1998–2007.
(3) Evans, D.; Osborn, J. A.; Wilkinson, G. J. Chem. Soc., A 1968, 3133–
3142.
(4) Esteruelas, M. A.; Oro, L. A. Chem. Rev. 1998, 98, 577–588.
(5) Fontecilla-Camps, J. C.; Volbeda, A.; Cavazza, C.; Nicolet, Y. Chem.
Rev. 2007, 107, 4273–4303.
(6) Tard, C.; Pickett, C. J. Chem. Rev. 2009, 109, 2245–2274.
(7) Raebiger, J. W.; Miedaner, A.; Curtis, C. J.; Miller, S. M.; Anderson,
O. P.; DuBois, D. L. J. Am. Chem. Soc. 2004, 126, 5502–5514.
(8) Mock, M. T.; Potter, R. G.; Camaioni, D. M.; Li, J.; Dougherty,
W. G.; Kassel, W. S.; Twamley, B.; DuBois, D. L. J. Am. Chem. Soc. 2009,
131, 14454–14465.
(18) DuBois, D. L.; Blake, D. M.; Miedaner, A.; Curtis, C. J.; DuBois,
M. R.; Franz, J. A.; Linehan, J. C. Organometallics 2006, 25, 4414–4419.
(19) Creutz, C.; Chou, M. H. J. Am. Chem. Soc. 2009, 131, 2794–2795.
r
pubs.acs.org/IC
Published on Web 03/24/2010
2010 American Chemical Society

Article
Inorganic Chemistry, Vol. 49, No. 8, 2010 3919
ligands play an important role in determining the reactivity of
the complexes. The steric and electronic effects can be under-
stood in terms of parameters such as the ligand cone angle Θ
and substituent parameters χ introduced by Tolman.29 In
addition to these parameters, Caseyand Whiteker introduced
the concept of the natural bite angle (NBA) for diphosphines
as a measure of the chelate angle preferred by a given ligand
backbone.30 This concept has been elaborated extensively by
van Leeuwen, Weigand, and others.31-38 Of particular re-
levance to the studies reported in this work are previous
thermodynamic studies of five-coordinate monohydride
complexes of the type HM(PP)2 (where M = Co or Rh
and PP is a chelating diphosphine ligand) and [HM0(PP)2]þ
(where M0 = Ni, Pd, and Pt). These studies have shown
the hydride donor abilities of these complexes depend on the
electron donor abilities of the substituents of the diphosphine
ligand,14 the nature of the metal,39 and the natural bite
angle (NBA) of the diphosphine ligand.7 In contrast to these
extensively studied HM(PP)2 and [HM0(PP)2]þ complexes,
the factors controlling the hydride donor abilities of [H2M-
(PP)2]þ complexes are not known. Just as for their five-
coordinate monohydride analogues, it is expected that
knowledge of the factors controlling the hydride donor
abilities of this class of hydrides will allow for a more rational
approach to the development of catalytic and stoichiometric
reactions. In this paper we interpret the thermodynamic
properties of [H2Rh(PP)2]þ and [HRh(PP)2(CH3CN)]2þ
complexes in terms of electronic substituent effects, ligand
cone angles, and the NBAs of the diphosphine ligands.
Scheme 1
Chart 1. List of Diphosphine (PP) Ligands and Their Abbreviations
These substrates include dihalogens,40-42 alkyl halides,43
dioxygen,41,42 and dihydrogen.12-14,41 In addition, these
cations can be reversibly protonated with appropriate acids
in a variety of solvents.12-14,41 Diphosphine ligands offer
flexibility in backbone and substituent selection, allowing
variation in the ligand bite angle as well as the steric and
electronic features of the substituents.
Results
Rhodium(I) cations of the type [Rh(PP)2]þ (where PP = a
diphosphine ligand) are convenient for equilibrium studies
because they lie near thermoneutrality for the oxidative
addition and reductive elimination of several substrates.
The determination of hydride donor abilities of transition
metal hydride complexes relies on the use of thermodynamic
cycles such as the one shown in Scheme 1. The sum of
reactions 1-3 is reaction 4, the heterolytic cleavage of the
Rh-H bond of [H2Rh(PP)2]þ to form [HRh(PP)2(CH3-
CN)]2þ, and the free energy associated with this reaction is
the sum of the free energies corresponding to reactions 1-3 as
shown in eq 5. The determinations of the equilibrium con-
stants associated with reaction 1 and the pKa values for
reaction 2 for complexes containing the diphosphine ligands
in Chart 1 are discussed below. The free energy for the
heterolytic cleavage of H2 in acetonitrile (reaction 3, 76.0
kcal/mol) was taken from reference 49. The stereochemistry
for reaction 1 is shown in reaction 10 below.
(20) Miedaner, A.; Raebiger, J. W.; Curtis, C. J.; Miller, S. M.; DuBois,
D. L. Organometallics 2004, 23, 2670–2679.
(21) Raebiger, J. W.; DuBois, D. L. Organometallics 2005, 24, 110–118.
(22) Price, A. J.; Ciancanelli, R.; Noll, B. C.; Curtis, C. J.; DuBois, D. L.;
DuBois, M. R. Organometallics 2002, 21, 4833–4839.
(23) Ciancanelli, R.; Noll, B. C.; DuBois, D. L.; DuBois, M. R. J. Am.
Chem. Soc. 2002, 124, 2984–2992.
(24) Curtis, C. J.; Miedaner, A.; Raebiger, J. W.; DuBois, D. L. Organo-
metallics 2004, 23, 511–516.
(25) Dedieu, A. Transition Metal Hydrides; John Wiley & Sons: New York, 1991.
(26) Cheng, T.; Bullock, R. M. Organometallics 1995, 14, 4031–4033.
(27) Cheng, T.; Bullock, R. M. Organometallics 2002, 21, 2325–2331.
(28) Cheng, T.; Bullock, R. M. J. Am. Chem. Soc. 1999, 121, 3150–3155.
(29) Tolman, C. A. Chem. Rev. 1977, 77, 313–348.
(30) Casey, C. P.; Whiteker, G. T. Isr. J. Chem. 1990, 30, 299–304.
(31) van der Veen, L. A.; Boele, M. D. K.; Bregman, F. R.; Kamer,
P. C. J.; van Leeuwen, P. W. N. M.; Goubitz, K.; Fraanje, J.; Schenk, H.; Bo,
C. J. Am. Chem. Soc. 1998, 120, 11616–11626.
(32) Dierkes, P.; Leeuwen, P. W. N. M. V. J. Chem. Soc., Dalton Trans.
1999, 1519–1530.
(33) van der Veen, L. A.; Keeven, P. H.; Schoemaker, G. C.; Reek,
J. N. H.; Kamer, P. C. J.; van Leeuwen, P. W. N. M.; Lutz, M.; Spek, A. L.
Organometallics 2000, 19, 872–883.
(34) van Leeuwen, P. W. N. M.; Kamer, P. C. J.; Reek, J. N. H.; Dierkes,
P. Chem. Rev. 2000, 100, 2741–2770.
(35) Freixa, Z.; Leeuwen, P. W. N. M. V. Dalton Trans. 2003, 1890–1901.
(36) van Zeist, W.; Visser, R.; Bickelhaupt, F. Chem.;Eur. J. 2009, 15,
6112–6115.
Determination of the Free Energy of Hydrogen Elimina-
tion (ΔG°-H2) from [H2Rh(PP)2]þ Complexes. It was
(37) Aguila, D.; Escribano, E.; Speed, S.; Talancon, D.; Yerman, L.;
Alvarez, S. Dalton Trans. 2009, ASAP.
€
(38) Niksch, T.; Gorls, H.; Weigand, W. Eur. J. Inorg. Chem. 2009, ASAP.
(42) Suzuki, T.; Isobe, K.; Kashiwabara, K. J. Chem. Soc., Dalton Trans.
1995, 3609–3616.
(43) Marder, T. B.; Fultz, W. C.; Calabrese, J. C.; Harlow, R. L.; Milstein,
D. J. Chem. Soc., Chem. Commun. 1987, 1543–1545.
(39) Curtis, C. J.; Miedaner, A.; Raebiger, J. W.; DuBois, D. L. Organo-
metallics 2004, 23, 511–516.
(40) Chatt, J.; Butter, S. A. Chem. Commun. (London) 1967, 501–502.
(41) Butter, S. A.; Chatt, J. J. Chem. Soc., A 1970, 1411–1415.

3920 Inorganic Chemistry, Vol. 49, No. 8, 2010
Wilson et al.
Table 1. 31P{1H} NMR Data and 1H NMR Hydride Data and Assignments for [Rh(PP)2]þ, [H2Rh(PP)2]þ, and [HRh(PP)2(CH3CN)]2þ Complexes
complex
31P NMR (δ)
1H NMR hydride peak (δ)
[Rh(dmpe)2]þ
[Rh(depe)2]þ
36.4, 1JRhP 123 Hz
63.2, 1JRhP 126 Hz
N/A
N/A
N/A
N/A
N/A
N/A
N/A
[Rh(dppe)2]þ
58.5, 1JRhP 133 Hz
[Rh(dcpe)2]þ
67.1, 1JRhP 131 Hz
[Rh(dppb)2]þ
[Rh(depp)2]þ
62.5, 1JRhP 134 Hz
5.0, 1JRhP 126 Hz
[Rh(dppp)2]þ
[Rh(depx)2]þa
[H2Rh(dmpe)2]þ
[H2Rh(depe)2]þ
[H2Rh(dppe)2]þ
[H2Rh(dcpe)2]þ
[H2Rh(dppb)2]þ
[H2Rh(depp)2]þ
8.4, 1JRhP 132 Hz
12.7, 1JRhP 134 Hz
38.0 br, 26.4 br
N/A
-10.3 br d, trans 2JPH 145 Hz
-10.7 br d, trans 2JPH 132 Hz
-9.0 br d, trans 2JPH 152 Hz
-12.5 br d, trans 2JPH 140 Hz
N/A
64.5 br, 47.5 br
63.3 br, 46.9 br
74.6 dt, 1JRhP 96 Hz, 2JPP 14 Hz, 50.5 br
N/A
15.0 dt, 1JRhP 94 Hz, 2JPP 30 Hz
-6.1 dt, 1JRhP 81 Hz, 2JPP 30 Hz
18.9 dt, 1JRhP 100 Hz, 2JPP 30 Hz
7.0 dt, 1JRhP 83 Hz, 2JPP 30 Hz
15.7 br, 3.4 ppm br
38.0, 1JRhP 88 Hz
-10.8 br d, trans 2JPH 140 Hz
[H2Rh(dppp)2]þ
-8. br d, trans 2JPH 141 Hz
[H2Rh(depx)2]þa
-10.8 br d, trans 2JPH 142 Hz
-18.5 dp, 1JRhH 14 Hz, 2JPH 14 Hz
-18.3 dp, 1JRhH 16 Hz, 2JPH 12 Hz
-15.9 dp, 1JRhH 11 Hz, 2JPH 11 Hz
-19.1 dp, 1JRhH 10 Hz, 2JPH 10 Hz
-14.6 dp,1JRhH 10 Hz, 2JPH 7.5 Hz
-17.6 dp, 1JRhH 15 Hz, 2JPH 14 Hz
-14.9 dp, 1JRhH 12 Hz, 2JPH 12 Hz
cis: -10.3, 2JPH 163 Hz
[HRh(dmpe)2(CH3CN)]2þ
[HRh(depe)2(CH3CN)]2þ
[HRh(dppe)2(CH3CN)]2þ
[HRh(dcpe)2(CH3CN)]2þ
[HRh(dppb)2(CH3CN)]2þ
[HRh(depp)2(CH3CN)]2þ
[HRh(dppp)2(CH3CN)]2þ
[HRh(depx)2(CH3CN)]2þa
58.4, 1JRhP 87 Hz
56.0, 1JRhP 92 Hz
63.9, 1JRhP 88 Hz
56.9, 1JRhP 93 Hz
4.8, 1JRhP 85 Hz
6.2, 1JRhP 91 Hz
cis: 26.4, 9.4, -2.0, -18.3
trans: 25.4 and 19.6
trans: -18.6 br s
a Data from reference 21.
previously reported that [H2Rh(dmpe)2]þ and [H2Rh-
(depe)2]þ reversibly lose hydrogen as shown in reaction
10, with equilibrium constants of 0.6 and 4.0 atm, respec-
tively.18 The complex [H2Rh(dppp)2]þ is also in equilib-
rium with [Rh(dppp)2]þ under 1.0 atm of hydrogen, and
the 31P{1H} NMR spectrum of [H2Rh(dppp)2]þ exhibits
two doublets of triplets indicating an AA0XX0M spin
system and a cis geometry. Integration of the 31P NMR
resonances of 1 and 2 can be used to determine an
equilibrium constant of 0.10 atm for reaction 1 when
the diphosphine ligand is dppp. A listing of the observed
31P{1H} NMR resonances and their assignments are
provided in Table 1 for the [Rh(PP)2]þ and [H2Rh-
(PP)2]þ complexes discussed in this paper.
[H2Rh(dcpe)2]þ (1) and [Rh(dcpe)2]þ (2), resonances were
observed at 92.6 ppm (1JRhP 186 Hz) and at 52.2 ppm with
no rhodium coupling. The former is tentatively attributed
to [Rh(dcpe)(CH3CN)2]þ and the latter to a ligand de-
composition product. Despite these additional species, a
reproducible equilibrium constant of 1.0 ꢀ 103 atm was
obtained for reaction 1 over the range of hydrogen
pressures studied (110-140 atm).
A solution of [Rh(dppb)2]þ was subjected to H2 pres-
sures greater than 130 atm, but no oxidative addition
product was observed; hence the dissociation free energy
of [H2Rh(dppb)2]þ is more negative than -4.4 kcal/mol.
At the opposite extreme, [Rh(depp)2]þ and [Rh(depx)2]þ
add H2 irreversibly at 1.0 atm H2. Because no equilibrium
is observed, the free energy of hydrogen elimination for
[H2Rh(depp)2]þ and [Rh(depx)2]þ must be derived indir-
ectly, as discussed below under the topic of hydride donor
abilities. The free energies (ΔG°-H2) associated with
reaction 1 for the [H2Rh(PP)2]þ complexes discussed in
the preceding paragraphs and those available for
[H2Rh(depx)2]þ from the literature21 are summarized in
column 5 of Table 2.
The complex [Rh(dppe)2]þ does not react with H2 at 1.0
atm; however, at 110 atm H2, sufficient [H2Rh(dppe)2]þ is
observed in equilibrium with [Rh(dppe)2]þ to permit
accurate integration of the two rhodium species. An
equilibrium constant of 5.7 ꢀ 102 atm was determined,
corresponding to a ΔG°-H2 value of -3.8 kcal/mol.
Similarly measurable amounts of [H2Rh(dcpe)2]þ were
produced from [Rh(dcpe)2]þ at H2 pressures greater than
130 atm. A broad resonance at 50.5 ppm is assigned to the
phosphorus atoms trans to the hydride ligands (PX and
Determination of pKa Values for [HRh(PP)2(CH3-
CN)]2þ Complexes. In addition to the equilibrium con-
stants for reaction 1, determination of the hydride donor
abilities of [H2Rh(PP)2]þ complexes using Scheme 1 re-
quires pKa values for the cationic rhodium [HRh(PP)2-
(CH3CN)]2þ complexes. The pKa values for the com-
plexes containing the ligands in Chart 1 were determined
by using Scheme 2. In a typical experiment, an acetonitrile
solution of [Rh(PP)2]þ was treated with a mixture of an
acid and its conjugate base, and monitored by 31P{1H}
NMR spectroscopy until equilibrium was achieved.
Equilibrium constants for eq 6 were determined by in-
tegration of the 31P NMR resonances of the equilibrating
rhodium species and mass balance of the acid/conjugate
0
PX of structure 1). This assignment is supported by the
HSQC NMR data, with the observation of a prominent
cross peak between the hydride at -12.5 ppm on the 1H
axis and the resonance at 50.5 ppm on the 31P NMR axis.
This signal can be assigned with confidence because of the
large (∼140 Hz) trans 2JPH. The area of the 31P resonance
at 50.5 ppm correlates with a relatively sharp doublet of
0
triplets at 74.6 ppm attributed to PA and PA of structure
0
1. The cis phosphorus atoms (PA and PA of structure 1)
did not display a cross peak with the hydride. This is
2
attributed to the smaller cis JPH (∼15 Hz) and H2
exchange. In addition to the resonances assigned to

Article
Inorganic Chemistry, Vol. 49, No. 8, 2010 3921
Table 2. Thermodynamic Data and Ligand Parameters for [Rh(PP)2]þ, [H2Rh(PP)2]þ, and [HRh(PP)2(CH3CN)]2þ Complexesa
[H2Rh(PP)2]þ ΔG°-H2
kcal/mol (eq 1)
[H2Rh(PP)2]þ ΔG°H-
kcal/mol (eq 5)
[HRh(PP)2(MeCN)]2þ ΔG°Hþ
kcal/mol (pKa) (eq 9)
PP
ΘBb (deg)
NBAc (deg)
TEMd Σ3χ
dmpe
depe
dppe
dcpe
depp
depx
dppb
dppp
107 (107)
117 (115)
125 (125)
142 (142)
119
121
125
127 (127)
86
86
86
86
93
100
86
93
7.8
6.2
11.2
2.8
6.2
6.2
12.9
11.2
þ0.44e
-0.87e
-3.8
50.4
52.3
59.6
63.1
26.0 (18.9)
22.8 (16.6)
12.3 (9.0)e
8.7 (6.0)
-4.2
þ5.1
61.3
19.8 (14.4)
þ11.4f
<-4.4
þ1.4
71.6f
<58.7
69.5
15.8 (11.5) f (6:1 cis to trans)
12.9 (9.4)e
7.9 (5.8)
a All values are for acetonitrile at 23 ( 2 °C. b From eq 15 as discussed in the Supporting Information, values in parentheses are from reference 29.
c Ref 37. d Tolman electronic parameters from ref 29, as discussed in the Supporting Information. e Ref 18. f Ref 21, weighted average of cis and trans
values.
Scheme 2
Scheme 3
base pair. (Details are discussed in the Experimental
Section.) These equilibrium constants, together with pub-
lished pKa data for acids in acetonitrile (eq 7),9,44-48 were
used to calculate the pKa values of the corresponding
[HRh(PP)2(CH3CN)]2þ complexes using eq 9. The result-
ing values and the corresponding free energies for depro-
tonation of [HRh(PP)2(CH3CN)]2þ are summarized in
column 7 of Table 2. It is worth noting that in reactions
6 and 8, deprotonation of the [HRh(PP)2(CH3CN)]2þ
complex is accompanied by loss of acetonitrile. For these
reactions, the activity of pure acetonitrile solvent is taken
as unity. This choice of standard states results in pKa
values that are internally consistent with pKa values
reported for complexes in acetonitrile that do not involve
solvent loss. However, the pKa values calculated accord-
ing to Scheme 2 cannot be readily transferred to solvents
other than acetonitrile.
Determination of the Hydride Donor Ability (ΔG°H-) of
[H2Rh(PP)2]þ Complexes. The equilibrium constants for
H2 loss and the pKa values for reaction 2 were used to
calculate the hydride donor abilities of the [H2Rh(PP)2]þ
complexes using Scheme 1, with the exception of [H2Rh-
(depp)2]þ. The ΔG°H- values are shown in Table 2. Be-
cause [Rh(depp)2]þ reacts irreversibly with H2, Scheme 1
could not be used to determine the hydride donor ability
of [H2Rh(depp)2]þ. In this case, the hydride donor ability
was determined using Scheme 3, where the free energy for
the heterolytic cleavage of H2 in acetonitrile (eq 12) was
taken from the literature.49 In contrast, [Rh(dppb)2]þ
does not react with H2 even at 140 atm. As a result, the
value of ΔG°H- for this complex can only be determined
to be less than 58.7 kcal/mol.
Discussion
The objective of this work is to delineate the ligand
parameters controlling various thermodynamic properties
of [H2Rh(PP)2]þ and [HRh(PP)2(CH3CN)]2þ complexes.
Previous studies of [HM(PP)2]þ complexes (where M =
Ni, Pd, and Pt) have shown that the hydride donor abilities,
pKa values, and homolytic solution bond dissociation free
energies depend on three factors. These are the nature of the
metal, the electron donor or acceptor ability of the substit-
uents on the diphosphine ligand, and the natural bite angle of
the ligand.39 In contrast, although thermodynamic hydride
donor abilities have been reported for a small number of six-
coordinate metal complexes,19,50-53 there has been no sys-
tematic study of this class. The ligands shown in Chart 1 were
selected because they provide a range of cone angles, natural
bite angles, and electronic parameters as shown in columns
2-4 of Table 2.
Reductive Elimination of H2 from [H2Rh(PP)2]þ Com-
plexes. The reductive elimination of H2 from distorted cis-
octahedral [H2Rh(PP)2]þ complexes results in the forma-
tion of [Rh(PP)2]þ complexes. The NBAs appear to be the
dominant factor influencing this reaction and its reverse,
the oxidative addition of H2 to [Rh(PP)2]þ. For example,
elimination of H2 is favored by 12.3 kcal/mol for [H2Rh-
(depe)2]þ (ΔG°-H2 = -0.87 kcal/mol) compared to [H2-
Rh(depx)2]þ (11.4 kcal/mol), which have NBAs of 86°
and 100°, respectively.37 This effect has been discussed
previously and can be understood in terms of simple
molecular orbital overlap arguments, in which the tetra-
hedral distortion from a square-planar geometry associated
(44) Kaljurand, I.; Kutt, A.; Soovali, L.; Rodima, T.; Maemets, V.; Leito,
I.; Koppel, I. A. J. Org. Chem. 2005, 70, 1019–1028.
(45) Kolthoff, I. M.; Chantooni, M. K.; Bhowmik, S. Anal. Chem. 1967,
39, 1627–1633.
(46) Menger, F. M.; Williams, R. F. J. Org. Chem. 1974, 39, 2131–2133.
(47) Kolthoff, I. M.; Bruckenstein, S.; Chantooni, M. K. J. Am. Chem.
Soc. 1961, 83, 3927–3935.
(48) Kutt, A.; Leito, I.; Kaljurand, I.; Soovali, L.; Vlasov, V. M.;
Yagupolskii, L. M.; Koppel, I. A. J. Org. Chem. 2006, 71, 2829–2838.
(49) Wayner, D. D. M.; Parker, V. D. Acc. Chem. Res. 1993, 26, 287–294.
(50) Ellis, W. W.; Ciancanelli, R.; Miller, S. M.; Raebiger, J. W.;
Rakowski DuBois, M.; DuBois, D. L. J. Am. Chem. Soc. 2003, 125,
12230–12236.
(51) Wei, M.; Wayland, B. B. Organometallics 1996, 15, 4681–4683.
(52) Fu, X.; Wayland, B. B. J. Am. Chem. Soc. 2004, 126, 2623–2631.
(53) Fu, X.; Li, S.; Wayland, B. B. Inorg. Chem. 2006, 45, 9884–9889.

3922 Inorganic Chemistry, Vol. 49, No. 8, 2010
Wilson et al.
with a larger NBA results in a decrease of the energy of the
σ acceptor orbital and an increase in the energies of
the π-donor orbitals of the [Rh(PP)2]þ complex.7,12,54
These changes in orbital energies favor the addition of
hydrogen to [Rh(PP)2]þ derivatives as the NBAs increase.
When considering the effect of the steric bulk of the
ligand substituents, it is not obvious whether sterically
bulky groups will favor either [H2Rh(PP)2]þ or [Rh-
(PP)2]þ. Tolman introduced the concept of cone angle
to quantify the bulkiness of various substituents, and
values have been published for the cone angles of the
diphosphine ligands with ethylene backbones used in this
study (shown in parentheses in Table 2). Tolman calcu-
lated the cone angle for diphosphine ligands (ΘB) using eq
15, where ΘM is the cone angle of the corresponding
monodentate PR3 ligand with the same R substituents as
the diphosphine ligands and R is the experimentally
observed PMP bond angle of the chelating diphosphine
ligand. We have extended this method to the other ligands
shown in Chart 1 using the experimentally determined
bite angles from the work of Aguila et al.37 and the cone
angles for the corresponding monodentate ligands used
by Tolman (see Supporting Information).29 Our calcu-
lated values are the same as those reported by Tolman for
the ligands previously reported with the exception of
depe for which we calculate a cone angle of 117° versus
Tolman’s 115°. This approach allows cone angles that are
consistent with those used by Tolman to be determined
for the additional ligands used in this work.
Figure 1. Plot of ΔG°-H2 versus the cone angle for the data in Table 2.
The diphosphine ligands of the corresponding cis-[H2Rh(PP)2]þ com-
plexes are shown adjacent to each data point with the green diamonds
corresponding to ligands with ethylene backbones, the blue squares to
propylene backbones, and the red triangle to that for depx which contains
four carbon atoms.
Figure 2. Plot of ΔG°Hþ for [HRh(PP)2(CH3CN)]2þ complexes versus
the cone angle of the corresponding diphosphine ligands. The straight line
is the best fit line through the data for the diphosphine ligands containing
ethylene backbones and alkyl substituents.
ΘB ¼ 2=3ðΘM þ 1=2RÞ
ð15Þ
A plot of ΔG°-H2 versus the cone angle for the data in
Table 2 is shown in Figure 1. It can be seen that the data
for the ligands with an ethylene backbone are nearly
colinear (see solid line), and increasing the steric bulk of
the substituents clearly favors reductive elimination of
H2. It is of interest to note that this steric effect is counter
to that observed for increasing the natural bite angle. It is
also apparent from Figure 3 that the data point for the
dppe complex deviates slightly from the value expected
based on pure steric considerations, and this small devia-
tion of -2.1 kcal/mol is attributed to electronic effects.
The poorer electron donating ability of the phenyl sub-
stituents compared to the alkyl substituents favors re-
ductive elimination of H2. This electronic effect has been
noted previously.55-57 In summary, oxidative addition of
H2 to [Rh(PP)2]þ complexes is favored by diphosphine
ligands with large NBAs, small cone angles, and electron
donating substituents, with the NBA being the dominant
factor.
Figure 3. Plotof ΔG°H- versusthe ligand coneangle. The solid lineisthe
best-fit line through the data for the ligands with ethylene backbones and
alkyl substituents.
the free energies (ΔG°Hþ) and pKa values associated with
deprotonation of the corresponding trans-[HRh(PP)2-
(CH3CN)]2þ complexes. These values are shown in col-
umn 7 of Table 2, and Figure 2 shows a plot of ΔG°Hþ
versus the cone angles of the diphosphine ligands.
The solid line represents the best-fit line for the dipho-
sphine ligands with an ethylene backbone and alkyl
substituents. The ΔG°Hþ value of [HRh(dmpe)2(CH3-
CN)]2þ is 17.3 kcal/mol larger than that of [HRh-
(dcpe)2(CH3CN)]2þ, indicating that the cone angles of
the diphosphine ligands play a major role in determining
the pKa values of these complexes. Steric interactions
between the phosphine substituents and the acetonitrile
ligand that coordinates upon protonation of the metal
pKa Values of [HRh(PP)2(CH3CN)]2þ Complexes. Pro-
tonation of [Rh(PP)2]þ cations in acetonitrile is accom-
panied by the coordination of a solvent molecule and
results in the formation of trans-[HRh(PP)2(CH3CN)]2þ
complexes. As a result, it is expected that factors influen-
cing the dissociation of the solvent ligand will influence
(54) Miedaner, A.; Haltiwanger, R. C.; DuBois, D. L. Inorg. Chem. 1991,
30, 417–427.
(55) Hartwig, J. F. Inorg. Chem. 2007, 46, 1936–1947.
(56) Abis, L.; Santi, R.; Halpern, J. J. Organomet. Chem. 1981, 215, 263–
267.
(57) Jones, W. D.; Kuykendall, V. L. Inorg. Chem. 1991, 30, 2615–2622.

Article
Inorganic Chemistry, Vol. 49, No. 8, 2010 3923
apparently results in less stable and more acidic trans-
[HRh(PP)2(CH3CN)]2þ complexes. An analogous expla-
nation, based on counterbalancing steric and electronic
effects, has been offered to account for the unexpected
similarity in the thermodynamic hydride donor abilities
of [HW(CO)4P(OMe3)]- (37 kcal/mol) and [HW(CO)4-
(PPh3)]- (36 kcal/mol).50
The observation that [H2Rh(dmpe)2]þ is a better hy-
dride donor than [H2Rh(dcpe)2]þ by 12.7 kcal/mol sug-
gests that the cone angle of the diphosphine ligand exerts
a significant influence on the hydride donor abilities of
these complexes. This is illustrated graphically in Figure 3
by the linear dependence of the plot of ΔG°H- versus the
ligand cone angle for the ligands with ethylene backbones
and alkyl substituents (solid line). Ligands with larger
cone angles give complexes with poorer hydride donor
abilities or larger ΔG°H- values. This can be understood in
terms of increased steric interactions between acetonitrile
and diphosphine ligands with larger cone angles. This
interaction disfavors the binding of acetonitrile that
accompanies hydride ligand loss. For the ethylene
bridged ligands, dppe deviates from the best-fit line for
the alkyl substituted ligands by 4.0 kcal/mol. This is
attributed to electronic effects with phenyl substituents
resulting in poorer hydride donors than alkyl substitu-
ents, and this electronic effect is comparable to the
difference in the ΔG°Hþ values for the [HRh(PP)2-
(CH3CN)]2þ complexes discussed in the preceding sec-
tion. The NBA also plays an important role in determin-
ing the hydride donor ability of [H2Rh(PP)2]þ complexes.
The complex [H2Rh(depe)2]þ is a better hydride donor
than [H2Rh(depx)2]þ by 19.3 kcal/mol. If this difference is
corrected for electronic effects of ethyl versus benzyl, this
difference would be expected to be slightly smaller (1-
2 kcal/mol). In summary, the hydride donor abilities of
[H2Rh(PP)2]þ complexes increase as the NBAs decrease,
the cone angles decrease, and the electron donor abilities
of the substituents increase. In this case the influence of
the NBAs and cone angles appear to be of comparable
magnitude and dominate electronic effects of the substit-
uents.
Comparison with Other Metal Hydride Systems. Pre-
viously reported thermodynamic studies of metal com-
plexes most relevant to those reported here include the
[H2Co(dppe)2]þ and [H2Pt(EtXantphos)2]2þ systems.
[H2Co(dppe)2]þ is reported to have a ΔG°-H2 of 4.6
kcal/mol while [H2Rh(dppe)]þ has a ΔG°-H2 of -4.0
kcal/mol. This indicates that oxidative addition of H2 to
[Co(dppe)2]þ is favored over the oxidative addition of H2
to [Rh(dppe)2]þ by 8.6 kcal/mol. Although this is only one
direct comparison, the conclusion that [Co(PP)2]þ com-
plexes will bind H2 more strongly than [Rh(PP)2]þ for the
same diphosphine ligand is likely to be generally valid.
The pKa value of 11.3 for [HCo(dppe)2(CH3CN)]2þ com-
pared to 9.0 for [HRh(dppe)2(CH3CN)]2þ indicates that
rhodium is slightly more acidic than cobalt in these
six coordinate structures. [H2Co(dppe)]þ (ΔG°H- =65.1
kcal/mol) is a significantly less powerful hydride donor
than [H2Rh(dppe)]þ (ΔG°H- = 59.6 kcal/mol), but this
difference is smaller than that observed between first
and second row metals in five coordinate monohydride
[HM(PP)2]þ complexes, for which palladium complexes
are generally 15 kcal/mol better hydride donors than the
corresponding nickel complexes.
Comparison of the values for depe, depp, and depx
complexes and of dppe and dppp complexes indicates that
an increase in the NBA also increases the acidity of these
complexes, but this effect is smaller than the effect of the
cone angle. Both trends are understandable in terms of
increased steric interactions between the diphosphine
ligands, and between the diphosphine ligands and acet-
onitrile, as the NBA or cone angle increases. In contrast,
as discussed above, oxidative addition of H2 to [Rh-
(PP)2]þ complexes is significantly favored by an increase
in the NBA of the diphosphine ligands, but disfavored by
an increase of the steric bulk of the ligands cone angle.
This difference in the dependence of the driving force for
the oxidative addition of H2 to [Rh(PP)2]þ and proton-
ation of [Rh(PP)2]þ, both of which result in an increase in
the formal oxidation state and the coordination number
of the metal by two, is thought to arise from the different
geometries (cis and trans) of the two products formed.
The value of ΔG°Hþ for trans-[HRh(dppe)2(CH3-
CN)]2þ, which has phenyl substituents, is significantly
smaller than expected for an alkyl-substituted complex
with the same cone angle. The -5.9 kcal/mol difference
between the expected and observed values for dppe is
attributed to the difference in the electron-donor ability
of phenyl groups compared to alkyl groups. For compar-
ison, the ΔG°Hþ value for [HM(depe)2]þ is 14 or 10 kcal/
mol larger than that of [HM(dppe)2]þ for M = Ni and Pt,
respectively. These values for nickel and platinum are
primarily determined by the electron donor abilities of the
substituents, not the NBAs or cone angles. Apparently
electronic substituent effects are relatively less important
(compared to steric effects) in determining the acidities
of [HRh(PP)2(CH3CN)]2þ complexes than those of the
[HM(PP)2]þ complexes of nickel and platinum; presum-
ably this is in large part a consequence of the sixth ligand
in the former. In summary, large pKa values for [HRh-
(PP)2(CH3CN)]2þ complexes are favored by small ligand
cone angles, small NBAs, and electron donating substit-
uents with both cone angles and electron donor abilities
playing a major role.
Hydride Donor Abilities of [H2Rh(PP)2]þ Complexes.
Column 6 in Table 2 lists the hydride donor abilities
(ΔG°H-) of [H2Rh(PP)2]þ complexes, which measure
the thermodynamic driving force for dissociation of a
hydride ligand from the dihydride to form [HRh-
(PP)2(CH3CN)]2þ. Smaller values of ΔG°H- indicate a
greater driving force for the delivery of a hydride by
[H2Rh(PP)2]þ complexes. As can be seen from Table 2,
the hydride donor abilities of the complexes range from
50.4 kcal/mol for [H2Rh(dmpe)2]þ to 71.6 kcal/mol for
[H2Rh(depx)2]þ. Because the pKa values of [HRh(PP)2-
(CH3CN)]2þ and the ΔG°-H2 values of [H2Rh(PP)2]þ are
used to calculate the ΔG°H- values of [H2Rh(PP)2]þ, it is
anticipated that both electronic and steric effects will
make significant contributions to the hydride donor
abilities of [H2Rh(PP)2]þ complexes.
Group 10 d6 bis(diphosphine) dihydrides are uncom-
mon; nonetheless one example, [H2Pt(EtXantphos)2]2þ
[EtXantphos = 9,9-dimethyl-4,5-bis(diethylphosphino)-
xanthene], has been observed by NMR.20 This complex
has a ΔG°-H2 of 9.3 kcal/mol. Consistent with the rho-
dium trends, this [H2Pt(PP)2]2þ species is one with an

3924 Inorganic Chemistry, Vol. 49, No. 8, 2010
Wilson et al.
extremely large bite angle. Despite the significant driving
force for the formation of [H2Pt(EtXantphos)2]2þ, it still
takes days for 1 atm of hydrogen to completely consume
[Pt(EtXantphos)2]2þ. In contrast, the reactions of H2 with
[Rh(PP)2]þ derivatives are fast (complete prior to taking
NMR spectra) in all cases reported here, even those with
very little driving force. While reaction kinetics have not
been explicitly evaluated in this work, it is clear that there
is a significant difference in the intrinsic activation bar-
riers for the addition of hydrogen to [Rh(PP)2]þ and
H2 from [H2Rh(PP)]þ complexes. We believe that these
concepts can be applied to a large range of proton and
hydride transfer reactions in which solvent coordination
is involved, and that they can help guide the develop-
ment of a more comprehensive and quantitative under-
standing of the thermodynamic properties of these com-
plexes.
Experimental Section
Syntheses. The [Rh(PP)2]þ complexes were prepared as yel-
low to orange powders following standard literature proce-
dures.18,58 In a representative preparation, a solution of the
ligand dppp (2.15 g, 5.21 mmol) in 20 mL of MeCN was added to
a 40 mL MeCN solution of [Rh(COD)2]OTf (1.22 g, 2.61 mmol),
and the reaction mixture was allowed to stir for 2 h at room
temperature. The volume of the solution was reduced to 10 mL
under vacuum, and 100 mL of Et2O added to produce a
precipitate. The bright orange solid was collected by filtration,
washed with 3 100 mL aliquots of petroleum ether, and dried in a
vacuum for 12 h. The NMR data for this complex match the
literature values reported for [Rh(dppp)2]OTf (2.73 g, 2.53
[Pt(PP)2]2þ
.
Comparison with Five-Coordinate HRh(PP)2 Com-
plexes. HRh(PP)2 complexes can be generated by the
deprotonation of [H2Rh(PP)2]þ, and the hydride donor
abilities of the five coordinate hydrides have been inves-
tigated for four different ligands; dmpe (ΔG°H- = 26.4
kcal/mol), depe (ΔG°H-=28.1 kcal/mol), dppb (ΔG°H-
=
33.9 kcal/mol), and depx (ΔG°H- =45 kcal/mol).18,21,22
Comparisons of these values with those of the corre-
sponding [H2Rh(PP)2]þ derivatives indicate that depro-
tonation of the dihydride complexes results in the
formation of a monohydride complex that is approxi-
mately a 25 kcal/mol better hydride donor. If both the
[H2Rh(PP)2]þ and HRh(PP)2 complexes are considered,
the hydride donor abilities demonstrated for HRh-
(dmpe)2 and [H2Rh(depx)2]þspan a range of 45 kcal/
mol. This range of hydride donor abilities can be accessed
readily from H2 gas and bases of various strengths.
1
mmol, 97% yield).58 The 31P{1H} NMR data and H NMR
hydride data also match the literature values for the remaining
[Rh(PP)2]þ complexes used in this study, and these values are
listed in Table 1.18,22,58
Equilibrium Measurements. All equilibrium measurements
were made at 23 ( 2 °C by recording a series of NMR spectra
on three to six different samples until the integration ratios
became constant within experimental error. The standard state
for H2 is 1.0 atm, and the standard state for acetonitrile was
taken as pure acetonitrile as discussed in the text.
Conclusions
Equilibria of [Rh(dppe)2][OTf] and [Rh(dcpe)2][OTf] with H2
Gas at High Pressure. [Rh(dppe)2][OTf] (7 mg, 0.007 mmol) and
0.5 mL of CD3CN were added to a sapphire NMR tube with a
volume of 1.6 mL and fitted with a check valve which could be
connected to a high pressure gas manifold. This tube was then
charged with 1,754 psi (119 atm, ∼4.8 mmol) of hydrogen with
the pressure measured by a Wikall Transmitter UT-10 pressure
gauge integrated into the manifold. Once charged, the NMR
tube was inverted several times over a period of minutes to
induce mixing. The reaction reached equilibrium within minutes
as indicated by a constant ratio of NMR signals for products,
[Rh(dppe)2][OTf] (31P{1H} NMR: 58.5 ppm, int. set to 1.0)
and [H2Rh(dppe)2][OTf] (31P{1H} NMR: 15.0 ppm, int. 0.104;
-6.1 ppm, int. 0.104). The NMR was followed for at least 24 h to
ensure the stability of the product ratio and vessel pressure.
Integration of each of the resonances of [H2Rh(dppe)2][OTf]
and [Rh(dppe)2][OTf] as well as the pressure of the H2 gas were
used to calculate an equilibrium constant for Reaction 1 of 5.7 ꢀ
102 atm. A total of three experiments were conducted, each
charged with hydrogen pressures between 1690 and 1760 psi
(115-120 atm, 4.6-4.8 mmol). The integrals were measured,
and an average ΔG°-H2 of -3.8 kcal/mol was calculated, with
none of the values for individual experiments deviating more
than 0.01 kcal/mol.
The thermodynamic studies described above demonstrate
that while both steric and electronic influences of dipho-
sphine ligands are factors, the steric constraints of the cone
and bite angle tend to dominate the thermodynamic proper-
ties of six-coordinate rhodium hydride complexes. As ex-
pected, electron donating substituents on the diphosphine
ligands increase pKa values of [HRh(PP)2(CH3CN)]2þ, cor-
relate with better hydride donor abilities of [H2Rh(PP)2]þ,
and disfavor reductive elimination of H2 from [H2Rh(PP)2]þ.
Larger NBAs decrease the pKa values of [HRh(PP)2-
(CH3CN)]2þ, correlate with poorer hydride donor abilities
for [H2Rh(PP)2]þ derivatives, and disfavor reductive elim-
ination of H2 from [H2Rh(PP)2]þ complexes. Finally, larger
cone angles decrease the pKa of [HRh(PP)2(CH3CN)]2þ
,
decrease the hydride donor abilities of [H2Rh(PP)2]þ deriva-
tives, and favor reductive elimination of H2 from [H2Rh-
(PP)2]þ. The coordination of acetonitrile that accompanies
the protonation of [Rh(PP)2]þ and hydride transfer from
cis-[H2Rh(PP)2]þ to form trans-[HRh(PP)2(CH3CN)]2þ re-
sults in increased steric interations between the coordinated
acetonitrile ligands and the diphosphine ligands. As a result,
these reactions are disfavored by an increase in both the NBA
and the cone angle. Our results suggest that if solvent
coordination is involved in hydride transfer and proton
transfer reactions, the observed trends can be understood
in terms of two different steric effects, NBAs, and cone
angles, and in terms of electronic effects. It is important to
emphasize that the two steric effects can be much larger
than electronic effects, and that increasing cone angles and
NBAs can have the same effect on thermodynamic pro-
perties, such as the acidity of [HRh(PP)2(CH3CN)]2þ
complexes and the hydricity of [H2Rh(PP)]þ complexes,
or opposite effects, as in the reductive elimination of
A similar series of three experiments were conducted for
[Rh(dcpe)2][OTf], with hydrogen pressures between 1470 and
2060 psi (100-140 atm, 4.6-4.8 mmol). In the pressurized
[Rh(dcpe)2][OTf] system, two major unknown side products
1
were produced (31P{1H} NMR: 92.6 ppm, d, JRhP 186 Hz;
and 52.2 ppm, s) among other minor products in addition to the
species attributed to [H2Rh(dcpe)2][OTf]. All species in the
reaction mixture appeared to reach thermodynamically stable
ratios within 24 h. The [H2Rh(dcpe)2][OTf] species was identi-
fied by a hydride signal in the 1H NMR spectrum at -12.5 ppm,
(58) Zhou, Z.; Facey, G.; James, B. R.; Alper, H. Organometallics 1996,
15, 2496–2503.

Article
Inorganic Chemistry, Vol. 49, No. 8, 2010 3925
m, and by the 31P{1H} NMR: 74.6 ppm, m (P cis to hydride),
required strong acids. In these cases acid/base buffers were
prepared and injected as a solution into the NMR tube contain-
ing the complex and 0.6 mL of CD3CN. A typical experiment
involved adding trifluoromethanesulfonic acid (HOTf) (795 mg,
5.30 mmol) to dimethylformamide (DMF, pKa = 6.1)45 (966
mg, 13.2 mmol) to create a buffer with a ratio of H(DMF)-
OTf/DMF of 0.67. Then a portion of the buffer (167 mg,
H(DMF)OTf, 0.51 mmol, DMF, 0.75 mmol) was injected into
the NMR tube containing [Rh(dppp)2][OTf] (20 mg, 0.019
mmol). The reaction reached equilibrium within minutes as
indicated by a constant ratio of products for more than 24 h.
The products included [Rh(dppp)2][OTf] (31P{1H} NMR: 8.4
2
50.5 ppm, m, JPH ∼140 Hz (P trans to hydride, assignment
supported by 31P-1H HSQC). The relative concentrations of
reactant and product were determined by integration of the
signal assigned to [Rh(dcpe)2][OTf] and the signals associated
with the phosphorus cis to hydride in [H2Rh(dcpe)2][OTf]. In
this series of experiments, an average ΔG°-H2 of -4.2 kcal/mol
was determined. None of the values for individual experiments
deviated more than 0.05 kcal/mol. When [Rh(dppb)2][OTf] was
exposed to high pressures, no reaction was observed.
Equilibrium of [Rh(dppp)2][OTf] with H2 Gas. [Rh(dppp)2]-
[OTf] (5 mg, 0.005 mmol) and 0.7 mL of CD3CN were added to
an NMR tube fitted with a septum. The tube was purged with a
slow stream of H2 for 10 min. The reaction reached equilibrium
within minutes as indicated by a constant ratio of products,
which persisted for more than 24 h. The species observed were
1
ppm, d, JRhP 132 Hz, int. 0.66) to [HRh(dppp)2(CH3-
1
CN)][OTf]2 (31P{1H} NMR: 6.2 ppm, d, JRhP 91 Hz, int. set
to 1.0). Considering the pKa of H(DMF)OTf (6.1)45 in acetoni-
trile and following the procedure set out above, a pKa of 6.5 and
ΔG°Hþ of 8.8 kcal/mol were calculated. This experiment was
conducted three times with a DMF buffer and three times with a
dimethyl sulfoxide buffer (DMSO pKa = 5.8)45 to find that
[HRh(dppp)2(CH3CN)][OTf]2 had an average pKa of 6.0 and
ΔG°Hþ of 8.7 kcal/mol with the maximum deviation for an
individual experiment of 2.2 kcal/mol. Regardless of whether
the acid was protonated DMSO or DMF, the same chemical
shifts were observed for [HRh(dppp)2(CH3CN)][OTf]2, suggest-
ing that [HRh(dppp)2(base)][OTf]2 is not present in appreciable
quantities.
1
[Rh(dppp)2][OTf] (31P{1H} NMR: 8.4 ppm, d, JRhP 132 Hz)
1
and [H2Rh(dppp)2][OTf] (31P{1H} NMR: 18.9 ppm, dt, JRhP
100 Hz, 2JPP 30 Hz; 7.0 ppm, dt, 1JRhP 83 Hz, 2JPP 30 Hz). The
signals centered at 8.4 ppm ([Rh(dppp)2][OTf]) and 7.0 ppm
(half of total signal for [H2Rh(dppp)2][OTf]) were not resolved,
so the combined integral of 1.2 was compared to the set integral
of 1.0 for the signal at 18.8 ppm (the other half of [H2Rh-
(dppp)2][OTf] total signal), giving
a [Rh(dppp)2][OTf]/
[H2Rh(dppp)2][OTf] ratio of 0.10. This value, combined with
the hydrogen pressure of 1 atm, was used to calculate an
equilibrium constant for Reaction 1 of 1.0 ꢀ 10-1 atm., and a
ΔG°-H2 of 1.4 kcal/mol. A total of three experiments were
conducted with no significant deviation in the integration ratios.
pKa Determination of [HRh(dmpe)2(CH3CN)][OTf]2 in Acet-
onitrile. In a typical experiment, [Rh(dmpe)2][OTf] (22.8 mg,
0.041 mmol), 1,8-bis(dimethylamino)naphthalene (proton
sponge, ps, 51 mg, 0.24 mmol), and the triflate salt of protonated
proton sponge (HpsOTf) (24 mg 0.066 mmol) were combined in
an NMR tube with 0.6 mL of CD3CN. The reaction was
followed by 31P NMR spectroscopy, and a constant ratio of
[HRh(dmpe)2(CH3CN)][OTf]2 (15.7 ppm, br; 3.4 ppm, br, int.
set to 1) to [Rh(dmpe)2][OTf] (36.4 ppm, int. 0.85) was observed
in less than 1 h. After correction for protonation of [Rh-
(dmpe)2][OTf], the ratio of H(ps)OTf/ps was determined to be
0.18 based on the initial mass added. Using these two ratios, an
equilibrium constant of 1.5ꢀ10-1 was calculated for reaction 6.
Combining this with the pKa of protonated proton sponge in
acetonitrile (18.2),9 a pKa of 18.9 can be calculated for [HRh-
(dmpe)2(CH3CN)][OTf]2 using eq 9. This corresponds to a
ΔG°Hþ of 26.0 kcal/mol. Three independent experiments were
performed in this manner, and [HRh(dmpe)2(CH3CN)][OTf]2
was found to have an average pKa of 18.9 and ΔG°Hþ of
26.0 kcal/mol with none of the experiments deviating from the
average by more than 0.16 kcal/mol.
pKa Determination of [HRh(dcpe)2(CH3CN)][OTf]2 in Acet-
onitrile. A similar series of experiments was conducted with
[Rh(dcpe)2][OTf], three runs with a DMF buffer, and three runs
with a dimethyl sulfoxide buffer. These experiments had the
added challenge that [Rh(dcpe)2][OTf] decomposed into species
identified as protonated dcpe and [HRh(dcpe)(CH3CN)3][OTf]2
on exposure to strong acid. The addition of excess dcpe com-
pletely prevented the formation of [HRh(dcpe)(CH3CN)3]-
[OTf]2. Ignoring these side equilibria and focusing on the
equilibrium between [Rh(dcpe)2][OTf], and [HRh(dcpe)2-
(CH3CN)][OTf]2, and the buffer, it was found that [HRh-
(dcpe)2(CH3CN)][OTf]2 had an average pKa of 5.8 and ΔG°Hþ
of 7.9 kcal/mol with the maximum deviation for an individual
experiment of 1.5 kcal/mol. Regardless of whether the acid was
protonated DMSO or DMF, the same chemical shifts were
observed for [HRh(dcpe)2(CH3CN)][OTf]2, suggesting that
[HRh(dcpe)2(Base)][OTf]2 is not present in appreciable quantities.
Equilibrium of [HRh(depp)2(CH3CN)][OTf]2 with H2 and
Base. A solution of [Rh(depp)2][OTf] (22 mg, 0.033 mmol),
anisidine (104 mg, 0.84 mmol), and anisidiniumOTf (31 mg,
0.11 mmol) (pKa = 11.9)44 in CD3CN was prepared and sealed
with a septum. This solution was bubbled with hydrogen twice a
day for 2 days before reaching a stable equilibrium. The ratio of
[H2Rh(depp)2][OTf] and [HRh(depp)2(MeCN)][OTf]2 was
determined from both the 31P NMR and the 1H NMR spectra.
31P{1H} NMR for [HRh(depp)2(MeCN)][OTf]2: 4.8 ppm, int
0.2, and for [H2Rh(depp)2][OTf]: 15.0 ppm, int. set to 1.0; -6.1
ppm int. 1.0. 1H NMR for [HRh(depp)2(MeCN)][OTf]2: -17.6
ppm, int 0.5; and for [H2Rh(depp)2][OTf]: -10.8 ppm, int. set to
1.0. These data were used to calculate a value for Keq of 1.3 ꢀ
10-2 for Reaction 10. The value of ΔG°H- for [H2Rh(depp)2]-
[OTf] was determined to be 59.9 kcal/mol. This experiment was
performed twice.
pKa Determination of [HRh(depe)2(CH3CN)][OTf]2 in Acet-
onitrile. A similar set of three experiments was conducted for
[HRh(depe)2(CH3CN)][OTf]2 using the same proton sponge
buffer system. For [HRh(depe)2(CH3CN)][OTf]2 an average
pKa of 16.6 and ΔG°Hþ of 22.8 kcal/mol were found; with
none of the experiments deviating from the average by more
than 0.05 kcal/mol.
pKa Determination of [HRh(depp)2(CH3CN)][OTf]2 in Acet-
onitrile. Using a anisidine buffer system (acetonitrile pKa = 11.9)44
prepared in an analogous way, another trio of experiments was
conducted for [HRh(depp)2(CH3CN)][OTf]2 to give an average
pKa value of 14.4 and ΔG°Hþ of 19.8 kcal/mol, with none of the
experiments deviating from the average by more than 0.04 kcal/
mol. Use of 2,4-dichloroanilinium triflate as the acid produced the
same chemical shifts as protonation with anisidinium triflate, sup-
porting the formation of [HRh(depp)2(CH3CN)][OTf]2 rather than
[HRh(depp)2(base)][OTf]2.
Another pair of experiments was performed with an aniline
buffer (pKa = 10.6)46 prepared in a similar way as the DMSO
and DMF buffers; however, the aniline was initially diluted in
CD3CN to prevent precipitation of the protonated amine. In
these experiments a solution of [Rh(depp)2][OTf] in CD3CN was
prepared and sealed with a septum. One sample was first purged
with hydrogen forming [H2Rh(depp)2][OTf] as the dominant
species. The buffer solution was added to this H2-saturated
solution, and an equilibrium mixture of [H2Rh(depp)2][OTf ]
and [HRh(depp)2(MeCN)][OTf]2 was formed over a period of
pKa Determination of [HRh(dppp)2(CH3CN)][OTf]2 in Acet-
onitrile. The species [Rh(dcpe)2][OTf] and [Rh(dppp)2][OTf ]

3926 Inorganic Chemistry, Vol. 49, No. 8, 2010
Wilson et al.
2 days. The other tube was first injected with the buffer solution
initially forming [HRh(depp)2(MeCN)][OTf]2 as the dominant
species. This buffered solution was then purged with hydrogen
for 10 min twice a day for 2 days, and progress toward
equilibrium monitored. By averaging the values for all four
experiments [H2Rh(depp)2][OTf] was found to have a ΔG°H- of
61.3 kcal/mol with none of the experiments deviating from the
average more than 1.1 kcal/mol.
would like to acknowledge the support of the Chemical Sciences
program of the Office of Basic Energy Sciences, Division of
Chemical Sciences, Biosciences and Geosciences of the Depart-
ment of Energy and by the National Science Foundation. The
Pacific Northwest National Laboratory is operated by Battelle
for the U.S. Department of Energy.
Supporting Information Available: Derivation of Tolman
cone angles for bidentate ligands used. This material is available
free of charge via the Internet at http://pubs.acs.org.
Acknowledgment. This research was funded by BP through
the Methane Conversion Cooperative (MC2). D. L. DuBois