Solvent influence on the thermodynamics for hydride transfer from bis(diphosphine) complexes of nickel

Article abstract of DOI:10.1039/c6dt00309e

The thermodynamic hydricity of a metal hydride can vary considerably between solvents. This parameter can be used to determine the favourability of a hydride-transfer reaction, such as the reaction between a metal hydride and CO₂ to produce formate. Because the hydricities of these species do not vary consistently between solvents, reactions that are thermodynamically unfavourable in one solvent can be favourable in others. The hydricity of a water-soluble, bis-phosphine nickel hydride complex was compared to the hydricity of formate in water and in acetonitrile. Formate is a better hydride donor than [HNi(dmpe)₂]⁺ by 7 kcal mol^-1 in acetonitrile, and no hydride transfer from [HNi(dmpe)₂]⁺ to CO₂ occurs in this solvent. The hydricity of [HNi(dmpe)₂]⁺ is greatly improved in water relative to acetonitrile, in that reduction of CO₂ to formate by [HNi(dmpe)₂]⁺ was found to be thermodynamically downhill by 8 kcal mol^-1. Catalysis for the hydrogenation of CO₂ was pursued, but the regeneration of [HNi(dmpe)₂] under catalytic conditions was unfavourable. However, the present results demonstrate that the solvent dependence of thermodynamic parameters such as hydricity and acidity can be exploited in order to produce systems with balanced or favourable overall thermodynamics. This approach should be advantageous for the design of future water-soluble catalysts.

Full text of DOI:10.1039/c6dt00309e

Dalton
Transactions
View Article Online
View Journal
PAPER
Solvent influence on the thermodynamics for
hydride transfer from bis(diphosphine) complexes
Cite this: DOI: 10.1039/c6dt00309e
of nickel†
Samantha J. Connelly Robinson,‡ Christopher M. Zall, Deanna L. Miller,
John C. Linehan and Aaron M. Appel*
The thermodynamic hydricity of a metal hydride can vary considerably between solvents. This parameter
can be used to determine the favourability of a hydride-transfer reaction, such as the reaction between a
2
metal hydride and CO to produce formate. Because the hydricities of these species do not vary consist-
ently between solvents, reactions that are thermodynamically unfavourable in one solvent can be favour-
able in others. The hydricity of a water-soluble, bis-phosphine nickel hydride complex was compared to
+
2
the hydricity of formate in water and in acetonitrile. Formate is a better hydride donor than [HNi(dmpe) ]
by 7 kcal mol− in acetonitrile, and no hydride transfer from [HNi(dmpe)
1
+
2
2
] to CO occurs in this solvent.
+
The hydricity of [HNi(dmpe)
CO
for the hydrogenation of CO
2
]
is greatly improved in water relative to acetonitrile, in that reduction of
+
was found to be thermodynamically downhill by 8 kcal mol−1. Catalysis
2
2
to formate by [HNi(dmpe) ]
2
was pursued, but the regeneration of [HNi(dmpe) ] under catalytic
2
Received 22nd January 2016,
Accepted 1st April 2016
conditions was unfavourable. However, the present results demonstrate that the solvent dependence of
thermodynamic parameters such as hydricity and acidity can be exploited in order to produce systems
with balanced or favourable overall thermodynamics. This approach should be advantageous for the
design of future water-soluble catalysts.
DOI: 10.1039/c6dt00309e
www.rsc.org/dalton
influence of reaction conditions, such as choice of solvent, will
require thermodynamic data in multiple solvents for the same
Introduction
While CO is a byproduct of fossil fuel consumption, it is also catalysts and catalytic intermediates. For complexes relevant to
2
1
–6
an attractive starting material for chemical transformations.
For energy applications, designing a synthetic carbon cycle dynamic data for the same system in multiple solvents.
involving efficient and sustainable reduction of CO to a liquid The hydrogenation of CO with one equivalent of molecular
2
the reduction of CO , there have been few reports of thermo-
2
1–24
2
2
fuel is an enticing prospect that could provide renewable fuels hydrogen produces formic acid, which is a thermodynamically
7
–10
for transportation.
The reduction of CO2 is thermo- unfavourable reaction in aqueous solution under standard
dynamically and kinetically challenging, requiring the develop- state conditions of 1 atm. pressure of gases and 1 M formate.
ment of efficient catalysts. In the design of new catalytic Increasing the solution pH, however, changes the thermodyn-
cycles, many different reaction parameters can be tuned, both amic bias and can make the reaction thermodynamically
for the catalyst (ligands, metal, oxidation state) and for the favourable (Scheme 1). The effect of non-standard state con-
reaction conditions (temperature, pressure, solvent choice, centrations of solutes and pressures of gases can be deter-
addition of a base). These changes can lead to the production mined using the equations in the right column, noting that
1
1–16
−1
of reduced species from CO2, such as formate
methanol.
or −RT ln K equals −1.364log K by conversion to kcal mol at
1
7–20
However, understanding and predicting the 25 °C and using base-10 log rather than natural log. More
specifically, the effect of pH upon the reaction free energy can
be calculated using the last term in Scheme 1. This term can
be simplified to ΔG° ꢀ 1:364 ꢁ pH, for which ΔG
°
is the
Pacific Northwest National Laboratory, P.O. Box 999, MS K2-57, Richland,
Washington 99352, USA. E-mail: aaron.appel@pnnl.gov
rxn
rxn
free energy at standard state (1 atm. for gases, 1 M formate).
Using this equation, it can be shown that hydrogenation of
†
Electronic supplementary information (ESI) available: All experimental details,
synthetic procedures, spectroscopic characterization information. CCDC
448783 and 1468872. For ESI and crystallographic data in CIF or other elec-
tronic format see DOI: 10.1039/c6dt00309e
Current address: Eaton Hall, Seattle Pacific University, Seattle, WA 98119, USA.
CO
2
to formate is favourable above a pH of 7.4.
1
The reaction of CO with a metal hydride is a key step in
the catalytic reduction of CO to formate. As such, the hetero-
2
‡
2
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
2
Scheme 1 Determination of the pH requirement for the favourable reduction of CO to formate.
lytic bond strength to cleave H− from metal hydrides (dis- tude of change in hydricity between metal hydrides and
cussed here as the hydricity) has been studied extensively in formate, the choice of solvent can be used to control the spon-
2
5–29
organic solvents.
The hydricity of formate has been taneity of the hydride transfer reaction.
−
1 30,31
reported in acetonitrile to be 44 kcal mol
,
as corrected
Nickel bis(diphosphine) complexes have been used exten-
3
2
by Bercaw et al. The favourability of the reaction between a sively in catalytic reactions, especially in electrocatalytic pro-
2
5,34
metal hydride and carbon dioxide can be predicted by compar- duction and oxidation of hydrogen,
ing the hydricities of formate and the metal hydride: the reac- formate,
oxidation of
3
5,36
37
and oxidation of alcohols. Catalyst development
tion will be thermodynamically favoured if the hydricity value using these complexes has been greatly aided by an under-
°
−1
(
ΔG
ꢀ
) of the M–H species is smaller than 44 kcal mol . In standing of their hydricity values, which have been well charac-
H
+
acetonitrile, hydride transfer from [HNi(dmpe)
2
] (dmpe = 1,2- terized in organic solvents. Recently, Yang et al. determined
bis(dimethylphosphino)ethane) to CO2 was reported to be the hydricity values of a nickel bis(diphosphine) complex,
−
1 31,32
+
22
thermodynamically uphill by 7 kcal mol
,
2
as illustrated in [HNi(DHMPE) ] in acetonitrile, DMSO, and water. Com-
Scheme 2. The favourability of this reaction can be modified pared to the hydricity of formate, the nickel hydride complex
−
1
using reaction additives. Bercaw et al. have established that the was a weaker donor by more than 12 kcal mol in acetonitrile
+
−1
reaction between [HNi(dmpe)
2
]
and CO
2
in organic solvent and DMSO, but only ∼6 kcal mol in water. Catalytic CO
2
3
2
can be made more favourable in the presence of boranes.
reduction with this complex is therefore unfavourable in all
The favourability of a reaction also depends strongly on three solvents, but the reaction is closer to favourable in water.
solvent, due to factors such as the solvent dependence of the The hydricities of Ni complexes can be tuned by variation of
acidity of a species. Noting the variation of pK
solvents, Creutz et al. first showed that the hydricities of metal hydride donor such as [HNi(dmpe) ] , which has the lowest
a
scales between the phosphine ligand, and we hypothesized that a strong
+
2
21
hydrides can also vary considerably with solvent. Recent hydricity of bis-diphosphine nickel complexes in acetonitrile,
2
3,24
22
reports by Berben et al.
and Yang et al. have also shown might promote favourable hydride transfer to CO
2
in water.
significant differences between the hydricities of metal Herein, we determine the hydricity (ΔG°
) of [HNi(dmpe) ] in
+
ꢀ
2
H
hydrides in organic vs. aqueous solutions. Recently, we pro- aqueous solution. Based on the relative hydricities of
+
posed revised thermodynamic parameters for the aqueous- [HNi(dmpe)
2
]
and formate, hydride transfer from
+
•
−
+
phase reactions of fundamental species such as H , H , H , [HNi(dmpe) ] to CO is predicted to be ∼8 kcal downhill. Con-
2
2
3
3
and
H
2
.
When used appropriately and consistently, sistent with this prediction, we observed the stoichiometric
these values enable quantifying the favourability of many transfer of a hydride from [HNi(dmpe)
+
2
2
] to CO to produce
important chemical reactions. For instance, the hydricity of formate in water. The catalytic conversion of CO to formate
2
°
formate is increased (ΔG
ꢀ
is decreased) in water relative to was investigated; however, only stoichiometric reactions were
H
°
ꢀ1
ꢀ1
acetonitrile (ΔG
ꢀ
¼ 24 kcal mol vs: 44 kcal mol
, respect- observed due to inadequate solution basicity to regenerate the
H
ively). In water, the Ru complexes reported by Creutz et al. hydride complex in the presence of CO . As a potential
2
undergo an even greater improvement in the hydricity than is pathway toward catalytic turnover, the related complex [HNi
2
1
+
observed for formate. Due to the discrepancy in the magni- (dmpbz) ] (dmpbz = 1,2-bis(dimethylphosphino)benzene) was
2
investigated, but the limited solubility of this complex in water
impeded catalysis.
Results
The complexes [Ni(dmpe)
synthesised as previously described,
2
](BF
4
)
2
and [HNi(dmpe)
2
](BF
4
) were
3
8,39
+
and they were used
2
Scheme 2 Hydride transfer from formate to generate [HNi(dmpe) ] in
acetonitrile.
for reactivity and thermochemical studies, as described below.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
Table 1 Measured equilibrium constants for eqn (1) and the resulting
+
2
hydricity of [HNi(dmpe) ]
ꢀ
ΔG°
ꢀ
H
eq. (atm.−
)
1
(kcal mol−1)
Entry # Base
2
P (H ) (atm.)
K
2−
2−
2−
−4
1
2
3
4
5
[CO
[CO
[CO
3
3
3
]
]
]
buffer
buffer
buffer 20.4
1.1
1.7
4.4 × 10
5.4 × 10
15.6
15.7
16.6
16.7
16.1
−4
−3
2.3 × 10−
8.2 × 10
3.3 × 10−
3
3
Et
Et
3
N
N
1.4
40.8
3
Average aqueous hydricity = 16.2 ± 1.0 kcal mol−
1
.
2
Scheme 5 Proposed catalytic cycle for reduction of CO to formate in
The
Ni(dmpe)
aqueous solutions of [Ni(dmpe)
equilibrium
involving
the
interconversion
of
2
+
+
aqueous solution.
[
2
]
and[HNi(dmpe) ] was studied by combining
2
2
+
2
]
with bases under various
pressures of hydrogen gas, as shown in eqn (1) and Table 1.
Sodium carbonate was used as a base under 1.1, 1.7, and
formate in aqueous solution, in which the hydride transfer
step is shown as Step 1.
2
2
0.4 atm. of H . Triethylamine was employed as a base at 1.4
Metal hydrides can be regenerated using heterolysis of H
2
atm. and 40.8 atm. of H . The reaction free energy was deter-
2
as a source of a hydride, but this reaction also produces a
mined from the Keq. values measured by NMR spectroscopy.
proton. The driving force for heterolysis of H is provided by
2
the sum of the hydride acceptor ability of the metal and the
basicity of the proton acceptor. To determine the feasibility for
regenerating the hydride (Step 2 in Scheme 5), the pH at which
the hydride generation is favourable was considered. The reac-
NiðdmpeÞ2 þ þ Base þ H2ðgÞ Ð HNiðdmpeÞ þ þ HBaseþ ð1Þ
2
2
As shown in Scheme 3, the equilibrium data for the overall
reaction (eqn (2)) was combined with the pK
−
1
a
for the conjugate
tion shown in eqn (8) has a free energy of +18 kcal mol
acid of the base employed (eqn (3)) and the heterolytic cleav-
under standard state conditions (eqn (4) minus eqn (5)).
Therefore, nickel hydride generation will only be favourable
above pH ∼ 13, and below this pH, protonation of the hydride
3
3
age free energy for hydrogen in water (eqn (4)) to determine
the hydricity (eqn (5), ΔG°
) of [HNi(dmpe) ] . The individual
ꢀ
2
H
+
−
1
measurements and the average value of 16 ± 1 kcal mol are
shown in Table 1. The data and detailed calculations are
included in the ESI.†
to eliminate H will be favourable (calculations shown in the
2
ESI†).
+
The free energy for the reaction of [HNi(dmpe) ] with CO
NiðdmpeÞ22þ þ H2ðgÞ Ð HNiðdmpeÞ2þ þ Hþ
2
2
ð8Þ
+
was calculated using the hydricity of [HNi(dmpe)
2
]
in con-
°
junction with the hydride acceptor ability (ꢀΔG
ꢀ
) of formate,
H
Phosphate and carbonate buffers were utilised to help
control the pH of the solution by mitigating acidification by
as illustrated in Scheme 4. Consistent with this predicted free
energy, the addition of 1 atm. CO to a solution containing
2
2
CO , but no catalytic turnover was observed. In phosphate
+
2+
[
HNi(dmpe)
and formate, as observed by H NMR spectroscopy. Scheme 5
shows a proposed catalytic cycle for hydrogenation of CO to
2
] in water resulted in formation of [Ni(dmpe)
2
]
31
1
buffers, the pH of the solution could be monitored by P{ H}
1
NMR spectroscopy, and could be observed to drop from pH 12
2
to pH 8 upon addition of CO
2
at 1 atm. The calibration curve
relating the solution pH with the chemical shift of the phos-
3
1
1
phate buffer at room temperature in the P{ H} NMR spec-
trum is included in the ESI.†
[
Ni(dmpbz) ](BF )
and the corresponding hydride
2
4 2
[HNi(dmpbz)
2
](BF
4
) were synthesised using methods analo-
gous to those employed for [Ni(dmpe)
2
](BF
4
)
2
and
[
HNi(dmpe) ](BF ). The complexes of dmpbz were originally
Scheme 3 Reactions and equations to determine hydricity of [HNi
2
4
+
40 31
(
dmpe)
2
] .
prepared with different anions by Warren & Bennett.
P and
and
1
2+
H
NMR spectroscopy data for [Ni(dmpbz)
2
]
+
[HNi(dmpbz) ] are included in the Experimental section,† as
2
this spectroscopy data has not been reported previously. The
−
2+
crystal structures of the BF
[
4
salts of both [Ni(dmpbz)
HNi(dmpbz)2] are included as Fig. 1 and 2, with detailed
information available in the ESI.† By X-ray diffraction, the
structure for [HNi(dmpbz) ][BF ] is nearly tetrahedral around
2
]
and
+
Scheme 4 Free energy for H− transfer in aqueous solution from [HNi
2
4
+
(
dmpe)
2
]
to CO
2
.
nickel, with no obvious coordination site for the hydride
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
+
2
insolubility of [HNi(dmpbz) ] in aqueous solution. The hydri-
city as well as reactivity towards CO could not be determined.
2
Discussion
Whether hydride transfer occurs between a metal hydride and
CO2 to give formate depends on the hydricity of the metal
complex relative to that of formate. Previously, it has been
+
determined that the hydricity of [HNi(dmpe)
2
] in acetonitrile
3
2
is insufficient to promote hydride transfer to carbon dioxide.
In fact, the hydride transfer reaction proceeds in the opposite
in acetonitrile to
generate the nickel hydride complex (Scheme 2). In contrast
to the observed hydride transfer from formate to the dicationic
2
+
direction: formate reacts with [Ni(dmpe)
2
]
3
1
2
+
metal complex in acetonitrile, dissolution of [Ni(dmpe)
HCO2 in water resulted in no observable reaction over a
2
]
and
−
Fig. 1 X-ray crystallographic structure of [Ni(dmpbz)
2
][BF
4
]
2
. Thermal
−
ellipsoid probability set at 30%. BF
omitted for clarity.
4
anions and hydrogen atoms period of days.
To estimate the hydricity of [HNi(dmpe)
mochemical cycle can be established using measured equili-
+
2
] in water, a ther-
3
3
bria, as well as previously reported constants.
By
] , as shown in
Scheme 3 and Table 1, the free energy for transferring a
+
determining the hydricity of [HNi(dmpe)
2
+
hydride from [HNi(dmpe)
2
]
2
to CO can be estimated, as
shown in Scheme 4. While the hydricity of [HNi(dmpe)
+
2
] is
3
2
inadequate in acetonitrile to transfer a hydride to CO2, the
hydricity is greatly improved in water relative to that of
−
1
formate. While this reaction is 7 kcal mol unfavourable in
3
2
−1
acetonitrile, the reaction was determined to be 8 kcal mol
favourable in aqueous solution.
The primary factor that enables the change in relative hydri-
cities is that the hydricity values reported here and those
2
1
reported previously by Creutz do not change consistently
between acetonitrile and water. This variable change between
solvents is expected to enable catalysis in water that was dis-
favoured thermodynamically in organic solvents. For the
+
present report, [HNi(dmpe)
2
] becomes a better hydride donor
]. Thermal by 35 kcal mol− (from 51 to 16 kcal mol−1) when moved from
1
Fig. 2 X-ray crystallographic structure of [HNi(dmpbz)
2
][BF
4
−
ellipsoid probability set at 30%. BF
4
anion and hydrogen atoms omitted
acetonitrile to water, whereas the hydricity value of formate
changes by only 20 kcal mol (from 44 to 24 kcal mol ). As
for clarity. Disorder in the location of the Ni metal centre is shown and
likely leads to the difficulty in locating the hydride ligand in the electron
density map.
−1
−1
such, the hydricity of formate is surpassed by that of
+
[
HNi(dmpe)
favoured.
The majority of the change in the relative hydricities of
2
] , and hydride transfer to generate formate is
ligand. The Ni centre is disordered over a special position, and formate and the metal complexes is likely due to stabilization
we anticipate that the hydride ligand induces this disorder. The of formate by hydrogen bonding in water. This explanation is
presence of the hydride ligand was confirmed by dissolving the consistent with the much larger than average change in the
31
crystalline material in CD
3
CN. Only one P-containing species acidities of carboxylic acids than for other protic species in
is observed, and the resonance for the hydride ligand integrates moving between water and acetonitrile, such as the nearly 19
1
appropriately in the H NMR spectrum (ESI†).
unit change in pK value for acetic acid (from 4.7 to 23.5),
was rather than the more typical change by 7–8 units. This
a
2
+
+
41
The synthesis of [Ni(dmpbz)
2
]
and [HNi(dmpbz)
2
]
+
driven by the expectation that the hydricity of [HNi(dmpbz)
2
]
∼11 unit discrepancy for carboxylic acids is within error of the
would be lower, and therefore regeneration of the hydride 15 kcal mol− change in the relative hydricities of formate and
1
+
would be more feasible in spite of solution acidification by [HNi(dmpe) ] in MeCN and water (using 1.364 as the conver-
2
(see discussion). However, thermochemical, reactivity, sion from a difference in pK values to a difference in free
a
and catalytic studies of [Ni(dmpbz)2] were limited by the energy in kcal mol ). The change in hydricity for
2
3
3
CO
2
+
−1
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
+
[HNi(dmpe)
2
]
is also considerably smaller than the change thermodynamically favourable in organic solvents. Studies of
reported for the analogous complex with a hydroxy substituent catalytic CO2 hydrogenation in aqueous solution with com-
2
2
at each methyl group, as this previously reported complex plexes of first row transition metals are in progress in our
−1
had a change in hydricity of 27.4 kcal mol between aceto- laboratory.
nitrile and water. The difference in change between these com-
plexes may be the result of hydrogen bonding effects for the
2
2
hydroxy substituents in water, similar to the difference Experimental section
observed for formate, as discussed above.
Syntheses and reactivity studies were carried out in air-free
Consistent with the favourable free energy for transfer of a
+
conditions using a N glove box or Schlenk techniques. NMR
2
2 2
hydride from [HNi(dmpe) ] to CO , the stoichiometric reac-
spectra were recorded on Varian 500 MHz spectrometers. High
tion is spontaneous, and therefore the first step in the catalytic
cycle illustrated in Scheme 5 is feasible. To ensure that the
4
3
pressure reactions involving CO
2
were carried out in poly-
ethyletherketone (PEEK) NMR tubes capable of withstanding
2
reaction of the hydride with CO is more favourable than its
4
4,45
the high pressures used in these experiments.
Safety
protonation to generate H , a high pH is required. Even at pH
2
warning: operators of high-pressure equipment such as that
required for these experiments should take proper precautions
to minimise the risk of personal injury.
1
1.8, the solution is acidic enough to protonate most of the
+
2+
[
HNi(dmpe)
protonation reaction to lose H2 and the hydride transfer to
CO to produce formate occur on competing kinetic time
2
] , forming [Ni(dmpe)
2
]
and hydrogen gas. The
The metal starting material, [Ni(CH
3
CN)
6
][BF
thesised according to literature procedure. Ligand dmpe (bis-
,2-dimethylphosphinoethane) was used as received, and
ligand dmpbz (bis-1,2-dimethylphosphinobenzene) was syn-
4 2
] , was syn-
2
4
6
scales (spectra are included in the ESI†).
While the individual steps of the catalytic cycle can be
made favourable, catalytic turnover was not achieved with
1
4
7
2
+
thesised according to a previously reported method.
[Ni(dmpe)
2
] . Turnover was impeded by acidification of the
Full experimental details for all compounds are included in
the ESI.†
reaction solution upon addition of CO , preventing hydride
regeneration. Under CO pressure with a phosphate buffer, the
2
2
solution pH equilibrated near 8, lower than the pH required to
regenerate the hydride.
To increase the favourability of Step 2 of the catalytic cycle,
a nickel hydride complex with a more electron-withdrawing
ligand was used, thereby increasing the stability of the metal
Synthesis of [Ni(dmpe) ][BF ]
2
4 2
The dmpe complex of nickel(II) was synthesised and purified
similarly to the previously published procedure. Characteri-
zation data in aqueous solution: H NMR spectrum (500 MHz,
38
1
2
98 K, D
2
O): 2.07 ppm (CH
2
, vt, 8H), 1.62 ppm (P(CH
3 2
) ,
hydride against protonation and loss of hydrogen. The hydri-
3
1
1
+
vt, 24H). P{ H} NMR spectrum (202 MHz, 298 K, D O):
city of [HNi(dmpbz)
2
] in acetonitrile has been calculated to be
2
−
1
+
43.3 ppm (s).
5
5.9 kcal mol , making it more stable than [HNi(dmpe) ]
2
1
−1 42
(
50.9 kcal mol− ) by ∼5 kcal mol
difference in aqueous solution, this increase in stability
should result in lower reactivity towards protonation as well as In previous reports, [HNi(dmpe)
CO hydrogenation. The free energy of hydride transfer to CO
tion of Ni(dmpe) (synthesised from a Ni(0) precursor). For
in water should still be favourable, with a change in driving this study, the material was obtained from reaction of
.
Assuming a similar
Synthesis of [HNi(dmpe) ][BF ]
2
4
+
2
] was obtained by protona-
3
9
2
2
2
−
1
2 4 2 2
force from −8 to −3 kcal mol . The greater stability against [Ni(dmpe) ][BF ] with excess CsHCO in acetonitrile. Over the
protonation and loss of hydrogen should allow catalysis near course of several minutes, the stirring suspension changes
buffer. Unfortunately, from yellow to orange in colour. The excess CsHCO can be
2
HNi(dmpbz)2] was insoluble in aqueous solutions, so removed by filtration. Characterization data in aqueous solu-
−
the pH of the CO
[
2
/HCO
3
+
1
aqueous hydride transfers could not be studied.
tion: H NMR spectrum (500 MHz, 298 K, D
CH , vt, 8H), 1.45 ppm (P(CH , vt, 24H), −13.6 ppm (NiH,
br) (signal disappears following exchange with the deuterated
2
O): 1.83 ppm
(
2
3 2
)
3
1
1
solvent). P{ H} NMR spectrum (202 MHz, 298 K, D O):
2
Summary and conclusions
24.9 ppm (s) (NiD); 25.0 ppm (s) (NiH).
The hydricities of metal hydrides are highly solvent depen-
dent, especially in comparison to the hydricity of formate.
Synthesis of [Ni(dmpbz) ][BF ]
2
4 2
+
Although hydride transfer from [HNi(dmpe)
2
]
to CO
2
is The dmpbz complex of nickel(II) was prepared by a method
−
1
40
unfavourable in acetonitrile (ΔG° = +7 kcal mol ), this same modified from the procedure reported by Warren & Bennett.
4 2
reaction proceeds spontaneously in water, resulting in the [Ni(CH CN) ][BF ] was employed as a Ni(II) precursor, and
3
6
generation of formate. Thermodynamic measurements were stirred with the dmpbz ligand to generate the desired complex.
performed to determine that the transfer of a hydride from The product was precipitated by layering the solution with di-
+
−1
2+
[HNi(dmpe)
2
] to CO
2
is favourable by 8 kcal mol in water. ethylether. [Ni(dmpbz)
2
]
was originally characterised only by
The observed dependence of the relative hydricities on the UV-vis spectroscopy and elemental analysis, so the NMR spec-
solvent can enable catalytic reactions in water that are not troscopic data are reported here in both water and acetonitrile.
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
1
H NMR Spectroscopy (500 MHz, 298 K, D
2
O): 2.05 ppm 14 R. Langer, Y. Diskin-Posner, G. Leitus, L. J. W. Shimon,
(
(
7
P(CH3)2 vt, 24 H), 7.90 ppm (Ph, 4H), 8.08 ppm (Ph, 4H)
Y. Ben-David and D. Milstein, Angew. Chem., Int. Ed., 2011,
50, 9948–9952.
500 MHz, 298 K, CD
3
3 2
CN): 1.97 ppm (P(CH ) vt, 24 H),
3
1
1
.88 ppm (Ph, 4H), 8.05 ppm (Ph, 4H) P{ H} NMR spec- 15 M. S. Jeletic, M. L. Helm, E. B. Hulley, M. T. Mock,
troscopy (202 MHz, 298 K, D O): 39.2 ppm (s) (202 MHz,
A. M. Appel and J. C. Linehan, ACS Catal., 2014, 4, 3755–3762.
16 J. F. Hull, Y. Himeda, W.-H. Wang, B. Hashiguchi,
R. Periana, D. J. Szalda, J. T. Muckerman and E. Fujita, Nat.
Chem., 2012, 4, 383–388.
2
2
3
98 K, CD CN): 41.8 ppm (s).
Synthesis of [HNi(dmpbz)
2 4
][BF ]
1
7 S. Wesselbaum, V. Moha, M. Meuresch, S. Brosinski,
K. M. Thenert, J. Kothe, T. V. Stein, U. Englert,
M. Hölscher, J. Klankermayer and W. Leitner, Chem. Sci.,
[
Ni(dmpbz) ][BF was reacted with CsHCO in acetonitrile to
2
4
]
2
2
produce the nickel hydride species by the same procedure as
used for [HNi(dmpe) ][BF ]. H NMR spectroscopy (500 MHz,
2
(
NMR Spectroscopy (202 MHz, 298 K, CD CN): 26.7 ppm (s).
This species was completely insoluble in water such that no
aqueous NMR spectroscopy data could be obtained.
1
2
4
2
015, 6, 693–704.
18 C. A. Huff and M. S. Sanford, J. Am. Chem. Soc., 2011, 133,
8122–18125.
98 K, CD
3 3 2
CN): 1.69 ppm (P(CH ) , vt, 24 H), −13.15 ppm
3
1
1
NiH, p, 1H), 7.66 ppm (Ph, 4H), 7.90 ppm (Ph, 4H). P{ H}
1
3
1
2
2
9 E. Balaraman, C. Gunanathan, J. Zhang, L. J. W. Shimon
and D. Milstein, Nat. Chem., 2011, 3, 609–614.
0 E. Alberico and M. Nielsen, Chem. Commun., 2015, 51,
6714–6725 (and references therein).
1 Y. Matsubara, E. Fujita, M. D. Doherty, J. T. Muckerman
Acknowledgements
and C. Creutz, J. Am. Chem. Soc., 2012, 134, 15743–15757.
This material is based upon work supported by the U.S. 22 C. Tsay, B. N. Livesay, S. Ruelas and J. Y. Yang, J. Am. Chem.
Department of Energy, Office of Science, Office of Basic Energy
Soc., 2015, 137, 14114–14121.
Sciences, Division of Chemical Sciences, Geosciences & Bio- 23 A. Taheri, E. J. Thompson, J. C. Fettinger and L. A. Berben,
sciences. Pacific Northwest National Laboratory is operated by
Battelle for the U.S. Department of Energy.
ACS Catal., 2015, 5, 7140–7151.
24 A. Taheri and L. A. Berben, Inorg. Chem., 2016, 55, 378–385.
2
2
5 D. L. DuBois, Inorg. Chem., 2014, 53, 3935–3960.
6 T. J. Schmeier, G. E. Dobereiner, R. H. Crabtree and
N. Hazari, J. Am. Chem. Soc., 2011, 133, 9274–9277.
7 S. Chakraborty, P. Bhattacharya, H. Dai and H. Guan, Acc.
Chem. Res., 2015, 48, 1995–2003.
References
2
1
M. Peters, B. Köhler, W. Kuckshinrichs, W. Leitner,
P. Markewitz and T. E. Müller, ChemSusChem, 2011, 4, 28 M. Rakowski DuBois and D. L. DuBois, Acc. Chem. Res.,
216–1240. 2009, 42, 1974–1982.
F. Barzagli, F. Mani and M. Peruzzini, Green Chem., 2011, 29 S. Chen, M.-H. Ho, R. M. Bullock, D. L. DuBois, M. Dupuis,
3, 1267–1274. R. Rousseau and S. Raugei, ACS Catal., 2013, 4, 229–242.
Carbon Dioxide as a Chemical Feedstock, ed. M. Aresta, 30 D. L. DuBois and D. E. Berning, Appl. Organomet. Chem.,
Wiley-VCH, Weinheim, Germany, 2010. 2000, 14, 860–862.
N. J. English, M. M. El-Hendawy, D. A. Mooney and 31 C. J. Curtis, A. Miedaner, W. W. Ellis and D. L. DuBois,
J. M. D. MacElroy, Coord. Chem. Rev., 2014, 269, 85–95. J. Am. Chem. Soc., 2002, 124, 1918–1925.
P. G. Jessop, F. Joó and C.-C. Tai, Coord. Chem. Rev., 2004, 32 A. J. M. Miller, J. A. Labinger and J. E. Bercaw, Organometal-
48, 2425–2442. lics, 2011, 30, 4308–4314.
C. Maeda, Y. Miyazaki and T. Ema, Catal. Sci. Technol., 33 S. J. Connelly, E. S. Wiedner and A. M. Appel, Dalton Trans.,
014, 4, 1482. 2015, 44, 5933–5938.
E. E. Benson, C. P. Kubiak, A. J. Sathrum and J. M. Smieja, 34 S. Raugei, M. L. Helm, S. Hammes-Schiffer, A. M. Appel,
1
2
3
4
5
6
7
8
1
2
2
Chem. Soc. Rev., 2009, 38, 89–99.
M. O’Hagan, E. S. Wiedner and R. M. Bullock, Inorg. Chem.,
2015, 55, 445–460.
G. A. Olah, A. Goeppert and G. K. Surya Prakash, Beyond Oil
and Gas: The Methanol Economy, Wiley-VCH, Weinheim, 35 B. R. Galan, J. Schöffel, J. C. Linehan, C. Seu, A. M. Appel,
Germany, 2009.
B. Hu, C. Guild and S. L. Suib, J. CO2 Util., 2013, 1, 18–27.
0 R. J. Lim, M. Xie, M. A. Sk, J.-M. Lee, A. Fisher, X. Wang
and K. H. Lim, Catal. Today, 2014, 233, 169–180.
1 Y. Inoue, H. Izumida, Y. Sasaki and H. Hashimoto, Chem.
Lett., 1976, 863–864.
J. A. S. Roberts, M. L. Helm, U. J. Kilgore, J. Y. Yang,
D. L. DuBois and C. P. Kubiak, J. Am. Chem. Soc., 2011, 133,
12767–12779.
9
1
1
1
1
36 C. S. Seu, A. M. Appel, M. D. Doud, D. L. DuBois and
C. P. Kubiak, Energy Environ. Sci., 2012, 5, 6480–6490.
37 C. J. Weiss, E. S. Wiedner, J. A. S. Roberts and A. M. Appel,
Chem. Commun., 2015, 51, 6172–6174.
2 P. G. Jessop, T. Ikariya and R. Noyori, Chem. Rev., 1995, 95,
2
59–272.
3 R. Tanaka, M. Yamashita and K. Nozaki, J. Am. Chem. Soc.,
009, 131, 14168–14169.
2 4 2
38 Prepared similarly to the preparation for [Ni(dmpp) ][BF ]
reported in: D. E. Berning, B. C. Noll and D. L. DuBois,
J. Am. Chem. Soc., 1999, 121, 11432–11447.
2
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
3
9 The synthesis of the PF −
6
salt was previously 43 M. S. Jeletic, M. T. Mock, A. M. Appel and J. C. Linehan,
reported in: A. Miedaner, D. L. DuBois, C. J. Curtis
and R. C. Haltiwanger, Organometallics, 1993, 12, 44 C. R. Yonker and J. C. Linehan, J. Organomet. Chem., 2002,
99–303. 650, 249–257.
0 L. F. Warren and M. A. Bennett, Inorg. Chem., 1976, 15, 45 C. R. Yonker and J. C. Linehan, Prog. Nucl. Magn. Reson.
126–3140. Spectrosc., 2005, 47, 95–109.
1 A. Kütt, I. Leito, I. Kaljurand, L. Sooväli, V. M. Vlasov, 46 R. A. Heintz, J. A. Smith, P. S. Szalay, A. Weisgerber,
J. Am. Chem. Soc., 2013, 135, 11533–11536.
2
4
4
3
L. M. Yagupolskii and I. A. Koppel, J. Org. Chem., 2006, 71,
829–2838.
K. R. Dunbar, K. Beck and D. Coucouvanis, Inorg. Synth.,
2002, 33, 75–83.
2
4
2 X.-J. Qi, Y. Fu, L. Liu and Q.-X. Guo, Organometallics, 2007, 47 E. P. Kyba, S. T. Liu and R. L. Harris, Organometallics, 1983,
6, 4197–4203. 2, 1877–1879.
2
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.