Incorporating amino acid esters into catalysts for hydrogen oxidation: Steric and electronic effects and the role of water as a base

Article abstract of DOI:10.1021/om300409y

Four derivatives of a hydrogen oxidation catalyst, [Ni(P^Cy ₂N^Bn-R₂)₂]²⁺ (Cy = cyclohexyl, Bn = benzyl, R = OMe, COOMe, CO-alanine-methyl ester, CO-phenylalanine-methyl ester), have been prepared to investigate steric and electronic effects on catalysis. Each complex was characterized spectroscopically and electrochemically, and thermodynamic data were determined. Crystal structures are also reported for the -OMe and -COOMe derivatives. All four catalysts were found to be active for H₂ oxidation. The methyl ester (R = COOMe) and amino acid ester containing complexes (R = CO-alanine-methyl ester or CO-phenylalanine-methyl ester) had rates slower (4 s^-1) than that of the parent complex (10 s^-1), in which R = H, which is consistent with the lower amine pK_a's and less favorable ΔG H₂'s found for these electron-withdrawing substituents. Dynamic processes for the amino acid ester containing complexes were also investigated and found not to hinder catalysis. The electron-donating methyl ether derivative (R = OMe) was prepared to compare electronic effects and has a catalytic rate similar to that of the parent complex. In the course of these studies, it was found that water could act as a weak base for H ₂ oxidation, although catalytic turnover requires a higher potential and utilizes a different sequence of catalytic steps than when using a base with a higher pK_a. Importantly, these catalysts provide a foundation upon which larger peptides can be attached to [Ni(P^Cy₂N ^Bn₂)₂]²⁺ hydrogen oxidation catalysts in order to more fully investigate and implement the effects of the outer coordination sphere.

Full text of DOI:10.1021/om300409y

Article
pubs.acs.org/Organometallics
Incorporating Amino Acid Esters into Catalysts for Hydrogen
Oxidation: Steric and Electronic Effects and the Role of Water as a
Base
Sheri Lense, Ming-Hsun Ho, Shentan Chen, Avijita Jain,† Simone Raugei, John C. Linehan,
John A. S. Roberts, Aaron M. Appel, and Wendy Shaw*
Pacific Northwest National Laboratory, Richland, Washington 99354, United States
S
* Supporting Information
ABSTRACT: Four derivatives of a hydrogen oxidation
catalyst, [Ni(PCy2NBn‑R2)2]2+ (Cy = cyclohexyl, Bn = benzyl,
R = OMe, COOMe, CO-alanine-methyl ester, CO-phenyl-
alanine-methyl ester), have been prepared to investigate steric
and electronic effects on catalysis. Each complex was
characterized spectroscopically and electrochemically, and
thermodynamic data were determined. Crystal structures are
also reported for the −OMe and −COOMe derivatives. All
four catalysts were found to be active for H2 oxidation. The
methyl ester (R = COOMe) and amino acid ester containing
complexes (R = CO-alanine-methyl ester or CO-phenyl-
alanine-methyl ester) had rates slower (4 s−1) than that of the
parent complex (10 s−1), in which R = H, which is consistent
with the lower amine pKa's and less favorable ΔGH 's found for these electron-withdrawing substituents. Dynamic processes for
2
the amino acid ester containing complexes were also investigated and found not to hinder catalysis. The electron-donating
methyl ether derivative (R = OMe) was prepared to compare electronic effects and has a catalytic rate similar to that of the
parent complex. In the course of these studies, it was found that water could act as a weak base for H2 oxidation, although
catalytic turnover requires a higher potential and utilizes a different sequence of catalytic steps than when using a base with a
higher pKa. Importantly, these catalysts provide a foundation upon which larger peptides can be attached to [Ni(PCy2NBn2)2]2+
hydrogen oxidation catalysts in order to more fully investigate and implement the effects of the outer coordination sphere.
a positioned amine.8,10 The R and R′ substituents can be varied
INTRODUCTION
■
to tune the catalysts for H2 production (Scheme 1, counter-
Hydrogenases catalyze the production or oxidation of hydro-
clockwise) or oxidation (Scheme 1, clockwise).10−13 In addition
gen, reactions which are of widespread interest for the
to the species implicated in the catalytic cycle, two noncatalyti-
production and utilization of renewable fuels. Using abundant
cally active isomers (endo-exo and exo-exo)14,15 of the
metals, Ni and/or Fe, hydrogenases catalyze this reaction with
high rates and low overpotentials.1−3 Catalysis at the
diprotonated tetrahedral Ni(0) product are observed which
have important consequences for catalysis.12,16 In more recent
hydrogenase active site is enhanced by the surrounding protein
derivatives, these catalysts meet or exceed the rates of the
enzymes,15,17 but at higher overpotentials. The combination of
high catalytic rates and low overpotentials, characteristic of
hydrogenases, is still not being achieved with catalyst
biomimics, indicating that the outer coordination sphere in
the enzyme plays an important role in achieving fast rates at
low overpotentials.
Effects of the second and outer coordination spheres on
molecular catalysts were assessed for hydrogen production
catalysts by attaching amino acids, amino acid esters, and
dipeptides to a derivative of the [Ni(PPh2NPh−R2)2]2+ proton
reduction catalyst.18 The resulting catalytic rates ranged over an
scaffold, including the second coordination sphere (amino acids
which interact with but do not bind to the metal) and the outer
coordination sphere (OCS) (amino acids which do not interact
with the metal). The stable secondary and tertiary structures of
the outer coordination sphere are thought to be vital to
enhancing catalysis by stabilizing the active site, controlling the
environment around the active site and mediating the transfer
of H2, protons, and electrons between the buried active site and
the exterior of the protein.1,4−7
Many structural and functional synthetic analogues of
hydrogenases have been made.8,9 The family of [Ni-
(PR NR′2)2]2+ complexes (Scheme 1) mimics the second
2
coordination sphere of [FeFe]-hydrogenases by including a
positioned pendant amine, resulting in rate enhancements of
several orders of magnitude in comparison to catalysts without
Received: May 14, 2012
Published: September 24, 2012
© 2012 American Chemical Society
6719
dx.doi.org/10.1021/om300409y | Organometallics 2012, 31, 6719−6731

Organometallics
Article
Scheme 1. Mechanism for H2 Production (Counterclockwise) and Oxidation (Clockwise) by [Ni(PR NR′2)2]2+ Complexes
2
Figure 1. Ligands used in this study to make Ni(PCy2NBn‑R2)2 complexes are derivatives of PCy2NBn2 (A), including PCy2NBn‑OMe2 (B), PCy2NBn‑COOMe
(C), PCy2NBn‑Ala‑OMe (D), and PCy2NBn‑Phe‑OMe (E).
2
2
2
order of magnitude, demonstrating the ability of the outer
coordination sphere to modulate the catalytic activity of a
molecular catalyst, despite the remoteness from the metal
center. Separate studies assessed the effect of para substituents
with varying electron-donating ability on [Ni(PPh2NPh‑R2)2]2+
and found a significant dependence of the catalytic rate on the
R substituents, despite their distance from the metal, due to
their effect on the pendant amine pKa.17
While several different effects of the second and outer
coordination sphere (amine pKa and amino acid functionality)
on hydrogen production catalysts have been reported,17,18 their
impact on hydrogen oxidation catalysts has been relatively
unexplored. In this work, we investigate the role of the second
and outer coordination spheres by modifying the groups on the
pendant amines of hydrogen oxidation catalysts, [Ni-
(PCy2NBn2)2]2+, using ether, ester, and amino acid ester
derivatives. The ligands for these catalysts are shown in Figure
1. To investigate electronic effects, the methyl ester (C) and the
similar methyl ether derivative (B) were incorporated into the
catalyst. Nonpolar ester-protected amino acid containing
complexes were also prepared by coupling the amino acid
esters alanine (Ala) and phenylalanine (Phe) to the hydrolyzed
methyl ester derivative (D and E), enabling the study of steric
effects in the absence of additional functional groups. The full
characterization of these new complexes is presented, and
insight into how these groups alter the thermodynamics and
catalytic activity of the parent complex is provided.
RESULTS AND DISCUSSION
■
Synthesis and Characterization of Ligands and Metal
Complexes. The PR NR′ ligands were synthesized by heating
2
2
bis(hydroxymethylcyclohexyl)phosphine and the appropriate
amine or ammonium hydrochloride salt precursor in ethanol, as
described previously.10 The ligands were characterized by mass
spectrometry and 31P{1H} and 1H NMR spectroscopy,
resulting in data that are consistent with the proposed
structures and with previous complexes of this type.10,18 The
ligand PCy2NBn‑OMe (1B) could not be synthesized cleanly or
2
purified by recrystallization; therefore, the impure ligand was
metallated to form [NiII(PCy2NBn‑OMe2)2](BF4)2 and then
reduced to [Ni0(PCy2NBn‑OMe2)2], which precipitates cleanly
from acetonitrile.19
6720
dx.doi.org/10.1021/om300409y | Organometallics 2012, 31, 6719−6731

Organometallics
Article
Amino acid ester complexes were prepared by coupling
(HOBt) resulted in oxidation of the phosphines, and
consequently its use was avoided.
methyl ester protected amino acid hydrochlorides (alanine
+
(Ala) and phenylalanine (Phe)) to PCy2NBn‑COO‑Li 2 to form the
The ligands were metallated by stirring with 0.5 equivalents
of [Ni(CH3CN)6](BF4)2 in acetonitrile. The complexes were
characterized by mass spectrometry, elemental analysis, 1H and
31P NMR spectroscopy, and, when possible, X-ray crystallog-
raphy. The data were consistent with previous data reported for
similar complexes.10 [Ni(PCy2NBn‑Ala‑OMe2)2]2+ and [Ni-
(PCy2NBn‑Phe‑OMe2)2]2+ were also characterized by 1H−1H
TOCSY NMR.
ligands PCy2NBn‑Ala‑OMe2 and PCy2NBn‑Phe‑OMe2, respectively, using
TBTU as the coupling reagent (Scheme 2b and Figure 1D,E).
Scheme 2. Synthesis of (a) PCy2NBnCOO‑Li+ and (b)
2
PCy2NBn‑R‑OMe (R = Ala, Phe)
2
The Ni(0) complexes were generated by reacting the Ni(II)
complexes with hydrogen followed by deprotonation with
tetramethylguanidine (TMG). Unfortunately, the Ni(0)
complexes of the amino acid ester coupled ligands could not
be generated in this way. Following addition of TMG, the
complex forms a precipitate shown by NMR spectroscopy to be
a complex mixture of products, presumably resulting from
deprotonation of the amide, which destabilizes the complex.
Attempts to form the amino acid ester containing Ni(0)
complexes with a weaker base, triethylamine, only generated
the Ni(II) hydride complexes rather than the twice
deprotonated Ni(0) complexes.
Structural Characterization: X-ray Crystallography. X-
ray-quality crystals of [Ni(PCy2NBn‑COOMe2)](BF4)2 and [Ni-
(PCy2NBn‑OMe2)](BF4)2 were obtained by slow diffusion of ether
into acetonitrile solutions of the complexes. As in the parent
complex, [Ni(PCy2NBn2)2]2+,10 all of the six-membered Ni−P−
C−N−C−P chelate rings are in boat configurations. The metal
is four-coordinate and displays a distorted square planar
geometry. The structures are shown in Figure 3, and the
structural parameters are compared to those of the parent
complex in Table 1. These show that the addition of electron-
withdrawing or -donating substituents to the benzyl group does
not alter the first coordination sphere significantly.
Attempts to directly synthesize the precursor ligand to which
amino acids would be attached, PCy2NBn‑COOH2, were
unsuccessful; therefore, this ligand was synthesized by
hydrolysis of PCy2NBn‑COOMe (Scheme 2a). In addition to the
2
1
standard characterization, H−1H TOCSY NMR spectroscopy
was used to identify the amide proton via coupling to the Cα
proton and other protons in the side chain (Figure 2 and Figure
The crystal structures of Ni0(PCy2NBn‑COOMe2)2,
Ni0(PCy2NBn2)2 and Ni0(PCy2NBn‑OMe2)2, shown in Figure 4,
are similar to each other as well as to other related
complexes.20,21 In all of these complexes, the Ni is in a
pseudotetrahedral geometry and each P2N2 ligand contains one
chelate ring in a chair configuration and one chelate ring in a
boat configuration. Structural parameters are given in Table 2.
As was observed for the Ni(II) complexes, the addition of
electron-withdrawing or -donating substituents to the benzyl
group has a negligible effect on the first coordination sphere of
the Ni(0) complexes.
1
S1 (Supporting Information)). Integration of the H NMR
spectra showed complete coupling of each amino acid ester to
PCy2NBn‑COO‑Li+2, in particular through the comparison of the
intensity of the amide and Cα protons to the benzyl protons on
the amine and methylene protons of the eight-membered P2N2
ring. In the case of PCy2NBn‑Phe‑OMe (1E), a large excess (5
2
equiv) of the amino acid ester was used during coupling to
increase the yield, as only ∼5% coupling occurred when the
amino acid ester was added in slight excess. Attempts to drive
the reaction forward by the addition of N-hydroxybenzotriazole
1
1
Figure 2. H NMR spectrum (left) and H−1H TOCSY spectrum (right) of PCy2NBn‑Phe‑OMe (1E) in CD2Cl2. The blue and red boxes indicate
2
coupling between the amide proton, the C∝ proton, and the side chain protons, used to confirm amino acid incorporation.
6721
dx.doi.org/10.1021/om300409y | Organometallics 2012, 31, 6719−6731

Organometallics
Article
(PCy2NBn‑Ala‑OMe2)2]2+ have exchange rates of ∼15 000 s−1.
Therefore, despite the hindered structural dynamics observed
for [Ni(PCy2NBn‑Ala‑OMe2)2]2+, the dynamics are still much faster
than catalysis at room temperature (vida infra).
H2 addition requires a correctly positioned pendant amine,
with one Ni−P−C−N−C−P ring in the boat configuration,22
and the bulky amino acid ester substituents could hinder the
chair-to-boat isomerization process. We are not able to measure
this process directly, but measurements of the intramolecular
proton exchange rates for the diprotonated isomers of
[Ni(PCy2NBn‑R2H)2]2+ (Scheme 1) provide a relative measure
of the chair-to-boat isomerization, since these isomerizations
were found to be the slowest steps in transferring the proton
from one amine to the other.23 To provide a measure of this
isomerization process, intramolecular proton transfer in
[Ni(PCy2NBn‑Phe‑OMe2H)2]2+ was compared to that in the parent
complex [Ni(PCy2NBn2H)2]2+.
At room temperature, the 31P NMR spectra of these
complexes contain a single resonance each for the endo-endo
and exo-exo isomers (Scheme 1) and two resonances for the
endo-exo isomer (Figure 6).24 As the temperature is decreased,
the 31P resonances for the endo-endo and endo-exo species
resolve as proton transfer between the amines is slowed.23
Analyzing the line shape of the 31P resonances as a function of
temperature, we determined the activation parameters for
proton transfer in the endo-exo isomer of [Ni-
(PCy2NBn‑Phe‑OMe2H)2]2+ and compared them to that of
[Ni(PCy2NBn2H)2]2+ (Table 3). For [Ni(PCy2NBn-Phe-OMe2H)2]2+,
proton transfer is slower and the activation enthalpy and free
energy are more positive than for [Ni(PCy2NBn2H)2]2+. The
data for the endo-endo isomer of [Ni(PCy2NBn‑Phe‑OMe2H)2]2+
could not be fit due to a second overlapping exchange process
(see below), but it can be seen qualitatively that proton transfer
for the endo-endo isomer is also slower than that of the parent
complex. The exo-exo isomer was not observed to exchange at
any temperature studied. The slower exchange rates observed
for the Phe-OMe substituted complex suggest that the chair-to-
boat isomerizations are more hindered in the complexes
substituted with the bulky amino acid ester ligands. However,
despite the reduction in rate, it is still several orders of
magnitude faster than catalysis at room temperature (vida
infra) and should not influence catalytic rates.
Figure 3. Crystal structures of [Ni(P2CyNBn‑COOMe2)2](BF4)2 (left) and
[Ni(P2CyNBn‑OMe2)2](BF4)2 (right), showing that the active sites of the
catalysts are similar despite derivatization of the Bn group. Nickel is
shown in green, nitrogen in blue, phosphorus in purple, oxygen in red,
and carbon in black. Solvent molecules, counterions, and hydrogen
atoms have been omitted for clarity.
NMR Spectroscopy: Structural and Chemical Dynam-
ics in the Amino Ester Derivatives. Addition of the amino
acid ester groups to the ligands would be expected to introduce
structural rigidity to the metal complexes, due to the steric bulk
of the amino acid ester groups and the potential for hydrogen
bonding. Indeed, the 31P NMR resonances for [Ni-
(PCy2NBn‑Ala‑OMe2)2]2+ and [Ni(PCy2NBn‑Phe‑OMe2)2]2+ are much
broader than those of the other [Ni(PCy2NBn‑R2)2]2+ complexes
at room temperature (Figure 5). These complexes are thought
to be distorted trigonal bipyramidal in solution, with the fifth
ligand situated in an equatorial position. The fifth ligand is
solvent, an agostic interaction with the electrons in the C−H
−
bond of the benzyl methylene group, or the BF4 counter-
ion.10,18 An averaged but broad peak is usually observed due to
exchange between the axial and equatorial positions for each
phosphorus atom. Comparison of the variable-temperature 31P
NMR spectroscopy of the amino acid ester derivative
[Ni(PCy2NBn‑Ala‑OMe2)2]2+ and that of the methyl ester derivative
[Ni(PCy2NBn‑COOMe2)2]2+ show that exchange is slower for the
amino acid ester derivative, with coalescence occurring
approximately 10 °C higher in temperature (Figure 5). At
their coalescence temperatures of approximately 10 and 20 °C,
respectively, [Ni(PC y 2 NB n ‑ C O O M e 2 )2 ]2 + and [Ni-
Unlike [Ni(PCy2NBn2H)2]2+, which contains only two
doublets at −40 °C for the endo-endo isomer,12,23 [Ni-
(PCy2NBn‑Phe‑OMe2H)2]2+ has four doublets at −40 °C. Addi-
tionally, the downfield endo-exo resonance of the parent
complex remains a single sharp peak,12,23 while that of
[Ni(PCy2NBn‑Phe‑OMe2H)2]2+ splits into two peaks at −20 °C.
Table 1. Structural Parameters of [Ni(PCy2NBn‑OMe2)2]2+, [Ni(PCy2NBn2)2]2+, and [Ni(PCy2NBn‑COOMe2)2]2+
a
[Ni(PCy2NBn‑OMe2)2]2+
2.1885(8)
[Ni(PCy2NBn2)2]2+
2.2048(13)
[Ni(PCy2NBn‑COOMe2)2]2+
Ni−P1 (Å)
Ni−P2 (Å)
Ni−P3 (Å)
Ni−P4 (Å)
Ni−N(1,2,3,4) (Å)
P1−Ni−P2, P3−Ni−P4 (deg)
dihedral angle (deg)
space group
2.1828(2)
2.1891(2)
2.1961(1)
2.2020(2)
3.31, 3.36, 3.27, 3.36
81.85, 82.57
40.62
2.1955(8)
2.2055(14)
2.1913(8)
2.1850(9)
3.376(2), 3.247(2), 3.283(2), 3.257(2)
3.28(7), 3.28(8), N/A, N/A
82.53(3), 82.66(3)
82.74(6), N/A
34.87
41.42
Pbca
C2/c
P21/c
R1
0.0579
0.0651
0.0819
a
Crystal structure obtained previously by Wilson et al.10
6722
dx.doi.org/10.1021/om300409y | Organometallics 2012, 31, 6719−6731

Organometallics
Article
Figure 4. Crystal structures of Ni0(PCy2NBn‑COOMe2)2, Ni0(PCy2NBn2)2, and Ni0(PCy2NBn‑OMe2)2, demonstrating the similarity between the substituted
and unsubstituted Ni0(PCy2NBn‑R2)2 derivatives. Nickel is shown in green, nitrogen in blue, phosphorus in purple, oxygen in red, and carbon in black.
Hydrogen atoms have been omitted for clarity.
Table 2. Structural Parameters of [Ni(PCy2NBn‑OMe2)2], [Ni(PCy2NBn2)2], and [Ni(PCy2NBn‑COOMe2)2]
[Ni(PCy2NBn‑COOMe2)2]
[Ni(PCy2NBn2)2]
[Ni(PCy2NBn‑OMe2)2]
Ni−P1,P2,P3,P4 (Å)
Ni−N1,N2,N3,N4 (Å)
2.1271(8), 2.1268(8), 2.1315(8),
2.1358(8)
2.1376(3), 2.1375(3), 2.1302(3),
2.1301(3)
2.1231(5), 2.1232(5), 2.1261(5),
2.1278(5)
3.783(2) 3.341(3) 3.800(2) 3.415(2)
3.7922(1) 3.3631(1)
3.7697(1), 3.4127(1), 3.7859(1),
3.4003(1)
P1−Ni−P2, P3−Ni−P4
86.06(3), 85.43(3)
85.450(9) 85.446(9)
85.854(19), 85.373(19)
(deg)
dihedral angle (deg)
R1
88.94
88.67
0.0364
C2/c
89.09
0.0622
0.0507
space group
P1
̅
P1
̅
Figure 5. Variable-temperature 31P NMR spectra of [Ni(PCy2NBn‑Ala‑OMe2)2]2+ (left) and [Ni(PCy2NCOOMe2)2]2+ (right) in acetonitrile, showing a
slower exchange process for the amino acid ester containing complex.
The parent complex was described as an AA′BB′ spin system.
The observations for [Ni(PCy2NBn‑Phe‑OMe2H)2]2+ are most
consistent with an ABMN system, or four unique phosphorus
atoms, pairs of which couple to each other. This interpretation
is based on the 31P−31P COSY spectrum at −40 °C, which
shows that each doublet couples only to a doublet in the
opposite pair (Figure S2 (Supporting Information)). A
dependence on field strength (300 and 500 MHz 1H
frequency) was also observed for the difference in frequency
between each pair of doublets, whereas the coupling for each
doublet was independent of field, confirming that the spectrum
for the endo-endo isomer is due to four chemically inequivalent
phosphorus atoms (Figure S3 (Supporting Information)).
These results suggest reduced conformational flexibility or a
minor structural change in comparison to the parent complex
upon protonation.
The resonance belonging to [HNi(PCy2NBn‑Phe‑OMe2)2]+ also
coalesces at temperatures warmer than that observed for the
parent complex. Variable temperature NMR spectra of
[HNi(PCy2NBn‑Phe‑OMe2)2]+ in butyronitrile from −80 to +25
°C are shown in Figure S4 (Supporting Information).
Combined, the structural and proton transfer dynamics suggest
reduced freedom and higher energy barrier processes when
amino acid esters are incorporated into the outer coordination
sphere, but these barriers are not high enough to affect catalysis.
6723
dx.doi.org/10.1021/om300409y | Organometallics 2012, 31, 6719−6731

Organometallics
Article
Figure 6. Variable-temperature 31P NMR spectra of [Ni(PCy2NBn‑Phe‑OMe2H)2]2+ (left) and [Ni(PCy2NBn2H)2]2+ (right) in CD3CN. The figure for
[Ni(PCy2NBn2H)2]2+ was taken from a paper previously published by O’Hagan et al.23 The spectra for [Ni(PCy2NBn‑Phe‑OMe2H)2]2+ show exchange
rates slower than those previously observed for [Ni(PCy2NBn2)2]2+.23
For the complexes [Ni(PCy2NBn‑R2)2] (R = OMe, COOMe),
the pKa of the pendant amine increases with increased electron-
donating character of the benzyl substituent (15.0 and 12.5,
Table 3. Proton Exchange Activation Parameters for the
Endo-Exo Isomer of [Ni(PCy2NBn‑Phe‑OMe2H)2]2+ and
[Ni(PCy2NBn2H)2]2+ in CD3CN
respectively), as expected (Table 4). The pKa values of the
ΔH⧧
ΔS‡ (cal/
ΔG⧧(298 K)
k(298 K)
b
pendant amine for [Ni(PCy2NBn‑Ala‑OMe2H)2]2+ and [Ni-
endo-exo
(kcal/mol)
(mol K))
(kcal/mol)
(s−1
)
2+
(PCy2NBn‑Phe‑OMe2H)2]
were found to be 12.8 and 12.5,
R = Bn-Phe-
OMe
17.4 0.3
18.1 0.9
12.1 0.3
8700
respectively, similar to that of [Ni(PCy2NBn‑COOMe2H)2]2+. From
these values and the potentials of the Ni(II/I) and Ni(I/0)
couples, we calculated the ΔGH and ΔGH values shown in
a
R = Bn
14.8 0.3
11.4 0.9
11.4 0.3
27000
Activation parameters measured previously by O’Hagan et al.23
Rates at 298 K are calculated from the activation parameters.
a
−
2
b
Table 4. The ΔGH value was also determined by
2
experimentally measuring the pKa of the hydride ([HNi-
Electrochemistry and Thermodynamic Measurements
(PCy2NBn‑OMe2)2]+), rather than using the Ni(I/0) couple. The
of the [Ni(PCy2NBn‑R2)2] Complexes. Under N2, Ni-
(PCy2NBn‑COOMe
) ) display reversible
and Ni(PCy2NBn‑OMe
pKa in benzonitrile was found to be 22.6, yielding a ΔGH value
2
2
2 2 2
of −3.6 kcal/mol, which is in good agreement with the value
calculated on the basis of the Ni(I/0) couple (−3.6 kcal/mol).
Previous pKa measurements in acetonitrile and benzonitrile
have shown similar values.25 The calculated free energies of
Ni(II/I) and Ni(I/0) couples in acetonitrile and benzonitrile
(Table S1 (Supporting Information)). The potential for the
Ni(II/I) redox couple becomes less negative as the electron-
donating nature of the benzyl substituent decreases (Table 4).
The Ni(I/0) couple is most negative for Ni(PCy2NBn‑OMe2)2 due
to the electron-donating group but similar for Ni-
hydrogen addition, ΔGH , for catalysts containing methyl ester
2
and amino acid ester substituents are less favorable than that of
the parent complex, whereas the methyl ether substituted
( P C y 2 N B n ‑ C O O M e
)
a n d N i ( P C y 2 N B n 2 ) 2 . [ N i -
2
2
(PCy2NBn‑Ala‑OMe2)2]2+ and [Ni(PCy2NBn‑Phe‑OMe2)2]2+ also display
reversible Ni(II/I) couples in acetonitrile when the scan is
reversed before the Ni(I/0) couple (ΔEpII/I ≈ 78 mV). In cyclic
voltammograms that scan through both the Ni(II/I) and Ni(I/
0) couples, both couples are irreversible due to insolubility of
the Ni(0) species in acetonitrile. As expected, the Ni(II/I)
couple shifts to more anodic potentials as the substituent
becomes more electron withdrawing (Table 4).
catalyst has a slightly more favorable ΔGH .
2
Cyclic Voltammograms (CV's) of the [Ni(PCy Bn‑R2)2]
2
Complexes under H2. Upon sparging solutions of [Ni-
(PCy2NBn‑R2)2]+ (R = COOMe, OMe) with H2, the Ep value for
the Ni(II/I) oxidation shifts to a slightly more negative
potential. The oxidation product, [NiII(PCy2NBn‑R2)2]2+, then
reacts with H2, forming [Ni0(PCy2NBn‑R2H)2]2+, which isomer-
a
Table 4. Electrochemical and Thermodynamic Data for Each of the Ni(PCy2NBn‑R2)2 Derivatives
II/I
I/0
ΔGH
(kcal/mol)
−
E1/2
(V)
E1/2
(V)
TOF(PhCN)
(s−1
TOF(MeCN/H2O)
(s−1
ΔGH
2
complex
Ni(PCy2NBn
)
)
pKa(amine) pKa(NiH)
(kcal/mol)
d
d
d
e
d
d
)
−0.80
−0.74
−0.85
−0.78
−0.78
−1.28
−1.29
−1.39
−1.30
−1.31
5.6 0.6
4.1 0.4
7.2 1.8
3.9 0.6
2.2 0.5
10 , 9
13.4
12.5
15.0
12.8
12.5
21.2
21.3
23.2
21.5
21.7
−3.1
−2.4
−3.6
−2.3
−1.9
60.7
2
2
f
b
b
b
b
b
b
Ni(PCy2NBn‑COOMe
)
2.1 0.2
7.1 0.1
3.4 0.1
3.9 0.2
61.3
2
2
bc
,
b
Ni(PCy2NBn‑OMe
)
59.1
2
2
b
b
b
[Ni(PCy2NBn‑Ala‑OMe2)2]2+
60.8
b
[Ni(PCy2NBn‑Phe‑OMe2)2]2+
60.8
a
b
c
Unless otherwise noted, all measurements were in acetonitrile. Calculated using equations in the Experimental Section. Confirmed using
experimentally measured pKa(NiH) in benzonitrile, as described in the Experimental Section. Previously reported by Wilson et al.10 Rate measured
d
e
f
in this study. Rate measured at the endo-exo [Ni(PCy2NBn‑COOMe2H)2]2+ oxidation potential.
6724
dx.doi.org/10.1021/om300409y | Organometallics 2012, 31, 6719−6731

Organometallics
Article
Table 5. Oxidation Potentials for [Ni(PR NR′2H)2]2+ in
MeCN
izes to the endo-endo, endo-exo, and exo-exo isomers (Scheme
1) on the time scale of the CV. The Ni(II/I) oxidation was not
seen when R′ = Ala-OMe or R′ = Phe-OMe, since it was
necessary to start in the [NiII(PCy2NBn‑R2)2]2+ oxidation state,
which immediately reacts with H2 to form [Ni-
(PCy2NBn‑R2H)2]2+. Each isomer oxidizes irreversibly at a
distinct potential at scan rates up to 1 V s−1. CV’s for the
methyl ether substituted complex, [Ni(I)(PCy2NBn‑OMe2)2]+,
illustrative of what is typically observed, are shown in Figure 7.
2
Δ exo-
exo vs
endo-
exo
Δ endo-
exo vs
endo-
endo
endo-
exo
endo-
c
exo-exo
endo
complex
(R, R′)
Ni(I/0) Ni(I/0) Ni(I/0)
Ni(II/I)
(V)
(V)
(V)
(mV)
(mV)
Cy, Bn-
−0.85
−0.74
−0.33
−0.04
410
350
390
400
290
250
250
230
a
OMe
Cy, Bn-
COOMe
−0.69
nd
−0.59
−0.66
−0.65
−0.24
−0.27
−0.25
0.01
−0.02
−0.02
a
Cy, Bn-Ala-
a
OMe
Cy, Bn-
nd
Phe-
a
OMe
b
tBu, Bn
−0.61
−0.29
−0.12
320
b
170
a
Oxidation peak potentials (Ep) of irreversible waves. E1/2 potentials
of quasi-reversible waves previously determined by Wiedner et al.16
c
Potential measured at half the maximum current of the wave.
These calculations indicate that there is increased hydrogen
bonding stabilization in the [Ni0(PCy2NMe2H)2]2+ complexes.
As can be inferred from Figure 8, because of the tetrahedral
symmetry of the Ni0 complexes, the substituent on the
Figure 7. CV’s of [Ni(PCy2NBn‑OMe2)2]+ under N2 (blue trace) and H2
(red trace), representative of CV’s observed for each of the
[Ni(PCy2NBn‑R2)2] complexes and showing that each of the three
conformational isomers formed upon the addition of H2 (shown in
Scheme 1) have a unique oxidation potential.
Understanding the electrochemistry of each of the isomers
may provide mechanistic insight and aid in future catalyst
design.16 Consistent with work by Weidner et al.,16 the
irreversible oxidation events following the Ni(II/I) oxidation
can be attributed to redox chemistry of the exo-exo (−0.74 V),
endo-exo (−0.33 V) and endo-endo (−0.04 V) isomers of
[Ni0(PCy2NBn‑OMe2H)2]2+. Provided that any kinetic potential
shifts perturb the oxidation current responses for these isomers
in a consistent way, we may assume that the order in the
oxidation peak potentials tracks the relative stabilities of the
Ni(I) vs Ni(0) states for each of the isomers. The hydrogen
bonds between Ni and H−N in the endo configuration
(Scheme 1) stabilize the Ni(0) versus the Ni(I) state, resulting
in a shift of the Ni(I/0) oxidation to more positive potentials.
The potentials for each of the complexes are shown in Table 5.
The first Ni···H hydrogen bond results in a shift of 350−410
mV (i.e. the endo-exo isomer), whereas the second Ni···H
hydrogen bond (i.e. the endo-endo isomer) results in a shift of
230−290 mV.
The differences in oxidation potentials observed here are
larger than those observed for the previously reported complex
[Ni0(PtBu2NBn2H)2]2+, wherein the E1/2 of the endo-endo
isomer was found to be 170 mV positive of the E1/2 for the
endo-exo isomer, which is 320 mV positive of the E1/2 for the
exo-exo isomer (Table 5).16 To understand the origin of the
difference in oxidation potentials, DFT calculations were used
to compare the Ni···H−N hydrogen-bonding distances for the
endo-endo isomers of [Ni0(PCy2NMe2H)2]2+ and
[Ni0(PtBu2NMe2H)2]2+. A methyl substituent on the nitrogen
was used instead of benzyl to reduce the computational cost.
Figure 8. Optimized structure of the [NiII(PCy2NMe2H)2]2+ (left) and
[NiII(PtBu2NMe2H)2]2+ (right) endo-endo isomers. The red dotted lines
show the NH···Ni hydrogen bond distance (in Å). The red arrows
indicate the movement of the pendant amine and of the cyclohexyl
group upon protonation. The steric interaction between the methyl
group of the tert-butyl substituent and the CH2 of the P−CH2−N
linker is schematically indicated with black and yellow lines.
phosphorus atom of the ligand has to accommodate the
protonated pendant amine of the other ligand, resulting in the
phosphorus atom substituent moving away from the pendant
amine upon protonation. In [Ni0(PCy2NMe2H)2]2+ the smaller
size of the cyclohexyl substituent allows the protonated amine
to get closer to the metal center upon protonation, and the Ni−
N distance decreases from 3.40 Å in the nonprotonated
complex to 3.18 Å in the protonated complex. In contrast, the
larger tert-butyl substituent is not able to accommodate the
protonated pendant amine because of the steric clash between
one of the methyl groups with the CH2 groups of the P−CH2−
N linker (Figure 8). Therefore, upon protonation to form
[Ni0(PtBu2NMe2H)2]2+, the Ni−N distance slightly increases
from 3.39 to 3.42 Å. As a result, the NH···Ni distances are
reduced for the cyclohexyl-containing substituent (2.47 Å) in
comparison to the tert-butyl substituent (2.89 Å). This is
consistent with the experimentally observed increase in metal
6725
dx.doi.org/10.1021/om300409y | Organometallics 2012, 31, 6719−6731

Organometallics
Article
hydrogen bonding stability of the endo-containing isomers of
the [NiII(PCy2NBn‑R2H)2]2+ complexes. These results are also
consistent with previous studies that have shown that increasing
steric bulk on the ammonium moiety decreases the strength of
metal hydrogen bonding in M···H−N bonds (M = Pt(II),
Co(−I)), as the ammonium moiety is pushed farther from the
metal coordination sphere.26−29
turquoise curves, Figure 9), occurring slightly negative of the
II/I couple, and is similar to previous studies of the parent
complex in acetonitrile.10 In comparison to the parent complex,
the rate decreases slightly with the methyl ester substituent and
increases slightly with the methyl ether substituent (Table 4).
This is the expected trend on the basis of the relative ΔGH
2
values if hydrogen addition is the rate-limiting step in catalysis.
In acetonitrile, the cyclic voltammograms for [Ni-
(PCy2NBn‑OMe2)2]+ displayed similar behavior, although in this
solvent the rate of H2 oxidation was slightly slower than for the
parent complex, [Ni(PCy2NBn2)2]2+. Consistent with previous
studies,17 the magnitude of the catalytic wave increases upon
the addition of water, resulting in a 10−50% increase in rate
(Table S2 (Supporting Information)), and the onset shifts to
slightly more negative potentials.
Catalytic Oxidation of H2 by the [Ni(PCy Bn‑R2)2]
2
Complexes. All complexes were observed to be catalytically
active for H2 oxidation, where the rate was measured both in
benzonitrile and acetonitrile with added water, using triethyl-
amine as the base (Figures 9 and 10, Figure S5 (Supporting
In acetonitrile, [Ni(PCy2NBn‑COOMe2)2]2+ exhibits a somewhat
more complex catalytic wave (Figure S5 (Supporting
Information)). In the presence of H2, the wave shows catalysis
at the endo-endo isomer (offscale), as was observed in
benzonitrile. Following addition of triethylamine under H2,
the catalytic current continues to increase at the endo-endo and
endo-exo isomer, but not at the Ni(II/I) couple as is seen for
the other complexes.30 Upon addition of water, the current
increase at ∼-0.3 V (the endo-exo isomer) becomes larger, with
the onset shifting to less positive potentials as more water is
added, resulting in a rate of ∼2 s−1 at the maximum current.
This wave corresponds to a catalytic pathway different from
that observed for the other complexes in the presence of H2,
triethylamine, and water.
Figure 9. CV's for [Ni(PCy2NBn‑OMe2)2] (0.75 mM) in PhCN with 0.2
M NBu4PF6, showing catalytic activity at the Ni(II/I) couple upon the
addition of H2 and Et3N. Data were taken at 50 mV s−1 and referenced
Hydrogen addition is calculated to be most favorable for
[Ni(PCy2NBn‑OMe2)2]2+ and least favorable for [Ni-
(PCy2NBn‑COOMe2)2]2+ (Table 4). In this series of complexes,
the rates of catalytic hydrogen oxidation measured in
+
vs FeCp2/FeCp2 using the potential of the Ni(I/0) couple.
benzonitrile have the expected (slight) increase as ΔGH
2
becomes more favorable and, relatedly, the amine pKa of
[Ni0(PCy2NBn‑R2H)2]2+ increases. However, in acetonitrile, the
methyl ether containing complex [Ni(PCy2NBn‑OMe2)2]2+,
expected to be fastest, is slower than [Ni(PCy2NBn2)2]2+. This
result is unexpected if the addition of H2 is the rate-limiting
step and may suggest that there is another step in the catalytic
cycle, such as deprotonation by the external base, that is
controlling the rate.
Information)), and Table 4). In benzonitrile, the large increase
in current at the potential assigned to the endo-endo isomer
observed in the absence of base (red curve, Figure 9) is scan
rate independent (Figure S6 (Supporting Information)) and is
indicative of catalysis occurring at the I/0 oxidation of the
endo-endo isomer of [Ni(PCy2NBn‑R2H)2]2+, as discussed in
detail below. Upon the addition of triethylamine, the onset of
catalysis shifts to more negative potentials (green, purple, and
Figure 10. CV's for [Ni(PCy2NBn‑Ala‑OMe2)2]2+ in PhCN (left) and MeCN with added water (right) with 0.2 M NBu4PF6. Data were taken at 50 mV/s
and referenced vs Fc/Fc+ using [CoCp*2](PF6); titrated with Et3N; [cat.] = 0.75 mM.
6726
dx.doi.org/10.1021/om300409y | Organometallics 2012, 31, 6719−6731

Organometallics
Article
Upon addition of triethylamine to the amino acid ester
containing complexes, the catalytic current increases in a
manner similar to that for the other complexes, suggesting a
similar mechanism (Figure 10). However, for the amino acid
ester containing complexes, the catalytic wave continues to
slope to increasingly negative currents rather than plateau
(Figure 10, left). This is consistent with H2 oxidation at the
Ni(I/0) endo-endo and endo-exo isomers as well as at the
Ni(II/I) couple. In acetonitrile, the addition of water results in
a catalytic wave at the Ni(II/I) couple which plateaus (Figure
10, right). This is thought to be due to the ability of water to
facilitate deprotonation of the endo-endo isomer to form the
hydride.17 The reported catalytic rates were determined by
measuring the icat. value at the II/I couple.
confirming that water is able to act as a base to facilitate
catalysis.
[Ni(PCy2NBn‑Ala‑OMe2)2]2+ and [Ni(PCy2NBn‑Phe‑OMe2)2]2+ have
ΔGH values similar to that of [Ni(PCy2NBn‑COOMe2)2]2+, and the
2
complexes have comparable rates of hydrogen oxidation. While
the 31P NMR spectroscopy results suggest that structural
rearrangements may be hindered by the steric bulk of the
amino acid ester complexes, these rates are still much faster
than catalysis and are likely not contributing to rates that are
slower than that for the parent complex. These results are
expected if H2 addition is the rate-limiting step and suggest that
the observed modifications in rate are explained by a second
coordination sphere effect (pKa of the pendant amine), rather
than effects in the outer coordination sphere.
The outer coordination sphere in enzymes is thought to
influence catalytic activity, due to both steric and electronic
factors. For example, Happe et al. found that amino acids that
interact with the dithiolate ligand in [FeFe]-hydrogenase
strongly impact the structure and activity of the enzyme.31
Additionally, hydrogen bonding by outer coordination sphere
serine and lysine residues to the active site CN− ligands and
steric effects from phenylalanine and other hydrophobic
residues are hypothesized to prevent rearrangement of the
[2Fe] subcluster during the catalytic cycle.32,33 For the
Figure 11. CV's for [Ni(PCy2NBn‑OMe2)2]+ in MeCN titrated with H2O,
showing that water is able to act as a base for the oxidation of H2,
albeit operating at a more positive potential. The wave at −1.94 V is
+/0
due to CoCp*2
.
At low water concentrations, the current enhancement was
located at the potential of the endo-endo isomer. At
intermediate water concentrations, a current enhancement
could also be seen at the oxidation potential of the endo-exo
isomer of [Ni(PCy2NBn‑OMe2H)2]2+ (Figure 11). At the highest
concentrations of water (4.5 M) the current enhancement
occurs only near the endo-exo oxidation potential, giving a
maximum ic/ip ratio of 8.5. This shift in potential can be
explained if isomerization from the endo-endo isomer to the
endo-exo isomer occurs faster than catalysis at these
concentrations of water, resulting in the oxidation and
subsequent deprotonation by water of the endo-exo isomer.
The maximum rate for H2 oxidation using water as a base was
8.8 s−1, similar to the rate when triethylamine is used as a base.
Upon addition of triethylamine to this solution, the catalytic
wave shifts cathodically to the Ni(II/I) couple, as expected in
the presence of a strong base (Figure S7 (Supporting
Information)).
Ni(PR NR′2)2 catalysts, a steric influence that could enhance
2
catalysis would be one which forced an endo conformation and
limited boat−chair conformational isomerizations of the Ni−
P−C−N−C−P chelate rings, a role which the protein scaffold
may serve in hydrogenases. While the amino acid ester ligands
here did slow boat to chair isomerizations, they were still orders
of magnitude faster than catalysis. This is likely due to a lack of
positioning of the appended amino acids, which undergo
unrestricted conformational movements. A well-positioned and
structurally less flexible outer coordination sphere such as that
seen in enzymes is necessary to achieve the positioning which
could result in the desired steric or electronic effects on the
active site.
Role of Water As Base. For the Ni complexes containing
methyl ester (Figure S5 (Supporting Information)), amino acid
ester (Figure 10), and methyl ether groups (Figures 7 and 9), it
was observed that under H2 a scan-rate-independent (Figure S6
(Supporting Information)) current enhancement was observed
at the Ni(0/I) oxidation for the endo-endo isomer, even in the
absence of added base. The most likely source of an external
base is residual water, and to test whether water could act as a
base at this potential, CV’s were collected as aliquots of water
were added to [Ni(PCy2NBn‑OMe2)2]+ in acetonitrile under H2
(Figure 11). As water was added, a scan-rate-independent
current enhancement at the endo-endo isomer was observed,
Water is unlikely to act as a base for [Ni(PCy2NBn‑OMe2H)2]2+
because of its low pKa value. Therefore, in contrast to the
situation where a strong base is used (Figure 9), oxidation of
[Ni(PCy2NBn‑OMe2H)2]2+ must occur before proton transfer.
The pendant amine of the resulting oxidized product,
[Ni(PCy2NBn‑OMe2H)2]3+, is expected to have a lower pKa,14
allowing water to remove a proton. Therefore, the sequence of
events when water acts as a base is altered from the sequence
shown in Scheme 1 and could be (1) e− H+ e− H+, (2) e− H+
H+ e−, or (3) e− e− H+ H+.
Similar current enhancements were seen upon titration of
[Ni(PCy2NBn2)2]+ and [Ni(PCy2NBn‑COOMe2)2]+ with water
under H2 (2.9 and 1.7 s−1, respectively), suggesting that using
water as a base for catalysis should be generalizable to this class
of hydrogen oxidation catalysts. However, the fastest rates were
observed with [Ni(PCy2NBn‑OMe2)2]2+. It is possible that the
methoxy group contained by the ligand assists in deprotonation
of the pendant amine by water. These results are extremely
encouraging for the development of molecular first row
transition metal containing H2 oxidation catalysts that use
6727
dx.doi.org/10.1021/om300409y | Organometallics 2012, 31, 6719−6731

Organometallics
Article
was then cooled to room temperature and brought into a glovebox
under a nitrogen atmosphere. The white precipitate was collected on a
fritted funnel by vacuum filtration, washed with 3 mL of additional
ethanol, and then dried under vacuum. Yield: 170.4 mg, 79%. 31P{1H}
water, an economical and environmentally friendly reagent, as a
base.
CONCLUSIONS
1
NMR (CD2Cl2): δ −60 ppm, broad. H NMR (CD2Cl2): δ (ppm)
■
7.96 (4H, d, J = 8 Hz, Ar-H), 7.44 (4H, d, J = 8 Hz, Ar-H), 3.96 (4H, s,
NCH2Ph), 3.88 (6H, s, CO2CH3), 3.13 (8H, s, PCH2N), 0.9−1.7 (22
In the derivatives studied here, rate enhancements were not
observed for catalysts derivatized with amino acid esters.
Although conformational flexibility was reduced for these
catalysts, rates for these processes were orders of magnitude
faster than catalysis and the slight variations in rate observed
could be explained by electronic effects in the second
coordination sphere. Several mechanistic observations will be
important to consider for future catalyst design. For example,
the sequence of catalytic events can be altered by using a weak
base, deprotonation may be the rate-limiting step in some cases,
and isomerization could control the potential at which the
catalyst operates.
H, m, Cy-H). APCI MS: m/z (PCy2NBn‑COOMe + H+) 611.319 (calcd
2
611.317).
PCy2NBn‑COO‑Li+2. PCy2NBn‑COOMe2 (250 mg, 0.4 mmol) was dissolved
in 16.9 mL of THF in a Schlenk flask. MeOH (8.4 mL) and 1 M
aqueous LiOH (2.4 mL) were added, and the reaction mixture was
heated to 50 °C for 5 h. The reaction mixture was then cooled and
transferred to a glovebox. The white precipitate product was collected
on a fritted funnel by vacuum filtration, washed with additional THF
(∼5 mL), and then dried under vacuum. Yield: 154.9 mg, 65%.
1
31P{1H} NMR (D2O): δ −68 ppm. H NMR (D2O): δ (ppm) 7.83
(4H, d, J = 8 Hz, Ar-H), 7.35 (4H, d, J = 8 Hz, Ar-H), 3.98 (4H, s,
NCH2Ph), 3.17 (4H, d, J = 14.5 Hz, PCH2N), 3.06 (4H, d, J = 14.5
Hz, PCH2N), 1.54−0.82 (22H, m, Cy-H). ESI MS: m/z
Ni(PR NR′
)
hydrogen production catalysts containing
2
2
2
(PCy2NBn‑COO‑Li+ − 2 Li+ + H+) 581.270 (calcd 581.271).
amino acids or amino acid esters in the outer coordination
sphere showed rates several times faster than that of the parent
catalyst.18 This may be due to the ability of these groups to
create a localized increase in proton concentration around the
catalyst due to their protic nature or to their ability to increase
water concentration around the catalyst.34 A localized increase
in protons would not be expected to facilitate hydrogen
oxidation, but rates can potentially be increased by adding
Brønsted bases in the outer coordination sphere to facilitate the
shuttling of protons from the active site to an external base. As
deprotonation becomes rate limiting, observed in one catalyst
reported here, a proton channel will be an essential feature to
enhance catalytic rates. The complexes developed here provide
the foundation to attach such a coordination sphere.
2
[Ni(PCy2NBn‑COOMe2)2](BF4)2. PCy2NBn‑COOMe (100 mg, 0.164
2
mmol) and [Ni(CH3CN)6](BF4)2 (39.2 mg, 0.082 mmol) were
combined in ∼3 mL of MeCN and stirred overnight. Recrystallization
by slow diffusion of ether into the MeCN solution gave dark reddish
purple blocklike crystals in quantitative yield. 31P{1H} NMR
1
(CH3CN): δ 10 ppm. H NMR (D2O): δ (ppm) 8.01 (8H, d, J = 8
Hz, Ar-H), 7.31 (8H, d, J = 8 Hz, Ar-H), 3.88 (12 H, s, CO2CH3), 3.66
(4H, d, J = 13 Hz, NCH2Ph), 3.43 (4H, d, J = 13 Hz, NCH2Ph), 3.10
(4H, d, J = 13 Hz, PCH2N), 2.90 (4H, d, J = 12 Hz, PCH2N), 2.79
(4H, d, J = 13 Hz, PCH2N), 2.61 (4H, d, J = 11 Hz, PCH2N), 2.14−
1.05 (m, 44H, Cy-H). Anal. Calcd for C68H96B2F8N4NiO8P4: C, 56.18;
H, 6.66; N, 3.85. Found: C, 56.10; H, 6.87; N, 3.88. ESI MS: m/z
{[Ni(PCy2NBn‑COOMe2)2](BF4)+} 1365.554 (calcd 1365.556).
Ni(PCy2NBn‑COOMe2)2. Tetramethylguanidine (19.8 mg, 0.1722
mmol) was added to [Ni(PCy2NBn‑COOMe2)2](BF4)2 (119.2 mg, 0.082
mmol) in ∼3 mL of MeCN. H2 was then bubbled through the solution
for 10 min. The reaction mixture was allowed to sit for several days,
during which time Ni(PCy2NBn‑COOMe2)2 precipitated as a yellow solid
(58 mg, 49% yield). When the reaction was run on a smaller scale
EXPERIMENTAL SECTION
■
General Procedures. Solution-state 1H and 31P NMR spectra
1
were recorded on Varian VNMR spectrometers (500 or 300 MHz H
(0.0318 mmol of [Ni(PCy2NBn‑COOMe2)2]2+), Ni(PCy2NBn‑COOMe
)
1
frequency). All H chemical shifts were internally calibrated to the
2
2
precipitated as X-ray-quality yellow-orange crystals. 31P{1H} NMR
(CH2Cl2): δ 11.1 ppm. 1H NMR (THF-d8): δ (ppm) 7.93 (8H, d, J =
6.5 Hz, Ar-H), 7.16 (8H, d, J = 7.5 Hz, Ar-H), 3.91 (12H, s, CO2CH3),
3.71 (8H, s, NCH2Ph), 2.67 (8H, d, J = 10.5 Hz, PCH2N), 2.41 (8H,
d, J = 11.5 Hz, PCH2N), 1.77−1.14 (44H, m, Cy-H). Anal. Calcd for
C68H96N4NiO8P4: C, 63.80; H, 7.56; N, 4.38. Found: C, 63.78; H,
7.60; N, 4.43.
monoprotic solvent impurity. Elemental analyses were performed by
Atlantic Microlab, Norcross, GA, with V2O5 as a combustion catalyst.
Electrospray ionization (ESI) and atomic pressure chemical ionization
(APCI) were collected at the Indiana University Mass Spectrometry
Facility on a Waters/Micromass LCT Classic using anhydrous solvents
and inert-atmosphere techniques.
Synthesis and Materials. All reactions were performed using
standard Schlenk techniques or in a glovebox under a N2 atmosphere.
Deoxygenated acetonitrile, diethyl ether, and THF were dried using an
Innovative Technology, Inc. PureSolv solvent purification system.
Ethanol, methanol, water, and triethylamine used in syntheses were
not dried but were degassed by sparging with N2 overnight prior to
use. O-(Benzotriazol-1-yl)-N,N,N′,N′-tetramethyluronium tetrafluoro-
borate (TBTU), L-alanine methyl ester hydrochloride (H-Ala-
OMe·HCl), and L-phenylalanine methyl ester hydrochloride (H-Phe-
OMe·HCl) were purchased from EMD Chemicals, Inc. Methyl 4-
(aminomethyl)benzonate hydrochloride, benzylamine, and p-methox-
ybenzylamine were purchased from SigmaAldrich. Liquid reagents
were deoxygenated by sparging with N2 for several hours, and solid
reagents were placed under alternating cycles of vacuum and N2 before
use. CyP(CH2OH)2 was synthesized according to a literature
procedure,21 as was [Ni(CH3CN)6](BF4)2.35
PCy2NBn‑Ala‑OMe2. PCy2NBn‑COO‑Li+2 (50 mg, 0.084 mmol) and TBTU
(59.4 mg, 0.185 mmol) were stirred together for 15 min in ∼2.5 mL of
CH2Cl2. H-Ala-OMe·HCl (24.7 mg, 0.176 mmol) and triethylamine
(0.03 mL) were combined in a separate flask in ∼5 mL of CH2Cl2, and
then added to the P2CyNBn‑COO‑Li+ and TBTU in a dropwise manner.
2
The mixture was stirred for an additional 20 h, followed by collection
of the white precipitate on a fritted glass funnel. The product was
washed with water (10 mL) and dichloromethane (10 mL) and dried
under vacuum. Yield: 54.5 mg, 86%. 31P{1H} NMR (THF): δ −60.3
1
ppm. H NMR (THF-d8): δ (ppm) 7.81 (4H, d, J = 7.5 Hz, Ar-H),
7.74 (2H, d, J = 7.5 Hz, amide-H), 7.41 (4H, d, J = 7.5 Hz, Ar-H), 4.70
(2H, m, C∝-H), 4.00 (4H, s, NCH2Ph), 3.67 (6H, s, CO2CH3), 3.19
(8H, s, PCH2N), 1.60 (6H, s, J = 7.2 Hz, C∝CH3), 1.87−1.00 (22H,
m, Cy-H). APCI MS: m/z (PCy2NBn‑Ala‑OMe + H+) 753.391 (calcd
2
753.391).
[Ni(PCy2NBn‑Ala‑OMe2)2](BF4)2. PCy2NBn‑Ala‑OMe (55.5 mg, 0.074
PCy2NBn‑COOMe2. Methyl 4-(aminomethyl)benzonate hydrochloride
(143.2 mg, 0.71 mmol) and triethylamine (0.11 mL, 0.78 mmol) were
combined in a Schlenk flask with 5 mL of ethanol and heated to 65 °C
until the reagents had dissolved. CyP(CH2OH)2 (125.0 mg, 0.71
mmol) dissolved in 3 mL of ethanol was then added dropwise over 25
min. The reaction mixture was further heated at 65 °C for 22 h, during
which time the product appeared as a white precipitate. The reaction
2
mmol) and [Ni(CH3CN)6](BF4)2 (17.6 mg, 0.037 mmol) were
combined in 2 mL of MeCN and stirred for 3 days. The deep purple
mixture was then filtered, and the material in the filtrate precipitated
by addition of diethyl ether (∼20 mL). The purple solid was collected
on a frit, washed with additional diethyl ether (∼10 mL), and then
dried under vacuum for several hours. Yield: 40 mg, 72%. 31P{1H}
6728
dx.doi.org/10.1021/om300409y | Organometallics 2012, 31, 6719−6731

Organometallics
Article
1
NMR (CD3CN): δ 0 ppm (broad). H NMR (CD3CN): δ (ppm)
7.86 (8H, d, J = 8 Hz, Ar-H), 7.68 (4H, d, J = 6.5 Hz, amide-H), 7.23
(8H, d, J = 7.5 Hz, Ar-H), 4.60 (4H, m, C∝-H), 3.73 (12H, s,
CO2CH3), 3.68 (8H, d, J=7 Hz, NCH2Ph), 3.02 (4H, d, J = 13 Hz,
PCH2N), 2.91 (4H, d, J = 12.5 Hz, PCH2N), 2.81 (4H, dd, J = 4.5 Hz,
13 Hz, PCH2N), 2.65 (4H, d, J = 11.5 Hz, PCH2N), 1.52 (12H, m,
C∝CH3), 2.02−1.11 (44 H, m, Cy-H). ESI MS: m/z {[Ni-
(PCy2NBn‑Ala‑OMe2)2](BF4)+}1649.700 (calcd 1649.706).
8.7 Hz, Ar-H), 6.93 (8H, d, J = 8.7 Hz, Ar-H), 3.79 (12H, s, OCH3),
3.61 (4H, d, J = 12.9 Hz, NCH2Ph), 3.45 (4H, d, J = 12.6 Hz,
NCH2Ph), 3.09 (4H, d, J = 13.8 Hz, PCH2N), 2.88 (4H, d, J = 12.6
Hz, PCH2N), 2.69 (4H, d, J = 13.5 Hz, PCH2N), 2.55 (4H, d, J = 12.5
Hz, PCH2N), 1.97−0.99 (44H, m, Cy-H). Anal. Calcd for
C67H100.5B2F8N5.5NiO4P4 ([Ni(PCy2NBn‑OMe2)2](BF4)2 + 1.5CH3CN):
C, 57.35; H, 7.22; N, 5.49. Found C, 57.22; H, 7.44; N, 5.46.
pKa Measurements. The pKa values for the pendant amine of
PCy2NBn‑Phe‑OMe2. PCy2NBn‑COO‑Li+2 (50 mg, 0.084 mmol) and TBTU
(59.4 mg, 0.185 mmol) were stirred together for 15 min in ∼4 mL of
CH2Cl2. H-Phe-OMe·HCl (90.6 mg, 0.42 mmol) and triethylamine
(0.07 mL) in ∼3 mL of CH2Cl2 were then added. The reaction
mixture was stirred for an additional 18 h at room temperature and
was then filtered. The solvent was removed under vacuum, and the
resulting residue was washed with H2O and MeCN (20 mL each) and
then dried under vacuum, giving a yield of 19.3 mg (25% yield).
[Ni(PCy2NR′2H)2]+, where R′ = Bn-COOMe, Bn-OMe, were
determined by first vortexing [Ni(PCy2NBn‑R′2)2] and 1.1 equiv of p-
anisidinium tetrafluoroborate in CD3CN to form the nickel hydride,
with a concentration <10 mM. This solution was then titrated with a
solution of p-anisidinium tetrafluoroborate in CD3CN, and the 1H and
31P NMR spectra were recorded after the addition of each aliquot
(∼0.1 equiv). The pKa was calculated using eq 1
1
31P{1H} NMR (CD2Cl2): δ 59 ppm (broad). H NMR (CD2Cl2): δ
−
pKa = pKa(anisidine) − log([HB+][A ]/[B][HA])
(1)
(ppm) 7.66 (4H, d, J = 8.4 Hz, Ar-H), 7.42 (4H, d, J = 8.4 Hz, Ar-H),
7.28 (6H, m, phenylalanine-H), 7.16 (4H, m, phenylalanine-H), 6.55
(2H, d, J = 7.5 Hz, amide-H), 5.02 (2H, ddd, J = 11.55 Hz, 6 Hz, 1.8
Hz, C∝-H), 3.94 (4H, s, NCH2Ph), 3.75 (6H, s, CO2CH3), 3.28 (2H,
dd, J = 8.1 Hz, 5.7 Hz, C∝CH2Ph), 3.19 (2H, dd, J = 8.1 Hz, 5.7 Hz,
C∝CH2Ph), 3.12 (8H, s, PCH2N), 0.90−1.65 (22 H, m, Cy-H).
where A− = [HNi(PCy2NBn‑R2)2]+, HA = [Ni(PCy2NBn‑R2H)2]2+, HB+ =
p-anisidinium, B = anisidine, and [HB+]/[B] = (δobs − δB)/(δHB
δobs).
−
+
The pKa of the pendant amine for [Ni(PCy2NBn‑R2H)2]2+, where R =
−COOMe, −Ala-OMe, −Phe-OMe, were calculated in the reverse
direction: [Ni(PCy2NBn‑R2)2]2+ was dissolved in CD3CN. The solution
was sparged with H2 to form [Ni(PCy2NBn‑R2H)2]2+ and then titrated
with p-anisidine to form [HNi(PCy2NBn‑R2)2]+. The pKa was then
calculated using the above equations. In the case of [Ni-
(PCy2NBn‑Ala‑OMe2)2]2+ and [Ni(PCy2NBn‑Phe‑OMe2)2]2+, the nickel hydride
appeared upon addition of H2, before addition of the base, likely due
either to residual H2O from the synthesis or the ligand itself acting as a
base. This results in an underestimation of the pKa value, with an
estimated error of 0.5 pKa units. Attempts to thoroughly dry these
two complexes by recrystallizing, washing with diethyl ether, and
placing the complexes under vacuum did not reduce the amount of
hydride observed initially.
[Ni(PCy2NBn‑Phe‑OMe2)2](BF4)2. PCy2NBn‑Phe‑OMe (19.3 mg, 0.021
2
mmol) and [Ni(CH3CN)6](BF4)2 (5.1 mg, 0.011 mmol) were
combined in ∼1 mL of acetonitrile. The reaction mixture was stirred
for 3 days and then filtered. The product was obtained by removal of
the solvent from the filtrate under vacuum. Yield: 0.0216 g, 99%.
31P{1H} NMR (CD3CN): δ 10 ppm (very broad). 1H NMR
(CD3CN): δ (ppm) 7.71 (8H, d, J = 7.5 Hz, Ar-H), 7.64 (4H, m,
amide-H), 7.15−7.26 (28H, m, Ar-H), 4.87 (4H, m, C∝-H), 3.68 (8H,
m, NCH2Ph), 3.64 (12H, s, CO2CH3), 3.19 (8H, m, C∝CH2Ph),
2.98−2.59 (16H, m, PCH2N), 2.10−0.88 (44H, Cy-H). ESI MS: m/z
[Ni(PCy2NBn‑Phe‑OMe2)2]2+ 933.896 (calcd 933.915).
Ni(PCy2NBn‑OMe2)2. CyP(CH2OH)2 (125.0 mg, 0.71 mmol) in 6 mL
of ethanol was added dropwise to 4-methoxybenzylamine in 6 mL of
ethanol at 65 °C. The reaction mixture was stirred at 65 °C overnight.
The reaction mixture was then cooled to room temperature, and the
ethanol was removed under vacuum. The crude ligand was combined
with [Ni(CH3CN)6](BF4)2 (84.9 mg, 0.1775 mmol) in ∼5 mL of
acetonitrile and stirred for 1 week. The reaction mixture was filtered,
and tetramethylguanidine (45.0 mg, 0.3905 mmol) was added to the
filtrate. Hydrogen was bubbled through the filtrate, and the reaction
mixture was allowed to sit for 4 days, during which time
The pKa for the metal hydride of [HNi(PCy2NBn‑OMe2)2]+ was
measured by titration of a benzonitrile solution of [Ni(PCy2NBn‑OMe2)2]
with N,N,N′,N′-tetramethylguanidinium tetrafluoroborate
1
(HTMG·BF4), where the H and 31P NMR spectra of the solution
were recorded after the addition of each aliquot (∼0.1 equiv).
Benzonitrile was used, since [Ni(PCy2NBn‑OMe2)2] is not soluble in
acetonitrile. The pKa was calculated in a fashion analogous to that used
for the pendant amine. Specifically, using eq 2
Ni(PCy2NBn‑OMe
) precipitated as a yellow solid. The product was
2
−
2
pKa = pKa(TMG) − log([HB+][A ]/[B][HA])
(2)
washed with additional acetonitrile (∼20 mL) and dried under
vacuum. Yield: 65.1 mg, 31.4%. 31P{1H} NMR (THF): δ (ppm) 10.1.
1H NMR. (THF-d8): δ (ppm) 6.97 (8H, d, J = 8.5 Hz, Ar-H), 6.80
(8H, d, J = 8.5 Hz, Ar-H), 3.79 (s, 12H, OCH3), 3.58 (s, 8H,
NCH2Ph), 2.68 (8H, d, J = 9.5 Hz, PCH2N), 2.34 (8H, d, J = 11 Hz,
PCH2N), 1.77−1.11 (m, 44H, Cy-H). Anal. Calcd for
C64H96N4NiO4P4: C, 65.81; H, 8.28; N, 4.80. Found: C, 65.78; H,
8.25; N, 4.75.
where A− = Ni(PCy2NBn‑OMe2)2, HA = [HNi(PCy2NBn‑OMe2)2]+, HB+ =
HTMG+, B = TMG, and [HB+]/[B] = (δobs − δB)/(δHB − δobs).
+
Electrochemical Studies. All cyclic voltammetry experiments
were carried out on a computer-aided CH Instruments 1100 A
potentiostat in a glovebox with an N2 atmosphere in 0.2 M
−
nBu4N+PF6 /acetonitrile or benzonitrile solutions at a scan rate of
50 mV/s, unless otherwise noted. The working electrode used was a
glassy-carbon disk, and the counter electrode was a glassy-carbon rod.
A silver wire was used as a pseudo reference electrode.
[Ni(PCy2NBn‑OMe2)2](BF4)2. [Ni(PCy2NBn‑OMe2)2] (51.0 mg, 0.044
+
−
mmol) and ferrocenium tetrafluoroborate (FeCp2 BF4 ) (24.1 mg,
0.088 mmol) were weighed into separate vials. Acetonitrile (1 mL) was
added to each vial, and the FcBF4 solution was then added to the
Ni(PCy2NBn‑OMe2)2 suspension, and 1 mL of additional acetonitrile was
used for transfer. The reaction mixture was stirred overnight, during
which time the majority of the solid dissolved. The solution was then
concentrated until ferrocene began precipitating as an orange solid.
The reaction mixture was stored at −35 °C overnight. The mother
liquor was removed by pipet, concentrated to dryness, and slurried
with diethyl ether overnight. The ether was decanted and the solid
dissolved in acetonitrile, followed by recrystallization by slow diffusion
of ether into the solution. The product crystallized as red-purple
needlelike crystals (30.0 mg, 51.2% yield). Recrystallization by slow
diffusion of ether into an acetonitrile solution of the product on a
smaller scale yielded X-ray-quality crystals. 31P{1H} NMR (CD3CN):
+
−
Decamethylcobaltocenium hexafluorophosphate (CoCp*2 PF6
,
+/0
−1.94 V vs FeCp2
in acetonitrile) or cobaltocenium hexafluor-
+
−
+/0
ophosphate (CoCp2 PF6 , −1.33 vs FeCp2 in acetonitrile) were
used as internal standards to reference the potentials of the Ni(II/I)
and Ni(I/0) redox couples to the ferrocenium/ferrocene couple. The
benzonitrile used for electrochemical measurements was purchased
from SigmaAldrich and filtered through activated alumina before use.
The triethylamine used for electrochemical measurements was distilled
over CaH2 and then degassed using the freeze−pump−thaw method.
Three trials were run for each type of catalyst in both benzonitrile and
acetonitrile, and the rates from each trial were averaged. At least one
trial in each set was run with either CoCp2+ or CoCp*2+ as an internal
reference, and at least one trial in each set was run without addition of
an internal reference in order to ensure that the internal reference did
not influence the rates measured. [NiI(PCy2NBn‑OMe2)2](BF4) for
1
δ (ppm) 9.2 (broad). H NMR (CD3CN): δ (ppm) 7.11 (8H, d, J =
6729
dx.doi.org/10.1021/om300409y | Organometallics 2012, 31, 6719−6731

Organometallics
Article
electrochemical measurements was synthesized by stirring 1 equiv each
of [Ni0(PCy2NBn‑OMe2)2] and [NiII(PCy2NBn‑OMe2)2](BF4)2 in the
electrolyte-containing acetonitrile solution immediately before meas-
urement.
Ab Initio Calculations. Ab initio calculations were performed in a
manner similar to those previously reported.16 The structures of the
[Ni0(PCy2NMe2H)2]2+ endo-endo isomer and of [Ni0(PCy2NMe2)2]
complex were optimized without symmetry constraints using the
B3P86 functional.46,47 The Stuttgart basis set with effective core
potential was used for the Ni atom, whereas the 6-31G* basis set was
used for all of the other atoms with one additional p polarization
function for the proton on the pendant amines. The optimized
structures were confirmed by frequency calculations. All of the
calculations were carried out with the program Gaussian 09.48
Calculations. For the majority of the [HNi(PCy2NBn‑R2)2]+
complexes, the hydride pKa values (pKa(NiH)) were calculated
using a correlation between the pKa(NiH) and the E1/2I/0 derived from
several similar [HNi(PR NR′2)2]+ complexes:36
2
I/0
pKa(Ni−H) = − 18.6 × E1/2 − 2.65
(3)
36
II/I
−
−
ΔGH was calculated using a correlation between ΔGH and E1/2
:
ASSOCIATED CONTENT
■
II/I
−
ΔGH (kcal/mol) = 20.8 × E1/2 + 76.8
S
(4)
* Supporting Information
CIF files giving crystallographic data for [Ni-
(PCy2NBn‑OMe2)2](BF4)2, [Ni(PCy2NBn‑COOMe2)2](BF4)2, [Ni-
(PCy2NBn‑OMe2)2], [Ni(PCy2NBn2)2], and [Ni(PCy2NBn‑COOMe2)2]
and figures and tables giving eedox characteristics of [Ni-
(PCy2NBn‑R2)2] complexes, turnover frequencies for the catalysts
with and without the addition of water, 1H NMR of
P2CyNBn‑Ala‑OMe2, scan rate dependence of peak current for the
oxidation of the endo-endo isomer of [Ni0(PCy2NBn‑R2H)2]2+,
electrocatalysis of [Ni(PCy2NBn‑COOMe2)2]+ in MeCN and
[Ni(PCy2NBn‑Phe‑OMe2)2]2+ in PhCN, electrochemistry of [Ni-
(PCy2NBn‑OMe2)2](BF4)2 with water followed by triethylamine,
31P NMR spectra of [Ni(PCy2NBn‑Phe‑OMe2)2]2+, 31P−31P COSY
spectra of [Ni(PCy2NBn‑Phe‑OMe2)2]2+ in CD3CN taken at −40
°C, and variable-temperature 31P NMR spectroscopy of
[HNi(PCy2NBn‑Phe‑OMe2)2]+. This material is available free of
charge via the Internet at http://pubs.acs.org.
ΔG was calculated using the thermodynamic cycle for H2 addition
H2
described previously:10
−
ΔGH (kcal/mol) = ΔGH + 1.364
2
pKa([Ni0(PCy2NBn−R2H)2]2+) − 76
(5)
In the case of [HNi(PCy2NBn‑OMe2)2]+, the pKa was also
experimentally measured (described above). Using the measured
pKa, ΔG(H−) was calculated using the thermodynamic cycle described
in Table 6.
Table 6. Calculation of ΔG(H−)
−
ΔGH
step
(kcal/mol)
[HNiII(PCy2NBn‑OMe2)2]+ → [Ni0(PCy2NBn‑OMe2)2] + H+
1.364 ×
pKa(NiH)37
[Ni0(PCy2NBn‑OMe2)2] → [NiI(PCy2NBn‑OMe2)2]+ + e−
[NiI(PCy2NBn‑OMe2)2]+ → [NiII(PCy2NBn‑OMe2)2]2+ + e−
H+ + 2e− → H−
23.06 ×
37
EI/0
AUTHOR INFORMATION
1/2
■
23.06 ×
37
EII/I
Corresponding Author
*E-mail: wendy.shaw@pnnl.gov.
1/2
79.638
Present Address
[HNiII(PCy2NBn‑OMe2)2]+ → [NiII(PCy2NBn‑OMe2)2] + H−
†Department of Chemistry, Indiana University of Pennsylvania,
Indiana, PA.
Notes
X-ray Crystallography. Crystals were mounted on a MiTeGen
MicroMounts pin using Paratone-N oil and cooled to the data
collection temperature (173 K for [Ni(PCy2NBn‑COOMe2)2], [Ni-
The authors declare no competing financial interest.
(PCy2NBn‑OMe2)2], and [Ni(PCy2NBn2)2]; 120
K for [Ni-
ACKNOWLEDGMENTS
(PCy2NBn‑OMe2)2](BF4)2 and [Ni(PCy2NBn‑OMe2)2](BF4)2). Data were
collected on a Bruker-AXS II CCD diffractometer with 0.710 73 Å Mo
Kα radiation. Cell parameters were retrieved using Bruker APEX II
software,39 raw data were integrated using SAINTPlus,40 and
absorption correction was applied using SADABS.41 The structures
were solved using either direct methods or the Patterson method and
refined by a least-squares method on F2 using the SHELXTL program
package. Space groups were chosen by analysis of systematic absences
and intensity statistics.42
■
We thank Daniel DuBois for useful discussions. This work was
funded by the DOE Office of Science Early Career Research
Program through the Office of Basic Energy Sciences (S.L. and
W.S.), by the Center for Molecular Electrocatalysis, an Energy
Frontier Research Center funded by the U.S. Department of
Energy, Office of Science, Office of Basic Energy Sciences (M.-
H.H., S.C., S.R., J.A.S.R.), and by the U.S. DOE Basic Energy
Sciences, Chemical Sciences, Geoscience and Biosciences
Division (A.M.A., A.J., J.C.L.). Computational resources were
provided by the National Energy Research Scientific Comput-
ing Center (NERSC) at Lawrence Berkeley National
Laboratory. Pacific Northwest National Laboratory is operated
by Battelle for the U.S. Department of Energy.
NMR Peak Fitting. The data was analyzed using the procedure
described by O’Hagan et al.,23 using gNMR for variable-temperature
31P line shape analysis using a two-site exchange model.43
Experimental spectra were processed within gNMR using a Lorentzian
function with 4 Hz line broadening. Rates were determined by
iteration of the simulated line widths, using both manual and gNMR
algorithm methods. Visual inspection of the overlaid experimental and
simulated spectra were used to determine the best fit.
The data were fit with the nonlinearized Eyring equation using
Profit, which allowed the application of individual error bars
determined for each data point in calculating error. Error bars for
individual data points were estimated to be 10−15% on the basis of
visual inspection of the gNMR fits.44 Rate constants at the coalescence
temperatures Tc were calculated using the equation kc = πΔν/√2,
where kc is the rate constant at the coalescence temperature and Δν is
the line separation in the low-temperature limit.45
REFERENCES
■
(1) Fontecilla-Camps, J. C.; Volbeda, A.; Cavazza, C.; Nicolet, Y.
Chem. Rev. 2007, 107, 4273.
(2) Pershad, H. R.; Duff, J. L. C.; Heering, H. A.; Duin, E. C.;
Albracht, S. P. J.; Armstrong, F. A. Biochemistry 1999, 38, 8992.
(3) Frey, M. ChemBioChem 2002, 3, 153.
(4) Cornish, A. J.; Gartner, K.; Yang, H.; Peters, J. W.; Hegg, E. L. J.
Biol. Chem. 2011, 286, 38341.
6730
dx.doi.org/10.1021/om300409y | Organometallics 2012, 31, 6719−6731

Organometallics
Article
(5) Dementin, S.; Burlat, B.; Lacey, A. L. D.; Pardo, A.; Adryanczyk-
Perrier, G.; Guigliarelli, B.; Fernandez, V. M.; Rousset, M. J. Biol.
Chem. 2004, 279, 10508.
(6) Marshall, N. M.; Garner, D. K.; Wilson, T. D.; Gao, Y.-G.;
Robinson, H.; Nilges, M. J.; Lu, Y. Nature 2009, 462, 113.
(7) Pandey, A. S.; Harris, T. V.; Giles, L. J.; Peters, J. W.; Szilagyi, R.
K. J. Am. Chem. Soc. 2008, 130, 4533.
endo-endo species. Upon addition of water, the irreversible electro-
chemical event at −0.6 V remains.
(31) Knorzer, P.; Silakov, A.; Foster, C. E.; Armstrong, F. A.; Lubitz,
W.; Happe, T. J. Biol. Chem. 2012, 287, 1489.
(32) Singleton, M. L.; Bhuvanesh, N.; Reibenspies, J. H.;
Darensbourg, M. Y. Angew. Chem., Int. Ed. 2008, 47, 9492.
(33) Peters, J. W.; Lanzilotta, W. N.; Lemon, B. J.; Seefeldt, L. C.
Science 1998, 282, 1853.
(8) Rakowski DuBois, M.; DuBois, D. L. Chem. Soc. Rev. 2009, 38,
(34) Jain, A.; Reback, M. L.; Lindstrom, M. L.; Thogerson, C. E.;
Helm, M. L.; Appel, A. M.; Shaw, W. J. Inorg. Chem. 2012, 51, 6592.
(35) Heintz, R. A.; Smith, J. A.; Szalay, P. S.; Weisgerber, A.; Dunbar,
K. R. Inorg. Synth. 2002, 33, 75.
62.
(9) Gloaguen, F.; Rauchfuss, T. B. Chem. Soc. Rev. 2009.
(10) 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.
(11) Pool, D. H.; DuBois, D. L. J. Organomet. Chem. 2009, 694, 2858.
(12) Wilson, A. D.; Shoemaker, R. K.; Miedaner, A.; Muckerman, J.
T.; DuBois, D. L.; Rakowski DuBois, M. Proc. Natl. Acad. Sci. U.S.A.
2007, 104, 6951.
(13) Yang, J. Y.; Bullock, R. M.; Shaw, W. J.; Twamley, B.; Fraze, K.;
Rakowski DuBois, M.; DuBois, D. L. J. Am. Chem. Soc. 2009, 131,
5935.
(14) Appel, A. M.; Pool, D. H.; O’Hagan, M.; Shaw, W. J.; Yang, J. Y.;
Rakowski DuBois, M.; DuBois, D. L.; Bullock, R. M. ACS Catal. 2011,
1, 777.
(15) Helm, M. L.; Stewart, M. P.; Bullock, R. M.; Rakowski DuBois,
M.; DuBois, D. L. Science 2011, 333, 863.
(36) Galan, B. R.; Schoffel, J.; Linehan, J. C.; Seu, C.; Appel, A. M.;
̈
Roberts, J. A. S.; Helm, M. L.; Kilgore, U. J.; Yang, J. Y.; DuBois, D. L.;
Kubiak, C. P. J. Am. Chem. Soc. 2011, 133, 12767.
(37) Wayner, D. D. M.; Parker, V. D. Acc. Chem. Res. 1993, 26, 287.
(38) Ellis, W. W.; Raebiger, J. W.; Curtis, C. J.; Bruno, J. W.; DuBois,
D. L. J. Am. Chem. Soc. 2004, 126, 2738.
(39) Bruker APEX 2, version 2010.3-0; Bruker AXS Inc., Madison, WI,
2009.
(40) SAINTPlus: Data Reduction and Correction Program; Bruker AXS
Inc., Madison, WI, 2004; Vol. 723a.
(41) SADABS: an Empirical Absorption Correction Program, version
2001/1; Bruker AXS Inc., Madison, WI, 2001.
(42) Sheldrick, G. M. SHELXTL: Structure Determination Software
Suite, version 6.14; Bruker AXS Inc., Madison, WI, 2004.
(43) Budzelaar, P. H. M. gNMR 5.0; Adept Scientific, Herts, U.K.,
2006.
(44) QuantumSoft, version 6.2; Fit, P., Ed.; QuantumSoft, Zurich,
Switzerland, 2010.
(45) Braun, S.; Kalinowski, H.-O.; Berger, S. 150 and More Basic
NMR Experiments: A Practical Course; Wiley-VCH: Weinheim,
Germany, 1998.
(46) Perdew, J. P. Phys. Rev. B 1986, 33, 8822.
(47) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(48) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
Revision B.01; Gaussian, Inc., Wallingford, CT, 2009.
(16) Wiedner, E. S.; Yang, J. Y.; Chen, S.; Raugei, S.; Dougherty, W.
G.; Kassel, W. S.; Helm, M. L.; Bullock, R. M.; Rakowski DuBois, M.;
DuBois, D. L. Organometallics 2012, 31, 144.
(17) Kilgore, U. J.; Roberts, J. A. S.; Pool, D. H.; Appel, A. M.;
Stewart, M. P.; Rakowski DuBois, M.; Dougherty, W. G.; Kassel, W. S.;
Bullock, R. M.; DuBois, D. L. J. Am. Chem. Soc. 2011, 133, 5861.
(18) Jain, A.; Lense, S.; Linehan, J. C.; Raugei, S.; Cho, H.; DuBois,
D. L.; Shaw, W. J. Inorg. Chem. 2011, 50, 4073.
(19) Smith, S. E.; Yang, J. Y.; DuBois, D. D. L.; Bullock, D. R. M.
Angew. Chem., Int. Ed. 2012, 51, 3152.
(20) Wiedner, E. S.; Yang, J. Y.; Dougherty, W. G.; Kassel, W. S.;
Bullock, R. M.; Rakowski DuBois, M.; DuBois, D. L. Organometallics
2010, 29, 5390.
(21) Yang, J. Y.; Chen, S.; Dougherty, W. G.; Kassel, W. S.; Bullock,
R. M.; DuBois, D. L.; Raugei, S.; Rousseau, R.; Dupuis, M.; Rakowski
DuBois, M. Chem. Commun. 2010, 8618.
(22) Raugei, S.; Chen, S.; Ho, M.-H.; Ginovska-Pangovska, B.;
Rousseau, R. J.; Dupuis, M.; DuBois, D. L.; Bullock, R. M. Chem. Eur.
J. 2012, 18, 6493.
(23) O’Hagan, M.; Shaw, W. J.; Raugei, S.; Chen, S.; Yang, J. Y.;
Kilgore, U. J.; DuBois, D. L.; Bullock, R. M. J. Am. Chem. Soc. 2011,
133, 14301.
(24) The spectrum of [Ni(PCy2NBn‑Phe‑OMe2H)2]2+ contains additional
resonances for the hydride at 13.27 ppm (18%) and an unknown
impurity at 3.37 ppm at 0 °C (∼2%) (Figure 6). The impurity is also
present in the [Ni(PCy2NBn‑Phe‑OMe2)2]2+ used to generate
[Ni(PCy2NBn‑Phe‑OMe2H)2]2+, moves downfield at cooler temperatures,
and broadens at higher temperatures. It was investigated by 31P−31P
EXSY at variable temperatures and was not found to exchange with
any of the other resonances present.
(25) Berning, D. E.; Noll, B. C.; DuBois, D. L. J. Am. Chem. Soc.
1999, 121, 11432.
(26) Wehman-Ooyevaar, I. C. M.; Grove, D. M.; Kooijman, H.; van
der Sluis, P.; Spek, A. L.; van Koten, G. J. Am. Chem. Soc. 1992, 114,
9916.
(27) Brammer, L. Dalton Trans. 2003, 3145.
(28) Brammer, L.; McCann, M. C.; Bullock, R. M.; McMullan, R. K.;
Sherwood, P. Organometallics 1992, 11, 2339.
(29) Zhao, D.; Ladipo, F. T.; Braddock-Wilking, J.; Brammer, L.
Organometallics 1996, 15, 1441.
(30) There is a current increase immediately following the Ni(II/I)
couple at ∼−0.6 V (green), but the current increases linearly with scan
rate, suggesting a stoichiometric rather than catalytic process. This
wave is tentatively assigned to oxidation of the hydride,
[HNi(PCy2NBn‑COOMe2)2]+, which arises from deprotonation of the
6731
dx.doi.org/10.1021/om300409y | Organometallics 2012, 31, 6719−6731