Synthesis, protonation, and reduction of ruthenium-peroxo complexes with pendent nitrogen bases

Article abstract of DOI:10.1021/ic3013987

Cyclopentadienyl and pentamethylcyclopentadienyl ruthenium(II) complexes have been synthesized with cyclic (RPCH₂NRCH₂)₂ ligands, with the goal of using these [Cp^RRu(P^R ₂N^R₂

Full text of DOI:10.1021/ic3013987

Article
pubs.acs.org/IC
Synthesis, Protonation, and Reduction of Ruthenium−Peroxo
Complexes with Pendent Nitrogen Bases
Tristan A. Tronic, Werner Kaminsky, Michael K. Coggins, and James M. Mayer*
Department of Chemistry, University of Washington, Box 351700, Seattle, Washington 98195-1700, United States
S
* Supporting Information
ABSTRACT: Cyclopentadienyl and pentamethylcyclopentadienyl
ruthenium(II) complexes have been synthesized with cyclic
(RPCH2NR′CH2)2 ligands, with the goal of using these [CpR′′Ru-
(PR NR′2)]+ complexes for catalytic O2 reduction to H2O (R = t-butyl,
2
phenyl; R′ = benzyl, phenyl; R″ = methyl, H). In each compound, the Ru is
coordinated to the two phosphines, positioning the amines of the ligand in the
second coordination sphere where they may act as proton relays to a bound
dioxygen ligand. The phosphine, amine, and cyclopentadienyl substituents
have been systematically varied in order to understand the effects of each of these parameters on the properties of the complexes.
These CpR″Ru(PR NR′2)+ complexes react with O2 to form η2-peroxo complexes, which have been characterized by NMR, IR,
2
and X-ray crystallography. The peak reduction potentials of the O2 ligated complexes have been shown by cyclic voltammetry to
vary as much as 0.1 V upon varying the phosphine and amine. In the presence of acid, protonation of these complexes occurs at
the pendent amine, forming a hydrogen bond between the protonated amine and the bound O2. The ruthenium−peroxo
complexes decompose upon reduction, precluding catalytic O2 reduction. The irreversible reduction potentials of the protonated
O2 complexes depend on the basicity of the pendent amine, giving insight into the role of the proton relay in facilitating
reduction.
molecules has been extensively demonstrated for H2
oxidation/H+ reduction with complexes of Fe,28−30 Ni,26,31−37
and Co.38−40 In particular, 1,5-diaza-3,7-diphosphacyclooctane
INTRODUCTION
■
The efficient reduction of dioxygen to water is critical to the
development of hydrogen as a fuel source, as this is the
cathodic half-reaction in a polymer electrolyte membrane
(PEM) fuel cell.1−3 This reaction, termed the oxygen reduction
reaction (ORR), is a proton-coupled electron transfer (PCET)
process, requiring the coordinated movement of four electrons
and four protons per molecule of O2. An ideal catalyst would
control the delivery of these electrons and protons and would
stabilize the many intermediates in this process.4,5 The
extensive research in this area has primarily focused on using
redox-active transition metal complexes to control the electron
delivery. Much progress has been made by studying iron,6−8
cobalt,9−11 and copper12−15 complexes with primary coordina-
tion spheres that resemble the active sites of biological oxidase
enzymes.16−18 In one recent elegant example, Collman,
Chidsey, and co-workers have attached ORR catalysts to a
gold electrode via a self-assembled monolayer (SAM), and
length of the SAM was varied to control the rate of electron
transfer to the catalyst.19 In contrast to the considerable
progress that has been made in controlling electron delivery,
only recently have complexes been developed that attempt to
control the delivery of protons.20−25 This report focuses on
developing an understanding of the design requirements for a
catalyst to control the delivery of protons to O2-derived
substrates, using amine bases in the second coordination sphere
that can act as proton relays.26,27
(PR NR′2) ligands have emerged as powerful ligands for H2
2
oxidation/H+ reduction catalysts,26,29,32−34,39,40 in part because
they bind to the metal center preferentially through the
phosphines, and their semirigid cyclooctane structure prevents
amine binding to the metal center but positions the amines
near the site of H2 binding/formation. Additionally, sub-
stitution of the R and R′ groups of these ligands is relatively
synthetically facile and allows for tuning of the metal’s redox
properties and pendent amine basicity. The PR NR′2 ligands
2
have been shown to dramatically improve catalyst overpotential
and turnover frequency in these systems. Recently, there has
been considerable interest in extending the utility of these
ligands to proton-coupled redox transformations of other small
molecules including CO2/HCOOH41,42 and N2/NH3.43,44
The utility of acidic proton relays in O2 reduction has been
explored previously with carboxylic acids rather than amines as
proton relays. Nocera and co-workers have pioneered the use of
proton relays with their “hangman” porphyrinoid complexes,
which position a carboxylic acid rigidly above the metal
center.22,23 They have shown with a cobalt “hangman”
porphyrin complex, selectivity for H2O production can be
increased from 48% to 71% in electronically similar
complexes.22 We have built on this work, showing that, with
The utility of positioned proton relays in molecular,
Received: June 28, 2012
transition metal-catalyzed redox transformations of small
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic3013987 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a series of iron (meso-tetra(carboxyphenyl))porphine catalysts,
positioning carboxylic acids near the oxygen binding site of the
complex dramatically improves catalyst stability and product
selectivity.25 While changes to the electronic structure of these
catalysts have been explored, no systematic study of the pKa of
the proton relay has been performed, nor has the mechanistic
role of the proton relay been elucidated in these systems. We
Scheme 1. Synthesis of Chloride Complexes
and others have begun exploring the use of PR NR′2 ligands for
2
O2 reduction, motivated in part by the relative synthetic ease
with which the pendent base may be modified without
significant changes to ligand structure. Yang et al. have
demonstrated that Ni(PR NR′2) complexes can reduce O2 to
2
H2O if a sufficiently basic pendent amine is used.45 However,
this system was not stable under catalytic conditions, and no
mechanistic data could be obtained. To address this lack of
mechanistic information, we have explored Ru complexes with
PR NR′2 ligands as potentially more stable systems, and recently
2
we have reported the synthesis and protonation of a
[Cp*Ru(PR NR′2)]+ (Cp* = η5-C5Me5) complex that binds
2
O2.46
The previously reported complex [Cp*Ru(PtBu2NBn2)]+ (tBu
= t-butyl, Bn = benzyl) reacts with O2 to make an η2-peroxo
complex.46 In the presence of acid, [Cp*Ru(PtBu2NBn2)(O2)]+
is protonated at a pendent amine, forming a hydrogen bond
with the O2 ligand. Both the protonated and unprotonated
species were characterized by X-ray crystallography, allowing
the interaction between the protonated pendent amine and the
O2 to be visualized. This species demonstrated a proton
dependent reduction potential, suggestive that the pendent
amine may be able to relay protons to the O2, but the reduction
mechanism could not be determined.
orange powder. The same synthetic procedure with PPh2NBn
2
gave a mixture of Cp*Ru(PPh2NBn2)Cl and other products, as
determined by NMR, that could not be separated by
recrystallization or column chromatography. These side
products may contain diphosphine ligands that bridge multiple
Ru centers, as has been observed in the synthesis of
Cp*Ru(dppm)Cl.47 In this case, phosphine ligand exchange
We have sought better mechanistic understanding of the
ability of these complexes to direct protons to O2 during
reduction by studying the effects of variations in the ligands.
Presented herein are the synthesis, protonation, and reduction
is a better procedure: refluxing Cp*Ru(PPh3)2Cl with PPh2NBn
2
of new [CpR″Ru(PR NR′2)(O2)]+ complexes with systematic
in toluene for two days gave Cp*Ru(PPh2NBn2)Cl as a yellow
2
variation in the ligands from the previously communicated
PtBu2NBn2 complex. A less basic pendent amine has been tested
powder in modest yield.
The Cp*Ru(PR NR′2)Cl compounds have been characterized
2
by synthesizing a complex with a PtBu2NPh ligand (Ph =
1
by H and 31P NMR spectroscopy and X-ray crystallography.
2
Their 31P NMR spectra contain only one sharp singlet
corresponding to the two equivalent phosphines. The X-ray
phenyl), and a less electron donating phosphine has been tested
by synthesizing a complex with a PPh2NBn2 ligand. Additionally,
the role of the Cp* has been tested by substituting this for a Cp
ligand (Cp = η5-C5H5). The effect of each of these changes on
the basicity and reduction potential of the complex has been
probed. These studies have shown that these complexes
decompose during reduction, likely due to oxidation of the
crystal structures show that the PR NR′2 ligands bind only
2
through the phosphines, with the third leg of the piano stool
occupied by a bound Cl− (Figure 1a,b; the structure of
Cp*Ru(PtBu2NBn2)Cl is reported in ref 46). The nitrogen bases
in each complex are positioned in the second coordination
sphere of the metal. The lone pair of the nitrogen closest to the
Cl− (N1) points away from electron-rich Cl−, and the lone pair
of N2 points away from the lone pair of N1. The binding of the
phosphine ligands is essentially the same in Cp*Ru(PtBu2NBn2)-
Cl46 and Cp*Ru(PtBu2NPh2)Cl, with average Ru−P distances of
2.309 0.004 Å and 2.304 0.009 Å and average P−Ru−P
angles of 78.0 0.1° and 78.4 0.1°, respectively (avg of two
independent molecules in each unit cell). With a phenyl
substituent on the phosphine (Cp*Ru(PPh2NBn2)Cl), the
phosphines, highlighting the limitations of applying PR NR′2
2
ligands to O2 reduction. The studies have also shown that the
potential for reduction of these species can be controlled via the
acidity of the pendent amine, highlighting the important role
that proton relays can play in improving O2 reduction catalysts.
RESULTS
■
Synthesis of Cp*Ru(PR NR′2)Cl and Related Com-
2
plexes. New complexes of Cp*Ru(II) with 1,5-diaza-3,7-
diphosphacyclooctane (PR NR′2) ligands were synthesized by
average Ru−P bond distance is shortened by 0.063
0.005
2
reaction of Cp*RuIICl precursors with the appropriate PR NR′2
Å (Table 1).
2
ligand (Scheme 1), similar to the synthesis of Cp*Ru-
(PtBu2NBn2)Cl reported in the previous communication.46
Cp*Ru(PtBu2NPh2)Cl was prepared by the room temperature
reaction of [Cp*RuCl]4 with PtBu2NPh2 in THF. The reaction is
The Cp derivative CpRu(PtBu2NBn2)Cl was obtained in good
yield by refluxing CpRu(PPh3)2Cl with PtBu2NBn2 in toluene for
1
two days. Its H NMR spectrum closely resembles that of
Cp*Ru(PtBu2NBn2)Cl, and it has a 31P{1H} NMR singlet
resonance that is 11.8 ppm downfield from that of Cp*Ru-
(PtBu2NBn2)Cl (Table 2). CpRu(PtBu2NBn2)Cl did not form
slow, due to the low solubility of PtBu2NPh in THF, but after
2
three days affords the desired product in modest yield as an
B
dx.doi.org/10.1021/ic3013987 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
acetonitrile complex closely resembles that of the Cp*Ru-
(PR NR′2)Cl complexes (Figure 1c).
2
A closely related Cp*Ru complex without a pendent amine,
Cp*Ru(dippp)Cl (dippp =1,3-bis(diisopropylphosphino)-
propane), was synthesized for comparison with the (PR NR′2)
2
derivatives. Its X-ray crystal structure (Figure 1d) shows that
the bulky isopropyl substituents and trimethylene backbone of
dippp resemble the t-butyl substituents and three-atom
backbone of PtBu2NBn2. Although the crystallographically
observed Ru−P bond lengths and bite angle are slightly larger
in Cp*Ru(dippp)Cl, this complex appears to be a good all-alkyl
analogue compared to Cp*Ru(PtBu2NBn2)Cl based on the
reactivity and redox potentials presented below.
Cp*Ru(PtBu2NBn2)Cl can be protonated, as observed by
NMR spectroscopy. Addition of 1 equiv of protonated
dimethylformamide triflate ([H+DMF][OTf], pKa in acetoni-
trile of 6.1),48 to a solution of Cp*Ru(PtBu2NBn2)Cl in CD2Cl2
1
formed a single new species by H and 31P NMR spectros-
1
copies. In the H NMR spectrum of Cp*Ru(PtBu2NBn2)Cl +
[H+DMF][OTf], one of the benzyl CH2 resonances is shifted
downfield and split into a doublet (JH−H = 5.5 Hz) due to its
proximity to the acidic NH. The NH proton appears as a broad
1
singlet at 11.1 ppm. The H NMR spectrum of this solution
Figure 1. ORTEPs of (a) Cp*Ru(PtBu2NPh2)Cl, (b) Cp*Ru-
(PPh2NBn2)Cl, (c) [CpRu(PtBu2NBn2)(MeCN)][PF6], and (d) Cp*Ru-
(dippp)Cl. Thermal ellipsoids are shown at 50% probability.
Hydrogen atoms and solvent molecules have been omitted for clarity.
Only the cationic portion of [CpRu(PtBu2NBn2)(MeCN)][PF6] and
only one of the two independent molecules of Cp*Ru(PtBu2NPh2)Cl in
the unit cell are shown.
closely resembles that of the protonated O2 complex [Cp*Ru-
(PtBu2NBn2H)(O2)][PF6] (see Supporting Information).46 This
suggests that protonation of the chloride species in CD2Cl2 is
similar to protonation of the O2 species, with the acidic proton
located on the pendent amine (see below).
Synthesis of Dioxygen Complexes. Dioxygen-bound
complexes are obtained by chloride abstraction from Cp*Ru-
(PR NR′2)Cl with TlX (X = OTf− or PF6 ) in CH2Cl2 or
−
2
acetone solutions open to air (eq 1). These reactions parallel
well-known analogues with simple bis(phosphine) ligands,49
and the chemistry of the dippp complexes is very similar. The
O2 complexes are stable species, and removal of the solvent
crystals suitable for X-ray diffraction. However, an X-ray
structure was obtained for a related complex with acetonitrile in
place of chloride, [CpRu(PtBu2NBn2)(MeCN)][PF6], synthe-
sized by chloride abstraction with TlPF6 in the presence of
acetonitrile. The geometry of the phosphine ligand of this
under vacuum yields [Cp*Ru(PR NR′2)(O2)][X] or [Cp*Ru-
2
(dippp)(O2)][X] as brown solids in high yield. Additionally,
a
Table 1. Selected Bond Lengths and Angles of Chloride, Acetonitrile, and Dioxygen Ligated Complexes
P1−Ru1
P2−Ru1
Cl1−Ru1
O1−Ru1
O2−Ru1
O1−O2
P1−Ru1−P2
b
Cp*Ru(PtBu2NPh2)Cl
2.3077(4),
2.3043(4)
2.3001(4),
2.3134(4)
2.4597(4),
2.4550(4)
78.272(15),
78.375(15)
Cp*Ru(PPh2NBn2)Cl
2.2492(4)
2.2435(4)
2.2914(4)
2.4412(3)
77.780(12)
79.643(12)
[CpRu(PtBu2NBn2)(MeCN)] 2.2868(4)
[PF6]
Cp*Ru(dippp)Cl
2.3271(3)
2.3271(3)
2.4608(3)
88.045(10)
[Cp*Ru(PtBu2NPh2)Cl]
2.3758(6),
2.3674(6)
2.3658(5),
2.3759(6)
2.3221(5),
2.3326(5)
77.388(18),
77.511(19)
b
[PF6]
[Cp*Ru(PtBu2NBn2)(O2)]
2.3591(3)
2.3660(3)
2.3053(8)
2.3264(9)
2.3668(3)
2.3774(3)
2.3022(8)
2.3286(9)
2.0229(7)
2.0190(7)
1.4009(11)
1.4063(13)
1.400(3)
76.572(9)
76.957(10)
76.83(3)
c
[BPh4]
[Cp*Ru(PtBu2NPh2)(O2)]
2.0280(8)
2.028(2)
2.027(2)
2.0227(8)
2.027(2)
2.025(3)
[BPh4]
[Cp*Ru(PPh2NBn2) (O2)]
[OTf]
[CpRu(PtBu2NBn2)(O2)]
[OTf]
1.404(4)
78.59(3)
[Cp*Ru(dippp)(O2)]
2.3777(3),
2.3865(3)
2.3812(3),
2.3920(3)
2.0244(10),
2.0312(9)
2.0294(8),
2.0245(8)
1.4003(12),
1.4064(12)
85.524(11),
85.030(10)
b
[PF6]
[Cp*Ru(PtBu2NBn2H)(O2)] 2.3574(5),
2.3674(5),
2.3744(4)
2.0393(14),
2.0260(11)
2.0362(14),
2.0350(11)
1.405(2),
1.4161(17)
78.412(16),
78.093(14)
c,d
[PF6]2
2.3580(4)
2.2974(9)
[Cp*Ru(PPh2NBn2H)(O2)]
2.3224(9)
2.025(2)
2.041(2)
1.414(4)
78.46(4)
[OTf]2
a
b
c
d
Distances are in Å, angles are in deg. Values are given for each independent molecule within the unit cell. Reference 46. Values are given for the
two different crystal structures obtained of this material (see ref 46).
C
dx.doi.org/10.1021/ic3013987 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a
1
Table 2. Selected H and 31P{1H} NMR Resonances and ν(O−O)
compd
31P{1H}
1H(C5CH3)5
1H(NH+)
ν(O−O) [ν(18O−18O)]
a
Cp*Ru(PtBu2NBn2)Cl
Cp*Ru(PPh2NBn2)Cl
Cp*Ru(PtBu2NPh2)Cl
CpRu(PtBu2NBn2)Cl
Cp*Ru(dippp)Cl
40.0
31.0
44.0
51.8
32.5
29.4
22.0
30.2
45.9
23.3
1.62
1.29
1.68
4.64 (C5H5)
1.63
b
[Cp*Ru(PtBu2NBn2)(O2)][PF6]
1.72
935 [880]
[Cp*Ru(PPh2NBn2)(O2)][OTf]
[Cp*Ru(PtBu2NPh2)(O2)][OTf]
[CpRu(PtBu2NBn2)(O2)][OTf]
[Cp*Ru(dippp)(O2)][PF6]
[Cp*Ru(PtBu2NBn2H)(O2)][PF6]2
[Cp*Ru(PPh2NBn2H)(O2)][PF6]2
[Cp*Ru(PtBu2NPh2H)(O2)][PF6]2
1.25
926 [879]
1.82
930 [880]
5.83 (C5H5)
1.77
d
d
b
c
30.3
1.81
7.68
8.29
9.98
905 [840]
16.6
1.24
d
e
41.6
2.01
∼910 [857]
a
b
c
d
e
Chemical shifts in δ (ppm); ν(O−O) in cm−1. Data from ref 46. Resonance is broad at room temperature. ν(O−O) not assigned. ν(O−O) is
obscured by absorbance of the CH2Cl2 solvent. An approximate value based on the observed ν(18O−18O) is provided, assuming harmonic oscillator
behavior for the O−O bond (see text).
[Cp*Ru(PtBu2NPh2)(O2)]+ and [Cp*Ru(dippp)(O2)]+ may be
−
obtained as tetraphenylborate (BPh4 ) salts by stirring the
respective Cl− complex with NaBPh4 in ethanol in the air. The
complexes have been characterized by X-ray crystallography
1
and IR spectroscopy, and by their well-resolved H and 31P
spectra. One sharp singlet is observed in the 31P{1H} spectra
due to the equivalent phosphines of the PR NR′2 ligand, shifted
2
∼10 ppm upfield of the singlet observed for the chloride
precursors (Table 2). These peroxo complexes appear to be
indefinitely stable in the solid state, and stable for several days
in CH2Cl2 solution before any decomposition is evident by
NMR. In solution they do not lose their O2 ligand under
vacuum or under a nitrogen atmosphere, though the O2 is
displaced by solvent in acetonitrile solution.
The X-ray crystal structures of the [Cp*Ru(PR NR′2)(O2)]+
2
and [Cp*Ru(dippp)(O2)]+ complexes show that the O2 ligand
is bound in an η2-fashion (Figure 2), with O−O bond distances
of 1.400−1.406 Å (Table 1). The O−O stretching frequencies
were obtained from spectra of KBr pellets and confirmed by
18O2-labeling; for instance, a value of 930 cm−1 (ν18O−18O = 880
cm−1) is observed for the PtBu2NPh2 derivative. These values are
in the typical range for [Cp*Ru(diphosphine)(O2)]+ com-
plexes,49 and make these species formally Ru(IV)−peroxo
complexes. Interestingly, all of the O−O stretching frequencies
are within 9 cm−1, and the O−O bond distances are the same
within error. From these data it can be concluded that there is
very little difference in the binding of O2 with changes in the
Figure 2. ORTEPs of (a) [Cp*Ru(PtBu2NPh2)(O2)][BPh4], (b)
[Cp*Ru(PPh2NBn2)(O2)][OTf], (c) [CpRu(PtBu2NBn2)(O2)][OTf],
and (d) [Cp*Ru(dippp)(O2)][PF6]. Thermal ellipsoids are shown
at 50% probability. Hydrogen atoms, counteranions, and solvent
molecules have been omitted for clarity. Only one of the two
independent molecules of [Cp*Ru(dippp)(O2)][PF6] in the unit cell
are shown.
[OTf], generated in situ from CpRu(PtBu2NBn2)Cl and TlOTf in
air, forming a yellow solution in minutes. The peroxo complex
was observed by NMR by the appearance of a singlet in the
31P{1H} NMR spectrum at 45.9 ppm, characteristically shifted
upfield from the CpRu(PtBu2NBn2)Cl resonance (51.8 ppm).
This peroxo complex decomposes completely over 1−2 h to an
unidentified brown oil, and some decomposition is apparent
even in the time required to prepare an NMR sample. Only one
[CpRu(phosphine)2(O2)]+ has been reported previously,50 so
instability is characteristic of these complexes. The high stability
of the related Cl− and MeCN complexes suggests that the O2
ligand. The conformation of the PR NR′2 ligands in the peroxo
2
species is identical to that in the chloride complexes, with the
amine adjacent to the O2 ligand positioned with its lone
electron pair pointed away from the O2 ligand.
The peroxo complex of the Cp derivative, [CpRu(PtBu2NBn2)-
(O2)]+, was obtained by reaction of [CpRu(PtBu2NBn2)]+ with
air but is not stable and could be isolated only in sufficient yield
to obtain a crystal structure. Formation of the O2 complex was
observed on stirring a CD2Cl2 solution of [CpRu(PtBu2NBn2)]-
D
dx.doi.org/10.1021/ic3013987 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
ligand plays a role in the decomposition. Quickly layering a
CH2Cl2 solution of [CpRu(PtBu2NBn2)(O2)][OTf] with hex-
anes and storing the solution at −20 °C for weeks afforded a
few X-ray quality crystals from which a structure was obtained.
NMR spectra of these crystals were identical to the spectra of
the compound prepared in situ. The X-ray crystal structure of
[CpRu(PtBu2NBn2)(O2)][OTf], to our knowledge the only
crystal structure of a Cp ligated Ru−O2 complex, is very similar
to that of the Cp* complexes (Figure 2c, Table 2). The O2 is
bound η2 with d(O1−O2) = 1.404(4) Å. While the esd on this
distance is large, the similarity to the O−O bond distances of
the Cp* derivatives is nonetheless striking considering the
differences in the stability of these species and electron
donating character of Cp* versus Cp (see electrochemical
data and Table 4 below). From a simplistic perspective, a more
electron donating Cp* ligand would be expected to give more
peroxo character and a longer O−O distance, but this is not
what is observed.51 The solid state structure of [CpRu-
(PtBu2NBn2)(O2)][OTf] contains a water molecule hydrogen-
bonded to the O2 ligand, with d(O2−O3) = 2.898(4) Å, which
may impart stability to this structure and explain why it was
able to be isolated despite the instability of solutions of this
compound.
[Cp*Ru(PPh2NBn2H)(O2)]+2 is stable for days in solution
and may be isolated as a red solid, similar to [Cp*Ru-
(PtBu2NBn2H)(O2)]+2. Red, X-ray quality crystals of [Cp*Ru-
(PPh2NBn2H)(O2)][OTf]2 were obtained from an acetone
solution of [Cp*Ru(PPh2NBn2)(O2)][OTf] and [H+DMF]-
[OTf] stored at −20 °C for one week. The X-ray structure
obtained from these crystals suffers from multiply disordered
triflate anions and acetone solvent, and is disordered in the
position of the benzyl arms of the ligand (Figure 3). The
Protonation of Peroxo Complexes. A. Effect of
Phosphine Substituent on Protonation. The ability of the
pendent amines to direct protons to the O2 ligand has been
Figure 3. ORTEPs of (a) [Cp*Ru(PPh2NBn2H)(O2)][OTf]2·(acetone)
and (b) [Cp*Ru(PtBu2NBn2H)(O2)][PF6]2·(H2O)0.25 (data from ref
46). Thermal ellipsoids are shown at 50% probability, except for
benzyl groups of [Cp*Ru(PPh2NBn2H)(O2)][OTf]2 for which both
disordered groups are shown. Hydrogen atoms except for the acidic
N−H, counteranions, and solvent molecules have been omitted for
clarity. The acidic hydrogens were not located in the difference map,
and were placed in geometrically idealized positions (a riding model).
tested by adding acid to solutions of [Cp*Ru(PR NR′2)(O2)]+.
2
We have reported previously that the t-butyl/benzyl derivative
can be protonated at the pendent amine forming a stable
complex [Cp*Ru(PtBu2NBn2H)(O2)]+2 in which there is an
intramolecular hydrogen bond between the protonated amine
and the O2 ligand.46 Protonation of [Cp*Ru(PPh2NBn2)(O2)]+,
which differs only in having a phenyl substituent at the
phosphine, is very similar. Addition of 1 equiv of [H+DMF]-
[OTf] to a CD2Cl2 solution of [Cp*Ru(PPh2NBn2)(O2)][OTf]
thermal parameters of the disordered phenyl rings were
restrained to be similar within each pair. However, no restraints
were imposed that would have affected the structure of the
complex other than the phenyl rings. The atomic positions of
the core of the cationic portion of the molecule show clear
evidence of protonation at the amine nearest the O2 ligand, and
there are strong similarities with the structure of [Cp*Ru-
(PtBu2NBn2H)(O2)]+2.46 Both amines are inverted from their
orientation in the unprotonated molecule, bringing nitrogen N1
within less than 2.9 Å from oxygen O2. This is characteristic of
an NH···O hydrogen bond,52 and bringing the lone pair of an
amine within such a short distance of the lone pair of an O2
ligand would be unlikely without protonation at the amine. The
NH···O2 interaction is asymmetric, with the protonated amine
substantially closer to one oxygen atom than the other,
d(N1···O2) = 2.716(4) Å, d(N1···O1) = 2.934(4) Å (Table 1).
The effect of varying the phosphine substituent on the
1
gives [Cp*Ru(PPh2NBn2H)(O2)][OTf]2 by H and 31P NMR
(eq 2). One benzylic resonance and the PCH2N resonances of
the ligand are shifted downfield, and a new, slightly broad
singlet appears at 8.29 ppm in the 1H NMR spectrum,
consistent with protonation of one of the benzylamine ligands.
Assignment of this new resonance at 8.29 ppm as the acidic
benzylammonium proton is supported by the signal dropping
in intensity when deuterium-labeled [D+DMF][OTf] is used to
protonate [Cp*Ru(PPh2NBn2)(O2)][OTf]. The chemical shift
of this benzylammonium is similar to the chemical shift of 7.68
ppm observed for protonated [Cp*Ru(PtBu2NBn2H)(O2)]+2,
evidence that the environment of the acidic proton is similar
between these two complexes. This chemical shift is
inconsistent with the proton bridging the two benzylamine
groups of the PtBu2NBn2 ligand, since protons in such structures
appear at ca. 15 ppm based on observations of protonated
basicity of [Cp*Ru(PR NR′2)(O2)]+ was tested by combining a
2
CD2Cl2 solution of the protonated t-butylphosphine complex
[Cp*Ru(PtBu2NBn2H)(O2)][PF6]2 with an equimolar quantity
of the unprotonated phenylphosphine compound [Cp*Ru-
(PPh2NBn2)(O2)][OTf]. 1H and 31P NMR spectra show distinct
resonances of both the protonated and unprotonated forms of
both the t-butyl- and phenyl-substituted complexes (eq 3,
nickel−PR NBn2 complexes.33 In the 31P{1H} NMR spectrum, a
2
new sharp singlet resonance appears, shifted upfield by 5.4 ppm
(Table 2). This upfield shift contrasts with the slight downfield
shift and substantial broadening observed in 31P{1H} spectra
upon protonation of [Cp*Ru(PtBu2NBn2)(O2)]+.46
E
dx.doi.org/10.1021/ic3013987 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 4). Thus, proton exchange between the complexes is
slow on the NMR time scale, and the equilibrium constant Keq
of 1 equiv of [H+DMF][OTf], this stretch is no longer present,
but no new band is observed (see Supporting Information).
Protonation of the 18O2-labeled material shows a new band at
857 cm−1, shifted −23 cm−1 from the band in the unprotonated
material. A similar shift for the 16O2 compound would place the
stretch underneath a solvent absorption, explaining why the
band was not observed. A ν(O−O) of ∼910 cm−1 is predicted
for the 16O2 compound by a harmonic oscillator approximation,
or a shift of ca. −20 cm−1 upon protonation. This shift in ν(O−
O) is a little smaller than the −30 cm−1 observed on
protonation of the related N-benzyl derivative [Cp*Ru-
(PtBu2NBn2H)(O2)][OTf]2.46
Density functional theory (DFT) calculations were per-
formed to demonstrate that protonation at the aniline and
hydrogen bonding to O2 is a reasonable structure for this
compound. Gas-phase geometry optimization of the unproto-
nated, cationic [Cp*Ru(PtBu2NPh2)(O2)]+ was performed at the
BP86/6-31G** (C, H, N, O, P) SDD (Ru) level of theory on
the singlet surface. This level of theory was chosen because of
its previously demonstrated ability to reproduce the geometries
of [Cp*Ru(PtBu2NBn2)(O2)]+and [Cp*Ru(PtBu2NBn2H)-
(O2)]+2.46 The DFT optimized geometry of [Cp*Ru-
(PtBu2NPh2)(O2)]+ agrees well with the X-ray structure (see
Table 3). A gas-phase optimization was also performed on the
Figure 4. 31P{1H} NMR spectrum of the reaction of [Cp*Ru-
Ph
+
(PtBu2NBn2H)(O2)]+2 ( ) with [Cp*Ru(P 2NBn2)(O2)] ( ) to form
■
▲
tBu
+
[Cp*Ru(PPh2NBn2H)(O2)]+2 ( ) and [Cp*Ru(P 2NBn2)(O2)] ( )
⧫
●
in CD2Cl2.
Table 3. Selected DFT-Calculated Bond Lengths and O−O
Stretching Frequencies for [Cp*Ru(PtBu2NPh2)(O2)]+ and
for this reaction is ∼0.04 (based on integrations of the t-butyl
1
and Cp* resonances in the H NMR and of the 31P{1H}
a
[Cp*Ru(PtBu2NPh2H)(O2)]+2
signals). From this it can be seen that the phosphine
substituent, despite its distance from the amine in the
[Cp*Ru(PtBu2NPh
)
2
PR NR′2 ligand, exerts a measurable effect on the basicity of
(O2)]+
(diff from expt)
[Cp*Ru(PtBu2NPh2H)
2
(O2)]+2
the complex, a difference in pKa of ∼1.4.
B. Effect of Amine Substituent on Protonation. Proto-
nation of the N-phenyl derivative [Cp*Ru(PtBu2NPh2)(O2)]-
[OTf] with 1 equiv of [H+DMF][OTf] to a CD2Cl2 solution
also forms a protonated species. The NMR spectra of this
complex indicate that protonation occurs at one pendent
amine, likely the amine near the O2, as for the complexes with
pendent benzylamines. The sharp singlet in the 31P{1H} NMR
spectrum of the unprotonated complex at δ = 30.2 ppm is
d(Ru−P)avg (Å)
d(Ru−O1) (Å)
d(Ru−O2) (Å)
d(O1−O2) (Å)
2.411 (+0.039)
2.043 (+0.015)
2.050 (+0.027)
1.404 (−0.002)
976 (+46)
2.419
2.044
2.085
1.418
956
ν(O−O) (cm−1
)
a
DFT calculations at the BP86/6-31G** (C, H, N, O, P) SDD (Ru)
level of theory.
1
shifted downfield by 11.5 ppm upon protonation. H NMR
spectra show a downfield shift of two of the PCH2N and the
aryl resonances, as well as the appearance of a broad singlet
integrating to 1H at 9.98 ppm. This new broad singlet has been
identified as a protonated pendent aniline by 1H−15N
correlated NMR spectroscopy of isotopically labeled [Cp*Ru-
protonated [Cp*Ru(PtBu2NPh2H)(O2)]+2, which gave a mini-
mum with the proton hydrogen bonding to the O2 (Figure 5).
The hydrogen bonding is asymmetric, with d(N1−O2) = 2.638
Å and d(N1−O1) = 2.998 Å. The calculated O−O stretching
frequency is reduced on protonation by 20 cm−1, which is quite
consistent with experiment.
(PtBu 15NPh2)(O2)][OTf]. The 1D 1H−15N HSQC spectrum of
2
[Cp*Ru(PtBu 15NPh2)(O2)][OTf] with 1 equiv of [H+DMF]-
2
[OTf] has one resonance at δ = 9.98 ppm, with a typical one-
bond coupling constant, J1H‑15N = 75 Hz. These data
demonstrate protonation of the NPh group, and show that
the proton does not bridge two anilines.53,54
The change in the basicity of the complex on changing the
pendent base was tested by NMR. A CD2Cl2 solution of the
protonated N-phenyl complex, prepared in situ from [Cp*Ru-
(PtBu2NPh2)(O2)][OTf] and ca. 1 equiv of [H+DMF][OTf],
was treated with ca. 1 equiv of [Cp*Ru(PtBu2NBn2)(O2)][OTf].
Complete deprotonation of the aniline complex and formation
of protonated benzylamine complex [Cp*Ru(PtBu2NBn2H)-
Solutions of [Cp*Ru(PtBu2NPh2H)(O2)]+2 showed multiple
decomposition products by NMR spectroscopy within a day.
The stability of the protonated complexes is thus greatly
reduced on decreasing the basicity of the pendent amine, since
solutions of [Cp*Ru(PtBu2NBn2H)(O2)]+2 and [Cp*Ru-
(PPh2NBn2H)(O2)]+2 showed no such signs of decomposition
after several days at room temperature. While crystals of
[Cp*Ru(PtBu2NPh2H)(O2)]+2 could not be obtained, IR spectra
provide clear evidence for hydrogen bonding between the
protonated pendent amine and the O2 ligand. IR spectra of
CH2Cl2 solutions of [Cp*Ru(PtBu2NPh2)(O2)]+ have an O−O
stretch at 930 cm−1 (ν(18O−18O) = 880 cm−1). Upon addition
1
(O2)][OTf]2 was observed by H and 31P NMR. The relative
integrations of the species in the product NMR spectra imply a
difference in pKa of at least ∼3. The large difference is not
surprising, given that the pKa of aniline in acetonitrile is 6.3 pKa
units less than that of benzylamine.55
The complex [Cp*Ru(dippp)(O2)][PF6], without a pendent
amine, does not protonate under the conditions tested. A
CD2Cl2 solution showed no change in its NMR spectrum when
1 equiv of [H+DMF][OTf] was added. The O2 ligand in this
F
dx.doi.org/10.1021/ic3013987 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
B. Reduction of Dioxygen Complexes. CVs of the peroxo
complexes [Cp*Ru(PR NR′2)(O2)]+ and [Cp*Ru(dippp)-
2
(O2)]+ in CH2Cl2 (ca. 1.5 mM, 0.1 M TBAPF6) show one
completely irreversible reduction wave at potentials negative of
−1 V vs Cp2Fe+/0 (see Supporting Information), as previously
shown for [Cp*Ru(PtBu2NBn2)(O2)]+.46 CVs were recorded in
CH2Cl2 rather than acetonitrile or other more polar solvents
that are typically used for CV because the O2 ligand is displaced
in coordinating solvents like acetonitrile. The potentials of the
peak cathodic currents (Epc) depend on the ligand, with a span
of 0.23 V.
In the presence of 1 equiv of [H+DMF][OTf], a new
irreversible reduction wave is observed for all [Cp*Ru-
(PR NR′2)(O2)]+, with a positive shift in Epc (Figure 6, Table
2
Figure 5. DFT (BP86/6-31G** (C, H, N, O, P) SDD (Ru))
optimized geometry of [Cp*Ru(PtBu2NBn2H)(O2)]+2. Hydrogen atoms
except the acidic hydrogen have been omitted for clarity. The
hydrogen bonding N1−H1···O2 interaction is indicated by a dashed
line to H1.
complex, despite being formally reduced to a peroxide, is still
less basic than dimethylformamide.
Electrochemical Characterization of CpR″Ru(PR NR′2)+
2
Complexes. A. Oxidation of Chloride Complexes. Cyclic
voltammograms (CVs) of all of the chloride complexes show a
reversible oxidation wave between −0.1 and −0.4 V versus
Cp2Fe+/0, with typical peak separations of 0.12−0.15 V in
CH2Cl2 with 0.1 M tetra-n-butylammonium hexafluorophos-
phate (TBAPF6). The E1/2 potentials are given in Table 4, and
Figure 6. Overlay plot of reductive cyclic voltammograms of ca. 1.5
mM [Cp*Ru(PtBu2NBn2)(O2)][OTf] (green, solid), [Cp*Ru-
(PtBu2NPh2)(O2)][OTf] (blue, long dashes), and [Cp*Ru(PtPh2NBn2)-
(O2)][OTf] (red, solid) in CH2Cl2 (0.1 M TBAPF6) in the presence
of 1 equiv of [H+DMF][OTf]. Scan rate is 0.1 V/s.
Table 4. Half-Wave Potentials (E1/2) for 1e− Oxidation of
Chloride Complexes from CVs in CH2Cl2 (0.1 M TBAPF6)
a
E1/2 vs Cp2Fe+/0 (V)
Table 5. Peak Irreversible Reduction Potentials (Epc, V)
from Cyclic Voltammograms of
Cp*Ru(PtBu2NBn2)Cl
Cp*Ru(PPh2NBn2)Cl
Cp*Ru(PtBu2NPh2)Cl
CpRu(PtBu2NBn2)Cl
−0.38
−0.38
−0.30
−0.12
−0.40
+0.22
[Cp*Ru(diphosphine)(O2)]+ in CH2Cl2 with and without
a
Acid
Epc with 1 equiv
Cp*Ru(dippp)Cl
Cp*Ru(PtBu2NBn2)Cl + H+DMF
Epc
[H+DMF][OTf]
ΔEpc (V)
[Cp*Ru(PtBu2NBn2)(O2)]+
[Cp*Ru(PPh2NBn2)(O2)]+
[Cp*Ru(PtBu2NPh2)(O2)]+
[Cp*Ru(dippp)(O2)]+
−1.44
−1.55
−1.32
−1.45
−0.77
−0.90
−0.56
−1.22
+0.67
+0.65
+0.76
+0.23
a
Estimated error in E1/2 is 0.01 V.
CVs are shown in the Supporting Information. The potentials
show only modest differences with ligand, with all Cp*
derivatives being within 0.1 V. The oxidation waves are
attributed to RuII → RuIII oxidation. To confirm this
assignment, Cp*Ru(PtBu2NBn2)Cl was chemically oxidized by
a
CVs with ca. 1.5 mM complex in CH2Cl2 with 0.1 M [n-Bu4N][PF6].
Potentials referenced to Cp2Fe+/0, with estimated errors 0.02 V.
1 equiv of [Cp2Fe+][PF6 ] in CH2Cl2, yielding dark green
5). This wave does not shift in potential with the addition of up
to 10 equiv of [H+DMF][OTf]. The magnitude of the shift,
ΔEpc (Table 5), is approximately +0.7 V for all three cations.
CVs of [Cp*Ru(dippp)(O2)][PF6], which lacks a pendent
amine, also show an irreversible reduction wave in the presence
of 1 equiv of [H+DMF][OTf] (Figure 7). Even though
[Cp*Ru(dippp)(O2)][PF6] does not react with [H+DMF]-
[OTf] (see above), this wave is shifted positive of the wave in
the absence of acid. However, the magnitude of this shift
(+0.23 V) is significantly smaller than for pendent amine
complexes, suggesting that the pendent amine does play a role
in facilitating reduction. The CV of [Cp*Ru(dippp)(O2)]-
[OTf] with [H+DMF][OTf] does appear to be qualitatively
−
crystals. The X-ray crystal structure (Figure S16 in the
Supporting Information) confirms the composition as the
Ru(III) complex [Cp*Ru(PtBu2NBn2)Cl][PF6] with no ligand
oxidation. The Ru−Cl bond length (average of two
independent molecules in the unit cell) is 0.115
0.005 Å
shorter in the RuIII species than the parent RuII complex.
Cyclic voltammetry was also performed on a solution of
Cp*Ru(PtBu2NBn2)Cl in the presence of 1 equiv of [H+DMF]-
[OTf]. With the complex protonated, the wave corresponding
to neutral Cp*Ru(PtBu2NBn2)Cl oxidation is no longer present.
Instead, a new reversible wave is observed, which is shifted from
the complex in the absence of acid by +0.60 V.
G
dx.doi.org/10.1021/ic3013987 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
of different PR NR′2 ligands in nickel hydrogenase catalysis, the
2
key parameters and design criteria for these ligands are only
beginning to be understood. In this work, we have examined
the effects of systematic variations in the of [Cp*Ru-
(PR NR′2)]+ framework in the context of dioxygen reduction.
2
Holding the other groups the same, the phosphorus substituent
has been varied from t-butyl to phenyl, the nitrogen substituent
has been changed from benzyl to phenyl, and the Cp* ligand
has been replaced with Cp.
Substituting phenyl for t-butyl at the phosphine was expected
to make the complexes less electron rich and therefore easier to
reduce, due to the more electron-withdrawing character of the
phenyl group. In previously studied Ni(PR NPh2)2 complexes,
2
Figure 7. Overlay plot of reductive cyclic voltammograms of ca. 1.5
mM [Cp*Ru(PtBu2NBn2)(O2)][OTf] (green, solid) and [CpRu-
(dippp)(O2)][PF6] (orange, dashed) in CH2Cl2 (0.1 M TBAPF6) in
the presence of 1 equiv of [H+DMF][OTf]. Scan rate is 0.1 V/s.
substitution of phenyl for n-butyl resulted in changes in
reduction potential (ΔE1/2’s) in acetonitrile of +0.09 V and
+0.21 V for the Ni(II/I) and Ni(I/0) couples, respectively.34
However, no difference was observed in CVs of the chloride
complexes Cp*Ru(PtBu2NBn2)Cl and Cp*Ru(PPh2NBn2)Cl, and
for the dioxygen complexes the t-butyl derivative [Cp*Ru-
(PtBu2NBn2)(O2)]+ is actually 0.11 V easier to reduce than the
phenyl analogue (Tables 4 and 5). This may be due to a steric
effect of the Cp* ring interacting with the t-butyl phosphines,
hampering their ability to bind and donate to the Ru, as is
reflected in the somewhat longer P−Ru bond distance for
Cp*Ru(PtBu2NBn2)Cl relative to Cp*Ru(PPh2NBn2)Cl (Table 1).
A larger effect is observed upon substitution of Cp for Cp*,
with an increase in E1/2 of +0.26 V. This value is exactly the
same as the difference between ferrocene and pentamethylfer-
somewhat different than for the PR NR′2 complexes, with a
2
much greater potential difference between the onset of cathodic
current and the peak cathodic current Epc, so this conclusion
must be treated with some caution. The shape and position of
this wave are not significantly affected by additional acid up to
10 equiv.
Cyclic voltammograms of [Cp*Ru(PR NR′2)(O2)]+ under an
2
O2 atmosphere in the presence of acid were also performed in
order to test for O2 reduction catalysis. Large catalytic currents
were observed under these conditions. Unfortunately, these
currents were also present at similar potentials in control
experiments performed under identical conditions but in the
absence of ruthenium complex. Thus, most of the current
observed likely corresponded to direct O2 reduction by the
glassy carbon electrode at these very negative potentials (see
Supporting Information).
rocene (CpCp*Fe) under similar conditions (ΔE1/2
=
+0.259(3)56).
Surprisingly, the effect of changing the amine substituent
from R′ = Bn to Ph is larger than that of changing the
phosphine substituent on the oxidation potential of the chloride
ligated species. Cp*Ru(PtBu2NBn2)Cl is 0.08 V easier to oxidize
than Cp*Ru(PtBu2NPh2)Cl (Table 4). A dependence of the
redox potential of the metal center on the basicity of the
To investigate the chemical processes occurring during
reduction of the O2 ligated complexes [Cp*Ru-
(PR NR′2)(O2)]+, reactions of these complexes with 4 equiv
pendent amine with PR NR′2 has previously been demonstrated
2
2
of Cp*2Fe (E1/2 = −0.53 V vs Cp2Fe)56 were followed by
NMR. Consistent with the Epc potentials observed by CV, no
reaction was observed in the absence of acid. Addition of 4
equiv of [H+DMF][OTf] to solutions of [Cp*Ru-
for Ni(PR NR′2)+2 complexes,36 with ΔE1/2’s for R′ = t-butyl
2
versus phenyl of +0.10 V and +0.17 V for the Ni(II/I) and
Ni(I/0) couples, respectively, of Ni(PPh2NR′2)+2 in acetoni-
trile.35 Still, it is surprising that the amine substituent four
bonds from the Ru could have a larger effect than the
phosphine substituent two bonds from the Ru, especially
considering the crystallographic similarity in phosphine binding
between these t-butyl phosphine complexes. The oxidation
potential of Cp*Ru(dippp)Cl, without a relay, is within error of
those of the N-benzyl derivatives Cp*Ru(PtBu2NBn2)Cl and
Cp*Ru(PPh2NBn2)Cl, supporting the suggestion that this
(PR NR′2)(O2)]+ and Cp*2Fe caused a rapid color change to
2
1
bright green, indicating the formation of [Cp*2Fe]+. The H
and 31P NMR spectra of these reaction solutions contain a large
number of peaks, none of which can be identified with
certainty, but which suggest decomposition of the complex.
With all the complexes, 31P{1H} NMR spectra showed
downfield resonances in the +50 to +80 ppm range, suggestive
of phosphine oxide formation, as well as resonances in the −50
to −20 ppm range suggestive of unbound phosphines.
complex is a good analogue electronically to the PR NR′2
2
complexes.
Potential oxygen-atom acceptors were also reacted with the
Not only do the amine substituents affect the ruthenium
redox potential, but also the phosphine substituents signifi-
cantly affect the amine basicity. [Cp*Ru(PPh2NBn2)(O2)]+ is 1.4
pKa units less basic than [Cp*Ru(PtBu2NBn2)(O2)]+ despite
having a more negative reduction potential. This may be due to
an electron-withdrawing effect of the phenyl phosphine on the
pendent benzylamine. The aniline-derived proton relay is
substantially less basic than the benzylamine relay, as expected.
protonated complexes [Cp*Ru(PR NR′2H)(O2)]+2. As sum-
2
marized in the Supporting Information, no reaction was
observed with most substrates, including norbornene, thio-
anisole, and triphenylarsine. With the more basic phosphines,
such as methyldiphenylphosphine, only deprotonation was
observed.
However, as observed with other PR NR′2 complexes,35 this
DISCUSSION
■
2
change in basicity by several pKa units was coupled to a change
in redox potential of the complex of ca. 0.1 V (Table 5). These
results show that the metal and its first coordination sphere are
significantly coupled to the proton relays in the second
Proton relays have been shown to be very valuable components
of catalysts for the hydrogenase reaction, 2H+ + 2e− ⇌ H2, and
are receiving increasing attention for other multielectron,
multiproton processes.22−46 While there has been much study
H
dx.doi.org/10.1021/ic3013987 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
coordination sphere. It is difficult to independently tune
properties of a complex and its pendent proton relays.
The nature of the peroxo compounds [Cp*Ru-
vibration of the protonated species also involves motion of the
2
versus the pendent benzylamine may explain the differences in
the shift in stretching frequency.
PR NR′2 ligand. Differences in the rigidity of the pendent aniline
(PR NR′2)(O2)]+ does not appear to be significantly affected
2
by changes in the phosphine, aniline, or Cp groups, on the basis
of the structural and spectroscopic parameters. This is
surprising, as it might be expected on the basis of the
electrochemical data obtained for the Cl− and O2 complexes
that [Cp*Ru(PtBu2NPh2)(O2)][BPh4] would be more easily
reduced than [Cp*Ru(PtBu2NBn2)(O2)][BPh4], and therefore
would not donate as much electron density to the bound O2,
resulting in a shorter O−O bond. Substituting Cp for Cp*
produced a complex with much lower stability, but this does
not appear to be due to changes in Ru−O2 bonding given the
similarity of the X-ray structures. The reduced stability could be
a steric effect, perhaps involving intermolecular reactions
between Cp complexes that are not possible with the less
accessible Cp* complexes.
Recent DFT calculations of Ru−polypyridyl water oxidation
catalysts indicate that Ru(IV) η1-peroxo species can be quite
similar in energy to their η2-peroxo analogues.57,58 We have
therefore calculated the energies of singlet and triplet η1-peroxo
[Cp*Ru(PtBu2NBn2)(O2)]+.59,60 In the optimized geometries for
both spin states (Figure S26 of the Supporting Information),
the calculated O−O bond lengths are quite short (1.304 and
1.317 Å for the singlet and triplet, respectively). The calculated
energies of these species (E − ZPE) are both 12 kcal mol−1
higher than the energy previously determined for the singlet η2-
peroxo species.46 These findings are consistent with a recent,
more thorough computational study of O2 binding to CpRu(II)
with different ligands, which also found η1 species to be at
higher energy (although in this study the η1 triplet species were
typically much lower in energy than the η1 singlet).61
Protonation of the pendent amine substantially facilitates
reduction of the peroxo complexes [Cp*Ru(PR NR′2)(O2)]+, as
2
the cathodic peak potentials Epc shift by 0.65−0.76 V (Table 5).
The waves in the CVs are irreversible, which complicates the
interpretation of these values. Still, the similarity between the
complexes studied suggests that the ΔEpc values may be
reasonably compared. Some of the shift in Epc is due to the
additional positive charge on the complex, but this alone is
unlikely to account for all of the ∼0.7 V shift. The shift in the
RuII/III couple of Cp*Ru(PtBu2NBn2)Cl in the presence and
absence of acid of +0.60 V gives an approximate value for how
the potential might shift due purely to the additional charge of
the proton. This is not a perfect model because different
electrochemical processes are being compared. However, each
electrochemical process involves the interconversion of a
monocation with a dication in the presence of protons, so
the comparison is not unreasonable. The Coulombic effect of a
proton at the pendent amine on the Ru center should be quite
similar between the Cl− and O2 complexes. From this
comparison it may be concluded that roughly +0.6 V of the
shift for the O2 complexes is due to the additional charge of the
proton, and +0.1 V can be attributed to the reductions being
proton-coupled processes.
It is probable that the first electron transferred occurs with
proton transfer, producing a Ru(III) hydroperoxo complex.
The protonated formally Ru(IV) η2-peroxo complexes appear
structurally primed to form such RuIIIOOH species, which
would be analogous to the isolated ruthenium(III) chloride
compound [Cp*Ru(PtBu2NBn2)Cl][PF6]. The dependence of
ΔEpc on the nature of the pendent amine is consistent with this
proposal. ΔEpc is 0.09 V larger for the aniline complex than for
the related benzyl-amine complex because aniline is the weaker
base, making proton transfer from aniline to the peroxide ligand
more thermodynamically favorable in the aniline complex. Still,
this shift is smaller than would have been expected, since 0.09 V
is equivalent to a change in Keq for electron transfer (ET) of
101.5 while aniline is 106.3 less basic than benzylamine (6.3 pKa
units in MeCN55).
Protonation of the peroxo complexes [Cp*Ru-
(PR NR′2)(O2)]+ proceeds in essentially the same manner for
2
all the compounds tested, despite the large differences in the
basicities of the pendent amines and in the overall complexes.
The proton is located on the amine near the O2 ligand, forming
a hydrogen bond to one of the O atoms in the O2. The
hydrogen bond should be stronger to a protonated pendent
aniline versus a protonated pendent benzylamine because of the
better pKa matching between the aniline and the bound O2.
This is observed in the O···H−N hydrogen bonding distances
predicted by DFT, with d(O2···N1) calculated to be 0.016 Å
shorter for the aniline compound (Table 3). The hydrogen
The importance of the proton relays is indicated by
comparing the peak reduction potentials for the [Cp*Ru-
(PR NR′2)(O2)]+ complexes versus that of [Cp*Ru(dippp)-
2
bond in [Cp*Ru(PR NR′2H)(O2)]+2 imparts a small amount of
(O2)]+, which does not contain a pendent amine. Protons do
facilitate this reduction, but give a ΔEpc of only 0.23 V, only
about a third of the shift for the compounds with the proton
relays. It is interesting that [H+DMF][OTf] does not protonate
the O2 ligand in the dippp complex (since the NMR spectra are
unaffected by this acid). The O2 ligand has very low basicity,
suggesting that the Ru−O2 interaction is quite covalent, so an
ionic Ru(IV)−peroxide picture of these species is not accurate.
From another perspective, the lack of protonation indicates that
the dicationic Ru(IV) hydroperoxide is a high energy species.
This analysis also illustrates the value of the proton relays, in
holding protons close to a ligand with low basicity.
2
η1-hydroperoxo character to the bound O2 ligand, as can be
seen by X-ray and in the DFT structures in the lengthening of
the Ru−O bond to the hydrogen bonded O atom relative to
the other Ru−O bond. Bonding to the positively charged H+
pulls some electron density out of the Ru and onto the O2. The
additional electron density from Ru is donated to the O2 π*
orbital, slightly weakening the O−O π bond. This is reflected in
lower energy O−O bond stretching frequencies and longer
bond lengths on protonation. It is therefore somewhat
unexpected that the complex with the pendent aniline has a
smaller shift in O−O stretching frequency upon protonation
(∼20 cm−1 versus 30 cm−1 for the benzylamine complex),
despite forming an apparently stronger hydrogen bond. DFT
calculations agree with this experimentally observed difference
in O−O stretching frequency, predicting a shift of 20 cm−1 for
the aniline complex versus 25 cm−1 for the benzylamine
complex.46 These calculations show that the O−O stretching
Despite the favorable interaction between the protonated
pendent amine and the O2, these complexes are not effective
catalysts for O2 reduction. Reduction of the peroxo complexes,
both protonated and not, is completely irreversible, and the
nature of the chemical processes occurring upon reduction is
not known. The lack of anodic current in the CVs suggests a
I
dx.doi.org/10.1021/ic3013987 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Bruker Optics Tensor27 FTIR spectrometer at room temperature as
KBr pellets or in CH2Cl2 solution between NaCl plates as indicated.
NMR spectra were recorded at 298 K on Bruker AV300, AV500, or
rapid loss of ligand, either O2 or phosphine, or rapid
decomposition. For the unprotonated compounds, the control
complex [Cp*Ru(dippp)(O2)][PF6] shows a similar irrever-
sible wave at a similar potential, so the relays are not key to the
decomposition. We have been unable to identify any of the
1
DRX500 spectrometers. H NMR chemical shifts are reported versus
TMS and referenced to residual solvent. 31P NMR chemical shifts are
reported relative to an 85% H3PO4 (aq) external standard. Mass spectra
were recorded on a Bruker Autoflex II MALDI-TOF. Elemental
analysis was performed by Atlantic Microlabs.
products of reduction of [Cp*Ru(PR NR′2)(O2)]+ with acid,
2
but 31P NMR spectra of reductions with Cp*2Fe are highly
suggestive of phosphine oxidation. The proposed RuIII−OOH
intermediate could undergo many undesirable reactions,
including ligand oxidation to form a phosphine oxide, which
has been proposed for these types of complexes.49e,62 The
stabilization of this possible hydroperoxo intermediate by the
pendent amine does not appear to be strong enough to prevent
this phosphine oxidation reaction from occurring. Phosphine
oxidation has been observed previously in O2 reduction with
Syntheses. Cp*Ru(PtBu2NPh2)Cl. The ligand PtBu2NPh (0.223 g,
2
0.54 mmol) was suspended in 30 mL of THF. A slurry of [Cp*RuCl]4
(0.145 g, 0.13 mmol) in 10 mL of THF was added over a period of 5
min. The dark orange solution was stirred at room temperature for 2
days. The solution was evaporated to dryness. The orange solid was
redissolved in 3 mL of toluene, filtered through Celite, and stored at
−20 °C for 2 days to give dark orange X-ray quality crystals of
Cp*Ru(PtBu2NPh2)Cl·(toluene). The toluene was removed under
vacuum to give Cp*Ru(PtBu2NPh2)Cl as an orange powder (0.200 g,
53% yield). NMR(CD2Cl2) 1H: 7.21 (m, 4H, Ar-H), 6.94 (m, 4H, Ar-
H), 6.76 (m, 2H, Ph-H), 4.36 (m, 2H, PCH2N), 3.81 (m, 2H,
PCH2N), 3.51 (m, 2H, PCH2N), 3.00 (m, 2H, PCH2N), 1.68 (s, 15H,
CpCH3), 1.39 (3-line pattern, 18H, tBu-CH3). 31P{1H}: 44.0 (s). Anal.
(Calcd) C 59.72 (59.51); H 7.54 (7.49); N 4.10 (4.08)
Ni(PR NR′2)+2 complexes,45 and remains a significant challenge
2
for the use of PR NR′2 ligands even for these relatively
2
substitutionally inert species in the presence of oxidizing
species. These results contrast with those of related
ruthenium−nickel thiolate complexes (L)Ru(NiS2N2) (L =
Cp* or η6-C6Me6, S2N2 is a dithiolate, diamine ligand).63 These
species bind O2 to form stable ruthenium(IV)−peroxo species,
and the hexamethylbenzene derivative is an active and stable
ORR catalyst that has been used in a molecular fuel cell.63c,d
Cp*Ru(PPh2NBn2)Cl. The ligand PPh2NBn (0.483 g, 1.0 mmol) was
2
suspended in 30 mL of toluene. To this suspension was added
Cp*Ru(PPh3)Cl (0.796 g, 0.98 mmol) slurried in 30 mL of toluene.
This mixture was refluxed for 10 min, at which time an orange solution
was formed. The orange solution was further refluxed for 36 h. It was
then cooled to room temperature and vacuum-dried to give an orange
oil. The orange oil was triturated with hexanes and vacuum-dried three
times to give a yellow-orange powder. This powder was collected by
filtration and washed once with 10 mL of acetone and then five times
with 5 mL of hexanes. The yellow-orange powder was dried under
vacuum to give Cp*Ru(PPh2NBn2)Cl (0.408 g, 55% yield). X-ray
quality crystals were obtained from CH2Cl2 layered with hexanes at
−20 °C. NMR(CD2Cl2) 1H: 7.60−6.93 (m, 20H, Ph-H), 4.06 (s, 2H,
PhCH2N), 3.63 (m, 2H, PCH2N), 3.43 (m, 2H, PCH2N), 3.42 6 (s,
2H, PhCH2N), 3.10 (m, 2H, PCH2N), 2.37 (m, 2H, PCH2N), 1.29 (s,
15H, CpCH3). 31P{1H}: 31.0 (s). Anal. (Calcd) C 64.27(63.69); H
6.44(6.28); N 3.71(3.71).
CONCLUSIONS
■
A series of Cp*Ru complexes with PR NR′2 ligands has been
2
prepared. All of the [Cp*Ru(PR NR′2)]+ complexes tightly bind
2
O2, partially reducing O2 to give formally a peroxide ligand. The
[Cp*Ru(PR NR′2)(O2)]+ complexes bind protons at the amine
2
near the O2 ligand, directing the proton to the O2 ligand
through a hydrogen bonding interaction. The same proton
directing is seen despite large differences in the basicity of the
pendent amines. These compounds thus allow a clear
visualization of one potentially key role of second-sphere
proton relays in the oxygen reduction reaction. The series of
compounds was made in order to adjust the properties of the
metal center, by varying the phosphine and Cp substituents,
and independently the properties of the relays, by varying the
amine substituents. However, it was found that the properties
of the metal center and the relays are intertwined, with changes
in phosphine substituents affecting the amine basicity and
change in amine substituent affecting the Ru redox potential.
The species formed upon reduction of the peroxo complexes
are unstable, and decomposition prevents these complexes from
acting as oxygen reduction catalysts. However, cyclic
voltammetry has shown that the positioned protons signifi-
cantly facilitate reduction of the peroxo complexes, perhaps by
protonating the O2 ligand upon reduction.
CpRu(PtBu2NBn2)Cl. The ligand PtBu2NBn (0.500 g, 1.1 mmol) was
2
dissolved in 30 mL of toluene. To this was added CpRu(PPh3)Cl
(0.730 g, 0.98 mmol) slurried in 30 mL of toluene. This mixture was
refluxed for 10 min, at which time an orange solution was formed. The
orange solution was further refluxed for 36 h. It was then cooled to
room temperature and vacuum-dried to give an orange oil. The orange
oil was triturated with hexanes and vacuum-dried three times to give a
yellow-orange powder. This powder was collected by filtration and
washed three times with 10 mL of hexanes. The yellow-orange powder
was dissolved in 5 mL of toluene, layered with 8 mL of hexanes, and
stored at −20 °C for two days, forming a yellow solid. This solid was
collected by filtration and dried under vacuum to give CpRu-
(PtBu2NBn2)Cl (0.452 g, 72% yield). NMR(CD2Cl2) 1H: 7.32−7.18 (m,
10H, Ph-H), 4.64 (s, 5H, Cp-H), 3.68 (s, 2H, PhCH2N), 3.61 (s, 2H,
PhCH2N), 3.37 (m, 2H, PCH2N), 2.68 (m, 2H, PCH2N), 2.53 (m,
2H, PCH2N), 2.42 (m, 2H, PCH2N), 1.03 (3-line pattern, 18H, tBu-
CH3). 31P{1H}: 51.8 Anal. (Calcd) C 58.09 (57.80); H 7.04 (7.16); N
4.31 (4.35).
EXPERIMENTAL SECTION
■
General Considerations. All manipulations were carried out
under a nitrogen atmosphere using standard glovebox and Schlenk
techniques unless otherwise specified. Solvents and reagents, including
15N-labeled aniline and 18O2 gas (97 atom % 18O), were purchased
from Aldrich unless otherwise noted. 1,3-Bis(diisopropylphosphino)-
propane (dippp) was purchased from Strem. CD2Cl2 was purchased
from Cambridge Isotope Laboratories and dried over CaH2 prior to
use. RuCl3·3H2O was purchased from Pressure Chemical Company.
[Cp*RuCl]4,64 Cp*Ru(PPh3)2Cl,65 CpRu(PPh3)2Cl,66 [H+DMF]-
Cp*Ru(dippp)Cl. To a solution of 1,3-bis(diisopropylphosphino)-
propane (dippp) (0.250 g, 0.90 mmol) in 15 mL of THF was added
[Cp*RuCl]4 (0.245 g, 0.23 mmol) slurried in 10 mL of THF. The
mixture was stirred overnight, forming an orange solution. This
solution was passed through a Celite plug while open to air to remove
a small amount of dark, insoluble material, and then dried under
vacuum to give a light orange powder. The orange powder was
dissolved in 3 mL of toluene, layered with 5 mL of hexanes, and stored
at −20 °C. On standing overnight, red crystals formed, as well as a
small amount of black oil. The red crystals of Cp*Ru(dippp)Cl were
isolated by filtration, with the black oil remaining on the glassware, and
washed with hexanes, giving (0.120 g, 24% yield) of Cp*Ru(dippp)Cl.
[OTf],67 and the ligands PtBu2NPh 40 and PPh2NBn were prepared
35
2
2
following literature procedures. Cp*Ru(PtBu2NBn2)Cl, [Cp*Ru-
(PtBu2NBn2)(O2)][PF6], and [Cp*Ru(PtBu2NBn2H)(O2)][PF6]2 were
prepared as reported previously.46 IR spectra were recorded using a
1
NMR(CD2Cl2) (The proton NMR spectrum is complex due to H
J
dx.doi.org/10.1021/ic3013987 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
1
and 31P coupling and has not been fully assigned.) H: 2.59 (m, 2H),
2.18 (m, 3H), 2.05 (m, 2H), 1.80 (m, 2H), 1.63 (s, 15H, CpCH3),
1.57 (m, 1H), 1.24−1.18 (m, 24H, CH(CH3)2). 31P{1H}: 32.5 (s).
Anal. (Calcd) C 54.99 (54.78); H 9.01 (8.93).
immediately turned deep blue, and white TlCl precipitated. After 5
min of stirring open to air, the blue color changed to a light yellow,
indicating formation of the O2 complex. The solution was stirred for
an additional 2 h, and then the TlCl was removed by filtration through
a plug of Celite. The solution was dried to a brown oil under vacuum,
triturated with hexanes, and dried again under vacuum to give
[Cp*Ru(dippp)(O2)][PF6] as a brown powder (0.090 g, 93%). X-ray
quality crystals were obtained from CH2Cl2 layered with hexanes at
−20 °C. NMR(CD2Cl2) (The proton NMR spectrum is complex due
to 1H and 31P coupling and has not been fully assigned.) 1H: 2.55 (m,
2H), 2.48 (m, 1H), 1.96−1.89 (m, 6H), 1.77 (s, 15H, CpCH3), 1.69
(m, 1H) 1.35 (m, 24H, CHCH3). 31P{1H}: 23.5 (s), −144.0 (m,
J19F−31P = 720 Hz). Anal. (Calcd) C 43.75 (43.54); H 7.35 (7.16).
X-ray Diffraction. Crystals were mounted on glass capillaries with
Paratone-N oil (Hampton Research) and frozen immediately in a cold
nitrogen gas stream at 100 K. X-ray diffraction data were collected on a
Bruker APEXII single crystal diffractometer coupled to a Bruker
APEXII CCD detector with graphite-monochromated Mo Kα
radiation (λ = 0.710 73 Å). The data was integrated and scaled
using SAINT, SADABS within the APEX2 software package by
Bruker.68 Solution by direct methods (SHELXS, SIR9769) produced a
complete heavy atom phasing model consistent with the proposed
structure. The structure was completed by difference Fourier synthesis
with SHELXL97.70,71 Scattering factors are from Waasmair and
Kirfel.72 All hydrogen atoms, including the hydrogen-bonding N−H
protons, were placed in geometrically idealized positions and
constrained to ride on their parent atoms with C−H and N−H
distances in the range 0.90−1.00 Å (a riding model). Isotropic thermal
parameters Ueq were fixed such that they were 1.2Ueq of their parent
atom Ueq for CH’s and NH’s, and 1.5Ueq of their parent atom Ueq in
the case of methyl groups. All non-hydrogen atoms were refined
anisotropically by full-matrix least-squares. Collection and refinement
data are presented in the Supporting Information. The structures of
Cp*Ru(PtBu2NPh2)Cl·(toluene) and [Cp*Ru(PPh2NBn2) (O2)]-
[OTf]·(hexane)1/3·(CH2Cl2)2/3 required restraints on the solvent
molecules. The structures of [Cp*Ru(PtBu2NBn2)Cl][PF6]·(acetone)1/2
and [CpRu(PtBu2NBn2)(O2)][OTf]·(CH2Cl2)2·(H2O) required re-
straints on the fluorines of the anions. The restraints used to fit the
structure of [Cp*Ru(PPh2NBn2H)(O2)][OTf]2·(acetone) are described
in the text.
[CpRu(PtBu2NBn2)(MeCN)][PF6]. To a solution of Cp*Ru(PPh2NBn
)
2
Cl (0.020 g, 0.03 mmol) in 2 mL of CH2Cl2 was added TlPF6 (0.11 g,
0.03 mmol) in 2 mL of CH2Cl2. After 30 min, the solution was passed
through Celite to remove TlCl, leaving a red solution. To this solution
was added an excess (approximately 5 μL) of acetonitrile, resulting in
an immediate color change to yellow. This solution was dried to a
yellow oil under vacuum, redissolved in acetonitrile, and layered with
diethyl ether. On standing for several days at −35 °C, large yellow
crystals of [CpRu(PtBu2NBn2)(MeCN)][PF6]·(diethyl ether) were
obtained. The ether could be removed under vacuum to give
[CpRu(PtBu2NBn2)(MeCN)][PF6] (0.008 g, 32% yield). NMR-
1
(CD2Cl2) H: 7.32−7.27 (m, 8H, Ph-H), 7.16 (d, 2H, Ph-H), 4.84
(s, 5H, Cp-H), 3.70 (s, 4H, PhCH2N), 3.06 (m, 2H, PCH2N), 2.58−
2.50 (m, 6H, PCH2N), 2.24 (s, 3H, NCCH3), 1.04 (3-line pattern,
18H, tBu-CH3) 31P{1H}: 52.6 (s), −144.0 (m, J19F−31P = 720 Hz).
Anal. (Calcd) C 49.85 (49.87); H 6.23 (6.09); N 5.33 (5.29).
[Cp*Ru(PtBu2NBn2)Cl][PF6]. To a solution of Cp*Ru(PtBu2NBn2)Cl
(0.100 g, 0.14 mmol) in 3 mL of CH2Cl2 was added a solution of
ferrocenium hexafluorophosphate (0.046 g, 0.14 mmol) in 8 mL of
CH2Cl2. The solution turned greenish-brown upon addition, and was
stirred for an additional 20 min. The solution was vacuum-dried to a
dark oil which was then triturated with 5 mL of toluene and filtered
through paper, separating the dark green solid from the orange,
ferrocene-containing filtrate. The green solid was then washed through
the paper with 5 mL of acetone. The green liquid was layered with 5
mL of hexane and stored at −20 °C overnight, affording large green
crystals of [Cp*Ru(PtBu2NBn2)Cl][PF6] (0.105 g, 87%). MS: (MALDI-
TOF, pyrene matrix) 714.2 [M+], 679.2 [M+ − Cl]. Anal. (Calcd) C
50.49 (50.38); H 6.52 (6.34); N 3.32(3.26).
[Cp*Ru(PtBu2NPh2)(O2)][OTf]. To an aerobic solution of Cp*Ru-
(PtBu2NPh2)Cl (0.053 g, 0.08 mmol) in 10 mL of acetone was added a
solution of TlOTf (0.027 g, 0.08 mmol) in 5 mL of acetone. The
yellow-orange solution immediately turned yellow-brown, and white
TlCl precipitated. The solution was stirred for an additional 2 h, and
then the TlCl was removed by filtration through a plug of Celite. The
solution was dried to a brown oil under vacuum, triturated with
hexanes, and dried again under vacuum to give [Cp*Ru(PtBu2NPh2)-
(O2)][OTf] as a yellow-brown powder (0.059 g, 93%). X-ray quality
crystals were obtained by evaporation of acetone solvent. NMR-
(CD2Cl2) 1H: 7.34 (m, 4H, Ar-H), 7.00 (m, 6H, Ar-H), 3.80 (m, 2H,
PCH2N), 3.74 (m, 2H, PCH2N), 3.57 (m, 2H, PCH2N), 3.24 (m, 2H,
PCH2N), 1.82 (s, 15H, CpCH3), 1.48 (3-line pattern, 18H, tBu-CH3).
31P{1H}: 30.2 (s). IR(KBr) ν(O−O) = 930 cm−1 [18O = 880 cm−1].
Anal. (Calcd) C 50.20 (50.47); H 6.20 (6.29); N 3.49 (3.36).
[Cp*Ru(PPh2NBn2)(O2)][OTf]. To an aerobic solution of Cp*Ru-
(PPh2NBn2)Cl (0.070 g, 0.09 mmol) in 10 mL of acetone was added a
solution of TlOTf (0.033 g, 0.09 mmol) in 5 mL of acetone. The
yellow-orange solution immediately turned yellow-brown, and white
TlCl precipitated. The solution was stirred for an additional 2 h, and
then the TlCl was removed by filtration through a plug of Celite. The
solution was dried to a brown oil under vacuum. This oil was
recrystallized from acetone/hexanes at −20 °C for several days to give
after filtration [Cp*Ru(PPh2NBn2)(O2)][OTf]·(acetone) as a yellow-
brown microcrystalline powder. The acetone could be removed in
vacuo to give [Cp*Ru(PPh2NBn2)(O2)][OTf] (0.065 g, 80%). X-ray
quality crystals were obtained from CH2Cl2 layered with hexanes at
−20 °C. NMR(CD2Cl2) 1H: 7.79 (m, 4H, Ph-H), 7.59−7.43 (m, 12H,
Ph-H), 6.93−6.85 (m, 4H, Ph-H), 4.06 (s, 2H, PhCH2N), 3.63 (m,
2H, PCH2N), 3.43 (m, 2H, PCH2N), 3.42 6 (s, 2H, PhCH2N), 3.10
(m, 2H, PCH2N), 2.37 (m, 2H, PCH2N), 1.29 (s, 15H, CpCH3).
31P{1H}: 22.0 (s). IR(KBr) ν(O−O) = 926 cm−1 [18O = 879 cm−1].
Anal. (Calcd) for [Cp*Ru(PPh2NBn2)(O2)][OTf]·(acetone) C 55.38
(55.11); H 5.62 (5.68); N 3.00 (2.92).
Electrochemistry. Cyclic voltammetry (CV) was performed under
N2 using a CH Instruments 600D apparatus equipped with glassy
carbon (1.0 mm diam) working electrode, glassy carbon rod auxiliary
electrode, and Ag/Ag(NO)3 (0.01 M in acetonitrile) reference
electrode. CV was carried out in dichloromethane with 0.1 M
[nBu4N][PF6] as supporting electrolyte. [nBu4N][PF6] was recrystal-
lized three times from ethanol and dried under vacuum prior to use.
Ferrocene or decamethylferrocene was added as internal standard, and
all potentials are reported versus the Cp2Fe+/0 couple. In all cases,
solutions were ca. 1−2 mM in concentration of Ru complex.
DFT Calculations. Calculations were performed with Gaussian09.60
Geometry optimizations were performed in the gas phase on cations
[Cp*Ru(PtBu2NBn2)(η1-O2)]+, [Cp*Ru(PtBu2NPh2H)(O2)]+2, and
[Cp*Ru(PtBu2NPh2)(O2)]+. Calculations were performed with the
BP86 functional, with the SDD basis set and effective core potential
for Ru, and the 6-31G** basis set for all other atoms. Optimization of
[Cp*Ru(PtBu2NPh2)(O2)]+ was performed starting for the geometry of
the crystal structure. All geometries were confirmed to be minima by
vibrational analysis (NImag = 0).
ASSOCIATED CONTENT
■
S
* Supporting Information
Additional NMR spectra, crystallographic refinement data and
CIFs, IR spectra of O2 complexes, cyclic voltammograms, DFT
structure of [Cp*Ru(PtBu2NPh2)(O2)]+, and coordinates of
DFT optimized molecular geometries. This material is available
free of charge via the Internet at http://pubs.acs.org
[Cp*Ru(dippp)(O2)][PF6]. To an aerobic solution of Cp*Ru(dippp)
Cl (0.075 g, 0.14 mmol) in 10 mL of acetone was added a solution of
TlPF6 (0.051 g, 0.15 mmol) in 5 mL of acetone. The orange solution
K
dx.doi.org/10.1021/ic3013987 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(25) Carver, C. T.; Matson, B. D.; Mayer, J. M. J. Am. Chem. Soc.
2012, 134, 5444.
(26) DuBois, M. R.; DuBois, D. L. Chem. Soc. Rev. 2009, 38, 62.
(27) Small, Y. A.; DuBois, D. L.; Fujita, E.; Muckerman, J. T. Energy
Environ. Sci. 2011, 4, 3008.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: mayer@chem.washington.edu.
Notes
(28) Liu, T.; Chen, S.; O’Hagan, M. J.; DuBois, M. R.; Bullock, R. M.;
DuBois, D. L. J. Am. Chem. Soc. 2012, 134, 6257.
The authors declare no competing financial interest.
́
(29) Lounissi, S.; Capon, J.-F.; Gloaguen, F.; Matoussi, F.; Petillon, F.
ACKNOWLEDGMENTS
Y.; Schollhammer, P.; Talarmin, J. Chem. Commun. 2011, 47, 878.
(30) Camara, J. M.; Rauchfuss, T. B. Nat. Chem. 2012, 4, 26.
(31) Helm., M. L.; Stewart, M. P.; Bullock, R. M.; DuBois, M. R.;
DuBois, D. L. Science 2011, 333, 863.
■
We thank Dr. John Roberts for assistance with electrochemical
measurements, Dr. Johanna Blacquiere for help with mass
spectrometry, and Dr. Mary Rakowski DuBois and Dr. Michael
Mock for synthetic advice and helpful discussion. This work is
supported as part of 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.
(32) Wiese, S.; Kilgore, U. J.; DuBois, D. L.; Bullock, R. M. ACS
Catal. 2012, 2, 720.
(33) Wilson, A. D.; Shoemaker, R. K.; Miedaner, A.; Muckerman, J.
T.; DuBois, D. L.; DuBois, M. R. Proc. Natl. Acad. Sci. U.S.A. 2007,
104, 6951.
(34) Kilgore, U. J.; Stewart, M. P.; Helm, M. L.; Dougherty, W. G.;
Kassel, W. S.; DuBois, M. R.; DuBois, D. L.; Bullock, R. M. Inorg.
Chem. 2011, 50, 10908.
REFERENCES
■
(35) Fraze, K.; Wilson, A. D.; Appel, A. M.; DuBois, M. R.; Dubois,
D. L. Organometallics 2007, 26, 3918.
(1) Winter, M.; Brodd, R. J. Chem. Rev. 2004, 104, 4245.
(2) Borup, R.; Meyers, J.; Pivovar, B.; Kim, Y. S.; Mukundan, R.;
Garland, N.; Myers, D.; Wilson, M.; Garzon, F.; Wood, D.; Zelenay,
P.; More, K.; Stroh, K.; Zawodzinski, T.; Boncella, J.; McGrath, J. E.;
Inaba, M.; Miyatake, K.; Hori, M.; Ota, K.; Ogumi, Z.; Miyata, S.;
Nishikata, A.; Siroma, Z.; Uchimoto, Y.; Yasuda, K.; Kimijima, K.-I.;
Iwishita, N. Chem. Rev. 2007, 107, 3904.
(3) Gewirth, A. A.; Thorum, M. S. Inorg. Chem. 2010, 49, 3557.
(4) Warren, J. J.; Tronic, T. A.; Mayer, J. M. Chem. Rev. 2010, 110,
6961.
(5) Saveant, J.-M. Chem. Rev. 2008, 108, 2348.
́
(6) Collman, J. P.; Boulatov, R.; Sunderland, C. J.; Fu, L. Chem. Rev.
2004, 104, 561.
(7) Shigehara, K; Anson, F. C. J. Phys. Chem. 1982, 86, 2776.
(8) Fukuzumi, S.; Okamoto, K.; Gros, C. P.; Guilard, R. J. Am. Chem.
Soc. 2004, 126, 10441.
(9) Wisener, K.; Ohms, D.; Neumann, V.; Franke, R. Mater. Chem.
Phys. 1989, 22, 457.
(10) Chang, C. J.; Deng, Y.; Shi, C.; Chang, C. K.; Anson, F. C.;
Nocera, D. G. Chem. Commun. 2000, 1355.
(11) Chang, C. J.; Loh, Z.-H.; Shi, C.; Anson, F. C.; Nocera, D. G. J.
Am. Chem. Soc. 2004, 126, 10013.
(12) McCrory, C. C. L.; Ottenwaelder, X.; Stack, T. D. P.; Chidsey,
C. E. D. J. Phys. Chem. A 2007, 111, 12641.
(13) McCrory, C. C. L.; Devadoss, A.; Ottenwaelder, X.; Lowe, R.
D.; Stack, T. D. P.; Chidsey, C. E. D. J. Am. Chem. Soc. 2011, 133,
3696.
(14) Ward, A. L.; Elbaz, L.; Kerr, J. B.; Arnold, J. Inorg. Chem. 2012,
51, 4694.
(36) Kilgore, U. J.; Roberts, J. A. S.; Pool, D. H.; Appel, A. M.;
Stewart, M. P.; DuBois, M. R.; Dougherty, W. G.; Kassel, W. S.;
Bullock, R. M.; DuBois, D. L. J. Am. Chem. Soc. 2011, 133, 5861.
(37) Yang, J. Y.; Chen., S.; Dougherty, W. G.; Kassel, W. S.; Bullock,
R. M.; DuBois, D. L.; Raugei, S.; Rousseau, R.; Dupuis, M.; DuBois, M.
R. Chem. Commun. 2010, 46, 8618.
(38) Lee, C. H.; Dogutan, D. K.; Nocera, D. G. J. Am. Chem. Soc.
2011, 133, 8775.
(39) Jacobsen, G. M.; Yang, J. Y.; Twamley, B.; Wilson, A. D.;
Bullock, R. M.; DuBois, M. R.; DuBois, D. L. Energy Environ. Sci. 2008,
1, 167.
(40) Wiedner, E. S.; Yang, J. Y.; Dougherty, W. G.; Kassel, W. S.;
Bullock, R. M.; DuBois, M. R.; DuBois, D. L. Organometallics 2010, 29,
5390.
(41) 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.
(42) Seu, C. S.; Appel, A. M.; Doud, M. D.; DuBois, D. L.; Kubiak, C.
P. Energy Environ. Sci. 2012, 5, 6480.
(43) Weiss, C. J.; Groves, A. N.; Mock., M. T.; Dogherty, W. G.;
Kassel, S. K.; Helm, M. L.; DuBois, D. L.; Bullock, M. R. Dalton Trans.
2012, 41, 4517.
(44) Mock, M. T.; Chen, S.; Rouseau, R.; O’Hagan, M. J.; Dougherty,
W. G.; Kassel, W. S.; DuBois, D. L.; Bullock, R. M. Chem. Commun.
2011, 47, 12212.
(45) Yang, J. Y.; Bullock, R. M.; Dougherty, W. G.; Kassel, W. S.;
Twamley, B.; DuBois, D. L.; DuBois, M. R. Dalton Trans. 2010, 39,
3001.
(46) Tronic, T. A.; DuBois, M. R.; Kaminsky, W.; Coggins, M. K.;
Liu, T.; Mayer, J. M. Angew. Chem., Int. Ed. 2011, 50, 10936.
(47) Lin, W.; Wilson, S. R.; Girolami, G. S. Organometallics 1996, 16,
2987.
(15) Fukuzumi, S.; Kotani, H.; Lucas, H. R.; Doi, K.; Suenobu, T.;
Peterson, R. L.; Karlin, K. D. J. Am. Chem. Soc. 2010, 132, 6874.
(16) Kaila, V. R. I.; Verkhovsky, M. I.; Wikstrom, M. Chem. Rev.
̈
2010, 110, 7062.
(17) Meunier, B.; de Visser, S. P.; Shaik, S. Chem. Rev. 2004, 104,
3947.
(18) Solomon, E. I.; Chen, P.; Metz., M.; Lee, S.-K.; Palmer, A. E.
Angew. Chem., Int. Ed. 2001, 40, 4570.
(19) Collman, J. P.; Devaraj, N. K.; Decreau, R. A.; Yang, Y.; Yan, Y.-
́
L.; Ebina, W.; Eberspacher, T. A.; Chidsey, C. E. D. Science 2007, 315,
1565.
(48) Izutsu, K. Acid−Base Dissociation Constants in Dipolar Aprotic
Solvents; Blackwell: Oxford, 1990.
(49) (a) Kirchner, K.; Mauthner, K.; Mereiter, K.; Schmid, R. J.
Chem. Soc., Chem. Comm. 1993, 892. (b) Mauthner, K.; Mereiter, K.;
Schmid, R.; Kirchner, K. Inorg. Chim. Acta 1995, 236, 95. (c) Sato, M.;
Asai, M. J. Organomet. Chem. 1996, 508, 121. (d) Lindner, E.;
Haustein, M.; Fawzi, R.; Steimann, M.; Wegner, P. Organometallics
1994, 13, 5021. (e) Jia, G.; Ng, W. S.; Chu, H. S.; Wong, W.-T.; Yu,
N.-T.; Williams, I. D. Organometallics 1999, 18, 3597. (f) de los Ríos,
(20) Shook, R. L.; Borovik, A. S. Inorg. Chem. 2010, 49, 3646.
(21) Collman, J. P.; Decrea
69, 3546.
́
u, R. A.; Zhang, C. J. Am. Chem. Soc. 2004,
I.; Jimen
Organometallics 1996, 15, 4565.
́
ez Tenorio, M.; Padilla, J.; Puerta, M. C.; Valerga, P.
(22) McGuire, R.; Dogutan, D. K.; Teets, T. S.; Suntivich, J.; Shao-
Horn, Y.; Nocera, D. G. Chem. Sci. 2010, 1, 411.
́
(50) Palacios, M. D.; Puerta, M. C.; Valerga, P.; Lledos, A.; Veilly, E.
Inorg. Chem. 2007, 46, 6958.
(23) Dogutan, D. K.; Stoian, S. A.; McGuire, R.; Schwalbe, M.; Teets,
T. S.; Nocera, D. G. J. Am. Chem. Soc. 2011, 133, 131.
(24) Shook, R. L.; Peterson, S. M.; Greaves, J.; Moore, C.; Rheingold,
(51) (a) Praetorius, J. M.; Allen, D. P.; Wang, R.; Webb, J. D.; Grein,
A. L.; Borovik, A. S. J. Am. Chem. Soc. 2011, 133, 5810.
F.; Kennepohl, P.; Crudden, C. M. J. Am. Chem. Soc. 2008, 130, 3724.
L
dx.doi.org/10.1021/ic3013987 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(b) Covelli, D. Probing the Electronic Structure of Dioxygen as a Ligand:
Using X-Ray Absorption Spectroscopy to Quantify Backbonding. Ph.D.
Thesis, University of British Columbia, Vancouver, British Columbia,
Canada, 2011.
(52) Jeffrey, G. A. An Introduction to Hydrogen Bonding; Oxford
University Press: Oxford, U.K., 1997.
(53) Piertzak, M.; Stefaniak, L.; Pozharskii, A. F.; Ozeryanskii, V. A.;
Nowicka-Scheibe, J.; Grech, E.; Webb, G. A. J. Phys. Org. Chem. 2000,
13, 35.
(54) Axenrod, T.; Pregosin, P. S.; Wieder, M. J.; Becker, E. D.;
Bradley, R. B.; Milne, G. W. A. J. Am. Chem. Soc. 1971, 93, 6536.
(55) Kaljurand, I.; Kutt, A.; Soovali, L.; Rodima, T.; Maemets, V.;
̈
̈
̈
Leito, I.; Koppel, I. A. J. Org. Chem. 2005, 70, 1019.
(56) Noviandri, I.; Brown, K. N.; Fleming, D. S.; Gulyas, P. T.; Lay,
P. A.; Masters, A. F.; Phillips, L. J. Phys. Chem. B 1999, 103, 6713.
(57) Concepcion, J. J.; Tsai, M.-K.; Muckerman, J. T.; Meyer, T. J. J.
Am. Chem. Soc. 2010, 132, 1545.
(58) Lin, X.; Hu, X.; Concepcion, J. J.; Chen, Z.; Liu, S.; Meyer, T.;
Yang, W. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 15669.
(59) Closed shell singlet and triplet η1-peroxo species were located as
minima by geometry optimization in the gas phase at the BP86/6-
31G** (C, H, N, O, P) SDD (Ru) level of theory using Gaussian09.60
(60) Frisch, M. J.; et al. Gaussian 09. See Supporting Information for
complete citation.
(61) Yu, H.; Fu, Y.; Guo, Q.; Lin, Z. Organometallics 2009, 28, 4443.
(62) Man, L. M.; Zhu, J.; Ng, S. M.; Zhou, Z.; Yin, C.; Lin, Z.; Lau, C.
P. Organometallics 2004, 23, 6214.
(63) (a) Reynolds, M. A.; Rauchfuss, T. B.; Wilson, S. R.
Organometallics 2003, 22, 1619−1625. (Addition: Organometallics
2003, 22, 4620.) (b) Canaguier, S.; Fontecave, M.; Artero, V. Eur. J.
Inorg. Chem. 2011, 1094−1099. (c) Kim, K.; Matsumoto, T.;
Robertson, A.; Nakai, H.; Ogo, S. Chem. Asian J. 2012, 7, 1394−
1400. (d) Matsumoto, T.; Kim, K.; Ogo, S. Angew. Chem., Int. Ed.
2011, 50, 11202−11205.
(64) Fagan, P. J.; Ward, M. D.; Calabrese, J. C. J. Am. Chem. Soc.
1989, 111, 1698.
(65) Boren, B. C.; Narayan, S.; Rasmussen, L. K.; Zhang, L.; Zhao,
H.; Lin, Z.; Jia, G.; Fokin, V. V. J. Am. Chem. Soc. 2008, 130, 9823.
(66) Bruce, M. I.; Hameister, C.; Swincer, A. G.; Wallis, R. C. Inorg.
Synth. 1982, 21, 78.
(67) Faviera, I.; Dunach, E. Tetrahedron Lett. 2004, 45, 3393.
̃
(68) Bruker APEX2 (Version 2.1-4), SAINT (Version 7.34A), SADABS
(Version 2007/4); BrukerAXS Inc.: Madison, WI, 2007.
(69) (a) Altomare, A.; Burla, C.; Camalli, M.; Cascarano, L.;
Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.;
Spagna, R. J. Appl. Crystallogr. 1999, 32, 115. (b) Altomare, A;
Cascarano, G; Giacovazzo, C; Guagliardi, A. J. Appl. Crystallogr. 1993,
26, 343.
(70) Sheldrick, G. M. SHELXL-97, Program for the Refinement of
Crystal Structures; University of Gottingen: Gottingen, Germany, 1997.
̈
̈
(71) Mackay, S.; Edwards, C.; Henderson, A.; Gilmore, C.; Stewart,
N.; Shankland, K.; Donald, A. MaXus; University of Glasgow:
Glasgow, Scotland, 1997.
(72) Waasmaier, D.; Kirfel, A. Acta Crystallogr., Sect. A 1995, 51, 416.
M
dx.doi.org/10.1021/ic3013987 | Inorg. Chem. XXXX, XXX, XXX−XXX