Aerobic and hydrolytic decomposition of pseudotetrahedral nickel phenolate complexes

Article abstract of DOI:10.1021/ic300551z

Pseudotetrahedral nickel(II) phenolate complexes Tp^R,MeNi-OAr (Tp^R,Me = hydrotris(3-R-5-methylpyrazol-1-yl)borate; R = Ph {1a}, Me {1b}; OAr = O-2,6-ⁱPr₂C₆H₃) were synthesized as models for nickel-substituted copper amine oxidase apoenzyme, which utilizes an N₃O (i.e., His₃Tyr) donor set to activate O₂ within its active site for oxidative modification of the tyrosine residue. The bioinspired synthetic complexes 1a,b are stable in dilute CH₂Cl₂ solutions under dry anaerobic conditions, but they decompose readily upon exposure to O₂ and H₂O. Aerobic decomposition of 1a yields a range of organic products consistent with formation of phenoxyl radical, including 2,6-diisopropyl-1,4-benzoquinone, 3,5,3′,5′-tetraisopropyl-4,4′-diphenodihydroquinone, and 3,5,3′,5′-tetraisopropyl-4,4′-diphenoquinone, which requires concurrent O₂ reduction. The dimeric product complex di[hydro{bis(3-phenyl-5-methylpyrazol-1-yl)(3-ortho-phenolato-5-methylpyrazol-1- yl)borato}nickel(II)] (2) was obtained by ortho C-H bond hydroxylation of a 3-phenyl ligand substituent on 1a. In contrast, aerobic decomposition of 1b yields a dimeric complex [Tp^Me,MeNi]₂(μ-CO₃) (3) with unmodified ligands. However, a unique organic product was recovered, assigned as 3,4-dihydro-3,4-dihydroxy-2,6-diisopropylcyclohex-5-enone on the basis of ¹H NMR spectroscopy, which is consistent with dihydroxylation (i.e., addition of H₂O₂) across the meta and para positions of the phenol ring. Initial hydrolysis of 1b yields free phenol and the known complex [Tp^Me,MeNi(μ-OH)]₂, while hydrolysis of 1a yields an uncharacterized intermediate, which subsequently rearranges to the new sandwich complex [(Tp^Ph,Me)₂Ni] (4). Autoxidation of the released phenol under O₂ was observed, but the reaction was slow and incomplete. However, both 4 and the in situ hydrolysis intermediate derived from 1a react with added H₂O₂ to form 2. A mechanistic scheme is proposed to account for the observed product formation by convergent oxygenation and hydrolytic autoxidation pathways, and hypothetical complex intermediates along the former were modeled by DFT calculations. All new complexes (i.e., 1a,b and 2-4) were fully characterized by FTIR, ¹H NMR, and UV-vis-NIR spectroscopy and by X-ray crystallography.

Full text of DOI:10.1021/ic300551z

Article
pubs.acs.org/IC
Aerobic and Hydrolytic Decomposition of Pseudotetrahedral Nickel
Phenolate Complexes
†
‡
‡
,†
Tapash Deb, Gregory T. Rohde, Victor G. Young, Jr., and Michael P. Jensen*
†
Department of Chemistry and Biochemistry, Ohio University, Athens, Ohio 45701, United States
‡
X-ray Crystallographic Facility, Department of Chemistry, University of Minnesota, Minneapolis, Minnesota 55455, United States
S Supporting Information
*
ABSTRACT: Pseudotetrahedral nickel(II) phenolate com-
R,Me
R,Me
plexes Tp Ni-OAr (Tp
= hydrotris(3-R-5-methylpyr-
azol-1-yl)borate; R = Ph {1a}, Me {1b}; OAr = O-
i
2
,6- Pr C H ) were synthesized as models for nickel-
2 6 3
substituted copper amine oxidase apoenzyme, which utilizes
an N O (i.e., His Tyr) donor set to activate O within its active
3
3
2
site for oxidative modification of the tyrosine residue. The
bioinspired synthetic complexes 1a,b are stable in dilute
CH Cl solutions under dry anaerobic conditions, but they
2
2
decompose readily upon exposure to O and H O. Aerobic decomposition of 1a yields a range of organic products consistent
2
2
with formation of phenoxyl radical, including 2,6-diisopropyl-1,4-benzoquinone, 3,5,3′,5′-tetraisopropyl-4,4′-diphenodihydroqui-
none, and 3,5,3′,5′-tetraisopropyl-4,4′-diphenoquinone, which requires concurrent O reduction. The dimeric product complex
2
di[hydro{bis(3-phenyl-5-methylpyrazol-1-yl)(3-ortho-phenolato-5-methylpyrazol-1-yl)borato}nickel(II)] (2) was obtained by
ortho C−H bond hydroxylation of a 3-phenyl ligand substituent on 1a. In contrast, aerobic decomposition of 1b yields a dimeric
Me,Me
complex [Tp
Ni] (μ-CO ) (3) with unmodified ligands. However, a unique organic product was recovered, assigned as 3,4-
2 3
1
dihydro-3,4-dihydroxy-2,6-diisopropylcyclohex-5-enone on the basis of H NMR spectroscopy, which is consistent with
dihydroxylation (i.e., addition of H O ) across the meta and para positions of the phenol ring. Initial hydrolysis of 1b yields free
2
2
Me,Me
phenol and the known complex [Tp
subsequently rearranges to the new sandwich complex [(Tp
Ni(μ-OH)] , while hydrolysis of 1a yields an uncharacterized intermediate, which
2
Ph,Me
) Ni] (4). Autoxidation of the released phenol under O was
2
2
observed, but the reaction was slow and incomplete. However, both 4 and the in situ hydrolysis intermediate derived from 1a
react with added H O to form 2. A mechanistic scheme is proposed to account for the observed product formation by
2
2
convergent oxygenation and hydrolytic autoxidation pathways, and hypothetical complex intermediates along the former were
1
modeled by DFT calculations. All new complexes (i.e., 1a,b and 2−4) were fully characterized by FTIR, H NMR, and UV−vis−
NIR spectroscopy and by X-ray crystallography.
1
.0. INTRODUCTION
been prepared, either by stoichiometric O addition to
2
1
9−23,28,29,37−39,43−45
Ni(I),
precursors.
or H O2 addition to Ni(II)
No biological role for nickel has
Dioxygen activation is fundamentally important in catalysis and
bioinorganic chemistry. Significant work has been directed at
enzymatic and biomimetic iron and copper centers that react
2
2
0,24−27,32−36,40−42
been established in humans, but nickel toxicology is of
46−48
1
−10
interest.
Elevated intracellular levels are associated with
with O2,
yet nickel/O chemistry and biochemistry are
2
mutagenesis, and one proposed mechanism involves catalytic
depletion of cellular antioxidants, which implies formation of
relatively limited in scope. Only one nickel-dependent oxidase
enzyme has been characterized so far, namely acireductone
dioxygenase (Ni-ARD), for which functional turnover
modeling has been elicited from synthetic Ni(II) complexes
of aryl-appended tris(pyridylmethyl)amine (TPA) ligands.
Synthetic Ni(II) complexes of electron-rich tris(oximato)amine
and diamido macrocyclic ligands catalyze aerobic substrate
48
1
1
reduced oxygen species.
Copper amine oxidases (CAOs) activate O to effect post-
2
1
2
translational oxidation of an active-site tyrosine residue in the
expressed apoenzyme to a catalytically essential 2,4,5-
trihydroxyphenylalanine quinone (i.e., TPQ) cofactor in the
49−52
1
3,14
active holoenzyme (Scheme 1).
Our present work was
oxidations,
and macrocyclic complexes of Ni(II) catalyze
inspired by reports that TPQ biogenesis is also observed in
apoenzymes reconstituted with nickel, comprising a second
oxygen atom transfer from suitable precursors to organic
substrates.
1
5−17
Stoichiometric oxygenation reactions of Ni-
53,54
1
8
19−26
example of biological dioxygenase activity for this metal.
X-
(
II)−thiolato complexes are also known. Superoxo,
25−31 31−42
ray crystallography revealed a pseudotetrahedral N O donor set
peroxo,
and oxo
complexes supported by diketimine
3
2
1,22,43
24,25,27,35,36,40−42
(
nacnac),
TPA,
hydrotris(pyrazolyl)-
3
0−34
borate (Tp),
tmc),
tetraazamacrocycles (e.g., tetramethylcyclam,
and related polydentate ligands
Received: March 14, 2012
Published: June 15, 2012
2
0,26,28,29
19,23,37−39,44
have
©
2012 American Chemical Society
7257
dx.doi.org/10.1021/ic300551z | Inorg. Chem. 2012, 51, 7257−7270

Inorganic Chemistry
Article
Scheme 1
5
8,59,65,66
in the apoenzyme active site, derived from three histidine
imidazoles and one tyrosine side chain, which is the TPQ
extremely toxic and must be properly handled and disposed of!).
,6-Diisopropylphenol (Alfa Aesar) was reacted with NaH in toluene
2
1
4
9
to afford the sodium salt of the conjugate base. H NMR data were
recorded on Bruker 300 Ultrashield and Varian Unity 500
spectrometers and processed using the MestReNova software suite
precursor. Following active-site binding of copper and O , a
2
Cu(II)−phenolate LMCT species intermediate absorbing at
3
50 nm was observed, which exhibited isosbestic decay to a 480
(
Mestrelab Research, Santiago de Compostela, Spain); spectra were
referenced internally to residual CHCl , CHDCl , and C CHD
2
nm chromophore characteristic of TPQ in the mature
holoenzyme. Other consensus intermediates include an
unobserved Cu(II)−peroxyquinone complex,
undergoes heterolysis to a structurally characterized dopaqui-
none intermediate, but the mechanistic details of the O2
activation step(s) remain to be elucidated.
5
0
3
2
6
D
5
solvents (7.24, 5.32, and 2.08 ppm, respectively). Solution magnetic
4
9−52
which
67
moments were determined by the Evans NMR method. FT-IR
spectra were recorded from KBr pellets on a Thermo-Electron Nicolet
380 spectrophotometer. UV−vis−NIR spectra were recorded on an
Agilent HP-8453 diode-array spectrophotometer. Elemental analyses
were performed by Atlantic Microlabs (Norcross, GA). Spectroscopic
data for complexes 1a,b are shown in Figures S1−S8 of the Supporting
Information and summarized below.
Ni−O-2,6- Pr C H (1a). A sample of
NiCl (150 mg, 0.26 mmol) dissolved in dichloromethane (10
mL) was added to a slurry of NaO-2,6- Pr C H (57 mg, 0.29 mmol)
in CH Cl (10 mL). The color changed from pale pink to dark green
when the solutions were combined. After stirring for 2.0 h, the solution
was filtered and evaporated to dryness. The filtrate was redissolved in a
minimal amount of CH Cl , and green crystals were obtained by slow
diffusion of n-hexane at −30 °C. Yield: 171 mg (0.24 mmol, 92%).
Anal. Calcd (found) for C H BN NiO, 1a: C, 70.13 (70.27); H, 6.31
49
TpCu(II)−OAr complexes with pseudotetrahedral N O
3
5
2,55
ligand fields have been reported as CAO models,
wherein
the Tp ligand models the facial array of imidazole donors and
Ph,Me
i
2
.1. Synthesis of Tp
the phenolate coligand completes an N O ligand field akin to
2 6 3
3
Ph,Me
tBu,R
Tp
the CAO active site. However, the complexes Tp Cu−O−
C H -4-F (R = Pr, Bu) were unreactive with O , while
i
i
t
52
2
6
3
6
4
2
2
2
analogues with ortho-disubstituted phenolates decomposed
55
even under inert atmosphere. Similar behavior was also
56
reported for (nacnac)Cu−OAr complexes. Compared to the
TpCu−OAr complexes, nickel analogues might be expected to
exhibit higher stability, owing to relatively cathodic Ni(I/II)
redox couples, but such complexes have not been reported.
2
2
4
2
45
6
1
(
5.88); N, 11.68 (11.69). H NMR (CD Cl , 293 K; δ, ppm): 74.1
2 2
Building on previous work with O -sensitive Ni(II)−
(3H, 4-pz); 42 (2H, br, −CHMe ); 39.3 (2H, meta); 14.4 (6H, br, 3-
2
2
R,Me
57−61
arylthiolate complexes Tp Ni−SAr,
we report herein
ortho); 8.9 (3H, 3-para); 7.1 (6H, 3-meta); 2.7 (12H, br, −CHMe );
2
R,Me
R,Me
the bioinspired complexes Tp Ni−OAr (Tp
= hydrotris-
−2.2 (9H, 5-Me); −19.9 (1H, B-H); −43.4 (1H, para). μeff = 2.72 μ
B
6
2,63
−1
−1
(
(
CDCl , 293 K). UV−vis−NIR (CH Cl ; λ , nm; ε, M cm ): 476
{
=
3-R-5-Me-pyrazol-1-yl}borate;
R = Ph (1a), Me (1b); Ar
3
2
2
max
−
1
i
1900); 612 (800); 995 (200). IR (KBr, cm ): 2543, ν(B−H).
2,6- Pr C H ). These are thermally stable in anaerobic
2 6 3
Me,Me
i
2
Tp
.2. Synthesis of Tp
Ni−O-2,6- Pr C H (1b). A sample of
2
6
3
solutions but decompose under O . Observed organic products
2
Me,Me
NiCl (150 mg, 0.38 mmol) dissolved in dichloromethane (10
were consistent with formation of phenoxyl radical and
i
mL) was added to a slurry of NaO-2,6- Pr C H (75 mg, 0.38 mmol)
2
6
3
concurrent reduction of O , and subsequent aromatic oxidation
2
in CH Cl (10 mL). The color changed from pale pink to dark orange
2
2
chemistry was observed, at either a 3-Ph pyrazole substituent of
a or the phenolato ring of 1b. Taken together, these reactions
immediately after the solutions were combined. After stirring for 2.0 h,
the solution was filtered and evaporated to dryness. The filtrate was
redissolved in a minimal amount of CH Cl , and orange crystals were
1
are analogous to oxidase and monooxygenase activities leading
to TPQ biogenesis in nickel-substituted apo-CAO, although the
reaction mechanism remains uncertain. Pro-oxidant chemistry
through formation of reduced oxygen species catalyzed by
interaction of nickel with biological phenols and thiols would
2
2
obtained by slow diffusion of n-hexane at −30 °C. Yield: 186 mg (0.35
mmol, 91%). Anal. Calcd (found) for C H BN NiO, 1b: C, 60.83
27
39
6
1
(60.79); H, 7.37 (7.36); N, 15.76 (15.50). H NMR (CD Cl , 293 K;
2
2
δ, ppm): 76.9 (3H, 4-pz); 30.3 (2H, meta); 26.5 (2H, br, −CHMe
);
2
48
also be of interest with respect to the toxicology of the metal.
3.9 (12H, br, −CHMe
2
); −0.8 (9H, 3-Me); −6.6 (9H, 5-Me); −19.0
1
(
7
1H, B-H); −28.0 (1H, para). H NMR (C D CD , 293 K; δ, ppm):
Recently, a biomimetic Cu(II) complex was reported to react
6
5
3
5.5 (3H, 4-pz); 31.5 (2H, meta); 28.0 (2H, br, −CHMe ); 4.0 (12H,
with O and transform a phenol ligand substituent into a TPQ
2
2
6
4
br, −CHMe ); −1.3 (9H, 3-Me); −6.5 (9H, 5-Me); −19.2 (1H, B-H);
−
(CH
(1000); 514 (330); 810 (80); 940 (120). IR (KBr, cm ): 2523, ν(B−
H).
2
analogue. Attention is called to analogous intramolecular
hydroxylations of ligand substituents upon addition of H O to
29.5 (1H, para). μ = 3.09 μ (CDCl , 293 K). UV−vis−NIR
eff
B
3
2
2
−1 −1
4
0−42
2
Cl
2
; λmax, nm; ε, M cm ): 326 (sh, 700); 421 (sh, 770); 449
Ni(II) complexes
and oxidative transformations of
substituted phenols by a discrete (nacnac)NiO complex.
−1
22
2
2
.3. Synthesis of Organic Standards. 3,5,3′,5′-Tetraisopropyl-
,4′-diphenoquinone and 2,6-diisopropyl-1,4-benzoquinone were
prepared by oxidation of 2,6-diisopropylphenol according to literature
procedures; reduced dihydroquinones were then obtained by
sequential addition of dilute hydrochloric acid and zinc dust to
solutions of quinone in aqueous methanol, followed by filtration and
removal of solvent under vacuum. Spectroscopic data are summarized
below and shown in Figures S9−S22 of the Supporting Information;
2.0. EXPERIMENTAL SECTION
4
All materials were obtained from commercial vendors and used as
received, except for drying of solvents by routine techniques. Syntheses
6
8
were carried out under prepurified argon, either in a glovebox
R,Me
(
(
MBraun Unilab) or using Schlenk techniques. Tp NiCl complexes
R,Me
R = Me, Ph) were prepared from anhydrous NiCl and TlTp
in
2
MeOH/CH Cl as previously described (Caution: Thallium salts are
2
2
7
258
dx.doi.org/10.1021/ic300551z | Inorg. Chem. 2012, 51, 7257−7270

Inorganic Chemistry
Article
Table 1. Summary of X-ray Crystallography
TpPh,MeNi−O-2,6- Pr C H
i
TpMe,MeNi−O-2,6- Pr C H
i
[(TpPh,Me*)Ni]2
(2)
[TpMe,Me]Ni (μ-CO )
[(TpPh,Me)2Ni]
2
6
3
2
6
3
2
3
compound
(1a)
(1b)
(3)
(4)
empirical formula
formula weight
temp (K)
crystal system
space group
a (Å)
C H BN NiO
C H BN NiO
C H B N Ni O
C H B N Ni O
C H B N Ni
4
2
45
6
27 39
6
60 54
2
12
2
2
31 44
2
12
2
3
60 56
2
12
719.36
123(2)
triclinic
533.16
1114.19
771.82
1025.50
123(2)
173(2)
173(2)
173(2)
monoclinic
tetragonal
monoclinic
P21
monoclinic
C2/c
P1
̅
P2 /n
1
P4 /n
2
10.7698(9)
11.111(1)
12.251(2)
84.405(1)
78.857(1)
65.717(1)
1845.8(3)
2
8.803(1)
29.373(3)
10.630(1)
90
19.682(2)
19.682(2)
13.541(1)
90
8.053(5)
30.94(2)
8.073(5)
90
18.0500(9)
13.7439(7)
21.985(2)
90
b (Å)
c (Å)
α (deg)
β (deg)
91.832(1)
90
90
113.820(6)
90
111.505(1)
90
γ (deg)
V (Å3)
90
2747.2(5)
4
5245.5(8)
4
1840(2)
2
5074.3
4
Z
3
density (calc, g/cm )
1.294
1.289
1.411
0.776
1.393
1.342
absorption coefficient
0.568
0.737
1.073
0.437
−
1
(
mm )
crystal color, morphology
crystal size (mm)
green, block
orange, irregular
green, needle
green, block
green, block
1.13 × 1.07 × 0.44
28593
0.22 × 0.12 × 0.10
22004
0.25 × 0.12 × 0.05
30891
0.45 × 0.12 × 0.10
60638
0.45 × 0.12 × 0.10
14999
reflections collected
independent reflections (Rint)
observed reflections
data/restraints/parameters
GoF
8349 (0.0310)
6959
6280 (0.0751)
4388
5365 (0.0500)
4145
6484 (0.0528)
5206
5818 (0.0221)
5209
8349/0/467
1.047
6280/0/339
1.037
5365/0/355
1.020
6484/1/464
1.061
5818/0/346
1.032
R1, wR2 [I > 2σ(I)]
R1, wR2 (all data)
0.0364, 0.0806
0.0471, 0.0859
0.596, −0.353
0.0476, 0.1056
0.0803, 0.1190
0.661, −0.503
0.0389/0.0983
0.0559/0.1088
0.436, −0.260
0.0525, 0.1114
0.0718, 0.1194
0.579, −0.627
0.0308, 0.0776
0.0356, 0.0805
0.268, −0.325
3
difference peak, hole (e/Å )
plots use a labeling scheme A−F, including a naphthalene standard
A) and others as indicated.
.3.1. 2,6-Diisopropylphenol (B). H NMR (300 MHz, CDCl , 293
coproducts remaining in the reddish-brown mother liquor were
identified and quantified by independent synthesis and GC-MS (vide
infra). The spectra of 2 are shown in Figures S30−S34 of the
(
1
2
3
K; δ, ppm): 7.05 (d, 7.6 Hz, 2H); 6.90 (t, 7.6 Hz, 1H); 4.76 (s, 1H);
Supporting Information. Anal. Calcd (found) for C60
2·2H
(CDCl
(1H); −3.6 (6H); −10.2 (1H); −25.5 (1H); −30.4 (1H); plus
unassigned peaks. μeff = 2.42 μ (CDCl
CH Cl ; λmax, nm; ε, M cm ): 268 (23000); 306 (sh, 8700); 317
H B N12Ni O ,
58 2 2 4
1
3
.15 (septet, 6.8 Hz, 2H); 1.27 (d, 6.8 Hz, 12H). UV−vis−NIR
2
O: C, 62.65 (63.05); H, 5.08 (4.91); N, 14.61 (14.62). H NMR
, 293 K; δ, ppm): 76.0 (1H); 64.6 (2H); 46.4 (1H); 30.2
−1
−1
(
CH Cl ; λ , nm; ε, M cm ): 273 (1600).
3
2
2
max
1
2
.3.2. 2,6-Diisopropyl-1,4-benzoquinone (C). H NMR (300 MHz,
CDCl , 293 K; δ, ppm): 6.45 (s, 2H); 3.05 (septet, 6.9 Hz, 2H); 1.11
B
3
, 293 K). UV−vis−NIR
3
−1
−1
−1
−1
(
(
(
3
d, 6.9 Hz, 12H). UV−vis−NIR (CH Cl ; λmax, nm; ε, M cm ):
2
2
2
2
−1
9000); 420 (3100); 598 (60); 755 (30); 935 (30). IR (KBr, cm ):
25 (320); 414 (40).
.3.3. 2,6-Diisopropyl-1,4-benzodihydroquinone (D). 1H NMR
2547, ν(B−H).
2
i
2
.5. Aerobic Decomposition of TpPh,MeNi−O-2,6- Pr C H
(
300 MHz, CDCl , 293 K; δ, ppm): 6.52 (s, 2H), 4.60 (s, 1H); 4.43 (s,
2
6
3
3
(1a): Identification and Quantification of Organic Coproducts.
1
H); 3.11 (septet, 6.8 Hz, 2H); 1.22 (d, 6.8 Hz, 12H). UV−vis−NIR
Ph,Me
−1 −1
The mother liquor from the sample of decomposed Tp
Ni−O-
2,6- Pr C H (1a) (vide supra) was evaporated to give a yellow-orange
2 6 3
(
CH Cl ; λ , nm; ε, M cm ): 289 (8400).
2 2 max
i
1
2
.3.4. 3,5,3′,5′-Tetraisopropyl-4,4′-diphenoquinone (E). H NMR
amorphous residue. Naphthalene was added as an internal standard,
and the solids were redissolved in dichloromethane (4.0 mL) and
filtered through a plug of silica gel. The solution was injected into a
Shimadzu GCMS-QP2010S instrument at 270 °C, with the column
initially at 50 °C, and then the temperature was ramped upward at 10
(
300 MHz, CDCl , 293 K; δ, ppm): 7.64 (s, 4H); 3.22 (septet, 6.9 Hz,
3
−1
4H); 1.20 (d, 6.9 Hz, 24H). UV−vis−NIR (CH Cl ; λmax, nm; ε, M
2
2
−1
cm ): 426 (71000).
.3.5. 3,5,3′,5′-Tetraisopropyl-4,4′-diphenodihydroquinone (F).
H NMR (300 MHz, CDCl , 293 K; δ, ppm): 7.17 (s, 4H); 4.74 (s,
2
1
3
°
C/min for 5 min, 15 °C/min for 10 min, and 20 °C/min for 3 min
2
H); 3.19 (septet, 6.9 Hz, 4H); 1.31 (d, 6.9 Hz, 24H). UV−vis−NIR
and then held at 310 °C (Figures S24 and S25 of the Supporting
Information). Peak assignments (except U2) were confirmed by
comparison to the authentic standards described above (Figures S9−
−1 −1
(
CH Cl ; λ , nm; ε, M cm ): 265 (17000).
2 2 max
Ph,Me
i
2
.4. Aerobic Decomposition of Tp
Ni−O-2,6- Pr C H
2
6
3
Ph,Me
(1a): Isolation of [(Tp
*)Ni] (2). A sample of 1a (96 mg, 0.13
1
2
S22). Qualitatively similar results were observed by H NMR
mmol) was dissolved in CH Cl (18 mL) and sealed in a vial with a
2
2
spectroscopy of CD Cl extracts (Figures S26−S29).
2
2
rubber septum under argon. The solution was purged for
Me,Me
i
2
.6. Aerobic Decomposition of Tp
Ni−O-2,6- Pr C H
2
6
3
approximately 5 s with a stream of O gas flowing from a needle
Me,Me
2
(1b): Isolation of [Tp
Ni] (μ-CO ) (3). A sample of 1b (42
2 3
after passage through a column of CaH , anhydrous CaSO , and
2
4
mg, 0.08 mmol) was dissolved in CH Cl (10 mL) and reacted with
2 2
MgSO to ensure removal of trace H O. The punctured septum was
4
2
O as described for 1a above. The solution changed color from orange
2
then covered with silicone grease to exclude atmospheric moisture.
The green color of the solution slowly deepened on standing
overnight. After 24 h, n-hexane was introduced by vapor diffusion.
Green crystals of product complex 2, identified as di[hydro{bis(3-
phenyl-5-methylpyrazol-1-yl)(3-ortho-phenolato-5-methylpyrazol-1-yl)
borato}nickel(II)], formed after standing for 3 days, and these were
recovered by filtration (yield: 40 mg, 0.04 mmol, 53%); the organic
to dark green within 3 h (Figure S35). Green crystals of
Me,Me
[Tp
Ni] (μ-CO ) (3) were grown by diffusion of n-hexane
2 3
(yield: 16 mg, 0.02 mmol, 51%). Spectra of 3 are shown in Figures
S45−S48. Anal. Calcd (found) for C H B N Ni O , 3: C, 48.24
3
1
44
2
12
2
3
1
(47.88); H, 5.75 (5.75); N, 21.78 (21.52). H NMR (CD Cl , 293 K;
2
2
δ, ppm): 43.4 (6H, 4-pz); 1.7 (18H, 3-Me); −4.7 (2H, B-H); −5.5
(18H, 5-Me). μ = 2.22 μ (CDCl , 293 K). UV−vis−NIR (CH Cl ;
eff
B
3
2
2
7
259
dx.doi.org/10.1021/ic300551z | Inorg. Chem. 2012, 51, 7257−7270

Inorganic Chemistry
Article
−
1
−1
λmax, nm; ε, M cm ): 412 (400); 657 (80); 839 (50). IR (KBr,
−1
cm ): 2504, ν(B−H); 1577, ν(CO ).
3
Ph,Me
i
2
.7. Hydrolytic Decomposition of Tp
Ni−O-2,6- Pr C H
2
6
3
Ph,Me
(
1a): Isolation of [(Tp
) Ni] (4). A sample of 1a (31 mg, 0.04
2
mmol) was dissolved under argon in dichloromethane (8 mL).
Degassed H O (30 μL, 1.7 mmol) was injected, and the dark green
2
solution turned to a pale blue-green color within 15 min. The solvent
was removed under vacuum, the residue was extracted with
dichloromethane, and the extracts were filtered. Pale blue-green
crystals of 4 were obtained by diffusion of n-hexane. Yield: 24 mg (0.02
mmol, 100%). Spectra of 4 are shown in Figures S57−S60. Anal. Calcd
(
1
found) for C H B N Ni, 4: C, 70.27 (70.15); H, 5.50 (5.65); N,
60 56 2 12
1
6.39 (16.35). H NMR (CDCl , 293 K; δ, ppm): 55.1 (3H, 4-pz); 7.3
3
(
3
6H, 3-Ph); 5.9 (9H, 3-Ph); −1.7 (18H, 5-Me); −8.0 (1H, B-H). μeff
=
Ph,Me
−
1
Figure 1. Thermal ellipsoid plots (50% probability) of Tp
Ni−O-
.16 μ (CDCl , 293 K). UV−vis−NIR (CH Cl ; λ , nm; ε, M
B
3
2
2
max
i
Me,Me
i
−1
2,6- Pr C H (1a, left) and Tp
Ni−O-2,6- Pr C H (1b, right).
cm ): 370 (sh, 15); 416 (10, sh); 603 (7.4); 762 (2.9); 992 (4.5). IR
(
2 6 3
2
6
3
−1
Hydrogen atoms are omitted for clarity. Relevant bond lengths (Å)
and angles (deg) for 1a: Ni1−N2, 2.046(2); Ni1−N4, 2.017(1); Ni1−
N6, 2.025(1); Ni1−O1, 1.821(1); N2−Ni1−N4, 91.52(6); N2−Ni1−
N6, 95.53(6); N4−Ni1−N6, 88.63(6); N2−Ni1−O1, 121.91(5);
N4−Ni1−O1, 130.14(6); N6−Ni1−O1, 119.72(5); Ni1−O1−C31,
KBr, cm ): 2549, ν(B−H).
2.8. DFT Calculations. A simplified TpNiOPh model was derived
from our previous TpNiSPh models by replacement of sulfur with
oxygen, adjustment of the resulting Ni−O and O−C bond lengths
ipso
59
to experimental values, and geometry optimization. Hypothetical O2
adducts were modeled by manipulating the N−Ni−OAr and Ni−O−
Ar angles within the mirror plane and inserting the oxygen atoms into
the resulting gap. All geometry optimizations were restrained to Cs
point symmetry. Spin-unrestricted calculations were performed using
the Amsterdam Density Functional software package (version 2008.01,
1
2
9
47.8(1). For 1b: Ni1−N1, 2.005(2); Ni1−N3, 1.982(2); Ni1−N5,
.014(2); Ni1−O1, 1.841(2); N1−Ni1−N3, 91.71(9); N1−Ni1−N5,
1.37(9); N3−Ni1−N5, 90.09(9); N1−Ni1−O1, 118.99(9); N3−
Ni1−O1, 124.41(9); N5−Ni1−O1, 129.92(9); Ni1−O1−C16,
1
38.9(2).
6
9,70
Scientific Computing and Modelling NV),
Wilk−Nusair LDA functional, the Becke−Perdew GGA correc-
tion,
library, with frozen atomic cores and default convergence criteria. A
solvation model and relativistic correction were not applied.
using the Vosko−
71
3.0. RESULTS
7
2,73
and the Slater-type TZP orbital basis set available in the ADF
3
g
.1. General Remarks. Photochemical generation of singlet
(1
Δ ) dioxygen in the presence of 2,6-diisopropylphenol results
in a one-electron redox reaction through net hydrogen atom
2.9. X-ray Crystallography. Suitable crystals were placed onto the
6
8
abstraction to yield superoxide and phenoxyl radicals. The
tips of 0.1 mm diameter glass capillaries and mounted on a Bruker
74
latter decomposes by competitive dimerization and O coupling
APEX-II CCD diffractometer. The data collection was carried out
using Mo Kα radiation (graphite monochromator) with a frame time
of 20−60 s and a detector distance of 6.0 cm. The data were corrected
for absorption and decay (SADABS). Final cell constants were
calculated from strong reflections from the actual data collection after
2
to yield a mixture of 3,5,3′,5′-tetraisopropyl-4,4′-diphenoqui-
none and 2,6-diisopropyl-1,4-benzoquinone, respectively rep-
resenting 2- and 4-electron oxidations per phenol precursor
(Scheme 2). We hypothesized that substitution of the phenolic
75
76
R,Me
+
integration (SAINT).
proton with paramagnetic [Tp Ni(II)] would provide the
3
−
Structures were solved by direct methods and difference Fourier
means to overcome the spin barrier of ground-state ( Σ ) O
g
2
18
techniques using SHELXS-97 (Sheldrick, 1990) and SHELXL-97
(
and obtain analogous thermal reactivity. Reduced O species
2
77
Sheldrick, 1997). Space groups were determined on the basis of
would then be captured for aromatic oxidation akin to the CAO
active site, although the phenolate ortho sites are blocked by
substituents. We accordingly prepared and characterized
pseudotetrahedral Tp Ni−OAr complexes (R = Ph, 1a;
Me, 1b), which are thermally stable under inert atmosphere but
systematic absences and intensity statistics. All non-hydrogen atoms
were refined with anisotropic displacement parameters. Hydrogen
atoms were placed in ideal positions and refined as riding atoms with
relative isotropic displacement parameters. Crystal and refinement
information is summarized in Table 1 and additional comments
specific to individual structure solutions are summarized below.
R,Me
decompose when exposed to O . Products consistent with
2
7
8
phenoxyl radical formation were observed, as well as aromatic
oxidation reactions, either on the supporting ligand of 1a or on
the phenolate ring of 1b. However, 2,6-disubstituted phenolates
will undergo free radical autoxidation to diphenoquinones
Thermal ellipsoid plots are shown in Figures 1, 3, 4, and 7; relevant
bond lengths and angles are given in the captions.
For 1b, the hydrogen atom on boron was located in a similar
manner to the non-hydrogen atoms and refined isotropically. The
dimeric structure of 2 sits on an inversion center, and only half of the
molecule is unique. The diffraction data of 3 were initially indexed to a
C-centered orthorhombic unit cell with twice the volume of the final
80
under alkaline conditions, and therefore hydrolyses of 1a,b
and subsequent oxidation reactions with O and H O were
2
2
2
also examined.
R,Me
monoclinic unit cell. XPREP suggested space group C222 , which is
1
3.2. Synthesis and Characterization of Tp Ni−O-
rare, and no solution could be found in it. The reassigned monoclinic
unit cell has nearly equivalent a- and c-axes, yet CHECKCIF and other
routines found no additional symmetry. The crystal was twinned, and
the twin element is a 180° rotation about [101]. A test was performed
to determine the twin law. There are two twin components related by
i
2
,6- Pr C H (1a,b). The phenolate complexes (R = Ph, 1a;
2
6
3
Me, 1b) were obtained as crystalline solids following
deprotonation of the phenol with sodium hydride and
metatheses with known Tp Ni−Cl precursor complexes.
Structures of 1a,b were determined by X-ray diffraction (Figure
). As expected, the N O ligand fields adopt pseudotetrahedral
geometries that are nearly identical for both complexes.
R,Me
65,66
[
0 0 1/0 1 0/1 0 0] in a ratio of 0.455:0.545. This implies the twins are
1
3
enantiomorphs. The refinement yielded a Flack parameter of 0.00(2).
The nickel atom of 4 is located on a special position (2-fold rotation
axis), so that half of the molecule is the asymmetric unit. The
hydrogen atom bonded to boron was located and refined with
isotropic displacement parameters. The structure of 4 is isomorphous
R,Me
Constrained Tp
ligand chelation yields average N−Ni−N
angles of 92(3)° in 1a and 91(1)° in 1b, resulting in umbrella
distortions with Ni−N bond vectors at respective angles of
123.9(3)° and 124.5(3)° from an ideal 3-fold axis. The
79
with the cobalt analogue.
7
260
dx.doi.org/10.1021/ic300551z | Inorg. Chem. 2012, 51, 7257−7270

Inorganic Chemistry
Article
Scheme 2
Figure 2. UV−vis−NIR spectra for decomposition of 1a (0.61 mM) in CH Cl under O (306 K). The inset shows traces at 476 nm (green trace,
2
2
2
left axis) and 426 nm (red trace, right axis), corresponding to the absorption maxima of 1a and 3,5,3′,5′-tetraisopropyl-4,4′-diphenoquinone,
respectively.
phenolate oxygen is then slightly displaced off this trigonal axis,
with N−Ni−O angles ranging from 119.72(5)° to 130.14(6)°
to the cases of both the chloride complex precursors and several
58,59
1
arylthiolate analogues.
The H NMR resonances of 1a,b
for 1a and from 118.99(9)° to 129.92(9)° for 1b, giving τ
accordingly exhibit significant contact shifts, but the spectra
4
8
1
values of 0.77 for 1a and 0.75 for 1b. The Ni−OAr bond
were nonetheless consistent with the assigned formulations
5
8,59
lengths are 1.821(1) and 1.841(2) Å, and the Ni−O−C
(Figures S7, S8). As with the arylthiolate analogues,
spin
ipso
bond angles are 147.8(1)° and 138.9(2)° in 1a,b, respectively.
The average Ni−N bond lengths of 2.03(2) and 2.00(2) Å in
a,b compare to values of 1.98(2) and 1.99(2) Å in the
delocalization by π-polarization leads to pronounced upfield
shifting of the phenolate para protons, with the meta and
isopropyl methine protons shifted downfield, and the isopropyl
methyl resonance nearer to the diamagnetic limit. The pyrazole
resonances of the supporting Tp ligands are unremarkable.
However, the borohydride resonances of 1a,b both exhibit
pronounced upfield shifts (−20 and −19 ppm at 293 K,
respectively) compared to those of the chloride complex
precursors (−14 and −13 ppm, respectively) and several
1
R,Me
66,82
respective Tp NiCl precursors.
The spectroscopic data of 1a,b are consistent with the solid-
state structures. The IR spectra of 1a,b exhibit ν(B−H) modes
−1
at 2543 and 2523 cm , respectively (Figures S1−S4 of the
3
R,Me
83
Supporting Information), indicative of κ -Tp
chelation.
The complexes exhibit intense green (1a) and orange (1b)
colors, reflecting visible phenolate−Ni(II) LMCT bands, as
well as weaker ligand field bands in the near-IR, with
extinctions consistent with the noncentrosymmetric geometries
5
8,59
arylthiolate analogues (typically −10 to −11 ppm),
reflecting the trend in the ligand field bands. All three
pyrazoles, as well as both sides of the phenolate rings, are
1
(Figures S5 and S6). The LMCT bands are similar to those of
spectroscopically equivalent in the H NMR spectra of 1a,b,
the arylthiolate analogues, with somewhat attenuated ex-
indicative of fluxionality in solution.
5
8,59
tinctions.
The ligand field bands are similar to those of
3.3. Reactivity: Oxygenation of 1a,b. The phenolate
complex 1a was indefinitely stable in a dilute solution of dry,
anaerobic CH Cl (Figure S23). However, rapid saturation of
both the arylthiolate analogues and the chloride complex
precursors, although small blue shifts in energies are evident in
2
2
Me,Me
the following order: −SAr > −Cl > −OAr; Tp
≥
the solution of 1a with O resulted in monotonic bleaching of
2
TpPh,Me.
58,59
In view of the low symmetry and lack of near-IR
the absorption bands (t1/2 ≈ 2 h at 306 K; Figure 2), coincident
with growth of an intense band at 426 nm, assigned to 3,5,3′,5′-
tetraisopropyl-4,4′-diphenoquinone (0.05 equiv, 10 mol %).
The induction period apparent in growth of this band reflects
bleaching of coincident CT absorption of 1a (Figure 2 inset),
and presumably the initial accumulation of the dihydroquinone
intermediate (Scheme 2), which disrupts the isosbestic points.
GC-MS analysis of a decomposed aerobic solution with added
data, rigorous assignments of the ligand field bands of 1a,b are
68
not offered, although such analyses were reported previously for
iPr,iPr
Me,Me
57,84
Tp
Ni−SC F and Tp
Ni−Cl.
Magnetic suscepti-
6
5
bilities in CDCl solutions, determined by the Evans NMR
3
6
7
method, gave values of 2.72 and 3.09 μ at 293 K for 1a,b,
B
respectively, indicative of an orbitally nondegenerate para-
magnetic ground state (i.e., μ = 2.83 μ for S = 1) and similar
S
B
7
261
dx.doi.org/10.1021/ic300551z | Inorg. Chem. 2012, 51, 7257−7270

Inorganic Chemistry
Article
naphthalene as an internal standard (A; t = 8.0 m, m/z = 128;
Figures S24 and S25) revealed a small fraction of unmodified
tate ligand binds Ni(II) in a distorted trigonal pyramidal
81
geometry (τ = 0.65) with an axial N2 donor atom on the
4
2
,6-diisopropylphenol (B; 9.6 m, m/z = 178; 14 mol %), as well
same tripodal arm as the equatorial phenolate (i.e., O1). The
Ni1−N2 bond is shorter than the Ni1−N4 and Ni1−N6 bonds,
2.000(2) vs 2.054(2) and 2.134(2) Å, respectively. The open
axial site is filled by the opposing phenolate oxygen (i.e., O1′)
in the assembled dimer, giving an N2−Ni1−O1′ angle of
164.81(8)°. However, the overall N O ligand field is best
as 2,6-diisopropyl-1,4-benzoquinone (C; 10.1 m, m/z = 192; 15
mol %) and overlapping 3,5,3′,5′-tetraisopropyl-4,4′-dipheno-
quinone (E; 17.80 m, m/z = 354!) and 3,5,3′,5′-tetraisopropyl-
4
,4′-diphenodihydroquinone (F; 17.83 m, m/z = 354; 46 mol %
combined with E); the combined yield of phenol equivalents
was 76 mol %. A peak corresponding to 2,6-diisopropyl-1,4-
benzodihydroquinone (D; 12.3 m, m/z = 194) was not
observed. Two additional unknown peaks were assigned as 3-
phenyl-5-methylpyrazole (U1; 12.4 m, m/z = 158, 0.20 equiv)
and 4-hydroxy-3,5-diisopropylphenyl-2,6-diisopropyl ether
3
2
85
described as a distorted square pyramid (τ = 0.32) with an
axial N4 donor. The central Ni O core is canted 12.6° relative
2
2
to the axial B−N1−N2−Ni1−O1′ least-squares planes, with the
two tripodal ligands disposed over opposite faces, which results
in diverging N4−Ni1−O1 and N6−Ni1−O1 bond angles of
122.82(8)° and 145.73(8)°.
(U2; 15.6 m, m/z = 354), the product of head-to-tail phenoxy
Ph,Me
Ph,Ph
radical coupling. The assignments of U1 (data not shown) and
B−F (Figures S18−S22) were confirmed by comparison to
authentic standards. Qualitatively similar results were observed
by H NMR spectroscopy of CD Cl extracts (Figures S26−
S29), except the diphenodihydroquinone was oxidized. Thus,
the organic product mixture is analogous to that previously
reported for oxidation of phenol by singlet oxygen, consistent
with formation of phenoxyl radical and obligating concomitant
generation of reduced oxygen species (Scheme 2). Further
evidence for a reactive oxygen intermediate was obtained from
isolation of product complex 2, with an oxidized ligand.
Complex 2, obtained as pale green crystals in 53% yield, was
characterized by X-ray crystallography (Figure 3). A dimeric
Six-coordinate Fe(III) complexes of Tp
* and Tp
*
8
6,87
ligands were obtained previously, by oxidation with O or
H O , and resonance Raman spectroscopy of the latter
2
2
2
1
2
2
derivative revealed vibrational modes of the ortho-phenolate
87
ring. The solid-state IR spectrum of 2 contains several
analogous bands, as well as a single ν(B−H) mode at 2547
6
8
−1
3
cm , consistent with κ -pyrazole ligation and ideal C point
2h
symmetry (Figures S30 and 31). The UV−vis−NIR spectrum
of 2 in CH Cl contains intense bands at 306, 317, and 420 nm,
2
2
which can be assigned as LMCT transitions arising from the
4
1,42,88
installed phenolate moieties,
as well as weaker ligand
field bands at longer wavelengths (Figure S32). The band at
420 nm exhibits a reversible, temperature-dependent red-shift
(
Figure S33), suggesting a solution-phase equilibrium.
Notwithstanding antiferromagnetic coupling of Ni(II) ions in
1
the dimeric complex (μeff = 2.42 μ in CDCl ), the H NMR
B
3
spectrum in CDCl exhibits signals over a significant range of
3
chemical shifts comparable to those for 1a,b (ca. −30 to 80
ppm, Figure S34). Thus, dissociation of 2 to monomeric
Ph,Me
(
Tp
*)Ni in solution is suggested. However, further analysis
was complicated by the low solubility of the isolated crystalline
solids, and the NMR spectrum was not fully assigned.
Aerobic decomposition of 1b appeared similar to that of 1a
by UV−vis−NIR spectroscopy (Figure S35), albeit at a ca. 10-
fold faster rate (at 293 K vs 303 K) and with a diminished yield
of diphenoquinone (ca. 6 mol %). A slight increase in optical
density broadly centered near 642 nm likely reflects
Figure 3. Thermal ellipsoid plot (30% probability) for dimeric
Ph,Me
[
(Tp
*)Ni] (2, left) and the monomeric asymmetric unit (right).
2
Hydrogen atoms are omitted for clarity. Relevant bond lengths (Å)
and angles (deg): Ni1−N2, 2.000(2); Ni1−N4, 2.054(2); Ni1−N6,
Me,Me
34
accumulation of [Tp
Ni(μ-OH)]2. As for decomposition
2
3
.134(2); Ni1−O1, 2.043(2); Ni1−O1′, 2.010(2); Ni1···Ni1′,
.1207(5); O1···O1′, 2.587(3); N2−Ni1−N4, 89.78(8); N2−Ni1−
of 1a, GC-MS analysis of the organic products revealed
formation of diphenoquinone with traces of benzoquinone and
unmodified phenol (Figures S36 and S37), and this was
N6, 89.40(8); N4−Ni1−N6, 90.97(8); N2−Ni1−O1, 85.55(8); N4−
Ni1−O1, 122.82(8); N6−Ni1−O1, 145.73(8); N2−Ni1−O1′,
1
confirmed by H NMR spectroscopy (Figures S38 and S39).
1
64.81(8); N4−Ni1−O1′, 99.54(8); N6−Ni1−O1′, 102.32(7); O1−
However, two significant unknown species U3 and U4 were
Ni1−O1′, 79.32(8); Ni1−O1−Ni1′, 100.68(8); Ni1−O1−C10,
30.0(2); Ni1−O1′−C10′, 126.8(2).
1
also observed. The unknown components were isolated by
1
chromatography on silica and characterized by H NMR
structure occupies a crystallographic inversion center, so only
half of the molecule is unique. Both nickel atoms are ligated
with a modified TpPh,Me ligand resulting from C−H bond
spectroscopy in CD Cl solution (Figures S40−S44). The data
2 2
for the major product U4 unambiguously indicate loss of ring
aromaticity and symmetry, as the three ring protons shifted to
6.77 (1H, d, J = 4.3 Hz), 3.72 (1H, d, J = 4.0 Hz), and 3.59
ppm (1H, dd). A fourth proton resonance, presumably arising
from phenol tautomerization, appeared at 3.67 ppm (1H, s).
Moreover, the isopropyl groups also lost equivalence, exhibiting
distinct methine septets at 2.80 and 1.88 ppm, each coupled to
pairs of diastereotopic methyl signals. This spectrum appears to
be consistent with assignment of U4 as 3,4-dihydro-3,4-
dihydroxy-2,6-diisopropylcyclohex-5-enone, which apparently
has not been reported previously. This product is consistent
with aromatic dihydroxylation of phenol by addition of H O or
hydroxylation at the ortho position of one 3-phenyl substituent
Ph,Me
(
(
i.e., Tp
*), a two-electron oxidation. The planar Ni-
II) (OR) core adopts a Ni···Ni separation of 3.1207(5) Å and
2
2
an O···O separation of 2.587(3) Å, with internal O−Ni−O and
Ni−O−Ni angles of 79.32(8)° and 100.68(8)°, respectively.
4
The Ni1−O1 bond length within the κ -chelate is longer than
the bridging Ni1−O1′ bond length, 2.043(2) vs 2.010(2) Å.
These core metrical parameters are similar to those of
32−34
previously characterized [TpNi(μ-OH)] complexes.
The
2
Ni1−O1−C10 and Ni1−O1′−C10′ angles are 130.0(2)°
ipso
ipso
2
2
and 126.8(2)°, respectively. The modified dianionic tetraden-
its equivalent across meta and para ring positions, a two-
7
262
dx.doi.org/10.1021/ic300551z | Inorg. Chem. 2012, 51, 7257−7270

Inorganic Chemistry
Article
electron oxidation distinct from the aromatic C−H bond
hydroxylation that yields 2. The spectrum of the minor product
U3 was not fully resolved, and a structure was not assigned.
Also isolated from a solution of aerobically decomposed 1b
decompositions already described are rate-limiting hydrolyses
due to introduction of H O. Rapid autoxidation of free phenol
2
might then yield H O , and precedent exists for peroxide
2
2
30−34
activation and oxidative chemistry at TpNi(II) centers.
was a dimeric carbonato-bridged complex with unmodified
Therefore, hydrolysis reactions were carried out in which
Me,Me
scorpionate ligands, [Tp
Ni] (μ-CO ) (3), as pale green
excess H O was deliberately added to solutions of 1a and 1b in
2
3
2
crystals in 51% yield. This complex presumably arises from the
CH Cl . This resulted in simple bleaching of the LMCT bands
in both cases (Figures 5 and S49). A UV−vis−NIR spectrum of
2
2
Me,Me
previously reported [Tp 3 Ni(μ-OH)]2 by CO2 capture
4,89
during workup under air.
Complex 3 was characterized
by X-ray diffraction (Figure 4). The nickel atoms of 3 lie on the
Figure 4. Thermal ellipsoid plot (30% probability) of [TpMe,MeNi] (μ-
2
Figure 5. UV−vis spectra demonstrating reaction of 1a (0.30 mM,
CO ) (3). Hydrogen atoms are omitted for clarity. Relevant bond
3
green) with added H O (270 mM) to form a hydrolyzed intermediate
2
lengths (Å) and angles (deg): Ni1−N2, 2.016(8); Ni1−N4, 2.009(7);
Ni1−N6, 2.021(6); Ni1−O1, 2.104(7); Ni1−O2, 2.021(7); Ni2−N8,
(
blue) that reacts rapidly with H O (67 mM) to form 2 (red).
2 2
2
2
.054(6); Ni2−N10, 1.955(8); Ni2−N12, 2.041(8); Ni2−O1,
.144(7); Ni2−O3, 1.963(6); C31−O1, 1.37(1); C31−O2, 1.27(1);
the hydrolyzed solution of 1b at equilibrium was consistent
C31−O3, 1.22(1); N2−Ni1−N4, 90.6(3); N2−Ni1−N6, 91.8(3);
Ni4−Ni1−N6, 91.6(3); N2−Ni1−O1, 146.9(3); N2−Ni1−O2,
Me,Me
with a mixture of previously reported [Tp
Ni(μ-OH)] (70
2
34
mol % based on reported extinctions) and an unknown minor
species that exhibits ligand field bands typical of a
pseudotetrahedral TpNi−X complex (Figure S49). Attempts
9
5.0(3); N4−Ni1−O1, 118.3(3); N4−Ni1−O2, 103.9(3); N6−
Ni1−O1, 102.3(2); N6−Ni1−O2, 162.9(3); O1−Ni1−O2, 64.2(2);
O1−C31−O2, 112.5(8); O1−C31−O3, 116.1(8); O2−C31−O3,
1
31.0(8); N8−Ni2−N10, 90.5(3); N8−Ni2−N12, 90.9(3); Ni10−
to isolate the immediate product complex from hydrolysis of
Ph,Me
Ni2−N12, 92.4(3); N8−Ni2−O1, 161.1(2); N8−Ni2−O3, 97.5(3);
1a, putatively the unknown monomeric analogue [Tp
Ni−
N10−Ni2−O1, 102.2(3); N10−Ni2−O3, 149.0(3); N12−Ni2−O1,
OH], instead gave the previously unreported and spectroscopi-
Ph,Me
1
02.3(3); N12−Ni1−O3, 117.2(3); O1−Ni2−O3, 64.5(3); Ni1−
cally distinct sandwich complex [(Tp
)2Ni] (4) as pale
O1−Ni2, 169.7(4).
blue-green crystals in quantitative yield (i.e., 0.5 equiv).
Subsequent addition of excess H O2 to the solution of
2
same side of the carbonato ligand plane, in contrast to the
previously reported Tp analogue, and are more trigonally
hydrolyzed 1a rapidly generated complex 2 in quantitative
iPr2
89
1
spectroscopic yield (Figure 5), as confirmed by H NMR
distorted (τ = 0.27 and 0.20 for Ni1 and Ni2, respectively, vs
(Figure S50) and FTIR (Figures S51 and S52) spectroscopy,
without accumulation of diphenoquinone. Treatment of 4 with
excess H O also resulted in conversion to 2, albeit much more
slowly (Figure S53).
Notwithstanding the above results, phenol oxidation
0
.16 and 0.14). The Ni−O bonds to the bridging oxygen are
longer than those to the unshared oxygens in 3, 2.104(7) and
2
2
Cy
2.144(7) Å vs 2.021(7) and 1.963(6) Å, similar to a Tp -
90
iPr2
supported dimer, but reversed from the Tp -supported
complex. The bridging Ni−O−Ni angle is correspondingly
decreased, from 178.0(1)° in the latter to 169.7(4)° in 3.
Within the carbonato ligand, the carbon−oxygen bond to the
bridging oxygen in 3 is longer than those to the two terminal
oxygens, 1.37(1) Å vs 1.27(1) and 1.22(1) Å. The IR spectrum
8
9
following hydrolyses of 1a,b and O addition was slow and
2
89
inefficient (Figures 6 and S54). Exposure of hydrolyzed 1a,b to
O resulted in relatively slow generation of weak absorption
2
bands at 426 nm, consistent with diminished accumulation of
diphenoquinone (≤3.0 mol %) and 2 (Figures 6 and S54,
insets). GC-MS analysis of the product mixtures after standing
24 h demonstrated that the balance of organic product was
unmodified 2,6-diisopropylphenol (Figures S55 and S56).
Autoxidation of free phenol is plausible, but accumulation of
−1
of 3 contains a band at 1577 cm assigned to the carbonato
ligand (Figures S45 and S46), compared to 1568 and 1581
−
1
iPr2
Cy
cm in the respective Tp - and Tp -supported ana-
8
9,90
1
logues.
The UV−visible−NIR (Figure S47) and H NMR
spectra of 3 (Figure S48) are also consistent with these prior
diphenoquinone, and hence reduction of O , is faster and more
2
8
9
results, with the latter exhibiting attenuated contact shifting
due to antiferromagnetic coupling through the carbonato
bridge (μeff = 2.22 μB).
extensive without added H O (Figures 2 and S35). The excess
2
H O, which is necessary to effect prompt hydrolyses of 1a,b,
2
might act as a significant inhibitor of phenol autoxidation,
R,Me
3.4. Reactivity: Hydrolyses of 1a,b and Subsequent
perhaps by compromising the efficacy of the [Tp Ni−OH]
Oxidations with O and H O . Given the basicity of the
coproduct as a base catalyst. Otherwise, it seems necessary to
2
2
2
phenolate ligands, complexes 1a,b are susceptible to hydrolysis.
We cannot exclude the possibility that the observed
ascribe the difference in O chemistry without added H O to
oxidation of intact 1a,b.
2
2
7
263
dx.doi.org/10.1021/ic300551z | Inorg. Chem. 2012, 51, 7257−7270

Inorganic Chemistry
Article
Figure 6. UV−vis−NIR absorption vs time (Δt = 1000 s) following addition of O to a hydrolyzed solution of 1a (0.44 mM in CH Cl , 306 K). The
2
2
2
inset shows a trace at 426 nm (red), corresponding to accumulation of 3,5,3′,5′-tetraisopropyl-4,4′-diphenoquinone, and compared to the data shown
in Figure 2 for O addition to intact 1a (green) and corrected for absorption of 1a (black).
2
A crystal structure of complex 4 confirmed the sandwich
in the Ni−N and Ni···B distances can be inferred from these
data, in which increased steric bulk of 3- and 5-pyrazole
substituents favors longer Ni−N bonds within a more compact
sandwich. The structure of 4 represents a new extreme, which is
accommodated by significant B−N−N−Ni torsion angles in
the pyrazole ring chelation, 15.2(1)−19.5(1)°. The intraligand
cis-N−Ni−N bond angles remain close to right angles,
Ph,Me
structure, [(Tp
)2Ni] (Figure 7). Despite the steric bulk of
8
8
1
9.17(4)−91.48(4)°, but the interligand angles diverge,
2.71(4)−97.25(4)°, and the trans angles depart from linearity,
69.84(4)−171.31(6)°. The Ni−N5 bond lengths of 2.226(1)
Å, disposed trans along one axis, are elongated compared to the
orthogonal Ni−N1 and Ni−N3 bond lengths, 2.149(1) and
2
.157(1) Å, respectively. The crystal lattice of 4 is isomorphous
Ph,Me
with that of the (Tp
)2Co analogue, which exhibits a nearly
identical structure with an average Co−N bond length of
7
9
2
.21(5) Å and Co···B distances of 3.10(1) Å.
Spectral data of 4 are consistent with the sandwich structure.
The IR spectrum of 4 contains a single ν (B−H) mode at 2549
as
−1
3
Ph,Me
Figure 7. Thermal ellipsoid plot (30% probability) of [(TpPh,Me)2Ni]
4). For clarity, hydrogen atoms are omitted and the carbon atoms of
cm (A under D symmetry), indicative of dual κ -Tp
2u 3d
(
ligation (Figures S57 and S58). The electronic spectrum of 4 in
opposing ligands are differentially shaded. Relevant bond lengths (Å)
and angles (deg): Ni1−N1, 2.149(1); Ni1−N3, 2.157(1); Ni1−N5,
CH Cl2 solution is typical of a pseudo-octahedral Tp Ni
2
2
9
1,100
sandwich (Figure S59).
Spin-allowed ligand field tran-
2
.226(1); N1−Ni1−N3, 89.96(4); N1−Ni1−N5, 91.48(4); N3−
3
3
3
sitions were observed at 990 ( A → T ) and 603 nm ( A
2g
2g
2g
Ni1−N5, 89.17(4); N1−Ni1−N1′, 96.32(6); N1−Ni1−N3′,
3
3
1
→
T ), and a spin-forbidden band ( A → E ) was observed
1g 2g g
1
69.84(4); N1−Ni1−N5′, 82.71(4); N3−Ni1−N3′, 85.00(6); N3−
−1
−1
at 762 nm (i.e., Δ = 10,100 cm ; B = 770 cm ). The third
O
Ni1−N5′, 97.25(4); N5−Ni1−N5′, 171.31(6), where the prime
symbol denotes a nitrogen atom on the opposing ligand.
3
3
spin-allowed band ( A → T {P}) was obscured by tailing of
2g
1g
4Bo
stronger UV bands. Spectra of Tp Ni and (Tp ) Ni are blue-
shifted, ΔO = 12,000 and 12,890 cm , respectively,
2
2
−1
91,100
opposing 3-phenyl substituents, the nickel ion resides in a
9
1,92
3
consistent with their shorter Ni−N bonds (vide supra).
The
pseudo-octahedral, ideally D N ligand field afforded by κ -fac
3d
6
solution magnetic susceptibility of 4 (3.16 μ ) is consistent
ligation of the two ligands. The average Ni−N bond length of
.18(4) Å and nonbonded Ni···B distances of 3.057(2) Å can
be compared to other reported sandwich structures:
B
3
100
1
with an ideal A ground state. The H NMR spectrum of 4
2
2g
(
Figure S60) exhibited equivalence of all pyrazoles, reflecting
4
Bo
91
ideal 2- and 3-fold symmetry of the complex. Notwithstanding
the steric hindrance evident in the structure, both edges of the
3-phenyl substituents also exhibit equivalence, consistent with
rotation.
(
Tp ) Ni, 2.087(4) and 3.161(6) Å; (Tp) Ni, 2.093(7)
2 2
92
Me
and 3.165(3) Å; (Tp ) Ni, molecule 1, 2.11(2) and
.140(7) Å; (Tp ) Ni, molecule 2, 2.096(6) and 3.16(1)
)2Ni, molecule 1, 2.10(1) and 3.13(1) Å;
) Ni, molecule 2, 2.11(1) and 3.143(7) Å;
Tp ) Ni, 2.12(1) and 3.089(6) Å; (Tp
2
9
3
Me
3
2
9
3
Me,Cl,Me
94
94
Å; (Tp
Me,Cl,Me
Np
(
(
Tp
3.5. DFT Calculations. Oxidation of intact phenolate
2
95
Me,Me
)2Ni, 2.13(2)
complexes with O was suggested as a reaction mechanism
2
2
9
6
CO2Et,Me
97
and 3.055(6) Å; (Tp
) Ni, 2.14(6) and 3.105(6) Å;
leading to phenoxyl radical formation, but complex inter-
mediates were not observed during aerobic decompositions of
1a,b. Therefore, geometric and electronic models were sought
2
3py
98
4bz
(
3
Tp ) Ni, 2.15(7) and 3.078(2) Å; (Tp ) Ni, 2.18(5) and
2 2
99
.115(7) Å; (4Bo, 4,5-fused benzo ring {i.e., indazolyl vs
pyrazolyl}; Np, neopentyl, CH C(CH ) ; 3-py, meta-C H N;
for hypothetical inner-sphere O adducts. Spin-unrestricted
2
3 3
5
4
2
4
bz, para-C H −CN). An inverse and counterintuitive trend
DFT calculations were performed first on a simplified TpNi−
6
4
7
264
dx.doi.org/10.1021/ic300551z | Inorg. Chem. 2012, 51, 7257−7270

Inorganic Chemistry
Article
OPh model (Figure 8A), related to the crystal structure of 1a,b
by replacement of scorpionate and phenolate ligand sub-
2.05(1) Å, the Ni−OPh bond length was 1.84 Å, and the Ni−
O−Cipso bond angle was 149°; comparable values in 1b are
2
.00(2) Å, 1.841(2) Å, and 138.9(2)°, respectively. Frontier
orbitals of interest included the five atomic d orbitals on nickel,
as well as the in-plane and out-of-plane phenolate oxygen donor
orbitals (i.e., β-spin orbitals 21a″ and 37a′, respectively), which
52
were separated by 1.8 eV (Figure 9). The electronic structure
of TpNiOPh was similar to prior calculations on the arylthiolate
analogues, except the relatively upright conformation of the
phenolate disposes the ligand for π overlap of both donor
orbitals, in contrast to the pseudo-σ interaction of bent
5
7,59
arylthiolates.
Consistent with the imposed spin state (S =
1
), the two lowest unfilled orbitals were β-spin (41a′ and 22a″),
composed predominantly of atomic d orbitals on nickel (69%
and 62%, respectively) and exhibiting σ* interactions with the
pyrazole N2 donor atoms and π* interaction with the
phenolate. The remaining β-spin Ni d orbitals (40a′, 39a′, and
2
0a″) were stabilized by an average of 1.8 eV and were filled, as
were all five α-spin analogues. The α-spin HOMO−LUMO gap
was 3.6 eV, and the latter was comprised entirely of pyrazole π*
interactions (not shown). The α- and β-spin HOMOs,
presumably the redox-active orbitals in a one-electron
oxidation, both exhibited major contributions from the out-
of-plane phenolate π donor orbital.
After benchmarking the TpNiOAr model against the
experimental data for 1a,b, we calculated possible structures
for a hypothetical O adduct, [TpNi(OPh)(O )] (Figure 8B−
2
2
D). Ordinary unrestricted calculations were performed, without
consideration for effects of magnetic coupling between spins
Figure 8. Optimized geometries of C -symmetric computational
s
localized on nickel and the O ligand. Four conformations are
models of TpNiOPh (A), TpNiOPh(O ) (B−D), and TpNiO (E−I).
2
2
2
possible under C symmetry, with the phenolate ligand
s
stituents with hydrogen. The O adducts [TpNi(OPh)(O )]
displaced cis or trans with respect to the axial pyrazole (defined
2
2
(
Figure 8B−D) and [TpNi(O )] (Figure 8E−I) were modeled
as occupying the mirror plane) and O introduced either end-
2
2
in turn, with the latter reflecting loss of phenoxyl radical
on or side-on into the opened coordination site. Calculations
with imposed low-spin (S = 0) states failed to converge, while
subsequent to O ligation. To simplify the computations and
2
their interpretation, geometry optimizations were constrained
high-spin (S = 2) states converged by dissociation of O and
2
to C symmetry in all cases. The calculated geometric
return of the TpNiOPh fragment toward its previously
described structure. In contrast, computations with intermedi-
ate spin (S = 1) converged in three of the four cases (Figure
8B−D), failing only for a trigonal prismatic geometry with an
s
parameters are summarized and compared (Table 2), and
noteworthy features of the electronic structures are discussed
below.
Geometry optimization of TpNiOPh yielded a sawhorse
configuration comparable to the experimental structure of 1b
equatorial phenolate and side-on O . Of the three other
2
models, the unique octahedral conformation with side-on O2
and an axial phenolate (Figure 8B) was predicted to exhibit
significant elongations of the Ni−OAr and trans Ni−N bonds
along an axial vector (Table 2), suggesting Jahn−Teller
(
Table 2, Figure 8A and Figure S61), with one large axial N−
Ni−O bond angle of 134° and two smaller N−Ni−O bond
angles of 121° (τ = 0.74). The average Ni−N bond length was
4
Table 2. Calculated Bond Lengths from Computational Models
a
b
b
b
model
structure
S
O−O (Å)
Ni−O (Å)
Ni−OPh (Å)
Ni−Nax (Å)
Ni−Neq (Å)
relative energy (eV)
2
TpNiOPh + O2
TpNiOPh(O2)
TpNiOPh(O2)
TpNiOPh(O2)
A
B
C
D
E
1 ± 1
1
1
1.23
1.33
1.28
1.28
1.31
1.35
1.34
∞
1.94
1.84
2.11
1.91
1.89
∞
∞
∞
2.05
2.14
2.02
2.06
2.01
2.03
2.08
2.06
1.99
2.09
2.06
2.06
2.07
2.04
0.00
+0.05
+0.18
−0.30
+0.98
+0.69
+0.73
2.09
1
1.91
•
3
1
TpNiO + PhO
/ ± /
1.87
2
2
2
2
2
TpNiO + PhO•
F
3
/ ± /
1
2.04
2
2
TpNiO + PhO•
G
3
/ ± /
1
1.99 (ax)
2
2
2
.11 (eq)
2.10 (ax)
.01 (eq)
1.89 (ax)
.84 (eq)
TpNiO + PhO•
H
I
1
/ ± /
1
1.35
1.37
∞
∞
2.06
1.96
2.03
2.08
+0.90
+0.46
2
2
2
2
2
TpNiO + PhO•
1
/ ± /
1
2
2
1
a
b
Figure 8. Axial positions defined as pyrazole occupying the mirror plane and trans coordination site.
7
265
dx.doi.org/10.1021/ic300551z | Inorg. Chem. 2012, 51, 7257−7270

Inorganic Chemistry
Article
Figure 9. Relevant spin-unrestricted frontier molecular orbitals in the C -symmetric TpNiOPh computational model (Figure 8A), circumscribing five
s
nickel d orbitals and the in-plane and out-of-plane phenolate donor orbitals. Red and blue orbitals and labels respectively signify symmetry (a′) and
antisymmetry (a″) with respect to the mirror plane.
Figure 10. Relevant spin-unrestricted frontier molecular orbitals in a C -symmetric TpNiOPh(O ) computational model (Figure 8B), circumscribing
s
2
five nickel d orbitals, the two phenolate donor orbitals, and two O π* orbitals. Red and blue orbitals and labels respectively signify symmetry (a′)
2
and antisymmetry (a″) with respect to the mirror plane.
7
1
distortion of a low-spin Ni(III) ion (d , S = / ). A relatively
the latter hole may facilitate phenoxide/superoxo and
phenoxyl/peroxide valence isomerism. Taken as a whole, the
octahedral conformation seems poised for facile homolysis of
phenoxyl radical. The end-on conformations (Figure 8C and
D) exhibited analogous electronic structures (Figures S62 and
S63), notwithstanding their trigonal bipyramidal geometries,
but the predicted O−O and Ni−OAr bond lengths were
shorter. The end-on conformations may be disposed toward
2
long O−O bond was also predicted (Table 2), presumably
facilitated by the side-on coordination. Four unoccupied
orbitals were found at low energy (Figure 10): a pair of α-
and β-spin σ* orbitals (25a″) delocalized over the NiO core;
one β-spin orbital (45a′) consisting predominantly of the axially
oriented dσ* orbital on nickel; and a third β-spin hole (24a″)
2
delocalized over both the O and OPh ligands. The nature of
2
7
266
dx.doi.org/10.1021/ic300551z | Inorg. Chem. 2012, 51, 7257−7270

Inorganic Chemistry
Article
51
intramolecular coupling of superoxo and phenoxo moieties,
decomposition under O . However, aerobic oxidation of the
2
free phenol was observed to be slow and inefficient. Unless
phenol autoxidation is inhibited by the presence of excess H O,
2
it seems necessary to propose that a majority of the oxidation
chemistry observed in dry CH Cl results from reaction of O
2
2
2
conformations, end-on (pseudotetrahedral) and side-on
with the intact phenolate complexes. Direct O reduction by
2
(
(
8
trigonal bipyramidal or square pyramidal), in both quartet
1a,b might entail one-electron transfer, either by an outer-
3
1
•+
S = / , Figure 8E−G) and doublet states (S = / , Figure
sphere mechanism to form superoxide and [TpNi−OAr]
2
2
H−I). Assignment of oxidation states was complicated by the
radical intermediates (mechanism iii) or by formation of an
high degree of NiO covalency, but calculated O−O bond
inner-sphere adduct [TpNi(OAr)(O )] (mechanism iv). In
2
2
lengths in the high-spin structures were typically consistent
with those of superoxo ligands (Table 2). Accordingly, three
vacant β-spin orbitals were found at low energy for the high-
spin side-on models, two predominantly nickel-centered dσ*
either case, the complex intermediate would decompose by
Ni(III)−OAr bond homolysis, with subsequent phenoxyl
radical and reduced O chemistry proceeding as already
2
67
described (Scheme 2).
orbitals and one O -centered π* orbital. However, these
Although nickel complex intermediates were not observed to
accumulate during aerobic decompositions of 1a,b, a number of
possible intermediates arising along competitive oxidation and
hydrolysis pathways of the TpNiOAr complexes can be
proposed (Scheme 3). Hydrolyses of the starting complexes
2
orbitals exhibited decreased nickel d character in the end-on
conformation, suggesting a shift toward univalent nickel
(Figures S64−S66). The low-spin calculations were more
opaque. The initially end-on model converged to a trigonal
bipyramidal geometry that appears to enforce antiferromagnetic
coupling in a superoxo-Ni(II) valence state (Figure S67), while
an initially trigonal bipyramidal model converged with
significantly contracted bond lengths, plausibly interpreted as
peroxo−Ni(III) (Figure S68). The square pyramidal con-
Scheme 3
formation failed to converge. Several Ni−O adducts exhibiting
2
low-spin EPR signals were recently reported, and DFT
calculations supported similarly divergent experimental assign-
ments of ground states, depending on the supporting
1
9−21,23,25,26
ligand.
4.0. DISCUSSION
Inspired by the dioxygenase activity of nickel-substituted apo-
5
4,55
CAO,
we prepared and characterized the pseudotetrahedral
R,Me
phenolate complexes Tp Ni−OAr (R = Ph, 1a; Me, 1b; Ar =
i
2
,6- Pr C H ). These novel complexes are thermally robust in
2 6 3
anaerobic, anhydrous CH Cl but decompose upon exposure to
2
2
O or H O to yield several novel organic and complex products
2
2
(
i.e., 2−4). The organic oxidation products are consistent with
formation of phenoxyl radical and reduction of O , which
2
results in distinct two-electron aromatic oxidation reactions,
either hydroxylation of an aromatic C−H bond on a 3-phenyl
pyrazole substituent of 1a or dihydroxylation of the phenol ring
of 1b. In contrast, oxygenation of hydrolyzed 1a,b leads to
relatively slow autoxidation of free phenol, although addition of
H O also gives the intramolecular aromatic hydroxylation to
form 2. Taken together, the O2 reduction and aromatic
oxidations may be relevant to TPQ generation in nickel-
substituted apo-CAO and to toxicology arising from oxidative
stress mediated by xenobiotic interaction of nickel with
intracellular thiols and phenols.
generate free phenol and intermediate [TpR,MeNi−OH] species
(Scheme 3, left). The latter give rise to the known dimeric
2
2
Me,Me
34
complexes [Tp
and [(Tp
Ni(μ-OH)] (70% spectroscopic yield)
2
Ph,Me
)2Ni] (4, 100% isolated yield). As demonstrated
by quantitative spectroscopic conversion to 2, the hydroxo
complexes may capture H O to form the oxidizing species
2
2
[TpNi(μ-O)] ; complex analogues of this intermediate have
2
Four different reaction mechanisms can be proposed that
been structurally characterized and are known to decompose by
32−34
explain the phenol oxidation and concurrent O reduction: (i)
intramolecular ligand oxidation.
However, phenol oxida-
2
homolysis of the Ni−OAr bond leading to phenoxyl radical and
tion, and thus reduction of oxygen, was inefficient in the
presence of excess water. Therefore, an alternative route to
[TpNi(μ-O)] involving direct reaction of O with 1a,b is
43−45
Ni(I), which captures O ;
(ii) hydrolysis of the Ni−OAr
2
bond and autoxidation of free phenol; (iii) outer-sphere
2
2
oxidation of intact phenolate complexes by O , followed by
proposed (Scheme 3, right). Possible electronic and geometric
2
Ni(III)−OAr bond homolysis; or (iv) inner-sphere ligation and
structures for an initial [TpNi(OPh)(O )] adduct, which was
not observed experimentally, were obtained by DFT calcu-
2
activation of O . With regard to initial Ni(II)−OAr bond
2
homolysis (mechanism i), we clearly demonstrated the thermal
lations (Figure 8B−D). As in proposed TPQ biogenesis
5
0−54
stability of 1a in dry CH Cl . Deliberate addition of H O to
mechanisms,
this intermediate might undergo intra-
2
2
2
1
a,b confirmed that the Ni−OAr bonds are susceptible to
molecular collapse, resulting in oxidative modification of the
phenolate ring. Competing phenoxyl radical homolysis, perhaps
assisted by enhanced steric contact between ortho-phenolate
hydrolysis (mechanism ii); moreover, addition of H O to
hydrolyzed 1a yielded the same product complex 2 as
2
2
7
267
dx.doi.org/10.1021/ic300551z | Inorg. Chem. 2012, 51, 7257−7270

Inorganic Chemistry
Article
and 3-pyrazole substituents in a high-coordinate adduct, would
reduction, the identity of the reactive nickel and reduced
oxygen intermediates, and how these yield the divergent
oxidative modifications of aromatic substrates.
generate a reactive TpNiO intermediate, for which a range of
2
computational structures was also obtained (Figure 8E−I).
Similar complexes were recently isolated and structurally
1
9−26
characterized;
in one case, an isolated (nacnac)NiO2
ASSOCIATED CONTENT
■
complex was found to carry out aromatic oxygenation of a
*
S
Supporting Information
Details of the reactions, product characterizations, and DFT
22
2
,6-disubstituted phenol. Following literature precedent, a
low-spin TpNiO complex, formally a radical species, might
2
calculations; and X-ray crystallographic data for 1a, 1b, and 2−
otherwise dimerize to generate [TpNi(μ-O )] , a bis(μ-1,2-
2
2
4
. This material is available free of charge via the Internet at
2
4,25
superoxo) intermediate akin to TPA-supported analogues.
http://pubs.acs.org.
O extrusion would then effect net superoxide dismutation to
2
yield a peroxo-bridged dimer [TpNi] (μ-O ) akin to an
analogue supported by tmc, and finally the [TpNi(μ-O)]2
2
2
AUTHOR INFORMATION
29
■
Corresponding Author
E-mail: jensenm@ohio.edu.
valence isomer, the same intermediate obtained by hydrolysis
34
*
and H O capture. However, any of the proposed activated
2
2
oxygen complexes (Scheme 3, right) might effect the observed
aromatic oxidations.
Notes
2
2,24
The authors declare no competing financial interest.
The [TpNi(μ-O)] intermediate is capable of two-electron
2
oxidation, giving a 50% theoretical yield of modified ligand
complex 2. Therefore, isolation of 2 in 53% yield likely reflects
minor turnover with residual H O (Scheme 3, center); the
ACKNOWLEDGMENTS
■
The authors acknowledge the donors of the American
Chemical Society Petroleum Research Fund (49296-DNI3)
for support of this research. We also thank the Ohio University
804 Fund for support in acquiring the hardware and software
for DFT computations.
2
2
yield of peroxide equivalents based on observed organic
products is 56 mol % (Scheme 2). The phenolate−nickel(II)
LMCT bands of 2 were partially obscured by organic products
in the oxygenation reaction of 1a (Figure 2), but the intensity
of the shoulder near 317 nm was consistent with a comparable
spectroscopic yield. Moreover, intramolecular C−H bond
hydroxylation was reported to be relatively slow for
1
REFERENCES
■
(
1) Sono, M.; Roach, M. P.; Coulter, E. D.; Dawson, J. H. Chem. Rev.
Me,Me
34
1996, 96, 2841−2888.
[
Tp
Ni(μ-O)]2, which may afford the ultimate oppor-
(
2
(
2) Costas, M.; Mehn, M. P.; Jensen, M. P.; Que, L., Jr. Chem. Rev.
004, 104, 939−986.
3) Kovaleva, E. G.; Neibergall, M. B.; Chakrabarty, S.; Lipscomb, J.
tunity for convergence of hydrolysis and oxygenation pathways,
as well as diversion of oxidation chemistry away from the
supporting Tp ligand and onto the phenol ring in the case of
D. Acc. Chem. Res. 2007, 40, 475−483.
1
Tp
b. In this case, the dimeric complex 3 with unmodified
(4) Krebs, C.; Fujimori, D. G.; Walsh, C. T.; Bollinger, J. M., Jr. Acc.
Chem. Res. 2007, 40, 484−492.
Me,Me
ligands was obtained in 51% isolated yield. The
complex yields are thus consistent with Scheme 3; however, the
mass balances of nickel complex(es) remaining after isolation of
(5) Tshuva, E. Y.; Lippard, S. J. Chem. Rev. 2004, 104, 987−1012.
(
1
(
6) Decker, A.; Solomon, E. I. Curr. Opin. Chem. Biol. 2005, 9, 152−
63.
7) Korendovych, I. V.; Kryatov, S. V.; Rybak-Akimova, E. V. Acc.
Chem. Res. 2007, 40, 510−521.
8) Mirica, L. M.; Ottenwaelder, X.; Stack, T. D. P. Chem. Rev. 2004,
04, 1013−1046.
9) Lewis, E. A.; Tolman, W. B. Chem. Rev. 2004, 104, 1047−1076.
10) Kim, E.; Chufan, E. E.; Kamaraj, K.; Karlin, K. D. Chem. Rev.
2
and 3 and yields of organic products arising from oxygenation
of 1b were not determined. Hence, other mechanistic pathways
are not excluded.
(
1
(
(
Regardless of the mechanistic details of O reduction, it is
2
clear this chemistry occurs at the expense of the phenolate
moiety during decomposition of 1a,b; both this chemistry and
the subsequent aromatic oxidations may be relevant to nickel-
catalyzed TPQ biogenesis in apo-CAO. Oxidation of 1a leads
to C−H bond hydroxylation of a 3-phenylpyrazole substituent
on the supporting TpPh,Me ligand, while oxidation of 1b leads to
a different modification of the phenolate ring. Thus, another
significant question remaining to be explored is the selectivity
of the oxidation chemistry in the case of 1a, where two different
aromatic substrates are present and a peroxide shunt was also
demonstrated.
́
2004, 104, 1077−1134.
(11) Ju, T.; Goldsmith, R. B.; Chai, S. C.; Maroney, M. J.; Pochapsky,
S. S.; Pochapsky, T. C. J. Mol. Biol. 2006, 363, 823−834.
12) Szajna, E.; Arif, A. M.; Berreau, L. M. J. Am. Chem. Soc. 2005,
27, 17186−17187.
13) Kimura, E.; Sakonaka, A.; Machida, R.; Kodama, M. J. Am.
Chem. Soc. 1982, 104, 4255−4257.
14) Edison, S. E.; Hotz, R. P.; Baldwin, M. J. Chem. Commun. 2004,
(
1
(
(
1
(
(
212−1213.
15) Koola, J. D.; Kochi, J. K. Inorg. Chem. 1987, 26, 908−916.
In summary, we prepared bioinspired TpNi−OAr complexes
16) Kinneary, J. F.; Albert, J. S.; Burrows, C. J. J. Am. Chem. Soc.
1
a,b that closely approximate the N O (His Tyr) ligand field of
1988, 110, 6124−6129.
3
3
apo-CAO. Similar to TPQ biogenesis (Scheme 1), 1a,b
decompose in aerobic solutions, leading to phenoxyl radical
(17) Nagataki, T.; Tachi, Y.; Itoh, S. Chem. Commun. 2006, 4016−
4
(
3
(
018.
18) Grapperhaus, C. A.; Darensbourg, M. Y. Acc. Chem. Res. 1998,
1, 451−459.
19) Fujita, K.; Schenker, R.; Gu, W.; Brunold, T. C.; Cramer, S. P.;
Riordan, C. G. Inorg. Chem. 2004, 43, 3324−3326.
20) Kieber-Emmons, M. T.; Annaraj, J.; Seo, M. S.; Van Heuvelen,
K. M.; Tosha, T.; Kitagawa, T.; Brunold, T. C.; Nam, W.; Riordan, C.
G. J. Am. Chem. Soc. 2006, 128, 14230−14231.
(21) Yao, S.; Bill, E.; Milsmann, C.; Wieghardt, K.; Driess, M. Angew.
Chem., Int. Ed. 2008, 47, 7110−7113.
generation, O reduction, and oxidation of aromatic substrates.
2
Hydrolysis products were identified, and subsequent phenol
autoxidation and H O activation were demonstrated. Having
2
2
identified several organic and complex products in a
complicated network of competitive reaction pathways
(
(Schemes 2 and 3), we are positioned to address fundamental
kinetic and mechanistic questions in future studies. Relevant
issues raised by this investigation include the nature of the O2
7
268
dx.doi.org/10.1021/ic300551z | Inorg. Chem. 2012, 51, 7257−7270

Inorganic Chemistry
Article
(
1
(
2
22) Company, A.; Yao, S.; Ray, K.; Driess, M. Chem.−Eur. J 2010,
6, 9669−9675.
23) Pietrzyk, P.; Podolska, K.; Mazur, T.; Sojka, Z. J. Am. Chem. Soc.
011, 133, 19931−19943.
24) Shiren, K.; Ogo, S.; Fujinami, S.; Hayashi, H.; Suzuki, M.;
Uehara, A.; Watanabe, Y.; Moro-oka, Y. J. Am. Chem. Soc. 2000, 122,
54−262.
25) Cho, J.; Furutachi, H.; Fujinami, S.; Tosha, T.; Ohtsu, H.; Ikeda,
(54) Samuels, N. M.; Klinman, J. P. Biochemistry 2005, 44, 14308−
14317.
(55) Fujisawa, K.; Iwata, Y.; Kitajima, N.; Higashimura, H.; Kubota,
M.; Miyashita, Y.; Yamada, Y.; Okamoto, K.; Moro-oka, Y. Chem. Lett.
1999, 739−740.
(56) Jazdzewski, B. A.; Holland, P. L.; Pink, M.; Young, V. G., Jr.;
Spencer, D. J. E.; Tolman, W. B. Inorg. Chem. 2001, 40, 6097−6107.
(57) Gorelsky, S. I.; Basumallick, L.; Vura-Weis, J.; Sarangi, R.;
Hodgson, K. O.; Hedman, B.; Fujisawa, K.; Solomon, E. I. Inorg. Chem.
2005, 44, 4947−4960.
(58) Chattopadhyay, S.; Deb, T.; Ma, H.; Petersen, J. L.; Young, V.
G., Jr.; Jensen, M. P. Inorg. Chem. 2008, 47, 3384−3392.
(59) Chattopadhyay, S.; Deb, T.; Petersen, J. L.; Young, V. G., Jr.;
Jensen, M. P. Inorg. Chem. 2010, 49, 457−467.
(60) Desrochers, P. J.; Cutts, R. W.; Rice, P. K.; Golden, M. L.;
Graham, J. B.; Barclay, T. M.; Cordes, A. W. Inorg. Chem. 1999, 38,
5690−5694.
(61) Nakazawa, J.; Ogiwara, H.; Kashiwazaki, Y.; Ishii, A.; Imamura,
N.; Samejima, Y.; Hikichi, S. Inorg. Chem. 2011, 50, 9933−9935.
(62) Rheingold, A. L.; Ostrander, R. L.; Haggerty, B. S.; Trofimenko,
S. Inorg. Chem. 1994, 33, 3666−3676.
(63) Trofimenko, S. J. Am. Chem. Soc. 1967, 89, 6288−6294.
(64) Tabuchi, K.; Ertem, M. Z.; Sugimoto, H.; Kunishita, A.; Tano,
T.; Fujieda, N.; Cramer, C. J.; Itoh, S. Inorg. Chem. 2011, 50, 1633−
1647.
(65) Santi, R.; Romano, A. M.; Sommazzi, A.; Grande, M.; Bianchini,
C.; Mantovani, G. J. Mol. Catal. A 2005, 229, 191−197.
(66) Uehara, K.; Hikichi, S.; Akita, M. J. Chem. Soc., Dalton Trans.
2002, 3529−3538.
(67) Evans, D. F.; Jakubovic, D. A. J. Chem. Soc., Dalton Trans. 1988,
2927−2933.
(68) Heyne, B.; Kohnen, S.; Brault, D.; Mouithys-Mickalad, A.;
Tfibel, F.; Hans, P.; Fontaine-Aupart, M.-P.; Hoebeke, M. Photochem.
Photobiol. Sci. 2003, 2, 939−945.
(69) ADF 2008.01; Scientific Computing and Modelling NV; Vrije
Universiteit: Amsterdam, Netherlands, 2008.
(70) te Velde, G.; Bickelhaupt, F. M.; Baerends, E. J.; Fonseca
Guerra, C.; van Gisbergen, S. J. A.; Snijders, J. G.; Ziegler, T. J.
Comput. Chem. 2001, 22, 931−967.
(71) Vosko, S. H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200−
1211.
(
2
(
O.; Suzuki, A.; Nomura, M.; Uruga, T.; Tanida, H.; Kawai, T.; Tanaka,
K.; Kitagawa, T.; Suzuki, M. Inorg. Chem. 2006, 45, 2873−2885.
(
26) Cho, J.; Sarangi, R.; Annaraj, J.; Kim, S. Y.; Kubo, M.; Ogura, T.;
Solomon, E. I.; Nam, W. Nat. Chem. 2009, 1, 568−572.
27) Cho, J.; Furutachi, H.; Fujinami, S.; Suzuki, M. Angew. Chem.,
Int. Ed. 2004, 43, 3300−3303.
28) Kieber-Emmons, M. T.; Schenker, R.; Yap, G. P. A.; Brunold, T.
C.; Riordan, C. G. Angew. Chem., Int. Ed. 2004, 43, 6716−6718.
29) Schenker, R.; Kieber-Emmons, M. T.; Riordan, C. G.; Brunold,
T. C. Inorg. Chem. 2005, 44, 1752−1762.
30) Hikichi, S.; Kobayashi, C.; Yoshizawa, M.; Akita, M. Chem.−
Asian J 2010, 5, 2086−2092.
31) Hikichi, S.; Okuda, H.; Ohzu, Y.; Akita, M. Angew. Chem., Int.
Ed. 2009, 48, 188−191.
32) Hikichi, S.; Yoshizawa, M.; Sasakura, Y.; Akita, M.; Moro-oka, Y.
J. Am. Chem. Soc. 1998, 120, 10567−10568.
33) Hikichi, S.; Yoshizawa, M.; Sasakura, Y.; Komatsuzaki, H.; Akita,
M.; Moro-oka, Y. Chem. Lett. 1999, 979−980.
34) Hikichi, S.; Yoshizawa, M.; Sasakura, Y.; Komatsuzaki, H.;
Moro-oka, Y.; Akita, M. Chem.−Eur. J 2001, 7, 5011−5028.
35) Itoh, S.; Bandoh, H.; Nagatomo, S.; Kitagawa, T.; Fukuzumi, S.
J. Am. Chem. Soc. 1999, 121, 8945−8946.
36) Itoh, S.; Bandoh, H.; Nakagawa, M.; Nagatomo, S.; Kitagawa,
T.; Karlin, K. D.; Fukuzumi, S. J. Am. Chem. Soc. 2001, 123, 11168−
1178.
37) Bag, B.; Mondal, N.; Rosair, G.; Mitra, S. Chem. Commun. 2000,
729−1730.
38) Mandimutsira, B. S.; Yamarik, J. L.; Brunold, T. C.; Gu, W.;
Cramer, S. P.; Riordan, C. G. J. Am. Chem. Soc. 2001, 123, 9194−9195.
39) Schenker, R.; Mandimutsira, B. S.; Riordan, C. G.; Brunold, T.
C. J. Am. Chem. Soc. 2002, 124, 13842−13855.
40) Honda, K.; Cho, J.; Matsumoto, T.; Roh, J.; Furutachi, H.;
(
(
(
(
(
(
(
(
(
(
1
(
1
(
(
(
Tosha, T.; Kubo, M.; Fujinami, S.; Ogura, T.; Kitagawa, T.; Suzuki, M.
Angew. Chem., Int. Ed. 2009, 48, 3304−3307.
(72) Becke, A. D. Phys. Rev. 1988, A38, 3098−3100.
(73) Perdew, J. P. Phys. Rev. 1986, B33, 8822−8824.
(74) APEX II; Bruker Analytical X-ray Systems: Madison, WI, 2001.
(75) An empirical correction for absorption anisotropy: Blessing, R.
H. Acta Crystallogr. 1995, A51, 33−38.
(
41) Kunishita, A.; Doi, Y.; Kubo, M.; Ogura, T.; Sugimoto, H.; Itoh,
S. Inorg. Chem. 2009, 48, 4997−5004.
42) Tano, T.; Doi, Y.; Inosako, M.; Kunishita, A.; Kubo, M.;
Ishimaru, H.; Ogura, T.; Sugimoto, H.; Itoh, S. Bull. Chem. Soc. Jpn.
(
2
(
(
010, 83, 530−538.
(76) SAINT+, V7.68; Bruker Analytical X-Ray Systems: Madison,
WI, 2003.
43) Yao, S.; Driess, M. Acc. Chem. Res. 2012, 45, 276−287.
44) Kieber-Emmons, M. T.; Riordan, C. G. Acc. Chem. Res. 2007, 40,
(77) SHELXTL, V2008/4; Bruker Analytical X-Ray Systems:
6
(
(
18−625.
Madison, WI, 2008.
(78) Mercury, 3.0; Cambridge Crystallographic Data Centre:
Cambridge, U.K., 2011.
45) Suzuki, M. Acc. Chem. Res. 2007, 40, 609−617.
46) Cheng, C.-C.; Rokita, S. E.; Burrows, C. J. Angew. Chem., Int. Ed.
Engl. 1993, 32, 277−278.
47) Liang, Q.; Ananias, D. C.; Long, E. C. J. Am. Chem. Soc. 1998,
20, 248−257.
48) Kasprzak, K. S.; Salnikow, K. Met. Ions Life Sci. 2007, 2, 619−
60.
49) Kim, M.; Okajima, T.; Kishishita, S.; Yoshimura, M.; Kawamori,
A.; Tanizawa, K.; Yamaguchi, H. Nat. Struct. Biol. 2002, 9, 591−596.
50) DuBois, J. L.; Klinman, J. P. Biochemistry 2005, 44, 11381−
1388.
51) Prabhakar, R.; Siegbahn, P. E. M. J. Am. Chem. Soc. 2004, 126,
996−4006.
52) Ghosh, S.; Cirera, J.; Vance, M. A.; Ono, T.; Fujisawa, K.;
Solomon, E. I. J. Am. Chem. Soc. 2008, 130, 16262−16273.
53) Okajima, T.; Kishishita, S.; Chiu, Y.-C.; Murakawa, T.; Kim, M.;
Yamaguchi, H.; Hirota, S.; Kuroda, S.; Tanizawa, K. Biochemistry 2005,
4, 12041−12048.
(79) Ruman, T.; Ciunik, Z.; Szklanny, E.; Wołowiec, S. Polyhedron
(
1
(
6
2002, 21, 2743−2753.
(80) Kharasch, M. S.; Joshi, B. S. J. Org. Chem. 1957, 22, 1439−1443.
(81) Yang, L.; Powell, D. R.; Houser, R. P. Dalton Trans. 2007, 955−
964.
(
(82) Desrochers, P. J.; LeLievre, S.; Johnson, R. J.; Lamb, B. T.;
Phelps, A. L.; Cordes, A. W.; Gu, W.; Cramer, S. P. Inorg. Chem. 2003,
42, 7945−7950.
(83) Northcutt, T. O.; Lachicotte, R. J.; Jones, W. D. Organometallics
1998, 17, 5148−5152.
(84) Desrochers, P. J.; Telser, J.; Zvyagin, S. A.; Ozarowski, A.;
Krzystek, J.; Vicic, D. A. Inorg. Chem. 2006, 45, 8930−8941.
(85) Addison, A. W.; Rao, T. N.; Reedijk, J.; van Rijn, J.; Verschoor,
G. C. J. Chem. Soc., Dalton Trans. 1984, 1349−1356.
(86) Fujisawa, K.; Tada, N.; Nishida, Y.; Miyashita, Y.; Okamoto, K.
Inorg. Chem. Commun. 2008, 11, 381−384.
(
1
(
3
(
(
4
7
269
dx.doi.org/10.1021/ic300551z | Inorg. Chem. 2012, 51, 7257−7270

Inorganic Chemistry
Article
(
87) Mehn, M. P.; Fujisawa, K.; Hegg, E. L.; Que, L., Jr. J. Am. Chem.
Soc. 2003, 125, 7828−7842.
(
(
88) Higgs, T. C.; Carrano, C. J. Inorg. Chem. 1997, 36, 298−306.
89) Kitajima, N.; Hikichi, S.; Tanaka, M.; Moro-oka, Y. J. Am. Chem.
Soc. 1993, 115, 5496−5508.
90) Trofimenko, S.; Rheingold, A. L.; Liable Sands, L. M. Inorg.
Chem. 2002, 41, 1889−1896.
91) Janiak, C.; Temizdemir, S.; Dechert, S.; Deck, W.; Girgsdies, F.;
(
(
Heinze, J.; Kolm, M. J.; Scharmann, T. G.; Zipffel, O. M. Eur. J. Inorg.
Chem. 2000, 1229−1241.
(
92) Bandoli, G.; Clemente, D. A.; Paolucci, G.; Doretti, L. Cryst.
Struct. Commun. 1979, 8, 965−970.
93) Cecchi, P.; Gioia Lobbia, G.; Marchetti, F.; Valle, G.; Calogero,
S. Polyhedron 1994, 13, 2173−2178.
94) Desrochers, P. J.; Brown, J. R.; Arvin, M. E.; Jones, G. D.; Vicic,
D. A. Acta Crystallogr. 2005, E61, m1455−m1458.
95) Calabrese, J. C.; Trofimenko, S. Inorg. Chem. 1992, 31, 4810−
814.
96) Santana, M. D.; Lop
Liu, M. Eur. J. Inorg. Chem. 2008, 4012−4018.
97) Hammes, B. S.; Luo, X.; Chohan, B. S.; Carrano, M. W.;
Carrano, C. J. J. Chem. Soc., Dalton Trans. 2002, 3374−3380.
98) Adams, H.; Batten, S. R.; Davies, G. M.; Duriska, M. B.; Jeffery,
(
(
(
4
(
́ ́
ez-Banet, L.; García, G.; García, L.; Perez, J.;
(
(
J. C.; Jensen, P.; Lu, J.; Motson, G. R.; Coles, S. J.; Hursthouse, M. B.;
Ward, M. D. Dalton Trans. 2005, 1910−1923.
(
99) Batten, S. R.; Duriska, M. B.; Jensen, P.; Lu, J. Aust. J. Chem.
2
(
1
(
007, 60, 72−74.
100) Jesson, J. P.; Trofimenko, S.; Eaton, D. R. J. Am. Chem. Soc.
967, 89, 3148−3158.
101) Fulmer, G. R.; Miller, A. J. M.; Sherden, N. H.; Gottlieb, H. E.;
Nudelman, A.; Stoltz, B. M.; Bercaw, J. E.; Goldberg, K. I.
Organometallics 2010, 29, 2176−2179.
7
270
dx.doi.org/10.1021/ic300551z | Inorg. Chem. 2012, 51, 7257−7270