Steric and Electronic Influence on Proton-Coupled Electron-Transfer Reactivity of a Mononuclear Mn(III)-Hydroxo Complex

Article abstract of DOI:10.1021/acs.inorgchem.6b01217

A mononuclear hydroxomanganese(III) complex was synthesized utilizing the N₅ amide-containing ligand 2-[bis(pyridin-2-ylmethyl)]amino-N-2-methyl-quinolin-8-yl-acetamidate (dpaq^2Me). This complex is similar to previously reported [Mn^III(OH)(dpaq^H)]⁺ [Inorg. Chem. 2014, 53, 7622-7634] but contains a methyl group adjacent to the hydroxo moiety. This α-methylquinoline group in [Mn^III(OH)(dpaq^2Me)]⁺ gives rise to a 0.1 ? elongation in the Mn-N(quinoline) distance relative to [Mn^III(OH)(dpaq^H)]⁺. Similar bond elongation is observed in the corresponding Mn(II) complex. In MeCN, [Mn^III(OH)(dpaq^2Me)]⁺ reacts rapidly with 2,2′,6,6′-tetramethylpiperidine-1-ol (TEMPOH) at -35 °C by a concerted proton-electron transfer (CPET) mechanism (second-order rate constant k₂ of 3.9(3) M^-1 s^-1). Using enthalpies and entropies of activation from variable-temperature studies of TEMPOH oxidation by [Mn^III(OH)(dpaq^2Me)]⁺ (ΔH^? = 5.7(3) kcal^-1 M^-1 ΔS^? = -41(1) cal M^-1 K^-1), it was determined that [Mn^III(OH)(dpaq^2Me)]⁺ oxidizes TEMPOH ～240 times faster than [Mn^III(OH)(dpaq^H)]⁺. The [Mn^III(OH)(dpaq^2Me)]⁺ complex is also capable of oxidizing the stronger O-H and C-H bonds of 2,4,6-tri-tert-butylphenol and xanthene, respectively. However, for these reactions [Mn^III(OH)(dpaq^2Me)]⁺ displays, at best, modest rate enhancement relative to [Mn^III(OH)(dpaq^H)]⁺. A combination of density function theory (DFT) and cyclic voltammetry studies establish an increase in the Mn^III/Mn^II reduction potential of [Mn^III(OH)(dpaq^2Me)]⁺ relative to [Mn^III(OH)(dpaq^H)]⁺, which gives rise to a larger driving force for CPET for the former complex. Thus, more favorable thermodynamics for [Mn^III(OH)(dpaq^2Me)]⁺ can account for the dramatic increase in rate with TEMPOH. For the more sterically encumbered substrates, DFT computations suggest that this effect is mitigated by unfavorable steric interactions between the substrate and the α-methylquinoline group of the dpaq^2Me ligand. The DFT calculations, which reproduce the experimental activation free energies quite well, provide the first examination of the transition-state structure of mononuclear Mn^III(OH) species during a CPET reaction.

Full text of DOI:10.1021/acs.inorgchem.6b01217

Article
pubs.acs.org/IC
Steric and Electronic Influence on Proton-Coupled Electron-Transfer
Reactivity of a Mononuclear Mn(III)-Hydroxo Complex
Derek B. Rice, Gayan B. Wijeratne, Andrew D. Burr, Joshua D. Parham, Victor W. Day,
and Timothy A. Jackson*
Department of Chemistry and Center for Environmentally Beneficial Catalysis, University of Kansas, Lawrence, Kansas 66045, United
States
*
S Supporting Information
ABSTRACT: A mononuclear hydroxomanganese(III) complex was
synthesized utilizing the N amide-containing ligand 2-[bis(pyridin-2-
5
2
Me
ylmethyl)]amino-N-2-methyl-quinolin-8-yl-acetamidate (dpaq ). This
III
H
+
complex is similar to previously reported [Mn (OH)(dpaq )] [Inorg.
Chem. 2014, 53, 7622−7634] but contains a methyl group adjacent to
III
the hydroxo moiety. This α-methylquinoline group in [Mn (OH)-
Me
+
(
dpaq2 )] gives rise to a 0.1 Å elongation in the Mn−N(quinoline)
III H +
distance relative to [Mn (OH)(dpaq )] . Similar bond elongation is
III
observed in the corresponding Mn(II) complex. In MeCN, [Mn (OH)-
Me
+
(
(
(
dpaq2 )] reacts rapidly with 2,2′,6,6′-tetramethylpiperidine-1-ol
TEMPOH) at −35 °C by a concerted proton−electron transfer
−1 −1
CPET) mechanism (second-order rate constant k of 3.9(3) M s ).
2
Using enthalpies and entropies of activation from variable-temperature
III
2Me
+
‡
−1
−1
‡
−1
−1
studies of TEMPOH oxidation by [Mn (OH)(dpaq )] (ΔH = 5.7(3) kcal M ; ΔS = −41(1) cal M K ), it was
III
2Me
+
III
H
+
III
determined that [Mn (OH)(dpaq )] oxidizes TEMPOH ∼240 times faster than [Mn (OH)(dpaq )] . The [Mn (OH)-
(
dpaq2 )] complex is also capable of oxidizing the stronger O−H and C−H bonds of 2,4,6-tri-tert-butylphenol and xanthene,
Me
+
III 2Me +
respectively. However, for these reactions [Mn (OH)(dpaq )] displays, at best, modest rate enhancement relative to
III
H
+
[
Mn (OH)(dpaq )] . A combination of density function theory (DFT) and cyclic voltammetry studies establish an increase in
III
II
III
2Me
+
III
H
+
the Mn /Mn reduction potential of [Mn (OH)(dpaq )] relative to [Mn (OH)(dpaq )] , which gives rise to a larger
III
2Me
+
driving force for CPET for the former complex. Thus, more favorable thermodynamics for [Mn (OH)(dpaq )] can account
for the dramatic increase in rate with TEMPOH. For the more sterically encumbered substrates, DFT computations suggest that
this effect is mitigated by unfavorable steric interactions between the substrate and the α-methylquinoline group of the dpaq
2
Me
ligand. The DFT calculations, which reproduce the experimental activation free energies quite well, provide the first examination
III
of the transition-state structure of mononuclear Mn (OH) species during a CPET reaction.
1
6,17
INTRODUCTION
environment and PCET reactivity.
For example, the
■
III
2+
III
[Mn (OH)(PY5)] complex of Stack and co-workers, which
features a neutral N ligand (PY5 = 2,6-bis(bis(2-pyridyl)-
methoxymethanepyridine)), can oxidize the relatively strong
Mn -hydroxo motifs are proposed in the catalytic cycles of Mn
1
−5
5
lipoxygenase (MnLOX)
OEC) of photosystem II. In MnLOX, a Mn -hydroxo
and the oxygen evolving complex
6
,7
III
(
C−H bonds of toluene (bond dissociation free energy (BDFE)
abstracts an H atom from a polyunsaturated fatty acid to initiate
its dioxygenation. This step proceeds with a large H/D kinetic
isotope effect (KIE) of 20−24, implying a concerted proton−
electron transfer (CPET). (CPET is a subclass of proton-
coupled electron-transfer (PCET) reactions, where the proton
and electron are transferred in the same kinetic step.) PCET
16
III
Me2
+
of 87 kcal/mol). However, the [Mn (OH)(S N (tren))]
4
−
complex of Kovacs and co-workers, which has a N S ligand
4
Me2
(
S
N (tren) = 3-((2-(bis(2-aminoethyl)-amino)ethyl)imino-
4
2
-methylbutane-2-thiolate), can only attack the weak O−H
bond of 2,2′,6,6′-tetramethylpiperidine-1-ol (TEMPOH; BDFE
1
7
III
III
= 66.5 kcal/mol). The substantial differences in Mn
environment for this pair of complexes obscure the impact of
from a Mn -hydroxo unit is also a possible pathway for
stepwise oxidation of the OEC during the water splitting
III/II
6
,7
general factors, such as Mn
reduction potentials and steric
cycle, although this mechanism remains disputed. Despite the
III
effects, on PCET reactivity.
importance of Mn -hydroxo complexes in PCET processes,
II
III
We have reported that the Mn complex of the anionic N5
ligand dpaq (1-H; dpaq = 2-[bis(pyridin-2-ylmethyl)]amino-N-
there are few examples of synthetic Mn -hydroxo complexes
known to mediate such reactions. This is in great contrast to
the well-studied reactivity of high-valent Mn(IV)-oxo centers in
8
−15
PCET processes.
Furthermore, of the known hydroxide
Received: May 18, 2016
III
complexes there is substantial variability in Mn coordination
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.6b01217
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
quinolin-8-yl-acetamidate) forms a hydroxomanganese(III)
immediately taken into an argon-filled glovebox and were stored in
tightly sealed Schlenk glassware. All experimental procedures were
performed under an argon atmosphere unless otherwise stated. The
species in the presence of O at ambient conditions in
2
1
8
III
+
MeCN (Scheme 1). The [Mn (OH)(dpaq)] (2-H)
purity of O gas used was greater than 99% and was further purified
2
using a gas purifier column packed with drierite and 5 Å molecular
sieves prior to use. TEMPOH, deuterated TEMPOH (TEMPOD),
and 2,4,6-tri-tert-butylphenol-d were synthesized as described in
a
Scheme 1
9
,28
1
literature procedures.
All H NMR spectra were collected on a
Bruker DRZ 400 MHz spectrometer. Mass spectrometry experiments
were performed using an LCT Primers MicroMass electrospray time
of-flight instrument. Electron paramagnetic resonance (EPR) experi-
ments were performed on a Bruker EMXPlus spectrometer with a
dual-mode cavity and an Oxford instruments cryostat.
Synthesis of Hdpaq2 . 2-Methylquinolin-8-amine (1.00 g, 6.32
mmol) and 2-bromoacetyl bromide (0.725 mL, 8.32 mmol) were
combined in acetonitrile (20 mL) along with excess Na CO (1.03 g,
Me
2
3
9
.7 mmol) under an inert atmosphere, and the resulting solution was
stirred for 30 min at 0 °C. The solution was then allowed to warm to
room temperature, filtered through diatomaceous earth, and
evaporated to dryness under reduced pressure. The solid product
was combined with dipicolyl amine (1.5 mL, 8.33 mmol) and Na CO
2
3
(1.03g, 9.7 mmol) in acetonitrile (40 mL) and was stirred overnight at
0
°C under an inert atmosphere. The final reaction mixture was filtered
through diatomaceous earth and evaporated to dryness. The crude
product was purified by flash chromatography on a silica column using
a MeOH/CH Cl mixture with an initial composition of 0.5% MeOH
aChemical structure of the HdpaqR ligand (top left). Electronic
2
2
absorption spectra of 2.5 mM MeCN solutions of 2-H and 2-Me after
that was gradually increased to 5% MeOH. The pure ligand was
II
full formation upon reaction of Mn complexes with O gas at 25° C
2
isolated in good yield (77%) as a brown oil and was characterized by
II
R
+
(
top right). Reaction scheme for the conversion of [Mn (dpaq )] to
Mn (OH)(dpaq )] in the presence of O (bottom). In the solid-
state structure of 1-H and 1-Me, the Mn coordination sphere is
completed by an amide oxygen from a separate Mn molecule.
1
13
H and C NMR methods and electrospray ionization mass
III
R
+
[
1
2Me
2
spectrometry (ESI-MS). H NMR data (400 MHz) for Hdpaq
II
(
(
CDCl , δ) = 11.52 (s, 1H), 8.70 (dd, J = 6.4, 2.7 Hz, 1H), 8.50−8.40
II
3
m, 2H), 8.02 (d, J = 8.3 Hz, 1H), 7.92 (d, J = 7.8 Hz, 2H), 7.54 (td, J
=
7.6, 1.9 Hz, 2H), 7.46−7.28 (m, 3H), 7.15−7.02 (m, 2H), 3.93 (s,
13
4
H), 3.45(s, 2H), 2.82 (d, J = 4.3 Hz, 3H). C NMR data (500 MHz)
complex abstracts hydrogen atoms from moderate-strength
O−H and C−H bonds (BDFE < 79 kcal/mol) by a CPET
mechanism. A dimeric bis(μ-oxo)manganese(III,IV) complex
supported by the dpaq ligand has also recently been shown
2Me
for Hdpaq (CDCl , δ) = 25.2 (s, CH Qu-), 59.6 (s, -CH2 CO−),
3
3
6
(
1
1.1 (s, -CH2Py), 116.6 (s, Qu6), 121.5 (s; Qu3, Qu5, or Qu7), 122.3
s; Qu3, Qu5, or Qu7), 122.4 (s, Py5), 123.1 (s, Py3), 126.3 (s,Qu10),
26.5 (s; Qu3, Qu5, or Qu7),133.8 (s, Qu8), 136.6 (s, Qu4), 136.8 (s,
1
9
capable of initiating hydrogen atom abstraction reactions.
Py4), 138.2 (s, Qu9), 149.2 (s, Py6), 156.9 (s, Qu2), 158.5 (s, Py2),
169.5 (s, CO). The numbering of the quinoline carbon atoms refers
to the standard numbering scheme, as illustrated in Figure S12. ESI-
Importantly, the dpaq ligand is easily amenable to
modification, as employed by Hitomi and co-workers to tune
20−24
2Me
+
the reactivity of Fe and Mn-NO complexes.
Here we
MS: {Hdpaq + Na] } m/z = 420.1766 (calc. = 420.1800)
II
2‑Me
Preparation of [Mn (dpaq )](OTf). To a stirred solution of
describe modification of the dpaq ligand through incorporation
2Me
2‑Me
103 mg (0.26 mmol) of Hdpaq in 2 mL of MeOH was added 113
mg (0.26 mmol) of Mn (OTf) in 2 mL of MeOH, followed by 25 mg
0.26 mmol) of NaO Bu in 2 mL of MeOH under an inert
atmosphere. The Mn (OTf) salt was generated using a previously
of a 2-methylquinoline moiety (dpaq
= 2-[bis(pyridin-2-
II
2
ylmethyl)]amino-N-2-methyl-quinolin-8-yl-acetamidate), pro-
viding steric bulk adjacent to the N-donor function and in
proximity to the reaction center (Scheme 1). Quinoline and
pyridine ligands with methyl substituents adjacent to the N-
donor support longer metal−N interactions, impacting metal
In this present case, this ligand
perturbation dramatically increases the rate of TEMPOH
oxidation by the Mn -hydroxo unit, while the rates of oxidation
of bulkier, or more rigid, substrates are largely unaffected.
Density functional theory (DFT) computations are used to aid
in understanding the electronic and steric factors accounting for
this interesting reactivity pattern.
t
(
II
2
29
reported method. The orange colored resultant solution was stirred
overnight, then filtered, and evaporated to dryness under vacuum. The
solid product was recrystallized by layering Et O onto a MeOH
2
25−27
II
electronic structure.
solution of the Mn product to yield yellow crystals of
[Mn (dpaq )](OTf). Crystals for X-ray diffraction analysis were
obtained by subsequent recrystallization of the final solid product in
II
2Me
III
II
2Me
the MeOH/Et O solvent system. [Mn (dpaq )](OTf) was further
2
characterized by ESI-MS (Figure S9), EPR spectroscopy (Figure S8),
X-ray crystallography (Figure S1 and Table S1), and elemental
II
2Me
analysis. Elemental analysis [Mn (dpaq )](OTf): C H F Mn-
2
5
22 3
N O S calcd. (%) C 50.01, H 3.69, N 11.66; found (%): C 49.15,
5
4
H 3.68, N 11.41.
METHODS
■
X-ray Diffraction Data Collection and Analysis for
II
2Me
Materials and Instrumentation. All chemicals and solvents were
obtained from commercial vendors at ACS grade or better and, unless
otherwise described, were used without further purification. Acetoni-
trile, methanol, and ether were dried and degassed using a Pure Solv
[Mn (dpaq )](OTf). A yellow colored single-domain crystal of
II 2Me
[Mn (dpaq )](OTf) was suspended with mineral oil on a MiteGen
MicroMount and placed on a goniometer head in a cold nitrogen
stream at 100 K for a single-crystal X-ray structure determination.
Monochromatic X-rays were provided by a Bruker diffractometer
equipped with Helios high-brilliance multilayer optics, a Platinum 135
CCD detector and a Bruker MicroStar microfocus rotating anode X-
ray source operating at 45 kV and 60 mA. Intensity data (6590 0.5°-
wide ω- or f-scan frames with counting times of 4−6 s each) were
(
2010) solvent purification system. These solvents were degassed in
airtight solvent reservoirs (4 L) by passing Ar gas at room temperature
for 20 min. Acetonitrile and ether were dried using airtight alumina
columns, and methanol was dried using a drierite column. Anhydrous
dichloromethane and deuterated acetonitrile (MeCN-d ) were
purchased from Acros Organics. All dried, degassed solvents were
3
3
0
collected with the Bruker program SMART, and diffracted intensities
B
DOI: 10.1021/acs.inorgchem.6b01217
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
3
1
32
33
were measured with the Bruker program SAINT. The space group
software package. The final structural model incorporated
anisotropic thermal parameters for all full- and major-occupancy
non-hydrogen atoms and isotropic thermal parameters for all
hydrogen atoms and the three minor-occupancy fluorine atoms.
and crystallographic data are summarized in Table S1. The Bruker
software package SHELXTL was used to solve the structure and locate
all non-hydrogen atoms with “direct methods” techniques. Each
methyl group was incorporated into the structural model as a fixed
rigid group (using idealized sp -hybridized geometry and a C−H bond
length of 0.98 Å) with idealized “staggered” geometry. The remaining
III
2Me +
Kinetics Studies of [Mn (OH)(dpaq )] with TEMPOH. For
3
III
2Me
+
each experiment, a 1.25 mM [Mn (OH)(dpaq )] acetonitrile
solution was prepared in an argon glovebox, placed in a gastight
cuvette, and sealed with a septum. The appropriate amount of
TEMPOH (10−40 equiv) was dissolved in 100 μL of acetonitrile and
sealed in a vial with a pierceable cap. The gastight syringe used to
inject the TEMPOH was first purged with argon before use.
Temperature was maintained using a Unisoku cryostat at −35 °C,
and the cuvette was allowed to equilibrate to the temperature for 10
min before injection took place. Kinetics measurements were made
using an Agilent 8453 spectrophotometer. Variable-temperature
reactions (−35−15 °C) were performed following the same procedure
as above. The final reaction product was analyzed by X-band EPR
spectroscopy at 5 K, which showed a signal consistent with a TEMPO
radical (Figure S8).
hydrogen atoms were included in the structural model as idealized
2
3
atoms (assuming sp or sp hybridization of the carbon and C−H
bond lengths of 0.95−0.99 Å). The isotropic thermal parameters of all
idealized hydrogen atoms were fixed at values 1.2 (nonmethyl) or 1.5
methyl) times the equivalent isotropic thermal parameter of the
carbon atom to which they are covalently bonded.
All three triflate anions are 50:50 disordered over two closely
separated sites in the unit cell. The first and third triflates are
disordered about crystallographic inversion centers at (1/2, 0, 0) and
1/2, 0, 1/2), respectively. The bond lengths and angles of the triflates
were mildly restrained to have similar values. All stages of weighted
(
(
2
o
full-matrix least-squares refinement were conducted using F data
3
3
III
2Me +
with the SHELXXTL XLMP v2013/4 software package. The final
structural model incorporated anisotropic thermal parameters for all
non-hydrogen atoms and isotopic thermal parameters for all hydrogen
Kinetics Studies of [Mn (OH)(dpaq )] with 2,4,6-tri-tert-
III
butylphenol. For each kinetics experiment, a 1.25 mM [Mn (OH)-
(dpaq2 )] solution was prepared in 2 mL of acetonitrile within an
argon-filled glovebox, placed in a gastight cuvette, and sealed with a
septum. The appropriate amount of 2,4,6-tri-tert-butylphenol (10−200
Me
+
atoms. Figure S1 shows two Mn complexes in the [Mn (dpaq )]-
II
II
2Me
(OTf) structure.
Preparation of [Mn (OH)(dpaq2Me)](OTf). A 2.5 mM
Mn (dpaq )](OTf) solution (3.0 mg in 2 mL of MeCN) was
prepared under an inert atmosphere and transferred to a gastight
cuvette sealed with a pierceable septum. An excess of O gas was then
delivered to the solution by means of a syringe, and the formation of
Mn (OH)(dpaq )](OTf) was monitored by electronic absorption
spectroscopy (Figure S2). The formation was complete in ∼250 min,
and the resulting dark red solution was evaporated to dryness under
III
equiv) was dissolved in 100 μL of CH Cl and sealed in a vial with a
2 2
II
2Me
[
pierceable cap. The gastight syringe used to inject the phenol was first
purged with argon before use. Temperature was maintained using a
Unisoku cryostat at 50 °C, and the cuvette was allowed to equilibrate
to the temperature for 10 min before injection took place. Kinetics
measurements were made using an Agilent 8453 spectrophotometer.
The final reaction product was analyzed by X-band EPR spectroscopy
at 5 K, which showed a signal consistent with a phenoxyl radical
(Figure S8).
2
III
2Me
[
reduced pressure. The solid residue was then recrystallized using
III
III
2Me +
MeCN/Et O (3.2 mg; 98% yield). Synthesis of [Mn (OH)-
Kinetics Studies of [Mn (OH)(dpaq )] with Xanthene. For
III 2Me +
2
Me
(
dpaq2 )](OTf) on a larger scale was undertaken to obtain suitable
each kinetics experiment, a 1.25 mM [Mn (OH)(dpaq )] solution
was prepared in 1.8 mL of acetonitrile within an argon-filled glovebox,
placed in a gastight cuvette, and sealed with a septum. Xanthene (250
material for X-ray crystallographic analysis. In this procedure, O gas
was passed through an acetonitrile solution of [Mn (dpaq )](OTf)
2
II
2Me
III
(
(
20 mg in 5 mL), and the complete formation of [Mn (OH)-
equiv) was dissolved in 300 μL of CH Cl and sealed in a vial with a
2 2
dpaq2 )](OTf) was confirmed by ESI-MS and electronic absorption
Me
pierceable cap. The gastight syringe used to inject the phenol was first
purged with argon before use. Temperature was maintained using a
Unisoku cryostat at 50 °C, and the cuvette was allowed to equilibrate
to the temperature for 10 min before injection took place. Figure S3
shows data for a representative reaction.
spectroscopy. The resulting solution was evaporated to dryness under
reduced pressure, and the solid residue was recrystallized using slow
diffusion of ether into MeCN. Dark red [Mn (OH)(dpaq )](OTf)
crystals of X-ray crystallographic quality were obtained by repetitive
recrystallization, and were further characterized by ESI-MS (Figure
S9), X-band EPR spectroscopy (Figure S8), Evans NMR spectroscopy,
III
2Me
III
H
+
III
Cyclic Voltammetry of [Mn (OH)(dpaq )] and [Mn (OH)-
(dpaq2 )] . Cyclic voltammetry was performed on a 1 mM solution
Me +
III
and elemental analysis. μ = 4.90 (by Evan’s NMR method).
Elemental analysis [Mn (OH)(dpaq )](OTf): C H F MnN O S
of each Mn -hydroxo complex in acetonitrile with 600 μL of H
2
O
eff
III
2Me
added to the solution. Experiments were performed under nitrogen
atmosphere, with the solution being purged with nitrogen gas for 1 h
prior to measurements. A glassy carbon working electrode, a platinum
auxiliary electrode, and a AgCl/Ag reference electrode were utilized
along with a 0.1 M Bu N(PF ) as the supporting electrolyte. Data were
referenced to the Ep,c for ferrocenium/ferrocene (Fc /Fc) in MeCN.
Electronic Structure Calculations. All DFT calculations were
25
23
3
5
5
calcd. (%) C 48.63, H 3.75, N 11.34; found (%): C 46.70, H 3.74, N
0.74.
X-ray Diffraction Data Collection and Analysis for [Mn (OH)-
1
III
III
2
Me
(
(
dpaq2 )](OTf). A deep red single-domain crystal of [Mn (OH)-
4
6
Me
+
dpaq )](OTf) was suspended with mineral oil oil on a MiteGen
MicroMount and placed on a goniometer head in a cold nitrogen
stream at 100 K for a single-crystal X-ray structure determination. Data
were collected and analyzed using the same methods as described
above for [Mn (dpaq )](OTf). The space group and crystallo-
graphic data are summarized in Table S2. Hydrogen atoms were
located from a difference Fourier and initially included in the structural
model as independent isotropic atoms whose parameters were allowed
to vary in least-squares refinement cycles. In the final least-squares
refinement cycles, the acetonitrile methyl group was included in the
3
4
performed using the ORCA 3.0.1 software package.
The
III
+
III
2Me
+
[Mn (OH)(dpaq)] and [Mn (OH)(dpaq )] complexes were
II
2Me
32
18
optimized starting from the available crystal structures.3 Geometry
5,36
optimizations were performed using the BP86 functional
with the
Ahlrichs TZVP basis set with the TZVP/J auxiliary basis set for Mn, N,
and O and the smaller SVP basis set with the SVP/J auxiliary basis set
For DFT geometry optimizations involving xanthene,
the TZVP basis set was also used for the C involved in HAT. The
resolution of identity approximation was used for all optimizations.
Solvent was implicitly accounted for in all calculations using the
COSMO continuum solvent model with the default parameters for
acetonitrile as found in ORCA. Numerical frequency calculations
were performed at the same level of theory as the geometry
optimizations. The frequency calculations provided the zero point
and thermal corrections to the energy as well as confirmed ground-
state structures by showing no imaginary frequencies. The transition
state was identified first by scanning along the O−H bond distance
3
7,38
for C and H.
3
39
structural model as an idealized sp hybridized rigid rotor (with C−H
bond lengths of 0.98 Å) that was allowed to rotate freely about its C−
C bond, and the isotropic thermal parameter of H(11a) was fixed at a
value 1.2 times the equivalent isotropic thermal parameter of carbon
4
0
atom C(11). The triflate CF group is 88:12 rotationally disordered
3
about the C−S bond, and the three minor-occupancy fluorine sites
were incorporated in the structural model with isotropic thermal
parameters. All stages of weighted full-matrix least-squares refinement
were conducted using Fo2 data with the SHELXTL XLMP v2013/4
C
DOI: 10.1021/acs.inorgchem.6b01217
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 1. Selected Manganese−Ligand Bond Lengths (Å) from the Crystal Structures of [Mn (dpaq2Me)](OTf) (1-Me),
II
II
III
2Me
III
[
Mn (dpaq)](OTf) (1-H), [Mn (OH)(dpaq )](OTf) (2-Me), and [Mn (OH)(dpaq)](OTf) (2-H)
II
R
III
R
[
Mn (dpaq )](OTf) (1-R)
[Mn (OH)(dpaq )](OTf) (2-R)
a
b
a
b
bond
1-Me
1-H
bond
2-Me
2-H
Mn−O1A
Mn−N1B
Mn−N2B
Mn−N3B
Mn−N4B
Mn−N5B
2.116(2)
2.268(3)
2.172(3)
2.317(3)
2.275(3)
2.286(3)
2.079(2)
2.214(3)
2.191(3)
2.314(3)
2.244(3)
Mn−O2
Mn−N1
Mn−N2
Mn−N3
Mn−N4
Mn−N5
1.819(3)
2.186(3)
1.979(3)
2.203(3)
2.148(3)
2.158(3)
1.806(13)
2.072(14)
1.975(14)
2.173(14)
2.260(14)
2.216(15)
2.286(3)
b
a
See Figures 1 and S1 for the atom numbering scheme. Taken from ref 18.
III
2Me
III
2Me
III
Figure 1. (left) ORTEP diagram of Mn (OH)(dpaq )](OTf). (right) ORTEP overlay of [Mn (OH)(dpaq )] (bold) and [Mn (OH)-
(
H
+
dpaq )] . Hydrogen atoms, solvent, and noncoordinating triflate counterions were removed for clarity.
between the manganese-bound hydroxide oxygen and the hydrogen
bound to the oxygen of TEMPOH (or, in the case of the xanthene
reactions, the C involved in HAT), then taking the highest-energy
structure and using the ORCA saddle point optimization function
following the transition-state mode. The transition state was confirmed
by having only one imaginary frequency, which corresponded to the
O−H vibration. Single-point energy calculations were performed on all
oxygen group from an adjacent 1-Me cation, creating the same
polymeric structure previously observed in crystals of
II
H
18
[
Mn (dpaq )](OTf) (1-H). The most notable differences
between 1-Me and 1-H are Mn−N (Mn−N1) and Mn−
quinoline
O
(Mn−O2) bond elongations in the 1-Me structure by
amide
0
.054 and 0.037 Å, respectively (see Table 1 for a structural
41
structures using the TPSSh functional and the Ahlrichs TZVPP basis
set with the TZVPP/J auxiliary basis set on all atoms. The RIJCOSX
comparison). When 1-Me is dissolved in MeCN, the electronic
absorption spectrum shows a weak band at 500 nm (ε = 24
42,43
−1
−1
approximation,
as well as COSMO, was used in all single-point
M
cm ), which is presumably a charge-transfer band
energy calculations. Cartesian coordinates for all DFT-optimized
structures considered in this work are included in Supporting
Information (Tables S7−S20).
II
associated with the Mn complex. The perpendicular-mode
X-band EPR spectrum of a frozen solution of 1-Me at 5 K
shows a large feature centered at g = 2.04, attributed to the
The rotational and vibrational entropies were accounted for using
the default values from numerical frequency calculations performed in
ORCA. The translational entropy was calculated using a correction
model developed by Mammen et al. to account for the volume
II
mononuclear Mn center (Figure S8). Hyperfine splitting for
18
this feature was not resolved, as was also observed for 1-H.
Upon addition of excess O , the MeCN solution of 1-Me
changes from orange to a dark purple, with absorption bands
2
44
available to the solute. For the calculation of the free volume, the van
der Waals radii were used to calculate the molecular volume, the
experimental value for the concentration of MeCN was used, and the
cubic model in a three-dimensional (3D) cubic array was chosen. For
the final translational entropy calculation, the model excluding the
entropy of condensation was used. The enthalpy was taken as the
single-point electronic energy plus the zero-point energy. The
activation free energy was calculated using the total free energy of
the infinitely separated reactants and the free energy of the transition
state. The thermodynamic driving force was calculated using the total
free energy of the infinitely separated products and the total free
energy of the infinitely separated reactants.
−1
−1
appearing at 515 nm (ε = 228 M cm ) and 770 nm (ε = 100
−1
−1
M
cm ; see Figure S2). These new absorption features are
similar to those of 2-H (λ = 550 and 780 nm, with
−1
−1
corresponding ε values of 320 and 130 M cm ; see Scheme
1
1
). The reaction of 1-Me with O shows full formation after
2
50 min, ∼5 times slower than the reaction of O with 1-H.
2
The purple solution formed upon oxygenation of 1-Me is EPR
silent in both perpendicular and parallel modes at X-band
frequency (Figure S8). Thus, 1-Me has been completely
consumed, and an EPR silent-species is formed. Examination of
the purple solution by the NMR method of Evans reveals an
effective magnetic moment (μeff) of 4.90, consistent with a
RESULTS AND DISCUSSION
■
II
III
III
Structure and Properties of Mn and Mn -Hydroxo
mononuclear, S = 2, Mn center. Such species are often EPR-
II
2Me
Complexes. The crystal structure of [Mn (dpaq )](OTf)
silent even in parallel-mode X-band experiments due to a larger
II
III
45
(
1-Me) shows a six-coordinate Mn center bound by the five
zero-field splitting of the Mn ion.
nitrogen donors of the dpaq ligand (Figure S1). The
coordination sphere is completed by binding of the amide
Definitive assignment for the purple chromophore comes
from X-ray diffraction experiments on crystals grown by ether
D
DOI: 10.1021/acs.inorgchem.6b01217
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
diffusion into the MeCN solution. The X-ray structure shows a
III
III
2Me
mononuclear Mn -hydroxo complex, [Mn (OH)(dpaq )]-
III
(
OTf), where the Mn coordination sphere is completed by
the five nitrogen donors of the dpaq
2
Me
ligand and with the
hydroxo bound trans to the amide group (Figure 1). The
structure also contains free triflate, ether, and acetonitrile
2
Me
groups. The dpaq
ligand binds in a similar fashion to that
H
previously reported for dpaq , although with a large 0.11 Å
elongation in the bond length of the equatorial quinoline (Mn−
N1; Table 1). This is presumably caused by the steric bulk of
the methyl group adjacent to the nitrogen donor. Similar
elongations of ∼0.1 Å have been observed for Fe complexes
when comparing α-methyl-appended and unsubstituted pyr-
II
II
idines, although for the Fe complexes the introduction of the
III
Figure 2. Electronic absorption spectra of 1.25 mM [Mn (OH)-
α-methyl substituents often results in a change from a low-spin
(dpaq2 )] upon the addition of 10 equiv TEMPOH at −35 °C in
MeCN (initial and final spectra are the red and blue traces,
respectively). (inset) The decay of the 770 nm band over time.
Me
+
25−27
(S = 0) to a high-spin (S = 2) ground state.
In addition to
the Mn−N1 elongation in 2-Me, there is also a contraction of
the two pyridine Mn−N bond distances of 0.1 and 0.06 Å for
the Mn−N4 and Mn−N5 bonds, respectively (see Figure 1 for
the numbering scheme and Table 1). Notably, the Mn−OH
bond distances of 2-Me and 2-H are identical, within the
experimental uncertainty (Table 1). The 2-Me structure also
displays an extended hydrogen-bonding interaction between
the hydroxo ligand and the amide oxygen of a different 2-Me
molecule, with an oxygen−oxygen separation of 2.787 Å, which
X-band EPR spectrum collected for frozen MeCN solutions
following the reaction shows an intense feature at g = 2.03,
which corresponds to the TEMPO radical (Figure S8). On the
basis of these results, we conclude that 2-Me reacts with
TEMPOH to produce [Mn (dpaq )] and the TEMPO
radical.
A series of reactions of [Mn (OH)(dpaq )] with 10−40
equiv of TEMPOH (pseudo-first-order conditions) were
performed at −35 °C. The reactions provided a linear
correlation between TEMPOH concentration and the
pseudo-first-order rate constant (Figure 3). The slope of this
II
2Me
+
III
2Me
+
18
is ∼0.05 Å longer than that seen in the 2-H structure.
The reduction potential of 2-Me in MeCN was investigated
experimentally using cyclic voltammetry (CV). Reproducible
CV data were achieved for experiments run with 600 μL of
added water. (In the absence of added water, multiple
reduction processes were observed. Previous investigations of
III
IV
Mn -hydroxo and Fe -oxo complexes have established the
importance of added water in obtaining ideal CV behavior by
eliminating competing decay processes following reduc-
4
6−48
tion.
We note that, under these conditions, 2-H shows a
slightly higher reduction peak potential than that previously
1
8
reported. ) Under these conditions, reduction of both 2-Me
and 2-H are irreversible processes, with peak potentials Ep,c of
−
0.62 and −0.73 V, respectively (Figure S4, all values vs Fc/
+
Fc in MeCN). Thus, 2-Me shows an increase in peak potential
of +0.09 V, indicating a more facile reduction.
Oxidative Reactivity of [Mn (OH)(dpaq
III
2Me) +
] . To
evaluate the effect of these minor perturbations in structure
and reduction potential on reactivity, 2-Me was reacted with
various substrates with weak O−H and C−H bonds under an
inert atmosphere. We first investigated the reactivity of 2-Me
with TEMPOH. Because of the weak BDFE of TEMPOH
−
1
Figure 3. Pseudo-first-order rate constants kobs (s ) as a function of
(
BDFE = 66.5 kcal/mol, in MeCN at 25 °C), as well as its
TEMPOH and TEMPOD concentration for a 1.25 mM solution of 2-
Me. The second-order rate constant k (M s ) was calculated from
49
−1 −1
propensity to participate in CPET reactions, this substrate is
commonly used to investigate the reactivity of midvalent
oxidants.
2
the slope of the linear correlation.
1
7,18,50
Treatment of 2-Me with 10 equiv of TEMPOH
−1
at 25 °C in MeCN led to complete bleaching of the electronic
relation gives a second-order rate constant (k ) of 3.9(3) M
2
III
−1
absorption signals of the Mn -hydroxo within 8 s. This
s
at −35 °C. Thus, for the reaction with TEMPOH, 2-Me has
apparently rapid reaction was unanticipated, as the reaction of
a second-order rate constant, at −35°C, that is 30 times faster
18
−1
2-H with 100 equiv of TEMPOH is complete within 10 min.
than the same reaction with 2-H at 25 °C (k = 0.13(1) M
2
−1
To monitor the kinetics of TEMPOH oxidation by 2-Me, the
s ). The rate is also much larger than that observed for the
III
III
Mn -hydroxo was treated with 10 equiv of substrate in MeCN
corresponding Mn -methoxy complex of the dpaq ligand,
III
+
51
at −35 °C. In this case, the disappearance of 2-Me was still
rapid (95% complete within 1 min), but the change in
absorbance versus time showed first-order behavior past five
half-lives, permitting a kinetics analysis (Figure 2). The final
spectrum after reaction is that of 1-Me in MeCN, which is
formed in nearly 100% yield (Figure 2). A perpendicular-mode
[Mn (OMe)(dpaq)] (k = 0.080(1)).
2
Using TEMPOD, the H/D KIE for the reaction with 2-Me
was determined to be 2.7 (Figure 3). This KIE suggests that
O−H/D bond cleavage is involved in the rate-determining step
for the reaction. We note that the intercepts for plots of k
obs
versus substrate concentration for both TEMPOH and
E
DOI: 10.1021/acs.inorgchem.6b01217
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
III
Table 2. Reaction Parameters for the Oxidation of TEMPOH, Xanthene, and 2,4,6-Tri-tert-butylphenol by [Mn (OR)
R
+
(
dpaq )] Complexes
TEMPOH
xanthene
2,4,6-tri-t-butylphenol
1
a
ΔH‡
b
ΔS‡
c
ΔG‡
b
kobs (s−1
d
k2 (M−1 s−1
e
complex
k (M− s−1
)
)
)
2
III
H
+
[
[
[
Mn (OH)(dpaq )]
0.13(1)
9.9(9)
11.4(5)
5.7(3)
−35(3)
−27(2)
−41(1)
20.3
19.4
17.9
0.0008
g
f
III
H
+
Mn (OMe)(dpaq )]
Mn (OH)(dpaq2Me)]+
0.080(1)
0.0052(1)
0.044(2)
III
h
31
0.000 25
a
b
c
d
Values at 25 °C. Units of kilocalories per mole. Units of calories per mole per Kelvin. Pseudo-first-order rate constant using 250 equiv of
xanthene at 50 °C. Values at 50 °C. The reaction of [Mn (OH)(dpaq ) with 2,4,6-tri-tert-butylphenol displays saturation kinetics; thus, the
reaction cannot be described by a second-order rate constant. Treatment of [Mn (OMe)(dpaq )] with xanthene led to a decay rate of the same
order as the self-decay rate of this complex; see ref 51. The second-order rate constant at 25 °C was determined using the activation parameters and
e
f
III
H
g
III
H
+
h
the corresponding rate at constant at −35 °C.
TEMPOD are slightly negative; however, the intercepts are
similar to each other, and both are indistinguishable from zero
within the experimental uncertainty.
dominates the observed reaction rate, demonstrating that the
increase in the strain of the transition state is overcome by the
more favorable electronics of 2-Me. TEMPOH is also not a
very bulky substrate and is likely to have little direct interaction
with the methyl group.
To determine the activation parameters for the reaction of
TEMPOH with 2-Me, an Eyring analysis was performed by
monitoring TEMPOH oxidation reactions from −35 to 15 °C.
III
To date, only two synthetic Mn -hydroxo species are known
‡
‡
−1
III
This provided ΔH and ΔS of 5.7(3) kcal mol and 41(1) cal
to abstract hydrogen atoms from C−H bonds. The [Mn (OH)
−1
−1
2+
mol K , respectively (Table 2 and Figure 4). This gives, at
(PY5)] complex of Goldsmith and Stack can attack the
relatively strong C−H bond of toluene (BDFE = 87 kcal/mol
16
in MeCN), while 2-H attacks the weaker C−H bond of
xanthene (BDFE = 73.3 kcal/mol in dimethyl sulfoxide), albeit
very slowly (with 250 equiv of xanthene, k = 8 × 10−
5
4
s
−1
at
obs
18
0 °C in MeCN). When 2-Me was treated with 250 equiv of
xanthene at 50 °C, the reaction showed a slow decay of the
III
electronic absorption signal from the Mn -hydroxo species
(
Figure S3). After 16 h, ∼85% of 2-Me had been consumed. A
−4 −1
pseudo-first-order rate constant of 2.5 × 10
s
was
determined, which is a factor of 3 smaller than that observed
for the reaction of 2-H with xanthene. Thus, the reactivity of 2-
Me with xanthene is comparable to that of 2-H, which is
unanticipated given the large rate differences with TEMPOH.
Because of the seemingly contradictory reactivity of 2-Me
toward TEMPOH and xanthene, we investigated the reactivity
of this complex using a substrate with a more sterically
hindered O−H bond. (We also explored the reactivity of 2-Me
with 1,4-cyclohexadiene, which has a less hindered C−H bond.
However, the kinetics profile for this reaction was complex.
Instead of the expected formation of 1-Me, this reaction yielded
a colored product that has eluded further characterization.) 2-
Me was reacted with 10 to 200 equiv of 2,4,6-tri-tert-
butylphenol at 50 °C under an argon atmosphere. For each
reaction, the electronic absorption features associated with the
−
1
Figure 4. Plot of ln(k/T) as a function of 1/T (K ) for the reaction of
.25 mM 2-Me with TEMPOH from −35 to 15 °C (blue ●).
Corresponding data for the reaction of 1.25 mM 2-H with TEMPOH
red ●) are included for comparison.
1
(
−1
2
5 °C, an activation free energy of 17.9 kcal mol , with the
entropic contribution being 12.2 kcal mol . Utilizing the
activation parameters, we can estimate a second-order rate
constant (k ) for 2-Me at 25 °C of 31 M s . Thus, the
reaction of 2-Me with TEMPOH shows a remarkable increase
−1
III
Mn -hydroxo decayed, while the characteristic 2,4,6-tri-tert-
butylphenoxyl radical feature (λ = 628 nm; ε = 400(10) M−
1
−
1
−1
max
2
−1
52
s ) grew in simultaneously (Figure 5). Perpendicular-mode
X-band EPR data collected following the reaction showed a
single derivative feature characteristic of the phenoxyl radical
(Figure S8). In addition, the absorbance intensity of the 2,4,6-
tri-tert-butylphenoxyl radical following the reaction indicates
90% formation of this radical relative to 2-Me. The second-
order rate constant for this reaction was determined to be
in the rate constant of ∼240-fold compared to that of 2-H (k =
2
1
−1
0
.13(1) M−
s
at 25 °C). The nature of the activation barrier
has also changed significantly with the addition of the methyl
group. A comparison of the kinetics parameters for TEMPOH
III
oxidation for 2-Me, 2-H, and the Mn -methoxy complex
III
+
[
Mn (OMe)(dpaq)] is provided in Table 2. Specifically, the
4.4(2) × 10−
2
M
−1 −1
s
from the linear correlation of the kobs
activation enthalpy has decreased by nearly a factor of 2, from
1
9
.9(9) kcal/mol for 2-H to 5.7(6) kcal mol− for 2-Me, while
and the substrate concentration (Figure 6). The H/D KIE for
this reaction was found to be 2.1, which is similar to that found
for the TEMPOH/TEMPOD reaction and suggests that the
reaction is undergoing a CPET mechanism for both substrates.
Phenol oxidation by 2-Me also shows a dramatic shift in the
kinetics model compared to the reaction of 2-H with phenols.
With that complex, the reaction was shown to involve
the entropic contribution, at 25 at 25 °C, is slightly increased
10 vs 12 kcal mol for 2-H and 2-Me, respectively). This
increase in the entropic contribution to the free energy of
activation could be accounted for by an increase in steric strain
brought about by the proximity of the 2-methyl substituent to
−1
(
‡
the reaction center. However, the decrease in the ΔH
F
DOI: 10.1021/acs.inorgchem.6b01217
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
similar, or even slower, reaction rates with larger, or rigid,
53
substrates (i.e., phenols and xanthene).
Electronic Structure Computations. Electronic structure
methods were used to analyze the basis for this intriguing
reactivity. DFT-optimized structures for both 2-H and the 2-
Me are in good agreement of the crystal structures (Tables 1
and S5). The free energy profiles for the reaction of each
complex with TEMPOH are shown in Figure 7. There is
Figure 5. Electronic absorption spectra of 1.25 mM 2-Me with the
addition of 100 equiv tri-tert-butylphenol at 50 °C in MeCN. (inset)
The decay of 2-Me and growth of the phenoxyl radical. The tri-tert-
butylphenoxyl radical product, which absorbs at 630 nm, reaches
maximum concentration at ∼200 s, and then slowly decays, as seen in
the decrease of the absorbance signals after this time point.
Figure 7. DFT-calculated reaction coordinate for 2-H (red) and 2-Me
(blue) with TEMPOH. Experimental values are in parentheses. All
values are in kilocalories per mole.
excellent agreement between the calculated free energy
18
activation barriers and the experimentally determined values.
This agreement, along with the smaller KIE values found
experimentally, suggests that tunneling does not play a large roll
in this CPET process. The reaction coordinate takes place
along the S = 2 surface, with the resulting TEMPO radical (S =
II
1
/2) and Mn product (S = 5/2) anti-ferromagnetically
coupled. The structures at the transition state are shown in
Figure 8A. These illustrate that the hydrogen atom is only
partially transferred between the substrate and the Mn-hydroxo,
with an O−H bond distance of 1.14 Å for 2-H and 1.15 Å for 2-
Me. The transition-state structure with 2-Me shows that the
TEMPOH is able to orient itself without interacting with the
methyl-quinoline group (Figure 8A).
−
1
Figure 6. Pseudo-first-order rate constant kobs (s ) as a function of
,4,6-tri-tert-butylphenol concentration for a 1.25 mM solution of 2-
Me (red ●) and corresponding data for deuterated phenol (blue ⧫).
2
−1 −1
The second-order rate constant k (M s ) was calculated from the
2
slope of the linear correlation.
For the CPET reaction, the electron-accepting orbital of both
2-H and 2-Me is the α-spin lowest unoccupied molecular
saturation kinetics, giving a nonlinear rate dependence on
2
substrate concentration (i.e., the k values leveled at high
orbital (LUMO), which consists of the Mn 3d orbital that is
obs
z
1
8
substrate concentration). This was attributed to the
stabilization of a hydrogen-bonded precursor complex. This
change in kinetics model for phenol oxidation by 2-Me could
arise from a destabilization of a hydrogen-bonded structure due
to interactions between the bulky tert-butyl groups on the
substrate and the methyl-functionalized quinoline of 2-Me.
This difference in phenol reactivity for 2-H and 2-Me
complicates a direct comparison of reaction rates, because the
strongly σ-antibonding with the hydroxide 2p orbital (Figure
z
8B). In the transition state, this MO is strongly mixing with the
nitrogen and oxygen orbitals of TEMPOH (Figure 8C), which
forms a superexchange pathway mediating electron to transfer
from substrate to the metal center. A previous computational
investigation of iron lipoxygenase also identified strongly
interacting substrate and Fe-hydroxo orbitals in the transition
state, with little or no localization of electron density on the
54
2
-H reaction cannot be properly described by a second-order
transferring proton. However, in that case the orbital
accepting the electron was a β-spin, Fe−OH π-antibonding
MO (i.e., a t orbital, using O symmetry labels). For the S = 2
III
rate constant. However, previous work with the Mn -methoxy
analogue of 2-H, [Mn (OMe)(dpaq)] , showed the standard
III
+
2g
h
51
second-order kinetics observed here for 2-Me. The second-
manganese(III)-hydroxo units considered here, the electron
order rate constant for that reaction was 5.2(1) × 10−
3
M
−1 −1
s ,
must be added to the α-spin 3d orbital (i.e., an e orbital),
2
z
g
54
which is slightly slower than that observed for 2-Me (4.4(2) ×
which is the LUMO for both 2-Me and 2-H. In the transition
state, the LUMO shows a similar bonding pattern to the
highest occupied molecular orbital (HOMO) but with a much
larger percent contribution from the TEMPO nitrogen and
oxygen (Figure 8C). The composition of the HOMO and
LUMO point to a strongly interacting system, with most of the
−2
−1 −1
10
M
s ). Thus, while 2-Me reacts with TEMPOH ∼390-
III +
fold faster than [Mn (OMe)(dpaq)] (Table 2), these two
complexes show more comparable reaction rates toward the
bulkier substrate 2,4,6-tri-tert-butylphenol. These results show
an increased reactivity of 2-Me with smaller substrates but
G
DOI: 10.1021/acs.inorgchem.6b01217
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
approach, maximizing overlap between the donor and acceptor
orbitals, would give an angle between the manganese, oxygen,
and transferring hydrogen (Mn−O−HTEMPOH) of nearly 180°.
However, the Mn−O−H
angle is ∼145° in both
TEMPOH
transition-state structures. This angle represents a compromise
between the electronic ideal of 180° and the steric limitations
created by the hydroxyl proton, which render a 180° angle
unfeasible. This places a severe steric constraint upon substrate
III
oxidation by a Mn -hydroxo species, as exemplified exper-
imentally by the reactivity of 2-Me with bulkier substrates.
The products of the DFT reaction coordinate calculations
II
R
+
are the manganese(II)−aqua complex [Mn (OH ) (dpaq )]
2
and the TEMPO radical. The driving force for the 2-Me
reaction is calculated to be almost 4 kcal mol− larger than that
of 2-H (Figure 7). This driving force represents the difference
in O−H BDFE between the manganese(II)−aqua complexes
and TEMPOH. CPET reactions, whether for metal-hydroxo or
1
-
oxo species, commonly show correlation between the
thermodynamic driving force and the activation barrier, with
55
more exergonic reactions proceeding with lower barriers. This
relationship holds true for the reaction of TEMPOH with 2-Me
and 2-H, at least in terms of the DFT results (Figure 7),
suggesting that the rate enhancement of 2-Me can be attributed
II
2Me
+
+
to an increase in the BDFE of the [Mn (OH )(dpaq )]
2
II
2Me
complex. Experimentally, the BDFE of [Mn (OH )(dpaq )]
2
III
II
could be calculated from the Mn (OH)/Mn (OH) reduction
potential and the pK for the manganese(III)−aqua complex
a
56
using the Bordwell relationship. Thus, a complete picture of
the difference in driving force for 2-H and 2-Me requires
knowledge of the reduction potentials and pKa values.
Unfortunately, the pK values of 2-H and 2-Me, and their
a
II
Mn -aqua analogues, are currently unobtainable experimentally.
None of the complexes are soluble in water at high enough
concentrations to perform CV experiments at different pH, and
III
the Mn -hydroxo species show rapid, and complex, decay upon
the addition of acid. In the absence of experimental pK values,
a
we estimated the shift in pK for 2-H and 2-Me using electronic
a
structure computations following a procedure developed by
Hammes-Schiffer (further details are available in Supporting
57
Information). To account for known deficiencies in the
treatment of the free energy of the electron, the potential of the
electrode in the reduction calculation, as well as in the free
III
energy of the proton in the pK calculation, the [Mn (OH)
(
a
PY5)]2+ complex was utilized as a reference system, as this
16
complex has a well-defined reduction potential and pKa. As a
test of the appropriateness of this method, we first calculated
E1 values, which were found to be −0.83 and −0.60 V (vs Fc/
/2
Figure 8. (A) Structural comparison of 2-H (left) and 2-Me (right)
transition states. (B) Qualitative MO diagram of 2-Me and TEMPOH
reactants showing the TEMPOH HOMO and manganese 3d
+
Fc ) for 2-H and 2-Me, respectively. These values agree quite
well with the experimental Ep,c values (−0.73 and −0.62 V).
III
The calculated pK values were quite similar, 21 and 20 for 2-H
a
manifold. Surface contour plots of the 2-Me α-spin LUMO (Mn
and 2-Me, respectively. With these E1/2 and pK parameters, the
2
a
d MO) and the α-spin TEMPOH HOMO are shown on the left- and
z
BDFEs for 2-H and 2-Me were calculated to be 64 and 68 kcal
right-hand sides, respectively. (C) Corresponding transition-state MO
diagram for 2-Me and TEMPOH, with surface countour plots of the
α-spin HOMO (left) and LUMO (right). The transition-state MO
plots depict the strong interaction between Mn-hydroxo and substrate
orbitals.
−1
mol , respectively, consistent with the 4 kcal/mol difference in
driving force from the reaction coordinate calculations (Figure
7). Importantly, these computations demonstrate that the
increased driving force for the reaction of 2-Me with TEMPOH
III
is due to a more favorable reduction of the Mn −OH complex,
character of the transferred electron residing on the manganese
center. The Loewdin spin density on the manganese center in
the transition state of both 2-Me and 2-H is also indicative of a
a conclusion that is reinforced by the experimental CV data.
If 2-Me has a stronger intrinsic driving force for CPET
reactions, why are the reaction rates with xanthene and tri-tert-
butylphenol more comparable to those of 2-H? To address this
question, we performed reaction coordinate calculations to
identify transition states for hydrogen-atom transfer from
II
Mn center.
III
The requirement for the electron to transfer into the Mn
2
3
d −OH 2p σ-antibonding MO means that an ideal angle of
z
z
H
DOI: 10.1021/acs.inorgchem.6b01217
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 9. Transition states for hydrogen-atom transfer from xanthene to 2-Me (A) and 2-H (B). (C) An overlay plot, where the two transition-state
structures were superimposed.
xanthene to 2-Me and 2-H. These calculations indicate a lower
reactivity of 2-Me is intrinsically favored over that of 2-H
because of a larger thermodynamic driving force, we propose
that steric restrictions can dampen the reactivity considerably.
‡
enthalpy of activation for 2-H compared to 2-Me (ΔH = 22.7
−1
and 25.6 kcal mol , respectively), with comparable entropies
‡
of activation at 50 °C (ΔS = −11.96 and −10.94, respectively).
Consequently, 2-H has a lower free-energy barrier for
hydrogen-atom transfer with xanthene (Table S6) and is
predicted to react with xanthene faster than 2-Me. This is
CONCLUSIONS
■
There are few synthetic hydroxomanganese(III) complexes
capable of oxidizing substrates via PCET, and, of those
reported, a wide range of ligand systems are used with large
consistent with the experimental rates (k of 8.0 × 10− and
4
obs
4
−1
2
.5 × 10−
s
for 2-H and 2-Me, respectively; see Table 2).
16−18
The transition-state structures for both systems are shown in
Figure 9. These structures indicate partial hydrogen-atom
transfer, with O−H bond distances of 1.13 and 1.12 Å for 2-Me
and 2-H respectively. The C−H bond in xanthene is also
significantly elongated in each transition state (Figure 9).
differences in oxidative ability.
These differences in the
ligand systems make it difficult to identify factors responsible
for changes in reactivity. To better elucidate the nature of these
reactions, small perturbations to the ligand systems can prove
very useful. The work in this paper is the first to explore how
such a change can affect the CPET reactivity of a
hydroxomanganese(III) species. A small change to the second
coordination sphere, by incorporating a 2-methyl-quinoline
group in the (dpaq) ligand, leads to a dramatic increase in the
rate of reaction with TEMPOH, which is related to an
increased thermodynamic driving force reflected in a lower
reduction potential. With the substrates tri-tert-butylphenol and
xanthene, however, there are modest changes in reaction rate,
likely due to steric clash caused by the combined effect of both
the 2-methyl substituent and the hydroxo proton. Such
interactions could lead to steric-controlled oxidation of
substrates. Further investigation is needed to explore the
III
Although the structural parameters within the Mn -hydroxo-
xanthene unit are comparable, the orientation of xanthene
III
relative to the Mn complex is different, as evident in the
overlay plot of the two transition states in Figure 9C. In the
transition state of 2-H, one ring of xanthene is located near the
quinoline ligand. Superimposing this xanthene location on the
transition state of 2-Me results in close contact between
xanthene carbons atoms and hydrogen atoms of the 2-methyl-
substituent (C···H contacts of 2.08 and 2.25 Å). These contacts
are illustrated in a space-filling model in Supporting
Information (Figure S6). These DFT-derived structures
provide strong evidence that xanthene must be oriented
differently in the 2-Me transition state to avoid the bulky
quinoline moiety. We presume that this reorientation has a
destabilizing effect, leading to the higher-energy transition state
of 2-Me as compared to 2-H.
III
oxidative limits of 2-Me and related Mn -hydroxo complexes.
Unfortunately, we were unable to further corrobarate this
model for the tri-tert-butylphenol reaction, as we could not
locate transition states for the reaction of this substrate with
ASSOCIATED CONTENT
■
*
S
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.6b01217.
either 2-H or 2-Me. Surface scans of the Mn−O···H
phenol
coordinate are endothermic and proceed uniformly uphill to
products. Nonetheless, stationary points along the Mn−O···
H
phenol
coordinate show the steric restrictions of this reaction.
Crystallographic information, crystal and DFT structures,
electronic absorption spectra, cyclic voltammogram,
description of method for calculating theoretical BDFE,
selected bond lengths, NMR spectrum, mass spectrum,
Cartesian coordinates for DFT-optimized structures
III
The phenol approaches the Mn -hydroxo complex along a
vector perpendicular to the quinoline moiety, similar to that
observed in the transition state with TEMPOH (cf. Figures S7
and 8A). This approach trajectory places one t-butyl arm in
close proximity with the 2-methyl substituent. An alternative
approach of phenol parallel with the quinoline moiety would
place the hydroxo proton in close proximity to the 2-methyl
substituent, also creating steric clash. Thus, although the CPET
(PDF)
X-ray crystallographic information (CIF)
X-ray crystallographic information (CIF)
I
DOI: 10.1021/acs.inorgchem.6b01217
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(
27) Kim, J.; Zang, Y.; Costas, M.; Harrison, G. R.; Wilkinson, C. E.;
AUTHOR INFORMATION
Corresponding Author
Phone: (785) 864-3968. Fax: (785) 864-5396. E-mail: taj@ku.
■
Que, L., Jr JBIC, J. Biol. Inorg. Chem. 2001, 6, 275.
28) Mader, E. A.; Davidson, E. R.; Mayer, J. M. J. Am. Chem. Soc.
007, 129, 5153.
29) Riedel, P. J.; Arulsamy, N.; Mehn, M. P. Inorg. Chem. Commun.
(
2
(
*
edu.
Notes
2011, 14, 734.
The authors declare no competing financial interest.
(30) Data Collection: SMART Software in APEX2 v2010.3−0 Suite;
Bruker: Billerica, MA, 2016.
ACKNOWLEDGMENTS
(31) Data Reduction: SAINT Software in APEX2 v2010.3−0 Suite;
Bruker: Billerica, MA, 2016.
■
This work was supported by the U.S. National Science
Foundation (CHE-1056470 to T.A.J). The U.S. NSF is also
acknowledged for funds used for the purchase of X-ray
instruments (CHE-0079282) and the EPR spectrometer
(
32) International Tables for Crystallography; International Union of
Crystallography, 1996; Vol. A.
33) Refinement: SHELXTL v2010.3−0; University of Gottingen:
Germany, 2013.
34) Neese, F. Wiley Interdiscip. Rev. 2012, 2, 73.
(
(
CHE-0946883). Support for the NMR instrumentation was
(
provided by NIH Shared Instrumentation Grant No.
S10OD016360, NSF Major Research Instrumentation Grant
No. 9977422, and NIH Center Grant No. P20 GM103418.
(35) Becke, A. D. J. Chem. Phys. 1986, 84, 4524.
(36) Perdew, J. P. Phys. Rev. B: Condens. Matter Mater. Phys. 1986, 33,
8822.
(
37) Schaf
̈
er, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97, 2571.
er, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100,
(38) Schaf
̈
REFERENCES
1) Su, C.; Oliw, E. H. J. Biol. Chem. 1998, 273, 13072.
2) Su, C.; Sahlin, M.; Oliw, E. H. J. Biol. Chem. 2000, 275, 18830.
3) Wennman, A.; Karkehabadi, S.; Oliw, E. H. Arch. Biochem.
■
5
829.
(
(
(
(39) Neese, F. J. Comput. Chem. 2003, 24, 1740.
(40) Sinnecker, S.; Rajendran, A.; Klamt, A.; Diedenhofen, M.;
Neese, F. J. Phys. Chem. A 2006, 110, 2235.
41) Staroverov, V. N.; Scuseria, G. E.; Tao, J.; Perdew, J. P. J. Chem.
Phys. 2003, 119, 12129.
Biophys. 2014, 555−556, 9.
4) Gaffney, B. J.; Su, C.; Oliw, E. H. Appl. Magn. Reson. 2001, 21,
11.
5) Wennman, A.; Oliw, E. H.; Karkehabadi, S.; Chen, Y. J. Biol.
Chem. 2016, 291, 8130.
6) Gilchrist, M. L.; Ball, J. A.; Randall, D. W.; Britt, R. D. Proc. Natl.
Acad. Sci. U. S. A. 1995, 92, 9545.
7) Hoganson, C. W.; Lydakis-Simantiris, N.; Tang, X.-S.; Tommos,
(
(
4
(
(
42) Izsak
́
, R.; Neese, F. J. Chem. Phys. 2011, 135, 144105.
(43) Neese, F.; Wennmohs, F.; Hansen, A.; Becker, U. Chem. Phys.
2
(
009, 356, 98.
44) Mammen, M.; Shakhnovich, E. I.; Deutch, J. M.; Whitesides, G.
M. J. Org. Chem. 1998, 63, 3821.
45) Telser, J.; Krzystek, J.; Ozarowski, A. JBIC, J. Biol. Inorg. Chem.
014, 19, 297.
46) Ghachtouli, S. E.; Guillot, R.; Aukauloo, A.; Dorlet, P.;
Anxolabehere-Mallart, E.; Costentin, C. Inorg. Chem. 2012, 51, 3603.
47) El Ghachtouli, S.; Lassalle-Kaiser, B.; Dorlet, P.; Guillot, R.;
(
(
(
2
(
C.; Warncke, K.; Babcock, G. T.; Diner, B. A.; McCracken, J.; Styring,
S. Photosynth. Res. 1995, 46, 177.
(
(
(
8) Borovik, A. S. Chem. Soc. Rev. 2011, 40, 1870.
9) Lansky, D. E.; Goldberg, D. P. Inorg. Chem. 2006, 45, 5119.
10) Leto, D. F.; Ingram, R.; Day, V. W.; Jackson, T. A. Chem.
́
̀
(
Anxolabehere-Mallart, E.; Costentin, C.; Aukauloo, A. Energy Environ.
Sci. 2011, 4, 2041.
Commun. 2013, 49, 5378.
(
11) Chen, J.; Lee, Y.-M.; Davis, K. M.; Wu, X.; Seo, M. S.; Cho, K.-
(48) Wang, D.; Ray, K.; Collins, M. J.; Farquhar, E. R.; Frisch, J. R.;
B.; Yoon, H.; Park, Y. J.; Fukuzumi, S.; Pushkar, Y. N.; Nam, W. J. Am.
Chem. Soc. 2013, 135, 6388.
Gomez, L.; Jackson, T. A.; Kerscher, M.; Waleska, A.; Comba, P.;
Costas, M.; Que, L. Chemical Science 2013, 4, 282.
(
12) Leeladee, P.; Baglia, R. A.; Prokop, K. A.; Latifi, R.; de Visser, S.
P.; Goldberg, D. P. J. Am. Chem. Soc. 2012, 134, 10397.
13) Baglia, R. A.; Durr, M.; Ivanovic-Burmazovic, I.; Goldberg, D. P.
Inorg. Chem. 2014, 53, 5893.
14) Parsell, T. H.; Yang, M.-Y.; Borovik, A. S. J. Am. Chem. Soc.
009, 131, 2762.
15) Wu, X.; Seo, M. S.; Davis, K. M.; Lee, Y.-M.; Chen, J.; Cho, K.-
B.; Pushkar, Y. N.; Nam, W. J. Am. Chem. Soc. 2011, 133, 20088.
16) Goldsmith, C. R.; Cole, A. P.; Stack, T. D. P. J. Am. Chem. Soc.
005, 127, 9904.
17) Coggins, M. K.; Brines, L. M.; Kovacs, J. A. Inorg. Chem. 2013,
2, 12383.
18) Wijeratne, G. B.; Corzine, B.; Day, V. W.; Jackson, T. A. Inorg.
Chem. 2014, 53, 7622.
19) Sankaralingam, M.; Jeon, S. H.; Lee, Y.-M.; Seo, M. S.; Ohkubo,
K.; Fukuzumi, S.; Nam, W. Dalton Transactions 2016, 45, 376.
20) Hitomi, Y.; Arakawa, K.; Funabiki, T.; Kodera, M. Angew. Chem.,
Int. Ed. 2012, 51, 3448.
21) Hitomi, Y.; Arakawa, K.; Kodera, M. Chem. - Eur. J. 2013, 19,
4697.
22) Hitomi, Y.; Arakawa, K.; Kodera, M. Chem. Commun. 2014, 50,
485.
23) Hitomi, Y.; Iwamoto, Y.; Kodera, M. Dalton Transactions 2014,
3, 2161.
24) Iwamoto, Y.; Kodera, M.; Hitomi, Y. Chem. Commun. 2015, 51,
539.
25) Zang, Y.; Kim, J.; Dong, Y.; Wilkinson, E. C.; Appelman, E. H.;
Que, L. J. Am. Chem. Soc. 1997, 119, 4197.
26) Chiou, Y.-M.; Que, L. J. Am. Chem. Soc. 1995, 117, 3999.
(49) Warren, J. J.; Tronic, T. A.; Mayer, J. M. Chem. Rev. 2010, 110,
6
961.
(
̈
́
́
(50) Porter, T. R.; Mayer, J. M. Chemical Science 2014, 5, 372.
(51) Wijeratne, G. B.; Day, V. W.; Jackson, T. A. Dalton Transactions
(
2
(
2
(
015, 44, 3295.
52) Manner, V. W.; Markle, T. F.; Freudenthal, J. H.; Roth, J. P.;
Mayer, J. M. Chem. Commun. 2008, 256.
53) We attempted to assess the hydrogen-atom transfer kinetics of
-H and 2-Me using the less bulky 2,4-di-tert-butylphenol as a
(
2
(
2
(
5
substate. However, treatment of either complex with this substrate in
MeCN at 50 °C results in immediate changes in the electronic
absorption spectra, showing the formation of a new chromophore
(
(
Supporting Information, Figure S10). This chromophore then rapidly
decays. We speculate that the less bulky 2,4-di-tert-butylphenol is able
(
to rapidly displace the hydroxide ligand and coordinate directly to the
III
III
(
Mn center. The newly formed Mn -phenolate species would be
expected to be colored, accounting for the new chromophore.
III
(
1
(
7
(
4
(
9
Presumably the Mn -phenolate species is unstable and undergoes
rapid decay. This proposed chemistry is consistent with an ESI-MS
II
analysis of the resulting reaction mixture, which shows Mn products
III
+
and a peak attributed to [Mn (phenolate)(dpaq)] (Figure S11).
III
Thus, the reactions of the Mn -hydroxo species with 2,4-di-tert-
butylphenol are not well-behaved, but they are fundamentally different
from the reaction with 2,4,6-tri-tert-butylphenol.
(54) Lehnert, N.; Solomon, E. I. J. Biol. Inorg. Chem. 2003, 8, 294.
(55) Mayer, J. M. Acc. Chem. Res. 2011, 44, 36.
(56) Bordwell, F. G.; Cheng, J.; Ji, G. Z.; Satish, A. V.; Zhang, X. J.
(
(
Am. Chem. Soc. 1991, 113, 9790.
J
DOI: 10.1021/acs.inorgchem.6b01217
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(
57) Solis, B. H.; Hammes-Schiffer, S. Inorg. Chem. 2014, 53, 6427.
K
DOI: 10.1021/acs.inorgchem.6b01217
Inorg. Chem. XXXX, XXX, XXX−XXX