General synthesis of Di-μ-oxo dimanganese complexes as functional models for the oxygen evolving complex of photosystem II

Article abstract of DOI:10.1021/ic0509940

A series of complexes with the formula [Mn^III/IV ₂(μ-O)₂(L)₂(X)₂]³⁺ have been prepared in situ from Mn^IILCl₂ precursors by a general preparative method (L = terpy, Cl-terpy, Br-terpy, Ph-terpy, tolyl-terpy, mesityl-terpy, tBu₃-terpy, EtO-terpy, py-phen, dpya, Me₂N-terpy, or HO-terpy, and X = a labile ligand such as water, chloride, or sulfate). The parent complex, where L = terpy and X = water, is a functional model for the oxygen-evolving complex of photosystem II (Limburg, et al. J. Am. Chem. Soc. 2001, 123, 423-430). Crystals of Mn^II(dpya) Cl₂, Mn^II(Ph-terpy)-Cl₂, Mn ^II(mesityl-terpy)Cl₂, and an organic-soluble di-μ-oxo di-aqua dimanganese complex, [Mn^III/IV₂(μ-O) ₂(mesityl-terpy)₂(OH₂)₂](NO ₃)₃, were obtained and characterized by X-ray crystallography. Solutions of the in situ-formed di-μ-oxo dimanganese complexes were characterized by electrospray mass spectrometry, EPR spectroscopy, and UV-visible spectroscopy, and the rates of catalytic oxygen-evolving activity were assayed. The use of Mn^IILCl₂ precursors leads to higher product purity of the Mn dimers while achieving the 1:1 ligand to Mn stoichiometry appropriate for catalytic activity assay. These methods can be used to screen the catalytic activity of other di-μ-oxo dimanganese complexes generated by using a ligand library.

Full text of DOI:10.1021/ic0509940

Inorg. Chem. 2005, 44, 7661−7670
General Synthesis of Di-µ-oxo Dimanganese Complexes as Functional
Models for the Oxygen Evolving Complex of Photosystem II
Hongyu Chen, Ranitendranath Tagore, Siddhartha Das, Christopher Incarvito, J. W. Faller,
Robert H. Crabtree,* and Gary W. Brudvig*
Department of Chemistry, Yale UniVersity, P.O. Box 208107,
New HaVen, Connecticut 06520-8107
Received June 17, 2005
/
IV
A series of complexes with the formula [MnIII
by a general preparative method (L terpy, Cl-terpy, Br-terpy, Ph-terpy, tolyl-terpy, mesityl-terpy, t Bu3-terpy,
EtO-terpy, py-phen, dpya, Me2N-terpy, or HO-terpy, and X a labile ligand such as water, chloride, or sulfate).
The parent complex, where L terpy and X water, is a functional model for the oxygen-evolving complex of
photosystem II (Limburg, et al. J. Am. Chem. Soc. 2001, 123, 423
430). Crystals of MnII(dpya)Cl2, MnII(Ph-terpy)-
Cl2, MnII(mesityl-terpy)Cl2, and an organic-soluble di- -oxo di-aqua dimanganese complex, [MnIII
terpy)2(OH2)2](NO3)3, were obtained and characterized by X-ray crystallography. Solutions of the in situ-formed
di- -oxo dimanganese complexes were characterized by electrospray mass spectrometry, EPR spectroscopy, and
UV
visible spectroscopy, and the rates of catalytic oxygen-evolving activity were assayed. The use of MnIILCl2
precursors leads to higher product purity of the Mn dimers while achieving the 1:1 ligand to Mn stoichiometry
(
2 µ-O)2(L)2(X)2]3+ have been prepared in situ from MnIILCl2 precursors
)
)
)
)
−
/
IV
µ
2(µ-O)2(mesityl-
µ
−
appropriate for catalytic activity assay. These methods can be used to screen the catalytic activity of other di-
dimanganese complexes generated by using a ligand library.
µ-oxo
Introduction
OEC, only a handful of di-µ-oxo manganese complexes have
been synthesized that contain coordinated water.5-10 Previ-
ously, we reported the O2-evolving reactions catalyzed by
[(H2O)LMnIII(µ-O)2MnIVL(OH2)]3+ (L ) dipic, 1a, and
terpy, 1b).7,11,12 These complexes are functional models for
the OEC with a biologically relevant Mn(µ-O)2Mn core. The
meridionally coordinating tridentate ligands allow water
Bioinorganic model chemistry has played an important role
in the understanding of the oxygen-evolving complex (OEC)1
of photosystem II (PS II),2,3 in which a tetrameric manganese
cluster is known to catalyze the oxidation of water to
molecular oxygen. Although it has been known for years
that the manganese tetramer in PS II contains di-µ-oxo
dimanganese units2,4 and that water is the substrate of the
(4) Penner-Hahn, J. E. Struct. Bonding 1998, 90, 1.
(5) Chen, H. Y.; Faller, J. W.; Crabtree, R. H.; Brudvig, G. W. J. Am.
Chem. Soc. 2004, 126, 7345.
* To whom correspondence should be addressed. E-mail:
gary.brudvig@yale.edu (G.W.B.); robert.crabtree@yale.edu (R.H.C.).
(1) Abbreviations used in this report: EPR, electron paramagnetic
resonance; ESI-MS, electrospray ionization mass spectrometry; OEC,
oxygen-evolving complex; PS II, photosystem II.; UV-vis, ultraviolet-
visible spectroscopy; bbp, 2,6-bis(benzimidazol-2′-yl)pyridine; bpy,
2,2′-bipyridine; Br-terpy, 4′-bromo-2,2′:6′,2′′-terpyridine; Cl-terpy, 4′-
chloro-2,2′:6′,2′′-terpyridine; dipic, 2,6-dipicolinate; dpya, dipyrido-
[4,3-b; 5,6-b]acridine; EtO-terpy, 4′-ethoxy-2,2′:6′,2′′-terpyridine; HO-
terpy, 4′-hydroxyl-2,2′:6′,2′′-terpyridine; Me2N-terpy, 4′-(N,N-di-
methylamino)-2,2′:6′,2′′-terpyridine; mesityl-terpy, 4′-mesityl-2,2′:
6′,2′′-terpyridine; Ph-terpy, 4′-phenyl-2,2′:6′,2′′-terpyridine; phen, 1,10-
phenanthroline; py-phen, 2-(2′-pyridyl)-1,10-phenanthroline; tBu3-
terpy, 4,4′,4"-tri-tert-butyl-2,2′:6′,2′′terpyridine; terpy, 2,2′:6′,2′′-
terpyridine; tolyl-terpy, 4′-(4-methylphenyl)-2,2′:6′,2′′-terpyridine.
(2) Yachandra, V. K.; Sauer, K.; Klein, M. P. Chem. ReV. 1996, 96, 2927.
(3) Manchanda, R.; Brudvig, G. W.; Crabtree, R. H. Coord. Chem. ReV.
1995, 144, 1.
(6) (a) Dave, B. C.; Czernuszewicz, R. S.; Bond, M. R.; Carrano, C. J.
Inorg. Chem. 1993, 32, 3593. (b) Reddy, K. R.; Rajasekharan, M. V.;
Padhye, S.; Dahan, F.; Tuchagues, J. P. Inorg. Chem. 1994, 33, 428.
(7) Limburg, J.; Vrettos, J. S.; Liable-Sands, L. M.; Rheingold, A. L.;
Crabtree, R. H.; Brudvig, G. W. Science 1999, 283, 1524.
(8) (a) Sarneski, J. E.; Thorp, H. H.; Brudvig, G. W.; Crabtree, R. H.;
Schulte, G. K. J. Am. Chem. Soc. 1990, 112, 7255. (b) Reddy, K. R.;
Rajasekharan, M. V.; Arulsamy, N.; Hodgson, D. J. Inorg. Chem. 1996,
35, 2283.
(9) Suzuki, M.; Hayashi, Y.; Munezawa, K.; Suenaga, M.; Senda, H.;
Uehara, A. Chem. Lett. 1991, 1929.
(10) Wieghardt, K.; Bossek, U.; Nuber, B.; Weiss, J.; Bonvoisin, J.;
Corbella, M.; Vitols, S. E.; Girerd, J. J. J. Am. Chem. Soc. 1988, 110,
7398.
(11) Limburg, J.; Vrettos, J. S.; Chen, H. Y.; de Paula, J. C.; Crabtree, R.
H.; Brudvig, G. W. J. Am. Chem. Soc. 2001, 123, 423.
(12) Limburg, J.; Brudvig, G. W.; Crabtree, R. H. J. Am. Chem. Soc. 1997,
119, 2761.
10.1021/ic0509940 CCC: $30.25
Published on Web 09/13/2005
© 2005 American Chemical Society
Inorganic Chemistry, Vol. 44, No. 21, 2005 7661

Chen et al.
Table 1. Structures of Tridentate Ligands Used in This Study
molecules to bind to the exchangeable sixth coordination site
on each Mn; this leads to much more rapid O2 evolution
than seen for coordinatively saturated complexes such as
[MnIII/IV2(µ-O)2(bpy)4]3+.
To identify the steric and electronic factors that promote
catalytic activity, it is important to expand the family of di-
µ-oxo dimanganese complexes with terpy-like ligands and
characterize their activity. So far, only two such complexes
(L ) dipic and terpy) have been characterized in solution
and only the parent terpy Mn dimer complex has been
successfully isolated. Attempts to expand this series using
substituted terpy ligands were largely unsuccessful, mainly
owing to the instability of the corresponding [MnIII/IV2(µ-
O)2(L)2(OH2)2]3+ complexes in solution. In the cases where
the Mn(III/IV) dimer complexes are stable (L ) Ph-terpy
and mesityl-terpy, reported in this study), the isolation
process is tedious and often leads to low yields. Fast
screening of a ligand library is, therefore, incompatible with
this procedure.
An alternative way of testing catalytic activity is to prepare
the Mn(III/IV) dimer complexes in situ, following the
previous procedure that used a solution containing a mixture
of ligand and MnCl2 for the initial tests of 1a and 1b.12
Because a wide range of organic species are known to be
oxidized by high-valent oxo-Mn complexes13 and because
our recent results have shown that O2 evolution is inhibited
by the presence of excess free ligand during the catalysis,14
Mn(III/IV) dimer solutions free of organic impurities and
with a strict 1:1 ligand-to-Mn ratio are necessary for
comparative analysis of the catalytic activity of a series of
di-µ-oxo dimanganese complexes. Therefore, an easy and
general synthesis of low-valent Mn complexes that can be
used for the in situ preparation of Mn(III/IV) dimer solutions
is needed.
An improved synthesis of MnIILCl2 was adapted from
literature procedures for the synthesis of Mn(terpy)Cl2,15,16
Mn(Ph-terpy)Cl2,17 and related complexes.18-23 It provides
an easy way both to produce complexes in good yield without
organic impurities and to achieve the desired ligand-to-Mn
stoichiometry. Crystals of MnII(Ph-terpy)Cl2, MnII(mesityl-
terpy)Cl2, and MnII(dpya)Cl2 were isolated and characterized
by X-ray crystallography. The extensive MnIILCl2 family
reported in this study could also be of interest for high-field
EPR studies16 and as bleaching catalysts for detergent
applications.19-21
In addition, we report the in situ preparation of a series of
di-µ-oxo dimanganese(III/IV) complexes [MnIII/IV2(µ-O)2(L)2-
(X)2]3+ (L ) terpy, 1b; Cl-terpy, 1c; Br-terpy, 1d; Ph-terpy,
1e; tolyl-terpy, 1f; mesityl-terpy, 1g; tBu3-terpy, 1h; EtO-
terpy, 1i; py-phen, 1j; dpya, 1k; Me2N-terpy, 1l; HO-terpy,
1m; and bbp, 1n and X ) a labile ligand such as water,
chloride, or sulfate) from MnIILCl2 precursors (2b-2n,
respectively). Table 1 shows the structures of the tridentate
ligands. The resulting in situ solutions of 1 were characterized
by EPR spectroscopy, UV-vis spectroscopy, and ESI-MS,
and the catalytic activity of each of the in situ prepared
solutions was determined. We obtained crystals of [MnIII/IV
-
2
(µ-O)2(mesityl-terpy)2(OH2)2](NO3)3, and characterization of
the structure by X-ray crystallography showed that the two
Mn ions are crystallographically inequivalent. This dimer is
very soluble in acetonitrile; having water molecules coor-
dinated to the Mn ions, it allows studies in nonaqueous
solutions where low-temperature experiments are possible
and the concentration of water can be varied.
(13) Sarneski, J. E.; Michos, D.; Thorp, H. H.; Didiuk, M.; Poon, T.;
Blewitt, J.; Brudvig, G. W.; Crabtree, R. H. Tetrahedron Lett. 1991,
32, 1153.
(14) Chen, H. Y. Ph.D. Thesis, Yale University, New Haven, 2004.
(15) Harris, C. M.; Lockyer, T. N.; Stephens, N. C. Aust. J. Chem. 1966,
19, 1741.
(16) Mantel, C.; Baffert, C.; Romero, I.; Deronzier, A.; Pecaut, J.; Collomb,
M. N.; Duboc, C. Inorg. Chem. 2004, 43, 6455.
(17) Constable, E. C.; Lewis, J.; Liptrot, M. C.; Raithby, P. R. Inorg. Chim.
Acta 1990, 178, 47.
(18) Wang, S. X.; Zhu, Y.; Zhang, F. J.; Wang, Q. Y.; Wang, L. F.
Polyhedron 1992, 11, 1909.
(19) Wieprecht, T.; Heinz, U.; Xia, J.; Schlingloff, G.; Dannacher, J. J.
Surfactants Deterg. 2004, 7, 59.
Experimental Section
Materials. Me2N-terpy,24 EtO-terpy,24 Ph-terpy,17 Br-terpy,25 py-
phen,26,27 and dpya28 were synthesized following literature proce-
dures (Table 1). Oxone (2KHSO5‚KHSO4‚K2SO4) was purchased
(20) Wieprecht, T.; Xia, J. T.; Heinz, U.; Dannacher, J.; Schlingloff, G. J.
Mol. Catal. A-Chem. 2003, 203, 113.
(21) Schlingloff, G.; Wieprecht, T.; Bachmann, F.; Dannacher, J.; Dubs,
M. J.; Hazenkamp, M.; Richter, G.; Schmidt, B.; Schneider, A.;
Weingartner, P. PCT Int. Appl. 2002.
(22) Hogg, R.; Wilkins, R. G. J. Chem. Soc. 1962, 341.
(23) Judge, J. S.; Reiff, W. M.; Intille, G. M.; Ballway, P.; Baker, W. A.
J. Inorg. Nucl. Chem. 1967, 29, 1711.
(24) Constable, E. C.; Thompson, A.; Tocher, D. A.; Daniels, M. A. M.
New J. Chem. 1992, 16, 855.
(25) Whittle, B.; Batten, S. R.; Jeffery, J. C.; Rees, L. H.; Ward, M. D. J.
Chem. Soc., Dalton Trans. 1996, 4249.
(26) Hung, C. Y.; Wang, T. L.; Shi, Z. Q.; Thummel, R. P. Tetrahedron
1994, 50, 10685.
(27) Riesgo, E. C.; Jin, X. Q.; Thummel, R. P. J. Org. Chem. 1996, 61,
3017.
7662 Inorganic Chemistry, Vol. 44, No. 21, 2005

General Synthesis of Di-µ-oxo Dimanganese Complexes
Scheme 1. Synthesis of Mesityl-terpy.
solution forms a colorless layer beneath the deep yellow acetonitrile
layer. Within minutes, crystals of 2g precipitate out and float
between the two layers. This mixture was refluxed for 3 h with
vigorous stirring and then cooled to room temperature. The resulting
yellow crystalline precipitate was filtered and washed with cold,
dilute aq MnCl2 (0.1 M, 3 × 1 mL), cold acetonitrile (3 × 1 mL),
and then with copious diethyl ether. Drying under vacuum overnight
gave MnII(mesityl-terpy)Cl2 with good purity (0.729 g, 79%, Table
2). Crystals of Mn(mesityl-terpy)Cl2 were selected directly from
this product for X-ray crystallographic analysis.
from Acros Organics and was standardized using iodometric
titrations. All other reagents were purchased from Aldrich and used
without further purification. NMR spectra were recorded on Bruker
DPX 400 and 500 MHz instruments. Elemental analyses were
performed by Atlantic Microlabs, Inc., Norcross, GA.
The Mn(bbp)Cl2 complex (2n) was prepared as previously
described.18
General Procedure for the in situ Preparation of [MnIII/IV
-
2
Synthesis of 4′-Mesityl-2,2′:6′,2′′-terpyridine. This ligand was
prepared following the synthesis of Ph-terpy (Scheme 1).17
2-Acetylpyridine (18.9 mL, 0.17 mol) was added dropwise to a
stirred mixture of mesitylaldehyde (10 g, 0.067 mol), EtOH (120
mL), and aqueous NaOH (1.5 M, 180 mL), was stirred at 70 °C
for 3 days, was cooled to room temperature, and the mixture was
extracted with CHCl3. The extractant was washed with water three
times and then evaporated to give a red oil. The red oil, presumably
composed of 3-mesityl-1,5-bis(2-pyridyl)-1,5-pentanedione by com-
parison with the synthesis of Ph-terpy, was mixed with NH4Ac (27
g, excess) and 300 mL of ethanol and then refluxed overnight with
passage of air. The resulting blue solution was evaporated to a
quarter of its original volume, water (100 mL) was added, and the
mixture was extracted with CHCl3 (3 × 100 mL). Neutral alumina
(30 g) was added to the combined CHCl3 extracts, and the resulting
mixture was filtered through a packed neutral alumina layer. After
the solid mixture was washed with copious diethyl ether, the
combined filtrate was evaporated to give a crude mesityl-terpy
product. Recrystalization from a 1:1 CH3CN/CHCl3 mixture gave
colorless crystals (13.2 g, overall yield 56% based on mesitylal-
(µ-O)2(L)2(X)2]3+ (1b-1n). A 0.75-2 mM solution of MnIILCl2
in a 1:1 acetonitrile-water mixture was cooled to 0 °C. To this
solution was added an equal volume of 0.56-1.5 mM (0.75 equiv)
oxone in water. The solution turned deep green within 10 min.
Preparation of [MnIII/IV2(µ-O)2(L)2(H2O)2](NO3)3, (L ) Ph-
terpy (1e) and mesityl-terpy (1g)). A mixture of L (0.282 mmol,
1.05 equiv), MnCl2‚4H2O (0.269 mmol, 20 mL, 1 equiv), and KNO3
(5 g) in a mixture of CH3CN and H2O (25 and 20 mL, respectively)
was stirred at 50 °C until all reactants were dissolved to form a
yellow solution. After the solution was cooled in an ice bath, oxone
(0.202 mmol, 0.75 equiv) dissolved in 5 mL of H2O was added
with stirring. The solution turned deep green in a few minutes and
was allowed to stir for an extra 5 min in the ice bath. The majority
of the CH3CN in solution was then removed in vacuo, as estimated
by the decrease of the total volume, and the resulting green
precipitate was quickly filtered off and washed first with a cold
1:5 acetonitrile-water mixture (3 × 0.5 mL) and then with copious
diethyl ether. Drying under vacuum overnight gave the desired
product.
1e. 105 mg of Ph-terpy and 64 mg of MnCl2‚4H2O gave 52 mg
of 1e (33% yield according to MnCl2‚4H2O). Anal. Calcd for
C42H34Mn2N9O13: C, 51.34; H, 3.49; N, 12.83. Found: C, 51.63;
H, 3.58; N, 12.82.
1g. 99 mg of mesityl-terpy and 53 mg of MnCl2‚4H2O gave 75
mg of 1g (52% yield according to MnCl2‚4H2O). Anal. Calcd for
C48H46Mn2N9O13: C, 54.04; H, 4.35; N, 11.82. Found: C, 54.32;
H, 4.33; N, 11.62.
1
dehyde). H NMR (400 MHz, CDCl3, 25 °C): δ 8.68 (m, 4H, Ar
H), 8.31 (s, 2H, Ar H), 7.87 (t, 2H, Ar H), 7.33 (t, 2H, Ar H), 6.94
(s, 2H, Ar H), 2.34 (s, 3H, CH3), 2.08 (s, 6H, CH3). 13C NMR
(400 MHz, CDCl3, 25 °C): δ 156.3, 155.6, 151.8, 149.2, 137.1,
136.8, 136.7, 135.1, 128.2, 123.8, 122.1, 121.3, 21.11, 20.77. Anal.
Calcd for C24H21N3: C, 82.02; H, 6.02; N, 11.96. Found: C, 81.80;
H, 5.99; N, 11.96.
Preparation of MnIILCl2 Complexes. Method A. The ligand
L (0.43 mmol) was completely dissolved in a minimum amount of
hot acetone (typically 20 mL, 50 °C), and a solution of MnCl2‚
4H2O (2.2 mmol) in methanol (5 mL) was then added. A yellow
precipitate formed immediately, except for L ) tBu3-terpy, where
the precipitate formed after several minutes. The resulting suspen-
sion was allowed to ripen for 0.5-24 h at 50 °C and cooled to
0 °C. The precipitate was filtered and washed first with a 1:4 v/v
methanol-acetone mixture (3 × 0.5 mL) and then with copious
diethyl ether. Drying the precipitate under vacuum overnight gave
MnIILCl2 with good purity and yields. Typical yields are 70-98%
(Table 2). Yellow crystals of Mn(Ph-terpy)Cl2 suitable for X-ray
analysis were obtained by slow evaporation of a solution of Mn-
(Ph-terpy)Cl2 in a 1:1 acetonitrile-methanol mixture. Orange
crystals of Mn(dpya)Cl2 suitable for X-ray analysis were obtained
by slow evaporation of a mixed solution of MnCl2 (2 mM) and
dpya (2 mM) in a 1:1 acetonitrile-water mixture.
Following similar procedures, but with 1.20 equiv of mesityl-
terpy, a 1 mM solution of 1g was prepared in 1:1 acetonitrile-
water mixture. Slow evaporation of this solution gave black-green
crystals of 1g suitable for X-ray analysis.
Crystal Structure Determination of Mn(Ph-terpy)Cl2 (2e),
Mn(mesityl-terpy)Cl2 (2g), Mn(dpya)Cl2 (2k), and [Mn2(µ-O)2-
(mesityl-terpy)2(OH2)2](NO3)3 (1g). All diffraction data were
collected on a Nonius KappaCCD diffractometer with graphite
monochromated Mo KR radiation. ORTEP diagrams are shown in
Figures 1-3, and the crystallographic data are presented in Table
3. Crystals of complexes with approximate dimensions 0.20 × 0.10
× 0.08 mm3 for 2e, 0.15 × 0.10 × 0.10 mm3 for 2g, 0.3 × 0.3 ×
0.3 mm3 for 2k, and 0.07 × 0.19 × 0.20 mm3 for 1g were mounted
with epoxy cement on the tips of fine glass fibers. All data were
corrected for Lorentz and polarization effects. The structures were
solved by direct methods and expanded using Fourier techniques.29
The non-hydrogen atoms were refined anisotropically. Hydrogen
atoms were included in calculated positions except for 2k for which
hydrogen atoms were refined isotropically.
Method B. Mesityl-terpy (0.676 g, 1.9 mmol) was suspended
in 5 mL of heated CH3CN, to which a concentrated aqueous MnCl2‚
4H2O (3 M, 5 mL, large excess) solution was added. Owing to the
high density and ionic strength of the MnCl2 solution, the aqueous
Electrospray Ionization Mass Spectrometry. ESI-MS spectra
were collected on a Waters/Micromass ZQ 4000 mass spectrometer.
(28) Hung, C. Y.; Wang, T. L.; Jang, Y. C.; Kim, W. Y.; Schmehl, R. H.;
Thummel, R. P. Inorg. Chem. 1996, 35, 5953.
(29) Vandersluis, P.; Spek, A. L. Acta Crystallogr. A 1990, 46, 194.
Inorganic Chemistry, Vol. 44, No. 21, 2005 7663

Chen et al.
Table 2. Preparation of MnIILCl2 Complexes
entry
no.
formula
method
A
yield
96%
elemental analysis
Mn(terpy)Cl2
2b
C15H11Cl2MnN3
calcd
found
calcd
found
calcd
found
calcd
found
calcd
found
calcd
found
calcd
found
calcd
found
calcd
found
calcd
found
calcd
found
calcd
found
C, 50.17; H, 3.09; N, 11.70
C, 50.03; H, 3.10; N, 11.70
C, 45.78; H, 2.56; N, 10.68
C, 45.81; H, 2.53; N, 10.44
C, 41.13; H, 2.30; N, 9.59
C, 40.86; H, 2.28; N, 9.58
C, 57.96; H, 3.47; N, 9.66
C, 57.95; H, 3.40; N, 9.65
C, 58.82; H, 3.81; N, 9.35
C, 58.40; H, 3.88; N, 9.31
C, 60.39; H, 4.43; N, 8.80
C, 60.28; H, 4.48; N, 8.94
C, 61.48; H, 6.69; N, 7.97
C, 61.56; H, 6.76; N, 7.86
C, 50.64; H, 3.75; N, 10.42
C, 50.54; H, 3.72; N, 10.46
C, 53.29; H, 2.89; N, 10.97
C, 53.52; H, 2.92; N, 11.04
C, 56.05; H, 2.72; N, 10.32
C, 56.05; H, 2.65; N, 10.34
C, 50.77; H, 4.01; N, 13.93
C, 50.67; H, 3.97; N, 14.01
C, 48.03; H, 2.96; N, 11.20
C, 48.11; H, 3.01; N, 11.27
Mn(Cl-terpy)Cl2
Mn(Br-terpy)Cl2
Mn(Ph-terpy)Cl2
Mn(tolyl-terpy)Cl2
Mn(mesityl-terpy)Cl2
Mn(tBu3-terpy)Cl2
Mn(EtO-terpy)Cl2
Mn(py-phen)Cl2
Mn(dpya)Cl2
2c
2d
2e
2f
C15H10Cl3MnN3
C15H10BrCl2MnN3
C21H15Cl2MnN3
C22H17Cl2MnN3
C24H21Cl2MnN3
C27H35Cl2MnN3
C17H15Cl2MnN3O
C17H11Cl2MnN3
C19H11Cl2MnN3
C17H16Cl2MnN4
C15H11Cl2MnN3O
A
A
A
A
B
A
A
A
A
A
A
96%
80%
96%
98%
79%
73%
77%
87%
85%
70%
90%
2g
2h
2i
2j
2k
2l
Mn(Me2N-terpy)Cl2
Mn(HO-terpy)Cl2
2m
Figure 1. ORTEP diagrams of MnII(Ph-terpy)Cl2 (2e) and MnII(mesityl-terpy)Cl2 (2g), showing 30% probability thermal ellipsoids. Hydrogen atoms are
omitted for clarity.
Owing to the weak signal at high mass, multiple scans over a narrow
mass range were required to obtain a good signal/noise ratio. To
obtain simulated spectra, a Gaussian peak was fitted to the highest
observed peak by varying mass (M), peak width, and peak area. A
group of Gaussian peaks was then generated at mass ) M ( n (n
) integers) with the same peak width and with peak areas adjusted
to the calculated isotope ratios. Isotope ratios were calculated using
Masslynx (V4.0) and the isotope distribution calculator at http://
www.sisweb.com/mstools.htm.
UV-Vis Spectroscopy. UV-vis spectra were recorded on a
Varian Cary-3E UV-visible spectrophotometer at room temperature
using a 1 cm path length cuvette.
EPR Spectroscopy. Perpendicular-mode EPR spectra were
collected on an X-band Bruker E-500 spectrometer equipped with
a SHQ cavity and an Oxford ESR-900 liquid helium cryostat. In
situ solutions of 1 were diluted with an equivalent volume of
NaOAc/HOAc buffer, pH ) 4.5; aliquots (250 µL) of this mixture
were transferred to EPR tubes, and were frozen in liquid nitrogen.
All in situ solutions were frozen within 10 min of mixing for the
EPR measurements. Spectra used for quantitation were collected
at 10 K with the following settings: microwave frequency, 9.38
GHz; microwave power, 1 mW; field modulation amplitude, 2 G;
field modulation frequency, 100 kHz; and time constant, 0.3 s.
Spectra shown in Figure 6 were collected with a field modulation
amplitude of 20 G.
Figure 2. ORTEP diagram of MnII(dpya)Cl2 (2k), showing 30% prob-
ability thermal ellipsoids. Hydrogen atoms are omitted for clarity. Selected
bond distances (Å) and bond angles (deg):
Mn(1)-N(1)
N(1)-C(5)
C(12)-C(16)
Mn(1)-N(3)
C(9)-N(2)
2.32
1.36
1.44
2.34
1.34
Mn(1)-N(2)
2.19
1.44
1.36
2.33
1.34
C(5)-C(9)
C(16)-N(3)
Mn(1)-Cl(2)
N(2)-C(12)
The quantitation of the yields of the [MnIII/IV2(O)2L2X2]3+
complexes obtained in situ from MnIILCl2 precursors was done by
measuring the peak-to-trough height of the lowest field hyperfine
N(1)-C(5)-C(9)
N(1)-C(12)-C(16)
117
116.6
C(5)-C(9)-N(2)
C(12)-C(16)-N(3)
116.1
116.6
7664 Inorganic Chemistry, Vol. 44, No. 21, 2005

General Synthesis of Di-µ-oxo Dimanganese Complexes
Figure 3. ORTEP diagram of [Mn2(µ-O)2(mesityl-terpy)2(OH2)2](NO3)3 (1g), showing 30% probability thermal ellipsoids. Hydrogen atoms, nitrate counterions,
and waters of crystallization are omitted for clarity. Selected bond distances (Å):
Mn(1)-N(1)ax
Mn(1)-N(3)ax
Mn(1)-O(2)eq
Mn(2)-N(4)ax
Mn(2)-N(6)ax
Mn(2)-O(2)eq
Mn(1)‚‚‚Mn(2)
2.22
2.19
1.81
2.04
2.06
1.76
2.72
Mn(1)-N(2)eq
Mn(1)-O(1)eq
Mn(1)-O(3)eq
Mn(2)-N(5)eq
Mn(2)-O(1)eq
Mn(2)-O(4)eq
2.05
1.84
2.00
2.04
1.81
1.99
Table 3. X-Ray Crystallographic Data for 1g‚6H2O, 2e, 2g, and 2k
1g‚6H2O
2e
2g
2k
empirical formula
fw
cryst syst
space group
a, Å
b, Å
c, Å
R, deg
â, deg
γ, deg
V, Å3
C48H58Mn2N9O19
1174.91
triclinic
C21H15Cl2MnN3
435.20
C24H21Cl2MnN3
477.28
C19H11Cl2MnN3
407.16
monoclinic
P2(1)/n(#14)
12.160(2)
9.756(2)
16.676(3)
90
106.68(3)
90
1895.0(7)
4
monoclinic
P2(1)/c(#14)
11.694(2)
12.843(3)
14.854(3)
90
97.59(3)
90
2211.2(8)
4
triclinic
P1h(# 2)
7.5455(15)
9.5624(19)
12.599(3)
95.91(3)
97.55(3)
110.33(3)
834.2(3)
2
P1h(# 2)
7.9563(3)
13.3147(4)
25.5000(7)
80.666(2)
86.254(2)
84.000(2)
2647.8(1)
2
Z
T, °C
-90(1)
0.71069
1.461
-100(2)
0.71073
1.525
-100(2)
0.71073
1.434
23(2)
0.71073
1.621
λ, Å
d
cal, g/cm3
F000
1202.00
884
980
410
no. of reflns
total/unique
no. observations
(I > 3.00σ(I))
refln/param ratio
R; Rw
30 050/12 043
7869/4662
10 103/5448
5404/3930
4921
4662
5448
3930
7.00
19.11
19.39
17.39
0.068a; 0.087b
0.041a; 0.081c
0.046a; 0.106c
0.040a; 0.078c
GOF
indicator
2.17
1.006
1.017
1.003
2
2
2
2
a R ) ∑||Fo| ∑|Fc||/∑|Fo|; b RW ) [∑w(|Fo|∑|Fc|)2/∑wFo ]1/2
;
c RW ) {∑[w(Fo ∑ Fc )2]/∑[w(Fo )2]}1/2
.
line in the 16-line EPR signal characteristic of di-µ-oxo Mn(III/
IV) compounds. This hyperfine line was chosen to minimize
interference from Mn(II) species in solution. The yield of each di-
µ-oxo Mn(III/IV) complex was determined by reference to the peak
to trough height of the 16-line EPR signal obtained for a solution
of 1g prepared from the pure complex with the same total
manganese concentration as in the in situ experiments. A solution
of 1g with tetrabutylammonium chloride added to a final concentra-
tion of two chlorides per manganese was also tested in the same
way to check for the effect of chloride ions on the EPR signal of
the [MnIII/IV2(O)2L2X2]3+ complex. The results for 1g with and
without added tetrabutylammonium chloride were identical within
experimental error.
Oxygen Evolution Assays. The rates of oxygen evolution from
solutions of the [MnIII/IV2(O)2L2X2]3+ complexes were measured
at 298 K by a YSI standard oxygen probe (Clark Electrode) attached
to a Cole-Parmer chart recorder using oxone as the oxidant as
previously described.11 The in situ experiments were done by mixing
a 7.2 mM solution of the Mn(II)LCl2 complex in a 1:1 (v/v)
acetonitrile-water mixture with an equal volume of a 5.4 mM (0.75
equiv) aqueous solution of oxone at 0 °C. After waiting a few
minutes for formation of the [MnIII/IV2(O)2L2X2]3+ complex, as
evidenced by the solution turning green, a 50 µL aliquot was
injected into 6 mL of an aqueous solution of 15 mM oxone stirring
beneath the Clark electrode. Thus, solutions of manganese com-
plexes were in a 1:3 acetonitrile-water mixture before injection
Inorganic Chemistry, Vol. 44, No. 21, 2005 7665

Chen et al.
into the Clark electrode cell, except in the case of 2e, which was
in a 3:1 acetonitrile-water mixture. This was done in order to
compare with the solution of the authentic dimer 1e, which was
not soluble in 1:3 acetonitrile-water but was soluble in 3:1
acetonitrile-water (see below). A 25% excess of mesityl-terpy
ligand and oxone (above what is calculated for stoichiometric
conversion of Mn(II) to Mn2(III/IV)) was used with 2g, as mesityl-
terpy was observed to undergo relatively fast oxidation in the
presence of oxone (data not shown). The pure complex 1b was
dissolved in a 1:3 acetonitrile-water mixture to a concentration of
3.6 mM total manganese. Fifty microliters of this solution was
injected into 6 mL of a 15 mM aqueous solution of oxone stirring
beneath the Clark electrode. The reported rates are normalized to
the rate obtained for this sample. A solution of pure 1e (3.6 mM
manganese in 3:1 acetonitrile-water) was also assayed for com-
parison to the di-µ-oxo dimanganese complex generated in situ from
a solution of 2e. Finally, a solution of pure 1g (3.6 mM manganese
in 1:3 acetonitrile-water) was prepared with 25% excess ligand
per manganese and excess oxone (as above, for 2g) and assayed
for comparison to the di-µ-oxo dimanganese complex generated in
situ from a solution of 2g.
yields. Following either method A or B, a small amount of
organic impurity and the excess MnCl2 are expected to
remain dissolved in the solvent mixture and to be removed
on isolation.
In contrast to the stability of MnII(terpy)Cl2 in aqueous
solution, the dissolution of many of the new MnIILCl2
complexes (L ) Cl-terpy, Br-terpy, ph-terpy, tolyl-terpy,
mesityl-terpy, tBu3-terpy, and bbp) in water leads to pre-
cipitation of the ligands within minutes. This is surprising
since it is generally believed that the ligation of tridentate
terpy-type ligands to Mn2+ is strong and that the substituent
groups at positions far from the metal binding site would
cause only minor changes in the chemical and physical
properties of the MnIILCl2 complexes. De-coordination may
be driven by precipitation of the ligand, however. Although
acetonitrile is generally not a good solvent for either the
ligands or the MnLCl2 complexes, all the MnIILCl2 com-
plexes are soluble and stable in a 1:1 mixture of acetonitrile
and water. Mn(II) complexes with less-hydrophobic ligands
(EtO-terpy, Me2N-terpy, HO-terpy, py-phen, and dpya) are
similar to Mn(terpy)Cl2 in solubility and stability in aqueous
solution.
Results and Discussion
Synthesis of Mesityl-terpy. The synthesis of mesityl-terpy
via Suzuki coupling of Br-terpy with a suitable boron reagent
was reported in the recent literature.30,31 By following the
procedure for synthesis of Ph-terpy,17 we were able to
synthesize mesityl-terpy by this better route (Scheme 1). The
reaction conditions were only slightly modified, although the
separation methods for products are different, owing to the
difference in solubility. Heating during the first step was
necessary to improve the overall yield (22% to 56%),
possibly by overcoming the sterically demanding second
Michael addition of 2-acetylpyridine to mesitylaldehyde.31
Using this method, a large-scale synthesis from commercially
available starting materials is achieved.
Synthesis and Characterization of MnIILCl2 (2). The
two preparative methods used for the syntheses of MnIILCl2
were adapted from previous work.15,17,18 Method A generally
provided pure product in high yield, as the ligand was fully
dissolved in acetone and MnCl2 in methanol while the
product MnIILCl2 has a low solubility in the final 4:1
acetone-methanol mixture. However, MnII(mesityl-terpy)-
Cl2 is very soluble in the 4:1 acetone-methanol mixture, so
method B was, therefore, used for its synthesis. This method
uses a minimum amount of acetonitrile-water mixture, in
which the mesityl-terpy was not completely dissolved. Long
reflux is necessary to ensure complete reaction. In both
methods, a large excess of MnCl2 was used to suppress [MnII-
(L)2]2+ formation and to decrease the solubility of MnIILCl2.
The MnIILCl2 complexes usually precipitate as fine particles,
and incubation at elevated temperature is needed to render
them suitable for filtration. In some cases, for example with
tBu3-terpy, a significant amount of the fine precipitate was
lost during filtration, even after ripening; this leads to lower
Figures 1 and 2 show the crystal structures of Mn(Ph-
terpy)Cl2 (2e), Mn(mesityl-terpy)Cl2 (2g), and Mn(dpya)-
Cl2 (2k). The major difference between the structures of 2e
and 2g is that the more sterically demanding mesityl group
of 2g tilts at a larger angle to the terpy plane (dihedral angles
between the central pyridine ring and its aromatic pendant
are 21.9° and 74.8° for 2e and 2g, respectively). This effect
appears to prevent close stacking of 2g in the crystal
structure; the inefficient stacking contributes to the smaller
calculated density of 2g (1.434 g/cm3) versus 2b, 2e, and
2k (1.614, 1.525, and 1.621 g/cm3, respectively), and may
contribute to the higher solubility of 2g in organic solvents.
Compared to the trigonal bipyramidal geometry of MnII-
(terpy)X2 (X ) Br (3) and I (4))16 and the distorted square
pyramidal geometry of 2b16 and 2n,18 the geometries of MnII
in 2e, 2g, and 2k are between the two geometries. This is
evident from Table 4; the X(1)-Mn(1)-N(2) and X(2)-
Mn(1)-N(2) angles of 2e, 2g, and 2k are between the
corresponding angles of 2b and the equivalent angles of 3
and 4. Interestingly, Cl(1) of 2g is disordered over two
positions with an occupancy factor ratio of 60:40. With Cl-
(1) at one position, the Cl(1)-Mn(1)-N(2) and Cl(2)-Mn-
(1)-N(2) angles are almost equivalent, while at the other
position they are not. The absence of ligand-field effects for
a d5 high-spin Mn(II) clearly allows packing forces to distort
those structures.
The ligand dpya differs from terpy only by two etheno
bridges (Table 1) which fuse the three pyridine rings and
appear to prevent the outer pyridine rings from tilting toward
the central pyridine ring. For example, the N(1)-C(5)-C(9),
C(5)-C(9)-N(2), N(2)-C(12)-C(16), and C(12)-C(16)-
N(3) angles of 2k (117.0°, 116.1°, 116.6°, and 116.6°,
respectively) are larger than the corresponding angles in MnII-
(terpy)Cl2 (2b, 115.4°, 114.9°, 114.4°, and 115.5°, respec-
tively). This leads to an increase of the N(1)-N(3) distances
from 2b (4.28 Å) to 2k (4.42 Å) even though the C-C bond
(30) Goodall, W.; Wild, K.; Arm, K. J.; Williams, J. A. G. J. Chem. Soc.,
Perkin Trans. 2 2002, 1669.
(31) Leslie, W.; Batsanov, A. S.; Howard, J. A. K.; Williams, J. A. G.
Dalton Trans. 2004, 623.
7666 Inorganic Chemistry, Vol. 44, No. 21, 2005

General Synthesis of Di-µ-oxo Dimanganese Complexes
Figure 6. Selected EPR spectra showing the preparation of [MnIII/IV2(µ-
O)2(L)2(X)2]3+ in situ from MnIILCl2 and oxone, with L ) (a) terpy, (b)
Cl-terpy, (c) dpya, (d) Me2N-terpy, and (e) HO-terpy. Mixed-valence Mn-
(III/IV) dimers were also prepared for L ) EtO-terpy, Br-terpy, Ph-terpy,
tolyl-terpy, mesityl-terpy, and tBu3-terpy; their corresponding 16-line spectra
were analogous to those in (a) and (b) and, therefore, are not shown here.
When L ) py-phen, the EPR spectrum of the corresponding in situ solution
of 1j was similar to that in (c), but with a slightly stronger 16-line signal.
The in situ yields of [MnIII/IV2(µ-O)2(L)2(X)2]3+ for all of the ligands are
listed in Table 5.
Figure 4. Selected UV-vis spectra (0 °C) of Mn(III/IV) and Mn(IV/IV)
dimers prepared in situ with meridionally coordinating ligand terpy (A)
and tolyl-terpy (B). Similar absorption bands were also observed for L )
Cl-terpy, Br-terpy, Ph-terpy, mesityl-terpy, tBu3-terpy, and EtO-terpy.
Shortly after the addition 0.75 equiv of oxone solution,
MnLCl2 solutions of complexes 2b-2i turned to a deep green
color characteristic of Mn(III/IV) dimers. The stability of
the green solutions varied significantly from species to
species, but all of them precipitated MnO2 after long periods
(>24 h) at room temperature. Complexes 1c and 1d (L )
Cl-terpy and Br-terpy) are among the least stable in solution;
the brown MnO2 precipitate was observed within 1 h even
when the solutions were kept at 0 °C. To assign the green
Mn complexes further, UV-vis spectroscopy, EPR spec-
troscopy, and ESI-MS studies of the in situ solutions were
carried out.
Despite the differences between the ligands involved in
the complexes, common UV-vis absorption features at 550
and around 650 nm (Figure 4) were observed in all eight
cases that are indicative of in situ formation of the di-µ-oxo
Mn(III/IV) dimers (1b-1i). These two features closely
resemble the absorptions of the previously characterized
[MnIII/IV2(O)2(terpy)2(OH2)2](NO3)3 complex,7,32 which were
assigned to a d-d transition and an oxygen-to-manganese
charge-transfer band, respectively, on the basis of the
assignments of similar absorptions of [MnIII/IV2O2(bpy)4]3+.33
Further addition of oxone oxidizes the Mn(III/IV) complexes
to the corresponding Mn(IV/IV) complexes, as is evident
from the increase of a new absorption shoulder in the range
420-450 nm (Figure 4). The similarity in the UV-vis
absorption bands suggests that the high-valent Mn dimer
complexes share similar structures.
Figure 5. Selected ESI-MS data showing [MnIII/IV2(µ-O)2(L)2Cl2]+ and
[MnIII/IV2(µ-O)2(L)2(SO4)]+ for L ) Cl-terpy and EtO-terpy. Dotted lines
show the simulated spectra in both cases. Corresponding peaks were also
observed for L ) terpy, Br-terpy, Ph-terpy, tolyl-terpy, mesityl-terpy, tBu3-
terpy, and py-phen.
distance between the adjacent pyridine rings decreases from
1.49 to 1.44 Å.
In Situ Preparation of [MnIII/IV2(O)2L2X2]3+. One major
difficulty in the synthesis of 1 was ligation of all free Mn2+
before the addition of an oxidant, otherwise the oxidation
of uncoordinated Mn2+ always led to rapid and irreversible
precipitation of MnO2. These initial attempts failed because
in aqueous solution most of the terpy derivatives do not form
stable complexes with Mn2+ (vide supra). A 1:1 acetonitrile-
water solvent mixture provides good solubility and stability
for MnIILCl2 complexes and is resistant to oxidation by high-
valent Mn complexes. To minimize O2 evolution catalyzed
by the initially formed 1 when oxone was added to oxidize
the Mn(II) complexes to the Mn(III,IV) dimers, the MnLCl2
solutions were cooled to 0 °C to slow O2 evolution and allow
complete mixing with the oxone solution.
The ESI-MS spectra of solutions of 1 show peaks from
the [MnIII/IV2(O)2L2]3+ core paired with the available coun-
terions. For Mn(III/IV) dimer complexes prepared in situ,
(32) Collomb, M. N.; Deronzier, A.; Richardot, A.; Pecaut, J. New J. Chem.
1999, 23, 351.
(33) Cooper, S. R.; Calvin, M. J. Am. Chem. Soc. 1977, 99, 6623.
Inorganic Chemistry, Vol. 44, No. 21, 2005 7667

Chen et al.
Table 4. Selected Angles and Distances for Complexes 2b,16 2e, 2g, 2k, 2n,18 3,16 and 416
2b (X ) Cl)
2e (X ) Cl)
2g (X ) Cl)b
2k (X ) Cl)c
2n (X ) Cl)
3 (X ) Br)
4 (X ) I)
T (°C)a
25
-100
115.8
129.2
115.0
2.36
-100
23
25
25
25
X(1)-Mn-N(2) (deg)
X(2)-Mn-N(2) (deg)
X(1)-Mn-X(2) (deg)
X(1)-Mn (Å)
X(2)-Mn (Å)
N(2)-Mn (Å)
104.6
144.2
111.2
2.37
2.35
2.21
109.7 (122.6)
128.6
114.0
133.2
112.8
2.35
2.33
2.19
105.9
141.1
112.9
2.38
2.34
2.28
122.0
122.0
116.1
2.50
2.50
2.20
122.8
122.8
114.3
2.72
2.72
2.23
121.6 (108.7)
2.29 (2.36)
2.36
2.36
2.21
2.21
a T, temperature of data collection; deg, degree. b The bond angles and distances involving the disordered Cl(1)′ in 2g are included in parentheses. c Numbering
scheme of nitrogen and halogen atoms in all complexes follows that of 2k in Figure 2, so that the X(1)-Mn-N(2) angle is smaller than that of X(2)-
Mn-N(2).
Table 5. Yields and O2-Evolving Activities of Solutions of Pure and in
the most prominent high-mass peaks were always found at
Situ Formed [MnIII/IV2(µ-O)2(L)2(X)2]3+ Complexes
m/z values corresponding to [MnIII/IV2(O)2L2Cl2]+ and
[MnIII/IV2(O)2L2(SO4)]+ (Figure 5), as expected because Cl-
O2 evolution rate
ligand, L (Mn dimer)a
yield (%)b
(% of control)c
2-
and SO4 are the most abundant counterions in solution.
tBu3-terpy (1h) - in situ
Ph-terpy (1e) - pure solution
Ph-terpy (1e) - in situ
EtO-terpy (1i) - in situ
terpy (1b) - pure solution
terpy (1b) - in situ
97
100
80
77
100
75
100
60
47
14
11
8
18
25
35
47
100
141
25
40
44
14
12
-
The two water ligands of 1 are probably lost during the
desolvation process in the MS, as was found in the ESI-MS
spectra of similar compounds.5,32 The assignment of the MS
peaks is constrained by the m/z values and isotope peak
ratios. On the basis of this information, charge considerations
of the assigned peaks require, in both of the MS peaks, the
existence of two oxo groups, two ligand molecules, and Mn-
(III) and Mn(IV) ions. This provides strong evidence that
[MnIII/IV2(O)2L2X2]3+ exists in solution (where X ) a labile
ligand such as water, chloride, or sulfate).
mesityl-terpy (1g) - pure solution
mesityl-terpy (1g) - in situ
tolyl-terpy (1f) - in situ
Cl-terpy (1c) - in situ
py-phen (1j) - in situ
Me2N-terpy (1l) - in situ
dpya (1k) - in situ
3
0
21
-
HO-terpy (1m) - in situ
The EPR spectra of the in situ-generated green species
show a 16-line EPR signal (Figure 6), which is characteristic
of mixed-valence Mn(III/IV) dimer complexes. Although
slightly different hyperfine splitting patterns were previously
reported for the [MnIII/IV2(O)2(dipic)2(OH2)2]- and [MnIII/IV2(O)2-
(bpy)4]3+ complexes,34,35 the 16-line signals observed for the
solutions of complexes 1b-1i are almost exactly the same
(Figure 6), with only small differences in the line width. The
replacement of terpy by structurally similar ligands does not
seem to significantly perturb the electronic structure of the
Mn(III/IV) dimer, less so than the replacement of terpy by
dipic or bpy. The UV-vis, EPR, and ESI-MS data suggest
that di-µ-oxo Mn(III/IV) dimer complexes with similar
structures were successfully synthesized in solution.
Following the same procedure, the oxidation of 2j, 2k,
2l, and 2m led to red solutions, while a brown MnO2
precipitate formed immediately after the oxidation of 2n.
Sixteen-line EPR signals of lower intensity were observed
in the case of 2j, 2k, and 2l (Figure 6), although ESI-MS
peaks corresponding to Mn(III/IV) dimers were observed
only in the case of 2j. The oxidized solutions of 2l and 2m
show significant MnII EPR signals, suggesting the reduction
of high-valent Mn species. Ligand oxidation, electronic
effects, or steric effects in the case of L ) bbp could
contribute to the failure of Mn(III/IV) dimer complex
formation.
a The MnLCl2 complexes were used as precursor complexes to generate
the corresponding Mn(III/IV) dimers in situ (see text). b Percentage yields
are calculated from EPR peak heights by normalizing to the peak height
obtained for a solution of pure 1g in 1:3 acetonitrile-water, diluted with
an equal volume of 0.1 M acetate buffer at pH 4.5 (see text). c Rates are
normalized to that obtained for a solution of pure 1b in 1:3 acetonitrile-
water (see text).
complex. The yields (Table 5) were obtained by measuring
peak heights of the EPR signals in reference to a solution of
1g as a standard (see Experimental Section). This procedure
gives a series of new complexes with structures similar to
that of [MnIII/IV2(O)2(terpy)2(OH2)2]3+. The preparation and
characterization of this family of complexes expands the
limited number of known cases in which labile coordination
sites are available on a di-µ-oxo-bridged dimanganese
complex and, more importantly, provides the basis for a fast-
screening assay for catalytic O2-evolution activity.
Measurements of O2-evolving activity are provided in
Table 5. The use of MnIILCl2 complexes as precursors
ensures no excess of free ligand (except in the case of 2g
where it was found to be desirable) or free Mn2+, and thus
enables comparative analyses of catalytic activity. In measur-
ing catalytic activity, the MnIILCl2 precursors were first
mixed with a measured amount of oxone that is just sufficient
to form the complexes 1 (see Experimental Section). After
allowing time for the assembly of 1 from the Mn(II)
precursors (as judged by the appearance of green color), an
aliquot of the solution was added to the assay solution
containing excess oxone and the O2-evolving activity was
measured. Overall, the in situ solutions show significant
activity, making the in situ assay a promising approach. A
comparison of activities of solutions of the pure complexes
1b, 1e, and 1g with those of solutions of the in situ-generated
Overall, the in situ synthesis of di-µ-oxo dimanganese-
(III/IV) complexes from the corresponding MnIILCl2 complex
is simple and, in most cases, gives a good yield of the desired
(34) Limburg, J.; Crabtree, R. H.; Brudvig, G. W. Inorg. Chim. Acta 2000,
297, 301.
(35) Cooper, S. R.; Dismukes, G. C.; Klein, M. P.; Calvin, M. J. Am. Chem.
Soc. 1978, 100, 7248.
7668 Inorganic Chemistry, Vol. 44, No. 21, 2005

General Synthesis of Di-µ-oxo Dimanganese Complexes
complexes obtained from 2b, 2e, and 2g, respectively, shows
that activities of the in situ solutions compare favorably with
their corresponding pure dimer solutions. The higher activi-
ties observed for some of the in situ solutions may be
ascribed to the differences in solution conditions between
the authentic dimer solutions and the in situ-assembled dimer
solutions due to the unavoidable presence of added oxone
in the latter case.
(µ-O)2(terpy)2(CF3CO2)2](ClO4) (5b) complexes are soluble
in acetonitrile, and the resulting solutions were studied by
electrochemistry.36,37
Following
the
preparation
procedures
for
[MnIII/IV2O2L2(H2O)2](NO3)3, but with a larger excess of
mesityl-terpy (20% excess), slow evaporation of the 1:1
acetonitrile-water solution led to a mixture of color-
less mesityl-terpy and black-green crystals of 1g suitable
for X-ray crystallography. An ORTEP diagram of
[MnIII/IV2O2(mesityl-terpy)2(H2O)2](NO3)3 (1g) is shown in
Figure 3. Like [MnIII/IV2O2(terpy)2(H2O)2](NO3)3 (1b),7 two
aqua ligands can be found coordinated to both Mn ions of a
Mn-di-µ-oxo-Mn core. The di-µ-oxo-bridged Mn-Mn
distance in 1g (2.72 Å) is almost the same as that in 1b (2.73
Å) and in other di-µ-oxo-bridged Mn dimers.3 In 1g, the
dihedral angle between the central pyridine ring and its
aromatic pendant is 71.0°, similar to that of the free ligand
(67.5°)31 and Mn(mesityl-terpy)Cl2 (74.8°). As in the case
of 2g, the sterically demanding mesityl groups in 1g may
prevent close packing of the molecules, leading to smaller
dcal (1.461 g/cm3, compared to dcal ) 1.614 g/cm3 of 1b),
and may contribute to the enhanced solubility of 1g in
acetonitrile compared to that of 1b and 1e.
Isolation of [MnIII/IV2(O)2L2(H2O)2](NO3)3. In contrast
to the short time scale of a catalytic run (<30 min), the
isolation and crystallization of Mn(III/IV) dimer complexes
takes a much longer time. Excess ligand is, therefore, used
to stabilize the Mn complexes in solution. To minimize
degradation, a low concentration of Mn(III/IV) dimer
complexes (1 mM) was synthesized at 0 °C and kept slightly
below this temperature as acetonitrile was removed
-
in situ. With a large excess of NO3 present, the
removal of acetonitrile led to the precipitation of
[MnIII/IV2(O)2L2(H2O)2](NO3)3. To avoid ligand precipitation
caused by the instability of the Mn(III/IV) dimer complexes
in pure water, the green precipitates were isolated by filtration
before the complete removal of acetonitrile. Difficulty in
controlling this step often led to limited yields. Following
this procedure, [MnIII/IV2(O)2(Ph-terpy)2(H2O)2](NO3)3 and
[MnIII/IV2(O)2(mesityl-terpy)2(H2O)2](NO3)3 were successfully
isolated and characterized by elemental analysis, while
similar green compounds with Cl-terpy and Br-terpy were
isolated but could not be obtained in a pure form. The EPR
and ESI-MS spectra of the four compounds in 1:1 water-
acetonitrile showed the presence of [MnIII/IV2(O)2L2X2]
species.
One major difference between the structures of 1b, 1g,
and 5b is that the two Mn ions of 1b are crystallographically
equivalent, while those of 5b and 1g are inequivalent. The
Mn(1) of 1g is typical of an axially elongated d4 Mn(III)
ion in that the Mn(1)-Nax distances (Mn(1)-N(1), 2.22 Å
and Mn(1)-N(3), 2.19 Å) are significantly longer than the
Mn(1)-Neq distance (2.05 Å). In contrast, the Mn(2)-Nax
distances (Mn(2)-N(4), 2.06 Å and Mn(2)-N(6), 2.04 Å)
are almost the same as the Mn(2)-Neq distance (2.04 Å),
suggesting that Mn(2) is a normal d3 Mn(IV) ion. All Mn-O
and Mn-N distances of Mn(2) are shorter than the corre-
sponding bond distances of Mn(1); this confirms the assign-
ment of Mn(1) as higher-valent Mn(IV). The inequivalency
of the two Mn ions in 1g is consistent with its 16-line EPR
signal, which is the result of hyperfine coupling between an
S ) 1/2 electron spin with two inequivalent 55Mn nuclei.
The different equivalency of the Mn ions in 1b versus 1g
and 5b is reminiscent of the known disorder phenomena in
the phen/bpy Mn(III/IV) dimers. The Mn ions are crystal-
lographically inequivalent in [MnIII/IV2O2(bpy)4](ClO4)3 (6a)
and [MnIII/IV2O2(phen)4](ClO4)3 (6b) but are equivalent in
[MnIII/IV2O2(bpy)4](PF6)3 (6c).38-40 On the basis of EPR
studies and the analysis of thermal displacement parameters
of 6c, the crystallographic equivalency of the Mn ions in 6c
has been concluded to be the result of disorder of inequiva-
lent MnIII and MnIV ions.38,39 In light of these previous results,
Like their corresponding in situ solutions, aqueous solu-
tions of the above four compounds are not stable. Within
minutes, the initial green solutions turn brown with the
formation of brown or white precipitates, depending on the
complex. With the 1:1 acetonitrile-water mixture, all four
complexes are soluble, with a particularly noticeable increase
in solubility for 1c, 1d, and 1e, although none of the resulting
four solutions are stable indefinitely. The hydrophobic
substituent groups (Cl-, Br-, Ph-, and mesityl-) on terpy
appear to increase the solubility of the corresponding
Mn(III/IV) complexes in CH3CN from that of 1b, which is
almost completely insoluble. Judging by the color of the
dissolved solutions, complex 1g is very soluble in acetoni-
trile, while the other three are only slightly soluble. The
significantly higher solubility of 1g in acetontrile versus 1e
may be owing to the difference in stacking of molecules in
crystal structure (vide infra). In contrast to the low stability
of 1g in the presence of water, a solution of 1g in dry
acetonitrile was stable for several days at room temperature
without color change. The high solubility of 1g in acetontrile
and the stability of the resulting solution affords the
opportunity to study the Mn(III/IV) complexes in non-
aqueous solutions. Previously, organic solubility of
[MnIII/IV2(µ-O)2(terpy)2(H2O)2]3+ complexes could only
be achieved by substituting the coordinated waters
of 1b with CF3COO-. The previously isolated
(36) Baffert, C.; Collomb, M. N.; Deronzier, A.; Pecaut, J.; Limburg, J.;
Crabtree, R. H.; Brudvig, G. W. Inorg. Chem. 2002, 41, 1404.
(37) Baffert, C.; Chen, H. Y.; Crabtree, R. H.; Brudvig, G. W.; Collomb,
M. N. J. Electroanal. Chem. 2001, 506, 99.
(38) Manchanda, R.; Brudvig, G. W.; Degala, S.; Crabtree, R. H. Inorg.
Chem. 1994, 33, 5157.
(39) Stebler, M.; Ludi, A.; Burgi, H. B. Inorg. Chem. 1986, 25, 4743.
(40) Plaksin, P. M.; Palenik, G. J.; Stoufer, R. C.; Mathew, M. J. Am. Chem.
Soc. 1972, 94, 2121.
[MnIII/IV2(µ-O)2(terpy)2(CF3CO2)2](CF3CO2) (5a) and [MnIII/IV
-
2
Inorganic Chemistry, Vol. 44, No. 21, 2005 7669

Chen et al.
it is likely that the equivalency of Mn ions in 1b is also due
solution. The oxidation of any of these species may lead to
undesired pathways. The current ligand library covers a range
of ligands with different electron-donating/withdrawing abil-
ity as well as ligands that restrict the rotation of the pyridine
rings of terpy (py-phen and dpya).
Crystals of MnII(dpya)Cl2, MnII(Ph-terpy)Cl2, MnII(mesityl-
terpy)Cl2, and [MnIII/IV2(O)2(mesityl-terpy)2(H2O)2](NO3)3
(1g) were isolated and characterized by X-ray crystal-
lography. As in 5b, the two Mn ions in 1g are crystallo-
graphically inequivalent; this provides structural evidence
that the di-µ-oxo-bridged Mn(III/IV) dimers are valence-
trapped. Complex 1g is a di-µ-oxo di-aqua dimanganese
complex that has good solubility in an organic solvent
(acetonitrile); this will enable further studies in acetonitrile
where the concentration of water can be varied.
to disorder of MnIII and MnIV ions.
Conclusion
In this study, we reported the general in situ synthesis of
[MnIII/IV2(O)2(L)2(X)2]3+ (1) complexes from MnIILCl2 (2)
precursors. Complexes 1 were generated and characterized
by UV-vis, EPR, ESI-MS and oxygen-evolving activity
assays. The yields and oxygen-evolving activities were found
to vary greatly depending on the particular complex. The
reasons are currently under investigation.
The preparation of this family of complexes has expanded
the limited number of known di-µ-oxo-bridged manganese
complexes that contain labile coordination sites that could
be transiently occupied by water in aqueous solution. This
type of complex is of importance because water is the
substrate of the oxygen-evolving complex of photosystem
II, where Mn-di-µ-oxo-Mn moieties are known to be part
of the active catalyst.2 In studies where synthetic compounds
were used to model this enzymatic process, water coordina-
tion on Mn was found to be the key element in successful
catalysts.7,11,12 Our simple in situ preparation of 1 enables a
fast-screening procedure for a catalytic assay for O2 evolu-
tion. The use of MnIILCl2 complexes as precursors is
advantageous, as it minimizes the presence of organic
impurities, excess ligand, or free Mn2+ in the catalytic
Acknowledgment. The authors thank the National Insti-
tutes of Health (GM32715) and the U.S. Department of
Energy (DE-FG02-84ER13297) for financial support and
Marie-Noe¨lle Collomb for communicating results prior to
publication and for helpful discussions.
Supporting Information Available: X-ray crystallographic data
of 2e, 2g, 2k, and 1g (pdf); X-ray crystallographic file for 2e, 2g,
2k, and 1g (cif). This material is available free of charge via the
Internet at http://pubs.acs.org.
IC0509940
7670 Inorganic Chemistry, Vol. 44, No. 21, 2005