Photoinduced electron transfer in Ruthenium bipyridyl complexes: Evidence for the existence of a cage with molecular oxygen

Article abstract of DOI:10.1021/jp048235a

Ruthenium complexes with three bipyridyl ligands, one of which is modified by attaching one or two hydroxamic acids groups (Ru-1 and Ru-2, respectively), were synthesized. Using EPR spectroscopy, we have found that photoexcitation leads to formation of nitroxyl radicals. The nitroxyl radical concentration in Ru-2 increased dramatically in the presence of spin traps DMPO (5,5-dimethyl-1-pyrroline-N-oxide) and PBN (N-tert-butyl-α-phenylnitrone) characterized by strong affinity to Superoxide radicals. We have attributed this behavior to the formation of a cage complex between Ru-2 and the Superoxide radical. This paper concerns the study of cages formed between ruthenium complexes and molecular oxygen and the effect of functional groups attached to modified bipyridyl ligands on cage formation. The complex between Ru-2 and O₂ was formed in the ground state, probably with participation of the hydroxamic acid groups. The equilibrium constant of this complex was determined by EPR as K_eq ～ 3 M^-1. The formation of the Ru-2-O₂ complex is supported by the temperature-dependent rate of appearance of the EPR signal in the presence of PBN. Additional evidence comes from observation of paramagnetic shifts of the peaks in the ¹H NMR spectrum of specific aromatic protons in the substituted bipyridyl ring upon exposure to O₂. Similar shifts were observed in the spectrum of Os-2, with osmium replacing ruthenium. Model compounds with functional groups that replace the hydroxamic acid or compounds without the metal center, but with the two hydroxamic acids, were synthesized. No shifts in the ¹H NMR spectra of these derivatives were observed in the presence of O₂. These results lead to the conclusion that both metal ions, Ru(II) or Os(II), and hydroxamic acid groups are essential components for the formation of the oxygen cage.

Full text of DOI:10.1021/jp048235a

9274
J. Phys. Chem. A 2004, 108, 9274-9282
Photoinduced Electron Transfer in Ruthenium Bipyridyl Complexes: Evidence for the
Existence of a Cage with Molecular Oxygen
Eylon Yavin,†,§ Lev Weiner,*,‡ Rina Arad-Yellin,† and Abraham Shanzer*,†
Department of Organic Chemistry and Department of Chemical Research Support,
Weizmann Institute of Science, RehoVot 76100, Israel
ReceiVed: April 22, 2004; In Final Form: July 30, 2004
Ruthenium complexes with three bipyridyl ligands, one of which is modified by attaching one or two
hydroxamic acids groups (Ru-1 and Ru-2, respectively), were synthesized. Using EPR spectroscopy, we
have found that photoexcitation leads to formation of nitroxyl radicals. The nitroxyl radical concentration in
Ru-2 increased dramatically in the presence of spin traps DMPO (5,5′-dimethyl-1-pyrroline-N-oxide) and
PBN (N-tert-butyl-R-phenylnitrone) characterized by strong affinity to superoxide radicals. We have attributed
this behavior to the formation of a cage complex between Ru-2 and the superoxide radical. This paper concerns
the study of cages formed between ruthenium complexes and molecular oxygen and the effect of functional
groups attached to modified bipyridyl ligands on cage formation. The complex between Ru-2 and O2 was
formed in the ground state, probably with participation of the hydroxamic acid groups. The equilibrium constant
of this complex was determined by EPR as Keq ∼ 3 M-1. The formation of the Ru-2-O2 complex is supported
by the temperature-dependent rate of appearance of the EPR signal in the presence of PBN. Additional evidence
1
comes from observation of paramagnetic shifts of the peaks in the H NMR spectrum of specific aromatic
protons in the substituted bipyridyl ring upon exposure to O2. Similar shifts were observed in the spectrum
of Os-2, with osmium replacing ruthenium. Model compounds with functional groups that replace the
hydroxamic acid or compounds without the metal center, but with the two hydroxamic acids, were synthesized.
No shifts in the 1H NMR spectra of these derivatives were observed in the presence of O2. These results lead
to the conclusion that both metal ions, Ru(II) or Os(II), and hydroxamic acid groups are essential components
for the formation of the oxygen cage.
Introduction
Hydroxamic acids are bidentate ligands found in many natural
products, especially in those that bind ferric ions. In neutral
solutions, the hydroxamic acid can be oxidized by hydroxyl and
superoxide radicals that yield nitroxyl radicals.30,31
Many efforts have been made to use the powerful oxidation
potential of photogenerated ruthenium(III) polypyridyl com-
plexes to catalyze oxidation of water,1-7 DNA, and RNA.8-15
Ruthenium complexes were also incorporated into specific sites
of proteins and nucleic acids and were used to study photoin-
duced electron-transfer processes in these biopolymeric sys-
tems.8,16-28
The synthesis and photochemical behavior of a ruthenium
tris bipyridine complex where one of its bipyridyl ligands was
modified with two arms, each bearing a hydroxamic acid group
(Ru-2, Scheme 1C),32 were reported previously. The electron-
and energy-transfer processes in the presence of molecular
oxygen were detected by EPR, following the kinetics of the
formation of nitroxyl radicals formed on the hydroxamic acid
side chain. In the absence of O2, nitroxyl radical did not form,
whereas in the presence of superoxide dismutase (SOD) radical
formation was enhanced.32 Thus, we have suggested that Ru-2
forms a genuine cage complex with a superoxide radical, which
can be released by adding molecules such as PBN and DMPO
with high affinity for superoxide radicals and that act as spin
traps. Consequently, the electron-transfer process, that is
otherwise dormant, is activated and leads to Ru(III) generation
(Scheme 1A).
Quenching the excited states of ruthenium tris bipyridine
complexes by molecular oxygen leads to either an energy-
transfer process that yields singlet oxygen and the Ru(II) ion
in its ground state (Scheme 1B) or an electron-transfer process
that yields superoxide radicals and Ru(III) ions (Scheme 1A).
Using transient absorption spectroscopy of Ru(II), Zhang and
Rodgers29 have suggested that a ‘cage’ complex consisting of
Ru(III) and superoxide radical is formed when the electron
transfer is the process of choice. For neutral pH, the ‘cage’
complex is not liberated; that is, the electron-transfer process
has a quantum yield of zero. In acidic conditions (3 N D2SO4),
the superoxide radical is released from the cage and Ru(III) is
formed, with a quantum yield of 0.55.29
To assess the role of the different components in Ru-2’s
affinity for molecular oxygen, we synthesized an analogous
ruthenium complex Ru-1 (Scheme 2) that has a single hydrox-
amic acid group on one of the bipyridyl ligands, and compared
the nitroxyl radicals’ formation by Ru-2 and Ru-1. Two
hydroxamic acid groups were indispensable for stabilization of
the cage complex formed between ruthenium in its trivalent
* To whom correspondence should be addressed: abraham.shanzer@
weizmann.ac.il; Phone: 972-8-9343954; Fax: 972-8-9432917. lev.weiner@
weizmann.ac.il; Phone: 972-8-9343410; Fax: 972-8-9346017.
§ California Institute of Technology, Department of Chemistry &
Chemical Eng., Pasadena, CA 91125.
† Department of Organic Chemistry, Weizmann Institute of Science.
‡ Department of Chemical Research Support, Weizmann Institute of
Science.
oxidation state and superoxide radical. The affinity for O2 (Kcage
)
was estimated from the EPR data.
10.1021/jp048235a CCC: $27.50 © 2004 American Chemical Society
Published on Web 09/18/2004

Electron Transfer in Ruthenium Bipyridyl Complexes
J. Phys. Chem. A, Vol. 108, No. 42, 2004 9275
SCHEME 1: Photoinduced Electron (A) and Energy (B) Transfer Processes in the Ru-2 Complex (C) in the Presence of
O2
SCHEME 2: Synthesis of Ru-1
Oxygen is a paramagnetic molecule that can induce chemical
shifts in the NMR signals of magnetically active atoms that are
in close contact with it33 and is highly solvable in hydrophobic
solvents. In general,34 and in the interior of cell membranes in
particular, this technique was used for probing the structure of
certain membrane proteins.35 We used NMR for tracking the
interactions between Ru-2 and molecular oxygen and observed
changes in chemical shifts of selective protons of the modified
bipyridyl ring in the presence of O2. This fact strengthened the
conclusion based on the EPR data. Model compounds that lack
the metal center and either one or two of the hydroxamic acid
groups did not reveal any significant shifts in the H NMR
signals in the presence of O2. This observation suggests that
the metal center and hydroxamic acids groups can be crucial
for the formation of the cage. The EPR and NMR data and the
study of various model compounds taken together, we suggest
that an oxygen cage is formed between Ru-2 and molecular
oxygen (Figure 1).
1
Experimental Section
1
Instrumentation. The H NMR spectra were measured on
an Avance DPX-400 MHz or DPX-250 MHz spectrometer
(Bruker) using the solvent deuterium signal as an internal
reference. All J values are given in Hertz. The IR spectra were
recorded on a Prote´ge´ 460 FTIR spectrometer. The UV-vis
spectra were measured with a Hewlett-Packard model 8450A
diode array spectrophotometer. MS-ES was performed at the
Weizmann Institute (Rehovot). Flash chromatography was
performed using Merck 230-400 mesh silica gel. Thin layer
chromatography (TLC) on 60 F-254 silica gel was visualized
by UV light and by one or more of the following reagents:
ninhydrin, basic KMnO4 solution, or iodine, or by FeCl3 in
MeOH. The X-band EPR spectra were recorded on an Electron
Figure 1. Schematic view (chem3D) of the cage formed between Ru-2
and O2 (red ) oxygen, blue ) nitrogen, gray ) carbon, and light blue
) hydrogen).

9276 J. Phys. Chem. A, Vol. 108, No. 42, 2004
Yavin et al.
Spin Resonance ER 200D-SRC spectrometer (Bruker) at room
temperature and at elevated temperatures.
cm-1 (CONO). UV λmax (ꢀ) ) 289 (47 200) and 477 (8800)
nm. MS-ES m/z ) 851 [M-H]1+
.
Chemicals. Chemicals and reagents were purchased from
Sigma. The cis-dichloro bis(2,2′-bipyridine) ruthenium(II) di-
hydrate was purchased from STREM chemicals. Dichloro-
methane (DCM) was dried by passing the solvent through a
basic alumina column. Tetrahydrofuran (THF) was distilled from
Na under argon. Double distilled water and spectroscopic grade
CH3CN were used for EPR experiments.
The spin traps (DMPO (5,5′-dimethyl-1-pyrroline-N-oxide)
and PBN (N-tert-butyl-R-phenylnitrone)) were purchased from
Sigma and purified as described before.32
EPR Experiments. The EPR experiments were carried out
in CH3CN/H2O (93:7) solutions. A quartz flat cell of 70 µL
was used in all experiments when the recording was ac-
companied by illumination. A 150 W lamp (Schott model KL
1500 LCD) adjusted with different filters was used as a light
source. The ruthenium complexes were illuminated using a blue
filter (380-500 nm). The osmium complex was illuminated
without a filter. The temperature was adjusted by a temperature
unit control (Eurotherm, B-VT 2000) with accuracy of (1 K.
The stable nitroxyl radical TEMPO was used as a standard.
Double integration of EPR peaks was used for estimating the
nitroxyl radical concentration.
The ruthenium complex with protected hydroxamic acids 3
was prepared by refluxing an 80% ethanolic solution of 2 (230
mg, 0.577 mmol) and Ru(bipy)2Cl2‚6H2O (355 mg, 0.60 mmol)
for 4 h under argon. The solvent was removed, and the product
was purified by column chromatography eluting with a solution
containing CH3CN/n-BuOH/0.4 M KNO3 (9:0.5:0.5). The Ru
complex 3 (308 mg) was obtained in 57% yield.
1H NMR (250 MHz, MeOH-d4) δ ) 9.08 (s, 1H, s-bipy 3),
8.64-8.71 (ov, 5H, bipy 3,3′ and s-bipy 3′), 8.12 (t, J ) 8 Hz,
4H, bipy 4,4′), 7.95 (d, J ) 6 Hz, 1H, s-bipy 6), 7.78-7.85
(ov, 5H, bipy 6,6′ and s-bipy 5), 7.64 (d, J ) 6 Hz, 1H, s-bipy
6′), 7.48 (m, 4H, bipy 5,5′), 7.37 (m, 1H, s-bipy 5′), 5.35 (br,
1H, THP, OCH), 4.95 (q, J ) 7 Hz, 1H, NHCH), 4.06 and
3.66 (m, 2H of THP, OCH2), 3.32 and 3.38 (s, 3H, NCH3),
2.58 and 2.60 (s, 3H, bipy-CH3), 1.65-1.84 (ov, 6H, THP,
OCHCH2CH2CH2), 1.50 (d, J ) 8 Hz, 3H, NHCHCH3). IR
(KBr) ν ) 1640 cm-1 (CONO).
Os(bipy)2Cl2. The Os(bipy)2Cl2 was prepared according to
literature protocol38 and was refluxed with 1 in ethanolic solution
to yield compound Os-1, which was converted to Os-2 as
described for Ru-2.32
1
Os-1. H NMR (250 MHz, MeOH-d4) δ ) 9.16 (m, 2H,
The abbreviations for the NMR spectra are as follows: bipy
) bipyridyl, s-bipy ) substituted bipyridyl, ov ) overlapping
proton peaks, d ) doublet, s ) singlet, and t ) triplet.
The synthetic protocol of Ru-2 was already published.32 The
Ru-1 was synthesized starting from 2,2′-bipyridyl-4-methyl-4′-
carboxylic acid36 following a similar protocol as described in
Scheme 2. The acid was suspended in SOCl2 and refluxed for
5 h, followed by removal of the solvent. The acyl chloride was
dissolved in dry THF, and triethylamine (415 µL, 3 mmol) was
added followed by 2-amino-N-methyl-N-(tetrahydropyran)-
propioxamic acid (1)37 (2 mmol) dissolved in THF. The pH
was adjusted to 8 by triethylamine, and the reaction mixture
was left to stir overnight at room temperature. The solvent was
removed, and the crude material was purified by column
chromatography, using a mixture of ethyl acetate/MeOH/25%
ammonia solution (97.5:2:0.5) as eluent. The product 2 (310
mg, 0.78 mmol) was obtained in 65% yield.
1H NMR (250 MHz, CDCl3) δ ) 8.78 (d, J ) 2.5 Hz, 1H,
bipy 3), 8.67 (m, 1H, bipy 3), 8.53 (m, 1H, bipy 6), 8.22 (m,
1H, bipy 5), 7.73 (m, 1H, bipy 6′), 7.42 (br, 1H, NH), 7.15 (m,
1H, bipy 5′), 5.15-5.8 (br, 1H of THP CHO), 5.10 (m, 1H,
NHCHCH3), 4.05 and 3.65 (m, 2H of THP OCH2), 3.40 (s,
3H, NCH3), 2.43 (s, 3H, bipy-CH3), 1.84 and 1.66 (br, 6H, THP
CHCH2CH2CH2), 1.48 (m, 3H, CHCH3). IR (CHCl3): ν ) 1640
cm-1 (CONO).
The Ru-1 was obtained by removal of the protecting group
from 3. It was done by heating 3 (308 mg, 0.33 mmol) for 1 h
to 60 °C in a 50% acetic acid solution. The solvent was
evaporated; the residue was dissolved in a minimum amount
of MeOH and dropped into a cold ether solution. Complete
precipitation was obtained after the solution had been left in
the refrigerator for the night. It was filtered and washed with
diethyl ether. Ru-1 (271 mg) was obtained in 97% yield.
1H NMR (400 MHz, MeOH-d4) δ ) 9.07 (s, 1H, s-bipy 3),
8.66-8.69 (ov, 5H, bipy 3,3′ and s-bipy 3′), 8.11 (m, 4H, bipy
4,4′), 7.93 (d, J ) 6 Hz, 1H, s-bipy 6), 7.78-7.83 (ov, 5H,
bipy 6,6′ and s-bipy 5), 7.63 (d, J ) 6 Hz, 1H, s-bipy 6′), 7.48
(m, 4H, bipy 5,5′), 7.36 (m, 1H, s-bipy 5′), 5.18 (q, J ) 7 Hz,
1H, NHCH), 3.23 (s, 3H, NCH3), 2.59 (s, 3H, bipy-CH3), 1.47
(dd, J ) 7.2, 1.6 Hz, 3H, NHCHCH3). IR (KBr): ν ) 1635
s-bipy 3,3′), 8.68 (m, 4H, bipy 3,3′), 8.00-7.92 (m, 6H, bipy
4,4′ and s-bipy 5,5′), 7.71 (m, 6H, bipy 6,6′ and s-bipy 6,6′),
7.45 (m, 4H, bipy 5,5′), 5.34 (br, 2H, THP), 4.96 and 5.05 (br,
2H, NHCH×2), 4.00 and 3.67 (m, 4H of THP), 3.34-3.36 (ov,
6H, NCH3×2), 1.67-1.82 (ov, 12H, THP), 1.44 (m, 6H,
CHCH3×2). IR (KBr): ν 1640 cm-1 (CONO). MS-ES m/z )
558.6 [M]2+
.
Os-2. 1H NMR (250 MHz, MeOH-d4) δ ) 9.19 (s, 2H, s-bipy
3,3′), 8.70 (m, 4H, bipy 3,3′), 7.95 (m, 6H, bipy 4,4′ and s-bipy
5,5′), 7.70 (m, 6H, bipy 6,6′ and s-bipy 6,6′), 7.41 (m, 4H, bipy
5,5′), 5.18 (m, 2H, NHCH×2), 3.23 (s, 6H, NCH3×2), 1.47 (d,
J ) 7 Hz, 6H, CHCH3×2). IR (KBr): ν ) 1635 cm-1 (CONO).
UV λmax (ꢀ) ) 290 (20 885) and 493 (4350) nm. MS-ES m/z )
474.5 [M]2+
.
Phen-1. The 4-carbomethoxy-benzyl bromide (540 mg, 2.35
mmol), prepared from the acid,39 and triphenylphosphine (667
mg, 2.58 mmol) were dissolved in toluene (8 mL) under an
inert atmosphere (argon), and the solution was warmed to 80
°C for 5 h. After cooling to RT the solution was cooled in an
ice bath and the resulting precipitate was filtered, washed with
toluene, and dried under high vacuum to afford 4-carbomethoxy-
benzyltriphenylphosphnium bromide (828 mg, 72% yield) as a
white solid. To this were added methyl-4-formylbenzoate (2.02
mmol) and sodium methoxide40 to yield the corresponding
stilbene derivative. The trans-stilbene isomer (200 mg) pre-
cipitated by cooling, and the cis isomer (100 mg) was recovered
from the mother liqueur and was purified by column chroma-
tography. The yield of both isomers was 60%.
Irradiation of a benzene solution of a mixture of cis- and
trans-dimethyl 4, 4′-stilbene-dicarboxylate (300 mg) yielded
dimethyl 3,6-phenanthrenedicarboxylate41 (163 mg, 0.55 mmol
in 55% yield). It was further hydrolyzed to the corresponding
3, 6-phenanthrene dicarboxylic acid42 and was further converted
to Phen-1 in a protocol similar to that described for Ru-2.32
Purification was done by column chromatography using 0-4%
MeOH in CHCl3 as eluent. The product Phen-1 (139 mg) was
obtained in 86% yield.
1H NMR (250 MHz, CDCl3) δ ) 9.37 (s, 2H), 8.08 (m, 2H),
7.92 (m, 2H), 7.81 (m, 2H), 7.59-7.65 (br, 2H, NH×2), 5.56
and 5.08 (br, 2H, THP), 5.26 (m, 2H, CHCH3×2), 4.08 and

Electron Transfer in Ruthenium Bipyridyl Complexes
J. Phys. Chem. A, Vol. 108, No. 42, 2004 9277
SCHEME 3: Synthesis of Ruthenium Complexes Ru-3,
Ru-4, and Ru-5
Compound 6 was purified by column chromatography with
MeOH/CHCl3 (5:95) as eluent and was obtained (290 mg, 0.58
mmol) in quantitative yield.
1H NMR (250 MHz, CDCl3) δ ) 8.77 (dd, J ) 5 and 0.7
Hz, 2H, bipy 5,5′), 8.67 (m, 2H, bipy 3,3′), 7.74 (dd, J ) 5 and
1.7 Hz, 2H, bipy 3,3′) 7.13 (m, 2H, NH×2), 3.73 (dt, J ) 6 Hz
and 6 Hz, 4H, NHCH2CH2×2), 2.60 (t, J ) 6 Hz, 4H,
NHCH2CH2×2), 1.48 (s, 18H, C(CH3)3×2).
The ruthenium complexes were obtained by stirring the 2,
2′-bipyridyl derivatives 4-6 with equivalent amount of Ru-
(bipy)2Cl2‚6H2O in 80% ethanolic solution for 4 h under argon
followed by removal of the solvent and purification by column
chromatography eluting with a CH3CN/n-BuOH/0.4 M KNO3
(9:0.5:0.5) solution.
t
The butyl ester groups were removed by stirring in TFA/
DCM (1:4) solutions for 1-2 h followed by evaporation of the
solvents. Ru-4 and Ru-5 were obtained in quantitative yields.
Complex Ru-3 (40 mg, 0.042 mmol) was obtained in 38% yield.
Ru-3. 1H NMR (250 MHz, MeOH-d4) δ ) 9.21 (s, 2H, s-bipy
3,3′), 8.73 (d, J ) 7.8 Hz, 4H, bipy 3,3′), 8.12-8.20 (m, 4H,
bipy 4,4′), 8.02 (d, J ) 5.8 Hz, 2H, s-bipy 5,5′), 7.82-7.91
(ov, 4H of bipy 6,6′ and 2H of s-bipy 6,6′), 7.48 (m, 4H, bipy
5,5′), 4.62-4.71 (m, 2H, CHCH3×2), 3.75 (s, 6H, OCH3×2),
1.55 (d, J ) 7.4 Hz, 6H, CHCH3×2).
3.66 (m, 4H, THP), 3.46 and 3.40 (s, 6H, NCH3×2), 2.28-
1.66 (br, 12H, THP), 1.55 (m, 6H, CHCH3×2).
The Phen-2 was obtained from Phen-1 after removal of the
THP protecting group by heating for 2 h at 60 °C in an AcOH/
H2O/THF (2:1:1) solution.43 It was obtained in quantitative
yield.
1H NMR (250 MHz, CDCl3) δ ) 9.12 (s, 2H), 8.48 (m, 2H,
NH×2), 7.95 (d, 2H, J ) 8.1 Hz), 7.69 (d, 2H, 8.1 Hz), 7.55
(s, 2H), 5.34 (m, 2H, CHCH3×2), 3.33 (s, 6H, NCH3×2), 1.56
(d, 6H, J ) 7 Hz). MS-ES [M + Na]+ ) 489.5.
Ru-3, Ru-4, and Ru-5 (4-6) were prepared by coupling the
corresponding amino derivatives with 2,2′-bipyridyl-4,4′-dicar-
boxylic acid chlorides.32
Ru-4. 1H NMR (250 MHz, MeOH-d4) δ ) 9.19 (s, 2H, s-bipy
3,3′), 8.73 (d, J ) 8.2 Hz, 4H, bipy 3,3′), 8.11-8.19 (m, 4H,
bipy 4,4′), 8.01 (d, J ) 5.8 Hz, 2H, s-bipy 5,5′), 7.81-7.89
(ov, 4H of bipy 6,6′ and 2H of s-bipy 6,6′), 7.46-7.54 (m, 4H,
bipy 5,5′), 4.59-4.68 (m, 2H, CHCH3×2), 1.56 (d, J ) 7.3
Hz, 6H, CHCH3×2).
Ru-5. 1H NMR (250 MHz, MeOH-d4) δ ) 9.10 (s, 2H, s-bipy
3,3′), 8.71 (d, J ) 8.0 Hz, 4H, bipy 3,3′), 8.10-8.18 (m, 4H,
bipy 4,4′), 7.98 (d, J ) 6.0 Hz, 2H, s-bipy 5,5′), 7.79-7.83
(ov, 6H, bipy 6,6′ and s-bipy 6,6′), 7.44-7.53 (m, 4H, bipy
5,5′) 3.67 (t, J ) 6.6 Hz, 4H, NHCH2CH2×2), 2.65 (t, J ) 6.6
Hz, 4H, NHCH2CH2×2).
Compound 4 was purified by column chromatography with
MeOH/CHCl3 (4:96) as eluent and was obtained (124 mg, 0.318
mmol) in 24% yield.
Results and Discussion
1. Synthesis. Ru-1. The metal complex, Ru-1, and all other
ruthenium complexes in this study were prepared according to
Schemes 2 and 3. The synthesis of the organic analogue (Phen-
2) is shown in Scheme 4.
2. How Does the Hydroxamic Acid Contribute to the
Stability of the Cage Complex? Comparison between Ru-1
and Ru-2. Irradiation of Ru-1 and Ru-2 in CH3CN/H2O
solutions by visible light (λmax ) 450 nm) leads to the formation
of radicals with a g factor of 2.0061 (Figure 2). The 12-line
EPR signal, corresponding to the nitroxyl radical (aN ) 6.83 G
and aH ) 7.87 G), is similar to that observed for the radical of
N-methyl-N-acetyl-hydroxylamine.44
1H NMR (250 MHz, CDCl3/MeOH-d4) δ ) 8.72 (dd, J ) 5
and 0.6 Hz, 2H, bipy 5,5′), 8.62 (d, J ) 0.6 Hz, 2H, bipy 3,3′),
7.97-8.00 (m, 2H, NH×2), 7.75 (dd, J ) 5 and 1.7 Hz, 2H,
bipy 3,3′), 4.69-4.75 (m, 2H, CHCH3×2), 3.73 (s, 6H,
OCH3×2), 1.50 (d, J ) 6.5 Hz, 6H, CHCH3×2).
Compound 5 was purified by column chromatography with
MeOH/CHCl3 (6:94) as eluent and was obtained (300 mg, 0.602
mmol) in 98% yield.
1H NMR (250 MHz, CDCl3) δ 8.77 (dd, J ) 5 and 0.6 Hz,
2H, bipy 5,5′), 8.67 (m, 2H, bipy 3,3′), 7.74 (dd, J ) 5 and 1.7
Hz, 2H, bipy 3,3′) 7.24-7.27 (m, 2H, NH×2), 4.65-4.77 (m,
2H, CHCH3×2), 1.50 and 1.53 (ov, 24H, CHCH3×2 and
C(CH3)3×2).
The kinetics of the radical formation was found to be very
modest, resulting from oxidation of the hydroxamic acid group
SCHEME 4: Synthesis of Phen-2

9278 J. Phys. Chem. A, Vol. 108, No. 42, 2004
Yavin et al.
SCHEME 5: Light-Dependent Mechanisms for the
Generation of Nitroxyl Radicals in the Presence of PBN
Figure 2. EPR spectrum of the nitroxyl radical formed on irradiation
of Ru-1 and Ru-2.
3. Spectroscopic Evidence Supporting Cage Formation
with Molecular Oxygen. 3a. EPR Kinetics of Nitroxyl Radical
Formation as a Function of Temperature. As can be seen from
the system of eqs 1-4 describing the kinetics of nitroxyl radical
generation under photoexcitation of our Ru(II)-2 complex, there
are a few temperature-dependent steps.
The effect of temperature on the direct and back electron-
transfer reactions requires a separate consideration. The electron-
transfer rate K in our system is described by the Marcus form-
ula:48
ln(K/K0) ) (∆Gi + λ)2/4λkT
(5)
Figure 3. Formation of nitroxyl radicals in 1 mM solutions of Ru-1
and Ru-2 in the presence of 50 mM PBN. The EPR signal was fixed
at 3273 G (doublet) and recorded with a receiver gain of 8 × 104 and
modulation amplitude of 1 G.
where ∆Gi is the Gibbs free-energy change in the overall
electron transfer (e-T) reaction and λ is the reorganization
energy. For known Ru complexes, the electron-transfer reaction
between Ru(II)* and O2 has ∆G(e-T) ranging from -0.04 to
-0.72 eV.29 On the other hand, ∆Gback(e-T) is in the range of
-1.78 to -2.46 eV (for estimation of ∆Gback(e-T) we used an
energy of light for excitation of our Ru(II)-2 ) 2.5 eV). From
the classical parabolic Marcus dependence of ln K versus ∆Gi
(see, for example, refs 48 and 49), it follows that the rate of
direct electron-transfer reaction is 102 to 103 orders of magnitude
higher than the back electron-transfer reaction, which means
that the reaction shifts significantly to the right side (see Scheme
5, eqs 1 and 2).
The bimolecular reaction between the ‘trapped’ superoxide
radical and PBN (Scheme 5, eq 3) is expected to speed up with
temperature and to be accompanied by an increase in concentra-
tion of nitroxyl radicals.50 However, taking into account
temperature dependence of electron transfer and back electron-
transfer reactions, such behavior should describe a system where
no equilibria between Ru(II)-2 and O2 are formed (eqs 1 and
2). When these equilibria and cage complexes are involved,
elevation of temperature is expected to cause a decrease in Keq.
As a result, concentrations of ‘bound’ oxygen, superoxide
radicals, and Ru(III) ions should decrease together with the
concentration of nitroxyl radicals.
The EPR spectra of ruthenium complexes Ru-2 and Ru-1
were recorded at different temperatures in the presence (Figure
4) and in the absence (Figure 5) of the spin trap PBN. The
decrease in the initial rate of formation of the nitroxyl radicals
with increasing temperature for both complexes supports the
existence of cage complexes (Figure 4).
Oxygen solubility, as with other gases, decreases with
increasing temperature.51 For the known values of oxygen
concentration within the temperature range under study, a
maximum effect could be 1.27. At the same time, an experi-
mentally observed decrease of nitroxyl radical generation rate
was 1.72-2. Note that an increase in 1O2 generation rate (Figure
5) with increasing temperature (and decreasing O2 concentration
in solution!) also indicates that concentration of oxygen is
insignificant for the kinetic behavior observed (Figure 4).
by singlet oxygen, produced by energy transfer from Ru(II)*
to molecular oxygen.32 Irradiating the solution of Ru-2 in the
presence of a spin trap, that is, PBN and DMPO, revealed a
dramatic enhancement in the concentration of nitroxyl radicals
(up to 30 times). Addition of superoxide dismutase also
stimulated the generation of nitroxyl radical, but less extensively
than spin traps.32 Removal of oxygen by argon bubbling
inhibited nitroxyl radical formation.32
This observation was attributed to the high reactivity of
Ru(III) ions, with a high redox potential {E0(Ru2+/3+) ) 1.29
V},45,46 which is responsible for the formation of the nitroxyl
radicals. The Ru(III) ion, produced in the irradiated solution,
is trapped in a cage complex with a superoxide radical
[Ru(III)‚‚‚O2-•]. A reverse electron-transfer reaction, Ru(III) +
-•
O2 f Ru(II) +O2, could play a role in the equilibrium. The
spin traps liberate the superoxide radical from the cage complex
and facilitate further reaction of Ru(III) with the hydroxamic
acid (See Scheme 1A).
To evaluate the effect of the two hydroxamic acid groups on
stability of the proposed cage complex, a second ruthenium
complex (Ru-1), with a single hydroxamic acid group on the
bipyridyl ring, was synthesized (Scheme 2). It was expected
that a weaker cage would be obtained with this derivative and
that the effect of the spin trap would be less pronounced. From
the EPR study, it was clear that for both complexes (Ru-2 and
Ru-1) the formation of the nitroxyl radical was accelerated by
spin trap molecules. However, the formation rate of the nitroxyl
radical was faster in Ru-2 than in Ru-1 solution (Figure 3).
Nitroxyl radical concentration decreases after several minutes
of irradiation of both compounds, probably because of the light-
induced photooxidation of the nitroxyl radical to the diamagnetic
N-oxo-ammonium ion by Ru(III).47
The proposed reaction mechanism is described in Scheme 5,
where Keq(1) ) Keq(2) ) K1/K-1, assuming that the complex in
its excited state has the same affinity for O2 as in its ground
state.

Electron Transfer in Ruthenium Bipyridyl Complexes
J. Phys. Chem. A, Vol. 108, No. 42, 2004 9279
Figure 4. Effect of temperature on the rate of nitroxyl radical formation in 1 mM solutions of Ru-2 (A) and of Ru-1 (B) in the presence of 50 mM
PBN. The EPR signal was fixed at 3170.1 G (triplet) and recorded with a receiver gain of 5 × 104 and a modulation amplitude of 2.2 G.
molecules are based only on collisions, nonselective shifts will
occur. In fact, we have observed a downfield shift of ∆δ ) 5.6
Hz of only two protons attached to one of the bipyridyl ligands
(i.e., the peaks of the protons at positions 3 and 3′ (Figure 6)).
Similar behavior was observed in the 1H NMR spectrum of Os-2
in the presence of O2 (data not shown).
To assess whether both hydroxamic acid groups are required
for the formation of the cage, we have looked for the formation
of a cage between Ru-1 and O2 by performing a similar NMR
experiment. A less pronounced downfield shift of ∆δ ) 1.6
Hz for H3 (Figure 7A) and no shift at all for H3′ were obtained
for the system with only one hydroxamic acid attached to the
bipyridyl ring. (Figure 7B)
The conclusion from the experiment is that though a cage is
formed when only one hydroxamic acid is present, both
hydroxamic acid groups are required for the formation of a stable
Figure 5. Effect of temperature on the rate of nitroxyl radical formation
cage with oxygen.
in a 1 mM solution of Ru-2. The EPR signal was fixed at 3164.3 G
A series of model complexes (Ru-3, Ru-4, and Ru-5) in
(doublet) and recorded with a receiver gain of 4 × 105 and a modulation
which the side chains are terminated with carboxylic acid (Ru-4
amplitude of 2.0 G.
and Ru-5) or ester (Ru-3) functional group was prepared. The
length of the side chain was also extended by replacing an
L-alanine amino acid (Ru-4) with a â-alanine (Ru-5). In all
cases, no shifts in the peaks of the protons in the 3, 3′-bipyridyl
positions were observed in the 1H NMR spectra in the presence
of oxygen, indicating no detectable affinity for O2. Other model
compounds that lack a metal center, including a free ligand and
a phenanthroline derivative (Phen-2) with two hydroxamic acid
groups attached to its skeleton with a similar mutual orientation
as in Ru-2, were also prepared and studied. No shifts in the 1H
NMR spectra of these compounds were observed in the presence
of O2. These observations imply that the metal (Ru(II) or
Os(II)) and the hydroxamic acid group are essential prerequisites
for cage formation.
We want to emphasize that the paramagnetic shift effects of
oxygen depend on the lifetime of the complex. For zeolites,
where the oxygen is absorbed within the cavities, the effect was
achieved by increasing the oxygen concentration and decreasing
the sample’s temperature.52-56 The selective low field shifts of
the 3, 3′ protons in Ru-2 are related to the relatively long
lifetime of the ‘caged’ oxygen. According to our knowledge,
the cage described in the present work is the first observation
of a complex between a metal complex and molecular oxygen
that is not chemically coordinated via the metal center, in
solution at room temperature.
In the absence of spin traps, the nitroxyl radicals result from
collisions between molecules of singlet oxygen (formed via
energy transfer from Ru(II)* to molecular oxygen (Scheme 1B))
and the hydroxamic acid group. Since in this type of bimolecular
reaction the collisions between the singlet oxygen molecule and
the hydroxamic acid are essential, raising the temperature should
increase the reaction rate. This type of reactivity was indeed
observed when a solution of Ru-2 was irradiated at elevated
temperatures (Figure 5).
3b. Effect of Oxygen on the Chemical Shifts of Specific
1
Protons in the H NMR Spectra of Ru-2 and Ru-1. Molecular
oxygen is a paramagnetic molecule and therefore is expected
to induce paramagnetic shifts in the NMR signals of ‘magneti-
cally active’ atoms and ions that are in its close vicinity.
Recently, O2 has been used as a shift reagent in solid state NMR
spectroscopy to probe accessibility of caption sites in zeolites.52-56
Also, O2 has been successfully applied as a paramagnetic
reporter of membrane protein topology, since it is inhomoge-
neously distributed in biological membranes and thus gives rise
to an exquisite range of position-dependent paramagnetic
effects.35 We have used this property to confirm the existence
of a cage structure between Ru-2 and the oxygen molecule in
1
acetonitrile by recording the H NMR spectra of the complex
in oxygen, air, and argon atmospheres. We expected to observe
changes in the chemical shifts for those protons that are in close
contact with the O2 molecule trapped in the cage. It is important
to emphasize that if interactions between Ru-2 and the oxygen
3c. Estimation of Keq for Ru(II) + O2 S Ru(II)-O2. The EPR
and NMR data indicate that Ru-2 can produce a complex with
molecular oxygen and the superoxide radical. To estimate the

9280 J. Phys. Chem. A, Vol. 108, No. 42, 2004
Yavin et al.
1
Figure 6. Peaks of protons 3 and 3′ in the H NMR spectrum of Ru-2 in O2, air, and argon atmospheres. The sample was dissolved in CD3CN
and the spectrum was recorded at 400 MHz.
1
Figure 7. Peaks of protons 3 (A) and 3′ (B) in the H NMR spectrum of Ru-1 in O2, air, and argon atmospheres, respectively. The sample was
dissolved in CD3CN and the spectrum was recorded at 400 MHz.
-•
equilibrium constant for Ru-2 with molecular oxygen, we
measured the rates of nitroxyl radical production for different
concentrations of ruthenium complex (see Figure 8). It is clearly
seen (see Figure 9) that dR/dt (rate of nitroxyl radical generation)
as a function of Ru-2 concentration follows eq 6:
106 M-1 s-1, and the reaction rate of O2 with DMPO was
calculated as K2 ) 10 M-1 s-1 (see ref 57). We used this value
in our calculations as the K2 value for the reaction of PBN and
superoxide, as there is no published value for PBN. As all other
parameters in eq 6 are known, we have inserted them in the
equation and estimated the value of Keq to be 3.1 M-1. A detailed
calculation of Keq is available as Supporting Information.
tan R ) K2 × Keq(2) × [O2]0 × [PBN]
(6)
4. Comparing the Photoinduced Processes in Ru-2 and
in Os-2. The similarity between Ru(II) and Os(II) in photo-
chemical behavior and redox properties led us to compare
osmium to ruthenium complexes. The Os(II) tris-polypyridyl
complexes have similar redox potentials to those of the Ru(II)
complexes, but their excited-state lifetime is shorter.58,59 There-
fore, it was expected that the energy and electron-transfer
Here, K2 is a bimolecular rate constant between the cage
complex and PBN, Keq(2) is the equilibrium constant for Ru-2
and molecular oxygen, [O2]0 is the concentration of molecular
oxygen in solution (in our case ∼0.5 mM), and [PBN] is the
concentration of the spin trap in solution (0.05 M).
Greenstock et al.50 have estimated the bimolecular rate
-•
constant for the reaction of PBN with O2 to be smaller than

Electron Transfer in Ruthenium Bipyridyl Complexes
J. Phys. Chem. A, Vol. 108, No. 42, 2004 9281
Figure 8. Rates of nitroxyl radical formation of Ru-2 in the presence
of 50 mM PBN for different concentrations of the complex. The EPR
signal was fixed at 3272.1 G (doublet) and recorded with a receiver
gain of 5 × 104 and modulation amplitude of 1.0 G.
Figure 11. Rate of nitroxyl radical formation in solutions (1 mM) of
Os-2 with/without PBN (50 mM). The EPR signal was fixed at 3182.2
G (doublet) and recorded with a receiver gain of 2 × 105 and a
modulation amplitude of 1.0 G.
Figure 9. Calculation of Keq from the Ru-2 concentration dependence
on nitroxyl radical formation.
Figure 12. Rate of nitroxyl radical formation in solutions (1 mM) of
Os-2 and Ru-2 in the presence of PBN (50 mM). The EPR signal was
fixed at 3182.2 G (doublet) and recorded with a receiver gain of 8 ×
104 and a modulation amplitude of 1.0 G.
In an attempt to trace the origin of this unique behavior, we
prepared a series of model compounds and studied them in the
presence and absence of molecular oxygen.
The NMR data indicate that carboxylic acids in the Ru-4
and Ru-5 compounds are not effective in forming the cage
complex with molecular oxygen. This may result from the
tendency of carboxylic acids (in close proximity) to readily form
intramolecular hydrogen bonding, thereby excluding their bind-
ing to O2.
To our surprise, the organic analogue (Phen-2), whose ligand
structure is similar to that of Ru-2, had no detectable affinity
for O2. This may be related to subtle differences in the aromatic
ring geometry. We believe, however, that the most likely reason
is stabilization of the cage complex by the positive metal ion
center that does not exist in this organic analogue. The detailed
nature of these interactions is not completely clear. Research
this direction is under way.
Figure 10. Rate of nitroxyl radical formation in 2 mM solutions of
Os-2 and Ru-2. The EPR signal was fixed at 3182.2 G (doublet) and
recorded with a receiver gain of 4 × 105 and a modulation amplitude
of 1.0 G.
processes would be less effective for Os(II) than for Ru(II)
complexes.
In fact, irradiation of Os-2 leads to the formation of the
nitroxyl radicals, with a similar EPR spectrum to that shown in
Figure 2. A comparison of the kinetics of nitroxyl radicals
formation in Os-2 and Ru-2 solutions shows that the Os-2 is
less effective than Ru-2. (Figure 10).
The effect of PBN on the rate of formation of nitroxyl radical
in Os-2 was also very modest (Figure 11) and less pronounced
than that observed for Ru-2 (Figure 12), which is in agreement
with the shorter lifetime of the Os excited state. In addition,
one cannot rule out the possibility that the geometry of Os-2 is
less favorable for binding molecular oxygen as compared to
that of Ru-2 or that there are other photophysical reasons for
the differences observed. Of crucial importance for us is the
fact that different photoactive metal complexes can produce
long-lived complexes (cage) with molecular oxygen.
Thus, using NMR we were able to show that the origin of a
unique behavior of ruthenium and osmium hydroxamate com-
plexes could be attributed to their ability to interact with free
oxygen, even in the ground state (triplet state). As a result, a
genuine cage complex can be formed upon irradiation.
Acknowledgment. The authors thank Dr. Leonid Konstan-
tinovski for his helpful advice in conducting NMR experiments
and Prof. Anatoly Burshtein for useful discussions. Support from
the Helen and Martin Kimmel Center for Molecular Design is
greatly acknowledged. A.S. holds the Siegfried and Irma
Ulmann professorial chair.
Supporting Information Available: A detailed calculation
of Keq for cage formation. This material is available free of
charge via the Internet at http://pubs.acs.org.
Conclusions
In this study, we present EPR data that suggests the formation
of a cage complex between the superoxide radical and Ru-2
and a less stable cage with Ru-1, where a single hydroxamic
acid group is attached to one of the bipyridyl ligands.
References and Notes
(1) Gilbert, J. A.; Eggleston, D. S.; Murphy, W. R. J.; Geselowitz, D.
A.; Gersten, S. W.; Hodgson, D. J.; Meyer, T. J. J. Am. Chem. Soc. 1985,
107, 3855.

9282 J. Phys. Chem. A, Vol. 108, No. 42, 2004
Yavin et al.
(2) Ghosh, P. K.; Brunschwig, B. S.; Chou, M.; Creutz, C.; Sutin, N.
J. Am. Chem. Soc. 1984, 106, 4772.
(31) Morehouse, K. M.; Flitter, W. D.; Mason, R. P. FEBS Lett. 1987,
222, 246.
(3) Shiroishi, H.; Nukaga, M.; Yamashita, S.; Kaneko, M. Chem. Lett.
2002, 4, 488.
(4) Shiroishi, H.; Yamashita, S.; Kaneko, M. J. Mol. Catal. A: Chem.
2001, 169, 269.
(5) Morris, N. D.; Mallouk, T. E. J. Am. Chem. Soc. 2002, 124, 11114.
(6) Yamada, H.; Koike, T.; Hurst, J. K. J. Am. Chem. Soc. 2001, 12775.
(7) Amouyal, E. Sol. Energy Mater. Sol. Cells 1995, 38, 249.
(8) Wagenknecht, H. A.; Stemp, E. D. A.; Barton, J. K. J. Am. Chem.
Soc. 2000, 122, 1.
(9) Ortmans, I.; Content, S.; Buotonnet, N.; Mesmaeker, A. K.-D.;
Bannawarth, W.; Constant, J.-F.; Defrancq, E.; Lhomme, J. Chem.sEur.
J. 1999, 5, 2712.
(32) Yavin, E.; Weiner, L.; Arad-Yellin, R.; Shanzer, A. J. Phys. Chem.
A 2001, 105, 8018.
(33) La Mar, G. N.; Budd, D. L.; Sick, H.; Gersonde, K. Biochim.
Biophys. Acta 1978, 537, 270.
(34) Barrette, W. C.; Johnson, H. W.; Sawyer, D. T. Anal. Chem. 1984,
56, 1890.
(35) Luchette, P. A.; Prosser, R. S.; Sanders, C. R. J. Am. Chem. Soc.
2002, 124, 1778.
(36) McCafferty, D. G.; Bishop, B. M.; Wall, C. G.; Hughes, S. G.;
Mecklenberg, S. L.; Meyer, T. J.; Erickson, B. W. Tetrahedron 1995, 51,
1093.
(37) Nakonieczna, L.; Chimiak, A. Synthesis 1987, 418.
(38) Kober, E. M.; Caspar, J. V.; Sullivan, B. P.; Meyer, T. J. Inorg.
Chem. 1988, 27, 4587.
(39) Hirai, Y.; Aida, T.; Inoue, S. J. Am. Chem. Soc. 1989, 111, 3062.
(40) Langer, M. E.; Khorshahi, F. Process for preparation of 4,4′-
stilbenedicarboxylates by Wittig reaction of 4-(halomehyl)benzoate with
4-formylbenzoate; Lever Brothers Co.: 1992.
(41) Rosa, J. C.; Galanakis, D.; Ganellin, C. R.; Dunn, P. M. J. Med.
Chem. 1996, 39, 4247.
(42) Khalaf, A. I.; Pitt, A. R.; Scobie, M.; Suckling, C. J.; Urwin, J.;
Waigh, R. D.; Fishleigh, R. V.; Young, S. C.; Wylie, W. A. Tetrahedron
2000, 56, 5225.
(43) Bernady, K. F.; Floyd, M. B.; Poletto, J. F.; Weiss, M. J. J. Org.
Chem. 1979, 44, 1438.
(44) Green, E. S. R.; Evans, H.; Rice-Evans, P.; Davies, M. J.; Salah,
N.; Rice-Evans, C. Biochem. Pharmacol. 1993, 45, 357.
(45) Kalyanasundaram, K. Coord. Chem. ReV. 1982, 46, 159.
(46) Erkkila, K. E.; Odom, D. T.; Barton, J. K. Chem. ReV. 1999, 99,
2777.
(10) Carter, P. J.; Cheng, C.-C.; Thorp, H. H. J. Am. Chem. Soc. 1998,
120, 632.
(11) O’Reilly, F.; Kelly, J.; Mesmaeker, A. K. Chem. Commun. 1996,
1013.
(12) Tossi, A. B.; Kelly, J. M. Photochem. Photobiol. 1989, 49, 545.
(13) Farrer, B. T.; Thorp, H. H. Inorg. Chem. 2000, 39, 44.
(14) Araki, K.; Silva, C. A.; Toma, H. E.; Catalani, L. H.; Medeiros,
M. H. G.; Di Mascio, P. J. Inorg. Biochem. 2000, 78, 269.
(15) Nunez, M. E.; Barton, J. K. Curr. Opin. Chem. Biol. 2000, 4, 199.
(16) Hamachi, I.; Tsukiji, S.; Shinkai, S.; Oishi, S. J. Am. Chem. Soc.
1999, 121, 5500.
(17) Wilker, J. J.; Dmochowski, I. J.; Dawson, J. H.; Winkler, J. R.;
Gray, H. B. Angew. Chem., Int. Ed. Engl. 1999, 38, 90.
(18) Liu, R. Q.; Hahm, S.; Miller, M.; Durham, B.; Millett, F. Biochem.
1995, 34, 973.
(19) Murphy, C. J.; Arkin, M. R.; Ghatlia, N. D.; Bossmann, S.; Turro,
N. J.; Barton, J. K. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 5315.
(20) Margalit, R.; Kostic, N. M.; Che, C. M.; Blair, D. F.; Chiang, H.
J.; Pecht, I.; Shelton, J. B.; Shelton, J. R.; Schroeder, W. A.; Gray, H. B.
Proc. Natl. Acad. Sci. U.S.A. 1984, 81, 6554.
(47) Ford, W. E.; Rodgers, M. A. J. J. Phys. Chem. B 1997, 101, 930.
(48) Markus, R. A. P.; Siders, P. J. Phys. Chem. 1982, 86, 622-630.
(49) Connolly, J. S.; Bolton, J. R. In Photoinduced Electron Transfer,
Part D; Fox, M. A. Chanon, M., Eds.; Elsevier: 1988; Chapter 6.2.
(50) Greenstock, C. L.; Wiebe, R. H. Can. J. Chem. 1982, 60, 1560.
(51) Battino, R.; Clever, H. L. Chem. ReV. 1966, 66, 395-463.
(52) Plevert, J.; de Menorval, L. C.; DiRenzo, F.; Fajula, F. J. Phys.
Chem. B 1998, 102, 3412.
(53) Accardi, R. J.; Lobo, R. F. J. Phys. Chem. B 2001, 105, 5883.
(54) Feuerstein, M.; Accardi, R. J.; Lobo, R. F. J. Phys. Chem. B 2000,
104, 10282.
(55) Liu, H.; Kao, H.-M.; Grey, C. P. J. Phys. Chem. B 1999, 103, 4786.
(56) Liu, H.; Grey, C. P. Microporous Mesoporous Mater. 2002, 53,
109.
(21) Aoudia, M.; Guliaev, A. B.; Leontis, N. B.; Rodgers, M. A. J.
Biophys. Chem. 2000, 83, 121.
(22) Fancy, D. A.; Denison, C.; Kim, K.; Xie, Y.; Holdeman, T.; Amini,
F.; Kodadek, T. Chem. Biol. 2000, 7, 697.
(23) Bishop, B. M.; McCafferty, D. G.; Erickson, B. W. Tetrahedron
2000, 56, 4629.
(24) Stemp, E. D. A.; Holmlin, R. E.; Barton, J. K. Inorg. Chim. Acta
2000, 297, 88.
(25) Dunn, A. R.; Dmochowski, I. J.; Winkler, J. R.; Gray, H. B. J.
Am. Chem. Soc. 2003, 125, 12450.
(26) Contakes, S. M.; Dunn, A. R.; Morales, N.; Winkler, J. R.; Gray,
H. B. J. Inorg. Biochem. 2003, 96, 120.
(27) Gross, Z.; Mahammed, A.; Abdales, M.; Weaver, J. J.; Gray, H.
B. J. Inorg. Biochem. 2003, 96, 140.
(28) Di Bilio, A. J.; Crane, B. R.; Wehbi, W. A.; Kiser, C. N.; Abu-
Omar, M. M.; Carlos, R. M.; Richards, J. H.; Winkler, J. R.; Gray, H. B.
J. Am. Chem. Soc. 2001, 123, 3181.
(57) Finkelstein, E.; Rosen, G. M.; Rauckman, E. J. J. Am. Chem. Soc.
1980, 102, 4994.
(58) Hauenstein, B. L.; Dressick, W. J.; Buell, S. L.; Demas, J. N.;
DeGraff, B. A. J. Am. Chem. Soc. 1983, 105, 4251.
(59) Sutin, N.; Cruetz, C. Pure Appl. Chem. 1980, 52, 2717.
(29) Zhang, X.; Rodgers, M. A. J. J. Phys. Chem. 1995, 99, 12797.
(30) Davies, M. J.; Donkor, R.; Dunster, C. A.; Gee, C. A.; Jonas, S.;
Willson, R. L. Biochem. J. 1987, 246, 725.