Effects of hydrogen bonding on metal ion-promoted intramolecular electron transfer and photoinduced electron transfer in a ferrocene-quinone dyad with a rigid amide spacer

Article abstract of DOI:10.1021/ja026441v

A ferrocene-quinone dyad (Fc-Q) with a rigid amide spacer and Fc-(Me)Q dyad, in which the amide proton acting as a hydrogen-bonding acceptor is replaced by the methyl group, are employed to examine the effects of hydrogen bonding on both the thermal and the photoinduced electron-transfer reactions. The hydrogen bonding of the semiquinone radical anion with the amide proton in Fc-Q^.- produced by the electron-transfer reduction of Fc-Q is indicated by the significant positive shift of the one-electron reduction potential of Fc-Q. The hyperfine coupling constants of Fc-Q^.- also indicate the existence of hydrogen bonding, agreeing with those predicted by the density functional calculation. The hydrogen-bonding dynamics in the photoinduced electron transfer from the ferrocene (Fc) to the quinone moiety (Q) in Fc-Q have been successfully detected in the femtosecond laser flash photolysis experiments. Thermal intramolecular electron transfer from Fc to Q in Fc-Q and Fc-(Me)Q also occurs efficiently in the presence of metal ions in acetonitrile at 298 K. The hydrogen bond formed between the semiquinone radical anion and the amide proton in Fc-Q results in remarkable acceleration of the rate of metal ion-promoted electron transfer as compared to the rate of Fc-(Me)Q in which hydrogen bonding is prohibited. The metal ion-promoted electron-transfer rates are well correlated with the binding energies of superoxide ion-metal ion complexes, which are derived from the g_zz values of the ESR spectra.

Full text of DOI:10.1021/ja026441v

Published on Web 12/07/2002
Effects of Hydrogen Bonding on Metal Ion-Promoted
Intramolecular Electron Transfer and Photoinduced Electron
Transfer in a Ferrocene-Quinone Dyad with a Rigid Amide
Spacer
Shunichi Fukuzumi,*,† Ken Okamoto,† Yutaka Yoshida,† Hiroshi Imahori,*,†,§
Yasuyuki Araki,‡ and Osamu Ito*,‡
Contribution from the Department of Material and Life Science, Graduate School of
Engineering, Osaka UniVersity, CREST, Japan Science and Technology Corporation (JST),
Suita, Osaka 565-0871, Japan, and Institute of Multidisciplinary Research for AdVanced
Materials, Tohoku UniVersity, CREST, Japan Science and Technology Corporation (JST),
Sendai, Miyagi 980-8577, Japan
Received April 5, 2002 ; E-mail: fukuzumi@ap.chem.eng.osaka-u.ac.jp
Abstract: A ferrocene-quinone dyad (Fc-Q) with a rigid amide spacer and Fc-(Me)Q dyad, in which the
amide proton acting as a hydrogen-bonding acceptor is replaced by the methyl group, are employed to
examine the effects of hydrogen bonding on both the thermal and the photoinduced electron-transfer
reactions. The hydrogen bonding of the semiquinone radical anion with the amide proton in Fc-Q•- produced
by the electron-transfer reduction of Fc-Q is indicated by the significant positive shift of the one-electron
reduction potential of Fc-Q. The hyperfine coupling constants of Fc-Q•- also indicate the existence of
hydrogen bonding, agreeing with those predicted by the density functional calculation. The hydrogen-
bonding dynamics in the photoinduced electron transfer from the ferrocene (Fc) to the quinone moiety (Q)
in Fc-Q have been successfully detected in the femtosecond laser flash photolysis experiments. Thermal
intramolecular electron transfer from Fc to Q in Fc-Q and Fc-(Me)Q also occurs efficiently in the presence
of metal ions in acetonitrile at 298 K. The hydrogen bond formed between the semiquinone radical anion
and the amide proton in Fc-Q results in remarkable acceleration of the rate of metal ion-promoted electron
transfer as compared to the rate of Fc-(Me)Q in which hydrogen bonding is prohibited. The metal ion-
promoted electron-transfer rates are well correlated with the binding energies of superoxide ion-metal ion
complexes, which are derived from the gzz values of the ESR spectra.
Introduction
studied extensively to understand the factors that control the
electron-transfer processes.6,7 A number of donor-acceptor
Electron-transfer reactions are known to be regulated through
noncovalent interactions such as hydrogen bonding which plays
an important role in biological electron-transfer systems, where
electron donors and acceptors are usually bound to proteins at
a fixed distance.1-5 Thus, electron-transfer reactions between
donor and acceptor molecules bound to proteins have been
linked systems with inert rigid spacers have also been developed
to study the electron-transfer reactions between the donor and
acceptor molecules at a fixed distance to understand the factors
to control the electron-transfer process.8-12 Because it would
be impossible to connect donor and acceptor molecules if an
electron transfer occurred thermally, photoexcitation of the donor
† Osaka University.
(5) (a) Cukier, R. I.; Nocera, D. G. Annu. ReV. Phys. Chem. 1998, 49, 337. (b)
Ghaddar, T. H.; Castner, E. W.; Isied, S. S. J. Am. Chem. Soc. 2000, 122,
1234.
‡ Tohoku University.
§ Present address: Kyoto University.
(1) (a) Scheiner, S. Hydrogen Bonding; Oxford Press: Oxford, 1997. (b)
Sessler, J. L.; Wang, B.; Springs, S. L.; Brown, C. T. In ComprehensiVe
Supramolecular Chemistry; Murakami, Y., Ed.; Pergamon: New York,
1996; Vol. 4, pp 311-336. (c) Nocek, J. M.; Zhou, J. S.; Forest, S. D.;
Priyadarshy, S.; Beratan, D. N.; Onuchic, J. N.; Hoffman, B. M. Chem.
ReV. 1996, 96, 2459.
(2) (a) The Photosynthetic Reaction Center; Deisenhofer, J., Norris, J. R., Eds.;
Academic Press: San Diego, 1993. (b) Anoxygenic Photosynthetic Bacteria;
Blankenship, R. E., Madigan, M. T., Bauer, C. E., Eds.; Kluwer Academic
Publishing: Dordrecht, 1995.
(6) (a) McLendon, G.; Hake, R. Chem. ReV. 1992, 92, 481. (b) McLendon, G.
Acc. Chem. Res. 1988, 21, 160. (c) Iseid, S. S.; Ogawa, M. Y.; Wishart, J.
F. Chem. ReV. 1992, 92, 381.
(7) (a) Moser, C. C.; Keske, J. M.; Warncke, K.; Farid, R. S.; Dutton, P. L.
Nature 1992, 355, 796. (b) Kirmaier, C.; Holton, D. In The Photosynthetic
Reaction Center; Deisenhofer, J., Norris, J. R., Eds.; Academic Press: San
Diego, 1993; Vol. II, pp 49-70. (c) Langen, R.; Chang, I.-J.; Germanas,
J. P.; Richards, J. H.; Winkler, J. R.; Gray, H. B. Science 1995, 268, 1733.
(d) Page, C. C.; Moser, C. C.; Chen, X.; Dutton, P. L. Nature 1999, 402,
47.
(3) (a) Niemz, A.; Rotello, V. M. Acc. Chem. Res. 1999, 32, 44. (b) Rotello,
V. M. In Electron Transfer in Chemistry; Balzani, V., Ed.; Wiley-VCH:
Weinheim, 2001; Vol. 4, pp 68-87.
(4) (a) Sessler, J. L.; Sathiosatham, M.; Brown, C. T.; Rhodes, T. A.;
Wiederrecht, G. J. Am. Chem. Soc. 2001, 123, 3655. (b) De Rege, P. J. F.;
Williams, S. A.; Therien, M. J. Science 1995, 269, 1409. (c) Myles, A. J.;
Branda, N. R. J. Am. Chem. Soc. 2001, 123, 177.
(8) (a) Gust, D.; Moore, T. A. In The Porphyrin Handbook; Kadish, K. M.,
Smith, K. M., Guilard, R., Eds.; Academic Press: San Diego, CA, 2000;
Vol. 8, pp 153-190. (b) Gust, D.; Moore, T. A.; Moore, A. L. Res. Chem.
Intermed. 1997, 23, 621. (c) Gust, D.; Moore, T. A.; Moore, A. L. Acc.
Chem. Res. 1993, 26, 198. (d) Kurreck, H.; Huber, M. Angew. Chem., Int.
Ed. Engl. 1995, 34, 849. (e) Gust, D.; Moore, T. A.; Moore, A. L. Acc.
Chem. Res. 2001, 34, 40.
9
10.1021/ja026441v CCC: $25.00 © 2003 American Chemical Society
J. AM. CHEM. SOC. 2003, 125, 1007-1013
1007

A R T I C L E S
Fukuzumi et al.
Scheme 1
or acceptor moiety was required to start the electron-transfer
reaction. As such, there has so far been no report on the effects
of hydrogen bonding on thermal intramolecular electron-transfer
reactions in donor-acceptor linked systems or on the hydrogen-
bonding dynamics. However, the thermal electron-transfer reac-
tion can be started by adding metal ions to the electron-transfer
reaction system, which would otherwise show no reactivity.13-16
Thus, the comparison of the metal ion-promoted electron-trans-
fer rates of a donor-acceptor linked system involving hydrogen
bonding and those without hydrogen bonding would provide
quantitative insight into the actual role of hydrogen bonding in
the thermal intramolecular electron-transfer reaction.
We report herein thermal intramolecular ET reactions of
donor-acceptor linked systems with and without a hydrogen-
bond acceptor, promoted by metal ions for the first time. A
ferrocene-quinone dyad (Fc-Q) linked with an amide group and
Fc-(Me)Q dyad, in which the amide proton acting as a
hydrogen-bond acceptor is replaced by the methyl group, are
employed to study the effects of hydrogen bonding between
the carbonyl oxygen of the Q•- moiety and N-H proton in the
thermal and photoinduced electron-transfer reactions.17 The
hydrogen-bonding formation results in remarkable acceleration
of the rate of metal ion-promoted electron transfer as compared
to the rate of Fc-(Me)Q in which no hydrogen bond is formed.
The successful detection of hydrogen-bonding dynamics in
intramolecular photoinduced electron transfer in Fc-Q is also
reported.
triflate [Sc(OTf)3 (99%, fw ) 492.16)] was obtained from Pacific
Metals Co., Ltd. (Taiheiyo Kinzoku). Lanthanum triflate [La(OTf)3]
was obtained from Aldrich as hexahydrate form. Yttrium triflate
[Y(OTf)3], europium triflate [Eu(OTf)3], ytterbium triflate [Yb(OTf)3],
lutetium triflate [Lu(OTf)3], and calcium triflate [Ca(OTf)2] were
prepared according to the literature.19 Metal triflates were dried under
vacuum evacuation at 403 K for 40 h prior to use. Magnesium
perchlorate [Mg(ClO4)2] and barium perchlorate [Ba(ClO4)2] were
obtained from Wako Pure Chemical Ind. Ltd., Japan. Chloroform-d
was obtained from EURI SO-TOP, CEA, France.
Synthesis. 4-Aminophenylferrocene was prepared according to the
literature.20 The synthetic route toward Fc-Q and Fc-(Me)Q is
summarized in Scheme 1.
1: A solution of 2,5-dimethoxybenzoic acid (1.54 g, 8.5 mmol) and
2-chloro-4,6-dimethoxy-1,3,5-triazine (1.50 g, 8.5 mmol) in THF (40
mL) was stirred at room temperature for 1 h under nitrogen. N-
Methylmorpholine (1.6 mL, 15 mmol) was then added dropwise at 0
°C. After the solution was stirred at 0 °C for 1 h, the solution was
further stirred at room temperature for 1 h. After filtration of the
mixture, the filtrate was added to a solution of 4-aminophenylferrocene
(210 mg, 0.73 mmol) and N-methylmorpholine (1.0 mL, 9.1 mmol) in
THF (10 mL). After the solution was stirred at 0 °C for 1 h, the solution
was further stirred at room temperature for 4 h. After removal of solvent,
the residue was purified by flash column chromatography (silica gel,
chloroform). Subsequent recrystallization from ethanol gave 1 as an
orange solid (280 mg, 0.63 mmol, 87%). 1H NMR (300 MHz,
CDCl3): δ 9.92 (s, 1H(N-H)), 7.85 (d, 1H(Q), J ) 2.9 Hz), 7.60 (d,
2H(Ph), J ) 8.4 Hz), 7.47 (d, 2H(Ph), J ) 8.4 Hz), 7.05 (dd, 1H(Q),
J ) 9.2, 3.1 Hz), 6.98 (d, 1H(Q), J ) 9.0 Hz), 4.63 (t, 2H(Fc)), 4.30
(t, 2H(Fc)), 4.04 (s, 5H(Fc)), 4.03 (s, 3H(MeO)), 3.85 (s, 3H(MeO)).
FABMS: m/z 441.
Fc-Q: A solution of 1 (885 mg, 2.0 mmol) in dichloromethane (60
mL) was stirred at 0 °C for 1 h under argon. A solution of boron
tribromide (BBr3) in the presence of oxygen and PbO2 (1.5 g, 6.28
mmol) in dichloromethane (0.20 M, 25 mL) was then added dropwise
at 0 °C. After the solution was stirred at 0 °C for 1 h, the solution was
further stirred at room temperature for 3 h. Water and saturated aqueous
sodium hydrogen carbonate were added to the solution. The organic
phase was dried over sodium sulfate and filtrated. After removal of
solvent, the residue was stirred in acetonitrile for 4 h, and the insoluble
ingredient was recovered. Subsequent recrystallization from acetonitrile
gave Fc-Q as a brown solid (20 mg, 0.29 mmol, 15%). 1H NMR (300
MHz, CDCl3): δ 10.30 (s, 1H(N-H)), 7.80 (s, 1H(Q)), 7.63 (d, 2H-
(Ph), J ) 8.6 Hz), 7.48 (d, 2H(Ph), J ) 8.6 Hz), 6.90 (s, 2H(Q)), 4.64
Experimental Section
General. 1H NMR spectra were measured on a JEOL JNM-AL300
spectrometer. Fast atom bombardment mass spectra (FAB-MS) were
obtained on a JEOL JMS-DX300 mass spectrometer. ESR spectra were
recorded on a JEOL X-band spectrometer (JES-ME-1X) with a quartz
ESR tube (1.2 mm i.d.).
Materials. Tris(2,2′-bipyridyl)ruthenium(III) hexafluorophosphate
[Ru(bpy)3(PF6)3] was prepared according to the literature.18 Scandium
(9) (a) Imahori, H.; Sakata, Y. AdV. Mater. 1997, 9, 537. (b) Imahori, H.; Sakata,
Y. Eur. J. Org. Chem. 1999, 2445, 5. (c) Guldi, D. M. Chem. Commun.
2000, 321. (d) Guldi, D. M.; Prato, M. Acc. Chem. Res. 2000, 33, 695. (e)
Fukuzumi, S.; Imahori, H. In Electron Transfer in Chemistry; Balzani, V.,
Ed.; Wiley-VCH: Weinheim, 2001; Vol. 2, pp 927-975.
(10) (a) Wasielewski, M. R. In Photoinduced Electron Transfer; Fox, M. A.,
Chanon, M., Eds.; Elsevier: Amsterdam, 1988; Part A, p 161. (b)
Wasielewski, M. R. Chem. ReV. 1992, 92, 435. (c) Jordan, K. D.; Paddon-
Row, M. N. Chem. ReV. 1992, 92, 395.
(11) (a) Chambron, J.-C.; Chardon-Noblat, S.; Harriman, A.; Heitz, V.; Sauvage,
J.-P. Pure Appl. Chem. 1993, 65, 2343. (b) Harriman, A.; Sauvage, J.-P.
Chem. Soc. ReV. 1996, 26, 41. (c) Blanco, M.-J.; Jime´nez, M. C.; Chambron,
J.-C.; Heitz, V.; Linke, M.; Sauvage, J.-P. Chem. Soc. ReV. 1999, 28, 293.
(d) Balzani, V.; Juris, A.; Venturi, M.; Campagna, S.; Serroni, S. Chem.
ReV. 1996, 96, 759.
(12) (a) Maruyama, K.; Osuka, A.; Mataga, N. Pure Appl. Chem. 1994, 66,
867. (b) Osuka, A.; Mataga, N.; Okada, T. Pure Appl. Chem. 1997, 69,
797.
(13) (a) Fukuzumi, S.; Ohkubo, K. Chem.-Eur. J. 2000, 6, 4532. (b) Fukuzumi,
S. Bull. Chem. Soc. Jpn. 1997, 70, 1. (c) Fukuzumi, S.; Itoh, S. In AdVances
in Photochemistry; Neckers, D. C., Volman, D. H., von Bu¨nau, G., Eds.;
Wiley: New York, 1998; Vol. 25, pp 107-172. (d) Fukuzumi, S. Bull.
Chem. Soc. Jpn. 1997, 70, 1.
(14) Fukuzumi, S. In Electron Transfer in Chemistry; Balzani, V., Ed.; Wiley-
VCH: Weinheim, 2001; Vol. 4, pp 3-67.
(15) (a) Fukuzumi, S.; Okamoto, T.; Otera, J. J. Am. Chem. Soc. 1994, 116,
5503. (b) Fukuzumi, S.; Inada, O.; Satoh, N.; Suenobu, T.; Imahori, H. J.
Am. Chem. Soc. 2002, 124, 9181. (c) Fukuzumi, S.; Mori, H.; Imahori, H.;
Suenobu, T.; Araki, Y.; Ito, O.; Kadish, K. M. J. Am. Chem. Soc. 2001,
123, 12458. (d) Fukuzumi, S.; Satoh, N.; Okamoto, T.; Yasui, K.; Suenobu,
T.; Seko, Y.; Fujitsuka, M.; Ito, O. J. Am. Chem. Soc. 2001, 123, 7756.
(16) Fukuzumi, S.; Okamoto, K.; Imahori, H. Angew. Chem., Int. Ed. 2002, 41,
620.
(19) (a) Forsberg, J. H.; Spaziano, V. T.; Balasubramanian, T. M.; Liu, G. K.;
Kinsley, S. A.; Duckworth, C. A.; Poteruca, J. J.; Brown, P. S.; Miller, J.
L. J. Org. Chem. 1987, 52, 1017. (b) Kobayashi, S.; Hachiya, I. J. Org.
Chem. 1994, 59, 3590.
(20) Imahori, H.; Tamaki, K.; Guldi, D. M.; Luo, C.; Fujitsuka, M.; Ito, O.;
Sakata, Y.; Fukuzumi, S. J. Am. Chem. Soc. 2001, 123, 2607.
(17) A preliminary report has appeared: Fukuzumi, S.; Yoshida, Y.; Okamoto,
K.; Imahori, H.; Araki, Y.; Ito, O. J. Am. Chem. Soc. 2002, 124, 6794.
(18) DeSimone, R. E.; Drago, R. S. J. Am. Chem. Soc. 1970, 92, 2343.
9
1008 J. AM. CHEM. SOC. VOL. 125, NO. 4, 2003

Electron Transfer in a Ferrocene-Quinone Dyad
A R T I C L E S
(t, 2H(Fc)), 4.33 (t, 2H(Fc)), 4.04 (s, 5H(Fc)). FAB-MS: m/z 411. TOF-
MS: m/z 411. HR-MS (EI): m/z 411.0539 (Calcd for C23H17NO3Fe
411.0509). Anal. Calcd for C23H17NO3Fe‚0.75H2O: C, 65.04; H, 4.39;
N, 3.30. Found: C, 65.05; H, 4.40; N, 3.29.
atures (258-298 K) using a Unisoku thermostated cell holder designed
for low-temperature experiments. Typically, an acetonitrile solution of
Fc-Q and Mg(ClO4)2 were transferred to the spectrophotometric cell.
Rates of intramolecular electron-transfer reaction from the ferrocene
moiety to the quinone moiety in the presence of metal ions were
monitored by the rise of the absorption band at 800 nm due to the
ferricenium ion moiety.
Electrochemical Measurement. Cyclic voltammetry (CV) measure-
ments of Fc-Q were performed on a BAS 100W electrochemical
analyzer in deaerated acetonitrile containing 0.1 M NBu4PF6 as
supporting electrolyte at 298 K. A conventional three-electrode cell
with a Pt working electrode (surface area of 0.3 mm2) and a Pt wire as
the counter electrode was utilized for the CV measurements. The Pt
working electrode (BAS) was polished with a BAS polishing alumina
suspension and rinsed with acetone before use. The measured potentials
were recorded with respect to the Ag/AgNO3 (0.01 M) reference
electrode. The potential (vs Ag/Ag+) was converted to values versus
SCE by adding 0.29 V.22
Fc-(Me)Q: A solution of 1 (0.16 g, 0.36 mmol) in THF (30 mL)
under nitrogen was treated sequentially with methyl iodide (65 µL, 1
mmol) and sodium hydride (65% oil dispersion, 0.015 g, 0.4 mmol) at
room temperature. The reaction mixture was stirred at room temperature
for 10 min (0 °C) and then was warmed at 60 °C under nitrogen for
1.5 h and again was stirred at room temperature for 2 h. After removal
of solvent, subsequent recrystallization from ethanol gave 2 (Scheme
1). A solution of 2 and PbO2 (1.5 g, 6.28 mmol) in dichloromethane
(30 mL) was stirred at -78 °C under oxygen. A solution of boron
tribromide (BBr3) in dichloromethane (5 mmol, 5 mL) was then added
dropwise at -78 °C. After the solution was stirred at -78 °C for 10
min, the solution was further stirred at room temperature overnight.
Water and saturated aqueous sodium hydrogen carbonate were added
to the solution. The organic phase was dried over sodium sulfate and
filtrated. After removal of solvent, subsequent recrystallization from
chloroform and hexane gave Fc-(Me)Q as a brown solid (42 mg, 0.10
Theoretical Calculations. Density functional (DFT) calculations
were performed using the Amsterdam density functional (ADF) program
version 1999.02 developed by Baerends et al.23 with various basis sets
on a COMPAQ DS20E computer. The gradient corrections of Becke88
(exchange) and Perdew86 (correlation) were included in the exchange-
correlation functional. The hfc values are derived from multiplying
506.82 and spin density values of each hydrogen.
1
mmol, 28%). H NMR (300 MHz, CDCl3): δ 7.4 (d, 2H(Ph), J ) 9
Hz), 6.9 (d, 2H(Ph), J ) 9 Hz), 6.54 (m, 1H(Q)), 6.51 (m, 2H(Q)), 4.5
(t, 2H(Fc), J ) 2 Hz), 4.3 (t, 2H(Fc), J ) 2 Hz), 3.9 (s, 5H(Fc)), 3.4
(s, 3H(N-Me)). FAB-MS: m/z 427 (+2H+). TOF-MS: m/z 427
(+2H+). HR-MS (EI): m/z 425.0705 (Calcd for C24H19NO3Fe 425.0666).
Time-Resolved Absorption Measurements. The picosecond laser
flash photolysis was carried out using SHG (388 nm) of mode-locked
Ti:sapphire laser (Clark-MXR, CPA-2001, fwhm of 150 fs) for an
exciting source of Fc-Q or Fc-(Me)Q in benzonitrile (5 × 10-4 M) at
295 K. A white pulse generated by focusing the fundamental of the
Ti:sapphire laser on a flowed D2O/H2O (1:1 volume) cell was used as
a monitoring light. The monitoring light transmitted through a sample
was detected with a Dual MOS detector (Hamamatsu Photonics C6140)
equipped with a polychromator (Acton Research SpectraPro 150). The
spectra were obtained by averaging 16 events on a microcomputer.
X-ray Structure Determination. Fc-Q was dissolved in chloroform
containing a little acetonitrile and recrystallized under an atmospheric
pressure of diethyl ether vapor. Data of X-ray diffraction were collected
by a Rigaku RAXIS-RAPID imaging plate two-dimensional area
detector using graphite-monochromated Mo KR radiation (λ ) 0.71069
Å) to 2θ max of 55.0°. All of the crystallographic calculations were
performed by using the teXsan software package of the Molecular
Structure Corp. [teXsan: Crystal Structure Analysis Package, Molecular
Structure Corp. (1985 and 1999)]. The crystal structure was solved by
direct methods and refined by full-matrix least squares. All non-
hydrogen atoms and hydrogen atoms were refined anisotropically and
isotropically, respectively. A summary of the crystal data and experi-
mental parameters for structure determinations is given in the Sup-
porting Information (S1).
Results and Discussion
Structure of Fc-Q Dyad. Single crystals of Fc-Q were
obtained by vapor diffusion of ether into an MeCN solution of
Fc-Q. The crystallographic data are summarized in the Sup-
porting Information (S1), and the ORTEP drawing is shown in
Figure 1a. The selected bond distances and angles are also
presented in the Supporting Information (S2). The closest
distance between the quinone oxygen atom and the amide
hydrogen is 2.17 Å, and the C-O bond lengths of two carbonyl
groups of quinone are eventually the same (1.22 Å). These
results indicate that there is no hydrogen bonding between the
quinone oxygen atom and the amide proton in the neutral form.
Hydrogen-Bonding Formation upon Electron-Transfer
Reduction. The cyclic voltammograms of Fc-Q exhibited two
reversible one-electron redox couples of two redox active
moieties at 0.39 and -0.16 V (vs SCE) as shown in Figure 2.
The former one-electron redox potential corresponds to the
Fc+/Fc couple, which agrees with the one-electron oxidation
Spectral Measurements. The formation of Fc+-Q was examined
from the change in the UV-vis spectrum of Fc-Q (1.0 × 10-4 M) in
3+
the presence of Ru(bpy)3 by using a Hewlett-Packard 8453 diode
array spectrophotometer with a quartz cuvette (path length ) 10 mm)
at 298 K. The formation of Fc-Q•- (Fc-(Me)Q•-) was examined by
measuring the ESR spectrum of Fc-Q (Fc-(Me)Q) (1.0 × 10-4 M) in
the presence of semiquinone radical anion (1.0 × 10-4 M) at 298 K.
The g values and hyperfine coupling constants (hfc) were calibrated
by using an Mn2+ marker. Tetramethylammonium hydroxide (TMAOH)
was used for generation of the semiquinone radical anion in the reaction
between hydroquinone and p-benzoquinone.21 The formation of com-
plexes of Fc+-Q•- (Fc+-(Me)Q•-) with each metal ion was examined
by the change in the UV-vis spectrum of an acetonitrile solution of
Fc-Q in the presence of each metal ion using a Hewlett-Packard 8453
diode array spectrophotometer and a Shimadzu UV-2200 spectropho-
tometer with a quartz cuvette (path length ) 10 mm) at 298 K. The
formation of complexes of Fc+-Q•- with Mg2+ was also examined by
measuring the ESR spectrum of an acetonitrile solution of Fc-Q in the
presence of Mg(ClO4)2 at 298 K.
0
potential of ferrocene (Eox ) 0.37 V).24 The latter potential
thereby corresponds to the Q/Q•- couple. The one-electron re-
duction potential of Q (Ered0 ) -0.16 V) is significantly shifted
to a positive direction as compared to that of p-benzoquinone
(-0.50 V) and more positive than the value of p-benzoquinone
Kinetic Measurements. Kinetic measurements of thermal intramo-
lecular electron-transfer reactions from the ferrocene moiety to the
quinone moiety in Fc-Q in the presence of metal ions were performed
on a UNISOKU RSP-601 stopped-flow rapid scan spectrophotometer
with the MOS-type high sensitive photodiode array at various temper-
(22) Mann, C. K.; Barnes, K. K. Electrochemical Reactions in Nonaqueous
Systems; Marcel Dekker: New York, 1990.
(23) (a) Baerends, E. J.; Ellis, D. E.; Ros, P. Chem. Phys. 1973, 2, 41. (b) Velde,
G. T.; Baerends, E. J. J. Comput. Phys. 1992, 99, 84.
(21) Fukuzumi, S.; Nakanishi, I.; Suenobu, T.; Kadish, K. M. J. Am. Chem.
Soc. 1999, 121, 3468.
(24) Fukuzumi, S.; Mochizuki, S.; Tanaka, T. Inorg. Chem. 1989, 28, 2459.
9
J. AM. CHEM. SOC. VOL. 125, NO. 4, 2003 1009

A R T I C L E S
Fukuzumi et al.
0
in Fc-Q (-0.16 V). Such a negative shift in the Ered value of
Q in Fc-(Me)Q as compared to that in Fc-Q indicates that the
radical anion (Q•-) is stabilized by the hydrogen bond formed
with the amide proton of the spacer.
The optimized geometry of Ph-Q•- in which Fc is omitted
using the Amsterdam density functional (ADF) calculation with
the V (small) basis set is shown in Figure 1b. The O-H distance
between the quinone oxygen atom of Q•- and the amide
hydrogen in Figure 1b is 1.56 Å, which is much shorter than
the distance in the X-ray structure of neutral Fc-Q (Figure 1a).27
This value is even shorter than the hydrogen-bonding distance
between semiquinone radical anion and water (1.78 Å).28 The
C-O bond length of the hydrogen-bonded carbonyl group (1.28
Å) becomes longer than the bond length of the other carbonyl
group (1.26 Å) due to the weakening of the C-O bond by the
hydrogen bonding with the amide proton. For comparison, the
optimized geometry of Ph-(Me)Q•- in which the N-H group
is replaced by the N-Me group is shown in Figure 1c. There
is no significant structural difference between Ph-Q•- and
Ph-(Me)Q•- except for the absence of the hydrogen bond in
the latter case.
To confirm the hydrogen-bonding formation of Q•- in the
dyad, Fc-Q•- was produced by the electron-transfer reduction
of Fc-Q by the semiquinone radical anion (see Experimental
Section). Because the one-electron oxidation potential (Eox0) of
0
the semiquinone radical anion, which is equivalent to the Ered
value of p-benzoquinone, is more negative than the Ered0 value
of Q in Fc-Q, an electron transfer from the semiquinone radical
anion to Q in Fc-Q is thermodynamically feasible. In fact, an
electron transfer from the semiquinone radical anion to Q occurs
to produce Fc-Q•- (eq 1).
Figure 1. (a) ORTEP drawing of Fc-Q. (b) Optimized structure of
Ph-Q•-. (c) Optimized structure of Ph-(Me)Q•-. The optimized structure is
obtained by the ADF calculation with the V (small) basis set.27
The ESR spectrum of Fc-Q•- thus produced is shown in
Figure 3a together with the computer simulation spectrum
(Figure 3b). The g value is determined as g ) 2.0055, which is
typical as the value of semiquinone radical anions.29 The
hyperfine coupling constants (hfc) are determined as 4.60 (1H),
2.05 (1H), and 1.75 G (1H). The hfc values obtained from the
result in Figure 3b can be well reproduced by the ADF
calculation with the V (small) basis set, which predicts the hfc
values as 4.11 (1H), 2.18 (1H), and 1.57 G (1H).30
The radical anion (Q•-) in Fc-(Me)Q•- without hydrogen
bonding is also produced by the electron-transfer reduction of
Fc-(Me)Q by semiquinone radical anion as shown in Figure
3c. The hyperfine coupling (hfc) constants are determined by
the computer simulation (Figure 3d) as 2.65 (1H), 2.25 (1H),
and 2.20 G (1H), which are quite different from those observed
in Fc-Q•- with hydrogen bonding. The hfc values can also be
Figure 2. Cyclic voltammograms of (a) Fc-Q (0.5 mM) and (b) Fc-(Me)Q
(0.5 mM) in MeCN containing 0.1 M Bu4NPF6.
(26) The positive shift of the Ered0 value of the quinone moiety of Fc-(Me)Q as
compared to p-benzoquinone is ascribed to the electron-withdrawing effect
of the carbonyl group in Fc-(Me)Q.
possessing an electron-withdrawing substituent (-0.38 V for
chloro-p-benzoquinone).25 On the other hand, the cyclic vol-
tammograms of Fc-(Me)Q (Figure 2b) exhibited two reversible
one-electron redox couples of two redox active moieties at 0.44
(27) Similar results were obtained using the B3LYP method. The ADF method
with small V core gave the spin densities consistent with the experimentally
determined hfc values in Figure 3b.
0
and -0.40 V (vs SCE).26 The Ered value of Q (-0.40 V) in
(28) O’Malley, P. J. Phys. Chem. A 1997, 101, 6334.
(29) Fukuzumi, S.; Nishizawa, N.; Tanaka, T. J. Org. Chem. 1984, 49, 3571.
(30) The hfc values were also calculated using the B3LYP method. Of all of
the computational methods and basis sets tested, the ADF calculation with
the V (small) set predicts the hfc values in closest agreement with
experimental results in Figure 3 (see Supporting Information S3).
Fc-(Me)Q is significantly more negative as compared to that
(25) Fukuzumi, S.; Koumitsu, S.; Hironaka, K.; Tanaka, T. J. Am. Chem. Soc.
1987, 109, 305.
9
1010 J. AM. CHEM. SOC. VOL. 125, NO. 4, 2003

Electron Transfer in a Ferrocene-Quinone Dyad
A R T I C L E S
Figure 4. Time-resolved absorption spectrum of Fc-Q dyad (5.0 × 10-4
M) in argon-saturated PhCN excited at 388 nm (delay time: 1 ns) at 298
K. (b) Visible absorption spectra of Fc-Q•- (0.40 mM, solid line) and
Fc-(Me)Q•- (0.40 mM, broken line) in deaerated MeCN at 298 K.
Figure 3. (a) ESR spectrum of Fc-Q•- (1.0 × 10-4 M) in deaerated MeCN
at 298 K and (b) the computer simulation spectrum. (c) ESR spectrum of
Fc-(Me)Q•- in deaerated MeCN at 298 K and (d) the computer simulation
spectrum.
Scheme 2
well reproduced by the ADF calculation, which predicts the hfc
values as 3.19 (1H), 1.88 (1H), and 1.77 G (1H).30
Photoinduced Electron Transfer in Fc-Q. Photoexcitation
of the Q moiety in Fc-Q in deaerated benzonitrile (PhCN)31
with a 388 nm femtosecond (150 fs width) laser light results in
the appearance of a new absorption band (λmax ) 580 nm) at 1
ns after the laser excitation as shown in Figure 4a. The
absorption band at 580 nm is significantly red shifted as
compared to the diagnostic absorption band of the semiquinone
radical anion at 422 nm.32 Such a red shift has been observed
when the semiquinone radical anion is bound to a hard acid
1 ps and decays with the first-order rate constant of ∼2 × 1011
s-1 (τ ≈ 5 ps) to the residual absorbance, accompanied by the
rise in absorption at 580 nm due to the hydrogen-bonded Q•-
(see Supporting Information S4).35 The absorption at 580 nm
decays at a longer time scale, obeying first-order kinetics with
a rate constant of 2.6 × 108 s-1. This indicates that electron
transfer from Fc to the singlet excited state of Q occurs rapidly
to produce Fc-Q•- without changing the conformation (<1 ps),
and then Q•- forms the hydrogen bonding with the amide proton
of the spacer (τ ≈ 5 ps), and the resulting radical ion pair decays
via a back electron transfer to the ground state as shown in
Scheme 2.
such as Mg2+ 33
Thus, the absorption band at 580 nm may be
.
assigned to Q•-, which is hydrogen bonded to the amide proton
of the spacer. No such absorption band was observed in the
case of Fc-(Me)Q, which has no hydrogen-bond acceptor. The
absorption spectra of the hydrogen-bonded Q•- in Fc-Q•- and
Q•- without hydrogen bonding in Fc(Me)Q•-, produced by the
electron-transfer reduction of Fc-Q and Fc-(Me)Q, are shown
in Figure 4b.34
The absorption band at 450 nm due to Q•- without hydrogen
bonding appears immediately upon the laser excitation within
Metal Ion-Promoted Thermal Electron Transfer in Fc-Q
and Fc-(Me)Q. No electron transfer from Fc to Q occurs in
(31) PhCN was used as a solvent for the laser flash photolysis, because good
spectral resolution was obtained as compared to MeCN.
(32) Fukuzumi, S.; Nakanishi, I.; Maruta, J.; Yorisue, T.; Suenobu, T.; Itoh, S.;
Arakawa, R.; Kadish, K. M. J. Am. Chem. Soc. 1998, 120, 6673.
(33) Fukuzumi, S.; Okamoto, T. J. Am. Chem. Soc. 1993, 115, 11600.
(34) The λmax value of the hydrogen-bonded Q•- varies slightly depending on
the system: 580 nm for Fc+-Q•- (Figure 4a) and 550 nm for Fc-Q•- (Figure
4b).
(35) The transient absorption due to the Fc+ part which should be formed together
with the Q•- part is not detected probably because of the low extinction
coefficient of Fc+. This is confirmed by formation of Fc+ (λmax ) 800
nm) by the chemical oxidation of Fc-Q with [Ru(bpy)3]3+ (bpy ) 2,2′-
bipyridine).
9
J. AM. CHEM. SOC. VOL. 125, NO. 4, 2003 1011

A R T I C L E S
Fukuzumi et al.
Figure 5. (a) ESR spectrum of Fc+-Q•- (4.0 × 10-4 M) in the presence
of Mg2+ (7.5 × 10-2 M) in deaerated MeCN at 298 K and (b) the computer
simulation spectrum.
Figure 6. Time course of the absorption change at 800 nm due to formation
of Fc+-Q•- by the intramolecular electron transfer in Fc-Q (1.0 × 10-4
M) in the presence of Mg(ClO4)2 (0.10 M) in deaerated MeCN at 298 K.
Inset: First-order plot.
Fc-Q thermally in MeCN at 298 K. However, addition of Mg2+
(0.10 M) results in formation of Fc+ as indicated by the
appearance of the absorption band due to Fc+ at 800 nm together
with the absorption band at 420 nm (see Supporting Information
S5). The absorption band at 420 nm which may be assigned to
Q•- bound to Mg2+ is blue shifted as compared to those of non-
hydrogen-bonded Q•- (450 nm) and hydrogen-bonded Q•- (580
nm). Such a significant blue shift has indeed been observed
when Q•- is bound to two Mg2+ ions: the absorption maximum
of the Q•--Mg2+ complex (590 nm) is blue shifted to 410 nm
Scheme 3
for Q•--2Mg2+ 33
.
To confirm the formation of a paramagnetic species, that is,
Q•- bound to both Mg2+ and the amide proton, the ESR
spectrum was measured after addition of Mg2+ ion (7.5 × 10-2
M) to an MeCN solution of Fc-Q (4.0 × 10-4 M). The observed
ESR spectrum is shown in Figure 5a together with the computer
simulation spectrum (Figure 5b). The hyperfine coupling (hfc)
constants are determined as 3.95 (1H), 2.30 (1H), and 1.60 G
(1H). The hfc values are similar to those observed for the
hydrogen-bonded Q•- in Fc-Q•- without Mg2+ ion in Figure
3b. However, the largest hfc value (3.95 G) is significantly
smaller than the value without Mg2+ ion (4.60 G) due to the
spin delocalization to the Mg nucleus. The g value is determined
as g ) 2.0057, which is slightly larger than the value of
Fc-Q•- without Mg2+ ion.
Table 1. Electron-Transfer Rate Constants (ket) of Mn+-Promoted
Electron Transfer in Fc-Q and Fc-(Me)Q in Deaerated MeCN at
298 K
ket, M-1 s-1
n+
M
Fc-Q
Fc-(Me)Q
Sc3+
Y3+
1.0 × 109
2.5 × 106
1.0 × 106
7.0 × 105
1.8 × 105
2.0 × 105
1.4 × 103
5.0
2.0 × 105
3.8 × 102
4.0 × 101
1.0 × 102
2.3 × 102
9.4 × 101
0.4
La3+
Eu3+
Yb3+
Lu3+
Mg2+
Ca2+
Ba2+
a
a
0.9
a Not determined.
The rate of formation of Fc+ monitored by the appearance
of the absorption band at 800 nm obeys first-order kinetics as
shown in Figure 6. The appearance and the observed-first-order
rate constant (kobs) increase linearly with increasing Mg2+
concentration (Supporting Information S6). The second-order
rate constant (ket) is determined from the slope of the linear
plot of kobs versus [Mg2+] as 1.4 × 103 M-1 s-1. When Fc-Q
is replaced by Fc-(Me)Q which contains no hydrogen-bond
acceptor, the ket value (0.4 M-1 s-1) is much smaller than the
ket value (1.4 × 103 M-1 s-1) of the electron transfer in Fc-Q,
which is promoted not only by Mg2+ but also by the hydrogen
bond formed between the Q•- moiety and the amide proton.
A variety of metal ions (Mn+: triflate salts) were employed
to promote the intramolecular electron transfer in Fc-Q and
Fc-(Me)Q (Scheme 3). The ket values are determined as listed
in Table 1.30 The promoting effects of metal ions vary signifi-
cantly depending on the Lewis acidity of metal ions. We have
recently reported that the binding energies (∆E) of metal ions
•-
with O2 can be evaluated from the gzz values of the
O2•--Mn+ complexes and that the ∆E values are well correlated
with the promoting effects of metal ions in intermolecular
electron-transfer reduction of p-benzoquinone as well as O2.13a
Figure 7 shows plots of log ket versus ∆E for the Mn+-promoted
intramolecular electron transfer in Fc-Q and Fc-(Me)Q. The
ket value of the Sc3+-promoted electron transfer in Fc-Q is the
largest among metal ions investigated, and this is 104 times
larger than the corresponding ket value of Fc-(Me)Q.36 The log
ket values of Mn+-promoted electron transfer in Fc-Q are linearly
correlated with the ∆E values, and each value is ca. 104 times
larger than the ket values of Fc-(Me)Q. The stabilization of
the Q•- moiety by hydrogen bonding with the amide proton in
Fc-Q•- results in the positive shift in Ered0 of Fc-Q as compared
(36) In the case of Fe-(Me)Q, no electron transfer occurred when a weaker
Lewis acid than Mg2+ such as Ca2+ or Ba2+ was employed.
9
1012 J. AM. CHEM. SOC. VOL. 125, NO. 4, 2003

Electron Transfer in a Ferrocene-Quinone Dyad
A R T I C L E S
Fc-(Me)Q is ascribed to the effect of hydrogen bonding formed
between the Q•- moiety and the amide proton in Fc+-Q•-.37
In conclusion, the present study has demonstrated that
formation of hydrogen bonding between the Q•- moiety and
the amide proton results in remarkable acceleration of the rate
of metal ion-promoted intramolecular electron transfer from Fc
to Q in Fc-Q. The hydrogen-bonding formation is coupled with
the metal ion-promoted thermal electron-transfer reaction in
Fc-Q. In contrast, photoexcitation of Fc-Q results in electron
transfer from Fc to the singlet excited state of Q to produce
Fc-Q•- without changing the conformation (<1 ps), and then
Q•- forms the hydrogen bonding with the amide proton of the
spacer (τ ≈ 5 ps).
Acknowledgment. This work was partially supported by a
Grant-in-Aid for Scientific Research Priority Area (No. 11228205)
from the Ministry of Education, Culture Sports, Science, and
Technology, Japan.
Figure 7. Plots of log ket versus ∆E in Mn+-promoted electron transfer in
Fc-Q and Fc-(Me)Q in MeCN at 298 K.
Note Added after ASAP: The ASAP version of this paper
published 12/7/2002 on the Web did not show the cation on
the ferrocene moieties after addition of the metal ions in Scheme
3. The final Web version published 12/30/2002 and the print
version are correct.
to that of Fc-(Me)Q in which the amide proton is replaced by
the methyl group. The 104 times difference in the ket values
corresponds to the difference in the one-electron reduc-
0
tion potential between Fc-Q (Ered vs SCE ) -0.16 V) and
Supporting Information Available: X-ray crystallographic
data (S1), selected bond lengths of Fc-Q (S2), comparison of
calculation methods of the hfc values (S3), time profile of
photoinduced electron transfer in Fc-Q (S4), Mg2+-promoted
Fc-(Me)Q (Ered0 vs SCE ) -0.40 V), because the ratio of the
rate constant is given by exp(0.24 (eV)/kBT), which is equal to
1.1 × 104 at 298 K (kB is the Boltzmann constant). Thus,
the remarkable difference in the ket values between Fc-Q and
electron transfer in Fc-Q (S5), and plot of kobs versus [Mg2+
]
(S6) (PDF). This material is available free of charge via the
Internet at http://pubs.acs.org.
(37) There is some structural difference between Ph-Q•- (Figure 1b) and
Ph-(Me)Q•- (Figure 1c), which may also affect the rate of metal
ion-promoted electron transfer of Fc-Q and Fc-(Me)Q in addition to the
effect of hydrogen bond.
JA026441V
9
J. AM. CHEM. SOC. VOL. 125, NO. 4, 2003 1013