Electron donor-bridge-acceptor molecules with bridging nitronyl nitroxide radicals: Influence of a third spin on charge- and spin-transfer dynamics

Article abstract of DOI:10.1021/ja0576435

Appending a stable radical to the bridge molecule in a donor-bridge- acceptor system (D-B-A) is potentially an important way to control charge- and spin-transfer dynamics through D-B-A. We have attached a nitronyl nitroxide (NN^.) stable radical

Full text of DOI:10.1021/ja0576435

Published on Web 03/15/2006
Electron Donor-Bridge-Acceptor Molecules with Bridging
Nitronyl Nitroxide Radicals: Influence of a Third Spin on
Charge- and Spin-Transfer Dynamics
Erin T. Chernick, Qixi Mi, Richard F. Kelley, Emily A. Weiss, Brooks A. Jones,
Tobin J. Marks,* Mark A. Ratner,* and Michael R. Wasielewski*
Contribution from the Department of Chemistry and Center for Nanofabrication and
Molecular Self-Assembly, Northwestern UniVersity, EVanston, Illinois 60208-3113
Received November 9, 2005; E-mail: m-wasielewski@northwestern.edu
Abstract: Appending a stable radical to the bridge molecule in a donor-bridge-acceptor system (D-B-
A) is potentially an important way to control charge- and spin-transfer dynamics through D-B-A. We have
attached a nitronyl nitroxide (NN•) stable radical to a D-B-A system having well-defined distances between
the components: MeOAn-6ANI-Ph(NN•)-NI, where MeOAn ) p-methoxyaniline, 6ANI ) 4-(N-piperidinyl)-
naphthalene-1,8-dicarboximide, Ph ) phenyl, and NI ) naphthalene-1,8:4,5-bis(dicarboximide). MeOAn-
6ANI, NN•, and NI are attached to the 1, 3, and 5 positions of the Ph bridge. Using both time-resolved
optical and EPR spectroscopy, we show that NN• influences the spin dynamics of the photogenerated
triradical states 2,4(MeOAn+•-6ANI-Ph(NN•)-NI-•), resulting in slower charge recombination within the
triradical compared to the corresponding biradical lacking NN•. The observed spin-spin exchange interaction
between the photogenerated radicals MeOAn+• and NI-• is not altered by the presence of NN•, which only
accelerates radical pair intersystem crossing. Charge recombination within the triradical results in the
formation of 2,4(MeOAn-6ANI-Ph(NN•)-3*NI), in which NN• is strongly spin-polarized. Normally, the spin
dynamics of correlated radical pairs do not produce a net spin polarization; however, net spin polarization
appears on NN• with the same time constant as describes the photogenerated radical ion pair decay. This
effect is attributed to antiferromagnetic coupling between NN• and the local triplet state 3*NI, which is
populated following charge recombination. This requires an effective switch in the spin basis set between
3
the triradical and the three-spin charge recombination product having both NN• and *NI present.
Introduction
to the latter, we have developed several approaches to control-
ling charge transport in organic donor-bridge-acceptor (D-
Electron-transfer reactions are important in applications
ranging from solar cells and light-emitting diodes to molecule-
based materials for electronics and photonics.1-16 With regard
B-A) arrays that take advantage of the speed and efficiency
of ultrafast photoinduced electron-transfer reactions.17-19 How-
ever, using spin dynamics to control charge and spin transport
properties within molecules is an important aspect of this
problem that has received only limited attention. Photogenerated
radical pairs are capable of exhibiting coherent spin motion over
microsecond time scales,20,21 which is considerably longer than
coherent phenomena involving photogenerated excited states.
This affords the possibility that coherent spin motion can provide
the basis for novel organic computational devices.22-32 However,
before efficient organic spintronic devices can be realized, a
(1) Wasielewski, M. R. Chem. ReV. 1992, 92, 435-461.
(2) Gust, D.; Moore, T. A.; Moore, A. L. Acc. Chem. Res. 2001, 34, 40-48.
(3) Wagner, R. W.; Lindsey, J. S.; Seth, J.; Palaniappan, V.; Bocian, D. F. J.
Am. Chem. Soc. 1996, 118, 3996-3997.
(4) de Silva, A. P.; Gunaratne, H. Q. N.; Gunnlaugsson, T.; Huxley, A. J. M.;
McCoy, C. P.; Radmacher, J. T.; Rice, T. E. Chem. ReV. 1997, 97, 1515-
1566.
(5) Willner, I.; Willner, B. J. Mater. Chem. 1998, 8, 2543-2556.
(6) Tour, J. M.; Kozaki, M.; Seminario, J. M. J. Am. Chem. Soc. 1998, 120,
8486-8493.
(7) Collier, C. P.; Matterstei, G.; Wong, E. W.; Luo, Y.; Beverly, K.; Sampaio,
J.; Raymo, F. M.; Stoddart, J. F.; Heath, J. R. Science 2000, 289, 1172-
1175.
(8) Davis, W. B.; Svec, W. A.; Ratner, M. A.; Wasielewski, M. R. Nature
1998, 396, 60-63.
(16) Jones, B. A.; Ahrens, M. J.; Yoon, M.-H.; Facchetti, A.; Marks, T. J.;
Wasielewski, M. R. Angew. Chem., Int. Ed. 2004, 43, 6363-6366.
(17) Gosztola, D.; Niemczyk, M. P.; Wasielewski, M. R. J. Am. Chem. Soc.
1998, 120, 5118-5119.
(9) Waldeck, D. H.; Beratan, D. N. Science 1993, 261, 576-577.
(10) Metzger, R. M. Acc. Chem. Res. 1999, 32, 950-957.
(11) Chen, J.; Reed, M. A.; Rawlett, A. M.; Tour, J. M. Science 1999, 286,
1550-1552.
(18) Hayes, R. T.; Wasielewski, M. R.; Gosztola, D. J. Am. Chem. Soc. 2000,
122, 5563-5567.
(12) Prathapan, S.; Yang, S. I.; Seth, J.; Miller, M. A.; Bocian, D. F.; Holten,
D.; Lindsey, J. S. J. Phys. Chem. B 2001, 105, 8237-8248.
(13) Yang, S. I.; Prathapan, S.; Miller, M. A.; Seth, J.; Bocian, D. F.; Lindsey,
J. S.; Holten, D. J. Phys. Chem. B 2001, 105, 8249-8258.
(14) Ik Yang, S.; Lammi, R. K.; Prathapan, S.; Miller, M. A.; Seth, J.; Diers, J.
R.; Bocian, D. F.; Lindsey, J. S.; Holten, D. J. Mater. Chem. 2001, 11,
2420-2430.
(15) del Rosario Benites, M.; Johnson, T. E.; Weghorn, S.; Yu, L.; Rao, P. D.;
Diers, J. R.; Yang, S. I.; Kirmaier, C.; Bocian, D. F.; Holten, D.; Lindsey,
J. S. J. Mater. Chem. 2002, 12, 65-80.
(19) Lukas, A. S.; Wasielewski, M. R. In Molecular Switches; Feringa, B. L.,
Ed.; Wiley-VCH: Weinheim, 2001; pp 1-35.
(20) Prisner, T.; Dobbert, O.; Dinse, K. P.; van Willigen, H. J. Am. Chem. Soc.
1988, 110, 1622-1623.
(21) Angerhofer, A.; Toporowicz, M.; Bowman, M. K.; Norris, J. R.; Levanon,
H. J. Phys. Chem. 1988, 92, 7164-7166.
(22) Rajca, A. Chem. ReV. 1994, 94, 871-893.
(23) Nakatsuji, S.; Anzai, H. J. Mater. Chem. 1997, 7, 2161-2174.
(24) Lahti, P. M., Ed. Magnetic Properties of Organic Materials; Marcel
Dekker: New York, 1999.
9
4356
J. AM. CHEM. SOC. 2006, 128, 4356-4364
10.1021/ja0576435 CCC: $33.50 © 2006 American Chemical Society

Stable NN• on a Donor
−Bridge−Acceptor Molecule
A R T I C L E S
greater fundamental understanding of the factors controlling spin
dynamics in complex organic donor-acceptor systems must be
obtained.
largely on its two N-O groups and not on the molecule to which
NN• is appended,50 so that attachment of NN• to B within
D-B-A does not result in significant spin delocalization onto
B itself. Modulating the degree of spin delocalization from the
appended radical onto B is potentially an important means to
control the electron-transfer dynamics through B, and the work
presented here is the first step toward this goal. We have chosen
to modify a previously well-characterized D-B-A system to
test these ideas, MeOAn-6ANI-Ph-NI (8),51,52 where MeOAn
) p-methoxyaniline, 6ANI ) 4-(N-piperidinyl)naphthalene-1,8-
dicarboximide, Ph ) phenyl, and NI ) naphthalene-1,8:4,5-
bis(dicarboximide).51 We have now synthesized an analogue to
this system, MeOAn-6ANI-Ph(NN•)-NI (7), in which
MeOAn-6ANI, NN•, and NI are attached to the 1, 3, and 5
positions, respectively, of the Ph bridge. Here we report on the
influence of NN• on both electron transfer and spin dynamics
in 7, studied using time-resolved optical and EPR spectroscopy.
The presence of additional unpaired spins provided by stable
free radicals and triplet-state molecules (e.g. oxygen) has been
shown to increase the rate of radical pair intersystem crossing
(RP-ISC) between photogenerated singlet and triplet radical
pairs.32-36 For molecules in which the stable radical and the
photoactive system are connected by a flexible covalent linkage,
the resulting large degree of conformational freedom translates
into a distribution of distances and spin-spin interaction
strengths that complicate the analysis of the spin dynamics.37-40
Our approach to studying the influence of spin dynamics on
charge and spin transport in D-B-A molecules strongly
restricts conformational freedom and controls electronic coupling
between each component of a D-B-A molecule having a stable
radical covalently bound to it. For example, we recently
demonstrated41 that covalently attaching a 2,2,6,6-tetramethyl-
1-piperidinyloxyl (TEMPO) stable free radical to the acceptor
terminus of a rigid D-B-A molecule perturbs charge recom-
bination rates via an enhanced intersystem crossing (EISC)
mechanism similar to that observed for intermolecular sys-
tems,32,33,35 while not altering the spin-spin exchange interaction
within the photogenerated radical ion pair.
These results, along with the fact that the structural and
electronic properties of the bridge molecule within a D-B-A
system are critical to determining the rate of electron (or hole)
transfer from D to A,42 prompted us to explore how a stable
paramagnetic center (a radical or radical ion) attached directly
to the bridge molecule of a D-B-A system affects the
dynamics of photoinduced charge and spin transport. Nitronyl
nitroxide (NN•) is a stable, easily functionalized radical that
has been studied extensively in the organic magnetic materials
field.43-49 The spin density distribution of NN• is localized
(25) Crayston, J. A.; Devine, J. N.; Walton, J. C. Tetrahedron 2000, 56, 7829-
7857.
(26) Sugawara, T., Sakurai, H., Izuoka, A., Eds. Electronically controllable high
spin systems realized by spin-polarized donors; Gordon and Breach:
Amsterdam, 2001.
(27) Wolf, S. A.; Awschalom, D. D.; Buhrman, R. A.; Daughton, J. M.; von
Molnar, S.; Roukes, M. L.; Chtchelkanova, A. Y.; Treger, D. M. Science
2001, 294, 1488-1495.
Experimental Section
The synthesis and characterization of compound 7 are described in
the Supporting Information. Compound 7 was purified by preparative
TLC on alumina. All solvents were spectrophotometric grade or distilled
prior to use.
Ground-state absorption measurements were made on a Shimadzu
(UV-1601) spectrophotometer. The optical density of all samples was
maintained between 0.3 and 0.6 at 420 nm (ꢀ6ANI,420 nm ) 7000 cm-1
M-1) for both femtosecond and nanosecond transient absorption
spectroscopy. Femtosecond transient absorption measurements were
made using the 420 nm frequency-doubled output from a regeneratively
amplified titanium-sapphire laser system operating at 2 kHz as the
(28) Miller, J. S.; Manson, J. L. Acc. Chem. Res. 2001, 34, 563-570.
(29) Awschalom, D. D.; Flatte, M. E.; Samarth, N. Sci. Am. 2002, 286, 67-73.
(30) Rajca, A. Chem. Eur. J. 2002, 8, 4834-4841.
(31) Itkis, M. E.; Chi, X.; Cordes, A. W.; Haddon, R. C. Science 2002, 296,
1443-1445.
(32) Buchachenko, A. L.; Berdinsky, V. L. Chem. ReV. 2002, 102, 603-612.
(33) Buchachenko, A. L.; Ruban, L. V.; Step, E. N.; Turro, N. J. Chem. Phys.
Lett. 1995, 233, 315-318.
(34) Vlassiouk, I.; Smirnov, S.; Kutzki, O.; Wedel, M.; Montforts, F.-P. J. Phys.
Chem. B 2002, 1-6, 8657-8666.
(35) Smirnov, S.; Vlassiouk, I.; Kutzki, O.; Wedel, M.; Montforts, F.-P. J. Am.
Chem. Soc. 2002, 124, 4212-4213.
(36) Volkova, O. S.; Taraban, M. B.; Plyusnin, V. F.; Leshina, T. V.; Egorov,
M. P.; Nefedov, O. M. J. Phys. Chem. A 2003, 107, 4001-4005.
(37) Mori, Y.; Sakaguchi, Y.; Hayashi, H. J. Phys. Chem. A 2002, 106, 4453-
4467.
(45) Matsuura, H.; Tamura, R.; Yamauchi, J. Synth. Met. 2003, 133-134, 605-
607.
(38) Mori, Y.; Sakaguchi, Y.; Hayashi, H. Bull. Chem. Soc. Jpn. 2001, 74, 293-
304.
(46) Yamaguchi, K.; Namimoto, H.; Fueno, T.; Nogami, T.; Shirota, Y. Chem.
Phys. Lett. 1990, 166, 408-414.
(39) Ishii, K.; Hirose, Y.; Kobayashi, N. J. Phys. Chem. A 1999, 103, 1986-
1990.
(47) Yamaguchi, K.; Okumura, M.; Fueno, T.; Nakasuji, K. Synth. Met. 1991,
43, 3631-3634.
(40) Ishii, K.; Hirose, Y.; Fujitsuka, H.; Ito, O.; Kobayashi, N. J. Am. Chem.
Soc. 2001, 123, 702-708.
(48) Hiraoka, S.; Okamoto, T.; Kozaki, M.; Shiomi, D.; Sato, K.; Takui, T.;
Okada, K. J. Am. Chem. Soc. 2004, 126, 58-59.
(49) Nagashima, H.; Inoue, H.; Yoshioka, N. J. Phys. Chem. B 2004, 108, 6144-
6151.
(41) Weiss, E. A.; Chernick, E. T.; Wasielewski, M. R. J. Am. Chem. Soc. 2004,
126, 2326-2327.
(42) Weiss, E. A.; Wasielewski, M. R.; Ratner, M. A. Top. Curr. Chem. 2005,
257, 103-133.
(50) Davis, M. S.; Kreilick, R. W.; Morokuma, K. J. Am. Chem. Soc. 1972, 94,
5588-5592.
(43) Moigne, J. L.; Gallani, J. L.; Wautelet, P.; Moroni, M.; Oswald, L.; Cruz,
C.; Galerne, Y.; Arnault, J. C.; Duran, R.; Garrett, M. Langmuir 1998, 14,
7484-7492.
(51) Lukas, A. S.; Bushard, P. J.; Weiss, E. A.; Wasielewski, M. R. J. Am.
Chem. Soc. 2003, 125, 3921-3930.
(44) Zhang, D.; Xu, Y.; Ding, L.; Liu, Y.; Zhu, D. Chem. Phys. Lett. 1999,
304, 236-240.
(52) Shaakov, S.; Galili, T.; Stavitski, E.; Levanon, H.; Lukas, A.; Wasielewski,
M. R. J. Am. Chem. Soc. 2003, 125, 6563-6572.
9
J. AM. CHEM. SOC. VOL. 128, NO. 13, 2006 4357

A R T I C L E S
Chernick et al.
excitation pulse and a white light continuum probe pulse as described
earlier.53 Samples were placed in a 2 mm path length glass cuvette and
stirred using a motorized wire stirrer to prevent thermal lensing and
sample degradation. The samples were irradiated with 0.5-1.0 µJ/pulse
focused to a 200 µm spot. The total instrument response time for the
pump-probe experiments was 150 fs. Transient absorption kinetics
were fit to a sum of exponentials with a Gaussian instrument function
using Levenberg-Marquardt least-squares fitting.
gave 2D complex spectra versus time and magnetic field. For each
kinetic trace, the signal acquired prior to the laser pulse was subtracted
from the data. Kinetic traces recorded off-resonance were considered
background signals, whose average was subtracted from all kinetic
traces. The spectra were subsequently phased into a Lorentzian part
and a dispersive part, and the former, also known as the imaginary
magnetic susceptibility ø′′, is presented.
High-power microwave pulses were generated by a 1-kW TWT
amplifier (Applied Systems Engineering 117X). The typical length of
a π/2 pulse was 8 ns. The resonator was fully overcoupled to achieve
Q < 200 and a dead time of ∼72 ns. All the quadrature-detected spectra
were properly phased, and ø′′ is presented. In the pulse EPR kinetic
measurements, the first microwave pulse was initially applied 200 ns
before the laser pulse and then was delayed incrementally relative to
the laser pulse. A kinetic trace was formed by measuring the intensity
of the free induction decay (FID) at different time delays of the
microwave pulse with respect to the laser pulse.
Samples for nanosecond transient absorption spectroscopy were
placed in a 10 mm path length quartz cuvette equipped with a vacuum
adapter and subjected to five freeze-pump-thaw degassing cycles.
The samples were excited with 5 ns, 1 mJ, 420 nm laser pulses
generated using the frequency-tripled output of a Continuum 8000 Nd:
YAG laser to pump a Continuum Panther OPO. The excitation pulse
was focused to a 5 mm diameter spot and matched to the diameter of
the probe pulse generated using a xenon flashlamp (EG&G Electro-
Optics FX-200). The signal was detected using a photomultiplier tube
with high voltage applied to only four dynodes (Hamamatsu R928).
The total instrument response time is 7 ns and is determined primarily
by the laser pulse duration. The sample cuvette was placed between
the poles of a Walker Scientific HV-4W electromagnet powered by a
Walker Magnion HS-735 power supply. The field strength was
measured by a Lakeshore 450 gaussmeter with a Hall effect probe.
Both the electromagnet and the gaussmeter were interfaced with the
data collection computer, allowing measurement and control of the
magnetic field to (1 × 10-5 T during data acquisition. Due to the
length of the sample runs (>3 h), a small amount of sample degradation
was observed, resulting in a decrease in the triplet yield at zero field,
∆A(B ) 0), over the course of the experiments. To compensate for
this, the magnetic field was reset to B ) 0 mT every three kinetic
traces and ∆A(B ) 0) was plotted and fit with a polynomial or series
of polynomials. These functions were used to calculate the relative
triplet yield or RP yield as a function of applied field strength. The
relative triplet yield is thus:
Results
Synthesis. The synthesis of 8 has been reported previously,51
and the complete synthesis of radical 7 is outlined in Scheme
1. Commercially available 3,5-dinitrobenzyl alcohol was pro-
tected with a triisopropylsilyl (TIPS) group, followed by
reduction of the nitro groups to yield 2 (71%). Imide coupling
of the known MeOAn-6ANI anhydride41 with diamine 2 yields
3 (80%). Due to the decrease in nucleophilicity of the remaining
amino group in 3 following the first imide coupling, the
subsequent imide coupling of the NI acceptor proved to be
inefficient, giving 4 in a modest 59% yield. Deprotection of
the TIPS group was accomplished with TBAF at -20 °C in
THF, giving alcohol 5 in 97% yield. We observed that
deprotection does not occur at -78 °C, while the NI group is
cleaved if the reaction is allowed to occur at room temperature.
Subsequent oxidation of the alcohol to the corresponding
aldehyde with fresh barium manganate yields 6 in quantitative
yield. Formation of the nitronyl nitroxide was accomplished
using slight modifications to Ullman’s procedure.55-58 2,3-Bis-
(hydroxamino)-2,3-dimethylbutane sulfate was condensed with
aldehyde 6 in dimethyl sulfoxide to form the tetramethylimi-
dazolidine intermediate. Attempts to use the typical methanol/
water solvent system for this reaction failed due to the
insolubility of the aldehyde precursor. The oxidation of the
tetramethylimidazolidine intermediate was accomplished with
sodium periodate in a biphasic system of water and methylene
chloride to give 7 in 22% yield.
∆A(B)
∆A(B ) 0)
T
T0
)
The results presented are averages of three or more experiments
conducted on separate days with freshly prepared samples in spectro-
photometric or freshly distilled ACS grade toluene.
For EPR measurements, a toluene solution of 7 (0.2 mM) was loaded
into a quartz tube (4 mm o.d. × 2 mm i.d.) and subjected to five freeze-
pump-thaw degassing cycles on a vacuum line (10-4 mbar). The tube
was then sealed using a hydrogen torch and kept in the dark when not
being used. The sample was excited using 416 nm, 1 mJ, 7 ns laser
pulses from the H2-Raman shifted output of a frequency-tripled,
Q-switched Nd:YAG laser (Quanta Ray DCR-2).
Steady-state EPR spectra, transient CW EPR spectra, and pulse EPR
spectra were measured using a Bruker Elexsys E580 X-band EPR
spectrometer with a variable-Q dielectric resonator (Bruker ER 4118X-
MS5) at room temperature. Steady-state CW EPR spectra were
measured under the conditions of 0.2-2 mW microwave power and
0.01-0.05 mT field modulation at 100 kHz. The g values of the spectra
were calibrated with a crystalline 2,2-diphenyl-1-picrylhydrazyl (DPPH)
standard (g ) 2.0036). Transient CW EPR measurements were carried
out under CW microwave irradiation (typically 2-20 mW) by
accumulating kinetic traces of transient magnetization following
photoexcitation. The field modulation was disabled to achieve a
response time τ ) Q/πν ≈ 30 ns,54 and microwave signals in emission
(e) and/or absorption (a) were detected in both the real and the
imaginary channels (quadrature detection). Sweeping the magnetic field
Steady-State Properties. The photophysical properties of
6ANI, NI, and 8 have been characterized previously in
detail.51,52,59,60 In summary, the ground-state optical spectrum
of 6ANI exhibits a broad charge-transfer (CT) band centered
at 390 nm, while that of NI exhibits three distinct absorptions
due to vibronic structure on the π-π* transition at 343, 363,
and 382 nm. In MeOAn-6ANI-Ph(NN•)-NI there is an
additional predominant absorption belonging to NN• at 368 nm,
which overlaps the absorptions of both 6ANI and NI (Figure
(55) Ullman, E. F.; Osiecki, J. H.; Boocock, D. G. B.; Darcy, R. J. Am. Chem.
Soc. 1972, 94, 7049-7059.
(56) Hirel, C.; Vostrikova, K. E.; Pecaut, J.; Ovcharenko, V. I.; Rey, P. Chem.
Eur. J. 2001, 7, 2007-2014.
(57) Osiecki, J. H.; Ullman, E. F. J. Am. Chem. Soc. 1968, 90, 1078-1079.
(58) Klyatskaya, S. V.; Tretyakov, E. V.; Vasilevsky, S. F. Russ. Chem. Bull.,
Int. Ed. 2002, 51, 128-134.
(53) Giaimo, J. M.; Gusev, A. V.; Wasielewski, M. R. J. Am. Chem. Soc. 2002,
124, 8530-8531.
(54) Schweiger, A.; Jeschke, G. Principles of pulse electron paramagnetic
resonance; Oxford University Press: Oxford, 2001.
(59) Greenfield, S. R.; Svec, W. A.; Gosztola, D.; Wasielewski, M. R. J. Am.
Chem. Soc. 1996, 118, 6767-6777.
(60) Debreczeny, M. P.; Svec, W. A.; Marsh, E. M.; Wasielewski, M. R. J.
Am. Chem. Soc. 1996, 118, 8174-8175.
9
4358 J. AM. CHEM. SOC. VOL. 128, NO. 13, 2006

Stable NN• on a Donor
−Bridge−Acceptor Molecule
A R T I C L E S
Scheme 1 a
a Reagents and conditions: (a) TIPSCl, imidazole, DMF, reflux, 71%; (b) Pd/C, H2, 32 psi, quantitative; (c) MeOAn-6ANI anhydride, pyridine, imidazole,
reflux, 80%; (d) NI-octyl anhydride, pyridine, imidazole, reflux, 59%; (e) TBAF, THF, -20 °C, 97%; (f) BaMnO4, CH2Cl2, quantitative; (g)
2,3-bis(hydroxamino)-2,3-dimethylbutane sulfate, DMSO; (h) NaIO4, H2O, CH2Cl2, 22%.
potential of NI is -0.53 V. These redox potentials are assumed
not to differ significantly in MeOAn-6ANI-Ph(NN•)-NI. The
one-electron potentials for oxidation and reduction of NN• are
approximately 0.82 and -1.25 V vs SCE, respectively.61,62
The energy levels of the excited and charge-separated species
for 8 have been calculated previously using the Weller equa-
tion.51 It is assumed that there is no significant change in these
levels when the nitronyl nitroxide free radical is appended to
the bridge, so that the energy level diagram displayed in Figure
2 is representative of both radical 7 and model compound 8.
Time-Resolved Charge Separation and Recombination
Dynamics. Femtosecond transient absorption spectroscopy
Figure 1. UV-vis spectrum of 7 in toluene.
1). There is also a weak, broad absorption due to NN• near 600
allows for the direct observation of the intermediates involved
nm. The redox potentials of MeOAn-6ANI-Ph-NI have been
described in detail elsewhere.51,59 Briefly, the first oxidation
potential of ANI occurs at 1.22 V and the first reduction
(61) Sakurai, H.; Izuoka, A.; Sugawara, T. J. Am. Chem. Soc. 2000, 122, 9723-
9734.
(62) Ziessel, R.; Ulrich, G.; Lawson, R. C.; Echegoyen, L. J. Mater. Chem.
1999, 9, 1435-1448.
9
J. AM. CHEM. SOC. VOL. 128, NO. 13, 2006 4359

A R T I C L E S
Chernick et al.
Figure 2. Energy level diagram for compound 7.
behind the MFE have been researched extensively67-70 and
applied to many donor-acceptor systems,51,63,71-77 including
biological68,78-81 systems. For example, photoexcitation of
model compound 8, which lacks NN•, initially produces the
singlet RP, which undergoes electron-nuclear hyperfine coupling-
induced RP-ISC to produce the triplet RP, 1(MeOAn+•-6ANI-
3
Ph-NI-•) f (MeOAn+•-6ANI-Ph-NI-•). The subsequent
charge recombination process is spin selective; i.e., the singlet
RP recombines to the singlet ground state and the triplet RP
recombines to yield the neutral local triplet MeOAn-6ANI-
Ph-3*NI. Application of a static magnetic field splits the RP
triplet levels, and variation of the field strength modulates the
efficiency of the ISC by adjusting the energies of triplet
sublevels relative to that of the singlet level. When the Zeeman
splitting of the triplet RP levels equals the intrinsic singlet-
triplet splitting, 2J, of the RP, there is an increase in intersystem
crossing rate. This increase translates into a maximum in triplet
RP production and therefore a maximum in local triplet
production upon recombination. By monitoring the yield of local
triplet production as a function of applied magnetic field, the
magnitude of the magnetic superexchange interaction, 2J, can
be measured directly.82,83
Figure 3. Femtosecond transient absorption spectra of 7 in toluene obtained
at the indicated times following a 420 nm, 100 fs laser pulse. Inset: A
kinetic trace obtained at 495 nm.
in charge separation (CS) (Figure 3). Photoexcitation of 6ANI
with a 420 nm, 100 fs laser pulse results in the reaction
MeOAn-1*6ANI-Ph(NN•)-NI
f
MeOAn+•-6ANI-•-
The same two rapid, nonadiabatic charge separation steps
occur in radical 7, even though the initial state is formally a
doublet, as will be discussed in detail below. Figure 4 shows
the MFE on the yield of MeOAn-6ANI-Ph(NN•)-3*NI
resulting from charge recombination. The data for 7 show a
Ph(NN•)-NI with a time constant τCS1 ) 11 ps, as indicated
by the formation of an absorption band (shoulder) due to
MeOAn+• near 510 nm. Subsequently, the absorption features
characteristic of NI-• at 480 and 610 nm appear with τCS2
)
31 ps, indicating that the reaction MeOAn+•-6ANI-•-Ph-
(NN•)-NI f MeOAn+•-6ANI-Ph(ΝΝ•)-NI-• has occurred.
The transient spectra are similar to those observed earlier for
8,51 and hence, there is no evidence from the transient absorption
data that NN• is either oxidized or reduced in this process. The
time constants for this sequential charge separation process are
comparable to those reported earlier for MeOAn-6ANI-Ph-
NI, where τCS1 ) 8 ps and τCS2 ) 40 ps.
(67) Weller, A.; Staerk, H.; Treichel, R. Faraday Discuss., Chem. Soc. 1984,
78, 271-278.
(68) Hoff, A. J.; Gast, P.; van der Vos, R.; Franken, E. M.; Lous, E. J. Z. Phys.
Chem. 1993, 180, 175-192.
(69) Till, U.; Hore, P. J. Mol. Phys. 1997, 90, 289-296.
(70) Steiner, U. E.; Ulrich, T. Chem. ReV. 1989, 89, 51-147.
(71) Schulten, K.; Staerk, H.; Weller, A.; Werner, H.-J.; Nickel, B. Z. Phys.
Chem. 1976, 101, 371-390.
(72) Tanimoto, Y.; Okada, N.; Itoh, M.; Iwai, K.; Sugioka, K.; Takemura, F.;
Nakagaki, R.; Nagakura, S. Chem. Phys. Lett. 1987, 136, 42-46.
(73) Sakaguchi, Y.; Hayashi, H. J. Phys. Chem. A 1997, 101, 549-555.
(74) Werner, U.; Kuhnle, W.; Staerk, H. J. Phys. Chem. 1993, 97, 9280-9287.
(75) Weiss, E. A.; Ratner, M. A.; Wasielewski, M. R. J. Phys. Chem. A 2003,
107, 3639-3647.
The magnetic field effect (MFE) on the rate and yield of
radical ion pair (RP) recombination directly reveals the mag-
nitude of electron spin-spin exchange interaction, 2J, between
(76) Tadjikov, B.; Smirnov, S. Phys. Chem. Chem. Phys. 2001, 3, 204-212.
(77) Tsentalovich, Y. P.; Morozova, O. B.; Avdievich, N. I.; Ananchenko, G.
S.; Yurkovskaya, A. V.; Ball, J. D.; Forbes, M. D. E. J. Phys. Chem. A
1997, 101, 8809-8816.
the spins within the RP, which is proportional to the square of
63-66
the donor-acceptor superexchange coupling, VDA
.
The
(78) Blankenship, R. E.; Schaafsma, T. J.; Parson, W. W. Biochim. Biophys.
Acta 1977, 461, 297-305.
mechanistic details of the RP-ISC mechanism and the theory
(79) Plato, M.; Mo¨bius, K.; Michel-Beyerle, M. E.; Bixon, M.; Jortner, J. J.
Am. Chem. Soc. 1988, 110, 7279-7285.
(63) Weiss, E. A.; Ahrens, M. J.; Sinks, L. E.; Gusev, A. V.; Ratner, M. A.;
Wasielewski, M. R. J. Am. Chem. Soc. 2004, 126, 5577-5584.
(64) Kobori, Y.; Sekiguchi, S.; Akiyama, K.; Tero-Kubota, S. J. Phys. Chem.
A 1999, 103, 5416-5424.
(80) Werner, H.-J.; Schulten, K.; Weller, A. Biochim. Biophys. Acta 1978, 502,
255-268.
(81) Norris, J. R.; Bowman, M. K.; Budil, D. E.; Tang, J.; Wraight, C. A.; Closs,
G. L. Proc. Natl. Acad. Sci. U.S.A. 1982, 79, 5532-5536.
(82) Anderson, P. W. Phys. ReV. 1959, 115, 2-13.
(65) Paddon-Row, M. N.; Shephard, M. J. J. Phys. Chem. A 2002, 106, 2935-
2944.
(83) Shultz, D. A.; Fico, R. M., Jr.; Bodnar, S. H.; Kumar, R. K.; Vostrikova,
K. E.; Kampf, J. W.; Boyle, P. D. J. Am. Chem. Soc. 2003, 125, 11761-
11771.
(66) Volk, M.; Haberle, T.; Feick, R.; Ogrodnik, A.; Michel-Beyerle, M. E. J.
Phys. Chem. 1993, 97, 9831-9836.
9
4360 J. AM. CHEM. SOC. VOL. 128, NO. 13, 2006

Stable NN• on a Donor
−Bridge−Acceptor Molecule
A R T I C L E S
Figure 4. Relative triplet yield of 7 in toluene as a function of magnetic
field strength.
Figure 6. Time-resolved EPR kinetics of the RP state (red) and transiently
polarized NN• (blue) of 7 after laser excitation (t ) 0), measured in toluene
at room temperature. The red curve is the difference between two transient
CW kinetic traces obtained at field values of 339.3 and 340.0 mT, as
indicated by the double-arrow lines in Figure 5, to cancel out the uniform
a signal from the polarized NN•, whereas the blue curve is measured by
the FID intensities of the polarized NN• at various times after laser excitation
(see text). Inset: The decay of the RP state matches the rise of polarization
at NN•.
not completely undetectable. Thus, time-resolved EPR (TREPR)
spectra (i.e., pulsed laser excitation, CW microwaves) were
measured around g ≈ 2, the region where RPs and stable radicals
are observed. Immediately after laser excitation, an intense
photogenerated RP signal with an emission, absorption (e a)
polarization pattern appears (Figure 5A), centered at g ) 2.0033,
similar to the g values of MeOAn•+ and NI•-.52 As the RP
decays, a new set of peaks with enhanced absorption grow in
(Figure 5B) which have g factors, hyperfine splittings, and line
widths identical to those measured for NN• in the steady state.
Thus, NN• is transiently polarized following charge recombina-
tion of the RP state.
To obtain a kinetic trace of the RP state alone, the raw data
traces from the TREPR spectra at 340.0 and 339.3 mT were
subtracted, so that the uniform a signal from the polarized NN•
cancels out, while the e a signal from the RP state gets doubled
(Figure 6). Furthermore, the kinetic behavior of the transient
polarization at NN• was monitored using the magnitude of its
FID with pulse EPR, a technique with better temporal fidelity
at long times than transient CW methods.86 Moreover, in contrast
to doublet-state radicals, the FID from a spin-correlated RP is
suppressed by a π/2 microwave pulse,87-89 and thus does not
interfere with selective observation of the polarized NN• radical.
The two kinetic traces in Figure 6 are both well fit with a rise-
decay biexponential model, yielding the charge recombination
lifetime of the RP state, τCR ) 101 ns, and the rise and decay
time of the NN• polarization τ1 ) 95 ( 7 ns and τ2 ) 2.3 (
0.2 µs, respectively. Notably, the decay of the RP state matches
the rise of polarization at NN•, the implications of which will
be discussed later in this paper, and the decay lifetime of the
Figure 5. Continuous wave (CW) EPR spectra of 7 in toluene at room
temperature, measured (A) 60 and (B) 320 ns after laser excitation, and
(C) in steady state in the dark. Note that spectrum C is given in the derivative
form. (D) Hyperfine dependence of the line widths (full width at half-
maximum, fwhm) of the individual peaks in spectrum C. Dashed line in
spectrum A is the numerical simulation to reveal ∆DQ ) 1.3 mT.
clear resonance at 1.5 ( 0.3 mT, which is very similar to that
observed for model system 8. The data indicate that the presence
of NN• on the Ph bridge does not perturb the magnitude of the
spin-spin exchange interaction between the MeOAn+• and NI-•
radicals.
EPR Spectroscopy. The steady-state EPR spectrum of 7
consists of five hyperfine lines centered at g ) 2.0062 due to
the two equivalent nitrogens of NN• having |aN| ) 0.745 mT
(Figure 5C), which is similar to that reported earlier for nitronyl
nitroxides.55 In contrast to unsubstituted nitronyl nitroxides, the
line widths of the hyperfine peaks of 7 are quadratically
dependent on the overall nuclear quantum number M (Figure
5D), signifying relatively slow rotational averaging of the g and
hyperfine anisotropies84 in the large D-B-A system.
At room temperature in toluene, molecular tumbling renders
the EPR spectra of triplet states broad and structureless,85 if
(86) McLauchlan, K. A. In Modern pulsed and continuous-waVe electron spin
resonance; Kevan, L., Bowman, M. K., Eds.; Wiley: New York, 1990; pp
285-363.
(87) Hasharoni, K.; Levanon, H.; Tang, J.; Bowman, M. K.; Norris, J. R.; Gust,
D.; Moore, T. A.; Moore, A. L. J. Am. Chem. Soc. 1990, 112, 6477-
6481.
(88) Timmel, C. R.; Hore, P. J. Chem. Phys. Lett. 1994, 226, 144-150.
(89) Weis, V.; van Willigen, H. J. Porphyrins Phthalocyanines 1998, 2, 353-
361.
(84) Carrington, A.; McLachlan, A. D. Introduction to magnetic resonance with
applications to chemistry and chemical physics; Harper & Row: New York,
1967.
(85) Yamauchi, S. Bull. Chem. Soc. Jpn. 2004, 77, 1255-1268.
9
J. AM. CHEM. SOC. VOL. 128, NO. 13, 2006 4361

A R T I C L E S
Chernick et al.
NN• polarization conforms to the spin-lattice relaxation time
T1 of nitroxides at room temperature.90
Discussion
Intersystem Crossing. The lifetime of MeOAn+•-6ANI-
Ph(NN•)-NI-• (τCR ) 101 ns) is significantly longer than that
of MeOAn+•-6ANI-Ph-NI-• (τCR ) 73 ns).51 In both cases,
charge recombination (CR) leading to the ground state lies deep
within the Marcus inverted region91 because -∆GCR . λ, where
λ ) 0.54 eV,92 and the rate of this process is therefore expected
to be relatively slow. The slower charge recombination in
MeOAn+•-6ANI-Ph(NN•)-NI-• compared to MeOAn+•-
6ANI-Ph-NI-• is attributed to EISC induced by the magnetic
field of NN•, which increases the rate of mixing between
2(MeOAn+•-6ANI-Ph(NN•)-NI-•) and 2,4(MeOAn+•-6ANI-
Ph(NN•)-NI-•).32,34,39,93 Similar effects have been reported in
D-B-A molecules having a TEMPO radical attached.34,38,93-95
The energy level diagram for radical 7 (Figure 2) shows that,
for the CR reaction, 2,4(MeOAn+•-6ANI-Ph(NN•)-NI-•) is
nearly isoenergetic with 2,4(MeOAn-6ANI-Ph(NN•)-3*NI),
so that an equilibrium between these two species is estab-
lished.52,96 EISC shifts the time-integrated population from
2(MeOAn+•-6ANI-Ph(NN•)-NI-•) to 2,4(MeOAn+•-6ANI-
Ph(NN•)-NI-•) in 7, which creates a bottleneck to overall
charge recombination due to the fact that formation of
2,4(MeOAn-6ANI-Ph(NN•)-3*NI) does not provide a fast exit
channel for the charge-separated state. Therefore, the overall
rate of charge recombination is slowed, similar to what we
observed previously for a MeOAn-6ANI-NI derivative in
which a TEMPO radical was rigidly bound to the terminus of
the NI acceptor group.41 This effect is analyzed in terms of the
spin dynamics of the system in more detail below.
Electron Spin-Spin Couplings in the Photogenerated
Triradical. Following photoexcitation of 7, the triradical
MeOAn•+-6ANI-Ph(NN•)-NI•- has three pairs of exchange
interactions, 2J12, 2J13, and 2J23, where the subscripts 1, 2, and
3 denote the spins on MeOAn•+, NI•-, and NN•, respectively.
Mori et al.37 studied a NI•--TEMPO• biradical anion with a
spin-spin distance of 8.44 Å and four σ bonds between the
two radicals and estimated the exchange coupling 2J to be ∼1
T (10-4 eV). The exchange interaction 2J23 between the two
paramagnetic centers NN• and NI•- in 7 is assumed to be similar
because these radicals are also 8.5 Å apart (center-to-center,
MM+, Hyperchem97) and have four intervening σ bonds. The
magnitude of this interaction is similar to that estimated
previously by our group for a MeOAn-6ANI-NI derivative
in which a TEMPO radical was rigidly bound to the terminus
of the NI acceptor group at a similar overall distance.41
Molecular modeling of 7 shows that the π systems of 6ANI
and NI are nearly perpendicular to that of the Ph bridge, so
enhanced spin-spin coupling via the π electrons is unlikely.51
Figure 7. Energy level diagram showing the spin manifolds of 7 after
charge separation and charge recombination in an external magnetic field
of about 0.35 T. Blue arrows denote charge separation forming the three-
spin RP state; orange arrows, charge recombination to the ground doublet
state; and magenta arrows, reversible D-D and Q-Q charge recombination
steps leading to the local triplet 3*NI, which is nearly isoenergetic with the
triradical state in toluene (see text). D1, D1′, and Q1 form a complete spin
basis set of the triradical state. Dashed lines mean less probable transitions.
Red double arrows stand for processes that mix certain Zeeman levels of
the relevant D and Q states. The size of the ellipse on each spin level
represents its population qualitatively.
Magnetic field effect51 and time-resolved EPR50 experiments
have shown that the exchange coupling 2J12 between MeOAn•+
and NI•- within the RP state of model compound 8 is 1.5 (
0.3 mT. MFE data presented in Figure 4 show that photoge-
neration of this same pair of radicals in 7 also yields 2J12
)
1.5 ( 0.3 mT. Finally, the exchange coupling 2J13 between
MeOAn•+ and NN• should also be ∼1 mT due to the similar
spin-spin distances and bonding framework linking MeOAn•+
and NN• compared to those for MeOAn•+ and NI•-.
These noncommutative exchange interactions remix the spin
manifold into two doublet states and a quartet state,32 shown in
Figure 7 as D1, D1′, and Q1 states. The energy splitting between
the D1 and D1′ (Q1) states is large due to the magnitude of 2J23,
while the splittings between the sublevels within D1′ and Q1
are small due to the values of 2J12 and 2J13. Thus, the well-
known S-T0 mixing mechanism51,86,98,99 of RP-ISC directly
translates into D′(1/2-Q(1/2 mixing in this context, and TREPR
spectra with the same polarization pattern should be observed.
This is a consequence of the fact that the Zeeman splitting
between the two spin sublevels of D1′, namely ∼0.35 T or
∼10-5 eV, is negligible compared to the overall free energy
change for charge separation, ∆G(*D f D1′) ) -0.8 eV (see
Figure 2), producing an unpolarized D1′ state (ignoring thermal
polarization), which is analogous to the singlet RP produced in
model systems such as 8 (Figure 7). The Q1 state is not
populated initially as a result of spin conservation. The D1 state,
which is energetically far removed from the D1′ and Q1 states,
does not mix with these states.
(90) Eaton, S. S.; Eaton, G. R. Biol. Magn. Reson. 2000, 19, 29-154.
(91) Marcus, R. A. J. Chem. Phys. 1965, 43, 679-701.
(92) Weiss, E. A.; Tauber, M. J.; Ratner, M. A.; Wasielewski, M. R. J. Am.
Chem. Soc. 2005, 127, 6052-6061.
(93) Kobori, Y.; Kawai, A.; Obi, K. J. Phys. Chem. 1994, 98, 6425-6429.
(94) Mori, Y.; Sakaguchi, Y.; Hayashi, H. J. Phys. Chem. A 2000, 104, 4896-
4905.
(95) Magin, I. M.; Shevel’kov, V. S.; Obynochny, A. A.; Kruppa, A. I.; Leshina,
T. V. Chem. Phys. Lett. 2002, 357, 351-357.
(96) Hasharoni, K.; Levanon, H.; Greenfield, S. R.; Gosztola, D. J.; Svec, W.
A.; Wasielewski, M. R. J. Am. Chem. Soc. 1995, 117, 8055-8056.
(97) Holl, N.; Port, H.; Wolf, H. C.; Strobel, H.; Effenberger, F. J. Chem. Phys.
1993, 176, 215-220.
The above reasoning is justified by TREPR and MFE results
for 7, which probe the D1′-Q1 splitting, ∆DQ, of the triradical
state.41 The experimental TREPR spectrum is simulated with a
two-state-mixing model similar to the one used for reference
(98) Hasharoni, K.; Levanon, H.; Greenfield, S. R.; Gosztola, D. J.; Svec, W.
A.; Wasielewski, M. R. J. Am. Chem. Soc. 1996, 118, 10228-10235.
(99) Closs, G. L.; Forbes, M. D. E.; Norris, J. R. J. Phys. Chem. 1987, 91,
3592-3599.
9
4362 J. AM. CHEM. SOC. VOL. 128, NO. 13, 2006

Stable NN• on a Donor
−Bridge−Acceptor Molecule
A R T I C L E S
compound 8,52 yielding ∆DQ ) 1.3 mT (Figure 5A). Also, the
sign of this splitting is confirmed to be positive, i.e., D1′ is higher
in energy than Q1, from the e a polarization pattern of the RP
spectrum, using the RP sign rule52,98 for field-swept spectra:
presence of the third spin on NN•, although it reshapes the spin
wave function, has negligible effects on orbital coupling and
the FCWD. Using the following spin-projection relationships
(see Supporting Information),
D0|D1′ 2 ) D2|D1 2 ) 3/4
D2|D1′ 2 ) D0|D1 2 ) 1/4
Q2|Q1 2 ) 1
(3a)
(3b)
(3c)
- e a
+ a e
Γ ) µ sign(∆DQ) )
(1)
{
where µ equals -1 or +1 for a doublet or a quartet precursor,
the former being the case. The magnitude of the D1′-Q1
splitting, |∆DQ| ) 1.5 mT ( 0.3 mT, is also obtained directly
from the maximum in the MFE plot shown in Figure 4, and is
in good agreement with the TREPR results. Since |∆DQ| for 7
and 2J for the photogenerated RP state in reference molecule 8
are the same within experimental error, the presence of the third
spin on NN• in the photogenerated triradical in 7 has essentially
no influence on the spin-spin exchange interaction 2J12 between
the two photogenerated radical ions.
the charge recombination rates from the three-spin state of 7
are related to kCR1 and kCR2 as shown in Figure 7. Note that eq
3c implies that the rate of charge recombination from the quartet
triradical state Q1 is the same as if the third spin does not exist:
kCR(Q1 f Q2) ) kCR2. In contrast, the total observed kCR from
the D1′ state turns into a weighted average of kCR1 and kCR2
:
Charge Recombination and Switching of the Spin Basis
Set. Based on previous studies of 851 and other closely related
molecules,41,75,100,101 RP-ISC in 7 is followed by charge
recombination, returning part of the population to the ground
state D0 and the remainder to a state having a local triplet
configuration on NI, 2,4(MeOAn-6ANI-Ph(NN•)-3*NI), which
is nearly isoenergetic with 2,4(MeOAn+•-6ANI-Ph(NN•)-
NI-•) in toluene. The neutral stable radical NN• is spin coupled
to NI-• with 2J23 ≈ 10-4 eV, and therefore it does not redirect
the charge recombination but establishes a new basis set of
doublet and quartet states, labeled in Figure 7 as D2 and Q2.
Again, no spin flip is considered during charge recombination,
and hence the Q2 state inherits the population of the (1/2
sublevels of Q1 generated by RP-ISC from D1′. However, the
D2 state becomes a mixture of the D1′ and D1 states, as pointed
out by Hoytink,102 because charge recombination is accompanied
by the rise of an extremely large 2J22, the local exchange
kCR(D1′ f D0 + D2) ) 3/4 kCR1 + 1/4 kCR2 (4)
compared to kCR1 for model compound 8. This recombination
rate is similar for the D1 state and also qualitatively correct if
an effective overall decay rate out of both D1 and D1′ states is
considered. Accordingly, overall charge recombination from the
triradical state is accelerated if kCR2 > kCR1 and otherwise
slowed. For 7, where 2,4(MeOAn+•-6ANI-Ph(NN•)-NI-•) is
in equilibrium with 2,4(MeOAn-6ANI-Ph(NN•)-3*NI), it has
been shown for closely related molecules that the net kCR2 is
slower than kCR1,
92,104 and thus the lifetime of the photogenerated
RP observed for 7 is extended by coupling to the third spin
(101 vs 73 ns), as determined here using both nanosecond
transient absorption and TREPR.
Dynamic Spin Polarization after Charge Recombination.
Based on the above analysis, the observed polarization of NN•
following charge recombination is unexpected (Figures 5B and
7), since all the charge separation and recombination steps as
well as the RP-ISC process itself in this system do not generate
a net oVerall spin polarization. The theory of chemically induced
dynamic electron polarization105 (CIDEP) has been well estab-
lished for doublet-doublet86,99,106 and triplet-doublet107-110
interactions, when these paramagnetic species diffuse in solution.
Generally, overall net spin polarization can be traced back to
nonadiabatic spin evolution, triggered by hyperfine interactions,
spin-orbit coupling, and/or zero-field splitting (ZFS). Gast and
Hoff111,112 observed that reduction of the ubiquinone electron
acceptor (UQ) in the primary electron transport cofactor
sequence of bacterial photosynthetic reaction center proteins,
(BChl)2-BPheo-UQ-•, followed by photogeneration of the
primary radical pair, (BChl)2+•-BPheo-•-UQ-•, results in a
net emissively polarized UQ-•, which they attributed to
polarization transfer from the nearby BPheo-• due to spin-
3
interaction between the two unpaired spins of *NI, which is
estimated to be 0.9 eV, the S1-T1 energy gap of NI.100 In other
words, D0, D2, and Q2 form a complete basis set for the
recombinant states, different from the other set, D1, D1′, and
Q1, which characterizes the triradical state.
Consequently, charge recombination from the RP state
requires switching or projection of the spin basis set. Typically,
charge recombination rates depend critically on the electronic
coupling matrix element V, including its spin component:
2
kCR ) (2π/p)|V| (FCWD)
(2)
where FCWD is the Franck-Condon weighted density of
states.103 Upon photoexcitation, model compound 8, which does
not have NN•, forms a two-spin RP with one singlet sublevel
1(MeOAn+•-6ANI-Ph-NI-•) and three triplet sublevels
3(MeOAn+•-6ANI-Ph-NI-•), which subsequently recombine
respectively to the ground-state singlet MeOAn-6ANI-Ph-
NI and the lowest triplet state MeOAn-6ANI-Ph-3*NI with
rates kCR1 and kCR2, respectively. Both recombination processes
conserve the spin wave function to the zeroth order. The
(104) Weiss, E. A.; Sinks, L. E.; Lukas, A. S.; Chernick, E. T.; Ratner, M. A.;
Wasielewski, M. R. J. Phys. Chem. B 2004, 108, 10309-10316.
(105) Muus, L. T.; Atkins, P. W.; McLauchlan, K. A.; Pedersen, J. B. Chemically
Induced Magnetic Polarization; Reidel: Dordrecht; Boston, 1977.
(106) Norris, J. R.; Morris, A. L.; Thurnauer, M. C.; Tang, J. J. Chem. Phys.
1990, 92, 4239-4249.
(107) Blattler, C.; Jent, F.; Paul, H. Chem. Phys. Lett. 1990, 166, 375-380.
(108) Kawai, A.; Okutsu, T.; Obi, K. J. Phys. Chem. 1991, 95, 9130-9134.
(109) Goudsmit, G. H.; Paul, H.; Shushin, A. I. J. Phys. Chem. 1993, 97, 13243-
13249.
(100) Wiederrecht, G. P.; Svec, W. A.; Wasielewski, M. R.; Galili, T.; Levanon,
H. J. Am. Chem. Soc. 2000, 122, 9715-9722.
(101) Greenfield, S. R.; Svec, W. A.; Wasielewski, M. R.; Hasharoni, K.;
Levanon, H. The Reaction Center of Photosynthetic Bacteria; Michel-
Beyerle, M.-E., Ed.; Springer: Berlin, 1996; pp 81-87.
(110) Blank, A.; Levanon, H. J. Phys. Chem. A 2001, 105, 4799-4807.
(111) Gast, P.; Hoff, A. J. Biochim. Biophys. Acta, Bioenergetics 1979, 548,
520-535.
(102) Hoytink, G. J. Acc. Chem. Res. 1969, 2, 114-120.
(103) Jortner, J. J. Chem. Phys. 1976, 64, 4860-4867.
(112) Hoff, A. J.; Gast, P. J. Phys. Chem. 1979, 83, 3355-3358.
9
J. AM. CHEM. SOC. VOL. 128, NO. 13, 2006 4363

A R T I C L E S
Chernick et al.
spin exchange coupling of BPheo-• with UQ-•. In addition,
Yamauchi85 has put forward a relaxation mechanism to account
for a two-phase polarization pattern where the sign of the overall
polarization switches from a to e or from e to a as time elapses,
but in this case a single rise and decay with no switching of the
sign of the transient polarization is observed. The experimental
observation that generation of dynamic spin polarization on NN•
is synchronous with the decay of the charge-separated state
slower charge recombination. The observed spin-spin exchange
interaction between the photogenerated radicals D+• and A-• is
not altered by the presence of the third spin, which only
accelerates radical pair intersystem crossing. Normally, the spin
dynamics of correlated radical pairs do not produce a net spin
polarization; however, net spin polarization appears on NN• with
the same time constant as describes the photogenerated radical
ion pair decay. This effect is attributed to antiferromagnetic
coupling between NN• and the local triplet state 3*NI, which is
populated following charge recombination. This requires an
effective switch in the spin basis set between the triradical and
the three-spin charge recombination product having both NN•
3
suggests that the recombinant local triplet *NI is the key to
the generation of net polarization. This can be explained by the
fact that the Hamiltonian that describes the ZFS of triplet states,
2
2
contains the double-quantum terms S+ and S- if the triplet
3
and *NI present. It remains to be seen what effect a radical
HZFS ) D(Sz2 - 1/3S2) + E(Sx2 - Sy2) )
having significant spin density delocalized onto the bridge has
on electron transfer. If the bridge within the D-B-A system
is itself a free radical, it is possible that enhanced charge and
spin transport over long distances s important for both molec-
ular electronics and photochemical energy conversion s may
result. Current studies are underway to develop a D-B-A
system that tests this limit.
D(Sz2 - 1/3S2) + 1/2E(S+2 + S- ) (5)
2
state is localized and E is significant. These two terms couple
the |D2 +1/2 sublevel, populated from the D1 and D1′ states,
and the vacant |Q2 -3/2 sublevel due to their proximity in
energy.108,109,113 Consequently, a net absorptive polarization
suggests that the energy level of Q2 lies above D2, corresponding
Acknowledgment. The work was supported by the National
Science Foundation under grant no. CHE-0415730 and by the
Defense Advanced Research Projects Agency. E.T.C. thanks
the Natural Sciences and Engineering Research Council of
Canada (NSERC). E.A.W. thanks the Presidential Fellowship
Program at Northwestern University for support of this work.
The Bruker E580 spectrometer was purchased with partial
support from NSF Grant No. CHE-0131048.
Supporting Information Available: Synthesis, 1H NMR,
EPR, and characterization details for all new compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
3
to antiferromagnetic coupling between *NI and NN•, so that
|D2 +1/2 gets depopulated by the S-2 term before the ground
D0 state inherits this spin polarization from D2 and remains
polarized until spin-lattice relaxation ensues.
Conclusion
We have demonstrated that the NN• radical within a
D-B(NN•)-A system having well-defined distances between
each component influences the spin dynamics of the photoge-
nerated triradical states 2,4[D+•-B(NN•)-A-•], resulting in
(113) Kobori, Y.; Takeda, K.; Tsuji, K.; Kawai, A.; Obi, K. J. Phys. Chem. A
1998, 102, 5160-5170.
JA0576435
9
4364 J. AM. CHEM. SOC. VOL. 128, NO. 13, 2006