Photo-initiated multi-step electron transfer in donor-acceptor systems using a novel bi-functionalized perylene chromophore

Article abstract of DOI:10.1016/j.cplett.2015.04.020

tThe excited state and redox properties of a new bi-functional perylene redox chromophore, 2,3-dihydro-1-azabenzo[cd]perylene (DABP), are described. Perylene has been widely used in electron donor-acceptormolecules in fields ranging from artificial photosynthesis to molecular spintronics. However, attachingmultiple redox components to perylene to carry out multi-step electron transfer reactions often produceshard to separate regioisomers, which complicate data analysis. The use of DABP provides a strategy toretain the electronic properties of perylene, yet eliminate regioisomers. Ultrafast photo-initiated single-and two-step electron transfer reactions in three linear electron donor-acceptor systems incorporatingDABP are described to illustrate its utility.

Full text of DOI:10.1016/j.cplett.2015.04.020

Chemical Physics Letters 629 (2015) 23–28
Contents lists available at ScienceDirect
Chemical Physics Letters
journal homepage: www.elsevier.com/locate/cplett
Photo-initiated multi-step electron transfer in donor–acceptor
systems using a novel bi-functionalized perylene chromophore
Scott M. Dyar, Amanda L. Smeigh, Steven D. Karlen, Ryan M. Young,
Michael R. Wasielewski∗
Department of Chemistry and Argonne-Northwestern Solar Energy Research (ANSER) Center, Northwestern University, Evanston, IL 60208-3113,
United States
a r t i c l e i n f o
a b s t r a c t
Article history:
The excited state and redox properties of a new bi-functional perylene redox chromophore, 2,3-dihydro-
1-azabenzo[cd]perylene (DABP), are described. Perylene has been widely used in electron donor–acceptor
molecules in fields ranging from artificial photosynthesis to molecular spintronics. However, attaching
multiple redox components to perylene to carry out multi-step electron transfer reactions often produces
hard to separate regioisomers, which complicate data analysis. The use of DABP provides a strategy to
retain the electronic properties of perylene, yet eliminate regioisomers. Ultrafast photo-initiated single-
and two-step electron transfer reactions in three linear electron donor–acceptor systems incorporating
DABP are described to illustrate its utility.
Received 5 March 2015
In final form 7 April 2015
Available online 21 April 2015
© 2015 Elsevier B.V. All rights reserved.
1. Introduction
a bi-functional perylene having a single regioisomer [10,11];
however, the synthesis of these derivatives often yields isomer
The rational design of molecular systems capable of photo-
initiated charge separation followed by charge and spin transport is
important in fields ranging from artificial photosynthesis to spin-
tronics. Perylene is an aromatic hydrocarbon that absorbs visible
light at 400–450 nm to produce a 2.85 eV excited singlet state (S1)
having both electron donor and acceptor characteristics [1,2]. The
perylene S1 state can provide significant energy to drive photo-
induced electron transfer reactions [3,4] that have proven useful to
solar fuels [5], artificial photosynthesis [6,7], and molecule-based
spintronics research [8]. The perylene S1 state has a prominent
absorption band near 700 nm [3], while its radical cation absorbs
near 550 nm [8]. These distinct spectroscopic features provide a
convenient way to identify these intermediates in photo-driven
charge transfer reactions involving perylene.
mixtures that are difficult to separate [12]. In this study, we
describe the properties of 2,3-dihydro-1-azabenzo[cd]perylene
(DABP, Scheme 1), a new bi-functional perylene that overcomes
the regioisomer problem, yet preserves the desirable electronic
properties of the perylene chromophore and can readily be incor-
porated into multi-step electron donor–acceptor systems. We also
describe photo-initiated electron transfer reactions in three elec-
tron donor–acceptor systems incorporating DABP (1–3, Scheme 1),
which were studied using femtosecond transient absorption (fsTA),
nanosecond transient absorption (nsTA), and transient electron
paramagnetic resonance (TREPR) spectroscopies.
2. Experimental
It is often desirable to construct photo-initiated charge separa-
tion systems having multiple electron transfer steps to prolong the
lifetime of the resulting radical ion pair (RP), a strategy inspired
by photosynthetic reaction center proteins [9]. Incorporation of
perylene into designs having strict control over donor–acceptor
distances and orientations is complicated by the need to use
2.1. Synthesis
The synthesis of 1–3 is summarized in Scheme 2 and is described
explicitly in Appendix A.
2.2. Electrochemistry
Electrochemical measurements were performed on a CH Instru-
ments model 660A electrochemical workstation. Samples were
measured in a solution of 0.1 M tetra-n-butylammonium hexaflu-
orophosphate (TBAPF6) in acetonitrile purged with N2 to remove
∗
Corresponding author at: Department of Chemistry, Northwestern University,
2145 Sheridan Rd., Evanston, IL 60208-3113, United States.
E-mail address: m-wasielewski@northwestern.edu (M.R. Wasielewski).
http://dx.doi.org/10.1016/j.cplett.2015.04.020
0009-2614/© 2015 Elsevier B.V. All rights reserved.

24
S.M. Dyar et al. / Chemical Physics Letters 629 (2015) 23–28
absorption (fsTA) measurements were made using the 130 fs,
416 nm frequency-doubled output from a 2 kHz regeneratively
amplified Ti:sapphire laser system as the pump [13,14]. A white
light continuum probe pulse was generated by focusing the IR fun-
damental into a 1 mm sapphire disk [15]. Samples were placed in a
2 mm path length quartz cuvette and irradiated with 1.0 ␮J, 416 nm
laser pulses focused to a 200 ␮m spot. Typically, 5–7 s of averaging
was used to obtain the transient spectrum at a given delay time.
The total instrument response for the pump-probe experiment was
180 fs. The optical density of all samples was maintained between
0.4 and 0.8 at 400 nm for both femtosecond and nanosecond tran-
sient absorption.
Transient absorption spectra were analyzed via singular value
decomposition (SVD) using lab-written Matlab [16] programs. SVD
deconvolutes the two-dimensional spectra to produce an orthonor-
mal set of basis spectra which describe the wavelength dependence
of the species and a corresponding set of orthogonal vectors which
describes the time dependent amplitude of the basis spectra [17].
A species-associated first order kinetic model [18] was fit to a lin-
ear combination of the time-dependent amplitude vectors and the
same linear combination of basis spectra was used to construct the
spectra for the chemical species.
Nanosecond transient absorption measurements were per-
formed using a previously described frequency-tripled Nd:YAG
laser (Continuum, Precision II 8000), coupled to an optical para-
metric oscillator (Continuum, Panther) [5]. Low temperature
experiments at 85 K were performed using a Janis VNF-100 N2
vapor-cooled variable temperature cryostat and held at the desired
temperature throughout the duration of the experiment. Samples
were prepared in a nitrogen-filled glove box and placed between
two quartz windows in a 2 mm path length lab-built sample holder.
Kinetic traces were collected from 430 to 800 nm at 5 nm inter-
vals, and spectra were constructed by merging the kinetic traces
(binned to 12-ns steps). Each kinetic trace used to construct the
Scheme 1. Structure of DABP and donor–acceptor molecules 1–3.
oxygen. A 1.0 mm diameter glassy carbon electrode, platinum wire
counter electrode, and silver wire reference electrode were used.
The ferrocene/ferrocinium couple was used as an internal refer-
ence.
2.3. Optical spectroscopy
Ground-state absorption measurements were made on
Shimadzu UV-1601 spectrophotometer. Femtosecond transient
a
Scheme 2. (a) 2,6-Diisopropylaniline, zinc acetate dihydrate, imidazole, H2O, 190 ◦C, 24 h, 61%; (b) bromine, CH2Cl2, reflux, 3 h, 69%; (c) p-(3,3,4,4-tetramethyl-2,5-
dioxaborolyl)nitrobenzene, K2CO3, Pd(PPh3)4, toluene, ethanol, H2O, 60 ◦C, 20 h, 92%; (d) (i) 1 M THF•BH3, THF, reflux, 48 h (ii) N-(2,5-di-t-butylphenyl)-naphthalene-
1,8-dicarboximide-4,5-dicarboxyanhydride, pyridine, 140 ◦C, 20 h, 19%; (e) (i) KOH, t-butanol, 100 ◦C, 3 h, (ii) 3,4,5-trimethoxyaniline, imidazole, 140 ◦C, 20 h, (iii) 1 M
THF•BH3, THF, reflux, 48 h (iv) N-(2,5-di-t-butylphenyl)-naphthalene-1,8-dicarboximide-4,5-dicarboxyanhydride, pyridine, 140 ◦C, 20 h, 4%; (f) (i) KOH, t-butanol, 100 ◦C,
3 h, (ii) N,N-dimethyl-p-phenylenediamine, imidazole, 140 ◦C, 20 h, (iii) 1 M THF•BH3, THF, reflux, 48 h, (iv) N-(2,5-di-t-butylphenyl)-naphthalene-1,8-dicarboximide-4,5-
dicarboxyanhydride, pyridine, 140 ◦C, 20 h, 2%.

S.M. Dyar et al. / Chemical Physics Letters 629 (2015) 23–28
25
spectra is representative of an average of 100 laser shots. In order
1
1.0
0.8
0.6
0.4
0.2
0.0
to obtain decay rates, single wavelength kinetics were fit with a
Levenberg–Marquardt nonlinear least squares fit to a sum of expo-
nentials convoluted with a Gaussian instrument response function.
Perylene
2.4. EPR spectroscopy
EPR measurements X-band (9.5 GHz) were made using a Bruker
Elexsys E680-X/W EPR spectrometer outfitted with a variable Q
dielectric resonator (ER-4118X-MD5-W1) at X-band and a cylindri-
cal resonator (EN-680-1021H) at W-band. For EPR measurements
at X-band, toluene solutions (∼10−4 M) were loaded into quartz
tubes (4 mm o.d. × 2 mm i.d.), subjected to four freeze-pump-thaw
degassing cycles on a vacuum line (10−4 Torr), and sealed using
a hydrogen torch. The EPR samples were stored in a dark freezer
when not in use.
300
400
500
600
Wavelength (nm)
Figure 1. UV/vis absorption spectra of 1 (blue) and perylene (red) in toluene.
(DMAA) before treatment with THF-BH3 and condensation with
NIA to form 2 or 3, respectively.
Transient EPR measurements were performed at X-band fol-
lowing photoexcitation with 7 ns, 3 mJ 416 nm pulses using the
output of an optical parametric oscillator (SpectraPhysics Basi-
scan), pumped with the 355 nm frequency-tripled output of a
Nd:YAG laser (SpectraPhysics Quanta-Ray Pro 350). Transient CW
EPR spectra were collected following photoexcitation, the kinetic
traces of the transient magnetization were acquired in quadrature
under CW irradiation (2–20 mW). Sweeping the magnetic field gave
2D spectra with respect to both time and magnetic field. For each
kinetic trace, the signal acquired prior to the laser pulse was sub-
tracted from the data. Kinetic traces recorded at magnetic field
values off-resonance were considered background signals, whose
average was subtracted from all kinetic traces.
3.2. Steady state measurements
Figure 1 shows the UV–vis absorption spectrum of 1 (blue),
which includes both DABP (400–500 nm) and NDI absorptions
(300–400 nm) [27]. The DABP absorption is red shifted by 22 nm
relative to that of unsubstituted perylene (red). Spectra of 2 and 3
are provided in Figure A1 (Appendix A). The corresponding emis-
sion spectrum of 1 in toluene is shown in Figure A2A (Appendix
A). The energy of the lowest excited singlet state of DABP (1*DABP),
obtained from the average of the (0,0) vibronic band energies of
the absorption and emission spectra, is 2.60 eV, which is somewhat
lower than that of unsubstituted perylene, 2.85 eV [2].
2.5. Computational methods
The one-electron redox potentials of 1–3 obtained by cyclic
voltammetry are given in Table 1. The NDI potentials are consis-
tent across 1–3; while the oxidation and reduction potentials of
DABP vary over a range of 26 mV depending on the secondary donor
attached to DABP through its nitrogen atom.
The geometry of the ground state singlet of 1–3 was ini-
tially relaxed using molecular mechanics with the MMFF294 force
field, as implemented in the Avogadro 1.1.0 software [19], then
subsequently relaxed using density functional theory (DFT), as
implemented in the TeraChem 1.5 K software [20]. All DFT calcu-
lations made use of the unrestricted B3LYP exchange-correlation
functional with a split-valence double zeta basis set with added
polarization functions (6-31G*). Images of the optimized structures
were generated with PyMol 1.2r1 [21]. Molecular orbitals were
printed through single point energy calculations with TeraChem,
using DFT (B3LYP/6-31G*), and visualized with VMD 1.9.2a27 [22].
3.3. Electron transfer dynamics
Femtosecond transient absorption (fsTA) was used to probe
the photo-initiated electron transfer dynamics of DABP systems
1–3 (Figure 2). Following selective excitation of DABP at 416 nm,
the initial transient absorption spectrum in each case exhibits an
absorption centered around 735 nm, characteristic of the perylene
excited state, S1 [2]. The Sn ← S1 absorption in all three cases decays
quickly due to the initial charge separation reaction (CS1).
3. Results and discussion
3.1. Synthesis
The three-dimensional transient absorption data sets were
analyzed using singular value decomposition (SVD) and global anal-
ysis (Figure A4, Appendix A) yielding, in the cases of 1 and 2,
only two species-associated components. The first component in
molecules 1 and 2 shows that 1*DABP decays in ꢀCS1 = 2.08 0.03
and 2.20 0.02 ps, respectively, concomitant with the rise of
absorptions at 480 and 610 nm characteristic of NDI•− [29], as well
N-(2,6-diisopropylphenyl)-PMI
one-pot monodecarboxylation
was
of
synthesized
perylene-3,4:9,10-bis-
by
(dicarboxyanhydride) and imide condensation of the remaining
anhydride with 2,6-diisopropylaniline (Scheme 2) [23]. This
species was mono-brominated using molecular bromine in
dichloromethane to selectively form N-(2,6-diisopropylphenyl)-
9-bromo-PMI. Suzuki coupling of the brominated PMI to
p-(3,3,4,4-tetramethyl-2,5-dioxaborolyl)nitrobenzene
yielded N-(2,6-diisopropylphenyl)-9-nitrophenyl-PMI.
derivative was subjected to exhaustive reduction of its
carbonyl groups with THF-BH3 complex [25] followed by
condensation with N-(2,5-di-t-butylphenyl)-naphthalene-1,8-
as a broad absorption from 500 to 650 nm assigned to DABP•+
,
which is consistent with spectroelectrochemistry on 2 (Figure A3,
Appendix A). For 1 and 2, the second species-associated compo-
nent represents DABP•+-NDI•− decay to ground state with lifetimes
of ꢀCR = 960 10 ps and ꢀCR = 1.51 0.03 ns, respectively. No hole
transfer reaction from DABP•+ to the TMA secondary donor was
observed in 2.
[24]
This
dicarboximide-4,5-dicarboxyanhydride
(NIA)
[26],
leaving
In the case of 3, global analysis yields three components:
the first species-associated component is assigned to 1*DABP
decay and the associated rise of an intermediate state, which
occurs in ꢀCS1 = 0.53 0.02 ps. The intermediate component, which
2,6-diisopropylaniline (An) in place on DABP to yield 1. The
remaining molecules were synthesized by first hydrolyzing
N-(2,6-diisopropylphenyl)-9-nitrophenyl-PMI to 9-nitrophenyl-
perylene-3,4-dicarboxyanhydride, then condensing it with either
3,4,5-trimethoxyaniline (TMA) or p-(N.N-dimethylamino)aniline
does not correspond with DABP•+, quickly decays to NDI•−
.
The NDI•− signal in this case lives significantly longer than

26
S.M. Dyar et al. / Chemical Physics Letters 629 (2015) 23–28
Table 1
One-electron redox potentials of 1–3a in V vs. SCE.
ERed (NDI0/−
)
ERed (NDI−/2−
)
ERed (DABP0/−
)
EOx (Donor0/+
)
EOx (DABP0/+
)
1
2
3
−0.57
−0.53
−0.55
−1.02
−0.96
−1.01
−1.77
−1.67
−1.53
−
0.94b
1.08b
1.20b
0.63b
0.11b
a
Recorded in acetonitrile with 0.1 M tetrabutylammonium hexafluorophosphate, a 1.0 mm diameter glassy carbon disk working electrode, a Pt wire counter electrode,
and Ag wire pseudo-reference electrode. The ferrocene/ferrocenium redox couple (0.4 V vs. SCE) [28] was used as an internal standard.
b
Irreversible couple.
C
B
A
0.08
0.06
0.04
0.02
0.00
-0.02
-0.04
-0.06
0.06
0.04
0.02
0.00
-0.02
-0.04
-0.06
0.06
0.04
0.02
0.00
-0.02
-0.04
-0.06
-0.08
-1 ps
200 ps
1000 ps
2000 ps
-1 ps
39 ps
-1 ps
200 ps
1000 ps
2000 ps
0.4 ps
4.4 ps
25 ps
0.3 ps
0.9 ps
2.4 ps
9 ps
300 ps
1500 ps
5000 ps
0.4 ps
4.4 ps
25 ps
500
600
700
800
500
600
700
800
500
600
700
800
Wavelength (nm)
Wavelength (nm)
Wavelength (nm)
Figure 2. Femtosecond transient absorption spectra of (A) 1, (B) 2, and (C) 3 in toluene, exciting at ꢁex = 416 nm (1.0 ␮J/pulse).
Table 2
clearly shows why the CS order is changed relative to what is
observed in 1 and 2. ꢂGCS1 to form DMAA–DABP•+–NDI•− lies well
above the 1*DABP energy. However, DMAA is easy to oxidize, so
that DMAA+•–DABP−•–NDI is accessible. After the initial CS1 step,
CS2 to form DMAA+•–DABP–NDI−• is energetically favorable.
Density functional theory (DFT) was applied to 1–3 to calculate
the frontier orbitals of 1–3, which are given in Appendix A. For 1–3,
the LUMO is consistently localized on NDI. For 1 and 2, the HOMO is
localized on DABP, which is expected for 1, and further supports the
inability of 2 to undergo CS2. For 3, the HOMO is localized on DMAA,
as expected, based on the observation of full charge separation after
photoexcitation.
Times of initial charge separation (CS1), charge shift (CS2), and charge recombina-
tion (CR) for 1–3, obtained via SVD and global fitting.
Molecule
ꢀCS1
(ps)
ꢀCS2
(ps)
ꢀCR
(ns)
1
2
3
2.08 0.03
2.20 0.02
0.53 0.02
–
–
0.96 0.01
1.51 0.03
5.23 0.19
4.76 0.08
Table 3
ꢂGIP of relevant CS states of 1–3.
Ion pair
ꢂGIP
(eV)
1
2
2
3
3
3
An–DABP+•–NDI−•
2.53
2.63
2.46
2.77
2.55
1.98
3.4. Spin dynamics of 3 at 85 K
TMA–DABP+•–NDI−•
TMA•+–DABP–NDI•−
DMAA–DABP+•–NDI−•
DMAA+•–DABP−•–NDI
DMAA+•–DABP–NDI−•
The photo-generated singlet RP, 1(DMAA+•–DABP–NDI−•) may
undergo radical pair intersystem crossing (RP-ISC) driven largely
by electron-nuclear hyperfine couplings in a few nanoseconds to
produce the triplet RP, 3(DMAA+•–DABP–NDI−•) [31,32]. The long
distance between the two radicals results in a small spin–spin
dipolar interaction (d) between the two radicals, which is mani-
fest in a small zero-field splitting of the triplet RP sublevels. The
dipolar interaction is averaged out in solution, but preserved when
the RP is dissolved in a glassy solid. In addition, the energy gap
between the nearly degenerate three triplet RP sublevels and the
singlet RP is determined by the spin–spin exchange interaction (2J,
Figure 3A). The subsequent RP recombination process is spin selec-
tive in that the singlet RP recombines to the singlet ground state,
whereas the triplet RP recombines to a locally excited neutral triplet
state, if this state is lower in energy than the triplet RP [33]. For
DMAA–DABP–NDI, the lowest excited triplet state resides on DABP,
which has a 1.5 eV energy similar to that of perylene [34]; thus, the
it does within the DABP•+–NDI•− state observed in
1 and
2. Based on these differences, the intermediate component is
assigned to DMAA•+–DABP•−, formed via electron transfer from
DMAA to 1*DABP in ꢀCS1 = 0.53 0.02 ps. From this intermediate
state, a secondary charge shift reaction (CS2) occurs, yielding
DMAA+•–DABP–NDI−• RP, which recombines to ground state in
ꢀCR = 5.23 0.19 ns. A summary of all CS and charge recombination
(CR) time constants is given in Table 2.
To better understand these electron transfer dynamics, the
free energy changes for the formation of relevant ion-pair states
(ꢂGIP) in toluene were estimated using the Weller model [30]
(Appendix A). These results (Table 3) provide interesting insights
into the CS and CR dynamics presented above. The 2.60 eV energy
of 1*DABP is only slightly higher than ꢂGIP for An-DABP•+–NDI•−
and is nearly isoenergetic with that of TMA–DABP•+–NDI•−, both of
which form after selective excitation of DABP in 1 and 2. While the
results shown in Table 3 suggest that CS2 to TMA•+–DABP–NDI•−
in 2 should be accessible, ꢂGCS2 is only −0.14 eV, which, in this
case results in a rate that is too slow to compete with CR of
TMA–DABP•+–NDI•−. Analysis of the calculations relevant to 3
∼2 eV energies of the triplet sublevels of 3(DMAA+•–DABP–NDI–•
)
are about 0.5 eV above 3*DABP, so that charge recombination via the
triplet channel can populate DMAA–3*DABP–NDI. It is worth not-
ing that the observed DMAA+•–DABP–NDI−• lifetime is only about a
factor of 3–5 longer than that of the various initial photo-generated
RPs that occur in 1–3. This mostly likely results from the fact that
ꢂG for the triplet RP recombination reaction occurs in the Marcus
normal region of the rate vs. free energy profile, in contrast to the

S.M. Dyar et al. / Chemical Physics Letters 629 (2015) 23–28
27
a)
b)
T+1
B(2J)
a
a
2J
ΦB
ΦA
e
e
T−1
Magnetic Field Strength (B)
Figure 3. (a) Zeeman splitting of RP energy levels (J > 0); (b) relative energies of the RP spin states. Radical pair energy levels and transitions in the high magnetic field limit
following S–T0 mixing to yield ˚A and ˚B, which are initially populated. The arrows indicate the microwave-induced transitions responsible for the observed spin-polarized
TREPR spectrum.
singlet RP recombination reaction that occurs deep into the Marcus
inverted region [35,36].
where ꢄ0, ge and ˇe are the vacuum permeability, electronic g-
˚
factor and Bohr magneton, respectively. In units of mT and A,
3
˚
TREPR spectroscopy using pulsed laser excitation and
continuous microwaves was employed to directly examine
DMAA+•–DABP–NDI–•. At the ∼340 mT magnetic field character-
istic of EPR measurements at X-band, the three RP triplet states
are Zeeman split resulting in the four state energy level diagram
illustrated in Figure 3B in which RP-ISC results in S-T0 mixing to
produce the coherent superposition states |˚Aꢀ and |˚Aꢀ [37].
d = −2785 mT A/r . When a RP undergoes RP-ISC by S–T0 mixing
followed by charge recombination to yield the neutral triplet state,
the six EPR transitions at the canonical (x,y,z) orientations relative
to the applied magnetic field of the recombination triplet spectrum
exhibit a unique (a,e,e,a,a,e) (low field to high field) polarization
pattern [38].
Nanosecond transient absorption (nsTA) was performed on
3 in a 2-methyltetrahydrofuran (2-MTHF) glass at 85 K (see
Appendix A). The DMAA+•–DABP–NDI−• RP forms within the
7 ns instrument response of the experiment and decays in
ꢀCR = 16.7 ␮s. Photoexcitation of 3 at 85 K yields a long-lived
spin-correlated RP with a TREPR spectrum visible immediately
following the instrument response (∼70 ns) having an (e,a) pattern
[39], which is superimposed on the spin-polarized spectrum of
DMAA-3*DABP-NDI (Figure 4A). The RP spectrum decays as the
neutral triplet spectrum appears and is followed by final decay
of the neutral triplet spectrum. Figure 4A gives the full spectrum
collected at 160 ns following the laser pulse, while Figure 4B
gives an expanded, baseline-corrected spectrum focusing on the
narrower RP signal from the same spectrum. Simulation of the
RP spectrum yields |J| = 0.7 mT and d = −0.2 mT. Using Eq. (1), the
Microwave-induced transitions between these states and the |T+1
ꢀ
and |T−1ꢀ states result in a spin-polarized EPR spectrum with four
lines having a symmetric (e,a,e,a) anti-phase pattern (where e
denotes emission and a denotes enhanced absorption, low to high
field), provided that J is positive and larger than d. If the g-factors
of the two radicals are similar and/or the transitions are split by
hyperfine coupling, the two doublets will overlap significantly and
appear as a distorted (e,a) signal [31,32]. The spin polarization
pattern (whether a transition is e or a) is also determined by the
sign rule, ꢃ = ꢄ·sign[2J–d(3cos2ꢅ − 1)], where ꢄ = −1 or +1, if the
spin-correlated RP is produced from a singlet or triplet precursor,
respectively, and ꢅ is the angle between the vector connecting
the two spins and the externally applied magnetic field [37]. For
ꢃ = (−) the observed phase is (e,a) whereas for ꢃ = (+) the phase is
(a,e). The magnitude of J depends exponentially on the distance, r,
between the individual radicals comprising the radical pair, while
the dipolar interaction d can be determined using the point dipole
approximation:
+•
˚
dipolar interaction yields a 24 A distance between DMAA and
NDI−• in 3, which is very close to the 23 A RP distance predicted
˚
using DFT (B3LYP/6-31G*) calculations (Table A2, Appendix A).
Simulation of the TREPR spectrum of DMAA-3*DABP-NDI yields
zero-field splittings D = 56.4 mT and |E| = 2.7 mT with sublevel
populations Tx = 0.45, Ty = 0.61, and Tz = 0.45. This simulation
shows that DMAA–3*DABP–NDI is formed by contributions from
3ꢄ0ge2ˇe2
8ꢆr3
d = −
(1)
B
A
3
Fit
3
Fit
1.0
0.5
1.0
0.5
0.0
0.0
-0.5
-1.0
-1.5
-0.5
-1.0
-1.5
330
335
340
345
350
355
280 300 320 340 360 380 400
Magnetic Field (mT)
Magnetic Field (mT)
Figure 4. (A) Full and (B) expanded spectra of 3 in 2-MTHF (85 K) 160 ns after excitation at ꢁex = 450 nm (1.8 mJ/pulse). The red, superimposed curves are simulations
performed with parameters given in the text.

28
S.M. Dyar et al. / Chemical Physics Letters 629 (2015) 23–28
both the RP-ISC and SO-ISC mechanisms. RP recombination of
DMAA+•–DABP–NDI–• via RP-ISC to DMAA–3*DABP–NDI (58%) is
accompanied by fast triplet formation via charge recombination
of the initial singlet RP state DMAA•+–DABP•−–NDI directly to
DMAA–3*DABP–NDI (42%) by the so-called spin–orbit charge
transfer (SOCT) mechanism, which has been observed previously
[39]. The data imply that at 85 K the quantum yield of complete
charge separation in 3 to yield DMAA+•–DABP–NDI−• is 58%.
[5] B.S. Veldkamp, W.S. Han, S.M. Dyar, S.W. Eaton, M.A. Ratner, M.R. Wasielewski,
Energy Environ Sci 6 (2013) 1917.
[6] S.M. Dyar, J.C. Barnes, M. Juricek, J.F. Stoddart, D.T. Co, R.M. Young, M.R.
Wasielewski, Angew. Chem. Int. Ed. Engl. 53 (2014) 5371.
[7] R.M. Young, S.M. Dyar, J.C. Barnes, M. Juricek, J.F. Stoddart, D.T. Co, M.R.
Wasielewski, J. Phys. Chem. A 117 (2013) 12438.
[8] L. Kobr, D.M. Gardner, A.L. Smeigh, S.M. Dyar, S.D. Karlen, R. Carmieli, M.R.
Wasielewski, J. Am. Chem. Soc. 134 (2012) 12430.
[9] R.E. Blankenship, Molecular Mechanisms of Photosynthesis, Blackwell Science,
Oxford, 2002.
[10] H.-F. Chen, D.M. Gardner, R. Carmieli, M.R. Wasielewski, Chem. Commun. 49
(2013) 8614.
[11] T. Miura, R. Carmieli, M.R. Wasielewski, J. Phys. Chem. A 114 (2010) 5769.
[12] A. Zinke, F. Linner, O. Wolfbauer, Chem. Ber. 58 (1925) 323.
[13] E.A. Weiss, M.J. Ahrens, L.E. Sinks, A.V. Gusev, M.A. Ratner, M.R. Wasielewski, J.
Am. Chem. Soc. 126 (2004) 5577.
[14] S.R. Greenfield, M.R. Wasielewski, Opt. Lett. 20 (1995) 1394.
[15] J.M. Giaimo, A.V. Gusev, M.R. Wasielewski, J. Am. Chem. Soc. 124 (2002) 8530.
[16] MATLAB, The MathWorks, Inc., Natick, MA, 2013.
[17] E.R. Henry, J. Hofrichter, Methods Enzymol. 210 (1992) 129.
[18] T.M. Zamis, L.J. Parkhurst, G.A. Gallup, Comput. Chem. (Oxford) 13 (1989) 165.
[19] M.D. Hanwell, K. Lutz, Abstracts of Papers of the American Chemical Society,
vol. 244, 2012.
4. Conclusions
The DABP chromophore can be synthesized from well-known
starting materials and is easily incorporated into systems compris-
ing common donors, acceptors, and chromophores. DABP performs
similarly to, but offers advantages over perylene, particularly
by avoiding regioisomers in the preparation of bi-functionalized
perylene-based donor–acceptor triads, by providing accessible
connectivity to a secondary donor, and by facilitating charge-shift
reactions to a secondary donor that incorporates the electron-rich
amine of DABP as part of its structure. Thus DABP is a new build-
ing block with significant advantages for the rational design of
electron-transfer systems having controlled distances and orien-
tations between their redox partners.
[20] I.S. Ufimtsev, T.J. Martinez, J. Chem. Theory Comput. 5 (2009) 2619.
[21] L.L.C. Schrodinger, The PyMOL Molecular Graphics System, Version 1.3r1, 2010.
[22] W. Humphrey, A. Dalke, K. Schulten, J. Mol. Graphics Modell. 14 (1996) 33.
[23] K.M. Lefler, K.E. Brown, W.A. Salamant, S.M. Dyar, K.E. Knowles, M.R.
Wasielewski, J. Phys. Chem. A 117 (2013) 10333.
[24] M.J. Ahrens, L.E. Sinks, B. Rybtchinski, W.H. Liu, B.A. Jones, J.M. Giaimo, A.V.
Gusev, A.J. Goshe, D.M. Tiede, M.R. Wasielewski, J. Am. Chem. Soc. 126 (2004)
8284.
[25] S.Y. Hsu, L.Y. Chiang, Synth. Commun. 19 (1989) 1885.
[26] R. Carmieli, Q.X. Mi, A.B. Ricks, E.M. Giacobbe, S.M. Mickley, M.R. Wasielewski,
J. Am. Chem. Soc. 131 (2009) 8372.
Acknowledgement
[27] M.T. Colvin, R. Carmieli, T. Miura, S. Richert, D.M. Gardner, A.L. Smeigh, S.M.
Dyar, S.M. Conron, M.A. Ratner, M.R. Wasielewski, J. Phys. Chem. A 117 (2013)
5314.
This work was supported by the National Science Foundation
under grant no. CHE-1266201.
[28] N.G. Connelly, W.E. Geiger, Chem. Rev. 96 (1996) 877.
[29] D. Gosztola, M.P. Niemczyk, W. Svec, A.S. Lukas, M.R. Wasielewski, J. Phys.
Chem. A 104 (2000) 6545.
Appendix A. Supplementary data
[30] A. Weller, Z. Phys. Chem. 133 (1982) 93.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.cplett.2015.04.020
[31] G.L. Closs, M.D.E. Forbes, J.R. Norris, J. Phys. Chem. 91 (1987) 3592.
[32] P.J. Hore, D.A. Hunter, C.D. Mckie, A.J. Hoff, Chem. Phys. Lett. 137 (1987) 495.
[33] E.A. Weiss, M.A. Ratner, M.R. Wasielewski, J. Phys. Chem. A 107 (2003) 3639.
[34] S.L. Murov, I. Carmichael, G.L. Hug, Handbook of Photochemistry, Marcel
Dekker, New York, 1993.
References
[35] R.A. Marcus, J. Chem. Phys. 24 (1956) 966.
[36] R.A. Marcus, J. Chem. Phys. 43 (1965) 679.
[37] P.J. Hore, in: A.J. Hoff (Ed.), Advanced EPR in Biology and Biochemistry, Elsevier,
Amsterdam, 1989, p. 405.
[38] H. Levanon, J.R. Norris, Chem. Rev. 78 (1978) 185.
[39] Z.E.X. Dance, S.M. Mickley, T.M. Wilson, A.B. Ricks, A.M. Scott, M.A. Ratner, M.R.
Wasielewski, J. Phys. Chem. A 112 (2008) 4194.
[1] C.R. Goldschmidt, M. Ottolenghi, J. Phys. Chem. 75 (1971) 3894.
[2] N. Sabbatini, M.T. Indelli, M.T. Gandolfi, V. Balzani, J. Phys. Chem. 86 (1982)
3585.
[3] W.B. Davis, M.A. Ratner, M.R. Wasielewski, Chem. Phys. 281 (2002) 333.
[4] K.E. Brown, B.S. Veldkamp, D.T. Co, M.R. Wasielewski, J. Phys. Chem. Lett. 3
(2012) 2362.