Spin-Allowed Transitions Control the Formation of Triplet Excited States in Orthogonal Donor-Acceptor Dyads

Article abstract of DOI:10.1016/j.chempr.2018.10.001

Reliance on triplet excited states (triplets) of molecules with heavy atoms, such as precious metals, limits their potential in technological applications. We envision that triplets of π-conjugated organic molecules could play bigger roles; however, their production without heavy atoms remains challenging. The direct, spin-forbidden conversion of singlet charge-separated states to triplets in an electron donor-acceptor (D-A) pair is a promising approach. Here, using a series of orthogonal D-A type boron dipyrromethene (BODIPY) derivatives as a model system, we show that the formation of triplets is largely controlled by the spin-allowed transitions. Yet, this spin-forbidden process can still proceed much faster than ordinary intersystem crossing between (π π*) states under favorable conditions because of stronger spin-orbit coupling. Our findings reveal a clear physical basis for this spin-forbidden process and provide guidelines for future molecular designs exploiting the process. Production of triplet excited states of π-conjugated organic molecules in high yields without using heavy atoms remains challenging. The direct formation of triplet excited states from singlet charge-separated states is a promising approach. Here, we show that spin-allowed electron-transfer reactions largely control such a formation, yet the spin-forbidden transition can outcompete the spin-allowed transitions under favorable conditions because of stronger spin-orbit coupling. Triplet excited states (triplets) serve as key intermediates in critical technologies and processes ranging from organic synthesis to biomedicine to molecular electronics. Production of triplets of π-conjugated organic molecules without heavy atoms remains challenging. Spin-orbit, charge-transfer intersystem crossing (SOCT-ISC) directly converts singlet charge-separated states to triplets in an electron donor-acceptor (D-A) pair. Here, using a series of orthogonal D-A type boron dipyrromethene (BODIPY) derivatives as a model system, we show that the formation of triplets is largely controlled by the spin-allowed transitions rather than by SOCT-ISC. Yet, the SOCT-ISC process can still proceed much faster than ordinary ISC between (π π*) states because the spin-orbit coupling of SOCT-ISC is 2 orders of magnitude stronger. We further show that such a process can produce triplets in a non-triplet-forming molecule, perylene. Our findings reveal a clear physical basis for this spin-forbidden process and provide guidelines for future molecular designs exploiting the process.

Full text of DOI:10.1016/j.chempr.2018.10.001

Article
Spin-Allowed Transitions Control the
Formation of Triplet Excited States in
Orthogonal Donor-Acceptor Dyads
Jason T. Buck, Andrew M.
Boudreau, Andre´ DeCarmine,
Reid W. Wilson, James
Hampsey, Tomoyasu Mani
tomoyasu.mani@uconn.edu
HIGHLIGHTS
Spin-allowed electron-transfer
reactions control the formation of
triplets
The Marcus inverted region is
used to suppress spin-allowed
transitions
Determination of spin-orbit
coupling constant
Production of triplet excited states of p-conjugated organic molecules in high
yields without using heavy atoms remains challenging. The direct formation of
triplet excited states from singlet charge-separated states is a promising
approach. Here, we show that spin-allowed electron-transfer reactions largely
control such a formation, yet the spin-forbidden transition can outcompete the
spin-allowed transitions under favorable conditions because of stronger spin-orbit
coupling.
Buck et al., Chem 5, 1–18
January 10, 2019 ª 2018 Elsevier Inc.
https://doi.org/10.1016/j.chempr.2018.10.001

Please cite this article in press as: Buck et al., Spin-Allowed Transitions Control the Formation of Triplet Excited States in Orthogonal Donor-
Acceptor Dyads, Chem (2018), https://doi.org/10.1016/j.chempr.2018.10.001
Article
Spin-Allowed Transitions Control
the Formation of Triplet Excited States
in Orthogonal Donor-Acceptor Dyads
Jason T. Buck,1,3 Andrew M. Boudreau,1,3 Andre´ DeCarmine,1,3 Reid W. Wilson,1 James Hampsey,1
and Tomoyasu Mani1,2,4,
*
SUMMARY
The Bigger Picture
Reliance on triplet excited states
(triplets) of molecules with heavy
atoms, such as precious metals,
limits their potential in
Triplet excited states (triplets) serve as key intermediates in critical technolo-
gies and processes ranging from organic synthesis to biomedicine to molecular
electronics. Production of triplets of p-conjugated organic molecules without
heavy atoms remains challenging. Spin-orbit, charge-transfer intersystem
crossing (SOCT-ISC) directly converts singlet charge-separated states to triplets
in an electron donor-acceptor (D-A) pair. Here, using a series of orthogonal D-A
type boron dipyrromethene (BODIPY) derivatives as a model system, we show
that the formation of triplets is largely controlled by the spin-allowed transitions
rather than by SOCT-ISC. Yet, the SOCT-ISC process can still proceed much
faster than ordinary ISC between (p, p*) states because the spin-orbit coupling
of SOCT-ISC is 2 orders of magnitude stronger. We further show that such a pro-
cess can produce triplets in a non-triplet-forming molecule, perylene. Our find-
ings reveal a clear physical basis for this spin-forbidden process and provide
guidelines for future molecular designs exploiting the process.
technological applications. We
envision that triplets of
p-conjugated organic molecules
could play bigger roles; however,
their production without heavy
atoms remains challenging. The
direct, spin-forbidden conversion
of singlet charge-separated states
to triplets in an electron donor-
acceptor (D-A) pair is a promising
approach. Here, using a series of
orthogonal D-A type boron
dipyrromethene (BODIPY)
INTRODUCTION
Triplet excited states (triplets)1 play essential roles in a wide range of technologies
and processes including phosphorescence-based light emitting diodes,2 visible
light photocatalysis for organic synthesis,3 oxygen sensing in vivo,4,5 and photody-
namic therapy in medicine.6,7 Because of their long lifetimes and associated high en-
ergy transport capability, triplets hold great promise in the development of new
generations of organic photovoltaics and photocatalysts, especially in a form that
exploits singlet fission8 with a potential to be more efficient than the existing systems
based on singlet excited states. Despite all the future promises, at present, triplets of
p-conjugated organic molecules are underutilized in technological applications.
derivatives as a model system, we
show that the formation of triplets
is largely controlled by the spin-
allowed transitions. Yet, this spin-
forbidden process can still
proceed much faster than ordinary
intersystem crossing between
(p, p*) states under favorable
conditions because of stronger
spin-orbit coupling. Our findings
reveal a clear physical basis for
this spin-forbidden process and
provide guidelines for future
molecular designs exploiting the
process.
Triplets of p-conjugated organic molecules are typically formed by intramolecular
intersystem crossing (ISC, S1(p, p*) / T1(p, p*)) upon photoexcitation to the singlet
excited states as the direct excitation to the low-lying triplet excited states from the
singlet ground state is spin-forbidden. ISC is typically 2–3 orders of magnitude
slower (i.e., occurring in micro- to milliseconds) than the radiative or non-radiative
decay of singlet excited states, which occurs in nanoseconds.1 Exceptions include
some heterocycles when the transition involves (n, p*) states, which can occur in sub-
nanoseconds.9 The relatively fast deactivation of singlet excited states severely
limits the quantum yields of formation of triplets. ISC can be accelerated using heavy
atoms such as Ir, Pt, bromine, and iodide,9 and these are often found in the present
applications mentioned above. However, the inclusion of these heavy atoms may
Chem 5, 1–18, January 10, 2019 ª 2018 Elsevier Inc.
1

Please cite this article in press as: Buck et al., Spin-Allowed Transitions Control the Formation of Triplet Excited States in Orthogonal Donor-
Acceptor Dyads, Chem (2018), https://doi.org/10.1016/j.chempr.2018.10.001
Figure 1. Photophysical Scheme and Molecular Structures of BODIPYs Examined in This Study
(A) Energy diagram of the photophysical pathway in a donor-acceptor (D-A) molecule leading to
the formation of triplets of the acceptor (T1A*) by way of charge separation (CS) and charge
recombination (CRT) through SOCT-ISC. The charge-separated state 1(D$+-A$ꢀ) can also recombine
into the ground state (CRS) or back to the initial excited state (bCR).
(B) Orthogonal molecular orbitals of D-A pairs that favor the SOCT-ISC transition. 4 is a dihedral
angle between the two MO planes.
(C) The general molecular structure of the D-A molecules examined in this study. Aryl molecules
and BD cores act as electron donor and acceptor, respectively. Ant, anthracene; MeAnt,
9-methylanthracene; Pyr, pyrene; Pery, perylene; Ph, benzene.
render unwanted side effects or increase the cost of the production of materials. Pro-
duction of triplets without heavy atoms would therefore open up new venues for
their utilization.
One promising approach to producing triplets of p-conjugated organic molecules is
by way of charge-separated (CS) states or radical pairs. Upon photoinduced electron
transfer from the singlet excited state in a donor-acceptor (D-A) type molecule, a
spin-correlated singlet CS (1CS) is formed. Although 1CS typically recombines selec-
tively into singlet ground or excited states, in some circumstances triplets can be
formed without spin evolution to a triplet by a direct (CRT) transition from 1CS by
charge recombination (Figure 1A).
This direct (1CS / T1) transition from a singlet CS to a triplet was first proposed by
El-Sayed in 1974,10 and Okada et al. reported such a molecular system in 1981.11
1Department of Chemistry, University of
Connecticut, Storrs, CT 06269-3060, USA
Subsequently, similar phenomena were reported in a limited number of D-A molec-
2Precursory Research for Embryonic Science and
Technology (PRESTO), Japan Science and
Technology Agency, 4-1-8 Honcho, Kawaguchi,
ular systems, notably by van Willigen and coworkers12 and Wasielewski and
coworkers.13–15 This spin-forbidden charge recombination transition, termed spin-
Saitama 332-0012, Japan
orbit, charge-transfer intersystem crossing (SOCT-ISC), is favored when the molec-
3These authors contributed equally
ular orbitals (MOs) of D-A pairs are perpendicular to each other, creating a large
4Lead Contact
enough orbital angular momentum to produce a spin flip (see Figure 1B illustrating
*Correspondence: tomoyasu.mani@uconn.edu
orthogonality of the D-A pairs when the p-type MOs lie in the planes of the mole-
https://doi.org/10.1016/j.chempr.2018.10.001
cules). A similar rationale applies for examples of intramolecular ISC involving
2
Chem 5, 1–18, January 10, 2019

Please cite this article in press as: Buck et al., Spin-Allowed Transitions Control the Formation of Triplet Excited States in Orthogonal Donor-
Acceptor Dyads, Chem (2018), https://doi.org/10.1016/j.chempr.2018.10.001
(n, p*) states; i.e., S1(n, p*) / T2(p, p*) reported for benzaldehyde, benzophenone,
and similar compounds, following the El-Sayed rules.9,16 Recently, roles of CS states
have drawn attention in producing triplets. Filatov and coworkers17 showed that the
incorporation of a CS state leads to an increased yield of triplets in an orthogonal
boron dipyrromethene (BODIPY or BD)-based D-A molecule. Other possible similar
BD-based systems were also reported.18–21 Damrauer and coworkers reported a
phenoxazine-based system in which the yield of triplet increases by including a CS
state as well.22 We would like to emphasize that although the orthogonality of the
MOs of D-A pairs together with the closely positioned D-A moieties strongly suggest
that SOCT-ISC is most likely operative, these recent studies17–22 lack clear support
such as magnetic field effects (MFEs) measurements showing that SOCT-ISC is the
main transition to populate triplet excited states.
All the previous studies clearly show that the inclusion of a CS state in the photophys-
ical pathway leading to triplets could enhance the triplets’ formation and that the
orthogonality of the MOs of D-A pairs is critical for the SOCT-ISC transition; a signif-
icant deviation from an ideal orthogonal conformation diminishes the quantum
yields of triplets.14 In select cases, the quantum yields of triplets can be very high
(>0.9).18,22 However, there remains a clear need for a deeper understanding of
this process to reveal how we can optimize molecular systems to properly control
quantum yields of triplets through such a process. Three distinct electron-transfer re-
actions are involved in the photophysical pathway to triplets by way of SOCT-ISC
(Figure 1A): spin-allowed initial charge separation from S1
/
1CS (CS), charge
recombination of 1CS to GS (CRS), and spin-forbidden charge recombination of
1CS to T1BD* (SOCT-ISC or more generally CRT). The experimental and computa-
tional work described here seeks to answer the following central question: how
are the three electron-transfer reactions related to one another in the production
of triplets?
Here, by using a series of BODIPY-based D-A dyads as a model molecular system
and maintaining the general orthogonal D-A motif, we systematically tune the reduc-
tion and oxidation potentials to investigate the dependence of Gibbs energy
changes associated with the three electron-transfer processes and to reveal how
they affect quantum yields of triplets. A series of dyads were designed according
to the photophysical scheme depicted in Figure 1A. Our study reveals that the
spin-allowed transitions (CS and CRS), and not the spin-forbidden transition (CRT),
largely control the formation of triplets. We determined the strength of the spin-
orbit coupling of 1CS/T1(p, p*) that is 2–3 orders of magnitude stronger than that
of ordinary S1(p, p*)/T1(p, p*), and such a relatively strong coupling makes the
SOCT-ISC process proceed faster than the spin-allowed transitions under favorable
conditions. General guidelines for molecular and materials design utilizing SOCT-
ISC are also presented. We then applied our guidelines to design and synthesize
a new orthogonal D-A molecule based on perylene, a non-triplet-forming molecule,
in which we observed the formation of triplets in a good yield, showing versatility of
the approach.
RESULTS
Production of Charge-Separated States Followed by the Formation of Triplets
Upon photoexcitation of the donor-acceptor type BODIPYs (1–9, whose structures
are shown in Figure 1C and whose absorption and emission spectra are reported
in Figure S1) at the absorption band of the BD core, the CS states (Aryl$+-BD$ꢀ
are produced from S1BD*. Note that throughout the text, S1 is reserved for local
)
Chem 5, 1–18, January 10, 2019
3

Please cite this article in press as: Buck et al., Spin-Allowed Transitions Control the Formation of Triplet Excited States in Orthogonal Donor-
Acceptor Dyads, Chem (2018), https://doi.org/10.1016/j.chempr.2018.10.001
Figure 2. Formation of the CS States and Triplets in the BODIPY D-A Dyads
Transient absorption spectra of 3 and 9 (A and C, respectively) at a time delay after the fs laser pulse (lex = 500 nm) in MeCN. Decay kinetics of 3 and 9
(B and D, respectively) were monitored at the indicated wavelengths (vertical colored bars).
singlet excited states, not CS states. Spectroscopic signatures of BD$ꢀ were
observed in the visible region (ꢁ550–580 nm) together with the signatures of aryl
radical cations (Aryl$+). Pulse radiolysis experiments23 provide an independent mea-
sure of the spectra of BD$ꢀ (Figure S2). The spectral signatures of Aryl$+ are in good
agreement with those reported in the literature.24 As the photoexcitation prepares
the singlet excited states, the CS states are produced as a singlet spin state (1CS).
1CS then decays either to the GS or to form the triplets of the BD unit (T1BD*) that
do not decay significantly up to ꢁ7.8 ns, the longest time point probed by our femto-
second transient absorption spectroscopy (fs-TA) setup. Representative spectra and
decay curves for 3 and 9 are shown in Figure 2 and others are shown in Figure S3.
Note that for the dyads 7–9, the absorption bands of BD$ꢀ and Pery$+ overlap
around the same spectral region.
Although all exhibit the formation of 1CS, the production of triplets greatly varies
from molecule to molecule. The rates of the initial charge separation (kCS) and
the total charge recombination (kCR), determined by global fitting of the fs-TA
4
Chem 5, 1–18, January 10, 2019

Please cite this article in press as: Buck et al., Spin-Allowed Transitions Control the Formation of Triplet Excited States in Orthogonal Donor-
Acceptor Dyads, Chem (2018), https://doi.org/10.1016/j.chempr.2018.10.001
Table 1. Photophysical Data of BODIPY Dyads in Acetonitrile
a
b
c
d
e
f
g
BD
fl
BD
fl
R
Aryl
labs
lfl
F
k
ꢀDGoCS
kCS
kCR
FT
fCRT
ꢀDGoCRT
kCRT
ꢀDGoCRS
kCRS
(3107
(nm) (nm)
(3108
(eV)
(31012 (3108
(eV)
(3108 (eV)
sꢀ1
)
sꢀ1
)
sꢀ1
3.8
1.8
3.7
2.1
2.5
2.4
14
)
sꢀ1
3.5
1.5
3.5
1.4
2.1
2.3
2.4
2.0
)
sꢀ1
2.3
2.8
2.1
6.7
3.8
)
1
2
H
Ant
Ant
501
525
507
537
508
534
508
536
507
538
512
505
532
512
0.01
0.55h
0.05
2.2 ꢀ0.13
6.0 ꢀ0.23
0.54
0.19
0.93
0.38
0.90
0.61
0.72
0.35
0.16
0.33
0.94
0.84
0.94
0.68
0.84
0.97
0.16
0.33
0.97
0.96
0.81
0.89
0.94
1.01
0.66
0.70
0.19
NA
2.59
2.57
2.42
2.49
2.55
2.61
2.27
2.30
1.81
NA
Et
H
3
MeAnt 501
MeAnt 525
2.0
0.053 1.5
4
Et
H
0.10
2.1 ꢀ0.14
0.16
5
Pyr
499
522
502
526
504
496
520
504
0.15
2.0 ꢀ0.083 0.11
6
Et
H
Pyr
0.64h
<0.01
0.01
7.1 ꢀ0.27
0.035
0.66
7
Pery
Pery
2.0
0.18
3.5
1.1 3102
31
8
Et
1.8
0.031 0.59
5.2
9
Ac Pery
0.01
>100i
0.63
5.6
NA
NA
NA
1.9 3102 0.005j 0.01j
0.97j
NA
1.9 3103
NA
10
H
Ph
Ph
0.74h
0.76h
0.37h
2.9 NA
NA
NA
NA
0
0
0
NA
NA
NA
11 Et
3.3 NA
5.6 NA
NA
NA
NA
NA
12 Ac Ph
NA
NA
NA
NA
NA, not applicable.
aQuantum yields of fluorescence of BODIPY core were measured by the reference method against rhodamine 6G in EtOH (Ffl = 0.94) unless otherwise noted.
bDetermined by the initial decays of S1BD* that are in agreement with the rise of 1CS.
cDetermined by the decay time of the 1CS states.
dQuantum yields of formation of triplets.
ꢀ
ꢁ
eEfficiency of CRT: fCRT = FT= 1 ꢀ FBD
.
fl
fkCRT = kCR 3 fCRT
.
gkCRS = kCR ꢀ kCRT
.
hAbsolute quantum yield.
iFaster than the detection limit of our instrument (ꢁ1010
s
ꢀ1).
jThe values are considered the upper limits.
data, are reported in Table 1. Photoexcitation of non-D-A type (or parent) BODIPYs
(10–12) leads to the formation of S1BD*, which decay back to the GS without form-
ing triplets, confirming that FT of the parent BODIPYs are negligible (Figure S4).
The aryl groups’ absorption bands lie energetically higher than the BODIPYs’
bands (Figure S1). When the dyads 1–9 are photoexcited at the donor’s absorption
band, an ultrafast energy transfer occurs to populate the singlet excited state of the
BD units (<1 ps), followed by the almost identical photophysical pathways as the
case with photoexcitation at the BD core. fs-TA spectra of 1 upon 390 nm photo-
excitation are shown in Figure S5 as an example. Such an ultrafast energy transfer
with orthogonal configuration was previously observed in similar BODIPY-based
dyads.25,26
Quantum Yields of Triplets
To quantify the formation of triplets in the dyads 1–9, we performed nanosecond
transient absorption spectroscopy (ns-TA) measurements. The formation of T1BD*
was reconfirmed by ns-TA (Figure 3), and the triplet-triplet (TT) spectra are similar
to those observed in the parent BODIPYs 10–12 (Figure S6). The decay rates of trip-
lets (ICT) were typically ꢁ105 sꢀ1 (1/kICT ꢁ 10 ms; Figure 3, inset). We did not observe
a clear formation of the triplets of electron donor moieties except for 7 and 8 where
the donor is perylene, whose triplet energy (1.57 eV)27 is slightly lower than that of
T1BD* (see below). In these cases, the triplet of perylene (T1Pery*) is populated from
T1BD* via triplet energy transfer and then the equilibrium is established, exhibiting
the spectral signatures of both species. Such a triplet energy transfer from T1BD*
to T1Pery* is routinely observed intermolecularly,7 and used as a way to form
the triplet of perylene whose innate quantum yield of triplet is only 0.01.28
Chem 5, 1–18, January 10, 2019
5

Please cite this article in press as: Buck et al., Spin-Allowed Transitions Control the Formation of Triplet Excited States in Orthogonal Donor-
Acceptor Dyads, Chem (2018), https://doi.org/10.1016/j.chempr.2018.10.001
Figure 3. Transient Absorption Spectra of 3 at 1 ms after the ns Laser Pulse (lex = 355 nm) in MeCN
The inset shows the decay kinetics at 430 nm.
Quantification of the formation of triplets by a relative actinometry method requires
TT of the compounds.29 We have determined
the knowledge of 3 3 TT for 10, 11, and
12 in acetonitrile (MeCN, dielectric constant 3 = 37.5) by using the energy-transfer
r
method; they are 3,600 G 300, 3,300 G 300, and 3,700 G 300 Mꢀ1 cmꢀ1 at 415 nm,
respectively, while
3
of perylene was determined to be ꢁ13,000 Mꢀ1 cmꢀ1 at
TT
480 nm, in good agreement with the published data.30
We assumed that 3 TT of the BD core in the dyads is the same as those of the parent
BODIPY molecule, which appears to be reasonable because the TT spectra of T1BD*
in the dyads do not change much from those in the parent BODIPYs (Figure S6).
Quantum yields of triplets were determined in MeCN and are tabulated in Table 1.
FT of 7 and 8 are reported as the sum of T1BD* and T1Pery*. Very little formation of
triplets was observed for 9 in MeCN (Figures 2C and 2D) and the TT absorption spec-
trum is not reported. The reported value of FT = 0.005 for 9 is thus an estimate. In
contrast, when the measurements were done in a non-polar solvent toluene ( 3 =
r
2.38), we observed a clear triplet formation in 9 with FT ꢁ 0.2 G 0.1 (Figure 4).
The absorption and emission spectra of 9 in toluene, and additional fs-TA spectra
(lex = 420 nm) are shown in Figure S7.
Diminished Quantum Yields of Fluorescence
Consistent with the formation of the CS state upon photoexcitation, the quantum
BD
fl
yields of fluorescence from the BD core ðF Þ are lower than those of the parent
BODIPYs (Table 1). This aligns well with many of the BODIPY compounds that un-
dergo photoinduced electron-transfer reactions.31
F
of 12 is not high (0.37) for
BD
fl
BODIPY, compared with other non-D-A BODIPYs including 10 and 11.32 A lack of
formation of triplets in 12 (Figure S4) indicates that an internal conversion (IC) is
BD
fl
more efficient in 12. The observed fluorescence decay rates ðk Þ are similar in
magnitude to the parent BODIPYs except for 9 where the fluorescence decay rate
is faster than the instrument response of TCSPC (time-correlated single photon
counting) setup. Similarly, because of the instrument limitation, we could not
observe a fast decay component of fluorescence that should be present correspond-
ing to the charge separation process. The observed lifetime can be considered as a
longer decay component. Some of the D-A dyads such as 1, 3, 7, and 8, although
6
Chem 5, 1–18, January 10, 2019

Please cite this article in press as: Buck et al., Spin-Allowed Transitions Control the Formation of Triplet Excited States in Orthogonal Donor-
Acceptor Dyads, Chem (2018), https://doi.org/10.1016/j.chempr.2018.10.001
Figure 4. Production of Triplets of 9 in Toluene
(A) Transient absorption spectra of 9 at a time delay after the fs laser pulse (lex = 500 nm) in toluene.
(B) Transient absorption spectra of 9 at 1 ms after the ns laser pulse (lex = 355 nm) in toluene. The inset shows the decay kinetics at 430 nm.
very weakly, exhibit clear charge-transfer (CT) emission (Figure S1). We did not
perform a detailed analysis of the CT emission here.
Energetics
The Gibbs energy change associated with the charge separation ðDGoCSÞ from the
lowest singlet excited state (S1) is given by the equation
ꢀ
ꢂ ꢁ
ꢀ ꢂ
ꢁ
DGCoS = Eo D$ + D ꢀ Eo A A$ꢀ + w ꢀ ES1
;
(Equation 1a)
00
where EoðD$ + =DÞ and EoðA=A$ꢀÞ are the standard oxidation and reduction potential
S1
00
of electron donor and acceptor and E is the electronic energy level of the first
excited states (S1BD*) that are determined from the cross-point of absorption and
emission spectra (Figure S1). The remaining term w is the electrostatic work term
for the effect of Coulombic attraction in the CS states, expressed as
e2
w = ꢀ
;
(Equation 1b)
4p 3 r3 rDA
0
where 3 , 3 , and rDA are the vacuum permittivity, the dielectric constant of the sol-
0
r
vent ( 3 = 37.5 for MeCN), and the center-to-center distance of the charged species
r
estimated by the computation, respectively. The standard oxidation and reduction
potentials are approximated by the half-peak potential (Eo = E1/2) and reported in
Table S1. The reduction potentials of acceptors vary depending on the substitution
R of the BD core as expected, and they stay relatively unchanged even with the pres-
ence of different electron donors. rDA was estimated as the electron-hole separation
calculated by the transition density matrix analysis of the CT transition and is re-
ported in Table S2. Our way of estimation of rDA gives smaller values over center-
to-center distances estimated solely on nuclear structures. The difference between
the two methods is ꢁ1–2 A, but the resulting difference in DGo is only about 0.01–
˚
0.02 eV and comparable to experimental error in the determination of reduction
and oxidation potentials by the electrochemical measurements. Gibbs energy differ-
T1
ences associated with the charge recombination of 1CS to BD* ðDGCoRTÞ and GS
ðDGoCRSÞ were computed in a similar manner with T1BD* and the GS energy, the latter
of which was set to 0. T1BD* of 1033 and 1134 was reported as 1.61 and 1.60 eV,
Chem 5, 1–18, January 10, 2019
7

Please cite this article in press as: Buck et al., Spin-Allowed Transitions Control the Formation of Triplet Excited States in Orthogonal Donor-
Acceptor Dyads, Chem (2018), https://doi.org/10.1016/j.chempr.2018.10.001
Figure 5. Production of Triplets of 13 in THF
(A) Transient absorption spectra of 13 at a time delay after the fs laser pulse (lex = 440 nm) in THF.
(B) Decay kinetics of 13 were monitored at the indicated wavelengths (vertical colored bars).
(C) Transient absorption spectra of 13 at 1 ms after the ns laser pulse (lex = 355 nm) in THF. Note that the bigger noise in the region between 400 and
450 nm is due to high absorbance of the molecule.
respectively, on the basis of phosphorescence measurements at low temperature.
T1BD* of 12 was estimated to be 1.62 eV according to a linear relationship between
the experimental and computational data of 10 and 11. The Gibbs energy differ-
ences are tabulated in Table 1. Some DGCoS values are positive, resulting in an uphill
electron transfer. Our computational calculations correctly predict the order of
S1BD* and 1CS to some extent (Table S2). uB97X-D better follows the observed
experimental transition energies for 1–4 while CAM-B3LYP follows 5–9. CAM-
B3LYP systematically places the energies of the CSs lower than uB97X-D. The
computed energies of S1BD* are on average ꢁ0.6 eV higher than the experimental
values. We assign 1CS on the basis of the CT character of the transition, which was
analyzed by the attachment-detachment35 densities (Figure S8). The computed
lowest triplet excited states are always of the BD core (T1BD*).
Orthogonality and Spin-Orbit Coupling
The computed ground-state geometries with uB97X-D show near-orthogonal con-
figurations of the donor and acceptor moieties (4 ꢁ 90ꢂ; Table S2). The combination
of the methyl substitution at the b position of the BD core and ‘‘bulky’’ aryl groups in
the dyads makes the energy barrier for the aryl rotation high, locking them to the
orthogonal configuration upon photoexcitation, in a manner similar to that of the
BODIPYs with bulky meso-substitution such as 4-tert-butylphenyl.36
The 1CS/T1(p, p*) spin-orbit couplings (HSO) computed for 1–9 range from
0.83–1.23 cmꢀ1, by using the method developed by Ou and Subotnik.37 Similarly,
HSO of S1(p, p*)/T1(p, p*) within the same molecules and for the parent BODIPYs
10–12 were computed. They are reported in Table S2. HSO of 1CS/T1(p, p*) are
1–3 orders of magnitude bigger than HSO of S1(p, p*)/T1(p, p*). On the other
hand, they are about an order of magnitude smaller than the reported HSO of
S1(n, p*)/T2(p, p*) for benzaldehyde (ꢁ26–45 cmꢀ1 depending on the geometry).37
Both uB97X-D and CAM-B3LYP give similar HSO values.
Triplet Formation in Perylene
Following the findings from the study of the BODIPY dyads (see Discussion), an
orthogonal D-A molecule based on perylene, 13, was designed and synthesized
(the structure and the synthetic scheme are shown in Figure 5A and Scheme S2,
respectively). Perylene here acts as an electron acceptor, not a donor, unlike the
8
Chem 5, 1–18, January 10, 2019

Please cite this article in press as: Buck et al., Spin-Allowed Transitions Control the Formation of Triplet Excited States in Orthogonal Donor-
Acceptor Dyads, Chem (2018), https://doi.org/10.1016/j.chempr.2018.10.001
Table 2. Photophysical Data of Molecule 13
Pery
fl
a
a
d
e
f
g
Solvent labs
(nm) (nm)
lfl
F
lCT
FCT
ꢀDGoCS kCSb 3
kCRc 3
FT
fCRT
ꢀDGoCRT kCRT
ꢀDGoCRS kCRS
(nm)
(eV)
1012 (sꢀ1
2.5
)
108 (sꢀ1
)
(eV)
0.84
1.07
(3107 sꢀ1
6.1
)
(eV)
2.38
2.52
(3108 sꢀ1
8.1
)
1
2
MeCN
THF
442
445
445
<0.001 650
<0.001 0.39
0.20
0.060i
8.7
7.4
0.07 0.07
0.20 0.20
h
h
–
–
537
0.30
15
5.9
aQuantum yields of fluorescence of perylene and CT emission were measured by the absolute quantum yield method.
bDetermined by the initial S1Pery* decays that are in agreement with the rise of 1CS.
cDetermined by the decay time of the 1CS states.
dQuantum yields of formation of triplets.
ꢀ
ꢁ
eThe efficiency of CRT: fCRT = FT= 1 ꢀ FBD
.
fl
fkCRT = kCR 3 fCRT
.
gkCRS = kCR ꢀ kCRT
.
hNot observed.
iEstimated with the reduction and oxidation potentials and consideration of the effects of ion pairs. The error is G0.1 eV.
molecules 7–9. Two methyl groups in 3,5 positions in the 3,5-dimethyl-N,N-dime-
thylaniline (Me2DMA) moiety locks the perylene and DMA’s MO planes to the
near-orthogonal configuration (Figure S9 and Table S3). The photophysical data
are summarized in Table 2 and reduction and oxidation potentials are reported in
Table S1. Experimental determination of the energy of the CS state (Me2DMAl$+
-
Pery$ꢀ) by Equation 1a shows that 1CS is energetically lower than S1Pery*, although
the density functional theory (DFT) calculation (uB97X-D or B3LYP) predicts other-
wise (Table S3). Photoexcitation of the molecule 13 in THF results in the formation
of 1CS state from S1Pery* (Figures 5A and 5B). The spectroscopic signature of Pery$ꢀ
was observed in the visible region (ꢁ590 nm).24 1CS then decays to form the triplets
of the perylene unit (T1Pery*) whose signature is visible around 490 nm (Figure 5C).30
The lack of fluorescence emission from perylene also supports a quenching of
S1Pery*. A clear CT emission was observed in THF (Figure S10). The quantum yield
of triplets in a relatively non-polar THF ( 3 = 7.52) is ꢁ0.2 G 0.05, whereas it is
r
only 0.07 G 0.05 in a polar MeCN (Figure S11). As a general trend, the formation
of triplets decreases with increasing solvent polarity (Figure S12).
DISCUSSION
Efficient Charge Separation in Orthogonal D-A BODIPY Pair
Here, we first discuss the study of BODIPYs. We confirmed that the production of
triplets proceeds through the pathway depicted in Figure 1A. Photoexcitation of
all the dyads 1–9 initially leads to a formation of the CS states. The energies of the
CS states are determined to be higher than S1BD* in some cases, giving a positive
Gibbs energy change. Our value of the CS state is higher than that reported earlier
for the molecule 1.18 This discrepancy most likely stems from the measurements of
reduction and oxidation potentials. Our electrochemical measurements were con-
ducted in the same solvent we used for the photophysical measurements (a polar
solvent MeCN), while the previous report performed the measurements in relatively
non-polar solvents and made the correction. The rates of charge separation exhibit
clear dependence on the changes in Gibbs energy (Figure 6). Amazingly, even with
significant uphill nature (DGoCS ꢁ0.1–0.2 eV; kBT ꢁ0.025 eV at room temperature),
charge separation can proceed with the rate constant >1010 ꢀ1, most likely because
s
of small rDA. At the same time, back electron transfer also becomes efficient, repo-
pulating S1BD* (bCR in Figure 1) and establishing the equilibrium between S1BD*
and 1CS. This leads to the following two consequences in some dyads: (1) reduction
of the overall yields of the CS (see below) and/or (2) the observed longer component
of fluorescence lifetime is not much shorter than the fluorescence lifetime of the
parent BODIPYs even with charge separation (Table 1). In general, an orthogonal
Chem 5, 1–18, January 10, 2019
9

Please cite this article in press as: Buck et al., Spin-Allowed Transitions Control the Formation of Triplet Excited States in Orthogonal Donor-
Acceptor Dyads, Chem (2018), https://doi.org/10.1016/j.chempr.2018.10.001
Figure 6. Electron-Transfer Rate Constant for the Initial Charge Separation, Charge
Recombination to the GS, and Charge Recombination to the Triplets in BODIPY D-A Dyads
The solid and dotted lines are fits of electron-transfer theory Equation 2 to the datasets.
Parameters: Hab = Hel = 177 cmꢀ1, lS = 0.17 eV, lV = 0.37 eV for the black solid line, and Hab = HSO
=
1.35 cmꢀ1 for the red solid line; Hab = HSO = 1.25 cmꢀ1, lS = 0.22 eV, and lV = 0.70 eV for the red
dotted line. All the fits were performed with ꢀhu = 1,500 cmꢀ1. The uncertainty in DG0 is G0.02 eV.
CS, charge separation; CRS, charge recombination to the GS; CRT, charge recombination to the
triplets.
configuration of p-type MOs between the donor and acceptor is not in favor of elec-
tron transfer, as such a configuration diminishes electronic coupling through p-type
MOs. Observed fast rate constants (kCS > 1010
contribution from s-type orbitals.38–40
s
ꢀ1) indicate that there is a significant
Branching of Charge Recombination
1CS decays either back to the GS (CRS) or to T1BD* via SOCT-ISC (CRT). We can
determine the individual rate constants of these two channels, kCRS and kCRT, on
the basis of the total kCR, FBflD, and FT by assuming that a charge separation process
is the only additional deactivation pathway of singlet excited states in the dyads 1–9
other than fluorescence (fl) and innate internal conversion (ICS). In other words, the
quantum yield of 1CS in the dyads is now given by fCS = 1 ꢀ FBD. This assumption is
fl
reasonable because the ICs in 10–12 are at least 2 orders of magnitude slower than
the measured rates of charge separation in the dyads: kICS = 7.4 3 107, 4.4 3 107,
and 3.5 3 108 sꢀ1 for 10, 11, and 12, estimated by kICS = ð1 ꢀ FfBlDÞ=kBD. We can
fl
then calculate kCRT and kCRS through kCRT = kCR 3 fCRT and kCRS = kCR ꢀ kCRT, where
fCRT = FT/fCS. kCRS and kCRT are tabulated in Table 1, along with fCRT. They are
plotted against the changes in the Gibbs energy (Figure 6; note that the x axis is
–DG0). We observed a change of 3 orders of magnitude in kCRS over 0.8 eV of Gibbs
energy change, exhibiting clear Marcus inverted region behavior. On the other
hand, the change in kCRT is relatively small and all of the points are ꢁ108
s
ꢀ1, as
observed in the form of ‘‘slow’’ growth of the signature of the triplets in the fs-TA
spectra (Figure 2). Within the dyads studied, we did not observe a clear indication
of the inverted region in CRT. Except for 7 and 8, many of the triplets of the donors
lie energetically higher than T1BD*,41 which is energetically accessible from 1CS. The
absence of the formation of the triplets of these donors indicates a preferential hole
movement from the Aryl$+ to BD$ꢀ, forming the triplet of the acceptor (T1BD*).
Although not known exactly, our computational analysis and others19 show that
10
Chem 5, 1–18, January 10, 2019

Please cite this article in press as: Buck et al., Spin-Allowed Transitions Control the Formation of Triplet Excited States in Orthogonal Donor-
Acceptor Dyads, Chem (2018), https://doi.org/10.1016/j.chempr.2018.10.001
T2BD* is slightly higher than S1BD* in energy. Although HSO of 1CS/T2(p, p*) may
have some contributions in the CRT pathway,14 we did not include the contribution
in our analysis.
In the present analysis, we did not account for the contribution to the production of
triplets by way of radical pair intersystem crossing (RP-ISC)42 within the CS followed
by triplet charge recombination (1CS / 3CS / T1) on the basis of the close prox-
imity of two radicals. Such pathways have been well documented previously.43–48
˚
Generally, a small rDA, as observed in our system <ꢁ5 A, makes the energy differ-
ence (or electronic coupling) between 1CS and 3CS sufficiently large so that the
RP-ISC, which typically occurs through a weak hyperfine coupling (ꢁ107–108
s
ꢀ1),
does not readily occur and only incoherent relaxation (ꢁ106 sꢀ1 49
)
occurs in the
absence of external magnetic fields.46,50,51 We further confirmed a lack of MFEs
on the formation of triplets by conducting ns-TA in the presence of external mag-
netic field up to 3,000 gauss (Figure S13). If RP-ISC is operative, we would expect
to observe an appreciable MFE on the formation of triplets as occurs in an unlinked
D/A system of DMA/pyrene, where we observed a ꢁ10% decrease in the formation
of pyrene’s triplets at high fields (Figure S13), as demonstrated earlier.42,52,53 The
lack of MFEs strongly supports the notion that the triplets are indeed formed via
SOCT-ISC.
Spin-Allowed versus Spin-Forbidden Processes
Marcus theory54–56 can be applied for calculating the electron-transfer rate contents
in the form of the following equation:
"
#
ꢃ
ꢄ
2
Sw
ðlS + DG0 + wZuÞ
4lSkBT
X
N
2p
1
2
kET
=
jHab
j
eꢀS
exp
ꢀ
;
(Equation 2)
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
w = 0
Z
w!
4plSkBT
lV
where Hab is the electronic coupling, S =
is the Huang-Rhys factor characterizing
Zu
the strength of e-v coupling, and u is only one high frequency of the coupled quan-
tum mechanical vibration modes. The vibrational quantum ꢀhu is set to 1,500 cmꢀ1
(0.186 eV) on the basis of aromatic C=C stretching vibrations.57 As the spin-orbit
coupling that dictates the transition is also weak, we can apply the Marcus theory
to the rate constant for CRT by using spin-orbit coupling in place of electronic
coupling: Hab = HSO.
58 We fitted the rate constants for CS and CRS together (black
solid line in Figure 6). Here, we assumed that there are ‘‘average’’ parameters for the
CS and CRS in all the combinations of BD cores and aryl donor moieties examined
here, neglecting the individual molecular details. We also assumed that CS and
CRS can be adequately described for all with the same parameters.59,60
We found the reorganization energy of the solvent molecules lS = 0.17 eV, relatively
small for the polar MeCN solution, which indicates that the alignment of the sur-
rounding solvent molecules does not significantly change upon CT movements.
The reorganization energy for high-frequency molecular vibrations lV = 0.37 eV is
comparable with lV = 0.45 found for intramolecular61 and lV = 0.4 eV found for inter-
molecular ET from biphenyl anion to BQ and other acceptors.56 This lV gives the
Huang-Rhys factor S ꢁ 2.0, suggesting a stronger e-v coupling comparable to those
reported previously.60,62 Deviation of the fitted curve from the data points in the up-
hill charge separation region ( ꢀ DGoCS < 0 eV) might be explained by the vibrationally
‘‘hot’’ or unrelaxed excited states created by ultrafast pulse of excitation. If a vibra-
tional relaxation within S1BD* is slower than the charge separation, the electron
transfer can occur from unrelaxed excited states, making the actual Gibbs energy
difference smaller than the ones calculated with Equation 1. We also note that the
inverted region was clearly observed in CRS because other than the triplets, there
Chem 5, 1–18, January 10, 2019
11

Please cite this article in press as: Buck et al., Spin-Allowed Transitions Control the Formation of Triplet Excited States in Orthogonal Donor-
Acceptor Dyads, Chem (2018), https://doi.org/10.1016/j.chempr.2018.10.001
are no low-lying states such as excited states of radical ions63,64 that could diminish
the effect of the inverted region.
We cannot adequately describe the rate constants of the charge recombination to
triplets (CRT) by using the same set of the parameters derived from the analysis of
CS and CRS. The rate constants for CRT can be fitted by using lS and lV obtained
from the fitting for CS and CRS, to determine HSO: HSO = 1.35 cmꢀ1 (red solid line
in Figure 6). This means that the reorganization energies for CRT may be comparable
with those for CRS. We also performed the fitting without any constraints to afford a
slightly better fitting with a similar coupling constant HSO = 1.24 cmꢀ1 (red dashed
line) and bigger lV value. Little change was observed in kCRT over the examined
range of ꢀDGoCRT (ꢁ0.2–1.0 eV), most likely because DGCoRT’s fall in the region near
the top of the Marcus inverted parabola. Because of the restriction imposed by
the photophysical scheme (Figure 1A), we could not readily study the region of
larger driving force for CRT, but the study of such molecular systems is planned for
the future. As previously shown, the values of Hab are not highly dependent on lS,
lV, and ꢀhu.59 While acknowledging the uncertainties associated with some of our
assumptions, our HSO value is among the very few experimental measures of spin-
orbit coupling strength. There have been no reports, to the best of our knowledge,
on the experimental assessment of HSO between the CS state and the local triplet
excited state. Our experimental result is in close agreement with our computational
value; HSO(comp) ꢁ 1.1 cmꢀ1
.
A striking difference in the Gibbs energy dependence between the spin-allowed and
spin-forbidden transitions comes from the coupling strength between the reactants
and products: Hel versus HSO. The obtained HSO ꢁ1.3 cmꢀ1 is 2 orders of magnitude
smaller than Hel = 177 cmꢀ1. Although the difference in coupling strengths between
H
el and HSO is not so surprising, we uniquely captured a 2-orders-of-magnitude dif-
ference in coupling results in a 4-orders-of-magnitude difference in rate constants,
as predicted by the Marcus-type equation. Obtained HSO of 1CS/T1(p, p*) is at least
1–2 orders of magnitude bigger than that of S1(p, p*)/T1(p, p*) that are <0.01 to
0.1 cmꢀ1. Together with the favorable Gibbs energy changes, this relatively large
spin-orbit coupling allows the spin-forbidden CRT process to compete against the
spin-allowed CRS transition. Our experimental and computational determination
of spin-orbit coupling of 1CS/T1(p, p*) provides us with a firm ground for the first
time to discuss SOCT-ISC in the context of other relevant transitions.
Design Criteria to Maximize FT
To maximize FT via SOCT-ISC, both CS and CRT need to be efficient as FT = fCS
3
fCRT. The efficiency of charge separation reaches almost unity even without signifi-
cant favorable Gibbs energy changes (Figure 7A), although too steep uphill electron
transfer does not work in favor. As mentioned above, this good efficiency of uphill
electron transfer is most likely due to the short distances between the donor and
the acceptor moieties. Such a close distance also makes SOCT-ISC more preferable
to RP-ISC (see above). We find that kCRT values are not particularly enhanced for
those with high fCRT; instead, kCRS values are suppressed because CRS values are
deep in the inverted region (Figure 6).54,65
We can attribute this change mostly to the Gibbs energy changes, as no clear depen-
dence of fCRT on a small deviation from the perfect orthogonal configuration was
observed (Figure S14). The fCRT shows dependences on the ratio of the Gibbs en-
ergy changes of DGoCRT and DGCoRS (Figure 7B). As the charge recombination to the
GS becomes more efficient with a smaller energy gap along the inverted curve,
12
Chem 5, 1–18, January 10, 2019

Please cite this article in press as: Buck et al., Spin-Allowed Transitions Control the Formation of Triplet Excited States in Orthogonal Donor-
Acceptor Dyads, Chem (2018), https://doi.org/10.1016/j.chempr.2018.10.001
Figure 7. Relations between the Gibbs Energy Changes and the Efficiencies of the Electron-Transfer Reactions
(A) Gibbs energy dependence of fCS. The uncertainty in fCS is within 5%.
(B) Dependence of fCRT on the ratio of DGCRT and DGCRS. The uncertainty is shown as error bars. All reported values are in MeCN.
the formation of triplets becomes inefficient. The two effects of charge separation
and recombination act in tandem to affect the production of triplets as they are parts
of the same photophysical pathway. For example, while the dyads 2 and 6 exhibit
low yields of charge separation, the CRT is very efficient. In this case, the limiting fac-
tor of forming triplets is initial charge separation. On the other hand, the initial
charge separation is almost unity in the dyad 9 thanks to a very favorable Gibbs en-
ergy change, but the triplets’ formation is negligible as the singlet charge recombi-
nation pathway is very efficient because of the small energy gap. However, when the
same dyad 9 is photoexcited in a non-polar solvent toluene, the CS state becomes
more destabilized. Although we do not know the exact energy of the CS state of 9 in
S1
00
toluene, the energy is most likely below E judging from the CT emission spectrum
(Figure S7). A higher-energy CS state makes DGoCS less favorable but makes a com-
bination of DGoCRT and DGCoRS more favorable for CRT, resulting in a fine formation of
triplets in toluene (FT ꢁ 0.2). This result indicates that the SOCT-ISC mechanism can
be realized in a wide range of environments of polarity, most likely both in solution
and solid states, by properly tailoring the molecular properties. Given that it could
depend on the selected pairs of donor and acceptor, especially the triplet energy
level of the acceptor, we propose the following as general guidelines to improve
FT along with the orthogonal D-A motif: (1) a close distance between donor and
acceptor and (2) adjusting the energy of CS state to suppress the CRS pathway,
which can most likely be realized by making the CS energetically close to S1. In
contrast, one can diminish FT while maintaining the orthogonal D-A motif by tuning
the potentials. In short, the balance between the initial charge separation and
recombination pathways is a key to controlling the production of triplets by way of
CS states.
Formation of Triplets in a Non-triplet-forming Molecule
Although FT of non-D-A type BODIPYs is close to 0 as discussed above, we can
enhance the triplet formation of BODIPYs by including heavy atoms such as iodine
(e.g., diiodo-BODIPYs66–68) as well as that of other p-conjugated molecules.9 Per-
ylene, on the other hand, is known for the lack of triplet formation itself (Figure S11),
and more importantly the lack of a heavy atom effect on the formation of triplets;
Chem 5, 1–18, January 10, 2019
13

Please cite this article in press as: Buck et al., Spin-Allowed Transitions Control the Formation of Triplet Excited States in Orthogonal Donor-
Acceptor Dyads, Chem (2018), https://doi.org/10.1016/j.chempr.2018.10.001
FT of bromoperylene is still 0.01.9,69 Therefore, the triplets of perylene are typically
produced by intermolecular triplet energy transfer7 or intramolecular triplet transfer
like we observed in 7–9. We have applied our guidelines to examine whether we
could produce the triplets of perylene directly through SOCT-ISC. The molecule
13 exhibits a good formation of triplets in relatively non-polar solvent THF (FT
ꢁ
0.20), a 20-fold increase compared with perylene and bromoperylene. The forma-
tion of triplets decreases with increasing solvent polarity. Although it still forms
triplets in a more polar solvent acetone, MeCN and dimethylsulfoxide (DMSO),
its quantum yield becomes lower as the polarity increases; they are 0.12, 0.07,
and 0.04, respectively (Figure S12). In all these solvents, the singlet excited state
of perylene is quenched by the charge separation to form 1CS with almost unity
(fCS ꢁ 0.99). Therefore, this difference of FT can be explained by the increased
fCRT, the same logic applied for the dyad 9; a higher-energy CS state makes the
combinations of DGos of the three electron-transfer reactions more favorable for
the production of triplets. On the other hand, a further non-polar solvent, toluene,
does not lead to an increase in the formation of triplets (FT ꢁ 0.18; Figure S12)
because of the unfavorable DGoCS for the initial formation of 1CS. In line with our
guidelines, we could enhance FT by suppressing the CRS (Table 2). A smaller FT
of 13 in tetrahydrofuran (THF), compared with some of the D-A BODIPYs described
here, may come from a smaller HSO of 1CS/T1(p, p*) (HSO(comp) ꢁ 0.6 cmꢀ1; Table
S2) and points to the possibility that further optimization is possible for the pery-
lene molecular system. Nonetheless, our result clearly demonstrates that our
guidelines can be applied to producing triplets of even typically non-triplet-form-
ing molecules. The two cases of BODIPY and perylene systems examined here
also show that it may be necessary to increase HSO of 1CS/T1(p, p*) to the level
in BODIPY system for efficient triplet formation in general, suggesting a way for
further improvement in addition to the orthogonal configuration of MO’s D-A pairs
and the favorable energetic conditions.
In summary, SOCT-ISC is an appealing mechanism that can enable the efficient
formation of triplets of p-conjugated organic molecules. Using a series of D-A
molecules based on BODIPYs as a model molecular system, we demonstrated
that the formation of triplets via this mechanism is largely controlled by the
spin-allowed transitions (i.e., initial charge separation and charge recombination
to the ground state) rather than by the spin-forbidden charge recombination it-
self. Yet, we can suppress the spin-allowed transitions by using a unique feature
of electron-transfer reactions (i.e., the Marcus inverted region) to make SOCT-
ISC a dominant process, resulting in triplets in an excellent yield. A lack of
appreciable MFEs on triplet formation strongly supports that SOCT-ISC is a
dominant process to form triplets. Spin-orbit coupling of 1CS/T1(p, p*) deter-
mined in this study allows us to discuss this unique transition relative to other
known transitions: it is 2–3 orders of magnitude bigger than that of S1(p, p*)/
T1(p, p*), which mediates ordinary ISC involving the only (p, p*) states of p-con-
jugated molecules, but an order of magnitude smaller than that of S1(n, p*)/
T2(p, p*) of benzaldehyde and similar triplet-forming compounds. Therefore,
the CRT via SOCT-ISC can proceed efficiently under favorable conditions but
is not necessarily always the preferred route like ISC in the compounds with
stronger spin-orbit couplings unless we could further increase the spin-orbit
coupling. We have further corroborated our findings by showing the modulation
of FT in a non-triplet-forming molecule. Our findings have provided a clearer
physical insight into this spin-forbidden process and reveal the general guide-
lines for future molecular and materials designs exploiting the process for a
wide range of technologies.
14
Chem 5, 1–18, January 10, 2019

Please cite this article in press as: Buck et al., Spin-Allowed Transitions Control the Formation of Triplet Excited States in Orthogonal Donor-
Acceptor Dyads, Chem (2018), https://doi.org/10.1016/j.chempr.2018.10.001
EXPERIMENTAL PROCEDURES
All solvents and reagents used were obtained from standard commercial sources
and used as received unless otherwise noted. Acetonitrile (MeCN) was purified
˚
with a Vacuum Atmospheres solvent purifier system. Silica gel (pore size 60 A,
230–400 mesh, SiliCycle) and neutral alumina (Type NW-3 Neutral, Sigma) were
used for column chromatography. 1H, 13C, and 19F NMR spectra were recorded
with a Bruker Avance III spectrometer operating at 400.14, 100.62, and 376.51
MHz, respectively. High-resolution mass spectra were obtained with QStar Elite
(AB Sciex) at the Laboratory of Mass Spectrometry and Omics Analysis at the Univer-
sity of Connecticut Department of Chemistry. UV-vis absorption spectra were
recorded with a Cary 50 Scan UV-vis spectrophotometer (Varian). Steady-state fluo-
rescence spectra were recorded using a FluoroMax (Horiba) spectrometer. Fluores-
cence quantum yields (Ffl) were measured either by a reference method against fluo-
rescence of rhodamine 6G in ethanol (EtOH, Ffl = 0.94)70 or by the absolute
measurement with a Quanta-4 (Horiba). All the photophysical and electrochemical
measurements were conducted at room temperature in MeCN unless otherwise
noted. The general methodology for fluorescence lifetime measurements, electro-
chemistry, pulse radiolysis, and the complete synthetic scheme, synthetic proced-
ures, and characterization of new compounds is described in the Supplemental
Information.
Femtosecond Transient Absorption Spectroscopy
The fs-TA setup was based on HELIOS FIRE (Ultrafast Systems), coupled to a femto-
second laser system (Coherent). The system consisted of an Astrella one-box
Ti:Sapphire amplifier. The pulse width was full width at half maximum (FWHM)
ꢁ35 fs at 800 nm with energy of ꢁ5.0 mJ/pulse at 1-kHz repetition rate. The output
of the amplifier is divided by beam splitters. One beam (ꢁ48% of the total power)
was used as the input for an OPerA Solo optical parametric amplifier (Coherent),
which provided a pump pulse in the UV-vis region. The energy of the pump beam
was adjusted to between 1 and 4 mJ, depending on the excitation wavelength
(370–520 nm). Another beam (ꢁ2% of the total power) was used to generate
white-light continuum probe pulses in the HELIOS spectrometer. For the UV contin-
uum (325–650 nm) the beam was focused into a CaF2 crystal, whereas the beam was
focused into a sapphire plate for the visible continuum (400–800 nm). A 1,024-pixel
CMOS sensor was used for detection. A quartz cell with a path length of 0.2 cm was
used. During the measurements, the sample was pseudo-randomly translated to
avoid photodegradation by using the translating sample holder (Ultrafast Systems).
Datasets were processed and analyzed with Surface Xplorer Software (Ultrafast Sys-
tems) by subtracting fluorescence background (if any), applying chirp correction,
performing single-value decomposition, and global fitting.
Nanosecond Transient Absorption Spectroscopy
The ns-TA setup was based on Edinburgh Instruments LP920, consisting of a third
harmonic (355 nm) of Nd:YAG as pump (Continuum Surelite I, pulse width FWHM
5–7 ns, 10-Hz repetition rate), Xe Model 920, 450-W xenon arc lamp as the probe
lamp, and a monochromator, with a P928 photomultiplier detector (Hamamatsu)
for decay measurements and iCCD camera (Andor Technology) for spectral
measurements, respectively. The energy of the pump beam was adjusted to ꢁ5–
15 mJ, depending on the samples, throughout the measurements. The samples
were deoxygenated by applying at least three pump-freeze-thaw cycles. A Pyrex
cuvette with a path length of 1.0 cm was used for the measurements. Quantum yields
of triplets (FT) were measured with a relative actinometry method compared with the
Chem 5, 1–18, January 10, 2019
15

Please cite this article in press as: Buck et al., Spin-Allowed Transitions Control the Formation of Triplet Excited States in Orthogonal Donor-
Acceptor Dyads, Chem (2018), https://doi.org/10.1016/j.chempr.2018.10.001
formation of triplets of benzophenone in MeCN (FT = 1.0).29 Absorption extinction
coefficients of triplet-triplet absorption ( 3 ) were measured by an energy-
TT
transfer method29 with a benzophenone as a triplet sensitizer ( 3 = 1,600 and
TT
6,500 Mꢀ1 cmꢀ1 at 415 and 500 nm, respectively).71 The values are the average of at
least three independent measurements. Datasets were processed and analyzed using
OriginPro 2016 or 2017 Software (OriginLab). Effects of external magnetic fields on trip-
lets’ formation were tested by integrating an electromagnet (3470, GMW Associates)
into the LP920 spectrometer to construct a magnetically affected reaction yield
(MARY) spectroscopy setup. The diagram is shown in the Supplemental Information.
In brief, the electromagnet has axial holes in each pole (8 mm), and is equipped with
a power supply (BOP50-8ML) and cooled with a recirculating chiller (Isotemp II, Fisher
Scientific). For the MARY spectroscopy measurements, the samples were deoxygen-
ated by bubbling Ar for at least 15 min before the measurements.
Computations
Computations were carried out with Q-Chem 4.2 or 5.0.72 The geometries were opti-
mized with uB97X-D73 functional in DFT calculations. CAM-B3LYP74 was used in sin-
gle-point calculations where noted. The geometry optimizations were performed
without symmetry constraints. The 6-31+G(d) basis set was used for the geometry
optimization and 6-31G(d) basis set was used for all other single-point calculations.
The conductor-like polarizable continuum model for solvation in MeCN ( 3 = 37.5
r
and index of refraction = 1.344) was used.75 Linear response time-dependent DFT
(TD-DFT) calculations were performed for low-lying excited states to determine tran-
sition energies at the ground-state geometries. Spin-orbit coupling constants (HSO
)
were determined within the framework of DFT, developed by Ou and Subotnik,37
and implemented in Q-Chem 5.0. Visualization was performed with VMD.
SUPPLEMENTAL INFORMATION
Supplemental Information includes Supplemental Experimental Procedures, 14 fig-
ures, 2 schemes, 3 tables, and 1 data file and can be found with this article online at
https://doi.org/10.1016/j.chempr.2018.10.001.
ACKNOWLEDGMENTS
This work was supported by start-up funds from the University of Connecticut and
from the Japan Science and Technology Agency through Precursory Research for
Embryonic Science and Technology (PRESTO, grant no. JPMJPR17GA) to T.M. A
part of this work, including use of the LEAF facility at the Brookhaven National Lab-
oratory (BNL) Accelerator Center for Energy Research and the computer cluster at
the Center for Functional Nanomaterials, was supported by the Division of Chemical
Sciences, Geosciences, & Biosciences of the Basic Energy Sciences program of the
US Department of Energy Office of Science under contract no. DE-SC0012704. We
thank Profs. Jing Zhao and Jessica Rouge of the University of Connecticut for kindly
letting us use their fluorometers. T.M. thanks Drs. John R. Miller and Marshall D.
Newton of the BNL for helpful discussions and Prof. Harry A. Frank of the University
of Connecticut for helpful comments on the manuscript.
AUTHOR CONTRIBUTIONS
T.M. designed the study. J.T.B., A.M.B., A.D., R.W.W., J.H., and T.M. performed the
experiments. T.M. analyzed data and wrote the manuscript.
DECLARATION OF INTERESTS
The authors declare no competing interests.
16
Chem 5, 1–18, January 10, 2019

Please cite this article in press as: Buck et al., Spin-Allowed Transitions Control the Formation of Triplet Excited States in Orthogonal Donor-
Acceptor Dyads, Chem (2018), https://doi.org/10.1016/j.chempr.2018.10.001
Received: January 22, 2018
Revised: July 25, 2018
Accepted: October 1, 2018
Published: October 25, 2018
REFERENCES AND NOTES
1. McGlynn, S.P., Azumi, T., and Kinoshita, M.
(1969). Molecular Spectroscopy of the Triplet
State (Prentice-Hall).
14. Dance, Z.E.X., Mickley, S.M., Wilson, T.M.,
Ricks, A.B., Scott, A.M., Ratner, M.A., and
Wasielewski, M.R. (2008). Intersystem crossing
mediated by photoinduced intramolecular
charge transfer: julolidine-anthracene
molecules with perpendicular pi systems.
J. Phys. Chem. A 112, 4194–4201.
24. Shida, T. (1988). Electronic Absorption Spectra
of Radical Ions (Elsevier).
25. Wiebeler, C., Plasser, F., Hedley, G.J.,
Ruseckas, A., Samuel, I.D.W., and Schumacher,
S. (2017). Ultrafast electronic energy transfer in
an orthogonal molecular dyad. J. Phys. Chem.
Lett. 8, 1086–1092.
2. Baldo, M.A., O’Brien, D.F., You, Y., Shoustikov,
A., Sibley, S., Thompson, M.E., and Forrest,
S.R. (1998). Highly efficient phosphorescent
emission from organic electroluminescent
devices. Nature 395, 151–154.
15. Colvin, M.T., Ricks, A.B., Scott, A.M., Co, D.T.,
and Wasielewski, M.R. (2012). Intersystem
crossing involving strongly spin exchange-
coupled radical ion pairs in donor-bridge-
acceptor molecules. J. Phys. Chem. A 116,
1923–1930.
26. Hedley, G.J., Ruseckas, A., Benniston, A.C.,
Harriman, A., and Samuel, I.D.W. (2015).
Ultrafast electronic energy transfer beyond the
weak coupling limit in a proximal but
orthogonal molecular dyad. J. Phys. Chem. A
119, 12665–12671.
3. Schultz, D.M., and Yoon, T.P. (2014). Solar
synthesis: prospects in visible light
photocatalysis. Science 343, 985.
4. Spencer, J.A., Ferraro, F., Roussakis, E., Klein,
A., Wu, J.W., Runnels, J.M., Zaher, W.,
Mortensen, L.J., Alt, C., Turcotte, R., et al.
(2014). Direct measurement of local oxygen
concentration in the bone marrow of live
animals. Nature 508, 269–273.
16. El-Sayed, M.A. (1968). Triplet state. its radiative
and nonradiative properties. Acc. Chem. Res.
1, 8–16.
27. Chrysochoos, J., and Grossweiner, L.I. (1968).
Flash photolysis of several aromatics by
ultraviolet light and visible light in the presence
of eosin Y. Photochem. Photobiol. 8, 193–208.
17. Filatov, M.A., Karuthedath, S., Polestshuk, P.M.,
Savoie, H., Flanagan, K.J., Sy, C., Sitte, E.,
Telitchko, M., Laquai, F., Boyle, R.W., et al.
(2017). Generation of triplet excited states via
photoinduced electron transfer in meso-
anthra-BODIPY: fluorogenic response toward
singlet oxygen in solution and in vitro. J. Am.
Chem. Soc. 139, 6282–6285.
28. Parker, C.A., and Joyce, T.A. (1966). Formation
efficiency and energy of the perylene triplet.
Chem. Commun. 108–109.
5. Rumsey, W.L., Vanderkooi, J.M., and Wilson,
D.F. (1988). Imaging of phosphorescence—a
novel method for measuring oxygen
distribution in perfused tissue. Science 241,
1649–1651.
29. Carmichael, I., and Hug, G.L. (1986). Triplet-
triplet absorption spectra of organic molecules
in condensed phases. J. Phys. Chem. Ref. Data
15, 1–250.
6. Celli, J.P., Spring, B.Q., Rizvi, I., Evans, C.L.,
Samkoe, K.S., Verma, S., Pogue, B.W., and
Hasan, T. (2010). Imaging and photodynamic
therapy: mechanisms, monitoring, and
18. Wang, Z., and Zhao, J. (2017). Bodipy-
anthracene dyads as triplet photosensitizers:
effect of chromophore orientation on triplet-
state formation efficiency and application in
triplet–triplet annihilation upconversion. Org.
Lett. 19, 4492–4495.
30. Bensasson, R., and Land, E.J. (1971). Triplet-
triplet extinction coefficients via energy
transfer. Trans. Faraday Soc. 67, 1904–1915.
optimization. Chem. Rev. 110, 2795–2838.
31. Sunahara, H., Urano, Y., Kojima, H., and
Nagano, T. (2007). Design and synthesis of a
library of BODIPY-based environmental
polarity sensors utilizing photoinduced
electron-transfer-controlled fluorescence ON/
OFF switching. J. Am. Chem. Soc. 129, 5597–
5604.
7. Zhao, J.Z., Xu, K.J., Yang, W.B., Wang, Z.J., and
Zhong, F.F. (2015). The triplet excited state of
Bodipy: formation, modulation and
19. Chen, K., Yang, W., Wang, Z., Iagatti, A.,
Bussotti, L., Foggi, P., Ji, W., Zhao, J., and Di
Donato, M. (2017). Triplet excited state of
BODIPY accessed by charge recombination
and its application in triplet–triplet annihilation
upconversion. J. Phys. Chem. A 121, 7550–
7564.
application. Chem. Soc. Rev. 44, 8904–8939.
8. Smith, M.B., and Michl, J. (2013). Recent
advances in singlet fission. Annu. Rev. Phys.
Chem. 64, 361–386.
32. Loudet, A., and Burgess, K. (2007). BODIPY
dyes and their derivatives: syntheses and
spectroscopic properties. Chem. Rev. 107,
4891–4932.
9. Turro, N.J., Ramamurhy, V., and Scaiano, J.C.
(2009). Principles of Molecular Photochemistry:
An Introduction (University Science Books).
20. Montero, R., Martı´nez-Martı´nez, V., Longarte,
A., Epelde-Elezcano, N., Palao, E., Lamas, I.,
Manzano, H., Agarrabeitia, A.R., Lo´ pez
33. Wu, W., Sun, J., Cui, X., and Zhao, J. (2013).
Observation of the room temperature
phosphorescence of Bodipy in visible light-
harvesting Ru(II) polyimine complexes and
application as triplet photosensitizers for
triplet-triplet-annihilation upconversion and
photocatalytic oxidation. J. Mater. Chem. C 1,
4577–4589.
10. El-Sayed, M.A. (1974). Effect of spin orbit
interactions on the dipolar nature of the
radiative microwave zero-field transitions in
aromatic molecules. J. Chem. Phys. 60, 4502–
4507.
Arbeloa, I., Ortiz, M.J., et al. (2018). Singlet
fission mediated photophysics of BODIPY
dimers. J. Phys. Chem. Lett. 9, 641–646.
21. Liu, Y., Zhao, J., Iagatti, A., Bussotti, L., Foggi,
P., Castellucci, E., Di Donato, M., and Han, K.-L.
(2018). A revisit to the orthogonal bodipy
dimers: experimental evidence for the
symmetry breaking charge transfer-induced
intersystem crossing. J. Phys. Chem. C 122,
2502–2511.
11. Okada, T., Karaki, I., Matsuzawa, E., Mataga,
N., Sakata, Y., and Misumi, S. (1981). Ultrafast
intersystem crossing in some intramolecular
heteroexcimers. J. Phys. Chem. 85, 3957–3960.
34. Galletta, M., Campagna, S., Quesada, M.,
Ulrich, G., and Ziessel, R. (2005). The elusive
phosphorescence of pyrromethene-BF2 dyes
revealed in new multicomponent species
containing Ru(II)-terpyridine subunits. Chem.
Commun. (Camb.), 4222–4224.
12. van Willigen, H., Jones, G., and Farahat, M.S.
(1996). Time-resolved EPR study of
22. Sartor, S.M., McCarthy, B.G., Pearson, R.M.,
Miyake, G.M., and Damrauer, N.H. (2018).
Exploiting charge-transfer states for
maximizing intersystem crossing yields in
organic photoredox catalysts. J. Am. Chem.
Soc. 140, 4778–4781.
photoexcited triplet-state formation in
electron-donor-substituted acridinium ions.
J. Phys. Chem. 100, 3312–3316.
35. Head-Gordon, M., Grana, A.M., Maurice, D.,
and White, C.A. (1995). Analysis of electronic
transitions as the difference of electron
attachment and detachment densities. J. Phys.
Chem. 99, 14261–14270.
13. Dance, Z.E.X., Mi, Q.X., McCamant, D.W.,
Ahrens, M.J., Ratner, M.A., and Wasielewski,
M.R. (2006). Time-resolved EPR studies of
photogenerated radical ion pairs separated by 23. Wishart, J.F., Cook, A.R., and Miller, J.R. (2004).
p-phenylene oligomers and of triplet states
resulting from charge recombination. J. Phys.
Chem. B 110, 25163–25173.
The LEAF picosecond pulse radiolysis facility at
Brookhaven National Laboratory. Rev. Sci.
Instrum. 75, 4359–4366.
36. Kee, H.L., Kirmaier, C., Yu, L., Thamyongkit, P.,
Youngblood, W.J., Calder, M.E., Ramos, L.,
Noll, B.C., Bocian, D.F., Scheidt, W.R., et al.
Chem 5, 1–18, January 10, 2019
17

Please cite this article in press as: Buck et al., Spin-Allowed Transitions Control the Formation of Triplet Excited States in Orthogonal Donor-
Acceptor Dyads, Chem (2018), https://doi.org/10.1016/j.chempr.2018.10.001
(2005). Structural control of the photodynamics
of boron-dipyrrin complexes. J. Phys. Chem. B
109, 20433–20443.
50. Weiss, E.A., Ahrens, M.J., Sinks, L.E., Gusev,
A.V., Ratner, M.A., and Wasielewski, M.R.
(2004). Making a molecular wire: charge and
spin transport through para-phenylene
the Marcus inverted region explained. J. Am.
Chem. Soc. 135, 9843–9848.
64. Zamadar, M., Cook, A.R., Lewandowska-
Andralojc, A., Holroyd, R., Jiang, Y., Bikalis, J.,
and Miller, J.R. (2013). Electron transfer by
excited benzoquinone anions: slow rates for
two-electron transitions. J. Phys. Chem. A 117,
8360–8367.
37. Ou, Q., and Subotnik, J.E. (2013). Electronic
relaxation in benzaldehyde evaluated via TD-
DFT and localized diabatization: intersystem
crossings, conical intersections, and
phosphorescence. J. Phys. Chem. C 117,
19839–19849.
oligomers. J. Am. Chem. Soc. 126, 5577–5584.
51. Rodgers, C.T., and Hore, P.J. (2009). Chemical
magnetoreception in birds: the radical pair
mechanism. Proc. Natl. Acad. Sci. U S A 106,
353–360.
65. Closs, G.L., and Miller, J.R. (1988).
38. Newton, M.D. (2000). Modeling donor/
acceptor interactions: combined roles of
theory and computation. Int. J. Quan. Chem.
77, 255–263.
52. Treichel, R., Staerk, H., and Weller, A. (1983).
Magnetic field effect measurements by time-
selective absorption sampling. Appl. Phys. B
31, 15–17.
Intramolecular long-distance electron transfer
in organic molecules. Science 240, 440–447.
66. Yogo, T., Urano, Y., Ishitsuka, Y., Maniwa, F.,
and Nagano, T. (2005). Highly efficient and
photostable photosensitizer based on BODIPY
chromophore. J. Am. Chem. Soc. 127, 12162–
12163.
39. Sachs, S.B., Dudek, S.P., Hsung, R.P., Sita, L.R.,
Smalley, J.F., Newton, M.D., Feldberg, S.W.,
and Chidsey, C.E.D. (1997). Rates of interfacial
electron transfer through p-conjugated
53. Schulten, K., Staerk, H., Weller, A., Werner,
H.J., and Nickel, B. (1976). Magnetic field
dependence of the geminate recombination of
radical ion pairs in polar solvents. Z. Phys.
Chem. 101, 371–390.
spacers. J. Am. Chem. Soc. 119, 10563–10564.
67. Zhou, Q., Zhou, M., Wei, Y., Zhou, X., Liu, S.,
Zhang, S., and Zhang, B. (2017). Solvent effects
on the triplet-triplet annihilation upconversion
of diiodo-Bodipy and perylene. Phys. Chem.
Chem. Phys. 19, 1516–1525.
40. Newton, M.D. (1991). Quantum chemical
probes of electron-transfer kinetics—the
nature of donor-acceptor interactions. Chem.
Rev. 91, 767–792.
54. Marcus, R.A. (1956). On the theory of oxidation-
reduction reactions involving electron transfer
.1. J. Chem. Phys. 24, 966–978.
55. Jortner, J. (1976). Temperature dependent
activation energy for electron transfer between
biological molecules. J. Chem. Phys. 64, 4860–
4867.
41. Montalti, M., Credi, A., Prodi, L., and Gandolfi,
M. (2006). Handbook of Photochemistry, Third
Edition (CRC Press).
68. Zhang, C., Zhao, J., Wu, S., Wang, Z., Wu, W.,
Ma, J., Guo, S., and Huang, L. (2013).
Intramolecular RET enhanced visible light-
absorbing bodipy organic triplet
42. Steiner, U.E., and Ulrich, T. (1989). Magnetic-
field effects in chemical-kinetics and related
phenomena. Chem. Rev. 89, 51–147.
56. Miller, J.R., Beitz, J.V., and Huddleston, R.K.
(1984). Effect of free energy on rates of electron
transfer between molecules. J. Am. Chem. Soc.
106, 5057–5068.
photosensitizers and application in
photooxidation and triplet-triplet annihilation
upconversion. J. Am. Chem. Soc. 135, 10566–
10578.
43. Klein, J.H., Schmidt, D., Steiner, U.E., and
Lambert, C. (2015). Complete monitoring of
coherent and incoherent spin flip domains in
the recombination of charge-separated states
of donor-iridium complex-acceptor triads.
J. Am. Chem. Soc. 137, 11011–11021.
57. Osuka, A., Noya, G., Taniguchi, S., Okada, T.,
Nishimura, Y., Yamazaki, I., and Mataga, N.
(2000). Energy-gap dependence of
69. Dreeskamp, H., Koch, E., and Zander, M.
(1975). Fluorescence of bromoperylenes and
the requirements of heavy-atom quenching.
Chem. Phys. Lett. 31, 251–253.
photoinduced charge separation and
subsequent charge recombination in 1,4-
phenylene-bridged zinc-free-base hybrid
porphyrins. Chem. Eur. J. 6, 33–46.
44. Mani, T., Niedzwiedzki, D.M., and Vinogradov,
S.A. (2012). Generation of phosphorescent
triplet states via photoinduced electron
transfer: energy and electron transfer dynamics
in Pt porphyrin-rhodamine B dyads. J. Phys.
Chem. A 116, 3598–3610.
70. Fischer, M., and Georges, J. (1996).
Fluorescence quantum yield of rhodamine 6G
in ethanol as a function of concentration using
thermal lens spectrometry. Chem. Phys. Lett.
260, 115–118.
58. Subotnik, J.E., Vura-Weis, J., Sodt, A.J., and
Ratner, M.A. (2010). Predicting accurate
electronic excitation transfer rates via Marcus
theory with Boys or Edmiston-Ruedenberg
localized diabatization. J. Phys. Chem. A 114,
8665–8675.
71. Bensasson, R.V., and Gramain, J.-C. (1980).
Benzophenone triplet properties in acetonitrile
and water. reduction by lactams. J. Chem. Soc.
Faraday Trans. 76, 1801–1810.
45. Chidsey, C.E.D., Takiff, L., Goldstein, R.A., and
Boxer, S.G. (1985). Effect of magnetic-fields on
the triplet-state lifetime in photosynthetic
reaction centers—evidence for thermal
59. Winkler, J.R., and Gray, H.B. (1992). Electron
transfer in ruthenium-modified proteins.
Chem. Rev. 92, 369–379.
repopulation of the initial radical pair. Proc.
Natl. Acad. Sci. U S A 82, 6850–6854.
72. Shao, Y., Gan, Z., Epifanovsky, E., Gilbert,
A.T.B., Wormit, M., Kussmann, J., Lange, A.W.,
Behn, A., Deng, J., Feng, X., et al. (2015).
Advances in molecular quantum chemistry
contained in the Q-Chem 4 program package.
Mol. Phys. 113, 184–215.
60. Sukegawa, J., Schubert, C., Zhu, X.Z., Tsuji, H.,
Guldi, D.M., and Nakamura, E. (2014). Electron
transfer through rigid organic molecular wires
enhanced by electronic and electron-vibration
coupling. Nat. Chem. 6, 899–905.
46. Wasielewski, M.R. (2006). Energy, charge, and
spin transport in molecules and self-assembled
nanostructures inspired by photosynthesis.
J. Org. Chem. 71, 5051–5066.
47. Rudolf, M., Feng, L., Slanina, Z., Akasaka, T.,
Nagase, S., and Guldi, D.M. (2013). A
metallofullerene electron donor that powers an
efficient spin flip in a linear electron donor-
acceptor conjugate. J. Am. Chem. Soc. 135,
11165–11174.
61. Miller, J.R., Calcaterra, L.T., and Closs, G.L.
(1984). Intramolecular long-distance electron
transfer in radical anions. The effects of free
energy and solvent on the reaction rates. J. Am.
Chem. Soc. 106, 3047–3049.
73. Chai, J.-D., and Head-Gordon, M. (2008).
Systematic optimization of long-range
corrected hybrid density functionals. J. Chem.
Phys. 128, 084106.
74. Yanai, T., Tew, D.P., and Handy, N.C. (2004).
A new hybrid exchange-correlation
functional using the Coulomb-attenuating
method (CAM-B3LYP). Chem. Phys. Lett. 393,
51–57.
62. Closs, G.L., Calcaterra, L.T., Green, N.J.,
Penfield, K.W., and Miller, J.R. (1986). Distance,
stereoelectronic effects, and the Marcus
inverted region in intramolecular electron
transfer in organic radical anions. J. Phys.
Chem. 90, 3673–3683.
48. Saha, A., Chen, M., Lederer, M., Kahnt, A., Lu,
X., and Guldi, D.M. (2017). Sulfur rich electron
donors—formation of singlet versus triplet
radical ion pair states featuring different
lifetimes in the same conjugate. Chem. Sci. 8,
1360–1368.
75. Truong, T.N., and Stefanovich, E.V. (1995). A
new method for incorporating solvent effect
into the classical, ab initio molecular orbital and
density functional theory frameworks for
arbitrary shape cavity. Chem. Phys. Lett. 240,
253–260.
63. Koch, M., Rosspeintner, A., Adamczyk, K.,
Lang, B., Dreyer, J., Nibbering, E.T.J., and
Vauthey, E. (2013). Real-time observation of the
formation of excited radical ions in bimolecular
photoinduced charge separation: absence of
49. Hisaharu, H., and Saburo, N. (1984). Theoretical
study of relaxation mechanism in magnetic
field effects on chemical reactions. Bull. Chem.
Soc. Jpn. 57, 322–328.
18
Chem 5, 1–18, January 10, 2019