Diphenylanthracene Dimers for Triplet-Triplet Annihilation Photon Upconversion: Mechanistic Insights for Intramolecular Pathways and the Importance of Molecular Geometry

Article abstract of DOI:10.1021/jacs.1c00331

Novel approaches to modify the spectral output of the sun have seen a surge in interest recently, with triplet-triplet annihilation driven photon upconversion (TTA-UC) gaining widespread recognition due to its ability to function under low-intensity, noncoherent light. Herein, four diphenylanthracene (DPA) dimers are investigated to explore how the structure of these dimers affects upconversion efficiency. Also, the mechanism responsible for intramolecular upconversion is elucidated. In particular, two models are compared using steady-state and time-resolved simulations of the TTA-UC emission intensities and kinetics. All dimers perform TTA-UC efficiently in the presence of the sensitizer platinum octaethylporphyrin. The meta-coupled dimer 1,3-DPA2 performs best yielding a 21.2% upconversion quantum yield (out of a 50% maximum), which is close to that of the reference monomer DPA (24.0%). Its superior performance compared to the other dimers is primarily ascribed to the longer triplet lifetime of this dimer (4.7 ms), thus reinforcing the importance of this parameter. Comparisons between simulations and experiments reveal that the double-sensitization mechanism is part of the mechanism of intramolecular upconversion and that this additional pathway could be of great significance under specific conditions. The results from this study can thus act as a guide not only in terms of annihilator design but also for the design of future solid-state systems where intramolecular exciton migration is anticipated to play a major role.

Full text of DOI:10.1021/jacs.1c00331

pubs.acs.org/JACS
Article
Diphenylanthracene Dimers for Triplet−Triplet Annihilation Photon
Upconversion: Mechanistic Insights for Intramolecular Pathways
and the Importance of Molecular Geometry
̊
Axel Olesund, Victor Gray, Jerker Martensson, and Bo Albinsson*
Cite This: J. Am. Chem. Soc. 2021, 143, 5745−5754
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Novel approaches to modify the spectral output of
the sun have seen a surge in interest recently, with triplet−triplet
annihilation driven photon upconversion (TTA-UC) gaining
widespread recognition due to its ability to function under low-
intensity, noncoherent light. Herein, four diphenylanthracene
(DPA) dimers are investigated to explore how the structure of
these dimers affects upconversion efficiency. Also, the mechanism
responsible for intramolecular upconversion is elucidated. In
particular, two models are compared using steady-state and time-
resolved simulations of the TTA-UC emission intensities and
kinetics. All dimers perform TTA-UC efficiently in the presence of
the sensitizer platinum octaethylporphyrin. The meta-coupled
dimer 1,3-DPA2 performs best yielding a 21.2% upconversion
quantum yield (out of a 50% maximum), which is close to that of the reference monomer DPA (24.0%). Its superior performance
compared to the other dimers is primarily ascribed to the longer triplet lifetime of this dimer (4.7 ms), thus reinforcing the
importance of this parameter. Comparisons between simulations and experiments reveal that the double-sensitization mechanism is
part of the mechanism of intramolecular upconversion and that this additional pathway could be of great significance under specific
conditions. The results from this study can thus act as a guide not only in terms of annihilator design but also for the design of future
solid-state systems where intramolecular exciton migration is anticipated to play a major role.
INTRODUCTION
■
The ability to manipulate incoming sunlight to improve the
spectral matching for different applications has seen great
progress in recent years. Shockley and Queisser famously
derived the intrinsic limit for p−n junction solar cell
efficiencies more than 50 years ago,1 and two different ways
of breaking this limit have gained extensive attention. To
mitigate thermalization losses due to excess energy in the
Figure 1. Jablonski diagram depicting the energy levels and transfer
photons arriving at the junction, singlet fission (SF), a process
steps involved in photon upconversion by triplet−triplet annihilation
in which a singlet excited state may be converted into two
(TTA-UC). S = sensitizer; A = annihilator.
triplet excited states, has been proposed.2 The reverse of SF,
that is, combining two low-energy photons into one high-
sensitizer (S) and an annihilator (A). The sensitizer molecule
energy excited state (exciton), is referred to as photon
absorbs one long wavelength photon and undergoes rapid
upconversion.3,4 This could be used to manage the losses
intersystem crossing (ISC). Dexter-type triplet energy transfer
associated with insufficient photon energy, thus expanding the
(TET)17 from the sensitizer subsequently populates the first
spectral range of the solar cell.5,6 The process may operate by
several different mechanisms, and triplet−triplet annihilation
Received: January 11, 2021
Published: April 9, 2021
upconversion (TTA-UC) is one possible mechanism (Figure
1) that may proceed utilizing low-intensity, noncoherent light.
Thus, TTA-UC is of special interest not only for photo-
voltaics7−10 but also in other solar energy conversion
applications, such as photocatalysis11−13 and photochemis-
try.14−16 Two compounds are needed in this process: a
© 2021 The Authors. Published by
https://doi.org/10.1021/jacs.1c00331
J. Am. Chem. Soc. 2021, 143, 5745−5754
American Chemical Society
5745

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
triplet excited state (3A*) of the annihilator molecule. This is
then followed by TTA between two triplet excited annihilators,
forming one ground-state (1A) and one singlet excited-state
(1A*) annihilator, of which the latter may emit one shorter-
wavelength photon.
The most efficient systems developed to date are liquid
solutions of S and A,18 allowing for swift diffusion of long-lived
triplets and thus promoting the short-range Dexter-type events.
For practical applications solid-state solutions are however
needed, introducing obstacles in terms of inefficient energy
transfer due to spatial separation between the molecules.19
One way to mitigate these obstructions is by allowing the TTA
event to proceed in an intramolecular fashion in addition to
the intermolecular counterpart, which would require excitons
to diffuse within the annihilator instead of relying on molecular
diffusion.20−33 Our group has previously studied dendrimeric
and oligomeric frameworks of covalently bonded 9,10-
diphenylanthracene (DPA) units, showing that intramolecular
TTA-UC in rigid environments is promoted by large
annihilator frameworks.23 This behavior was further inves-
tigated using porphyrin−anthracene complexes, showing that
parasitic back energy transfer from the anthracene unit to the
covalently attached porphyrin is subdued as the anthracene
unit increases in size.24,32 Several recent contributions have
investigated different types of dimeric annihilators based on
DPA29,31 and tetracene,22,33 respectively, in which intra-
molecular TTA (intra-TTA) is hypothesized to occur. A full
mechanistic picture of intramolecular upconversion is however
lacking, and different models for the energy transfer events
have been proposed.23,29,31,33
In this study, four novel dimeric compounds based on DPA
have been synthesized and investigated. The dimeric nature
allows these annihilators to hold two triplets (triplet excited
states) simultaneously, thus enabling intra-TTA in addition to
conventional intermolecular TTA. The importance of structure
and molecular geometry has been thoroughly investigated for
SF and has been shown to play an important role in governing
the SF efficiency and rate.34−40 In contrast, the structural
effects of intra-TTA have not been fully understood. These
dimers allow us not only to investigate the mechanism of
intramolecular upconversion but also to draw conclusions with
regard to how structural motifs of the annihilators relate to the
upconversion performance.
Figure 2. Absorption (dashed) and emission (dot−dash) spectra of
the investigated annihilators and the sensitizer PtOEP.
and rapid intersystem crossing.41 The molar absorptivity of the
dimers is roughly two times higher than for DPA, which is
expected given the dimeric nature of these compounds. The
vibronic progression is clearly visible in the absorption features
of all dimers, with relative intensities of the vibronic peaks
being the major difference between different spectra. Only
minor shifts of the peak positions were observed. More
pronounced differences were found upon investigation of the
fluorescence properties. All compounds have a high
fluorescence quantum yield close to unity in deaerated toluene,
thus fulfilling one important prerequisite to perform well as an
annihilator in TTA-UC.4 The fluorescence lifetimes are slightly
shorter than for DPA for all dimers except 1,2-DPA2. Although
not quantitatively reflected in the increased molar absorptiv-
ities, the faster radiative rates of the dimers show that the
linked DPA chromophores interact electronically. Too weak
emission at 410 nm from 1,2-DPA2 at room temperature
hindered us from properly monitoring the lifetime of
monomeric fluorescence, i.e., fluorescence originating from
the isolated, noninteracting DPA moieties. The tabulated
lifetime at 410 nm was instead measured at 93 K where short-
wavelength fluorescence dominates. The findings are summar-
ized in Table 1. A substantial red shift is seen for the
fluorescence of 1,2- DPA2, with only a small fraction of the
fluorescence taking place at shorter wavelengths at room
temperature (Figure S11). This behavior is ascribed to the
formation of an excited dimer, or excimer, upon excitation and
RESULTS AND DISCUSSION
■
Design and Synthesis of DPA Dimers. Synthesis details
(Figure S1) as well as the proton and carbon NMR spectra for
the dimers (Figures S2−S9) are found in the Supporting
Information. The dimers consist of two anthracene units which
are connected in four different ways. Three of the dimers, 1,2-
bis(10-phenylanthracen-9-yl)benzene (1,2-DPA2), 1,3-bis(10-
phenylanthracen-9-yl)benzene (1,3-DPA2), and 1,4-bis(10-
phenylanthracen-9-yl)benzene (1,4-DPA2), have a central
phenyl ring, and the numeric indices indicate at which
positions the anthracenes connect to the ring. In comparison,
10,10′-diphenyl-9,9′-bianthracene (9,9′-PA2) lack the central
phenyl ring connector, and instead the anthracene moieties are
directly connected at their respective 9-positions.
Photophysical Characterization. Figure 2 presents the
absorption and fluorescence spectra of DPA and the dimer
compounds, alongside that of platinum octaethylporphyrin
(PtOEP). The latter was employed as the sensitizer during
TTA-UC owing to its suitable energetics, strong absorption,
5746
https://doi.org/10.1021/jacs.1c00331
J. Am. Chem. Soc. 2021, 143, 5745−5754

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
ϕUC = f × ϕISC × ϕTET × ϕTTA × ϕ
Table 1. Photophysical Properties of DPA and Dimers
(1)
f
a
b
c
E0−0 (nm)
λem (nm)
ϕf
τf (ns)
6.91
6.56
7.32 , 44.0
5.85
Here, ϕISC is the intersystem crossing quantum yield of the
sensitizer, ϕTET is the quantum yield of triplet energy transfer
from sensitizer to annihilator, ϕTTA is the quantum yield of
triplet−triplet annihilation, and ϕf is the fluorescence quantum
yield of the annihilator. The spin statistical factor, f, defines the
fraction of excited annihilator triplets that go on to create an
emissive singlet excited state following TTA.18 Evaluating each
of the terms in eq 1 is not viable, and instead the method of
relative actinometry (eq 2) is used:
DPA
393 (13.6)
402 (35.0)
403 (17.3)
398 (27.7)
398 (26.7)
409
431
511
411
416
1.00
1.00
0.95
0.96
0.95
9,9′-PA2
1,2-DPA2
1,3-DPA2
1,4-DPA2
d
e
4.56
a
Wavelength of 0 → 0 transition and molar absorptivity, ε (×103 M−1
b
cm−1). Wavelength of maximum emission upon 377 nm excitation at
295 K. Fluorescence lifetime probed at 410 nm and 295 K. Probed
at 410 nm and 93 K. Probed at 510 nm and 295 K.
c
d
e
η2
Ar FUC
r
ϕUC = ϕ
r AUC
F
ηU2C
r
(2)
has previously been observed for a number of different
annihilator molecules.42−44 The creation of excimers is
normally highly dependent on chromophore concentration as
it is formed through interactions between a ground-state and
an excited-state molecule. The fluorescence from 1,2-DPA2
does not show this dependence on concentration. The long-
wavelength emission is dominating also for low-concentration
samples, and thus, the excimer is believed to form by means of
intramolecular interactions.42,45,46 The behavior and kinetics of
1,2-DPA2 were investigated thoroughly but are not the main
focus of this paper, and the interested reader is referred to
Section 3 of the Supporting Information. It is important to
point out that, at room temperature, the lifetime of the initially
excited singlet state of 1,2-DPA2 is very short (<10 ps) due to
quantitative transformation into the singlet excited excimer,
which is why all emission emanates from the excimer
irrespectively if populated through TTA or direct excitation.
A large fraction of this emission lies at longer wavelength than
532 nm and is, thus, not strictly speaking upconverted.
Upconversion Study. The upconversion potential of the
four dimers has been evaluated in a series of experiments. The
upconverting samples consisted of 6.6 μM PtOEP and 1 mM
(DPA) or 0.5 mM (dimers) annihilator, which upon 532 nm
excitation produced bright upconverted fluorescence (Figures
S16−S17). No upconverted emission could be detected when
PtOEP was absent from the samples.
Here, ϕr is the known fluorescence quantum yield of a
reference compound, Ai is the absorption at the excitation
wavelength, Fi is the integrated emission intensity, and ηi is the
refractive index. Subscripts r and UC denote reference sample
and upconversion sample, respectively. Rhodamine 6G in air-
saturated ethanol was employed as the reference compound
(ϕf = 0.95)48 during UC measurements.
All dimers exhibit efficient upconversion upon 532 nm
excitation (Table 2). 9,9′-PA2, 1,2-DPA2, and 1,4-DPA2
perform similarly, yielding a quantum yield of around 15%.
Interestingly, 1,3-DPA2 outperforms the other dimers, showing
an impressive 21.2% ϕUC which is comparable with 24.0% for
DPA. Dimerization in the meta-position has previously yielded
higher UC efficiencies than the corresponding para- and ortho-
connected annihilators,31 and our results indicate that
connections in the meta-position are beneficial at least for
DPA-type dimer annihilators. The calculated thermodynamic
driving forces for TTA, i.e., 2 × E(T1) − E(S1), are similar for
all annihilators and do not explain the differences in our
results.
In order to understand why 1,3-DPA2 exhibits a higher ϕUC
than the other dimers, further experiments aimed at under-
standing the kinetics involved were performed. Efficient triplet
energy transfer between sensitizer and annihilator is a feature
present in all high-performing UC systems and may be
investigated using Stern−Volmer quenching (eq 3):
A figure of merit which is of big importance when evaluating
upconverting systems is the upconversion quantum yield, ϕUC
.
I0
I
= 1 + kTETτ0[A]
It is defined as the number of emitted high-energy photons
compared with the number of absorbed low-energy photons
and can thus take a maximum value of 50%.47 The efficiency is
dependent on the quantum yield of all steps leading up to the
emission of photons in accordance with eq 1:
(3)
I and I0 are the quenched and unquenched donor emission,
respectively, kTET is the rate constant for TET from donor to
acceptor, τ0 is the lifetime of the unquenched donor, and [A] is
Table 2. Experimentally and Computationally Determined Parameters Important to the Upconversion Process
b
c
d
e
f
g
h
j
Ith
kTET (×109
kTTA (×109
τT
kT (×103
E(S1)
E(T1)
2 × E(T1) − E(S1)
a
ΦUC
(mW/cm2)
M−1 s−1
)
M−1 s−1
)
(ms)
s−1
)
(eV)
(eV)
(eV)
i
DPA
0.240
0.150
0.140
0.212
0.149
15
1.78
0.99
0.95
1.04
0.96
3.01
3.73
2.89
2.81
4.00
5.5
0.18
1.79
1.25
0.21
3.44
3.05/3.15
2.86/3.08
2.85/3.08
3.04/3.12
3.06/3.12
1.72
0.29
0.36
0.34
0.32
0.32
9,9′-PA2
1,2-DPA2
1,3-DPA2
1,4-DPA2
605
142
44
0.56
0.80
4.7
1.72
1.71
1.72
1.72
1343
0.29
a
b
Upconversion quantum yield relative to a theoretical maximum of 0.5 (50%). Threshold intensity. Individual values have been normalized with
c
d
respect to slight deviations in [S] between samples. Rate constant for triplet energy transfer from PtOEP. Rate constant for triplet−triplet
e
f
g
annihilation. First triplet excited-state lifetimes. Rate constant for intrinsic triplet decay. Energy of the first singlet excited state as calculated with
TD-DFT (B3LYP/6-31G**)/calculated from the 0 → 0 transition of the absorption spectra. See the Supporting Information for calculation details.
h
i
j
Energy of the first triplet excited state as calculated with TD-DFT. Experimental literature value50 is 1.77 eV. Thermodynamic driving force for
TTA.
5747
https://doi.org/10.1021/jacs.1c00331
J. Am. Chem. Soc. 2021, 143, 5745−5754

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
the quencher concentration. The quenching of PtOEP with the
five different annihilators was determined by monitoring the
phosphorescence of PtOEP, and the resulting Stern−Volmer
plots are presented in Figure 3. The unquenched lifetime of
upconverting samples. Only the rate of intrinsic triplet decay of
the annihilator, kT, and the rate of TTA, kTTA, will thus
influence differences in Ith between annihilators. The Ith value
defines the crossing point between a quadratic region where
the supply of annihilator triplets is the limiting factor and a
linear region where UC proceeds with the highest possible
efficiency. For practical applications it is desirable to have as
low an Ith as possible, as this facilitates possible applications
using solar energy (the power density from the sun between
470 and 550 nm, i.e., the absorption range for the PtOEP Q-
band, is approximately 15 mW/cm2 under AM1.5 solar
irradiation). The excitation power density dependence on
the UC emission intensity of our systems is presented in Figure
4. 1,2-DPA2 and 1,3-DPA2 have Ith values reasonably close to
what is usually seen for compounds based on DPA (142 and
44 mW/cm2, respectively) but are significantly higher than the
acquired value for DPA (15 mW/cm2, see also Table 2). The
less congested dimers 9,9′-PA2 and 1,4-DPA2 have Ith values of
605 and 1343 mW/cm2, respectively, which is more than 1
order of magnitude higher than that of DPA. At closer
examination of Figure 4B it becomes obvious that the linear
regime (slope 1) has not been fully reached by either 9,9′-PA2
or 1,4-DPA2 at our setups maximum power density, and the
presented values thus represent a lower estimate of their
respective Ith value. The spread in Ith values is a bit surprising
given that other studies on DPA-based molecules have shown a
narrow distribution of threshold intensities and often close to
that of DPA.29,31,51 These unexpected results were further
analyzed by examining eq 4 and seeking to determine the
parameters involved.
Figure 3. Stern−Volmer plots of TET from PtOEP (3.4 μM) to the
five annihilator compounds.
PtOEP was determined to be 95 μs in deaerated toluene
(Figure S20), and TET from PtOEP to the annihilators is
found to be very efficient in all cases. The kTET rate constants
of the dimers are roughly half that of DPA (measured in DPA
subunit concentrations, Table 2), which is reasonable given
that the lower diffusivity of the dimers is compensated for by a
larger collision radius according to the Smoluchowski
equation.49 All TET efficiencies are 99% or higher for DPA
subunit concentrations of [A] = 1 mM, and it was concluded
that the TET step is not the determining factor when
evaluating these dimers against each other.
Vital to an efficient upconversion process are long-lived
annihilator triplet states, as this allows for diffusion-mediated
annihilation to proceed more efficiently. Using time-resolved
emission measurements and a fitting procedure based on eq 5,
the annihilator triplet lifetimes, τT, and related kT rate
constants could be determined (Table 2). The annihilator
triplet state may decay by both first- and second-order
channels, and the observed kinetics of the upconverted
fluorescence will thus depend on the annihilator triplet
concentration:52
Another important figure of merit when evaluating
upconverting systems is the threshold intensity, Ith. This is
the excitation intensity at which ϕUC reaches 50% of a systems’
maximum and is related to system parameters as in eq 4:
kT2
Ith
=
2kTTAα[1S]
(4)
The absorption cross section, α, and ground-state
concentration, [1S], of the sensitizer are the same in all
Figure 4. (A, B) Double-logarithmic plots of upconversion emission intensity versus excitation power density (532 nm). The threshold intensity Ith
for each annihilator is indicated by vertical dashed lines, and quadratic and linear slopes (solid lines) for each compound are included for facile
evaluation of Ith. Individual plots have been vertically shifted for clarity purposes. (C) UC quantum yield versus excitation power density for the
annihilators.
5748
https://doi.org/10.1021/jacs.1c00331
J. Am. Chem. Soc. 2021, 143, 5745−5754

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
2
DPA2 are consistent with higher Ith values. The longer lifetime
of the triplet state in 1,3-DPA2 could potentially be explained
by the decreased electronic coupling between the moieties due
to linkage in the meta-position,31,53−55 thus better retaining the
properties of the DPA monomer. This notion was
strengthened by calculations of the electronic coupling
between the triplet excited states of the dimers (see Supporting
Information, Section 4.4). The second parameter that affects
the threshold intensity is the rate of annihilation. kTTA was
determined for all compounds using nanosecond transient
absorption, monitoring UC samples at 430 nm and at 646 nm
following 532 nm excitation. The transient signals are
presented in Figures S26−S30, and a global fitting procedure
was used to obtain the rate constants of interest (the full
procedure is explained in the Supporting Information, Section
4.3). The kTTA rates are quite similar (see Table 2), but
interestingly 9,9′-PA2 and 1,4-DPA2 show slightly elevated
rates, even compared to DPA.
i
j
j
y
z
z
z
1 − β
exp(t/τT) − β
3
2
3
I(t) ∝ [ A*] = [ A*]
j
0
j
j
z
z
(5)
k
{
Here, I(t) is the emission intensity, β is a dimensionless
parameter between 0 and 1 expressing what fraction of initial
decay is governed by second-order channels (with β = 1
meaning all initial decay is of second order), and t is time. Full
details are found in the Supporting Information. As was the
case with Ith, these compounds show a rather broad
distribution of triplet lifetimes, ranging from 291 μs for 1,4-
DPA2 to 5.5 ms for DPA (Figures S21−S24 and Table 2). 1,3-
DPA2 (Figure 5) has the most long-lived triplet out of the
Intramolecular Upconversion in Solution. The design
of our annihilators was in part based on the possibility of
populating the molecules with two triplets simultaneously, as
this enables intramolecular upconversion (hereon after referred
to as intra-UC). Instead of relying on molecular diffusion, this
mechanism may rely on the diffusion of excitons within a
molecular system. Several recent studies exploiting dimer-
ic,22,29,31,33 polymeric,20,21,26−28,30 and oligomeric23 structures
have sought to investigate the influence of this additional
pathway. In a recent study, Gao et al. studied structurally
similar DPA derivatives which were separated with additional
phenyl groups.31 The coupling pattern, para-, ortho-, meta-,
was however the same as in the present study. Interestingly, the
ϕUC trend follows what we observe here: DPA > meta- >
ortho- and para-. However, in their case no significant
difference in the triplet lifetime was observed between the
derivatives and can thus not explain this trend. The difference
Figure 5. Time-resolved, delayed upconverted fluorescence from 1,3-
DPA2 in the presence of PtOEP. The emission is measured at 430 nm
and at different pump excitation intensities (532 nm). Solid lines are
best global fits to eq 5.
dimers, which explains the high ϕUC as well as the relatively
low Ith, while the shorter triplet lifetimes of 9,9′-PA2 and 1,4-
a
Scheme 1. Intermolecular Upconversion vs Two Suggested Models for Intramolecular Upconversion: TETA and DS
a
Schematic of suggested pathways for triplet−triplet annihilation upconversion with dimeric annihilator compounds (here represented by 1,3-
DPA2). Designations: S = sensitizer, A = annihilator moiety (spin multiplicity denoted by left superscript), * = excited state. Upon light absorption
(hν1) and rapid intersystem crossing (ISC), the sensitizer populates the triplet excited state of one annihilator moiety through triplet energy transfer
(TET). The TTA event forms a singlet excited state, and one high-energy photon (hν2, ν2 > ν1) is emitted. (1) Conventional intermolecular TTA
between two triplet excited dimers. (2a) TET between annihilators (TETA) model: The triplet excited dimer becomes doubly excited following
TET from another triplet excited annihilator. (2b) Double sensitization (DS) model: The ground-state moiety of the singly triplet excited dimer is
populated with another triplet following TET from a sensitizer molecule.
5749
https://doi.org/10.1021/jacs.1c00331
J. Am. Chem. Soc. 2021, 143, 5745−5754

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 6. Simulation results for the (A, E) TETA model, (B, F) DS model, (C, G) model where intra-TTA contributions are disallowed, and (D,
H) experimental kinetic traces from delayed UC fluorescence emanating from (A−D) samples with [1S]0 = 5 μM and [1A]0 = 1 mM (i.e., [1A]0/
[1S]0 = 100), and (E−H) samples with [1S]0 = 100 μM and [1A]0 = 5 μM (i.e., [1A]0/[1S]0 = 0.05). Emission monitored at 430 or 510 nm (for 1,2-
DPA2).
was instead assigned to greater intra-TTA contributions for the
meta-coupled dimer, which were indicated by means of
magnetic-field-dependent measurements of the UC perform-
ance. It should be noted that their reported triplet lifetime of
DPA was roughly five times shorter than that reported herein.
This discrepancy is likely caused by differences in the type of
measurement and fitting procedures, as Gao et al. used
transient absorption without taking second-order channels into
account during fitting. Figure S25 highlights that a single-
exponential tail fit significantly underestimates the triplet
lifetime, especially in systems with high TTA efficiencies.
In this section, the quest to fully understand intra-UC is
continued, as there are still disagreements on how the actual
mechanism of intra-UC proceeds in solution-based systems.
Two different mechanisms are predominantly considered in
the literature, and these are schematically presented in Scheme
1 alongside the intermolecular upconversion (inter-UC) route
(pathway 1 in Scheme 1) always present in solution-based
systems. The triplet energy transfer between annihilators model
(TETA model, pathway 2a in Scheme 1) has been suggested
by several authors29,31 and is based on the interactions between
The double-sensitization model (DS model, pathway 2b in
Scheme 1) relies not on TET between annihilators but rather
on a second TET step from a triplet excited sensitizer (3S*) to
3
the ground-state moiety of A*−1A, ultimately leading to a
3A*−3A* annihilator which can proceed to perform intra-TTA.
This model has been employed in a multitude of other
studies,22,23,28,33 but the conclusions of different investigations
in low-viscosity media vary significantly. In pursuit of
elucidating the nature of intra-UC, steady-state and time-
resolved simulations of the presented models were performed.
The experimentally determined parameters for each annihilator
(i.e., kTET, kT, and kTTA) were used during simulations, and the
detailed description of the kinetic model is presented in the
Supporting Information.
The two models have been evaluated for two different
sample conditions: typical UC conditions (i.e., [1A]0 = 1 mM,
[1S]0 = 5 μM), which will be referred to as high annihilator to
sensitizer ratios ([1A]0/[1S]0), and conditions with low [1A]0/
[1S]0 ratios (i.e., [1A]0 ≈ 10 μM, [1S]0 = 100 μM). This has
been done in order to investigate the expected influence from
each intra-UC mechanism separately and to draw conclusions
on in which regimes each model is valid. Time-resolved
simulations were performed, which are presented in Figure 6.
At high [1A]0/[1S]0 ratios, the TETA model predicts that the
dimer UC emission kinetics are slightly slower than in the DS
model. However, both models predict that the kinetics get
slower upon lowering [1A]0. The largest difference between the
models is the evolution of early time kinetics, which are related
to the TET events. No major difference between the models is
discernible at high [1A]0/[1S]0 ratios (Figure 6A, B), but upon
going to low [1A]0/[1S]0 it is clear that the DS model predicts
that dimers will have a much faster rise time relative to that of
DPA, a behavior not predicted by the TETA model (Figure
6E, F). Given the nature of the DS model, the faster early time
kinetics would be expected as the intra-UC pathway depends
3
two multichromophoric triplet excited annihilators, A*−1A
(in this section, annihilators are assumed to have the capacity
of holding two triplets simultaneously and are then designated
3A*−3A*). Normally the interaction between two 3A* leads to
an intermolecular TTA event where the f factor (see eq 1)
determines the yield of singlet excited states. However, when
3
two A*−1A interact there is the additional possibility of a
triplet energy transfer step from the triplet moiety of one
annihilator to the ground-state moiety of the other, creating
one 3A*−3A* and one 1A−1A annihilator. An intra-TTA event
between two triplets then rapidly gives the singlet excited state
1A*−1A and subsequent emission of an upconverted photon. It
has been suggested that the TET step between two singly
excited dimeric annihilators proceeds statistically, meaning that
3
3
half of these interactions lead to A*−3A*, while only 25%
on TET from S*, an event most likely to take place early on.
proceeds through the common intermolecular TTA pathway
To further elucidate the differences between the two models,
31
1
to produce the singlet excited state A*−1A.
steady-state simulations were also performed. Figure S33
5750
https://doi.org/10.1021/jacs.1c00331
J. Am. Chem. Soc. 2021, 143, 5745−5754

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
shows the expected steady-state delayed UC fluorescence
following 532 nm continuous-wave (cw) excitation for high
[1A]0/[1S]0 ratios, with [1A]0 ranging from 0 to 5 mM. It is
clear that the potential contribution from intra-UC is
indiscernible at higher [1A]0 according to the DS model,
3
which is expected as generated S* will quickly transfer their
1
energy to surrounding ground-state A, thus swiftly depleting
3
the S* population and hindering subsequent TET events
needed to produce the doubly excited dimers (Figure S34). On
the contrary, intra-UC contributions are expected to increase
with [1A]0 in the TETA model, as this pathway ultimately
depends mainly on [3A*−1A] (Figure S33B). Conversely,
increasing [1S]0 to 100 μM gives rise to interesting behavior at
lower [1A]0 (<10 μM), with the DS model predicting that
intra-UC will dominate, allowing annihilator dimers to exhibit
stronger UC fluorescence than their corresponding monomer
under such conditions (Figure S35A). The TETA model
predicts the resulting UC intensity to be lower than that of
DPA also under these conditions (Figure S35B).
Samples of low and high [1A]0/[1S]0 ratios were prepared,
and the behavior observed experimentally was compared with
that predicted by each model. Time-resolved measurements of
the UC emission display a marked difference in the kinetics
when going from high (Figure 6D) to low (Figure 6H) [1A]0/
[1S]0. It is clearly seen that the rise time of the UC emission of
the dimers at low [1A]0/[1S]0 is faster relative to that of DPA
(red line), even to such an extent that some dimers develop the
UC emission faster than DPA. The evolution of the rise time
kinetics in the dimers is compatible with the DS model only,
which indeed predicts that the dimer UC emission will develop
faster than that of DPA at low [1A]0/[1S]0 (Figure 6F).
Interestingly, the kinetic evolution of the individual dimers
differs slightly as well, with 1,2-DPA2 in particular showing a
substantial shortening of its rise time compared with the other
annihilators, indicating a stronger influence from intra-UC.
While this could potentially be ascribed to differences in the
intra-TTA event, simulations firmly establish that the observed
differences are in fact determined by the second sensitization
step (Figure S37). We find no reason to believe that the ortho-
coupling in 1,2-DPA2 would cause a more efficient second
sensitization step and ascribe the observed differences between
individual dimers to experimental uncertainty.
To quantify our analysis, the mean value of the rise times
and decay times of our dimers was compared with that of DPA
(Figure 7). The experimental ratios (red symbols) were then
compared to the ratios predicted by the DS model (black
symbols) and the TETA model (blue symbols) and for a
scenario where intra-UC is disallowed (purple symbols). This
was made at the two previously used [1A]0/[1S]0 ratios, and
the results show a striking agreement between our
experimental data and the model where no intra-UC occurs.
The experimental data also show good agreement with the DS
model, while the accordance to the TETA model is rather
poor. Even though our quantified time-resolved data may be
explained without involving the intra-UC pathway, such a
description is not sufficient to explain the appearance of the
kinetics in Figure 6H. Specifically, the immediate emergence of
UC emission from the dimers during the first tens of
microseconds is most likely a result of the DS mechanism
and is particularly evident in the traces of 1,2-DPA2 (green)
and 1,4-DPA2 (blue, see Figure 6F and 6G for a comparison of
the DS model kinetics and the kinetics expected if no intra-UC
contributions are present). It should be noted that the
Figure 7. Comparison of relevant time constants. The mean value for
the dimeric annihilators is compared to the value of DPA at [1A]0/
[1S]0 = 0.05 or 100. The rise times (τrise) and decay times (τdecay) are
evaluated at the peak and 1/e values of the normalized emission
traces, respectively.
remarkable agreement between experimental data and the
DS model at high [1A]0/[1S]0 ratios (Figure 7) is caused by the
fact that no intra-UC is expected under such conditions, and
the data is thus expected to coincide with that of the model
where intra-UC is disallowed completely.
Figure S38 shows the results from the steady-state
measurements, where [1A]0 has been varied while keeping
[1S]0 at 100 μM throughout, except for the rightmost data
point which is given for [1A]0 = 1 mM, [1S]0 = 6.6 μM. No
clear-cut interpretation is readily obtained from these data,
with the only prominent feature being the relative 1,4-DPA2
emission increasing as the [1A]0/[1S]0 ratio gets lower. This
could indicate that there are some contributions from intra-
TTA in 1,4-DPA2 (given that the DS mechanism is active).
However, 9,9′-PA2 and 1,3-DPA2 do also show signs of
increased relative emission, although not as systematically as
1,4-DPA2, while 1,2-DPA2 exhibits similar behavior independ-
ent of [1A]0. The steady-state measurements are rather
sensitive to experimental errors, with the evaluation of precise
intensities being subject to substantial inner-filter effects, high
sensitivity to exact sample concentrations, and possible oxygen
contamination. Similar measurements on tetracene dimers by
Pun et al. have however proved useful previously, specifically in
contexts where low kTTA rates are measured for the
upconverting materials.33 In this case the impact of the DS
mechanism is manifested also at relatively high [1A]0/[1S]0
ratios, thus facilitating comparisons between annihilators
where intra-TTA is allowed and forbidden, respectively.
Despite the fact that our experiments did not yield any
obvious signatures indicating the presence of the TETA
mechanism, a closer examination of our steady-state data might
shine some additional light on this matter. Although the data
presented in Figure S38 show differences between the
performances of individual dimers, our time-resolved measure-
ments indicate that all investigated dimeric annihilators
perform intra-TTA to some extent. It can therefore be helpful
to look closer at the dimers as a group to understand the role
5751
https://doi.org/10.1021/jacs.1c00331
J. Am. Chem. Soc. 2021, 143, 5745−5754

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
of the intra-UC mechanism. Comparing the experimental
mean performance of the dimers to the two proposed models
then indicates which model explains the UC performance of
the dimers as a group. We have previously established that the
annihilator triplet lifetime is a crucial parameter which dictates
much of the UC performance, but it cannot on its own explain
the observed differences in UC efficiencies. In fact, the DS and
TETA models predict slightly different UC efficiencies at
different conditions, with the former complying better with our
experimental findings at low [1A]0/[1S]0 ratios but the latter
agreeing more with experiments at high [1A]0/[1S]0 ratios
(Figure 8). Even though this transition from conditions under
multichromophoric annihilators are hampered. These implica-
tions are strengthened by the results from the study by Dzebo
et al., where large DPA annihilator frameworks showed better
UC performance than DPA when put inside a rigid polymer
matrix.23 First, the performance of these compounds improved
with molecular (dendrimeric/oligomeric) size and even
outperformed a reference sample consisting of the DPA
monomer, despite the fact that the molecular concentration of
DPA was almost 10 times higher in the reference sample than
in the oligomeric frameworks. Second, it was found that the
best performance was achieved when the sensitizer concen-
tration was on the same order as that of the annihilator.
Together, this forms unequivocal proof that it is the DS
mechanism that is active in such systems and that intra-UC in
fact may enhance UC performance in solid-state systems.
Major challenges in facilitating the energy transfer events that
are needed remain, particularly how to precisely control the
spatial association and interactions between the sensitizer and
annihilator. Efforts to understand and control the complicated
assembly of high-performing solid-state UC systems are
currently undergoing in our lab.
CONCLUSIONS
■
In this study, four novel dimers based on DPA have been
synthesized and evaluated as annihilators for TTA-UC. The
upconversion efficiencies of these molecules are high
throughout, with the meta-coupled 1,3-DPA2 in particular
performing on par with DPA. This is primarily ascribed to the
long triplet lifetime of this dimer, which is at least five times
longer than that of the other dimers and similar to the triplet
lifetime of DPA. The importance of the triplet lifetime to
achieve efficient TTA-UC is thus reinforced, and a facile
method to accurately measure this important parameter has
been utilized. Furthermore, the mechanism of intramolecular
upconversion has been investigated by comparison between
simulations and experiments. Our results from time-resolved
UC emission measurements firmly ascertain that the DS
mechanism is active in systems with dimeric annihilators and
that this additional pathway could be of great importance
under specific conditions, e.g., at extremely low annihilator to
sensitizer concentration ratios or in diffusionally restricted
media. While the full picture of intra-UC is still lacking,
especially in terms of how efficiently intra-UC may proceed,
our results may act as a guide for future solid-state designs
where exciton migration is envisioned as a crucial component.
Figure 8. Comparison between the UC emission of DPA and dimers
as a group. The abscissa signifies the ratio between ground-state
annihilator and sensitizer concentrations, and the ordinate the mean
relative UC emission of the dimers compared to that of DPA. Purple
circles are the experimental mean values of dimer UC emission
divided by that of DPA and are compared with the corresponding
intensities predicted by the DS (black solid line) and TETA (blue
solid line) models. The red dotted line represents the UC emission of
DPA and is included as a reference point.
which the DS mechanism dominates to one where TETA
dominates is not unambiguous, this change is expected given
that the intra-UC mechanism in the DS and TETA models
depends on [3S*] and [3A*-1A], respectively, which are high at
each extreme of Figure 8. The correspondence between
simulation and experiment is however not without fault, and
the relatively low UC efficiencies of the dimers compared to
DPA at high [1A]0/[1S]0 ratios (rightmost point in Figure 8)
could possibly be due to differences in the intramolecular spin-
statistical factor. This factor has not been included in our
models but has been suggested to be rather low in compounds
similar to ours.31 Moreover, the TETA mechanism which
dictates intra-UC at high [1A]0/[1S]0 ratios will in fact be
competing with the ubiquitous inter-UC pathway. If the
efficiency of the intra-UC is lower than that of inter-UC, this
would then be detrimental to UC performance, as indicated by
our results.
While these results are interesting on their own, one must
consider in what settings intra-UC is anticipated to be of
importance. As solid-state solutions are a desirable path
moving forward, it is in the context of restricted molecular
diffusion these results must be interpreted.19,56−61 With respect
to this, it is improbable that the TETA mechanism will be
active in future solid-state systems where interactions between
ASSOCIATED CONTENT
■
sı
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/jacs.1c00331.
Detailed description of the synthesis of novel com-
pounds, proton and carbon NMR spectra, descriptions
of experimental setups, additional spectroscopic and
modeling data, full model descriptions, and full details
on the temperature-dependent photophysics of 1,2-
DPA2 (PDF)
AUTHOR INFORMATION
■
Corresponding Author
Bo Albinsson − Department of Chemistry and Chemical
Engineering, Chalmers University of Technology, 412 96
5752
https://doi.org/10.1021/jacs.1c00331
J. Am. Chem. Soc. 2021, 143, 5745−5754

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(14) Awwad, N.; Bui, A. T.; Danilov, E. O.; Castellano, F. N. Visible-
Light-Initiated Free-Radical Polymerization by Homomolecular
Triplet-Triplet Annihilation. Chem. 2020, 6, 3071−3085.
(15) Börjesson, K.; Dzebo, D.; Albinsson, B.; Moth-Poulsen, K.
Photon upconversion facilitated molecular solar energy storage. J.
Mater. Chem. A 2013, 1, 8521−8524.
Gothenburg, Sweden; orcid.org/0000-0002-5991-7863;
Email: balb@chalmers.se
Authors
Axel Olesund − Department of Chemistry and Chemical
Engineering, Chalmers University of Technology, 412 96
Gothenburg, Sweden; orcid.org/0000-0003-1202-7844
Victor Gray − Department of Chemistry and Chemical
Engineering, Chalmers University of Technology, 412 96
Gothenburg, Sweden; Department of Chemistry, Ångström
Laboratory, Uppsala University, 751 20 Uppsala, Sweden;
orcid.org/0000-0001-6583-8654
(16) Khnayzer, R. S.; et al. Upconversion-powered photoelec-
trochemistry. Chem. Commun. 2012, 48, 209−211.
(17) Dexter, D. L. A theory of sensitized luminescence in solids. J.
Chem. Phys. 1953, 21, 836−850.
(18) Hoseinkhani, S.; Tubino, R.; Meinardi, F.; Monguzzi, A.
Achieving the photon up-conversion thermodynamic yield upper limit
by sensitized triplet−triplet annihilation. Phys. Chem. Chem. Phys.
2015, 17, 4020−4024.
(19) Gray, V.; Moth-Poulsen, K.; Albinsson, B.; Abrahamsson, M.
Towards efficient solid-state triplet-triplet annihilation based photon
upconversion: Supramolecular, macromolecular and self-assembled
systems. Coord. Chem. Rev. 2018, 362, 54−71.
(20) Lee, S. H.; Ayer, M. A.; Vadrucci, R.; Weder, C.; Simon, Y. C.
Light upconversion by triplet-triplet annihilation in diphenylanthra-
cene-based copolymers. Polym. Chem. 2014, 5, 6898−6904.
(21) Sittig, M.; et al. Fluorescence upconversion by triplet-triplet
annihilation in all-organic poly(methacrylate)-terpolymers. Phys.
Chem. Chem. Phys. 2020, 22, 4072−4079.
(22) Imperiale, C. J.; Green, P. B.; Miller, E. G.; Damrauer, N. H.;
Wilson, M. W. B. Triplet-Fusion Upconversion Using a Rigid
Tetracene Homodimer. J. Phys. Chem. Lett. 2019, 10, 7463−7469.
(23) Dzebo, D.; Börjesson, K.; Gray, V.; Moth-Poulsen, K.;
Albinsson, B. Intramolecular Triplet−Triplet Annihilation Upconver-
sion in 9,10-Diphenylanthracene Oligomers and Dendrimers. J. Phys.
Chem. C 2016, 120, 23397−23406.
(24) Gray, V.; et al. Porphyrin-Anthracene Complexes: Potential in
Triplet-Triplet Annihilation Upconversion. J. Phys. Chem. C 2016,
120, 19018−19026.
(25) Gao, C.; et al. Tetraphenylethene 9,10-Diphenylanthracene
Derivatives − Synthesis and Photophysical Properties. ChemPlusChem
2019, 84, 746−753.
̊
Jerker Martensson − Department of Chemistry and Chemical
Engineering, Chalmers University of Technology, 412 96
Gothenburg, Sweden
Complete contact information is available at:
https://pubs.acs.org/10.1021/jacs.1c00331
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Professor Kasper Moth-Poulsen and M. Sc. Fredrik Edhborg
are gratefully acknowledged for valuable discussions. We are
grateful to The Swedish Energy Agency (contracts 46526-1
and 36436-2) for providing financial support.
REFERENCES
■
(1) Shockley, W.; Queisser, H. J. Detailed Balance Limit of
Efficiency of p-n Junction Solar Cells. J. Appl. Phys. 1961, 32, 510.
(2) Smith, M. B.; Michl, J. Recent Advances in Singlet Fission. Annu.
Rev. Phys. Chem. 2013, 64, 361−386.
(3) Gray, V.; Dzebo, D.; Abrahamsson, M.; Albinsson, B.; Moth-
Poulsen, K. Triplet−triplet annihilation photon-upconversion:
towards solar energy applications. Phys. Chem. Chem. Phys. 2014,
16, 10345−10352.
(26) Boutin, P. C.; Ghiggino, K. P.; Kelly, T. L.; Steer, R. P. Photon
upconversion by triplet-triplet annihilation in Ru(bpy)3- and DPA-
functionalized polymers. J. Phys. Chem. Lett. 2013, 4, 4113−4118.
(27) Lee, S. H.; Thévenaz, D. C.; Weder, C.; Simon, Y. C. Glassy
poly(methacrylate) terpolymers with covalently attached emitters and
sensitizers for low-power light upconversion. J. Polym. Sci., Part A:
Polym. Chem. 2015, 53, 1629−1639.
(28) Tilley, A. J.; Robotham, B. E.; Steer, R. P.; Ghiggino, K. P.
Sensitized non-coherent photon upconversion by intramolecular
triplet-triplet annihilation in a diphenylanthracene pendant polymer.
Chem. Phys. Lett. 2015, 618, 198−202.
(4) Singh-Rachford, T. N.; Castellano, F. N. Photon upconversion
based on sensitized triplet-triplet annihilation. Coord. Chem. Rev.
2010, 254, 2560−2573.
(5) Gholizadeh, E. M.; et al. Photochemical upconversion of near-
infrared light from below the silicon bandgap. Nat. Photonics 2020, 14,
585−590.
(6) Albinsson, B.; Olesund, A. Untapping solar energy resources.
Nat. Photonics 2020, 14, 528−530.
(29) Matsui, Y.; Kanoh, M.; Ohta, E.; Ogaki, T.; Ikeda, H. Triplet−
Triplet Annihilation-Photon Upconversion Employing an Adaman-
tane-linked Diphenylanthracene Dyad Strategy. J. Photochem. Photo-
biol., A 2020, 387, 112107.
(7) Dilbeck, T.; Hanson, K. Molecular Photon Upconversion Solar
Cells Using Multilayer Assemblies: Progress and Prospects. J. Phys.
Chem. Lett. 2018, 9, 5810−5821.
(8) Cheng, Y. Y.; et al. Improving the light-harvesting of amorphous
silicon solar cells with photochemical upconversion. Energy Environ.
Sci. 2012, 5, 6953−6959.
(30) Simon, Y. C.; Weder, C. Low-power photon upconversion
through triplet-triplet annihilation in polymers. J. Mater. Chem. 2012,
22, 20817−20830.
(9) Monguzzi, A.; et al. Broadband up-conversion at subsolar
irradiance: Triplet-triplet annihilation boosted by fluorescent semi-
conductor nanocrystals. Nano Lett. 2014, 14, 6644−6650.
(10) Hill, S. P.; Dilbeck, T.; Baduell, E.; Hanson, K. Integrated
Photon Upconversion Solar Cell via Molecular Self-Assembled
Bilayers. ACS Energy Lett. 2016, 1, 3−8.
(31) Gao, C.; et al. Intramolecular versus Intermolecular Triplet
Fusion in Multichromophoric Photochemical Upconversion. J. Phys.
Chem. C 2019, 123, 20181−20187.
(32) Edhborg, F.; Kucu̧ köz, B.; Gray, V.; Albinsson, B. Singlet
̈ ̈
Energy Transfer in Anthracene−Porphyrin Complexes: Mechanism,
Geometry, and Implications for Intramolecular Photon Upconversion.
J. Phys. Chem. B 2019, 123, 9934−9943.
(11) Huang, L.; et al. Highly Effective Near-Infrared Activating
Triplet−Triplet Annihilation Upconversion for Photoredox Catalysis.
J. Am. Chem. Soc. 2020, 142, 18460−18470.
(12) Yu, T.; et al. Triplet−Triplet Annihilation Upconversion for
Photocatalytic Hydrogen Evolution. Chem. - Eur. J. 2019, 25, 16270−
16276.
(33) Pun, A. B.; Sanders, S. N.; Sfeir, M. Y.; Campos, L. M.;
Congreve, D. N. Annihilator dimers enhance triplet fusion
upconversion. Chem. Sci. 2019, 10, 3969−3975.
(34) Papadopoulos, I.; et al. Perylene-Monoimides: Singlet Fission
Down-Conversion Competes with Up-Conversion by Geminate
Triplet-Triplet Recombination. J. Phys. Chem. A 2020, 124, 5727−
5736.
(13) Ravetz, B. D.; et al. Photoredox catalysis using infrared light via
triplet fusion upconversion. Nature 2019, 565, 343−346.
5753
https://doi.org/10.1021/jacs.1c00331
J. Am. Chem. Soc. 2021, 143, 5745−5754

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(35) Chen, M.; et al. Quintet-triplet mixing determines the fate of
the multiexciton state produced by singlet fission in a terrylenediimide
dimer at room temperature. Proc. Natl. Acad. Sci. U. S. A. 2019, 116,
8178−8183.
(57) Gharaati, S.; et al. Triplet−Triplet Annihilation Upconversion
in a MOF with Acceptor-Filled Channels. Chem. - Eur. J. 2020, 26,
1003−1007.
(58) Barbosa de Mattos, D. F.; et al. Covalent incorporation of
diphenylanthracene in oxotriphenylhexanoate organogels as a quasi-
solid photon upconversion matrix. J. Chem. Phys. 2020, 153, 214705.
(59) Sakamoto, Y.; Tamai, Y.; Ohkita, H. Sensitizer-host-annihilator
ternary-cascaded triplet energy landscape for efficient photon
upconversion in the solid state. J. Chem. Phys. 2020, 153, 161102.
(60) Saenz, F.; et al. Nanostructured Polymers Enable Stable and
Efficient Low-Power Photon Upconversion. Adv. Funct. Mater. 2021,
31, 2004495.
(36) Tayebjee, M. J. Y.; et al. Quintet multiexciton dynamics in
singlet fission. Nat. Phys. 2017, 13, 182−188.
(37) Basel, B. S.; et al. Unified model for singlet fission within a non-
conjugated covalent pentacene dimer. Nat. Commun. 2017, 8, 1−8.
(38) Korovina, N. V.; et al. Singlet Fission in a Covalently Linked
Cofacial Alkynyltetracene Dimer. J. Am. Chem. Soc. 2016, 138, 617−
627.
(39) Sanders, S. N.; et al. Quantitative Intramolecular Singlet Fission
in Bipentacenes. J. Am. Chem. Soc. 2015, 137, 8965−8972.
(40) Lukman, S.; et al. Tuneable Singlet Exciton Fission and Triplet-
Triplet Annihilation in an Orthogonal Pentacene Dimer. Adv. Funct.
Mater. 2015, 25, 5452−5461.
(61) Perego, J.; et al. Engineering Porous Emitting Framework
Nanoparticles with Integrated Sensitizers for Low-Power Photon
Upconversion by Triplet Fusion. Adv. Mater. 2019, 31, 1903309.
(41) Bansal, A. K.; Holzer, W.; Penzkofer, A.; Tsuboi, T. Absorption
and emission spectroscopic characterization of platinum-octaethyl-
porphyrin (PtOEP). Chem. Phys. 2006, 330, 118−129.
(42) Benniston, A. C.; Harriman, A.; Howell, S. L.; Sams, C. A.; Zhi,
Y. G. Intramolecular excimer formation and delayed fluorescence in
sterically constrained pyrene dimers. Chem. - Eur. J. 2007, 13, 4665−
4674.
(43) Zhao, W.; Castellano, F. N. Upconverted Emission from Pyrene
and Di-tert-butylpyrene Using Ir(ppy) 3 as Triplet Sensitizer. J. Phys.
Chem. A 2006, 110, 11440−11445.
̊
(44) Ye, C.; Gray, V.; Martensson, J.; Börjesson, K. Annihilation
Versus Excimer Formation by the Triplet Pair in Triplet-Triplet
Annihilation Photon Upconversion. J. Am. Chem. Soc. 2019, 141,
9578−9584.
(45) Chandross, E. A.; Dempster, C. J. Dempster. Intramolecular
Excimer Formation and Fluorescence Quenching in Dinaphthylal-
kanes. J. Am. Chem. Soc. 1970, 92, 3586−3593.
(46) Nishiuchi, T.; Uno, S.; Hirao, Y.; Kubo, T. Intramolecular
Interaction, Photoisomerization, and Mechanical C−C Bond
Dissociation of 1,2-Di(9-anthryl)benzene and Its Photoisomer: A
Fundamental Moiety of Anthracene-Based π-Cluster Molecules. J.
Org. Chem. 2016, 81, 2106−2112.
(47) Zhou, Y.; Castellano, F. N.; Schmidt, T. W.; Hanson, K. On the
Quantum Yield of Photon Upconversion via Triplet−Triplet
Annihilation. ACS Energy Lett. 2020, 5, 2322−2326.
(48) Kubin, R. F.; Fletcher, A. N. Fluorescence quantum yields of
some rhodamine dyes. J. Lumin. 1982, 27, 455−462.
(49) Lakowicz, J. R. Principles of Fluorescence Spectroscopy; Springer:
New York, 2006.
(50) Chattopadhyay, S. K.; Kumar, C. V.; Das, P. K. Triplet-related
photophysics of 9,10-diphenylanthracene. A kinetic study of reversible
energy transfer from anthracene triplet by nanosecond laser flash
photolysis. Chem. Phys. Lett. 1983, 98, 250−254.
(51) Gray, V.; et al. Photophysical characterization of the 9,10-
disubstituted anthracene chromophore and its applications in triplet-
triplet annihilation photon upconversion. J. Mater. Chem. C 2015, 3,
11111−11121.
(52) Kondakov, D. Y.; Pawlik, T. D.; Hatwar, T. K.; Spindler, J. P.
Triplet annihilation exceeding spin statistical limit in highly efficient
fluorescent organic light-emitting diodes. J. Appl. Phys. 2009, 106,
124510.
(53) Arroyo, C. R.; et al. Quantum interference effects at room
temperature in OPV-based single-molecule junctions. Nanoscale Res.
Lett. 2013, 8, 234.
(54) Solomon, G. C.; et al. Understanding quantum interference in
coherent molecular conduction. J. Chem. Phys. 2008, 129, 054701.
(55) Gantenbein, M.; et al. Quantum interference and hetero-
aromaticity of para-and meta-linked bridged biphenyl units in single
molecular conductance measurements. Sci. Rep. 2017, 7, 1−9.
(56) Vadrucci, R.; et al. Nanodroplet-Containing Polymers for
Efficient Low-Power Light Upconversion. Adv. Mater. 2017, 29,
1702992.
5754
https://doi.org/10.1021/jacs.1c00331
J. Am. Chem. Soc. 2021, 143, 5745−5754