BODIPY based A-D-A molecules: Effect of CF3 group substitution at meso phenyl group

Article abstract of DOI:10.1016/j.dyepig.2020.108289

Two pairs of A-D-A molecules have been synthesized with fluorene and benzodithiophene as the central donor subunits and terminal BODIPY units, functionalized with either a 4-methylphenyl or 4-trifluoromethylphenyl group at the meso position. The effect of the para substituent of the meso phenyl group on the photophysical properties of these molecules is studied through steady state absorption and fluorescence spectroscopy as well as femtosecond transient absorption and time resolved fluorescence spectroscopy techniques. Applicability of these molecules as donors in solution processed solar cell active layers was investigated through time resolved microwave conductivity measurements on blends with PC₆₀BM acceptor, which shows a varying yield of charge transfer with choice of substituent. Transient absorption spectroscopy is then employed to investigate the role of the 4-trifluoromethylphenyl group in altering the efficiency of charge transfer from these A-D-A molecules to PC₆₀BM. The results show a consistent picture of picosecond charge transfer and a component of a few hundred ps geminate recombination that results in a small yield of long-lived free charges optimized for the methylphenyl derivatives.

Full text of DOI:10.1016/j.dyepig.2020.108289

Journal Pre-proof
BODIPY based A-D-A molecules: Effect of CF group substitution at meso phenyl
3
group
Gourav Tarafdar, Justin C. Johnson, Bryon W. Larson, Praveen C. Ramamurthy
PII:
S0143-7208(19)32571-9
DOI:
https://doi.org/10.1016/j.dyepig.2020.108289
DYPI 108289
Reference:
To appear in: Dyes and Pigments
Received Date: 2 November 2019
Revised Date: 14 February 2020
Accepted Date: 15 February 2020
Please cite this article as: Tarafdar G, Johnson JC, Larson BW, Ramamurthy PC, BODIPY based A-D-
A molecules: Effect of CF group substitution at meso phenyl group, Dyes and Pigments (2020), doi:
3
https://doi.org/10.1016/j.dyepig.2020.108289.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
© 2020 Published by Elsevier Ltd.

BODIPY based A-D-A molecules: Effect of CF3 group substitution at meso
phenyl group
Gourav Tarafdar1, Justin C. Johnson2*, Bryon W. Larson2, Praveen C Ramamurthy1*
1. Interdisciplinary Centre for Energy Research, Indian Institute of Science, Bangalore 560012
2. National Renewable Energy Laboratory, Golden CO, USA 80401
Email: justin.johnson@nrel.gov; praveen@iisc.ac.in
Abstract: Two pairs of A-D-A molecules have been synthesized with fluorene and benzodithiophene
as the central donor subunits and terminal BODIPY units, functionalized with either a 4-methylphenyl
or 4-trifluoromethylphenyl group at the meso position. The effect of the para substituent of the meso
phenyl group on the photophysical properties of these molecules is studied through steady state
absorption and fluorescence spectroscopy as well as femtosecond transient absorption and time
resolved fluorescence spectroscopy techniques. Applicability of these molecules as donors in solution
processed solar cell active layers was investigated through time resolved microwave conductivity
measurements on blends with PC60BM acceptor, which shows a varying yield of charge transfer with
choice of substituent. Transient absorption spectroscopy is then employed to investigate the role of the
4-trifluoromethylphenyl group in altering the efficiency of charge transfer from these A-D-A
molecules to PC60BM. The results show a consistent picture of picosecond charge transfer and a
component of a few hundred ps geminate recombination that results in a small yield of long-lived free
charges optimized for the methylphenyl derivatives.
1. Introduction
4,4-Difluoro-4-bora-3a,4a-diaza-s-indacene or boron-dipyrromethene (commonly abbreviated
as BODIPY) dyes have good π electron delocalization across the core, exhibit high molar
extinction coefficient, and offer good photostability[1,2]. These properties, coupled with the
ease of changing substituents at different positions of the core, make BODIPY a widely
explored building block for organic photovoltaic (OPV)[3–5] . BODIPY is an electron
deficient species and can act as acceptor subunits in D-A type architectures. Both small
molecules and polymers, with BODIPY incorporated in their backbone, have been used as
both donors[6,7] and acceptors [8,9] in organic solar cells.

One way of tuning the properties and functionality of BODIPY dyes, and hence BODIPY
conjugates, is by changing the substituents at the meso position [10–12]. Trifluoromethyl
groups (CF3), for example, have been shown to affect the energy levels of BODIPY
molecules[13] and lead to stronger intermolecular interactions.[14] Trifluoromethyl
substituted BODIPY has also been reported for solar cell applications.[15] However, to our
best knowledge, the effect of trifluoromethyl substitution at the para position of the meso
phenyl substituent, on the photophysical properties of BODIPY based OPV materials has not
been studied yet.
To investigate the effect of CF3 on the photophysical, and in particular charge transfer,
properties of BODIPY based OPV materials, we have designed two pairs of A-D-A
molecules with terminal BODIPY units (Figure 1). Within each pair, the chemical structure
differentiation is located at the meso substituent of the BODIPY, where X is either a methyl
or trifluoromethyl group. The central donor core subunits are chosen to be fluorene in the first
pair of molecules (S1 and S2) and benzodithiophene (BDT) in the second pair (S3 and S4).
Figure 1: Structures of the A-D-A molecules
2. Experimental section
2.1. Synthesis
Starting with 4-methylbenzaldehyde or 4-trifluoromethylbenzaldehyde, two similar BODIPY
dyes MeBDP (3a) and CF3BDP (3b) were synthesized and subsequently brominated to obtain
monobrominated BODIPY units following previously reported synthetic routes [16,17]
(Scheme 1). These were coupled with diboronic esters of dihexylfluorene (11) and
distannylated benzodithiophene (10) through Suzuki and Stille coupling respectively

(Scheme 2), to obtain the A-D-A molecules (S1-S4). 11 was purchased from Sigma Aldrich
and 12 was synthesized following literature reports [18]. The molecules were purified
through column chromatography and characterized using 1H & 13C NMR spectroscopy.
Scheme 1: Synthesis of monomers (i) Pyrrole, TFA, RT, 15 mins (ii) DDQ; TEA, BF3.Et2O (iii) N-
bromosuccinimide (iv) (COCl)2 (v) Et2NH (vi) n-BuLi, –78°C (vii) n-BuLi, –78°C; 2-Ethylhexyl
bromide (viii) n-BuLi, –78°C; SnCl2/HCl (ix) n-BuLi, –78°C; SnMe3Cl.

Scheme 2: Coupling reactions (i) Pd2dba3, P(o-tolyl)3, K2CO3, toluene, reflux, 6 h (ii) Pd2dba3, P(o-
tolyl)3, toluene, reflux, 6 h.
3. Results and Discussion
3.1. Optical characterization:
The absorption and emission spectra of the molecules were recorded in chloroform solution
(Figure 2). The absorption spectra of the A-D-A triads do not retain the characteristic
absorption peaks of the monomer units, thereby confirming strong electronic coupling among
the chromophores. The molecules show broad intramolecular charge transfer bands around
600 nm with maxima at 585 nm, 600 nm, 625 nm and 640 nm for S1, S2, S3 and S4
respectively. The absorption spectra for S1-S4 show a trend of narrowing band gaps
(evidenced by gradually red-shifted absorbance onsets), consistent with our predicted values
from DFT calculations, vide infra. The steady state photoluminescence emission maxima
(Figure 2b) for S1, S2 and S3 were recorded at 670 nm, 685 nm and 667 nm, respectively.
The CF3-substituted molecules exhibit weaker emission than their CH3-substitued analogues,
and S4 is practically non-emissive. A large stokes shift (~ 85 nm) is observed in the fluorene-
core molecules S1 and S2 as compared to only ~ 27nm for BDT-core molecule S3. This
suggests that unlike the BODIPY-BDT molecules (S3 and S4), the excited state geometry is
very different from the ground state geometry in the case of the BODIPY-fluorene molecules
(S1 and S2).

Figure 2: (a) Absorption spectra of the monomers (in dotted lines) and A-D-A compounds (solid
lines) in chloroform. (b) Emission spectra of the compounds upon excitation at the S1 absorption
maxima.
3.2. Theoretical Calculations:
DFT calculations were performed using Gaussian09 software at the B3LYP level with 6-
31g(d) basis sets to predict the optimized molecular geometry, HOMO and LUMO energies
and band gaps for the A-D-A molecules. Unlike the fluorene molecules (S1 and S2), the
benzodithiophene molecules (S3 and S4) have a more planar structure with the dihedral
angles between the BODIPY and BDT units close to 175°. The optimized geometry of the
cations of the molecules were also obtained using similar calculations. The cations of the
fluorene molecules (S1 and S2) show a greater degree of reorganization of the geometry (ESI
Table S5). While the dihedral angles between the BODIPY and fluorene units are ~ 157° in
the neutral molecules, in case of the cation the dihedral angle is ~ 180° for S1 and S2. In case
of S3 and S4, the dihedral angle between the BODIPY and BDT units remain unaffected
when the molecules carry a positive charge. However, the dihedral angle between the
backbone and the thiophene rings on the BDT changes from ~ 122° to ~ 132°.The calculated
HOMO and LUMO energies are tabulated in Table 1 Lower band gap is predicted for the
trifluoromethylated molecules (S2 and S4) than their methylated counterparts (S1 and S3
respectively). Furthermore, the more planar structure for the BDT molecules (S3 and S4)
leads to better electron delocalization and hence narrower band gaps compared to fluorene
molecules (S1 and S2). These predictions are borne out in red-shifted bands for S2 and S3
(Figure 2) that lie well below the absorptions of the constituent BODIPY of BDT units.

TD-DFT calculations were performed for the molecules to obtain the simulated absorption
spectra of the molecules (ESI Fig S11-S14). The calculated spectra match the experimental
data more closely for the fluorene cored molecules (S1 and S2) compared to the BDT
molecules (S3 and S4). Nevertheless, the TD-DFT results suggest that the intense broad
absorption bands in the 600 nm - 700 nm region indeed arise due to intramolecular charge
transfer from the core donor subunit to the terminal acceptor BODIPY units.
Figure 3: Electron distribution in the frontier molecular orbitals of the A-D-A molecules optimized
using DFT calculations.
3.3. Cyclic voltammetry:
The HOMO and the LUMO energy levels of the molecules were calculated using cyclic
voltammetry measurements. The experiments were conducted using a Gamry Interface 1000
potentiostat. A 0.1 M solution of TBAP (tetrabutylammonium perchlorate) in o-
dichlorobenzene: acetonitrile (4:1) was taken as the supporting electrolyte and ferrocene was
chosen as the external standard. The electrochemical cell comprised of a glassy carbon
working electrode, Pt-wire counter electrode and Ag-wire as pseudo-reference electrode. The
spectra were recorded at a scan rate of 50 mV/s and under N2 environment. The HOMO and
the LUMO were calculated from the E1/2 of reversible oxidation and the reduction peaks
respectively using the following relations.[19,20]
HOMO = – [4.8 + (Eox – E1/2Fc)] eV; LUMO = – [4.8 + (Ered – E1/2Fc )] eV

Fc
where E1/2 is the measured potential corresponding to the reversible oxidation peak of
ferrocene. S1 and S2 show irreversible oxidation peaks (Figure 4). The HOMO values for S1
and S2 are therefore calculated by deducting the optical band gap values from the LUMO
values calculated from electrochemical measurements. As predicted from the DFT
calculations (shown in Table 1), the LUMO of the molecules with a trifluoromethyl
substituent (S2 and S4) are more stabilized (– 3.9 eV) than those with a methyl substituent,
S1 and S3 (~ – 3.7 eV). However, the relation used above to obtain the energy levels values
from the cyclic voltammograms is empirical and does not account for solvation energy of the
molecules. Hence the values obtained are qualitative. For a better comparison, the LUMO of
PC60BM was also measured from CV under similar conditions and calculated to be –3.71 eV.
Figure 5 shows the energy offset of the A-D-A molecules with PC60BM. The HOMO value
for PC60BM taken as –6.1 eV from literature.[21]
Table 1: Energy levels calculated from CV measurements and DFT calculations. * The
HOMO for S1 and S2 were calculated by subtracting the optical band gap from LUMO obtained from
electrochemical measurements.
HOMO
(DFT)
eV
LUMO
(DFT)
eV
Eg
HOMO
LUMO
(from CV)
eV
Eg
Eg
(predicted) (from CV)
(measured) (optical)
eV
eV
eV
—
eV
−5.21
−5.41
−4.97
−5.14
−2.79
−3.08
−2.87
−3.14
2.42
2.33
2.10
2.00
−5.33*
−5.39*
−5.36
−5.31
−3.52
1.81
1.74
1.65
1.56
S1
S2
S3
S4
−3.65
—
−3.66
1.7
1.6
−3.71

Figure 4: Cyclic voltammograms for the A-D-A molecules and the band edge matching with
PC60BM. The data has been plotted by setting E1/2(Fc/Fc+) as 0.
3.4. Time resolved photoluminescence and absorption
Time resolved photoluminescence (TRPL) experiments were carried out on the A-D-A
molecules in chloroform solution. The decay profile was fitted with a single exponential
convoluted with the instrument response function to obtain the emission lifetimes for the
molecules. The fluorescence lifetimes for the BDT-based molecules (S3 and S4) are shorter
than that of the fluorene-based molecules (S1 and S2), which is consistent with the weaker
emission from the BDT molecules observed from steady state fluorescence studies. Among
each pair, the molecules with the CF3 substituent (S2 and S4) show shorter lifetimes for the
excited state than their counterparts with CH3 substituent (S1 and S3). We attribute this
observation to more efficient intramolecular charge transfer in the CF3-substituted analogues.

Figure 5: Time resolved photoluminescence spectra of the A-D-A molecules in chloroform solution
integrated over the range of 590-870 nm for S1 & S2 and over the range of 640-920 nm for S3 & S4.
Table 2: Lifetimes obtained for the TA decay associated spectra and time-resolved PL decay
for the A-D-A molecules in solution
τ1 (fs)
τ2 (ps)
τ3 (ps)
τPL (ps)
692
183 ± 140
345 ± 120
578 ± 190
570 ± 130
5.9 ± 1.1
5.8 ± 1.0
8.3 ± 4.8
11.6 ± 6.3
685.7 ± 22
264.0 ± 8.3
106.3 ± 9.0
40.6 ± 6.3
S1
S2
S3
S4
283
128
57
The excited state dynamics of the A-D-A compounds were studied using femtosecond
transient absorption (TA) spectroscopy experiments. Global fitting yielded three exponential
decay components for each of the four compounds (shown in ESI Figure S40). The spectra
with the longest lifetime (τ3) most likely represent the radiative decay pathway and match
with the photoluminescence lifetimes obtained from TRPL measurements. (Table 2) The
shorter time components can be attributed to Frank-Condon and solvent-related
*
reorganization from S1 to S1, which is typical of conjugated molecules that have excited
states with charge-transfer character.

3.5. Time resolved microwave conductivity (TRMC)
To investigate the effect of the trifluoromethyl group on photoinduced charge transfer in the
A-D-A donor molecules, PC60BM was chosen as a bulk heterojunction acceptor based on
favourable alignment of energy levels (Figure 5). Blended thin films of these A-D-A
molecules with PC60BM were prepared from solution in chlorobenzene solvent at a 1:1
weight ratio by spin coating onto 25x11x1 mm quartz plates for TRMC analysis. TRMC is a
contactless pulsed pump-continuous microwave probe technique that detects generation and
local mobility of free charge carriers in photoactive materials. A full description of the
technique and the instrument specifications at NREL can be found elsewhere [22]. Here, we
discuss the yield-mobility product (φΣµ), which is the main figure of merit for TRMC
measurements, defined as the product of the quantum efficiency of photogeneration of free
charge carriers and the sum of the electron and hole mobilities in the active layer.
A higher φΣµ value is obtained for the CH3-substituted molecules compared to the ones with
X=CF3 (Figure 6), yet there is no trend between the BDT-core pair and the fluorene-core pair
when exciting the films at 640 nm. The S3:PC60BM blend shows the largest values for φΣµ,
more than a factor of 3 larger than the other blends. However, in all of the blends the peak
φΣµ values are quite low, which is likely indicative of a low yield of long-lived (greater than
the 5 ns laser pulse width) mobile free charges, a supposition that is further corroborated by
the transient absorption data discussed below. The photoconductivity transients recorded over
500 ns (Figure 6) were fitted with a biexponential function convolved with the instrument
response to extract the average weighted free carrier lifetime for each blend (shown Table 3).

Figure 6: Decay of TRMC signal for blends of the different A-D-A molecules with PC60BM.
Table 3: Summary of TRMC data for S1-S4 blends with PC60BM.
A1
τ1
A2
τ2
τ
φΣµ
Absorbed
flux
S1:PC60BM
S2:PC60BM
S3:PC60BM
S4:PC60BM
3.97793
5.12982
4.7201
2.87E-08 3.85781
1.34E-08 11.0415
2.35E-08 1.85348
1.12E-08 3.2897
8.76E-08 5.77E-08 1.56E-05 1.67E+14
9.13E-08 6.66E-08 3.3E-05 1.66E+14
1.52E-07 5.98E-08 1.31E-04 1.37E+14
9.03E-07 1.22E-08 1.55E-05 1.22E+14
2790.19
3.6.Transient absorption spectra of blends
Femtosecond transient absorption spectroscopy experiments were also conducted on blends
of the A-D-A molecules with PC60BM, to gain insights into the nature of photoinduced
absorption species at very early timescales. Neat as well as blend films were prepared on
sapphire substrates by spin coating from chlorobenzene solution in the glove box and
subsequently sealed under inert atmosphere. The films for the fluorene molecules (S1 and S2)
were pumped at 600 nm and the BDT molecules (S3 and S4) at 700 nm. The films were
probed in the visible region (400 – 850 nm) and the NIR region (850 -1600 nm) separately.
The transient absorption spectra of the neat and blend films of S1-S4 are shown in Supporting
Information (Figure S39).

The positive peaks at 500 nm in S1 and S2 neat films can be assigned to the singlet excited
state absorption and appear at a timescale shorter than 1 ps. The singlet excited state has a
shorter lifetime in S2 than in S1 which is consistent with the observations from solution-
phase TRPL measurements. The positive peaks formed at early times near 1000 nm and also
at 1200-1500 nm can be assigned to the excited singlet state absorption, which may contain a
mixture of excitonic and polaronic character. In the case of S3 and S4, a broad negative peak
from 600 nm to 800 nm corresponds to the ground state bleach overlapping with the
stimulated emission. Two positive peaks observed at 480 nm and 590 nm are due to excited
state absorption. The decay of these peaks is much faster in the case of S4 compared to S3.
The broad positive peaks in the NIR region due to excitons and polarons also have faster
decay in the case of S4 compared to S3 and were fitted with single exponents (ESI Table S6).
In blends with PC60BM, S1-S3 show slightly longer lifetimes for the singlet excited states
compared to neat films, which is discussed below. Differences are also observed in the
evolution of the absorption for the blends of S1, S2 and S3 with PC60BM compared to their
neat films. However, the spectrum for S4:PC60BM film resembles closely that of the neat S4
film. While the lifetime of the singlet excited state is the same (~ 40 ps) in both the neat and
blend for S4, the lifetime of the polarons measured at 1328 nm is slightly shortened in case of
the blend. This suggests formation of very few free charge carriers from delocalized polaron
pairs (DPP). The same was also concluded from the low yield mobility product of
S4:PC60BM from TRMC experiments.
In the case of S1, S2 and S3, the ground state bleach recovery is slower with the
PC60BM blends as compared to neat films. Furthermore, growth of a positive peak is
observed at 1030 nm (within 1 ps delay) for the blends with S1, S2 and S3 (c.f., the spectral
evolution occurring in Figure 7b but not in Figure 7a). This can be attributed to formation of
PC60BM anion radical, which is known to show a characteristic peak at ~ 1030 nm [23]. This
peak is completely absent in case of S4 (Figure 7c-d). Similar conclusions about the relative
amount of long-lived population in each film can be drawn from monitoring the residual
bleach in the visible TA spectrum (ESI Figure S39). A residual singlet population is also
found (absorbing near 600 nm and 1400 nm) that outlives the same population in the neat
films, likely because of the possibility that long-lived polaron pair recombination due to
charge-transfer between A-D-A molecules and PCBM repopulates the singlet. The dynamics
of evolution of the spectra at 1030 nm can be fitted with three exponentials where the longest
lifetime is larger than the instrument resolution. (ESI Table S6). The amplitude of this

component, likely representing free charges, is highest for S3. The intermediate lifetime,
which may represent the geminate recombination time of polaron pairs, is longest for S1 and
S3. These trends are roughly consistent with the TRMC results.
Figure 7: Comparison of the transients in NIR region for the neat and blend films of S3 and S4.
While PC60BM anion peak is observed at ~1030 nm in case of S3:PC60BM blend, the same is missing
for S4:PC60BM blend.

Figure 8: Global analysis DADS (Decay associated difference spectra) of transient absorption
measurements for the films of blends of A-D-A molecules with PC60BM. The peak near 1030 nm is
present primarily in S1 and S3 blends but less prominent in S2 and absent in S4.
SVD (Singular value decomposition) analysis of the NIR-TA spectra was performed for the
neat and the blend films using three components (two for S4). Global fits for the neat samples
are shown in Supporting information (Figure S25). The global fits for the blends suggest the
presence of a long-lived species (lifetime greater than 1 ns) in the case of S1, S2 and S3
(Figure 8). From the spectral assignments made above, this can be attributed to free polaron
pairs (FPP). The longest lifetime is observed in the case of S3:PC60BM, which is consistent
with the TRMC results. The component with the shortest lifetime in all the cases corresponds
to the singlet exciton, and the corresponding decay associated spectra have similar spectral
features in neat and blend films with a characteristic peak around 980 nm. The lifetimes of
the singlet exciton are in the range of 1-5 ps, and the DADS show a negative feature near
1030 nm suggesting the concomitant rise of the PCBM anion. We observe that the lifetime is
longer in the CF3 substituted A-D-A molecules compared to the methyl substituted
counterpart. The production of the polaron pairs at the donor-acceptor interfaces gives rise to

the next decay step from 12-140 ps, which is likely a combination of geminate recombination
and exciton-charge annihilation due to coexistence of both PCBM anion and exciton features.
The final long-lived component arises from polaron pairs that escape fast geminate
recombination and give rise to the free polaron pairs. In addition to the 1030 nm PCBM
anion absorption, there remain broad singlet PIA features into the ns timescale that could
represent A-D-A exciton species re-formed upon polaron recombination. An additional
pathway of recombination of the geminate polaron pairs could be through formation of
triplet excitons,[24] although we do not clearly resolve triplet features.
Figure 9: Dynamics of photoinduced charge transfer in the A-D-A:PC60BM blends.
3.7. Investigation of morphology
The microstructure of the blends could be at least partially dictated by the different intermolecular
forces due to the substitution of the CF3 group, and thus unique optoelectronic properties could result.
In order to gauge the structure of different films on a nm scale, tapping mode AFM measurements
were performed on films of the S1-S4:PCBM (1:1) blends on quartz substrates. The films were
prepared by spin-coating from 1:1 solution of the ADA molecules with PC60BM in dichlorobenzene.
The films were annealed at 100 °C for 10 min to match the processing conditions for TRMC samples.
The roughness of the films for S1 and S2 was found to be similar (7.44 nm and 7.52 nm respectively).
Much higher roughness is obtained for the films for the benzodithiophene molecules (S3 and S4).
Among the two, S4 has a higher roughness (114 nm) than S3 (73.6 nm). The phase images of the
blend films are shown in Figure 10. While S1 and S2 have very similar morphology, a much greater
degree of phase separation is seen in the case of S4. This could be a result of aggregation of the S4
molecules due to stronger intermolecular forces. However, the lack of strong distinction between
films of S1 and S2 suggest that the CF3 group alone does not significantly modify the microstructure.

Figure 10: AFM phase images of the PC60BM blend films of (a) S1 (b) S2 (c) S3 (d) S4 for 1µm × 1
µm area.
4. Summary
In conclusion, four A-D-A small molecules (S1-4) with terminal BODIPY units were
synthesized, and the effects of trifluoromethyl groups on the optoelectronic properties of the
molecules were studied. As expected, we observed that LUMO energy levels are more
stabilized in case of the CF3 substituted A-D-A molecules (S2 and S4) resulting in narrower
band gaps than the methyl substituted derivatives (S1 and S3). This substitution strategy can
be used for tuning the driving force for electron transfer. In this case S2 and S4 therefore
have smaller driving forces for electron transfer to PC60BM. As observed from the solution
state transient absorption spectra, the excited state dynamics of the A-D-A compounds is also
affected by the CF3 substituent. The lifetime of the excited state that contains CT character
due to intramolecular charge transfer from the fluorene subunit (in S1 and S2) and BDT
subunit (in S3 and S4) to BODIPY, is reduced by the trifluoromethyl substitution. The
lifetime of the polarons, formed in neat films of these molecules, are shorter in case of the

trifluoromethylated A-D-A compounds, suggesting faster recombination in the presence of
CF3 group. Significant differences are also observed in photoinduced charge transfer from
these molecules into PC60BM acceptors. From TRMC measurements it was observed that the
A-D-A molecules with methyl substituents at the BODIPY meso positions (S1 and S3)
exhibit more efficient charge transfer to PC60BM acceptors than their counterparts with CF3
groups (S2 and S4). S3:PC60BM exhibits the highest φΣµ value among all the blends. This
was further supported by results from transient absorption spectroscopy wherein longer lived
free polaron pairs are observed in methylated A-D-A molecules (S1 and S3) compared to
their trifluoromethylated counterparts. Based on these results we can conclude that the
presence of the trifluoromethyl group at the para position of the meso phenyl substituent on
the BODIPY unit leads to increased recombination and is detrimental to photoinduced charge
transfer from this class of A-D-A donor molecules. This is in contrast with the effect of an
electron withdrawing group at the meso position in BODIPY based D-A-D type donor
molecules [12], where the better performance is attributed to better electronic communication
between the meso substituent and the core as well as better phase separation. In case of the
present set of A-D-A molecules, the meso substituents are not seen to affect the
delocalization of orbitals, as concluded from the DFT calculations. Considering that better
charge transfer efficiency is seen in S1 as compared to S2, even though their morphology on
a nm scale is very similar, we attribute the difference in charge transfer to the electronic
properties of the CF3 substituent. Thus, we conclude that BODIPY based donor molecules
with methyl substituents may have more potential utility in in certain types of optoelectronic
devices than those with trifluoromethyl groups.
5. Acknowledgement
The synthesis part of this work was conducted at Indian Institute of Science, Bangalore. We thank the
NMR Research Centre, IISc for the NMR characterization We are also thankful to Mr. Sandeep B. S.
for conducting the AFM measurements at the MNFC facility at CENSE, IISc. The spectroscopic
characterization was conducted at NREL as a part of the Fulbright-Kalam fellowship for climate
change (2018-19). This work was authored in part by the National Renewable Energy Laboratory,
operated by Alliance for Sustainable Energy, LLC, for the U.S. Department of Energy (DOE) under
Contract No. DE-AC36-08GO28308 with funding provided by the Office of Science, Basic Energy
Sciences, Division of Chemical Sciences, Biosciences, and Geosciences for transient spectroscopic
characterization. The views expressed in the article do not necessarily represent the views of the DOE
or the U.S. Government. The U.S. Government retains and the publisher, by accepting the article for

publication, acknowledges that the U.S. Government retains a nonexclusive, paid-up, irrevocable,
worldwide license to publish or reproduce the published form of this work, or allow others to do so,
for U.S. Government purposes.
6. References
[1] Kowada T, Maeda H, Kikuchi K. BODIPY-based probes for the fluorescence imaging of
biomolecules
in
living
cells.
Chem
Soc
Rev
2015;44:4953–72.
https://doi.org/10.1039/C5CS00030K.
[2] Kue CS, Ng SY, Voon SH, Kamkaew A, Chung LY, Kiew LV, et al. Recent strategies to
improve boron dipyrromethene (BODIPY) for photodynamic cancer therapy: an updated
review. Photochem Photobiol Sci 2018;17:1691–708. https://doi.org/10.1039/C8PP00113H.
[3] Xiao L, Wang H, Gao K, Li L, Liu C, Peng X, et al. A-D-A Type Small Molecules Based on
Boron Dipyrromethene for Solution-Processed Organic Solar Cells. Chem – Asian J
2015;10:1513–8. https://doi.org/10.1002/asia.201500382.
[4] Liao J, Zhao H, Xu Y, Cai Z, Peng Z, Zhang W, et al. Novel D–A–D type dyes based on
BODIPY platform for solution processed organic solar cells. Dyes Pigments 2016;128:131–40.
https://doi.org/10.1016/j.dyepig.2016.01.020.
[5] Bulut I, Huaulmé Q, Mirloup A, Chávez P, Fall S, Hébraud A, et al. Rational Engineering of
BODIPY-Bridged Trisindole Derivatives for Solar Cell Applications. ChemSusChem
2017;10:1878–82. https://doi.org/10.1002/cssc.201700465.
[6] Srinivasa Rao R, Bagui A, Hanumantha Rao G, Gupta V, Singh SP. Achieving the highest
efficiency using a BODIPY core decorated with dithiafulvalene wings for small molecule based
solution-processed
organic
solar
cells.
Chem
Commun
2017;53:6953–6.
https://doi.org/10.1039/C7CC03363J.
[7] Liao J, Zhao H, Cai Z, Xu Y, Qin FGF, Zong Q, et al. BODIPY-based panchromatic
chromophore for efficient organic solar cell. Org Electron 2018;61:215–22.
https://doi.org/10.1016/j.orgel.2018.05.049.
[8] Bucher L, Desbois N, Koukaras EN, Devillers CH, Biswas S, Sharma GD, et al. BODIPY–
diketopyrrolopyrrole–porphyrin conjugate small molecules for use in bulk heterojunction solar
cells. J Mater Chem A 2018;6:8449–61. https://doi.org/10.1039/C8TA01291A.
[9] Poe AM, Della Pelle AM, Subrahmanyam AV, White W, Wantz G, Thayumanavan S. Small
molecule BODIPY dyes as non-fullerene acceptors in bulk heterojunction organic photovoltaics.
Chem Commun 2014;50:2913–5. https://doi.org/10.1039/C3CC49648A.
[10] Valeur B, Leray I. Design principles of fluorescent molecular sensors for cation recognition.
Coord Chem Rev 2000;205:3–40. https://doi.org/10.1016/S0010-8545(00)00246-0.

[11] Bañuelos J, Arroyo-Córdoba IJ, Valois-Escamilla I, Alvarez-Hernández A, Peña-Cabrera E, Hu
R, et al. Modulation of the photophysical properties of BODIPY dyes by substitution at their
meso position. RSC Adv 2011;1:677–84. https://doi.org/10.1039/C1RA00020A.
[12] Marques dos Santos J, Jagadamma LK, Latif NM, Ruseckas A, Samuel IDW, Cooke G.
BODIPY derivatives with near infra-red absorption as small molecule donors for bulk
heterojunction solar cells. RSC Adv 2019;9:15410–23. https://doi.org/10.1039/C9RA01750J.
[13] Kyeong M, Lee J, Lee K, Hong S. BODIPY-Based Conjugated Polymers for Use as Dopant-
Free Hole Transporting Materials for Durable Perovskite Solar Cells: Selective Tuning of
HOMO/LUMO
https://doi.org/10.1021/acsami.8b05956.
[14] Choi S, Bouffard J, Kim Y. Aggregation-induced emission enhancement of a meso-
trifluoromethyl BODIPY via J-aggregation. Chem Sci 2014;5:751–5.
https://doi.org/10.1039/C3SC52495G.
Levels.
ACS
Appl
Mater
Interfaces
2018;10:23254–62.
[15] Li T, Meyer T, Ma Z, Benduhn J, Körner C, Zeika O, et al. Small Molecule Near-Infrared Boron
Dipyrromethene Donors for Organic Tandem Solar Cells. J Am Chem Soc 2017;139:13636–9.
https://doi.org/10.1021/jacs.7b07887.
[16] Prasannan D, Raghav D, Sujatha S, Hareendrakrishna kumar H, Rathinasamy K, Arunkumar C.
Synthesis, structure, photophysical, electrochemical properties and antibacterial activity of
brominated BODIPYs. RSC Adv 2016;6:80808–24. https://doi.org/10.1039/C6RA12258B.
[17] Hayashi Y, Yamaguchi S, Cha WY, Kim D, Shinokubo H. Synthesis of Directly Connected
BODIPY Oligomers through Suzuki–Miyaura Coupling. Org Lett 2011;13:2992–5.
https://doi.org/10.1021/ol200799u.
[18] Kim J-H, Kim HU, Kang I-N, Lee SK, Moon S-J, Shin WS, et al. Incorporation of Pyrene Units
to Improve Hole Mobility in Conjugated Polymers for Organic Solar Cells. Macromolecules
2012;45:8628–38. https://doi.org/10.1021/ma301877q.
[19] Chen C-P, Chan S-H, Chao T-C, Ting C, Ko B-T. Low-Bandgap Poly(Thiophene-Phenylene-
Thiophene) Derivatives with Broaden Absorption Spectra for Use in High-Performance Bulk-
Heterojunction Polymer Solar Cells.
https://doi.org/10.1021/ja801877k.
J
Am Chem Soc 2008;130:12828–33.
[20] Li C-Z, Chien S-C, Yip H-L, Chueh C-C, Chen F-C, Matsuo Y, et al. Facile synthesis of a 56π-
electron 1,2-dihydromethano-[60]PCBM and its application for thermally stable polymer solar
cells. Chem Commun 2011;47:10082–4. https://doi.org/10.1039/C1CC14446D.
[21] Zhou Y, Ding L, Shi K, Dai Y-Z, Ai N, Wang J, et al. A Non-Fullerene Small Molecule as
Efficient Electron Acceptor in Organic Bulk Heterojunction Solar Cells. Adv Mater
2012;24:957–61. https://doi.org/10.1002/adma.201103927.

[22] Reid OG, Rumbles G. Quantitative Transient Absorption Measurements of Polaron Yield and
Absorption Coefficient in Neat Conjugated Polymers. J Phys Chem Lett 2013;4:2348–55.
https://doi.org/10.1021/jz401142e.
[23] Guo J, Ohkita H, Benten H, Ito S. Charge Generation and Recombination Dynamics in Poly(3-
hexylthiophene)/Fullerene Blend Films with Different Regioregularities and Morphologies. J
Am Chem Soc 2010;132:6154–64. https://doi.org/10.1021/ja100302p.
[24] Snedden EW, Monkman AP, Dias FB. Kinetic Studies of Geminate Polaron Pair
Recombination, Dissociation, and Efficient Triplet Exciton Formation in PC:PCBM Organic
Photovoltaic Blends. J Phys Chem C 2012;116:4390–8. https://doi.org/10.1021/jp208820g.

Highlights of “BODIPY based A-D-A molecules: Effect of CF3 group substitution at meso
position”
•
Two pairs of new A-D-A dyes based on BODIPY have been synthesized and
characterized.
•
•
•
The effect of trifluoromethyl substituent on properties of BODIPY based dyes studied.
Electron transfer studied using TRMC and TA measuements.
Mechanism of electron transfer from BODIPY based A-D-A molecules to PC60BM
proposed.
•
Trifluoromethyl substituent lowers electron transfer.

Author Statement
Gourav Tarafdar: Conceptualization, Methodology, Validation, Investigation, Formal
analysis, Writing-Original Draft, Visualization
Justin Johnson: Methodology, Investigation, Formal Analysis, Resources, Writing- Review
and Editing, Supervision
Bryon W Larson: Methodology, Investigation, Formal Analysis, Resources, Writing- Review
and Editing.
Praveen C Ramamurthy: Conceptualization, Writing- Review and Editing, Supervision,
Project administration, Funding acquisition

Declaration of interests
☒ The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.
☐The authors declare the following financial interests/personal relationships which may be considered
as potential competing interests: