Spirobifluorene based small push-pull molecules for organic photovoltaic applications

Article abstract of DOI:10.1016/j.dyepig.2017.01.034

Four analogous push-pull systems have been synthesized. If the latter all involve the same electron rich diphenylamine termination (D) and π-conjugating spacer (p) they differ from their electron withdrawing groups (A) and more importantly by their linear or 3D structure. Indeed, two push-pull spirobifluorene derivatives, which present two perpendicular D-p-A systems by molecule, are compared to their linear analogues. After description of their syntheses, spectroscopic and electrochemical properties, comforted by theoretical calculations, are discussed and compared. Then, a preliminary evaluation of compounds as active materials in organic solar cells is presented and demonstrates the potential interest of spiro-based derivatives for organic photovoltaics.

Full text of DOI:10.1016/j.dyepig.2017.01.034

Accepted Manuscript
Spirobifluorene based small push-pull molecules for organic photovoltaïc applications
Clément Dalinot, György Szalóki, Chloé Dindault, Olivier Segut, Lionel Sanguinet,
Philippe Leriche
PII:
S0143-7208(16)30808-7
10.1016/j.dyepig.2017.01.034
DYPI 5733
DOI:
Reference:
To appear in: Dyes and Pigments
Received Date: 26 September 2016
Revised Date: 13 January 2017
Accepted Date: 13 January 2017
Please cite this article as: Dalinot C, Szalóki G, Dindault C, Segut O, Sanguinet L, Leriche P,
Spirobifluorene based small push-pull molecules for organic photovoltaïc applications, Dyes and
Pigments (2017), doi: 10.1016/j.dyepig.2017.01.034.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. 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.

ACCEPTED MANUSCRIPT
Spirobifluorene based small push-pull molecules for organic photovoltaïc applications
Clément Dalinot, György Szalóki, Chloé Dindault, Olivier Segut, Lionel Sanguinet* and Philippe
Leriche*
MOLTECH – Anjou UMR CNRS 6200, Groupe systèmes conjugués linéaires, University of
Angers, 2 boulevard Lavoisier, 49045 Angers, France
philippe.leriche@univ-angers.fr, 33241735010; lionel.sanguinet@univ-angers.fr, 33241735374
Abstract
Four analogous push-pull systems have been synthesized. If the latter all involve the same
electron rich diphenylamine termination (D) and π-conjugating spacer (p) they differ from their
electron withdrawing groups (A) and more importantly by their linear or 3D structure. Indeed,
two push-pull spirobifluorene derivatives, which present two perpendicular D-p-A systems by
molecule, are compared to their linear analogues. After description of their syntheses,
spectroscopic and electrochemical properties, comforted by theoretical calculations, are
discussed and compared. Then, a preliminary evaluation of compounds as active materials in
organic solar cells is presented and demonstrates the potential interest of spiro-based
derivatives for organic photovoltaics.
Keywords
Organic electronics, Organic solar cells, Spirobifluorene, Fluorene, Diphenylamine
1. Introduction
The molecular engineering of π conjugated systems is today a subject in strong expansion
due to continuous progresses in organic electronics. Indeed, in addition to incontestable
technological advances that allow optimizations of organic devices, another more fundamental
approach consists in exploring structure-properties relationships of compounds and materials
still not studied in the domain.1, 2 As example, we and others have demonstrated few years ago
that small propeller shape organic systems based on the triphenylamine family already known
for their semiconductor properties are promising candidates as chromophores for photovoltaic
applications after their proper functionalization with electron withdrawing groups.3-5 Based on
these results, numerous other tridimensional systems have been synthesized and incorporated in
photovoltaic cells.1, 6 In this context, our group was involved in the development of oligomers
or polymers based on 2D or 3D nodes such as triphenylamine,7-9 triazine,10 cruciform
quaterthiophene11 and more recently spirobifluorene (SBF).12, 13 Indeed, the spiro-concept was

2
ACCEPTED MANUSCRIPT
originally introduced in order to favour obtainment of amorphous materials presenting more
homogeneous transport properties by improving the morphological stability of low molecular
mass systems while retaining their electronic properties.14, 15 Generally presented as two π
conjugated systems linked together via a common sp3-hybridized atom, spiro derivatives
exhibit higher solubility than corresponding non-spiro-linked parent compounds,14 and the
orthogonality of both molecular halves associated to their rigidity lead to entanglement in the
amorphous state and hinder crystallization.16 Concerning this last motif, the uses of SBF
moieties are generally focused on the preparation of materials as semiconductors or fluorescent
systems.14 At the opposite, the use of this platform to elaborate donor17-19 or acceptor20, 21 layer
materials for OPV is up to now only very seldom explored. However, the direct influence of
this platform on the device efficiency remains nearly absent from the OPV literature and, to the
best of our knowledge, is limited to bulk-heterojunction (BHJ) solar cells.22, 23
Thus, in order to determine if this motif really deserves attention and allows straightforward
access to small molecules suitable for OPV applications, we developed two simple models
which, for comparison, were competed with their fluorene linear analogues. In this context, two
sets of different push-pull systems incorporating either a SBF moiety as π conjugated system or
its corresponding non-spiro-linked parent have been prepared. After discussion of their
spectroscopic and electrochemical properties supported by theoretical results, derivatives have
been incorporated as donor materials in planar heterojunction (PHJ) solar cells as a first
approach to evaluate their potential as material for OSC.
2. Experimental
2.1. General
Flash chromatography was performed with analytical grade solvents using Aldrich silica gel
(technical grade, pore size 60 Å, 230-400 mesh particle size). Flexible plates ALUGRAM®
Xtra SIL G UV254 from MACHEREY-NAGEL were used for TLC. Compounds were
detected by UV irradiation (Bioblock Scientific) or staining with I2, unless stated otherwise.
The NMR spectra were recorded with a Bruker AVANCE III 300 (1H, 300 MHz and 13C, 75
MHz) using CDCl3 as solvent. Chemical shifts are reported in ppm relative to the solvent
1
residual value: δ = 7.26 for H NMR and δ = 77.16 for 13C NMR. Coupling constants are
reported in Hz and rounded to the nearest 0.1 Hz.
Infrared spectra were recorded on a Bruker Vertex 70 spectrometer. UV-Vis spectra were
performed with a Perkin Elmer 950 spectrometer in CHCl3 (HPLC grade) solutions. Mass
spectra were recorded using BIFLEX III Bruker Daltonics spectrometer (MALDI-TOF,

3
ACCEPTED MANUSCRIPT
dithranol matrix). Photoluminescence properties of compounds were investigated by
fluorescence emission spectroscopy recorded on Photon Technology International LPS 220
spectrofluorimeter using CHCl3 (HPLC grade) as solvent and Fluorescein as standard for
quantum yields determination.
The electrochemistry studies were performed on dichloromethane solution (HPLC grade)
with tetra-n-butylammonium hexafluorophosphate (TBAP, electrochemical grade from Fluka)
as the support electrolyte (0.1M) under a dry and oxygen-free atmosphere (H2O < 1 ppm, O2 <
1 ppm). Cyclic voltammetry (CV) was performed in a three-electrode cell equipped with a
platinum millielectrode as working electrode, a platinum wire counter-electrode and a silver
wire used as a quasi-reference electrode. The voltammograms were recorded on a
potentiostat/galvanostat (BioLogic – SP150) driven by the EC-Lab software with positive
feedback compensation and were all calibrated versus Ferrocene/Ferrocenium oxidation
potential (+0.405V vs SCE or +0.425V vs Ag/AgCl).24
As commonly admitted, the energy levels and the electrochemical gaps of the compounds
were evaluated from oxidation (Eox) and reduction (Ered) potentials.25
2.2. Devices Preparation and Characterization
Indium-tin oxide (ITO) coated glass slides were patterned using a 37% hydrochloric acid
solution and zinc powder. The substrates were then washed with a diluted Deconex solution
(5% in water) and cleaned by subsequent ultrasonic treatments for 20 min in acetone, ethanol
and isopropanol. Once dried under a steam of nitrogen, a UV-ozone plasma treatment
(UV/Ozone ProCleaner Plus, Bioforce Nanosciences) was performed for 15 min. A filtered
aqueous solution of poly(3,4-ethylenedioxythiophene)-poly(styrenesulfonate) (PEDOT:PSS;
Clevios P VP. AI 4083) through a 0.45 mm PTFE membrane (Millex®) was spun-cast onto the
patterned ITO surface at 5000 rpm for 40 s before being baked a 140°C for 20 min. A solution
of molecular donor in chloroform (10 mg/mL) was spun-cast at different spin-speeds onto the
PEDOT: PSS layer. Finally, devices were completed by the successive thermal deposition of
fullerene C60 (20 nm) and aluminum (120 nm) through a shadow mask defining six cells of 27
mm² each. J vs V curves were recorded in the dark and under illumination using a Keithley 236
source-measure unit and a home-made acquisition program. The light source was an AM1.5
Solar Constant 575 PV simulator (Steuernagel Lichttecknik, equipped with a metal halogen
lamp). The light intensity was measured by a broad-band power meter (13PEM001, Melles
Griot). EQE were measured under ambient atmosphere using a halogen lamp (Osram) with an

4
ACCEPTED MANUSCRIPT
Action Spectra Pro 150 monochromator, a lock-in amplifier (PerkineElmer 7225) and a S2281
photodiode (Hamamatsu).
Space charge limited current (SCLC) measurements were carried out from single layer
devices fabricated as followed process: a solution of molecular donor (10 mg/mL) in CHCl3
was spun cast at 1000 rpm on the above described PEDOT: PSS substrate to provide the
organic layer. Gold cathode (60 nm) was thermally evaporated under vacuum, through a
shadow mask defining actives area of 12.60 mm², 3.10 mm², and 0.78 mm² per substrate. Hole
mobilities, µh, were evaluated using the Mott-Gurney law, i.e. JSCLC=(9/8)ε0εrµh(V2/d3) where
εr is the static dielectric constant of the medium (εr = 3) and d, the thickness of the active
layer.26
2.3. Synthesis
2,2'-(2,2'-bis(diphenylamino)-9,9'-spirobi[fluorene]-7,7'diyl)bis
(methanylylidene)dimalononitrile
SCN;
7-(diphenylamino)-9,9-dimethyl-9H-fluorene-2-
carboxaldehyde27
1
and 2,2'-(2,2'-bis(diphenylamino)-9,9'-spirobi[fluorene]-7,7'diyl)bis
carboxaldehyde28 2 were synthesized according to previously reported methods.
(5-((7-(diphenylamino)-9,9-dimethyl-9H-fluoren-2-yl)methylene)-3-ethyl-2-
thioxothiazolidin-4-one (FRh).
Under argon atmosphere 200 mg (0.51 mmol) of 7-(diphenylamino)-9,9-dimethyl-9H-
fluorene-2-carboxaldehyde 1 and 95.2 mg (0.59 mmol, 1.15 equiv.) of 3-ethylrhodanine were
added to 25 mL of acetic acid and the solution was refluxed for 10h in the presence of 395 mg
of ammonium acetate (5.1 mmol, 10 equiv.). After cooling to room temperature, the mixture
was poured into iced water. The precipitate was filtered and washed with distilled water. The
product was purified by column chromatography (pentane / methylene chloride 1/4) and
isolated as red solid (220mg, 80%). 1H NMR (300 MHz, CDCl3): δ 7.81 (s, 1H), 7.69 (d, J =
8.5 Hz, 1H), 7.59 (d, J = 8.3 Hz, 1H), 7.47 (m, 2H), 7.36 – 7.26 (td, J = 6.6, 1.8 Hz, 4H), 7.22 –
7.11 (m, 5H), 7.08 (d, J = 7.3 Hz, 2H), 7.03 (dd, J = 8.4, 1.8 Hz, 1H), 4.22 (q, J = 7.1 Hz, 2H),
1.44 (s, 6H), 1.31 (t, J = 7.1 Hz, 3H).13C NMR (75 MHz, CDCl3): δ 193.3, 167.8, 156.2,
154.6, 148.9, 147.7, 142.4, 134.1, 132.1, 131.4, 131.2, 129.5, 124.8, 124.6, 123.4, 122.7, 121.7,
121.2, 120.1, 117.5, 47.0, 39.9, 27.1, 12.4. IR (neat): ῡ = 1704 (C=O), 1578 (C=C), 1240(C=S),
M.P: 124°C, MS (MALDI-TOF) m/z (%): 532.3 (100) [M]+●, HRMS (FAB): m/z calcd. for
C33H28N2OS2: 532.1643 [M] +●; found: 532.1640

5
ACCEPTED MANUSCRIPT
2-((7-(diphenylamino)-9,9-dimethyl-9H-fluoren-2-yl)methylene)malononitrile (FCN).
7-(diphenylamino)-9,9-dimethyl-9H-fluorene-2-carboxaldehyde 1 (200 mg, 0.51 mmol),
malononitrile (135 mg, 2.05 mmol), and basic aluminum oxide (716 mg) were stirred in toluene
(20 mL) for 24 h at 100°C under N2. After cooling to room temperature, the reaction solution
was filtered. The filtrate was subjected to flash column chromatography (silica gel, methylene
chloride). A red solid was obtained (70 mg, 31%). 1H NMR (300 MHz, CDCl3): δ 7.98 (d, J =
1.2 Hz, 1H), 7.81 (dd, J = 8.2, 1.5 Hz, 1H), 7.75 (s, 1H), 7.69 (d, J = 8.0 Hz, 1H), 7.60 (d, J =
8.4 Hz, 1H), 7.30 (td, J = 7.2, 1.8 Hz, 4H), 7.16 (dd, J = 7.8, 0.9 Hz, 5H), 7.09 (td, J = 7.2, 1.2
Hz, 2H), 7.03 (dd, J = 8.4, 2.1 Hz, 1H), 1.43 (s, 6H).13C NMR (76 MHz, CDCl3): δ 159.8,
157.0, 154.6, 150.0, 147.5, 146.5, 131.9, 131.1, 129.6, 128.9, 125.2, 124.5, 123.9, 122.4, 122.2,
120.0, 116.7, 114.7, 113.7, 47.1, 26.9. IR (neat): ῡ = 2224 (CN), 1571 (C=C), 1555 (C=C),
1297, M.P: 205°C, MS (MALDI-TOF) m/z (%): 437.4 (100) [M]+●, HRMS (FAB+): m/z calcd.
for C31H23N3 :437.1892 [M]+●; found: 437.1902.
2,2'-bis(diphenylamino)-9,9'-spirobi[fluorene]-7,7'-diyl)bis(methanylylidene))bis(3-ethyl-
2-thioxothiazolidin-4-one) (SRh).
Under argon atmosphere 50 mg (0.070 mmol) of 2,2'-bis(diphenylamino)-9,9' spirobi[fluorene]
-7,7'-dicarboxaldehyde 2 and 23.95 mg (0.15 mmol, 2.1 equiv.) of 3-ethylrhodanine were
added to 10 mL of acetic acid and the solution was refluxed for 6 hours in the presence of 54
mg of ammonium acetate (0.70 mmol, 10 equiv.). After cooling to room temperature, the
mixture was poured into iced water. The precipitate was filtered and washed with distilled
water. After drying under vacuum, the precipitate was recrystallized from ethanol to give a red
compound (50 mg, 71%).1H NMR (300 MHz, CDCl3): δ 7.77 (d, J = 8.0 Hz, 2H), 7.67 (d, J =
8.4 Hz, 2H), 7.59 (s, 2H), 7.47 (dd, J = 8.1, 1.5 Hz, 2H), 7.20 (m, 8H), 7.07 – 6.94 (m, 14H),
6.80 (d, J = 0.9 Hz, 2H), 6.62 (d, J = 1.9 Hz, 2H), 4.15 (q, J = 7.1 Hz, 4H), 1.24 (t, J = 7.0 Hz,
6H). 13C NMR (76 MHz, CDCl3): 193.2, 167.6, 150.1, 149.3, 149.2, 147.4, 144.3, 134.9,
133.4, 132.1, 131.1, 129.4, 125.9, 124.4, 124.3, 123.4, 122.2, 121.9, 120.6, 118.8, 65.5, 39.9,
12.4. IR (neat): ῡ = 1718 (C=O), 1570 (C=C), 1205 (C=S), M.P: 175°C, MS (MALDI-TOF)
m/z (%): 992.2 (100) [M]+●, HRMS (MALDI-TOF): m/z calcd. for C61H44N4O2S4 : 992.2347
[M]+●; found: 992.2333 .

6
ACCEPTED MANUSCRIPT
2,2'-((2,2'-bis(diphenylamino)-9,9'-spirobi[fluorene]-7,7' diyl)bis(methanylylidene))
dimalononitrile (SCN).
2,2'-bis(diphenylamino)-9,9'-spirobi[fluorene]-7,7'-dicarboxaldehyde 2 (125 mg, 0.17 mmol),
malononitrile (46.73 mg, 0.70 mmol), and basic aluminum oxide (238 mg) were stirred in
toluene (5 mL) for 24 h at 100°C. After cooling to room temperature, the reaction solution was
filtered. As SCN appears fairly stable on stationary phases, the residue was subjected to several
short flash column chromatographies (silica gel, methylene chloride). A red solid was obtained
(67 mg, 47%). 1H NMR (300 MHz, CDCl3): δ 7.97 (dd, J = 8.3, 1.5 Hz, 2H), 7.77 (d, J = 8.1
Hz, 2H), 7.67 (d, J = 8.4 Hz, 2H), 7.55 (s, 2H), 7.22 (t, J = 7.8 Hz, 8H), 7.10 (d, J = 1.4 Hz,
2H), 7.03 (m, 14H), 6.60 (d, J = 2.0 Hz, 2H). 13C NMR (76 MHz, CDCl3): δ 159.2, 150.2,
150.1, 148.8, 148.1, 147.1, 133.8, 131.3, 129.5, 126.2, 124.7, 123.9, 122.9, 120.6, 117.8, 114.3,
113.1, 80.4, 65.2, M.P: 135°C, IR (neat): ῡ = 2224 (CN), 1556 (C=C), 1463, 1273. HRMS
(FAB +): calcd. for C57H34N6 : 802.2845[M]+●.; Found 802.2846.
3. Results and discussion
3.1. Synthesis
As shown on scheme 1, the four targets prepared for this study can be considered as simple
push-pull systems using a common diphenylamine node as donor part but exhibiting either a
different π conjugated system (linear fluorene vs tridimensional spirobifluorene) or a different
acceptor moiety (dicyanovinyl vs 3-ethyl-rhodanine). The modification of the π conjugated
system required the most synthetic effort but can be easily overcome. In fact, both
corresponding aldehydes can be obtained in two steps form commercial compounds consisting
in formylation of bromo derivatives followed by a classical Buchwald N-arylation (Scheme 1)
as described in the literature.27, 28 The introduction of the different electron withdrawing groups
was not straightforward and each of them required specific experimental conditions to carry out
the condensation. Based on previous studies,27 FCN and SCN were efficiently synthesized in
refluxing toluene in presence of basic alumina, but application of these conditions to 3-ethyl-
rhodanine led to only traces of desired compounds. In this context, more sensitive FRh and
SRh have to be prepared in boiling acetic acid in presence of acetate ammonium. It is worth to
note that the nature of π conjugated system has a strong impact on the condensation efficiency.
As reactions from fluorene starting derivative appear straightforward and allow a clean
obtainment of targets, the total consumption of starting spirobifluorene aldehydes appears
difficult and derivatives are generally obtained as mixtures of mono condensate and target.

7
ACCEPTED MANUSCRIPT
Moreover, SCN and SRh seem moderately stable and, as consequence, their purification by
chromatography appears tricky due to their rapid decomposition on almost all classical
stationary phases (silica, acidic, neutral and basic alumina). As consequence, the purification of
SCN (SRh is purified by recrystallization in a non protic solvent) should be performed through
several flash chromatographies (silica gel). However, the different compounds are stable and
can be stocked either in solution or in solid state after the purification. As consequence, the four
targets were isolated and fully characterized despite these limitations.
(i)
(ii)
, Al2O3, toluene, ∆
, AcO(NH4), AcOH, ∆
(i)
(ii)
1/ TiCl4
(i)
2/ CH3OCHCl2
CH2Cl2
(ii)
Scheme 1. Representation of the synthetic pathways to obtain the push-pull spiro derivatives and their non-
spiro-linked parents.
3.2. Molecular electronic properties
The main optical characteristics of the four compounds, such as example the maximum
absorption wavelength (λmax) and the maximum emission wavelength (λem), determined in
solution, are collected in Table 1.
Stokes shift
(eV)
E0-0
(eV)
ε
λmax
(nm)
λem
(nm)
φ
(%)
Compound
(M-1.cm-1)
49 200
66 100
310
487
624
626
630
639
0.55
0.55
0.65
0.65
2.25
2.25
2.26
2.24
9.8
27.4
5.4
SCN
FCN
SRh
FRh
29 400
44 000
312
488
42 200
61 400
25 100
39 300
306
474
310
478
15.5
Table 1. Absorption and emission characteristics of the different four compounds in chloroform solution.
The optical band gap corresponding to E0-0 was obtained from the intersection of the absorption and
emission spectra.

8
ACCEPTED MANUSCRIPT
As we can see on figure 1, all compounds present analogous absorption spectra composed in
two characteristic absorption bands. Starting from photophysical studies on similar fluorene
derivatives,29 and assuming that SBF and Fluorene present an absorption at low energy
wavelength centred on 309 and 301 nm respectively,30, 31 we can reasonably assign the first
band ranging from 306 to 312 nm to π-π* transitions. The second absorption band (in the 474-
488 nm range), may result from an internal charge transfer between the electron rich and
deficient parts of the conjugated systems as confirmed by calculation.
a)
b)
Figure 1. UV-visible (solid line) and emission (λexc= 480 nm; dotted line) spectra of compounds in
chloroform solutions (respectively at 10-5 and 10-6 M) of spiro (black) and fluorene (grey) functionalized by
a dicyanovinyl (left, a) or a rhodamine (right, b) as electron withdrawing group.
Indeed, ab initio quantum calculations were performed with the Gaussian 09 package of
programs at a hybrid density functional theory (DFT) level and carried out using the PBE1PBE
as functional and 6-311++G(2df,2pd) as basis. For all four compounds, the geometries and
electron distributions are shown in supplementary data. Concerning the fluorene derivatives,
the simulation of the different absorptions by TD-DFT confirms the presence of two main
absorption bands with a lower electronic transition which corresponds to a 99% HOMO ꢀ
LUMO transition with a strong charge transfer (CT) character shown by relative localizations
of levels. At the opposite, the second band corresponds to a higher energy π-π* transition
involving higher and lower molecular orbitals as resumed in table 2. In this context, the
observation of a bathochromic shift of the CT band around 10 nm between rhodanine based
derivatives and dicyanovinyl terminated compounds is not surprising and in agreement with
experimental observations reported elsewhere for such structural modification.32 In the same
time, a slight hyperchromic effect is also observed. Both effects are commonly observed in
push-pull systems by modification of the acceptor strength and result from a lowering of the

9
ACCEPTED MANUSCRIPT
internal charge transfer energy and are well reproduced on simulated spectra. At the opposite,
the high energy band is logically almost unaffected by the modification of the electron
withdrawing group due to its poor charge transfer character.
Wavelength
(nm)
Oscillator
strength
Compound
Excitation
Energy (eV)
Transition
72% HOMO-1ꢀLUMO
20% HOMO-1ꢀ LUMO+1
59% HOMO-1ꢀLUMO+1
36% HOMOꢀ LUMO+1
S0ꢀS3
499
2.48
0.582
SCN
S0ꢀS2
S0ꢀS1
509
560
2.43
2.21
0.163
0.005
99% HOMOꢀLUMO
56% HOMO-1ꢀLUMO
35% HOMO-1ꢀLUMO+1
S0ꢀS2
S0ꢀS1
339
494
3.65
2.51
0.363
0.766
FCN
SRh
FRh
99% HOMOꢀLUMO
S0ꢀS3
S0ꢀS2
S0ꢀS1
503
507
541
2.47
2.45
2.29
0.762
0.378
0.012
86% HOMO-1 ꢀLUMO
53% HOMO ꢀLUMO+1
35% HOMO-1 ꢀLUMO+1
98% HOMOꢀLUMO
S0ꢀS2
S0ꢀS1
384
504
3.23
2.46
0.528
0.928
96% HOMO-1ꢀLUMO
99% HOMOꢀLUMO
Table 2. Main characteristics of electronic transitions obtained by TD-DFT calculation for the four
different compounds
.
As shown on figure 1, the passage from linear to spiro compounds has only a negligible
effect on experimental absorption spectra and both SCN and SRh compounds exhibit two
absorption bands similar to those of their corresponding linear parent (table 1). Nevertheless, it
can be pointed out that all bands of spiro derivatives are slightly hypsochromically shifted
compare to those of fluorene systems. In these conditions, we can reasonably assume that both
molecular halves linked via the sp3 hybridized carbon act independently what could explain the
strong hyperchromic effect (by a factor 1.5-1.6) observed whatever the nature of the electron
withdrawing group and the absorption band. Associating two identical chromophorous motif
acting independently in a perpendicular manner do not lead to a two-fold increase of the molar
extinction coefficient. Indeed, the value of the absorption coefficient is essentially dependent
on two factors, which are, the percentage of orbital overlap between the initial and final wave
functions, but also the transition moment within the considered transition. Due to geometric
consideration in symmetrical push-pull bis-substituted spiro, as in our present cases, the global
dipole moment and transition moment can be approximated by considering them as the vector
sum of the two dipoles and transition moment of the two orthogonally linked fragment
respectively.33 Due to the relation of Chasles, we could thus expect an enhancement of the

10
ACCEPTED MANUSCRIPT
absorption coefficient by a factor 1.4, when comparing the spiro derivatives to their linear
parents. As we can see, this simple assumption allows to provide results very closed to the
experimental observation (1.5-1.6).
Complementary to the experimental observations, the TD-DFT calculations performed on
spiro derivatives did not follow this trend and let presume that the system is pseudo-
degenerated due to symmetrical purposes. Indeed, HOMO/HOMO-1 and LUMO/LUMO+1
couples differ only from their localization on each perpendicular conjugated system and are
separated by only 0.11 and 0.07eV for SCN and SRh respectively. This feature conducts to
generate three different absorption bands in the visible range. However, the first S0ꢀS1
energies at 560 and 541 nm for SCN and SRH respectively corresponding to HOMOꢀLUMO
transitions are not observed due to negligible oscillator strengths. These low values of the latter
at 0.005 and 0.012 can be attributed to geometric purposes. In fact, the HOMO level is
localized on one conjugated system as the LUMO one is confined on its perpendicular pendant.
As consequence, the overlap between the two wave functions is too low and the transition
probability close to zero. In contrary, the oscillator strengths of HOMO-1ꢀLUMO and
HOMOꢀLUMO+1 which are respectively localized on the same conjugated system are in both
case significant. Accompanied with HOMO-1ꢀLUMO+1 transitions they allow the
observation of calculated bands in the 500 nm area. If these calculations reproduce well the
slight bathochromic shift observed experimentally between spiro derivatives and linear parents,
the predicted enlargement of the ITC band as consequence of the two electronic transitions at
close wavelengths is not observed.
As observed for the absorption properties, the emission wavelengths of both spiro
compounds are almost identical to their corresponding linear parent and, in each case, the
emission spectra present structureless bands at close wavelengths. Indeed, SRh and FRh
derivatives present maxima of emission at 630 and 639 nm somewhat bathochromically shifted
compared to SCN and FCN at 624 and 626 nm due to 0.1 eV higher stoke shifts. As linear
compounds FCN and FRh present high fluorescence quantum yields calculated against
fluorescein at 27 and 15%, fluorescence properties of their spiro analogues SCN and SRh
appear partially quenched as they present threefold lower quantum yields at 10 and 5%
respectively. This decrease could be explained by the strong influence of the polarity of the
molecular environment on the quantum yield as reported earlier.33 At the opposite to their
linear parent, in push-pull spiro-derivatives each molecular halve induces to the other one a
strong electric field which was estimated in the case of SCN to 12 MV.cm-1 at a 5Å distance.33

11
ACCEPTED MANUSCRIPT
The substitution of the π conjugated system by a spiro node should impact the packing, and
the processability of the molecule. For these reasons, the absorption properties of films of the
different compounds spun-cast onto ITO/PEDOT transparent supports have been studied. The
direct comparison between normalized solution and film absorption spectra for all four
derivatives is presented below (Figure 2).
1.0
1.0
0.8
0.8
0.6
0.6
0.4
0.4
0.2
0.2
0.0
0.0
400
500
600
700
800
400
500
600
700
800
(a)
(b)
1.0
0.8
0.6
0.4
0.2
0.0
1.0
0.8
0.6
0.4
0.2
0.0
400
500
600
700
800
400
500
600
700
800
(c)
(d)
Figure 2. UV-Visible spectra of FCN (a), FRh (b), SCN (c) and SRh (d) solution (solid line) and thin film
(dotted line).
In all cases, very close maxima and an enlargement of the charge transfer absorption band as
well as the appearance of a new very weak band around 600 nm are observed. If such
modifications on solid state are regularly observed for push-pull systems and could be
rationalized using Kasha exciton model,34 we can observe that these effects lead only to a few
nm shift varying with the nature of the electron withdrawing group and the π conjugated
system.
Noteworthy, the almost constancy of the optical bandgap E0-0 (2.24-2.26 eV) observed in
solution whatever the nature of the acceptor group and/or the π conjugated (Table 1) system
does not remain on solid state. Indeed, the bandgap in the solid state calculated from the edge
of the main energy band (λonset) of dicyanovinyl based compounds are identical (577nm) and
appear 25 nm bathochromically shifted compared to those of SRh and FRh (552 nm) what

12
ACCEPTED MANUSCRIPT
probably indicates better intermolecular interactions in materials based on the smaller electron
withdrawing motif.
opt
Eg
λ
max (nm)
λonset(nm)
thin film
Donor
thin film
thin film
(eV) c
SCN
FCN
SRH
FRH
487
487
470
472
577
577
552
552
2.15
2.15
2.25
2.25
Table 3. Absorption characteristics of the four different compounds in thin film spun-cast onto
ITO/PEDOT transparent supports. (c) Calculated via E=1240/λ with λ=λonset
.
As commonly required for OPV materials, the electrochemistry properties of the four
different compounds were investigated by cyclic voltammetry and the results are gathered in
table 4. As expected, all cyclic voltammograms present two redox processes, one reversible in
oxidation leading to a stable bis-radical cation (in the case of the spiro derivatives) and another
one, in reduction, more or less reversible depending on the structure of the molecule. In
accordance with spectroscopic data, the difference between the oxidation and reduction
potentials are close in the 2.20-2.25 eV area.
Compound Epa
(V)
Epc EHOMO-(DFT) ELUMO-(DFT)
(V)
(eV)
(eV)
SCN
FCN
SRH
FRH
0.59 -1.60
0.55 -1.69
0.55 -1.70
0.49 -1.71
-5.80
-3.06
-5.74
-5.54
-5.49
-2.86
-2.74
-2.65
Table 4. Electrochemical characteristics of the four different compounds in CH2Cl2 (with TBAP 0.1M as
electrolyte) and energy of frontier orbitals calculated by DFT calculation.
As suggested by spectroscopic data, the lower acceptor strength of the rhodanine in
comparison to dicyanovinyl is well confirmed by the lower oxidation potential and higher
reduction potential of SRh and FRh. Similarly, the small potential shift between spiro
derivatives and their linear parents confirms the poor influence of the modification of the π
conjugated system. In this context, all four compounds exhibit low HOMO levels which should
be beneficial for obtaining a reasonable Voc value.35

13
ACCEPTED MANUSCRIPT
3.4. Devices characterizations
Consequently, all derivatives have been evaluated as donor material in planar heterojunction
(PHJ) solar cells ITO/PEDOT-PSS(40nm)/donor/C60(20nm)/Al(120nm). The open-circuit
voltages (Voc), short-circuit current densities (Jsc), fill factors (FF) and power conversion
efficiencies (PCE) of best cells of 0.27 cm2 active area after optimization of the deposition
condition of the donor were determined under AM 1.5 simulation solar illumination and are
collected in table 5.
PCE
rpm
Voc max
Jscmax
(mA.cm-²)
FF max
(%)
µh
(cm²V-1s-1)
Donor
max/average
(tr.min-1) (V)
(%)
FCN
FCN
5000
5000
0.86
/
3.17
/
33.0
/
1.07/0.95
/
1.9 x 10-4
4.4 x 10-5
SCN
SCN
4000
4000
0.90
0.73
2.51
5.35
34.9
35.7
0.94/0.82
1.67/1.64
FRh
FRh
3000
3000
0.79
0.79
1.90
1.94
32.5
32.3
0.58/0.53
0.59/0.56
1.0 x 10-4
1.8 x 10-4
SRh
SRh
5000
5000
0.74
0.65
1.19
3.08
21.0
31.3
0.22/0.20
0.75/0.71
Table 5: Photovoltaic characteristics of OSC before (first line) and after (second line) optimal annealing
treatment elaborated from FCN, SCN, FRh and SRH respectively.
Due to the isostructurality of conjugated systems, cells based on spiro and fluorene
derivatives exhibit similar open circuit voltages before annealing. As commonly admitted, the
Voc of an OPV device is strongly related to the relative HOMO level of donor and LUMO level
of acceptor materials. In this context, it is not surprising to observe a higher Voc with SCN and
FCN than SRh and FRh respectively. In addition, devices elaborated from fluorene derivatives
exhibit analogous FF but reveal higher Jsc than spiro ones in the absence of any thermal
treatment. As consequence, FCN provides the best PCE at 1.07% as expected.
As thermal treatment has no effect on, or leads to degradation of, fluorene based cells,
annealing allows doubling or tripling the PCE of SBF containing devices. Indeed, power
conversion efficiencies of SRh and SCN devices increase from averages of 0.20 and 0.82 to
0.71 and 1.64% respectively. These annealing are accompanied with a slight decrease of Voc
and, except for SRh, a relative stability of FF that remain around 30%. Thus, most
improvements seem imputable to Jsc increases, which reach a maximum improvement factor of
2.6 (from 1.19 to 3.08 mA.cm-2) in the case of SRh cells. As the measurement of the hole
mobilities by space charge limited current technic performed on all four compounds (table 5)

14
ACCEPTED MANUSCRIPT
did not reveal any clear trend between spiro derivatives and linear parents, this improvement of
PCE seems attributed to interface phenomenons.
In order to compare the respective contributions of derivatives on the PCE, the external
quantum efficiencies (EQE) of the best performing cells have been recorded and are presented
on Figure 3. Each photoresponse exhibits a first broad band extending from 311 to 420 nm
associated with the contributions of both fullerene C60 and donors followed by a second broad
band from 420 to 570 nm corresponding to the internal charge transfer bands of the donors. In
accordance with Jsc data, the EQE spectrum of SCN shows a first maximum of 76% at 356 nm
and a second maximum of 45% at 492 nm higher than for FCN (respectively at 55% and 33%).
Moreover, EQE spectra of FCN and SCN present edges bathochromically shifted compared to
rhodanine based devices, in agreement with the absorption properties of films.
2
80
60
40
20
0
0
-2
-4
-6
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
300
400
500
600
700
V (V)
λ (nm)
2
0
80
60
40
20
0
-2
-4
-6
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
300
400
500
600
700
V (V)
λ (nm)
Figure 3. J-V (left) and EQE (right) curves of the photovoltaic devices prepared using dicyanovinyl (top) or
rhodamine (bottom) terminated compounds incorporating either a spiro node (solid line) or a fluorene
(dashed line) as
π conjugated system after thermal treatment.

15
ACCEPTED MANUSCRIPT
In this context, the enlarged photoactivity of dicyanovinyl based derivatives combined with
higher EQE maxima for SBF compounds, coherent with spectroscopic data which indicate
higher extinction molar coefficients for the spiro compounds and lower band gap for SCN and
FCN explain the better performances of SCN which allows a PCE of 1.67% with the higher Jsc
in the series at 5.35 mA/cm2.
4. Conclusion
In conclusion, after synthesis, the properties of two spirobifluorene push-pull derivatives
have been compared with those of their fluorene linear analogues. Spectroscopic properties of
all compounds in solution are analogous but little decay are observed in cyclic voltammetry.
Indeed, in accordance with theoretical data, spiro compounds present bandgap analogous to
those of fluorene derivatives but lower HOMO and LUMO levels. Analogously, it should be
noted that dicyanovinyl terminated derivatives present lower HOMO and LUMO levels than
rhodanine compounds. Additionally, if the DFT calculations confirm the internal charge
transfer character of lower energy transitions, the TD-DFT calculations show the existence of
forbidden transitions due to the orthogonality of the π-systems.
After incorporation in planar heterojunction (PHJ) solar cells, best results are obtained with
SCN derivative which conjugates a higher and larger EQE spectrum due to better absorbance
properties and lower bandgap. These results unambiguously confirm the interest of short spiro
derivatives compared to analogous fluorene systems for OPV application. In fact, this study is
the first direct comparison between them performed on PHJ devices which is certainly one of
the most basic and reliable architecture OPV device. However, it must be pointed out that the
PCE registered here remain modest and that targeted compounds present limited stabilities on
stationary phases which complicate their purification. Thus, optimization aiming to increase
both stabilities and performances of chromophores are in due course.
Acknowledgments. Clément Dalinot thanks the University of Angers for granting. Authors
thank the Johnson Matthey Company for their generous providing of palladium salts (necessary
for the syntheses of starting aldehydes). P.L. and L.S. are grateful to Dr Clément Cabanetos for
fruitful and heady discussions but also the SCLC measurements.
References
1.
Roncali, J.; Leriche, P.; Cravino, A. Adv. Mater. 2007, 19 (16), 2045-2060.

16
ACCEPTED MANUSCRIPT
2.
3.
4.
5.
6.
7.
8.
9.
Roncali, J.; Leriche, P.; Blanchard, P. Adv. Mater. 2014, 26 (23), 3821-3838.
Thelakkat, M. Macromol. Mater. Eng. 2002, 287 (7), 442-461.
Luponosov, Y. N.; Solodukhin, A. N.; Ponomarenko, S. A. Polym. Sci. Ser. C 2014, 56 (1), 104-134.
Mahmood, A. Sol. Energy 2016, 123, 127-144.
Mishra, A.; Bäuerle, P. Angew. Chem. Int. Ed. 2012, 51 (9), 2020-2067.
Ripaud, E.; Rousseau, T.; Leriche, P.; Roncali, J. Adv. Energy Mater. 2011, 1 (4), 540-545.
Leliège, A.; Blanchard, P.; Rousseau, T.; Roncali, J. Org Lett 2011, 13 (12), 3098-3101.
Leliège, A.; Grolleau, J.; Allain, M.; Blanchard, P.; Demeter, D.; Rousseau, T.; Roncali, J. Chem-Eur
J 2013, 19 (30), 9948-9960.
10. Leriche, P.; Piron, F.; Ripaud, E.; Frère, P.; Allain, M.; Roncali, J. Tetrahedron Lett 2009, 50 (40),
5673-5676.
11. Roquet, S.; De Bettignies, R.; Leriche, P.; Cravino, A.; Roncali, J. J Mater Chem 2006, 16 (29),
3040-3045.
12. Jeux, V.; Dalinot, C.; Allain, M.; Sanguinet, L.; Leriche, P. Tetrahedron Lett 2015, 56 (11), 1383-
1387.
13. Yassin, A.; Mallet, R.; Leriche, P.; Roncali, J. ChemElectroChem 2014, 1 (7), 1219-1225.
14. Saragi, T. P. I.; Spehr, T.; Siebert, A.; Fuhrmann-Lieker, T.; Salbeck, J. Chem. Rev. 2007, 107 (4),
1011-1065.
15. Mauch, R. H.; Gumlich, H. E., Inorganic and organic electroluminescence. Wissenschaft und
Technik Verlag: 1996.
16. Pudzich, R.; Fuhrmann-Lieker, T.; Salbeck, J., Spiro Compounds for Organic Electroluminescence
and Related Applications. In Emissive Materials Nanomaterials, Springer Berlin Heidelberg: 2006; Vol.
199, pp 83-142.
17. Chan, C. Y.; Wong, Y. C.; Wong, H. L.; Chan, M. Y.; Wing-Wah Yam, V. J. Mater. Chem. C 2014, 2
(36), 7656-7665.
18. Liu, S. W.; Lin, C. F. J. Mater. Res. 2013, 28 (11), 1442-1448.
19. Bulut, I.; Chávez, P.; Fall, S.; Méry, S.; Heinrich, B.; Rault-Berthelot, J.; Poriel, C.; Lévêque, P.;
Leclerc, N. Rsc Adv 2016, 6 (31), 25952-25959.
20. Yan, Q.; Zhou, Y.; Zheng, Y.-Q.; Pei, J.; Zhao, D. Chem. Sci. 2013, 4 (12), 4389-4394.
21. Wang, J.; Dai, S.; Yao, Y.; Cheng, P.; Lin, Y.; Zhan, X. Dyes Pigments 2015, 123, 16-25.
22. Ma, S.; Fu, Y.; Ni, D.; Mao, J.; Xie, Z.; Tu, G. Chem Commun 2012, 48 (97), 11847-11849.
23. Dang, D.; Zhou, P.; Zhi, Y.; Bao, X.; Yang, R.; Meng, L.; Zhu, W. J. Mater. Sci. 2016, 51 (17), 8018-
8026.
24. Gritzner, G.; Kuta, J. Pure Appl Chem 1984, 56 (4), 461-466.
25. Cardona, C. M.; Li, W.; Kaifer, A. E.; Stockdale, D.; Bazan, G. C. Adv. Mater. 2011, 23 (20), 2367-
2371.
26. Karl, N. Synthetic Met 2003, 133-134, 649-657.
27. Singh, V.; Wang, S.; Kool, E. T. J Am Chem Soc 2013, 135 (16), 6184-6191.
28. Chiang, C.-L.; Shu, C.-F.; Chen, C.-T. Org Lett 2005, 7 (17), 3717-3720.
29. Morales, A. R.; Frazer, A.; Woodward, A. W.; Ahn-White, H. Y.; Fonari, A.; Tongwa, P.; Timofeeva,
T.; Belfield, K. D. J Org Chem 2013, 78 (3), 1014-1025.
30. Thiery, S.; Tondelier, D.; Declairieux, C.; Seo, G.; Geffroy, B.; Jeannin, O.; Rault-Berthelot, J.;
Metivier, R.; Poriel, C. J. Mater. Chem. C 2014, 2 (21), 4156-4166.
31. Poriel, C.; Liang, J.-J.; Rault-Berthelot, J.; Barriere, F.; Cocherel, N.; Slawin, A. M. Z.; Horhant, D.;
Virboul, M.; Alcaraz, G.; Audebrand, N.; Vignau, L.; Huby, N.; Wantz, G.; Hirsch, L. Chem-Eur J 2007, 13
(36), 10055-10069.
32. Luponosov, Y. N.; Min, J.; Solodukhin, A. N.; Kozlov, O. V.; Obrezkova, M. A.; Peregudova, S. M.;
Ameri, T.; Chvalun, S. N.; Pshenichnikov, M. S.; Brabec, C. J.; Ponomarenko, S. A. Org Electron 2016, 32,
157-168.
33. Chiang, C. L.; Tseng, S. M.; Chen, C. T.; Hsu, C. P.; Shu, C. F. Adv Funct Mater 2008, 18 (2), 248-
257.
34. Kasha, M.; Rawls, H. R.; El-Bayoumi, M. A. Pure Appl Chem 1965, 11 (3-4), 371-392.

17
ACCEPTED MANUSCRIPT
35. Brabec, C. J.; Cravino, A.; Meissner, D.; Serdar Sariciftci, N.; Fromherz, T.; Rispens, M. T.;
Sanchez, L.; Hummelen, J. C. Adv Funct Mater 2001, 11 (5), 374-380.

18
ACCEPTED MANUSCRIPT
Spirobifluorene based small push-pull molecules for organic photovoltaïc applications.
Clément Dalinot, György Szalóki, Chloé Dindault, Olivier Segut, Lionel Sanguinet* and Philippe
Leriche*
MOLTECH – Anjou UMR CNRS 6200, Groupe systèmes conjugués linéaires, University of Angers, 2 boulevard
Lavoisier, 49045 Angers, France
philippe.leriche@univ-angers.fr, 33241735010; lionel.sanguinet@univ-angers.fr, 33241735374
Supporting information
Figure S1: Geometries and electron distributions of all four chromophores.

19
ACCEPTED MANUSCRIPT
Figure S2: Cyclic Voltammetry of SCN (solid line) and FCN (dotted line) in
dichloromethane (1 and 3 10-5M) with TBAP (0.1M) on Pt at 100mV.s-1.
6
4
2
0
-2
-4
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
E (V vs Fc/Fc+)

ACCEPTED MANUSCRIPT
Highlights.
Spirobifluorene based chromophores
Comparison of 3D and linear semi-conductors for OPV
Donor-Pi-Acceptor compounds
Chromophores for organic solar cells