Preparation and redox properties of fluorinated 1,3-diphenylisobenzofurans

Article abstract of DOI:10.1016/j.electacta.2019.134659

Eleven fluorinated 1,3-diphenylisobenzofurans, materials that are promising to perform efficient singlet fission, have been synthesized and their electrochemical reduction and oxidation in non-aqueous dimethylformamide examined. The redox properties are n

Full text of DOI:10.1016/j.electacta.2019.134659

Journal Pre-proof
Preparation and redox properties of fluorinated 1,3-diphenylisobenzofurans
Jiří Kaleta, Ludmila Šimková, Alan Liška, Daniel Bím, Jenica M.L. Madridejos, Radek
Pohl, Lubomír Rulíšek, Josef Michl, Jiří Ludvík
PII:
S0013-4686(19)31527-0
DOI:
https://doi.org/10.1016/j.electacta.2019.134659
EA 134659
Reference:
To appear in: Electrochimica Acta
Received Date: 13 June 2019
Revised Date: 1 August 2019
Accepted Date: 8 August 2019
Please cite this article as: Jiří. Kaleta, L. Šimková, A. Liška, D. Bím, J.M.L. Madridejos, R.
Pohl, Lubomí. Rulíšek, J. Michl, Jiří. Ludvík, Preparation and redox properties of fluorinated
1,3-diphenylisobenzofurans, Electrochimica Acta (2019), doi: https://doi.org/10.1016/
j.electacta.2019.134659.
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.
© 2019 Published by Elsevier Ltd.

Preparation and Redox Properties of Fluorinated 1,3-Diphenylisobenzofurans
c
Jiří Kaleta a,b, Ludmila Šimková , Alan Liška c, Daniel Bím a,c, Jenica M. L. Madridejos a,
Radek Pohl a, Lubomír Rulíšek a, Josef Michl a,b and Jiří Ludvík c*
a Institute of Organic Chemistry and Biochemistry, Czech Academy of Sciences, Flemingovo nám.
2, 16610 Prague 6, Czech Republic, jiri.kaleta@uochb.cas.cz
b Department of Chemistry and Biochemistry, University of Colorado, Boulder, CO 80309-0215,
United States, josef.michl@colorado.edu
c J. Heyrovský Institute of Physical Chemistry, Czech Academy of Sciences, Dolejškova 2155/3,
182 23 Prague 8, Czech Republic. jiri.ludvik@jh-inst.cas.cz ; *corresponding author
ABSTRACT
Eleven fluorinated 1,3-diphenylisobenzofurans, materials that are promising to perform efficient
singlet fission, have been synthesized and their electrochemical reduction and oxidation in non-
aqueous dimethylformamide examined. The redox properties are not changed continuously with
increasing number of fluorine substituents but there are some qualitative turns in behavior for
specific types of substitution revealing a discontinuity requiring deeper investigation. The DFT
calculations confirm the trends observed experimentally by correlation of experimental and
calculated data. The differences between the geometries of the neutral species and their radical ions
were computed and rationalized.
KEYWORDS
Fluorinated 1,3-Diphenylisobenzofurans; singlet fission; synthesis; oxidation/reduction; DFT
calculations.
1
INTRODUCTION
1,3-Diphenylisobenzofuran (1) is one of the few materials that are known to perform highly
efficient singlet fission,1,2 a photophysical process in which a singlet exciton is converted into two
triplet excitons. Singlet fission has attracted increasing attention because it promises3 to increase
the maximum theoretical efficiency of inexpensive single-junction solar cells beyond the Shockley-
Queisser limit4 of 1/3. So far, it has not found its way into practice because none of the materials in
which it proceeds efficiently are sufficiently sturdy. This is certainly true of 1, which is very
1

sensitive to a combination of light and air. Quantum theory has been quite helpful in efforts to
guide the search for more efficient materials for singlet fission use, both when it comes to the
choice of chromophores and when it comes to their mode of packing.
The most important criterion in the selection of chromophores, E(S1) ≥ 2 E(T1), assures that
the singlet fission process is isoergic or even better, somewhat exoergic. This criterion is rarely
fulfilled. A class of compounds that are likely to meet the requirement are biradicaloids, species
formally derived from perfect biradicals by sufficient but not excessive covalent perturbation.5
This was the consideration that led to the original choice of 1 as a molecule suitable for singlet
fission.6
The critical importance of molecular packing was underscored upon closer examination of
1. This material exists in two crystalline forms. While one of them performs singlet fission very
efficiently, the other performs it hardly at all.7 General theoretical guidance toward optimization of
crystal packing has been formulated,8,9,10 but is still in early stages.
One way to test the theoretical notions concerning crystal packing that is optimal for singlet
fission would be to study singlet fission on additional chromophores11,12 that crystallize in two or
more crystal forms. We have decided to try another approach and to take advantage of the fact that
certain substituents have little effect on electronic excitation energies and presumably other
properties relevant to singlet fission, yet are capable of modifying crystal packing dramatically.
This should provide relatively easy access to a larger pool of data. Starting with 1, we have
examined the effect of alkylation, but found that it tends to produce amorphous materials.13 In
order to investigate the effect of fluorination, which also does not modify the electronic spectra of
aromatics much but tends to affect crystal packing strongly,14,15,16 we needed to gain access to a fair
number of fluorinated derivatives of 1, few of which are known.
2

Figure 1. Structures of 1 and its partially fluorinated derivatives 2 - 12.
Presently, we report the preparation and characterization of eleven new compounds of this
type (Figure 1). Since redox properties of the chromophores play a significant role in understanding
singlet fission due to the usually dominant mechanistic importance of virtual charge-separated
states in the process, we have primarily examined their electrochemical reduction and oxidation
using various electrode materials in non-aqueous DMF, with focus on the potentials of the first
uptake and loss of an electron. The experimental data were compared with those of density
functional theory (DFT) calculations. Besides the redox characterization, the other task for the
electrochemical treatment of these polyfluorinated derivatives was to solve the question, whether
the redox properties are changed continuously with increasing number of fluorine substituents or if
there are some qualitative turns in behavior for specific types of substitution which would reveal a
discontinuity relevant to the desired properties and requiring deeper investigation.
2
EXPERIMENTAL
2.1
Electrochemistry
DC-polarography and cyclic voltammetry were carried out in N,N-dimethylformamide
(DMF, purified and dried by double distillation17) containing 0.1 M tetrabutylammonium
3

hexafluorophosphate (TBAPF6) as the supporting electrolyte and 1×10-3 M solutions of 1 - 12
deoxygenated with argon. A dropping mercury electrode (DME) with a controlled drop time (1 s)
was used in DC-polarography. A platinum, glassy carbon disk (GC, area ca. 1 mm2), boron doped
diamond (BDDE) stationary electrode, or a hanging mercury drop electrode (HMDE) were used for
cyclic voltammetry, measured at four different scan rates (50, 100, 200 and 500 mV.s–1). The
reduction potentials obtained at the four different electrodes differed by less than the experimental
error (±10 mV). The oxidation peaks were broader and their positions again agreed within the
experimental uncertainty of ±20 mV on three of the electrodes. The values obtained with the GC
electrode were systematically less positive by about 30 mV and are presumably the closest to the
thermodynamic values.
The auxiliary electrode was made of a platinum wire and a saturated calomel electrode
(SCE) separated from aprotic solution by a salt bridge served as the reference electrode. All
experiments were carried out in an undivided 10 mL cell and were conducted by the analog
potentiostat PA4 with an XY recorder, both Laboratorní přístroje Praha and by the computer driven
digital potentiostat PGSTAT101 (Autolab-Metrohm) using software NOVA.
2.2
Calculations
Standard density functional theory calculations were performed using the Turbomole 7.1
program.18 First, all structures were optimized at the RI-BP86-D3/def-TZVP level of
theory19,20,21,22,23 using implicit solvation conductor-like screening model (COSMO)24 with εr= 37.5,
corresponding to the dielectric constant of DMF. After each geometry optimization, single-point
calculations using RI-B3LYP-D3/def2-TZVPD25,26 and RI-PBE-D3/def2-TZVPD27 methods were
carried out to obtain in vacuo Eel energies as defined in Equation 2. The solvation free energy
difference Gsolv(T) was calculated using the conductor-like screening model for real solvents
(COSMO-RS) with the radii-based isosurface cavity ($cosmo_isorad keyword) and the
COSMOtherm parameter set BP_TZVPD_FINE_C30_1701.ctd as available in COSMOtherm17.28
The COSMO-RS calculations were carried out following the recommended protocol: the RI-BP86-
D3/def2-TZVPD/COSMO(εr = ∞) ≡ ECOSMO,∞ and RI-BP86-D3/def2-TZVPD/in vacuo ≡ Ein vacuo
single-point calculations (on top of geometry optimizations with COSMO) are used to calculate
Gsolv(T) = (ECOSMO,∞ - Ein vacuo) + ꢀCOSMO-RS(T), where ꢀ(T) is the temperature-dependent COSMO-
RS chemical potential.29 The last term in Equation 2, ꢀ(EZPVE + pV - RT lnQ), was obtained from
frequency calculations with the rigid-rotor/harmonic-oscillator approximation (for p = 1 bar),
considering low-frequency vibration modes (≤100 cm-1) to contribute to Q as hindered rotors.30 For
frequency calculations, the gas-phase optimized structures at RI-BP86-D3/def-TZVP level of
4

theory were used.
Molecular orbitals were calculated at the RI-B3LYP-D3/def2-TZVPD level of theory.
2.3
X-ray Diffraction
Crystallographic data for 21 were collected on Bruker D8 VENTURE Kappa Duo
PHOTON100 by IꢁS micro-focus sealed tube with CuKα radiation (λ = 1.54178 Å) at 150(2) K.
The structures were solved by direct methods (XT)31 and refined by full matrix least squares based
on F2 (SHELXL2014).32 The hydrogen atoms on carbon atoms were fixed into idealized positions
(riding model) and assigned temperature factors of Hiso(H) = 1.2 Ueq(pivot atom).
2.4
Materials
All reactions were carried out under argon atmosphere with dry solvents freshly distilled
under anhydrous conditions, unless otherwise noted. Standard Schlenk and vacuum line techniques
were employed for all manipulations of air- or moisture-sensitive compounds. Yields refer to
isolated, chromatographically and spectroscopically homogenous materials, unless otherwise stated.
THF and ether were dried over sodium with benzophenone and distilled under argon prior to
use. All other reagents were used as supplied unless otherwise stated.
2.5
Procedures
Analytical thin-layer chromatography (TLC) was performed using precoated TLC
aluminum sheets (Silica gel 60 F254). TLC spots were visualized using either UV light (254 nm) or
a 5% solution of phosphomolybdic acid in ethanol, and heat (200 °C) as a developing agent. Flash
chromatography was performed using silica gel (high purity grade, pore size 60 Å, 70 - 230 mesh).
Melting points are reported uncorrected. Infrared spectra (IR) were recorded in KBr pellets.
Chemical shifts in 1H, and 13C spectra are reported in ppm on the δ scale relative to CHCl3 (δ = 7.26
ppm for 1H NMR), and CHCl3 (δ = 77.0 ppm for 13C NMR) as internal references. Splitting
patterns are assigned s = singlet, d = doublet, t = triplet, m = multiplet, br = broad signal.
2.6
Syntheses
1-(4-Fluorophenyl)-3-phenylisobenzofuran (2). A previously published procedure14 was
adapted as follows: An argon filled 50 mL two-necked flask equipped with stir bar and a gas
condenser was charged with magnesium turnings (535 mg, 22.000 mmol). After three successive
vacuum/argon cycles, ether (40 mL) and one grain of iodine were added, and a brownish
suspension was stirred for 15 min at room temperature. Subsequently 1-bromo-4-fluorobenzene
(2.35 mL, 21.404 mmol) was added dropwise. An exothermic reaction was followed by changes of
5

the color of the reaction mixture from original brown, through colorless to final grayish. This
suspension was stirred 120 min at room temperature and then 3-phenylisobenzofuran-1(3H)-one
(19) (3.000 g, 14.269 mmol) was added and stirring was continued for 16 h. A dense grayish solid
precipitated. Subsequently trifluoroacetic acid (2 mL) was added dropwise and the reaction
mixture was stirred for additional 15 min. The black reaction mixture was diluted with ether (50
mL), washed with a saturated aqueous solution of NaHCO3 (3 × 30 mL), and dried over MgSO4.
Solvent was removed under reduced pressured and column chromatography on silica gel
(hexane/ethyl acetate - 4:1) gave 2 as an yellow crystalline solid (1.260 g, 4.370 mmol, 31%).
1
Mp 113.5 - 114.2 °C. H NMR (500 MHz, CDCl3): δ 6.99 - 7.04 (m, 2H), 7.17 - 7.20 (m,
2H), 7.29 - 7.32 (m, 1H), 7.48 - 7.51 (m, 2H), 7.76 (m, 1H), 7.83 (m, 1H), 7.90 (m, 2H), 7.93 (m,
2H). 13C {1H} NMR (125 MHz, CDCl3): δ 116.0 (d, JC,F = 21.9 Hz), 119.8, 120.2, 121.7 (d, JC,F
=
1.3 Hz), 122.0, 124.7, 125.1, 125.2, 126.4 (d, JC,F = 7.9 Hz), 126.9, 128.0 (d, JC,F = 3.4 Hz), 128.9,
131.5, 142.9 (d, JC,F = 1.2 Hz), 143.7 (d, JC,F = 1.9 Hz), 161.7 (d, JC,F = 247.7 Hz). 19F NMR (470
MHz, CDCl3): δ -110.73 (tt, 1JF,H = 8.6 Hz, 2JF,H = 5.4 Hz, 1F). IR (KBr): 3053, 1662, 1627, 1597,
1573, 1547, 1511, 1490, 1464, 1447, 1438, 1404, 1367, 1296, 1282, 1225, 1203, 1153, 1130, 1095,
1071, 1028, 1013, 993, 980, 937, 908, 833, 813, 765, 750, 742, 721, 711, 691, 659, 642, 632, 617,
608, 575, 552, 511, 487 cm-1. MS, m/z (%): 288.1 (100, M), 270.1 (6), 259.1 (43), 239.1 (12),
183.1 (12), 165.1 (14), 144.0 (13), 95.0 (16), 77.0 (9). HRMS, (EI) for (C20H13FO+): calcd
288.0950, found 288.0947. Anal. Calcd. for C20H13FO: C, 83.32; H, 4.54. Found: C, 83.13; H,
4.65.
General Procedure for Synthesis of Asymmetric Fluorinated DPIBF Derivatives
(GP1). An argon filled 50 mL two-necked flask equipped with a stir bar and a gas condenser was
charged with magnesium turnings (129 mg, 5.300 mmol). After three successive vacuum/argon
cycles, THF (15 mL) and one grain of iodine were added, and the brownish suspension was stirred
for 15 min at room temperature. Subsequently fluorinated bromobenzene (5.200 mmol) was added.
An exothermic reaction was followed by changes of the color of the reaction mixture from original
brown, through colorless to final grayish. This suspension was stirred 2 h at room temperature and
then 3-phenylisobenzofuran-1(3H)-one (19) (1.000 g, 4.757 mmol) was added and stirring
continued for 16 h. Then Ac2O (4 mL) was added dropwise and the reaction mixture was refluxed
at the temperature of the oil bath 70 °C for additional 30 min. A dense yellowish precipitate was
formed. The yellow suspension was cooled to room temperature, diluted with ether (100 mL), and
washed with a saturated aqueous solution of NaHCO3 (3 × 15 mL). The deep yellow organic phase
was dried over MgSO4. Solvents were removed under reduced pressure and column
chromatography on silica gel (hexane/ethyl-acetate - 4:1) afforded products as yellow crystalline
6

solids.
1-(3-Fluorophenyl)-3-phenylisobenzofuran (3) was synthesized from 1-bromo-3-
fluorobenzene (14) (581 ꢁL, 5.200 mmol), magnesium (129 mg, 5.300 mmol), and 19 (1.000 g,
4.757 mmol) in THF (15 mL) according to GP1. Compound 3 was obtained as a bright yellow
crystalline solid (1.216 g, 4.218 mmol, 89%).
1
Mp 130.9 - 132.0 °C. H NMR (400 MHz, CDCl3): δ 6.94 - 7.01 (m, 1H), 7.01 - 7.10 (m,
2H), 7.29 - 7.35 (m, 1H), 7.40 - 7.47 (m, 1H), 7.47 - 7.53 (m, 2H), 7.60 - 7.65 (m, 1H), 7.71 - 7.74
(m, 1H), 7.80 - 7.87 (m, 2H), 7.93 - 7.97 (m, 2H). 13C {1H} NMR (125 MHz, CDCl3): δ 111.3 (d,
JC,F = 23.6 Hz), 113.5 (d, JC,F = 21.5 Hz), 119.8, 120.2 (d, JC,F = 2.8 Hz), 120.3, 122.1, 122.7,
125.0, 125.2, 125.8, 127.2, 129.0, 130.5 (d, JC,F = 8.6 Hz), 131.3, 133.5 (d, JC,F = 8.6 Hz), 142.3 (d,
JC,F = 3.2 Hz), 144.4, 163.3 (d, JC,F = 245.1 Hz). 19F NMR (470 MHz, CDCl3): δ -108.92 (ddd,
1JF,H = 10.3 Hz, 2JF,H = 8.6 Hz, 3JF,H = 6.0 Hz, 1F). IR (KBr): 3051, 3039, 1625, 1608, 1579, 1543,
1518, 1503, 1491, 1458, 1451, 1443, 1414, 1397, 1315, 1274, 1259, 1213, 1194, 1160, 1141, 1111,
1096, 1067, 1028, 1007, 994, 963, 951, 902, 877, 860, 854, 838, 829, 773, 762, 747, 744, 736, 723,
678, 665, 659, 599, 554, 523, 490 cm-1. MS, m/z (%): 288.1 (100, M), 270.1 (5), 259.1 (23), 239.1
(11), 183.1 (10), 165.1 (9), 144.0 (10). HRMS, (EI) for (C20H13FO+): calcd 288.0950, found
288.0948. Anal. Calcd. for C20H13FO: C, 83.32; H, 4.54. Found: C, 83.01; H, 4.45.
1-(3,5-Difluorophenyl)-3-phenylisobenzofuran (4) was synthesized from 1-bromo-3,5-
difluorobenzene (15) (599 ꢁL, 5.200 mmol), magnesium (129 mg, 5.300 mmol), and 19 (1.000 g,
4.757 mmol) in THF (15 mL) according to GP1. Compound 4 was obtained as a bright yellow
crystalline solid (1.033 g, 2.867 mmol, 60%).
1
Mp 118.2 - 119.5 °C. H NMR (400 MHz, CDCl3): δ 6.67 - 6.73 (m, 1H), 7.02 - 7.11 (m,
2H), 7.32 - 7.44 (m, 3H), 7.48 - 7.53 (m, 2H), 7.73 - 7.76 (m, 1H), 7.82 - 7.86 (m, 1H), 7.91 - 7.94
(m, 2H). 13C {1H} NMR (125 MHz, CDCl3): δ 101.7 (t, JC,F = 25.8 Hz), 106.9, 119.4, 120.5, 122.2,
123.3, 125.1, 125.3, 126.3, 127.6, 129.0, 131.0, 134.1 (t, JC,F = 10.7 Hz), 141.2 (d, JC,F = 3.6 Hz),
145.0, 163.5 (dd, 1JC,F = 247.2 Hz, 2JC,F = 13.5 Hz). 19F NMR (470 MHz, CDCl3): δ -105.76 (m,
2F). IR (KBr): 3081, 3025, 1621, 1589, 1547, 1504, 1483, 1453, 1443, 1403, 1327, 1280, 1221,
1195, 1184, 1142, 1115, 1073, 1004, 986, 952, 903, 852, 840, 835, 821, 760, 744, 734, 704, 685,
671, 618, 599, 532, 511, 504 cm-1. MS, m/z (%): 306.1 (100), 277.1 (21), 257.1 (12), 229.0 (10),
201.1 (11), 165.1 (10), 153.0 (9). HRMS, (EI) for (C20H12F2O+): calcd 306.0856, found 306.0854.
Anal. Calcd. for C20H12F2O: C, 78.42; H, 3.95. Found: C, 78.12; H, 3.86.
1-Phenyl-3-(3,4,5-trifluorophenyl)isobenzofuran (5) was synthesized from 5-bromo-
1,2,3-trifluorobenzene (16) (621 ꢁL, 5.200 mmol), magnesium (129 mg, 5.300 mmol), and 19
7

(1.000 g, 4.757 mmol) in THF (15 mL) according to GP1. Compound 5 was obtained as a bright
yellow crystalline solid (1.078 g, 3.324 mmol, 70%).
1
Mp 129.5 - 131.0 °C. H NMR (400 MHz, CDCl3): δ 7.02 - 7.12 (m, 2H), 7.32 - 7.37 (m,
1H), 7.45 - 7.53 (m, 4H), 7.68 - 7.71 (m, 1H), 7.82 - 7.86 (m, 1H), 7.89 - 7.93 (m, 2H). 13C {1H}
NMR (125 MHz, CDCl3): δ 108.2 (m), 119.1, 120.5, 122.2, 122.9 (d, JC,F = 1.3 Hz), 125.1, 125.4,
126.5, 127.5 (td, 1JC,F = 9.1 Hz, 2JC,F = 4.7 Hz), 127.6, 129.1, 131.0, 138.4 (dt, 1JC,F = 252.5 Hz,
2JC,F = 15.6 Hz), 140.3 (d, JC,F = 2.8 Hz), 145.1, 151.8 (ddd, 1JC,F = 249.9 Hz, 2JC,F = 10.4 Hz, 3JC,F
= 4.5 Hz). 19F NMR (470 MHz, CDCl3): δ -158.89 (tt, JF,F = 20.4 Hz, JF,H = 6.4 Hz, 1F); -130.19
(dd, JF,F = 20.4 Hz, JF,H = 8.7 Hz, 2F). IR (KBr): 3092, 3055, 1630, 1615, 1590, 1551, 1524, 1500,
1451, 1440, 1434, 1412, 1382, 1338, 1309, 1291, 1282, 1255, 1227, 1211, 1198, 1184, 1100, 1073,
1044, 1035, 1014, 951, 903, 866, 859, 855, 844, 833, 822, 762, 743, 699, 687, 664, 656, 611, 590,
563, 556, 508 cm-1. MS, m/z (%): 324.1 (100, M), 306.1 (5), 295.1 (20), 275.1 (13), 247.0 (5),
162.0 (8). HRMS, (EI) for (C20H11F3O+): calcd 324.0762, found 324.0765. Anal. Calcd. for
C20H11F3O: C, 74.07; H, 3.42. Found: C, 73.83; H, 3.34.
1-Phenyl-3-(2,4,6-trifluorophenyl)isobenzofuran (6). An argon filled 50 mL two-necked
flask equipped with stir bar and a gas condenser was charged with magnesium turnings (535 mg,
22.000 mmol). After three successive vacuum/argon cycles, ether (40 mL) and one grain of iodine
were added, and the brownish suspension was stirred for 15 min at room temperature.
Subsequently 1-bromo-2,4,6-trifluorobenzene (2.52 mL, 21.404 mmol) was added dropwise. The
reaction mixture was stirred for 16 h at room temperature. Then 3-phenylisobenzofuran-1(3H)-one
(19) (3.000 g, 14.269 mmol) was added and stirring continued for 5 h. A brownish oily phase was
formed. Subsequently trifluoroacetic acid (2 mL) was added dropwise and the reaction mixture was
stirred for additional 15 min. The black reaction mixture was diluted with ether (50 mL), washed
with a saturated aqueous solution of NaHCO3 (3 × 30 mL), and dried over MgSO4. Solvent was
removed under reduced pressured and column chromatography on silica gel (hexane/ethyl acetate -
4:1) provided 6 as an yellow crystalline solid (460 mg, 1.418 mmol, 10%).
1
Mp 112.0 - 113.1 °C. H NMR (500 MHz, CDCl3): δ 6.84 - 6.87 (m, 2H), 7.02 - 7.06 (m,
2H), 7.31 - 7.34 (m, 1H), 7.41 - 7.44 (m, 1H), 7.48 - 7.51 (m, 2H), 7.85 - 7.87 (m, 1H), 7.94 - 7.95
(m, 2H). 13C {1H} NMR (125 MHz, CDCl3): δ 101.0 (m), 106.0 (td, 1JC,F = 18.7 Hz, 2JC,F = 5.1
Hz), 119.8 (t, JC,F = 4.2 Hz), 119.9, 121.1, 125.00, 125.04, 125.1, 125.27, 127.31, 128.9, 131.4,
132.3, 146.2, 160.0 (ddd, 1JC,F = 253.9 Hz, 2JC,F = 14.8 Hz, 3JC,F = 9.5 Hz), 162.5 (dt, 1JC,F = 251.8
Hz, 2JC,F = 15.4 Hz). 19F NMR (470 MHz, CDCl3): δ -103.81 (m, 2F), -103.76 (m, 1F). IR (KBr):
3091, 3057, 1639, 1595, 1577, 1508, 1495, 1451, 1442, 1400, 1352, 1266, 1208, 1177, 1161, 1126,
8

1112, 1092, 1068, 1054, 1004, 994, 978, 897, 839, 760, 744, 736, 729, 729, 718, 692, 681, 664,
644, 635, 618, 608, 581, 575, 560, 540, 514 cm-1. MS, m/z (%): 324.1 (100, M), 295.1 (41), 275.1
(29), 193.1 (6), 165.1 (14), 162.0 (11), 137.5 (6), 105.0 (5), 77.0 (14). HRMS, (EI) for
(C20H11F3O+): calcd 324.0762, found 324.0761. Anal. Calcd. for C20H11F3O: C, 74.07; H, 3.42.
Found: C, 74.24; H, 3.61.
1-(Perfluorophenyl)-3-phenylisobenzofuran (7). An argon filled 50 mL two-necked flask
equipped with stir bar and a gas condenser was charged with magnesium turnings (535 mg, 22.000
mmol). After three successive vacuum/argon cycles, ether (40 mL) and one grain of iodine were
added, and brownish suspension was stirred for 15 min at room temperature. Subsequently 1-
bromopentafluorobenzene (2.67 mL, 21.404 mmol) was added dropwise during 15 min. An
exothermic reaction was followed by changes of the color of the reaction mixture from original
brown, through colorless to final grayish. This suspension was stirred 60 min at room temperature
and then the 3-phenylisobenzofuran-1(3H)-one (19) (3.000 g, 14.269 mmol) was added and stirring
continued for 16 h. Subsequently Ac2O (4 mL) was added dropwise and the reaction mixture was
refluxed at the temperature of the oil bath 70 °C for additional 30 min. A dense yellowish
precipitate was formed. Then a saturated and degassed aqueous solution of NaHCO3 (15 mL) was
added from a syringe and the aqueous phase was carefully removed with a syringe. The precipitate
was filtered off on frit, washed with hexane (2 × 10 mL) and thoroughly dried under reduced
pressure (600 mTorr, 25 °C). Compound 7 was obtained as a yellow crystalline solid (2.748 g,
7.627 mmol, 53%).
1
Mp 167.8 - 169.3 °C (dec.). H NMR (400 MHz, CDCl3): δ 7.06 - 7.11 (m, 2H), 7.34 - 7.38
(m, 1H), 7.41 - 7.44 (m, 1H), 7.49 - 7.53 (m, 2H), 7.87 - 7.90 (m, 1H), 7.93 - 7.96 (m, 2H). 13C
{1H} NMR (100 MHz, CDCl3): δ 106.9 (td, 1JC,F = 16.5 Hz, 2JC,F = 4.0 Hz), 119.4 (t, JC,F = 4.7 Hz),
120.1, 121.2, 125.3, 126.1, 126.2, 127.9, 129.0, 130.0, 130.9, 138.1 (m), 140.5 (m), 143.6 (m),
147.6. 19F NMR (376 MHz, CDCl3): δ -285.67 - -285.53 (m, 2F), -278.87 - -287.76 (m, 1F), -
263.83 - -263.75 (m, 2F). IR (KBr): 3092, 3064, 3024, 1685, 1632, 1599, 1521, 1496, 1451, 1421,
1380, 1368, 1323, 1199, 1185, 1166, 1159, 1143, 1130, 1103, 1072, 1033, 1021, 987, 948, 907,
828, 784, 764, 742, 724, 701, 688, 660, 636, 579, 552, 488 cm-1. MS, m/z (%): 360.1 (100, M),
340.0 (10), 331.1 (14), 311.0 (21), 293.1 (12), 255.0 (8), 180.0 (5), 165.1 (6). HRMS, (EI) for
+
(C20H9OF5 ): calcd 360.0574, found 360.0573. Anal. Calcd. for C20H9OF5: C, 66.67; H, 2.52.
Found: C, 66.50; H, 2.52.
General Procedure for Synthesis of Symmetric Fluorinated Derivatives of 1 (GP2).
An argon filled 50 mL two-necked flask equipped with stir bar and a gas condenser was
9

charged with magnesium turnings (165 mg, 6.800 mmol). After three successive vacuum/argon
cycles, THF (15 mL) and one grain of iodine were added, and brownish suspension was stirred for
15 min at room temperature. Subsequently fluorinated bromobenzene (6.700 mmol) was added.
An exothermic reaction was followed by changes of the color of the reaction mixture from original
brown, through colorless to final grayish. This suspension was stirred 2 h at room temperature and
then cooled to -50 °C. A white solid precipitated. Subsequently, methyl 2-formylbenzoate (20)
(500 mg, 3.046 mmol) was added into vigorously stirred suspension. Cooling was stopped and the
reaction mixture was slowly allowed to reach room temperature and then stirred for additional 16 h.
The suspension dissolved at ca -5 °C leaving dark red solution, which then slowly changed color to
orange and the dense yellowish solid slowly precipitated during subsequent stirring. Then Ac2O (4
mL) was added dropwise (the suspension dissolved leaving a clear yellow/orange solution) and the
reaction mixture was refluxed at the temperature of the oil bath 70 °C for additional 30 min. A
dense yellowish precipitate was formed again. Yellow suspension was cooled to room temperature,
diluted with ether (100 mL), and washed with a saturated aqueous solution of NaHCO3 (3 × 15
mL). Deep yellow organic phase was dried over MgSO4. Solvents were removed under reduced
pressure and column chromatography on silica gel (hexane/ethyl-acetate - 4:1) afforded products as
yellow crystalline solids.
1,3-Bis(4-fluorophenyl)isobenzofuran (8)14 was synthesized from 1-bromo-4-
fluorobenzene (13) (736 ꢁL, 6.700 mmol), magnesium (165 mg, 6.800 mmol), and 20 (500 mg,
3.046 mmol) in THF (15 mL) according to GP2. Compound 8 was obtained as a bright yellow
crystalline solid (419 mg, 1.368 mmol, 45%).
1
Mp 138.9 - 139.4 °C. H NMR (400 MHz, CDCl3): δ 7.00 - 7.03 (m, 2H), 7.16 - 7.20 (m,
4H), 7.74 - 7.76 (m, 2H), 7.86 - 7.89 (m, 4H). 13C {1H} NMR (100 MHz, CDCl3): δ 116.0 (d, JC,F
= 22.0 Hz), 119.8, 121.6, 125.3, 126.4 (d, JC,F = 7.9 Hz), 127.9 (d, JC,F = 3.4 Hz), 142.9, 161.7 (d,
JC,F = 247.8 Hz). 19F NMR (376 MHz, CDCl3): δ -238.11 (m, 2F). IR (KBr): 3055, 1628, 1598,
1550, 1514, 1504, 1443, 1410, 1402, 1299, 1291, 1229, 1223, 1201, 1152, 1098, 1092, 1001, 974,
935, 824, 799, 748, 744, 733, 712, 705, 695, 653, 636, 603, 558, 550, 499 cm-1. MS, m/z (%):
306.1 (100, M), 277.1 (50), 257.1 (20), 211.1 (5), 183.1 (14), 153.0 (12), 128.5 (7), 95.0 (12), 75.0
+
(11). HRMS, (EI) for (C20H12F2O2 ): calcd 306.0856, found 306.0857. Anal. Calcd. for
C20H12F2O2: C, 78.42; H, 3.95. Found: C, 78.35; H, 4.08.
1,3-Bis(3-fluorophenyl)isobenzofuran (9) was synthesized from 1-bromo-3-fluorobenzene
(14) (748 ꢁL, 6.700 mmol), magnesium (165 mg, 6.800 mmol), and 20 (500 mg, 3.046 mmol) in
THF (15 mL) according to GP2. Compound 9 was obtained as a bright yellow crystalline solid
10

(637 mg, 2.080 mmol, 68%).
1
Mp 139.8 - 141.2 °C. H NMR (500 MHz, CDCl3): δ 7.00 (m, 2H), 7.04 – 7.09 (m, 2H),
7.44 (m, 2H), 7.60 (m, 2H), 7.70 (m, 2H), 7.78 – 7.83 (m, 2H). 13C {1H} NMR (125 MHz, CDCl3):
δ 111.5 (d, JC,F = 23.7 Hz), 113.9 (d, JC,F = 21.5 Hz), 119.9, 120.4 (d, JC,F = 2.8 Hz), 122.7, 125.8,
130.5 (d, JC,F = 8.6 Hz), 133.2 (d, JC,F = 8.6 Hz), 142.9 (d, JC,F = 3.2 Hz), 163.2 (d, JC,F = 245.4 Hz).
19F NMR (470 MHz, CDCl3): δ -108.73 (ddd, 1JF,H = 10.3 Hz, 2JF,H = 8.4 Hz, 3JF,H = 6.0 Hz, 2F).
IR (KBr): 3084, 3033, 1625, 1610, 1581, 1541, 1502, 1487, 1464, 1451, 1270, 1261, 1198, 1192,
1171, 1159, 1108, 1067, 998, 995, 879, 869, 856, 850, 830, 778, 774, 750, 743, 684, 562, 527 cm-1.
MS, m/z (%): 306.1 (100, M), 288.1 (76), 277.1 (21), 257.1 (20), 239.1 (10), 211.1 (11), 183.1 (19),
153.0 (10), 144.0 (7). HRMS, (EI) for (C20H12F2O+): calcd 306.0856, found 306.0859. Anal.
Calcd. for C20H12F2O: C, 78.42; H, 3.95. Found: C, 78.21; H, 3.78.
1,3-Bis(3,5-difluorophenyl)isobenzofuran (10) was synthesized from 1-bromo-3,5-
difluorobenzene (15) (772 ꢁL, 6.700 mmol), magnesium (165 mg, 6.800 mmol), and 20 (500 mg,
3.046 mmol) in THF (15 mL) according to GP2. Compound 10 was obtained as a bright yellow
crystalline solid (395 mg, 1.153 mmol, 38%).
1
Mp 197.0 - 198.2 °C. H NMR (600 MHz, CDCl3): δ 6.72 - 6.80 (m, 2H), 7.12 - 7.16 (m,
2H), 7.41 - 7.45 (m, 4H), 7.79 - 7.82 (m, 2H). 13C {1H} NMR (150 MHz, CDCl3): δ 102.6 (t, JC,F
=
25.7 Hz), 107.4 (m), 119.7, 123.3, 126.6, 133.5 (t, JC,F = 10.5 Hz), 142.4 (t, JC,F = 3.7 Hz), 163.5
(dd, 1JC,F = 248.1 Hz, 2JC,F = 13.4 Hz). 19F NMR (470 MHz, CDCl3): δ -105.23 (m, 4F). IR (KBr):
3097, 1620, 1590, 1547, 1499, 1475, 1455, 1443, 1406, 1323, 1310, 1261, 1235, 1223, 1192, 1118,
1107, 1003, 986, 863, 854, 840, 828, 820, 744, 729, 678, 653, 645, 533, 521, 508 cm-1. MS, m/z
(%): 342.1 (100), 313.1 (19), 293.1 (14), 229.0 (13), 201.1 (16), 171.0 (8). HRMS, (EI) for
(C20H10F4O+): calcd 342.0668, found 342.0671. Anal. Calcd. for C20H10F4O: C, 70.18; H, 2.94.
Found: C, 69.92; H, 2.83.
1,3-Bis(3,4,5-trifluorophenyl)isobenzofuran (11) was synthesized from 5-bromo-1,2,3-
trifluorobenzene (16) (800 ꢁL, 6.700 mmol), magnesium (165 mg, 6.800 mmol), and 20 (500 mg,
3.046 mmol) in THF (15 mL) according to GP2. Compound 11 was obtained as a bright yellow
crystalline solid (486 mg, 1.285 mmol, 42%).
1
Mp 205.8 - 207.4 °C. H NMR (500 MHz, CDCl3): δ 7.12 - 7.15 (m, 2H), 7.47 - 7.52 (m,
4H), 7.72 - 7.76 (m, 2H). 13C {1H} NMR (125 MHz, CDCl3): δ 108.7 (m), 119.4, 122.9, 126.7,
126.8 (td, 1JC,F = 8.9 Hz, 2JC,F = 5.0 Hz), 138.9 (dt, 1JC,F = 254.0 Hz, 2JC,F = 15.6 Hz), 141.6 (td,
1JC,F = 3.2 Hz, 2JC,F = 1.0 Hz), 151.8 (ddd, 1JC,F = 249.9 Hz, 2JC,F = 10.5 Hz, 3JC,F = 4.5 Hz). 19F
NMR (470 MHz, CDCl3): δ -157.41 (tt, JF,F = 20.4 Hz, JF,H = 6.4 Hz, 2F), -129.53 (m, 4F). IR
11

(KBr): 3098, 1630, 1618, 1592, 1552, 1518, 1469, 1449, 1434, 1426, 1332, 1278, 1250, 1230,
1191, 1042, 1023, 998, 874, 858, 850, 834, 824, 771, 749, 742, 676, 634, 626, 590, 557, 519 cm-1.
MS, m/z (%): 378.0 (100, M), 349.0 (20), 342.1 (12), 331.1 (18), 329.0 (17), 311.1 (10), 247.0 (9),
189.0 (11). HRMS, (EI) for (C20H8F6O+): calcd 378.0479, found 378.0482. Anal. Calcd. for
C20H8F6O: C, 63.50; H, 2.13. Found: C, 63.28; H, 2.12.
1,3-Bis(perfluorophenyl)isobenzofuran (12). An argon filled 25 mL two-necked flask
equipped with stir bar and a gas condenser was charged with magnesium turnings (51 mg, 2.100
mmol). After three successive vacuum/argon cycles, ether (5 mL) and one grain of iodine were
added, and a brownish suspension was stirred for 15 min at room temperature. Subsequently 1-
bromopentafluorobenzene (249 ꢁL, 1.999 mmol) was added dropwise. An exothermic reaction was
followed by changes of the color of the reaction mixture from original brown, through colorless to
final grayish. This suspension was stirred 20 min at room temperature and then a solution of 3-
(perfluorophenyl)isobenzofuran-1(3H)-one (19) (300 mg, 0.999 mmol) in ether (10 mL) was added
dropwise and stirring continued for 4 h. A dense white solid precipitated. Subsequently
trifluoroacetic acid (1 mL) was added dropwise and the reaction mixture was stirred for additional
10 min. The reaction mixture was diluted with ether (50 mL), washed with a saturated aqueous
solution of NaHCO3 (2 × 30 mL), and dried over MgSO4. Solvent was removed under reduced
pressured and column chromatography on silica gel (hexane/ethyl acetate - 3:1) yielded 12 as an
yellow crystalline solid (360 mg, 0.800 mmol, 80%).
1
Mp 124.6 - 125.8 °C (dec.). H NMR (500 MHz, CDCl3): δ 7.13 - 7.16 (m, 2H), 7.44 - 7.48
(m, 2H). 13C {1H} NMR (125 MHz, CDCl3): δ 106.2 (td, 1JC,F = 16.3 Hz, 2JC,F = 4.2 Hz), 119.2 (t,
JC,F = 4.2 Hz), 125.4, 126.3, 133.7, 138.1 (m), 141.0 (m), 143.8 (m). 19F NMR (376 MHz, CDCl3):
δ -263.08 - -263.16 (m, 4F), -276.73 - -276.84 (m, 2F), -284.91 - -285.05 (m, 4F). IR (KBr): 1651,
1557, 1525, 1517, 1497, 1447, 1396, 1379, 1371, 1342, 1316, 1222, 1199, 1168, 1145, 1115, 1106,
1045, 898, 847, 842, 802, 789, 782, 749, 742, 727, 667, 632, 615, 579, 436 cm-1. MS, m/z (%):
450.0 (100, M), 431.0 (8), 421.0 (7), 403.0 (42), 383.0 (28), 372.0 (21), 352.0 (10), 334.0 (11),
255.0 (52), 205.0 (10), 186.0 (11), 167.0 (19), 117.0 (12). HRMS, (EI) for (C20H4F10O+): calcd
450.0102, found 450.0099. Anal. Calcd. for C20H4F10O: C, 53.35; H, 0.90. Found: C, 53.25; H,
1.09.
3-(Perfluorophenyl)isobenzofuran-1(3H)-one (21). An argon filled 50 mL two-necked
flask equipped with stir bar and a gas condenser was charged with magnesium turnings (177 mg,
7.300 mmol). After three successive vacuum/argon cycles, ether (10 mL) and one grain of iodine
were added, and brownish suspension was stirred for 15 min at room temperature. Subsequently 1-
12

bromopentafluorobenzene (873 ꢁL, 7.059 mmol) was added dropwise during 15 min. An
exothermic reaction was followed by changes of the color of the reaction mixture from original
brown, through colorless to final grayish. This suspension was stirred 60 min at room temperature
and then cooled to -40 °C. Then methyl 2-formylbenzoate (20) (500 mg, 3.046 mmol) was added
and stirring continued for 2 h at 0 °C. A grayish precipitate was formed. Subsequently
trifluoroacetic acid (1 mL) was added dropwise and the reaction mixture was stirred for additional
10 min at room temperature. A dense yellowish precipitate was formed. The reaction mixture was
diluted with ether (150 mL), washed with a saturated aqueous solution of NaHCO3 (2 × 30 mL),
and dried over MgSO4. Solvent was removed under reduced pressured and column
chromatography on silica gel (hexane/ethyl acetate - 2:1) afforded 21 as a white crystalline solid
(728 mg, 2.425 mmol, 80%).
1
Mp 114.5 - 115.8 °C (dec.). H NMR (400 MHz, CDCl3): δ 6.73 (s, 1H), 7.33 - 7.36 (m,
1H), 7.60 - 7.64 (m, 1H), 7.69 - 7.73 (m, 1H), 7.99 - 8.01 (m, 1H). 13C {1H} NMR (100 MHz,
CDCl3): δ 71.9, 110.3, 122.0, 125.7, 126.1, 130.1, 134.7, 139.0 (m), 140.8 (m), 144.2 (m), 146.8,
169.3. 19F NMR (376 MHz, CDCl3): δ -263.06 - -263.14 (m, 2F), -276.71 - -276.82 (m, 1F), -
284.89 - -285.03 (m, 2F). IR (KBr): 3088, 3066, 3033, 1803, 1774, 1726, 1655, 1611, 1601, 1524,
1507, 1475, 1468, 1427, 1388, 1349, 1338, 1308, 1286, 1260, 1211, 1191, 1161, 1139, 1127, 1116,
1102, 1052, 1001, 956, 893, 869, 807, 760, 750, 736, 702, 692, 662, 646, 595, 580, 523, 494 cm-1.
MS, m/z (%): 301.0 (100, M + H). HRMS, (APCI) for (C14H5O2F5 + H+): calcd 301.02825, found
301.02829. Anal. Calcd. for C14H5F5O2: C, 56.02; H, 1.68. Found: C, 55.81; H, 1.82.
3
RESULTS
3.1
Synthesis
Both the asymmetrically (2 - 8) and symmetrically (9 - 12) substituted fluorinated
derivatives of 1 were prepared by one-pot reaction33 shown in Scheme 1. Grignard reagents
obtained from 13 - 18 were treated with 3-phenylisobenzofuran-1(3H)-one (19) and the reaction
mixtures were quenched with CF3COOH14 or Ac2O13 to afford the asymmetric compounds 2 - 8.
The symmetrically substituted derivatives 9 - 12 were synthesized under almost identical reaction
conditions, but using methyl 2-formylbenzoate (20) instead of 19. Relatively low yields were
attributed mostly to a combination of low stability of some of the fluorinated Grignard reagents and
the general instability of the products, which in some cases rapidly decomposed during purification.
13

Scheme 1. One-pot syntheses of asymmetrically (2 - 8) and symmetrically (9 - 12) substituted
fluorinated derivatives of 1.
A two-step approach was tested on the synthesis of 12. A low-temperature reaction of 20 with a
slight excess of the Grignard reagent derived from 18 afforded 21 in a high yield (Scheme 2).
Addition of a second equivalent of the same Grignard reagent resulted in nearly quantitative
formation of 12 based on 1H NMR. The low stability of the product caused losses during isolation
and purification and reduced the yield to 80%.
Scheme 2. A two-step synthesis of 12.
14

3.2
Electrochemistry
The choice of material of the working electrode, Hg, Pt, glassy carbon (GC), or boron doped
diamond (BDDE) had negligible effect on the results, suggesting the absence of electrode-assisted
(adsorption) effects. The measured currents were linear with substrate concentration and with
square root of scan rate, as expected for diffusion-controlled redox processes.
The reduction of compounds 1 – 12 involves several steps (Table 1), suggesting a
complicated mechanism. Three reduction patterns have been found. Compounds 2-4, 6, 8-10, with
up to four fluorine atoms in the molecule (except 5), behave like the parent 1. Their first reduction
steps are reversible and one-electron, and are followed by another one-electron step. The
reversibility and formation of a stable radical anion were proven by cyclic voltammetry and in-situ
EPR spectroelectrochemistry. The values of all redox potentials vary regularly and additively with
the number and position of the fluorine atoms (Table 1).
Compounds 5 and 11 follow a second pattern, where after the first one-electron reduction a
two electron process occurs. A third pattern is observed with 7 and 12, whose molecules contain
one or two fully fluorinated phenyl rings, and where an irreversible first reduction step involves two
electrons. Only a very weak structureless EPR signal is observed at the first reduction step,
suggesting that the initial reduction product is unstable.
In the accessible potential range all compounds 1 - 12 are oxidized similarly in a single
irreversible step (Figure 2), therefore only the first anodic peak potentials and currents were
considered. From comparison of anodic peak currents with those of the first reduction steps of 2 - 4,
6, and 8 – 10 follows that the first oxidation step is a two-electron process.
Although the really correct correlation of experimental data with the quantum chemical
calculations is possible only for reversible processes (valid for the first reduction), we tried to
correlate the calculations also with the first oxidation step, though it behaves irreversibly.
Considering that the first anodic peak potentials were practically constant with increasing scan rate
(50-500mV.s-1), it points to the fact, that the overpotential is not significant, the irreversibility
(absence of the counter peak) is caused by fast follow-up reaction whereas the primary oxidation
potential is close to the reversible one. This interpretation was supported by very good correlation
of experimental oxidation potential with the calculated electrochemical oxidation potentials (Fig.
S6a in supporting information) as well as with calculated HOMO energies (Fig. S7a).
15

Figure 2. Cyclic voltammograms of 1 - 12 in DMF at 25 °C, glassy carbon electrode, 0.2 V/s, 0.1
M [nBu4N][PF6] vs SCE
16

Table 1: Table of experimental electrochemical data of 1 – 12 (conc. 1mM) on glassy carbon
electrode taken from CV at 100 mV.s-1 (in V relative to SCE).
oxidn.
first reduction step
other reduction stepsd
Epa(1)a Epc(1) Epa(1)b E0(1)c Epc(2) Epc(3) Epc(4) Epc(5) Epc(6)
Cmpd.
0.79
0.80
0.87
0.93
0.91
0.99
1.12
0.82
0.94
1.05
1.04
1.39
-1.89 -1.81
-1.90 -1.82
-1.85
-1.86
-2.32
-2.32
-2.81
1
-2.77 -2.98
-2.73 -2.98
-2.67 -2.99
-2.66 -3.00
-2.70 -2.98
-2.66
2
-1.81 -1.74 -1.775 -2.23
3
-1.72 -1.64
-1.75 -1.67
-1.68
-1.71
-2.15
-2.05
4
-2.15
-2.11
5
-1.90 -1.83 -1.865 -2.29
6
e
-1.69
–
-2.03
-2.33
-2.16
-2.42
7
-1.90 -1.82
-1.73 -1.65
-1.86
-1.69
-2.77 -3.05
-2.68 -2.99
-2.60 -2.99
-2.63 -3.00
-2.53
8
9
-1.56 -1.49 -1.525 -1.99
10
11
12
-1.60 -1.51 -1.555 -1.90
-1.99
-2.25
e
-1.59
–
-1.94
a
Epa are the first anodic peak potentials.
b Epa are the counter peak potentials of the first reduction step.
c E0 are the first reduction potentials (reversible) obtained from cyclic voltammetry as (Epc + Epa)/2
d Epc are the cathodic peak potentials. Accuracy (reproducibility and repeatability) is ± ≤10mV.
e
Different reduction pattern: two-electron step, where anodic counter peak is missing.
3.3
Calculations
3.3.1 Interatomic Distances (Figure 3 and Table S1). In the neutral compounds 1 - 12,
calculated bond lengths are not significantly affected by fluorine substitution. The length of the C-
17

O bonds a and e is 137.0 - 137.6 pm, that of the C=C bonds b and d 139.2 - 140.1 pm, that of the
aromatic C-C bonds c 145.1 - 145.4 pm, and that of the exocyclic C-C bonds f and g, 144.5 - 145.1
pm. In the radical anions of 1 - 12, the C-O bonds a and e are noticeably longer (139.8 - 140.3 pm)
and the C=C bonds b and d slightly longer (141.0 - 141.7 pm). In contrast, the length of bond c
remains almost unchanged (145.2 - 145.9 pm) and the bonds f and g are 2 pm shorter (142.1 -
143.3 pm). In the radical cations of 1 - 12, the calculated length of C-O bonds a and e (136.9 -
137.3 pm) is essentially the same as in the neutrals, but the C=C bonds b and d are significantly
elongated (141.3 - 142.2 pm), and the bond c is shortened (143.7 - 143.9 pm). The exocyclic C-C
bonds f and g (142.9 - 143.9 pm) are slightly shorter (144.5 - 145.1 pm).
Figure 3. Labels for interatomic distances and torsion angles (see Table S1).
3.3.2 Torsion angles. Substitution of hydrogen by fluorine atoms in meta and para position of
phenyl rings Ph(1) and Ph(2) (Figure 3) does not affect the calculated torsion angles with the plane
of the isobenzofurane moiety, ϕ1 and ϕ2. Not surprisingly, fluorine atoms in the ortho positions of
these phenyl rings increase the torsion angle ϕ1 in 6 and 7, and both ϕ1 and ϕ2 in 12.
In the neutral compounds that do not carry fluorine atoms in ortho positions, the angles ϕ1 and
ϕ2 are 18.5 - 23.5°. The radical anions are nearly planar, with torsion angles of 0.4 - 8.1°. In
contrast, in cation radicals the torsion angles ϕ1 and ϕ2 (15.7 - 21.7°) are nearly identical with
those observed in the neutral compounds.
The fluorine atoms in the ortho positions of 6, 7, and 12 cause the phenyl rings to twist more,
to 36.9 - 45.0°. This twist angle is slightly reduced in the radical anions (32.0 - 35.3°) and cations
(32.8 - 35.5°).
18

3.3.3 Standard Reduction Potentials. Table 2 compares the values obtained from DFT
calculations for the first oxidation and reduction potentials using two different choices of
functionals with the averages of the potentials measured on all four electrodes used. The
calculations relied on a standard procedure: the one-electron reduction potential reflects Gibbs free
energy change upon one-electron oxidation of a solute,
E° [V] = - ꢀGred/ox - E°abs[reference],
(1)
where ꢀGred/ox (= Gred – Gox) is the free energy difference between the reduced and oxidized
state of a solute and E°abs[reference] is the absolute potential of a reference electrode. Since we are
not aware of any standard recommendation for the absolute potential of SCE in DMF solution, we
+/0
estimated the E°abs[SCE(DMF)] from E°abs[FeCp2+/0(DMF)] (= 4.97 eV; FeCp2
=
ferrocenium/ferrocene redox couple) 34, and FeCp2+/0 to SCE scale conversion of 0.45-0.47 eV - see
Table 1 in ref. 35. Similarly, one may adopt the absolute potential of standard hydrogen electrode
(SHE) in DMF (= 4.29 eV) with SHE to SCE scale conversion of 0.24 eV (reported for an aqueous
solution in ref. 36). Both approaches lead to a similar result for SCE in DMF, with an absolute
potential of E°abs[SCE(DMF)] = 4.50-4.53 eV. Herein, we adopted a rounded value of 4.5 eV. The
ꢀGred/ox values were evaluated from:37
ꢀGred/ox = ꢀEel + ꢀGsolv + ꢀ(EZPVE + pV – RT lnQ)
(2)
where ꢀEel (= Eel,red – Eel,ox) is the energy difference between the reduced and oxidized structures in
vacuum, ꢀGsolv (= ꢀGgsolv,red − ꢀGsolv,ox) is the difference in their solvation energies, and ꢀ(EZPVE
+
pV – RT lnQ) is the difference in the thermal enthalpic and entropic contributions. As documented
in several instances, such methodology is expected to provide relative values of reduction potentials
within the accuracy of ca. 50 mV 36. An excellent relative agreement between the experimental and
calculated electrochemical potentials is evidenced in Table 2 and in Figure S6a,b in the Supporting
Information. We note in passing that a correlation between the energy of HOMO (LUMO) in a
series of compounds 1-12 and their oxidation (reduction) potentials also agrees well and is plotted
in Figure S7a,b.
Table 2. Experimental data a taken from CV at 100 mV.s-1 and calculated reduction potentials for
“radical cation – neutral” (E+/0) and “neutral – radical anion” (E0/-) couples (in V against SCE in
DMF at 298 K).
19

E+/0
E0/-
Exp.a
(E0
Exp.a
Cmpd.
B3LYPb PBEc
B3LYPb PBEc
(Ep,a)
)
red
0.79
0.80
0.87
0.93
0.91
0.99
1.12
0.82
0.94
1.05
1.04
1.39
0.58
0.57
0.64
0.72
0.73
0.73
0.87
0.56
0.71
0.87
0.88
1.16
0.965
0.66
0.63
0.72
0.77
0.76
0.78
0.89
0.60
0.77
0.89
0.85
1.10
0.956
-1.85
-1.86
-1.74
-1.73
-1.65
-1.54
-1.53
-1.79
-1.58
-1.73
-1.51
-1.36
-1.31
-1.37
0.962
-1.54
-1.54
-1.47
-1.38
-1.39
-1.61
-1.43
-1.56
-1.35
-1.23
-1.21
-1.26
0.962
1
2
-1.775
-1.68
3
4
-1.71
5
-1.865
-1.69e
-1.86
6
7
8
-1.69
9
-1.525
-1.555
-1.59e
10
11
12
R2 d
a Experimental values for oxidation are anodic peak potentials using GC electrode (data from
BDDE and Pt electrodes are very close to those of GC), the data for reduction are the first
reversible potentials E0 obtained from cyclic voltammetry on GC as (Epc + Epa)/2. b RI-B3LYP-
D3/def2-TZVPD. c RI-PBE-D3/def2-TZVPD. d R is the regression coefficient. e Cathodic peak
potentials due to absence of anodic counter peak and thus different reduction pattern: two-electron
irreversible step.
4
DISCUSSION
20

4.1
Synthesis. Preparation of all 11 fluorinated DPIBF derivatives was a straightforward
process utilizing previously developed reaction conditions13,14 and does not call for much comment.
Two different quenching procedures were tested, using either CF3COOH or Ac2O, and produced
nearly identical isolated yields. Probably the most noticeable difference was observed immediately
after the addition of these reagents: CF3COOH usually caused the crude reaction mixture to turn
black, while it turned deep yellow when Ac2O was used. The difficult part was the isolation and
purification. Although the bright to deep yellow solid products can be stored under an argon
atmosphere in a freezer for several months without noticeable decomposition, their solution
stability is poor. They require strictly oxygen-free solvent and even then some of them decompose
1
within a few hours. H NMR revealed that 1 and 2 are the most stable in CDCl3 solution and did
not show any signs of decomposition even after several days in the dark at room temperature.
4.2
Molecular Geometry. Single-crystal X-ray diffraction structures reported for the two
crystal forms of neutral 17 yield essentially identical molecular geometries, in good agreement with
the results shown in Table S1, and an effort to determine the geometries of the neutral species 2 -
12 is underway. They will presumably all be very similar except possibly for the phenyl twist
angles. As explained in the introduction, the main interest is in determining crystal packing.
Geometries of the radical ions are much harder to access experimentally, and at this time, only
the calculated values listed in Table S1 are available. The bond lengths differ from those in the
neutral species in a way that is easily understood by consideration of the distribution of nodal
planes in the highest occupied (HOMO) and lowest unoccupied (LUMO) molecular orbitals of
isobenzofuran, which is the same in all twelve compounds (Figure S1) and is illustrated for the
parent compound 1 in Figure 4. Approximate positions of the nodal planes can be deduced easily
from a consideration of the MOs of the [9]annulenide anion perimeter of isobenzofuran. The two
degenerate HOMOs of this perimeter have two nodal planes perpendicular to the molecular plane,
with nodal points along the perimeter 9/4 of a bond length apart, and the two LUMOs have three,
with nodal points 3/2 bond lengths apart. When an oxygen atom replaces one of the carbons to
yield oxonine, the energies of the MOs that have a node going through the oxygen atom will not
respond to its high electronegativity in the first approximation, whereas those that do not will be
stabilized. Therefore, the HOMO will have a node going through the oxygen and the LUMO will
not. When a cross-link is introduced to convert oxonine to isobenzofuran, the HOMO will be
destabilized and LUMO will be stabilized, and their status as the frontier orbitals reinforced. Once
the locations of the nodal points on the isobenzofuran perimeter have been established, the relative
amplitudes of the MOs at various perimeter atoms follow - the closer the atom is to a nodal plane,
the smaller the amplitude.
21

Removal of an electron from HOMO will halve its bonding contribution between atoms that
are not separated by a node and halve its antibonding contribution between atom that are, and bond
lengths will respond correspondingly. Addition of an electron to the LUMO will similarly
introduce additional bonding between atoms that are not separated by a node and antibonding
between atoms that are, and bond lengths will respond. The magnitude of the effect will reflect the
amplitude of the MO on the atoms in question. Thus, oxidation is expected to extend the b and d
bonds and shorten f, g, and c bonds, whereas reduction is expected to extend the a and e bonds and
shorten the f and g bonds, exactly as is observed in Table S1.
Figure 4. Frontier molecular orbitals of 1: (a) HOMO and (b) LUMO.
4.3
Redox Potentials. The relative values of the first oxidation and first reduction potentials of
compounds 1 - 12 are well reproduced by DFT calculations (Table 2), with high regression
coefficients. However, a large shift of the regression line is also observed, suggesting that our
theoretical model does not fit the measured electrochemical potentials absolutely. The shift from
the experimental values in average corresponds to –0.24 V and +0.19 V for the B3LYP method for
“radical cation – neutral” and “neutral – radical anion” reduction couples, or –0.21 V and +0.34 V
for the PBE method, respectively. The shift is not particularly surprising as the scaling of the
experimental values to the absolute potential depends on the experimental setup, calibration of
electrode used, surface potential, etc.
It is useful to obtain qualitative intuitive insight as well, since it is easily transferred to
related compounds. This is obtained by considering the two phenyl substituents as a perturbation of
the electronic structure of isobenzofuran and using standard notions for the description of
22

substituent effects. By extending the π system, the phenyl groups act as substituents with both
resonance and inductive effects, increasing the energy of the HOMO and lowering that of the
LUMO of isobenzofuran, to a degree that depends on the phenyl twist angle. The fluorine atoms on
the phenyl rings modulate their inductive and resonance effects. The latter can occur in two ways,
directly when the fluorine substituents are in ortho or para positions, and indirectly via their steric
effects on the phenyl twist angles when they are in ortho positions. Comparison of results for 1 - 12
shows that every fluorine atom in the meta position on a phenyl substituent makes the first
reduction easier by ~80 mV, and this can be attributed to its inductive effect. In contrast, para and
ortho fluorine atoms on the phenyl substituents have practically no effect on the first reduction
potentials. In the case of para substitution, this can be understood as a result of compensation of the
now weaker electron-withdrawing inductive effect of the fluorine by its weak electron-donating
resonance effect. For ortho fluoro substitution the inductive effect would then be expected to
dominate over the resonance effect, facilitating the first reduction, but it is apparently just
compensated by the sterically induced increase in the phenyl twist angle that makes the reduction
harder.
The results for anodic potentials can be understood similarly. The inductive effect of every
meta fluorine makes the oxidation harder by 50-60 mV, which is consistent with its effect on
reduction, and the negligible influence of a para fluorine on the oxidation potential can be
understood in the same way as in the case of the reduction potential. Now, however, the inductive
and the sterically induced reduction of the resonance effect due to the two ortho fluorine
substituents at each phenyl do not compensate each other but add, making the oxidation more
difficult by ~180 mV.
These simple considerations may be intuitively satisfying but are strictly qualitative. A
more detailed comparison of the magnitudes of substituent effects on the oxidation and reduction
potentials would require a knowledge of the relative amplitudes of the HOMO and the LUMO in
the position of attachment of the perturbing phenyl substituent and hence a computation. For such
quantitative work we consider it preferable to perform a DFT calculation as discussed above.
5
SUMMARY
A series of fluorinated 1,3-diphenylisobenzofurans was synthesized and characterized by
23

standard electrochemical methods. For most of the compounds, the first reduction step is reversible
and the formation of an intermediate radical anion was proved by cyclic voltammetry and EPR
spectroelectrochemistry. The redox potentials obtained showed significant variation and will be
useful in studies of singlet fission. The observed regularities were interpreted qualitatively by
consideration of substituent effects and quantitatively by comparison with DFT calculations. The
bond length changes expected upon one-electron oxidation or reduction were computed and
rationalized.
Four members of the series with the highest numbers of fluorine atoms undergo reduction
by a more complicated mechanism that has not been investigated further.
ASSOCIATED CONTENT
Supporting Information. Copies of 1H, 13C NMR spectra of all new compounds, ORTEP view
of a single molecule and packing in 21, calculated geometries of 1 - 12 and their radical cations and
anions, correlation between the experimental oxidation (reduction) potentials of compounds 1-12,
their calculated energies HOMO (LUMO) and calculated electrochemical potentials, respectively.
ACKNOWLEDGEMENT
Work in Prague was supported by the grant GAČR 19-22806S, institutional supports of the J.
Heyrovský Institute of Physical Chemistry (RVO 61388955), and Institute of Organic Chemistry
and Biochemistry (RVO: 61388963). We are grateful to Dr. Ivana Císařová for performing a
single-crystal X-ray analysis of compound 21. Work in the USA was supported by the U.S. DoE
BES, Division of Chemical Sciences, Biosciences, and Geosciences (DE-SC0007004).
REFERENCES
1. M. B. Smith, J. Michl, Singlet Fission, Chem. Rev. 110 (2010) 6891.
2. M. B. Smith, J. Michl, Recent Advances in Singlet Fission, Ann. Rev. Phys. Chem. 64 (2013)
361.
3. M. C. Hanna, A. J. Nozik, Solar Conversion Efficiency of Photovoltaic and Photoelectrolysis
Cells with Carrier Multiplication Absorbers, J. App. Phys. 100 (2006) 074510.
4. W. Shockley, H. J. Queisser, Detailed Balance Limit of Efficiency of p-n Junction Solar Cells,
24

J. App. Phys. 32 (1961) 510.
5. I. Paci, J. C. Johnson, X. Chen, G. Rana, D. Popović, D. E. David, A. J. Nozik, M. A. Ratner,
J. Michl, Singlet Fission for Dye-Sensitized Solar Cells: Can a Suitable Sensitizer Be Found?
J. Am. Chem. Soc. 128 (2006) 16546.
6. J. C. Johnson, A. J. Nozik, J. Michl, High Triplet Yield from Singlet Fission in a Thin Film of
1,3-Diphenylisobenzofuran, J. Am. Chem. Soc. 132 (2010) 16302.
7. J. Ryerson, J. N. Schrauben, A. J. Ferguson, S. C. Sahoo, P. Naumov, Z. Havlas, J. Michl, A. J.
Nozik, J. C. Johnson, Two Thin Film Polymorphs of the Singlet Fission Compound 1,3-
Diphenylisobenzofuran, J. Phys. Chem. C. 118 (2014) 12121.
8. Z. Havlas, J. Michl, Guidance for Mutual Disposition of Chromophores for Singlet Fission, Isr. J.
Chem. 56 (2016) 96.
9. E. A. Buchanan, J. Michl, Packing Guidelines for Optimizing Singlet Fission Matrix Elements in
Non-Covalent Dimers, J. Am. Chem. Soc. 139 (2017) 15572.
10. E. A. Buchanan, Z. Havlas, J. Michl, Singlet Fission: Optimization of Chromophore Dimer
Geometry, in: J. R. Sabin, E. J. Brändas (Eds.), Eds.Advances in Quantum Chemistry: Ratner
Volume, Vol. 75, Elsevier, Cambridge, MA, 2017, p. 175.
11. R. J. Dillon, G. B. Piland, C. J. Bardeen, Different Rates of Singlet Fission in Monoclinic
versus Orthorhombic Crystal Forms of Diphenylhexatriene, J. Am. Chem. Soc. 135 (2013) 17278.
12. D. H. Arias, J. L. Ryerson, J. D. Cook, N. H. Damrauer, J. C. Johnson, Polymorphism
Influences Singlet Fission Rates in Tetracene Thin Films, Chem. Sci. 7.2 (2016) 1185.
13. P. I. Dron, J. Michl, J. C. Justin, Singlet Fission and Excimer Formation in Disordered Solids of
Alkyl-Substituted 1, 3-Diphenylisobenzofurans, J. Phys. Chem. A 121 (2017) 8596.
14. Ch. Dai, P. Nguyen, T. B. Marder, A. J. Scott, W. Cleggc, Ch. Viney, Control of single crystal
structure and liquid crystal phase behaviour via arene-perfluoroarene interactions, Chem. Commun.
24 (1999) 2493.
15. M. L. Renak, G. P. Bartholomew, S. Wang, P. J. Ricatto, R. J. Lachicotte, G. C. Bazan,
Fluorinated Distyrylbenzene Chromophores: Effect of Fluorine Regiochemistry on Molecular
Properties and Solid State Organization, J. Am. Chem. Soc.121 (1999) 7787.
16. J. C. Collings, J. M. Burke, P. S. Smith, A. S., Batsanov, J. A. K. Howard, T. B. Marder, The
synthesis and crystal structures of halogenated tolans p-X-C6H4-C C-C6F5 and p-X-C6F4-C C-
C6H5 (X = F, Cl, Br, I), Org. Biomol. Chem. 2 (2004) 3172.
17. A. Liška, P. Vojtíšek, A. J. Fry, J. Ludvík, Electrochemical and Quantum Chemical
Investigation of Tetranitrocalix[4]arenes: Molecules with Multiple Redox Centers, J. Org. Chem.
78 (2013) 10651.
18. R. Ahlrichs, M. Bar, M. Haser, H. Horn, C. Kolmel, Electronic Structure Calculations on
Workstation Computers: The Program System Turbomole, Chem. Phys. Lett. 162 (1989) 165.
19. K. Eichkorn, O. Treutler, H. Öhm, M. Häser, R. Ahlrichs, Auxiliary Basis Sets to Approximate
25

Coulomb Potentials, Chem. Phys. Lett. 240 (1995) 283.
20. J. P. Perdew, Density-Functional Approximation for the Correlation Energy of the
Inhomogeneous Electron Gas, Phys. Rev. B 33 (1986) 8822.
21. A. D. Becke, Density-Functional Exchange-Energy Approximation With Correct Asymptotic
Behavior, Phys. Rev. A 38 (1988) 3098.
22. S. Grimme, J. Antony, S. Ehrlich, H. Krieg, A Consistent and Accurate ab initio
Parametrization of Density Functional Dispersion Correction (DFT-D) for the 94 Elements H-Pu, J.
Chem. Phys. 132 (2010) 154104.
23. F. Weigend, R. Ahlrichs, Balanced Basis Sets of Split Valence, Triple Zeta Valence and
Quadruple Zeta Valence Quality for H to Rn: Design and Assessment of Accuracy, Phys. Chem.
Chem. Phys. 7 (2005) 3297.
24. A. Klamt, G. Schuurmann, COSMO: A New Approach to Dielectric Screening in Solvents
With Explicit Expressions for the Screening Energy and its Gradient, J. Chem. Soc., Perkin Trans.
(1993) 799.
25. A. D. Becke, Density Functional Thermochemistry. III. The Role of Exact Exchange, J. Chem.
Phys. 98 (1993) 5648.
26. D. Rappoport, F. Furche, Property-Optimized Gaussian Basis Sets for Molecular Response
Calculations, J. Chem. Phys. 133 (2010) 134105.
27. J. P. Perdew, K. Burke, M. Ernzerhof, Fluid Vesicles in Shear Flow, Phys. Rev. Lett. 77 (1996)
3865.
28. COSMOtherm, C3.0, release 1701, COSMOlogic GmbH & Co KG, http://www.cosmologic.de.
29. A. Klamt, Conductor-like Screening Model for Real Solvents: A New Approach to the
Quantitative Calculation of Solvation Phenomena, J. Phys. Chem. 99 (1995) 2224.
30. S. Grimme, Supramolecular Binding Thermodynamics by Dispersion Corrected Density
Functional Theory, Chem. Eur. J. 18 (2012) 9955.
31. G. M. Sheldrick, SHELXT - Integrated Space-Group and Crystal-Structure Determination, Acta
Cryst. A71 (2015) 3.
32. G. M. Sheldrick, Crystal Structure Refinement With SHELXL, Acta Cryst. C71 (2015) 3.
33. T. Hamura, R. Nakayama, A One-pot Preparation of 1,3-Diarylisobenzofuran, Chem. Lett. 42
(2013) 1013.
34. P. P. Romańczyk, G. Rotko, S. S. Kurek, The Redox Potential of the Phenyl Radical/Anion
Couple and the Effect Thereon of the Lithium Cation: A Computational Study, Electrochem.
Commun. 48 (2014) 21.
35. N. G. Connelly, W. E. Geiger, Chemical Redox Agents for Organometallic Chemistry, Chem.
Rev. 96 (1996) 877.
26

36. A. V. Marenich, J. Ho, M. L. Coote, C. J. Cramer and D. G. Truhlar Computational
electrochemistry: prediction of liquid-phase reduction potentials. Phys. Chem. Chem. Phys. 16 (2014)
15068.
37. Several relevant conformers/rotamers for each of the oxidized and reduced structure were taken
into account. The reduction potential E° is calculated as the difference between the lowest-lying
conformers as predicted by the free energy calculations.
27