K. Funabiki et al. / Journal of Fluorine Chemistry 127 (2006) 257–262
261
fluorescein hardly have an influence on the short-circuit
photocurrent. On the other hand, fluorine atoms at 20,40,50,70-
position slightly caused the reduction of not only the Jsc but Voc.
Highly fluorinated fluoresceins 4c also decreased the values of
the short-circuit photocurrent. In other words, as shown in
Table 1, fluorination of 20,40,50,70-position of fluorescein (4a:
fluorophthalic anhydride was purchased from Acros organics.
All chemicals were of reagent grade and, if necessary, were
purified in the usual manner prior to use. 3,5-Difluororesorci-
nol, 4,5,6,7-tetrafluorofluorescein, and 4,5,6,7, 20,40,50,70-octa-
fluorofluorescein were prepared according to literature [11].
E
red = ꢀ1.63) resulted in a lowering of the Ered (LUMO) than
4.3. Procedure for the synthesis of 20,40,50,70-
tetrafluorofluorescein (4a)
that of 4,5,6,7-position (4b: Ered = ꢀ1.68), and further
fluorination enormously lowers the Ered of the dye 4c (4c:
E
red = ꢀ1.39). That is, the photochemical properties, such as
The mixture of phthalic anhydride (1 mmol, 0.150 g),
methanesulfonic acid (10 ml), 2,4-difluoroesorcinol (2 mmol,
0.292 g) was heated at 80–85 8C for 40 h under argon. The
resultant mixture was quenched with water (70 ml). The
precipitate was corrected by filtration and dried under vacuum.
The precipitate was purified by column chromatography on
silica gel using hexane–ethyl acetate (v/v = 1/3) to give
20,40,50,70-tetrafluorofluorescein in 74% yield (0.304 g).
Yield 74%; mp > 300 8C; Rf 0.50 (hexane/ethyl acetate = 1/
Voc and Jsc, may have a crucial relation to the LUMO of the
dyes. Fluorination of fluorescein may cause a lowering of the
Ered (LUMO) as well as a reduction of the photochemical
properties.
3. Conclusion
In summary, four kinds of ring-fluorinated fluoresceins were
prepared and employed as a sensitizer for ZnO solar cell.
Among the dyes, 4,5,6,7-tetrafluorofluorescein gave highest
solar-light-to-electricity conversion efficiency (h), which is less
efficient than that of fluorine-free fluorescein. As an anchoring
group, the carboxy group is better than sulfo group.
1
3); H NMR (DMSO-d6) d 6.50 (2H, dd, J = 11.22, 1.95 Hz),
7.38 (1H, d, J = 7.69 Hz), 7.75 (1H, t, J = 7.69 Hz), 7.81 (1H, t,
J = 7.69 Hz), 8.01 (1H, d, J = 7.69 Hz), 11.2 (2H, s); 13C NMR
(DMSO-d6) d 168.15 (s), 151.36 (s), 148.52 (dd, J = 240.67,
4.96 Hz), 140.58 (dd, J = 244.80, 6.62 Hz), 136.63 (d,
J = 9.10 Hz), 136.25 (dd, J = 17.37, 12.41 Hz), 135.85 (s),
130.60 (s), 125.49 (s), 125.20 (s), 123.98 (s), 108.87 (d,
J = 7.45 Hz), 108.24 (d, J = 21.50 Hz), 80.28 (s); 19F NMR
(DMSO-d6) d ꢀ57.26 (2F, t, J = 11.22 Hz), ꢀ74.84 (2F, dd,
J = 11.22, 1.95 Hz); HRMS (EI): m/z calcd. for C20H8O5F4
(M), 404.0308; found, 404.0305.
4. Experimental
4.1. Instruments
Melting points were obtained on a Yanagimoto MP-S2
micro-melting point apparatus and are uncorrected. Infrared
spectra (IR) were recorded on a Perkin-Elmer System 2000 FT-
4.4. Procedure for the synthesis of 20,40,50,70-
tetrafluorosulfofluorescein (4d)
1
IR spectrometer using KBr. H NMR spectra were measured
with a JEOL a-400 (400 MHz) FT-NMR spectrometers in
CDCl3 or DMSO-d6 solutions with tetramethylsilane as the
internal standard. 13C NMR spectra were obtained on a JEOL
a-400 (100 MHz) FT-NMR spectrometers in CDCl3 or DMSO-
d6 solutions. High resolution mass spectra (HRMS) were taken
on a JEOL JMS-700 mass spectrometer operating at an
ionization potential of 70 eV. UV–vis absorption and fluores-
cence spectra were taken on Hitachi U-3500 and F-4500
spectrophotometers, respectively. Electrochemical measure-
ments of dyes were measured with a model 800 electrochemical
workstation (CH Instruments, Austin, TX, USA). One-step
cathode electrodeposition was taken on Hokuto-Denko HSV-
100 potentiostat system. The photoelectrochemical measure-
ments (two-electrode system) of solar cells were performed on
a Bunko-Keiki CEP-2000 system (AM 1.5).
The mixture of phthalic anhydride (1 mmol, 0.150 g),
methanesulfonic acid (10 ml), 2,4-difluoroesorcinol (2 mmol,
0.292 g) was heated at 80–85 8C for 2 days under argon. The
resultant mixture was quenched with water (70 ml), extracted
with ethyl acetate (50 ml ꢁ 3 ml), dried over Na2SO4, and
concentrated under vacuum. The precipitate was purified by
column chromatography on silica gel using methanol–
dichloromethane (v/v = 1/5) to give 20,40,50,70-tetrafluorosulfo-
fluorescein (4d) in 31% yield (0.136 g).
Yield 37%; mp > 300 8C; Rf 0.03 (MeOH–CH2Cl2 = 1/5);
1H NMR (DMSO-d6) d 6.25 (2H, d, J = 12.39 Hz), 7.13 (2H, d,
J = 7.59 Hz), 7.48 (1H, t, J = 7.59 Hz), 7.56 (1H, t, J =
7.39 Hz), 7.96 (1H, d, J = 7.59 Hz); 13C NMR (DMSO-d6) d
109.99 (d, J = 21.49 Hz), 113.66 (d, J = 9.04 Hz), 127.60 (s),
128.27 (s), 129.27 (s), 129.08 (s), 129.39 (s), 129.83 (s), 139.95
(s), 140.18 (dd, J = 244.07, 7.83 Hz); 19F NMR (DMSO-d6) d
ꢀ55.44 (2F, dd, J = 19.46, 12.20 Hz), ꢀ86.28 (2F, d,
J = 19.46 Hz); HRMS (FAB): m/z calcd. for C19H9 O6F4S
(M + H), 441.0056; found, 441.0057.
4.2. Materials
The isolation of pure products was carried out by column
chromatography using silica gel (Wakogel C-200, 100–200
mesh, Wako Pure Chemical Ind., Ltd.). Analytical TLC was
done on Merck precoated (0.25 mm) silica gel 60 F254 plates.
Ten percent Pd on carbon was purchased from Merck Co. Ltd.
2,3,4,5-Tetrafluoronitrobenzene and tetrafluorosulfobenzoic
anhydride were purchased from Tokyo Kasei Kogyo. Tetra-
4.5. Preparation of dye-sensitized solar cells with ZnO
Preparation of ZnO thin film as well as the dye-sensitized
solar cells was prepared as described in the previous paper