Diarylindenotetracenes via a selective cross-coupling/C-H functionalization: Electron donors for organic photovoltaic cells

Article abstract of DOI:10.1021/ol300098p

A direct synthesis of new donor materials for organic photovoltaic cells is reported. Diaryindenotetracenes were synthesized utilizing a Kumada-Tamao-Corriu cross-coupling of peri-substituted tetrachlorotetracene with spontaneous indene annulation via C-H

Full text of DOI:10.1021/ol300098p

ORGANIC
LETTERS
2012
Vol. 14, No. 6
1390–1393
Diarylindenotetracenes via a Selective
Cross-Coupling/CÀH Functionalization:
Electron Donors for Organic Photovoltaic
Cells
Xingxian Gu,† Wade A. Luhman,‡ Elisey Yagodkin,† Russell J. Holmes,*,‡ and
Christopher J. Douglas*,†
Department of Chemistry, Department of Chemical Engineering and Materials Science,
University of Minnesota, Minneapolis, Minnesota 55455, United States
rholmes@umn.edu; cdouglas@umn.edu
Received January 14, 2012
ABSTRACT
A direct synthesis of new donor materials for organic photovoltaic cells is reported. Diaryindenotetracenes were synthesized utilizing a
KumadaÀTamaoÀCorriu cross-coupling of peri-substituted tetrachlorotetracene with spontaneous indene annulation via CÀH activation.
Vacuum deposited planar heterojunction organic photovoltaic cells incorporating these molecules as electron donors exhibit power conversion
efficiencies exceeding 1.5% with open-circuit voltages ranging from 0.7 to 1.1 V when coupled with C60 as an electron acceptor.
Cross-coupling reactions of aryl nucleophiles with aryl
electrophiles, such as the KumadaÀTamaoÀCorriu
reaction,1,2 are now ubiquitous transformations in syn-
thetic organic chemistry. Recently, Scott discovered that
peri-substituted dihaloacenes can undergo an indene an-
nulation reaction via a cross-coupling/CÀH activation cas-
cade reaction, albeit with high catalyst loadings (40 mol %
palladium).3À5 Given our interest in acene materials,6,7 we
thought that 5,6,11,12-tetrachlorotetracene (1a)8 might
undergo similar types of transformations. Hypothetically,
reaction of phenyl magnesium bromide with 1a could lead
to rubrene (1b), 9,10-diphenylindeno[1,2,3-fg]tetracene
(2a), or diindenotetracene (3) (Scheme 1). In some ways,
2a might be more difficult to prepare from 1a, since
disparate reactivity between the two sets of peri-chlorides
is required. Products 2a and 3 are candidates for organic
† Department of Chemistry.
‡ Department of Chemical Engineering and Materials Science.
(1) Corriu, R. J. P.; Masse, J. P. J. Chem. Soc., Chem. Commun. 1972,
144a.
(6) Yagodkin, E.; Douglas, C. J. Tetrahedron Lett. 2010, 51, 3037.
(7) Yagodkin, E.; Xia, Y.; Kalihari, V.; Frisbie, C. D.; Douglas, C. J.
J. Phys. Chem. C 2009, 113, 16544.
(2) Tamao, K.; Sumitani, K.; Kumada, M. J. Am. Chem. Soc. 1972,
94, 4374.
(3) Quimby, J. M.; Scott, L. T. Adv. Synth. Catal. 2009, 351, 1009.
(4) Wegner, H. A.; Scott, L. T.; de Meijere, A. J. Org. Chem. 2003, 68,
883.
(8) 1a can be prepared in multigram quantities by a three-step,
chromatography-free synthesis from 1,4-naphthalenediol and phthaloyl
chloride. See ref 4b and Yagodkin, E.; McGarry, K. A.; Douglas, C. J.
Org. Prep. Proc. Int. 2011, 43, 360.
(5) Wang, L.; Shevlin, P. B. Org. Lett. 2000, 2, 3703.
r
10.1021/ol300098p
Published on Web 03/01/2012
2012 American Chemical Society

major product (Scheme 2).21 Additional optimization of the
reaction conditions allowed us to lower the catalyst loading
(from 12 to 3 mol %) and reaction temperature (from 80 to
40 °C) with a negligible decrease in yield (82%). These lower
catalyst loadings, when compared to prior work,3 are valu-
able in the preparation of pure materials for device studies.
Prior work with 2a was reported before modern structural
characterization methods were available.12 Our data for 2a
are in agreement with the previous reports, and we were able
to confirm the structure by X-ray crystallography.
Scheme 1. Potential Products upon Treatment of 1a with a
Phenyl Nucleophile and Pd Catalyst
Scheme 2. Synthesis of 9,10-Diphenylindeno[1,2,3-fg]tetracene 2aa
a Conditions A: PhMgBr (12 equiv), PEPPSI-IPR (10 mol %), dioxane,
80 °C, 24 h (84%). Conditions B: PhMgBr (10 equiv), PEPPSI-IPR
(3 mol %), Et2O, 40 °C, 16 h (82%). Yields are of recrystallized material.
electronic devices, given their relationship to rubrene, a
benchmark organic semiconductor.9À11 Although the syn-
theses of 2a12À14 and 315 were reported decades ago, they
require harsh reaction conditions and, in our hands, were
low yielding.
We explored the reaction scope by incorporating groups
that might alter optoelectronic properties. The reac-
tion tolerated both electron-donating and -releasing sub-
stituents at the para position of the Grignard reagents
(entries 2À6, Table 1). An aryl Grignard reagent bearing 3,
5-dimethoxy groups reacted with 1a to provide 2g in good
yield (73%, entry 7). Albeit in diminished yields, an aryl
Grignard reagent with even more substituents was tolerated
under the reaction conditions (entry 8). The lower yields of 2h
and 2f are partially attributed to difficulties in the purification
(entries 6 and 8). Grignard reagents bearing ortho-alkyl
groups (entries 9 and 10) were not effective, likely because
additional steric hindrance slows the cross-coupling step. We
were unable to isolate any of the corresponding diarylindeno-
[1,2,3-fg]tetracene in an attempted reaction of 1a with 4-tri-
fluoromethylphenyl magnesium bromide (not shown).
To probe whether indene formation precedes or follows
the standard cross-coupling events, we prepared two po-
tential intermediates 4 and 5 (Scheme 3), which were
separately subjected to our above reaction conditions.
Indene 2a was formed in both experiments, indicating that
either 4 or 5 is a potential intermediate. Compound 4
proceeded cleanly to 2a (44%). By contrast, the conversion
of 5 to 2a was very low yielding (<5%), likely because of
the poor solubility of 5. If 5 is a viable intermediate in the
conversion of 1a to 2a, it must not build up in sufficient
amounts to precipitate from solution.
Herein, we report a new synthesis of 9,10-diarylindeno-
[1,2,3-fg]tetracenes via a chemoselective KumadaÀTamaoÀ
Corriu coupling/CÀH activation cascade reaction. In our
cascade, one pair of peri-chlorides undergoes standard
KumadaÀTamaoÀCorriu coupling while the other pair
undergoes the CÀH activation cascade. We reduced the
catalyst loading for the CÀH activation by an order of
magnitude. Our route is easily scaled and allows for late-
stage diversification. To the best of our knowledge, these
diarylindeno[1,2,3-fg]tetracenes constitute a new class of
acene materials for organic photovoltaic cells (OPVs).16À18
Our success in the cross-coupling reaction of 1a with
MeMgBr as a nucleophile6 using the PEPPSI-IPR19,20
catalyst encouraged us to explore the reaction of PhMgBr
with 1a. We did not isolate any rubrene (1b) or di-indene
(3) from this attempt. Rather, we isolated 9,10-
diphenylindeno[1,2,3-fg]tetracene (2a) in 84% yield as the
(9) Taima, T.; Sakai, J.; Yamanari, T.; Saito, K. Jpn. J. Appl. Phys.
2006, 45, L995.
(10) Perez, M. D.; Borek, C.; Forrest, S. R.; Thompson, M. E. J. Am.
Chem. Soc. 2009, 131, 9281.
(11) Najafov, H.; Lee, B.; Zhou, Q.; Feldman, L. C.; Podzorov, V.
Nat. Mater. 2010, 9, 938.
(12) IR spectrum: Naerland, A. Acta Chem. Scand. 1958, 12, 224.
(13) UVÀvis spectrum: Dufraisse, C.; Amiard, G. Compt. Rend.
1945, 220, 695.
(14) Melting point and combustion analysis: Dufraisse, C.; Badoche,
M. Compt. Rend. 1931, 193, 529.
(15) Lang, K. F.; Theiling, E. A. Chem. Ber. 1956, 89, 2734.
(16) Yoo, S.;Domercq, B.; Kippelen, B. Appl. Phys. Lett. 2004, 85, 5427.
(17) Yoo, S.; Domercq, B.; Kippelen, B. J. Appl. Phys. 2005, 97, 103706.
(18) Chu, C.; Shao, Y.; Shrotriya, V.; Yang, Y. Appl. Phys. Lett.
2005, 86, 243506.
Thin film absorption spectra can indicate the degree of
conjugation of the various functional groups (Figure 1).
Figure 1a compares the thin film absorption spectra of
indenes with R = H, R = F, and R = OMe. Figure 1b
compares changes in the thin film absorption for indenes
containing either three, six, or nine OMe groups. Increased
conjugation of the oxygen lone pairs (R = OMe) may
(19) Review: Kantchev, E. A. B.; O’Brien, C. J.; Organ, M. M. G.
Angew. Chem., Int. Ed. 2007, 46, 2768.
(20) Use in KumadaÀTamaoÀCorriu reactions: Organ, M. G.;
Abdel-Hadi, M.; Avola, S.; Hadei, N.; Nasielski, J.; O’Brien, C. J.;
Valente, C. Chem.;Eur. J. 2007, 13, 150.
(21) A sample of 2a was bench stable for up to 6 months in a
N2-flushed, clear glass scintillation vial, as determined by 1H NMR.
Org. Lett., Vol. 14, No. 6, 2012
1391

efficiencies were found for structures containing a 10À
15-nm-thick donor layer and a 40-nm-thick acceptor layer
of C60. All structures were capped with a 10-nm-thick
exciton blocking layer of 2,9-dimethyl-4,7-diphenyl-1,
10-phenanthroline (BCP),23 and a 70-nm-thick Al cathode
defining an active area that is 1 mm in diameter.
Table 1. Reaction Scopea
entry
R
product
% yieldb
1
2
3
4
5
6
7
8
9
10
H
2a
2b
2c
2d
2e
2f
82
4-F
4-Me
32
73c
76c
40d
31
4-nBu
4-tBu
4-OMe
3,5-di-OMe
3,4,5-tri-OMe
2-Me
2g
2h
2i
73
31
e
À
e
2,6-di-Me
2j
À
a Conditions: 1a, PEPPSI-IPR (3 mol %), ArMgBr (10 equiv), Et2O
(0.05 M of 1a), sealed vessel, 40 °C, 16 h. b Yields after purification
by alumina chromatography and recrystallization. c Purification by
alumina chromatography only. d PEPPSI-IPR (12 mol %), ArMgBr
(12 equiv), dioxane, sealed vessel, 80 °C, 24 h. Purification by silica gel
chromatography. e 2i or 2j was not observed in the mixture of products.
Figure 1. Thin film absorption spectra for compounds (a) 2a, 2b,
2f and (b) 2f, 2g, 2h.
cause a red shift observed in the spectra of compounds
2fÀh compared to 2a (R = H). Compound 2h (R = 3,4,
5-tri-OMe) is blue-shifted relative to 2g (R = 3,5-di-OMe)
even though 2h has additional methoxy groups. The all-
ortho substitution pattern likely prevents the oxygen lone
pairs from donating effectively to the π-system of 2h.
The photovoltaic performance of 2a, 2b, 2f, 2g, and 2h as
electron donors was evaluated by fabricating devices on
indiumÀtin oxide (ITO) coated glass substrates. A layer of
poly(3,4-ethylenedioxythiophene) poly(styrenesulfonate)
(PEDOT:PSS) was spun-cast on the substrate at a speed
of 7000 rpm for 45 s.22 The organic semiconductor materials
were deposited by high vacuum thermal sublimation
(<10À7 Torr) at a rate of 0.2 nm sÀ1. Optimum device
Scheme 3. Possible Intermediates in the Formation of 2aa
a Conditions A: PhMgBr (12 equiv), PEPPSI-IPR (10 mol %),
dioxane, 80 °C, 24 h. Conditions B: PhMgBr (10 equiv), PEPPSI-IPR
(3 mol %), Et2O, 40 °C, 16 h.
Figure 2. External quantum efficiency (ηEQE) spectra for devices con-
taining donor material compounds (a) 2a, 2b, 2f and (b) 2f, 2g, 2h.
(22) Roman, L. S.; Mammo, W.; Pettersson, L. A. A.; Andersson,
€
(23) Rand, B. P.; Li, J.; Xue, J.; Holmes, R. J.; Thompson, M. E.;
Forrest, S. R. Adv. Mater. 2005, 17, 2714.
M. R.; Inganas, O. Adv. Mater. 1998, 10, 774.
1392
Org. Lett., Vol. 14, No. 6, 2012

Device performance was characterized by measuring
current densityÀvoltage characteristics under AM 1.5G
solar simulated illumination. All of the donor (D) materi-
als exhibited a large open-circuit voltage (VOC) when used
with a C60 acceptor (A) due to their low dark current
characteristics, with the largest VOC of >1 V obtained for
the fluorinated compound (2b, Table 2). This is significant
as it is comparable to the highest VOC values reported for
singlejunctionOPVs.24,25 Peakvaluesofthefillfactor(FF)
and power conversion efficiency (ηP) exceed 0.5 and 1.5%,
respectively. This performance is comparable to that pre-
viously reported for planar heterojunction OPVs based on
the DÀA pairing of rubrene-C60.10
in the optical field in the device depending on the optimized
donor layer thickness.26,27
In conclusion, a direct and selective catalytic synthesis of
new donor materials for organic photovoltaic cells was
reported. The synthesis uses readily available 1a and aryl
Grignard reagents in a KumadaÀTamaoÀCorriu cross-
coupling/CÀH functionalization cascade. A variety of
intermediates can account for the formation of the pro-
ducts. These materials show high VOC and power conver-
sion efficiencies approaching 1.6% as donors in bilayer
OPVs. Further improvements in efficiency could be ex-
pectedina mixed DÀAÀheterojunction OPV architecture.
Acknowledgment. This work was supported by the
National Science Foundation (NSF) DMR-0819885 and
DMR-1006566, University of Minnesota: Initiative for
RenewableEnergyand the Environment (IREE), DuPont,
and 3M. We thank Dr. J. Dalluge, Dr. S. Francis, and
Mr. S. Murray (UMN) for assistance with mass spectro-
metry and Prof. G. Georg (UMN) for translating ref 15.
Table 2. Device Performance under AM 1.5G Solar Simulated
Illumination (44 mW/cm2)
JSC
(mA/cm2)a
VOC
ηP
(%)
donor
(V)
FF
2a (10 nm)
2b (15 nm)
2f (15 nm)
2g (10 nm)
2h (10 nm)
1.295
1.259
1.341
1.314
1.177
0.955
1.120
0.810
0.702
0.834
0.563
0.437
0.381
0.536
0.408
1.58
1.40
0.94
1.12
0.91
Supporting Information Available. Complete experi-
mental details and spectral data for all compounds de-
scribed (including FID files), crystallographic data (.cif
file) for 2a, and current densityÀvoltage characteristics
for devices containing donor material compounds 2a, 2b,
2f, 2g, and 2h. This material is available free of charge via
the Internet at http://pubs.acs.org.
a Short-circuit current density.
A clear response from each donor is observed in the
range of 500À700 nm (Figure 2). Slight variations in the
external quantum efficiency (ηEQE) corresponding to the
activity of C60 (λ = 350À400 nm) are attributed to changes
(27) Preliminary experiments indicate that 9,10-diphenylindeno-
[1,2,3-fg]tetracene (2a) reacts slowly with C60 in an NMR tube experi-
ment. The structure of the product of this reaction, and the implications
of this experiment with regard to devices fabricated from 2a and C60, is
currently under investigation. We thank a reviewer for suggesting this
course of investigation.
(24) Cheyns, D.; Rand, B. P.; Heremans, P. Appl. Phys. Lett. 2010,
97, 033301.
(25) Pandey, R.; Holmes, R. J. Adv. Mater. 2010, 22, 5301.
(26) Pettersson, L. A. A.; Roman, L. S.; Inganas, O. J. Appl. Phys
1999, 86, 487.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 6, 2012
1393