Influence of alkyl chain length on the solid-state packing and fluorescence of 1,4,5,8-tetra(alkyl)anthracenes

Article abstract of DOI:10.1080/15421400701617350

1,4,5,8-Tetra(alkyl)anthracenes (alkyl=methyl, ethyl, n-propyl, and n-hexyl) were prepared by a sequence of reactions of 1,2,4,5-tetrabromobenzene and 2,5-dialkylfurans in the presence of n-BuLi, hydrogenation, and treatment with acid. The influence of alkyl chain length on the packing patterns in the crystals and the fluorescent properties in the solid state was investigated. X-ray analysis revealed that the molecular structures can be classified into plane, semi-chair, and chair forms and that the packing patterns can be categorized into two-dimensional (herringbone) and one-dimensional (slipped-parallel) arrangements, in both of which there is no - stacking. In the case of the methyl, ethyl, and n-propyl derivatives, the wave shapes of the fluorescence spectra in the solid state resemble each other; on the other hand, the n-hexyl derivative displayed a slightly red-shifted and broader spectrum. The absolute quantum yield depended on the transition dipole moments because of the packing patterns and crystal rigidity. The n-propyl derivative demonstrated the highest quantum yield of f=0.85 among the tetra(alkyl)anthracenes.

Full text of DOI:10.1080/15421400701617350

Mol. Cryst. Liq. Cryst., Vol. 474, pp. 119–135, 2007
Copyright # Taylor & Francis Group, LLC
ISSN: 1542-1406 print=1563-5287 online
DOI: 10.1080/15421400701617350
Influence of Alkyl Chain Length on the
Solid-State Packing and Fluorescence
of 1,4,5,8-Tetra(alkyl)anthracenes
Chitoshi Kitamura
Yasushi Abe
Nobuhiro Kawatsuki
Akio Yoneda
Department of Materials Science and Chemistry, Graduate School of
Engineering, University of Hyogo, Himeji, Hyogo, Japan
Kohei Asada
Takashi Kobayashi
Hiroyoshi Naito
Department of Physics and Electronics, Graduate School of Engineering,
Osaka Prefecture University, Sakai, Osaka, Japan
1
,4,5,8-Tetra(alkyl)anthracenes (alkyl ¼ methyl, ethyl, n-propyl, and n-hexyl) were
prepared by a sequence of reactions of 1,2,4,5-tetrabromobenzene and 2,5-dialkyl-
furans in the presence of n-BuLi, hydrogenation, and treatment with acid. The
influence of alkyl chain length on the packing patterns in the crystals and the flu-
orescent properties in the solid state was investigated. X-ray analysis revealed that
the molecular structures can be classified into plane, semi-chair, and chair forms
and that the packing patterns can be categorized into two-dimensional (herring-
bone) and one-dimensional (slipped-parallel) arrangements, in both of which there
is no p–p stacking. In the case of the methyl, ethyl, and n-propyl derivatives, the
wave shapes of the fluorescence spectra in the solid state resemble each other; on
the other hand, the n-hexyl derivative displayed a slightly red-shifted and broader
spectrum. The absolute quantum yield depended on the transition dipole moments
because of the packing patterns and crystal rigidity. The n-propyl derivative
demonstrated the highest quantum yield of U
anthracenes.
f
¼ 0.85 among the tetra(alkyl)-
Keywords: alkyl-substituted anthracene; crystal structure; packing; quantum yield;
solid-state fluorescence
Address correspondence to Chitoshi Kitamura, Department of Materials Science and
Chemistry, Graduate School of Engineering, University of Hyogo, 2167 Shosha, Himeji,
Hyogo 671-2280, Japan. E-mail: kitamura@eng.u-hyogo.ac.jp
119

1
20
C. Kitamura et al.
INTRODUCTION
Solid-state packing effects play an important role in the performance
of electronic and photonic materials [1]. Although there have been
numerous reports on molecular modification by the introduction of
substituents into organic fluorophores to control fluorescent properties
in solution, to date, there has been relatively little research on the cor-
relation between solid-state packing patterns and fluorescent proper-
ties [2–7]. Thus, molecular design to control solid-state fluorescence is
not fully understood [4–6].
We report here the synthesis, crystal structures, and solid-state flu-
orescent properties of 1,4,5,8-tetra(alkyl)anthracenes 1a-d (Fig. 1). We
expected that by varying the alkyl chain length, both the molecular
packing behavior and solid-state fluorescence would be varied easily.
The former would depend on the balance of the intermolecular inter-
actions between aromatic moieties, such as face-to-face (p–p stacking)
and edge-to-face (CH-p) interactions, and the self-assembling ability of
alkyl side chains. Anthracene was selected because it is a representa-
tive fluorescent molecule as well as belonging to the acene family,
which has been receiving particular attention because of its impor-
tance as an organic semiconducting material [8]. It is well known that
unsubstituted oligoacenes prefer the herringbone structure because of
edge-to-face interactions between neighboring molecules in two-
dimensional (2-D) layers. As shown later, a large transition dipole
moment, which is one of the important factors leading to large fluores-
cence quantum yields, cannot be obtained in the herringbone (2-D)
packing pattern of anthracene moieties. It is possible that the lateral
alkyl chains act as spacers and modulate the spatial arrangement of
anthracene moieties. Accordingly, the fluorescent behavior in the solid
state can be adjusted by the molecular ordering of alkyl-substituted
anthracene molecules. Longer alkyl chains tend to form a lamellar
structure by van der Waals interactions. In the tetrathiafulvalene
(
TTF) series, it is well known that the longer alkyl chains serve as
FIGURE 1 1,4,5,8-Tetra(alkyl)anthracene 1a–d.

Solid-State Packing and Fluorescence
121
molecular fasteners [9], which form the molecular packing and stack-
ing of the TTF molecules, leading to high conductivity. We thought
that the intermolecular interactions between anthracene moieties
can be spatially tuned by the length of alkyl chains, and as a result,
the solid-state fluorescence can be chemically controlled.
EXPERIMENTAL
General
THF and toluene were distilled from LiAlH and CaH , respectively,
4
2
prior to use. Commercially available reagents containing 2,5-dimethyl-
furan (3a) were used as supplied unless otherwise stated. All reactions
were carried out under a nitrogen atmosphere unless otherwise noted.
1
,2,4,5-Tetrabromobenzene (4) was prepared according to the litera-
ture procedure [10]. Analytical thin-layer chromatography (TLC)
was performed on Merck silica-gel 60 F254 0.25 mm aluminium plates.
Column chromatography was performed on Wako silica gel C-300 (45–
7
5 mm). Melting points were determined on a Yanaco melting-point
1 13
apparatus and are uncorrected. H and
recorded on a Bruker-Biospin DRX500 FT spectrometer at 500 and
26 MHz, respectively. IR spectra were recorded on a Shimadzu
C NMR spectra were
1
FTIR-8400 spectrometer as KBr pressed pellets. Elemental analysis
was carried out on a Yanaco MT-5 CHN corder.
General Procedure for the Synthesis of 2,5-Dialkylfurans 3
The procedure described by Cook et al. [11] was modified. To an ice-
cooled mixture of furan 2 (2.0 mL, 27.6 mmol) and TMEDA (9.4 mL,
62.3 mmol), a solution of n-BuLi in n-hexane (40 mL, 1.54 M,
61.6 mmol) was added slowly, and then the mixture was refluxed for
1 h. During the heating time, the mixture changed to a brown suspen-
sion. The solution was allowed to be cooled to rt and cooled with an ice
bath. A solution of 1-bromoalkane (82.9 mmol) in THF (20 mL) was
added dropwise to the suspension. The mixture was stirred at rt for
2
1.5 h. After quenching with water, the crude product was extracted
with Et O, washed with brine, and dried over Na SO . After evapor-
2
2
4
ation of the solvent, the residue was subjected to silica-gel chromato-
graphy (CHCl =n-hexane ¼ 1:1) and dried under vacuum with a
3
mild heating to remove unreacted 1-bromoalkane and monosubsti-
tuted furan. The product obtained was used without further purifi-
cation in the next reaction except that the purification of 3b was
ꢀ
ꢀ
carried out by distillation (bp 138 C, lit. [11] 138 C).

1
22
C. Kitamura et al.
2
,5-Diethylfuran (3b)
1
Yield, 51%. H NMR (500 MHz, CDCl ) d 1.21 (t, J ¼ 7.4 Hz, 6H,
3
2
CH ), 2.60 (t, J ¼ 7.4 Hz, 4H, 2CH ), 5.85 (s, 2H, 3,4-H).
3
2
2
,5-Di(n-propyl)furan (3c)
1
Yield, 40%. H NMR (500 MHz, CDCl ) d 0.95 (t, J ¼ 7.4 Hz, 6H,
3
2
CH ), 1.60–1.67 (m, 4H, 2CH ), 2.54 (t, J ¼ 7.4 Hz, 4H, 2CH ), 5.85
3
2
2
(
s, 2H, 3,4-H).
2
,5-Di(n-hexyl)furan (3d)
1
Yield, 66%. H NMR (500 MHz, CDCl ) d 0.88 (t, J ¼ 6.8 Hz, 6H,
3
2
CH ), 1.26–1.37 (m, 4H, 2CH ), 1.55–1.63 (m, 12H, 6CH ), 2.56
3 2 2
(
t, J ¼ 7.6 Hz, 4H, 2CH ), 5.83 (s, 2H, 3,4-H).
2
General Procedure for the Synthesis of syn/anti
,4:5,8-Diepoxy-1,4:5,8-tetra(alkyl)anthracenes 5
1
A mixture of 3 (13.3 mmol) and 4 (2.10 g, 5.34 mmol) in toluene (45 mL)
ꢀ
was cooled to ꢁ30 C. To the mixture, a solution of n-BuLi in n-hexane
(10mL, 1.58 M, 15.8mmol) was added dropwise over 1 h. Then the mix-
ture was warmed up to rt over 2 h and stirred at rt for additional 5 h. After
quenching with water, the aqueous layer was extracted with CHCl . The
3
combined organic layer was washed with brine and dried over Na SO .
2
4
After evaporation, the residue was subjected to silica-gel chromato-
graphy (CHCl =n-hexane ¼ 1:1 ! CHCl ! CHCl =AcOEt ¼ 10:1) to
3
3
3
afford syn=anti bis(furan)adducts 5 as a mixture of a viscous orange oil
and a white solid (in the case of 5d, only a brown oil). In the case of
syn=anti 5c, washing the adducts with n-hexane provided pure anti 5c,
1
which was confirmed by X-ray analysis and was characterized by H
1
NMR [12]. H NMR of 5 showed that the syn=anti ratios were ca. 1:1.
syn/anti 1,4:5,8-Diepoxy-1,4:5,8-tetramethylanthracene (5a)
1
Yield, 70%. H NMR (500 MHz, CDCl ) d 1.86 (s, 12H, 4CH , anti),
3
3
1
2
.87 (s, 12H, 4CH , syn), 6.77 (s, 4H, 2,3,6,7-H, syn), 6.78 (s, 4H,
,3,6,7-H, anti), 6.96 (s, 2H, 9,10-H, anti), 6.97 (s, 2H, 9,10-H, syn).
3
syn/anti 1,4:5,8-Diepoxy-1,4:5,8-tetraethylanthracene (5b)
1
Yield, 49%. H NMR (500 MHz, CDCl ) d 1.15 (t, J ¼ 7.5 Hz, 24H,
3
8
CH , syn and anti), 2.21–2.34 (m, 16H, 8CH , syn and anti), 6.78 (s,
3 2
4
H, 2,3,6,7-H, anti), 6.79 (s, 4H, 2,3,6,7-H, syn), 6.92 (s, 2H, 9,10-H,
syn), 6.93 (s, 2H, 9,10-H, syn).
syn/anti 1,4:5,8-Diepoxy-1,4:5,8-tetra(n-propyl)anthracene (5c)
1
Yield, 66%. H NMR (500 MHz, CDCl ) d 1.04 (t, J ¼ 7.4 Hz,
3
2
4H, 8CH , syn and anti), 1.55–1.65 (m, 16H, 8CH , syn and anti),
3 2

Solid-State Packing and Fluorescence
123
2
.10–2.28 (m, 16H, 8CH , syn and anti), 6.75 (s, 4H, 2,3,6,7-H, anti),
2
6
.76 (s, 4H, 2,3,6,7-H, syn), 6.89 (s, 4H, 9,10-H, syn and anti).
syn/anti 1,4:5,8-Diepoxy-1,4:5,8-tetra(n-hexyl)anthracene (5d)
1
Yield, 42%. H NMR (500 MHz, CDCl ) d 0.90 (t, J ¼ 6.8 Hz, 24H,
3
8
2
CH , syn and anti), 1.32–1.63 (m, 64H, 32CH , syn and anti), 2.15–
3 2
.26 (m, 16H, 8CH , syn and anti), 6.74 (s, 4H, 2,3,6,7-H, anti), 6.76
2
(
s, 4H, 2,3,6,7-H, syn), 6.89 (s, 4H, 9,10-H, syn and anti).
General Procedure for the Synthesis of 1,4,5,8-
Tetra(alkyl)anthracene 1
syn=anti 5 (0.48 mmol) in EtOH (20 mL) and n-BuOH (1 mL) were
hydrogenated over 10% Pd=C (30 mg) under atmospheric pressure at
rt for 3 h. The catalyst was removed by filtration, and the filtrate
was evaporated under reduced pressure. To the residue, an ice-cooled
solution of conc. HCl (4 mL) and Ac O (20 mL) was added. The mixture
2
was stirred at rt for 3 h. After cooling with ice, water was added into
the mixture. The resultant mixture was extracted with CHCl , and
3
the extract was washed with aqueous Na CO and brine and dried
2
3
over Na SO . After evaporation of the solvent, column chromato-
2
4
graphy on silica gel with (CHCl =n-hexane ¼ 1:1) gave the correspond-
3
ing 1,4,5,8-tetra(alkyl)anthracene 1.
1
,4,5,8-Tetramethylanthracene (1a)
Compound 1a was obtained in 73% yield from 5a as a pale yellow
ꢀ
ꢀ
1
solid (from n-hexane), mp 219–221 C; lit. [13] 219 C. H NMR
(
(
500 MHz, CDCl ) d 2.81 (s, 12H, 4CH ), 7.22 (s, 4H, 2,3,6,7-H), 8.64
s, 2H, 9,10-H).
3
3
1
,4,5,8-Tetraethylanthracene (1b)
Compound 1b was obtained in 18% yield from 5b as a pale yellow
ꢀ
solid (from n-hexane), mp 120–121 C. The yield was the lowest
because some unidentified products were formed.
1
H
NMR
(
500 MHz, CDCl ) d 1.48 (t, J ¼ 7.4 Hz, 12H, 4CH ), 3.25 (q,
3
3
J ¼ 7.4 Hz, 8H, 4CH ), 7.26 (s, 4H, 2,3,6,7-H), 8.83 (s, 2H, 9,10-H).
2
1
3
C NMR (126 MHz, CDCl ) d 14.81, 26.11, 119.59, 123.63, 129.98,
3
1
8
9
38.28. FTIR (KBr) 2960.5, 2935.5, 2873.7, 1618.2, 1460.0, 1296.1,
ꢁ1
42.8 cm . Anal. calcd. for C H : C, 90.98; H, 9.02%. Found: C,
2
2
26
0.73; H, 9.20%.
1
,4,5,8-Tetra(n-propyl)anthracene (1c)
Compound 1c was obtained in 74% yield from 5c as a pale yellow
ꢀ
1
solid (from n-hexane), mp 106–108 C. H NMR (500 MHz, CDCl3)

1
24
C. Kitamura et al.
d 1.08 (t, J ¼ 7.3 Hz, 12H, 4CH ), 1.85–1.92 (m, 8H, 4CH ), 3.16 (t,
3
2
J ¼ 7.7 Hz, 8H, 4CH ), 7.22 (s, 4H, 2,3,6,7-H), 8.79 (s, 2H, 9,10-H).
2
13
C NMR (126 MHz, CDCl ) d 14.44, 23.80, 35.61, 119.96, 124.68,
3
1
8
30.06, 136.77. FTIR (KBr) 2956.7, 2929.7, 2869.9, 1629.7, 1465.8,
73.7, 835.1 cm . Anal. calcd. for C H : C, 90.11; H, 9.89%. Found:
26 34
ꢁ1
C, 90.10; H, 10.04%.
1
,4,5,8-Tetra(n-hexyl)anthracene (1d)
Compound 1d was obtained in 72% yield from 5d as a pale yellow
ꢀ
1
solid (from n-hexane), mp 94–96 C. H NMR (500 MHz, CDCl ) d
3
0
.90 (t, J ¼ 6.9 Hz, 12H, 4CH ), 1.31–1.40 (m, 8H, 4CH ), 1.48–1.53
3
2
(
m, 16H, 8CH ), 1.81–1.87 (m, 8H, 4CH ), 3.17 (t, J ¼ 7.6 Hz, 8H,
2
2
13
4
CH ), 7.22 (s, 4H, 2,3,6,7-H), 8.80 (s, 2H, 9,10-H).
C NMR
2
(
126 MHz, CDCl ) d 14.16, 22.79, 29.84, 30.93, 31.93, 33.70, 120.02,
3
1
8
8
24.62, 130.07, 137.06. FTIR (KBr) 2956.7, 2925.8, 2850.6, 1467.7,
ꢁ1
73.7 cm . Anal. calcd. for C H : C, 88.65; H, 11.35%. Found: C,
3
8
58
8.93; H, 11.40%.
Crystal Structure Determinations
Crystals of compounds 1a–d were grown from n-hexane. The single-
crystal data were collected on a Rigaku=Mercury CCD diffractometer
˚
with Mo-Ka radiation (k ¼ 0.71070 A). The structure was solved by a
direct method using SIR92 [14] and refined by the full matrix least-
2
squares method on F with anisotropic temperature factors for non-
hydrogen atoms. Hydrogen atoms were refined isotropically. All of
the calculations were performed using the teXsan program packages
[
15]. Full crystallographic details, excluding structure factors, for
the structures of compounds 1a (CCDC 296864), 1b (CCDC 296865),
c (CCDC 296866), and 1d (CCDC 296867) have been deposited with
1
the Cambridge Crystallographic Data Centre. Copies of the data can
be obtained, free of charge, via http://www.ccdc.cam.ac.uk/conts/
retrieving.html (or form the Cambridge Crystallographic Data Centre,
1
2 Union Road, Cambridge, CB2 1EZ, UK; Fax: þ44 1223 336033;
e-mail: deposit@ccdc.ca.ac.uk).
Crystal Data for 1a
C H , M ¼ 234.34, monoclinic, P2 =c, a ¼ 9.880(2), b ¼ 5.0470(5),
1
8
18
1
ꢀ
3
˚
˚
c ¼ 13.239(2) A, b ¼ 101.501(7) , V ¼ 646.9(2) A , Z ¼ 2, T ¼ 223 K,
ꢁ3
ꢁ1
D ¼ 1.203 g cm , l (MoKa) ¼ 0.67 cm , 5172 reflections measured,
c
4
936 unique (Rint ¼ 0.025), GoF ¼ 0.89, 118 parameters refined, final
R-factors R ¼ 0.053 for 1164 reflections with I > 2r(I), wR ¼ 0.213
1
2
for all data.

Solid-State Packing and Fluorescence
125
Crystal Data for 1b
C H , M ¼ 290.45, monoclinic, P2 =n, a ¼ 10.208(1), b ¼ 5.1793(5),
2
2
26
1
ꢀ
3
˚
˚
c ¼ 15.681(2) A, b ¼ 92.587(7) , V ¼ 828.2(2) A , Z ¼ 2, T ¼ 223 K,
ꢁ3
ꢁ1
D ¼ 1.165 g cm , l (MoKa) ¼ 0.65 cm , 6381 reflections measured,
c
1
835 unique (Rint ¼ 0.023), GoF ¼ 1.58, 152 parameters refined, final
R-factors R ¼ 0.052 for 1627 reflections with I > 2r(I), wR ¼ 0.155
1
2
for all data.
Crystal Data for 1c
C H , M ¼ 346.55, triclinic, P 1, a ¼ 6.5551(1), b ¼ 7.9150(2),
ꢀ
2
6
34
ꢀ
˚
c ¼ 10.9768(1) A, a ¼ 108.01(1), b ¼ 72.29(1), c ¼ 102.05(2) , V ¼
3
ꢁ3
ꢁ1
˚
5
3
1
11.80(5) A , Z ¼ 1, T ¼ 223 K, D ¼ 1.124 g cm , l (MoKa) ¼ 0.62 cm
,
c
792 reflections measured, 2205 unique (Rint ¼ 0.051), GoF ¼ 1.61,
18 parameters refined, final R-factors R ¼ 0.063 for 2043 reflec-
1
tions with I > 2r(I), wR ¼ 0.217 for all data.
2
Crystal Data for 1d
ꢀ
C H , M ¼ 514.88, triclinic, P 1, a ¼ 4.731(4), b ¼ 12.81(1),
3
8
58
ꢀ
3
˚
˚
c ¼ 14.13(1) A, a ¼ 78.45(2), b ¼ 77.32(2), c ¼ 84.94(3) , V ¼ 817(1) A ,
ꢁ3
ꢁ1
Z ¼ 1, T ¼ 173 K, D ¼ 1.046 g cm , l (MoKa) ¼ 0.58 cm , 6439
c
reflections measured, 3638 unique (R ¼ 0.048), GoF ¼ 1.30, 288
int
parameters refined, final R-factors R ¼ 0.095 for 2310 reflections
1
with I > 2r(I), wR ¼ 0.186 for all data.
2
Fluorescence Measurements in the Solid State
The crystals were excited by He-Cd laser (Kimmon, IK3301R-G), and
the spectra were recorded using a calibrated optical multichannel ana-
lyzer (Hamamatsu Photonics, PMA11). The measurement of the absol-
ute quantum yield was performed using an integrating sphere
(
Labsphere, IS-040-SF).
RESULTS AND DISCUSSION
Synthesis and Optical Properties in Solution
The synthesis of 1,4,5,8-tetramethylanthracene from 2,5-dimethyl-
furan and transient bis(aryne) using the Diels–Alder reaction has
been reported [12]. We improved the method to obtain alkyl-substi-
tuted anthracenes simply (Scheme 1). First, we prepared 2,5-dialkyl-
0
furans 3 by direct dilithiation of furan 2 in the presence of N,N,N ,
0
N -tetramethylethylenediamine (TMEDA) in 40–66% yield, although

1
26
C. Kitamura et al.
SCHEME 1 Reagents and conditions: i, n-BuLi (2.2 equiv.), TMEDA (2.2
equiv.), n-hexane, reflux; ii, 1-bromoalkane (3 equiv.), THF, rt, 40–66%
ꢀ
(
H
two-steps, from 2); iii, n-BuLi (3 equiv.), toluene, ꢁ30 C to rt, 42–70%; iv,
2
, 10% Pd=C, 20:1 EtOH–n-BuOH; v, conc. HCl, Ac
2
O, 18–74% (two-steps,
from 5).
3
could be obtained by a two-step lithiation=alkylation procedure that
was time-consuming [11]. Next, 1,2,4,5-tetrabromobenzene (4) in the
presence of 2.5 equiv. of 3 was treated with 3 equiv. of n-BuLi to give
the corresponding anthracene intermediate 5 in 42–70% yield as a
mixture of syn and anti stereoisomers. The syn=anti ratios in 5 were
1
ca. 1:1 as revealed by H NMR. In the case of 5c, the anti isomer could
be isolated as a white solid purely by filtration and washing with
n-hexane and was characterized by X-ray crystallography [16]. On
1
the basis of the H NMR spectrum of isolated anti-5c, the assignment
of signals of the syn isomer in a syn=anti mixture became possible.
Thus, understanding the chemical shifts of syn and anti isomers led
1
to the differentiation between syn and anti isomers in H NMR. Trans-
formation of 5 into anthracene 1 was achieved by a facile sequence of
catalytic hydrogenation in alcohols and then dehydration of 6 with
acid in 18–74% yield. The solubility of compounds 1a–d in organic sol-
vents was investigated, and they showed good solubility in hydro-
carbon solvents such as n-hexane.
The absorption and fluorescence spectra of 1a–d in n-hexane are
shown in Fig. 2, and the spectral data are summarized in Table 1.

Solid-State Packing and Fluorescence
127
FIGURE 2 (a) Absorption and (b) fluorescence spectra in n-hexane of 1a
ꢁ
5
(
red), 1b (blue), 1c (green), and 1d (black). Concentration: 1a, 5.53 ꢂ 10
ꢁ5 ꢁ5 ꢁ5
M; 1b, 4.48 ꢂ 10 M; 1c, 3.79 ꢂ 10 M; 1d, 4.55 ꢂ 10 M.
As we expected, there is no marked differences in the wave shape, the
wavelengths of the absorption and fluorescence peaks, and the fluores-
cence quantum yield (U values are around 0.3) among 1a–d because
f
the molecules exist in an almost monodispersed state in a dilute sol-
ution. Both the absorption and fluorescence spectra showed the
vibrational structure, in which the band assigned as the 0–1 transition
was the highest, and their peaks shifted slightly to longer wavelengths
with increasing alkyl chain length.
TABLE 1 UV–Vis Absorption and Fluorescence Properties of 1a–d in
n-Hexane
a
Fluorescence spectral data
Absorption spectral
data , kmax (log e) (nm)
a
b
c
Compound
k
em (nm)
U
f
1
1
1
1
a
b
c
350 (3.71), 368 (3.88), 388 (3.83)
352 (3.74), 369 (3.92), 391 (3.87)
355 (3.78), 373 (3.97), 394 (3.92)
356 (3.71), 374 (3.90), 395 (3.85)
395, 415, 440
399, 420, 444
401, 423, 448
402, 425, 450
0.32
0.25
0.31
0.36
d
a
Peaks based on the 0–0, 0–1, and 0–2 transitions.
Excited at 355 nm.
b
c
Fluorescence quantum yields were determined using 9,10-diphenylanthracene as the
standard.

1
28
C. Kitamura et al.
X-ray Crystal Structures
To investigate the effect of changing the length of the alkyl chain on
the packing pattern in the crystal, we carried out X-ray crystallo-
graphic analysis of 1a–d. With an increase in the alkyl chain length,
the interactions between alkyl chains were expected to induce stron-
ger molecular ordering. All single crystals were obtained by slow evap-
oration from their n-hexane solutions. The crystal systems were the
monoclinic space group P2 =c with Z ¼ 2 for 1a, monoclinic space
1
ꢀ
group P2 =n with Z ¼ 2 for 1b, triclinic space group P 1 with Z ¼ 1
1
ꢀ
for 1c, and triclinic space group P 1 with Z ¼ 1 for 1d. The molecular
packing patterns in the crystals are depicted in Fig. 3. All of the mole-
cules of 1a–d possess a center of symmetry, and halves of the formula
units are crystallographically independent because of the high sym-
metry of the molecules. The anthracene units are strictly planar.
Interestingly, longer alkyl side chains than ethyl groups produced a
FIGURE 3 Packing diagrams for (a) 1a, (b) 1b, (c) 1c, and (d) 1d.

Solid-State Packing and Fluorescence
129
striking change in the packing patterns. When viewed down the long
molecular axis, both 1a and 1b appear to be stacked in the herring-
bone structure (Figs. 3a and 3b) in a similar manner to nonsubstituted
anthracene [17], indicating 2-D arrangements. In addition, the pack-
ing patterns do not adopt p–p stacking (face-to-face overlap of anthra-
cene moieties). The interplanar tilt angles between anthracene rings
ꢀ
ꢀ
in two adjacent columns are 91.6 for 1a and 85.4 for 1b, respectively.
On the other hand, the anthracene rings in both 1c and 1d adopt a
slipped-parallel pattern without p–p stacking along the anthracene
column direction (Figs. 3c and 3d), exhibiting one-dimensional (1-D)
arrangements. Further, the anthracene units in 1c and 1d form an
anthracene plane.
In the crystals of 1a and 1b, the alkyl side chains are essentially
planar with the anthracene ring, as shown in Fig. 4a. In the crystal
of 1c, a pair of two n-propyl groups at the 1 and 5 positions take a
coplanar conformation with the anthracene ring, while another pair
of two n-propyl groups at the 4 and 8 positions extend upward and
downward toward the out-of-plane anthracene directions (Fig. 4b).
In addition, 1d in the crystal is a stair-like molecule in which two
n-hexyl groups at the 1 and 8 positions extend upward and another
two n-hexyl groups at the 4 and 5 positions extend downward out of
the anthracene plane (Fig. 4c). The alkyl chains of 1c and 1d take
an all-trans conformation like other hydrocarbon chains frequently
found in X-ray analysis. We refer to the molecular structures of 1b,
1
c, and 1d as plane, semi-chair, and chair forms, respectively. The for-
mation of three different conformations results from the torsion
degrees of freedom in the alkyl chain.
Figure 5 shows the stacking patterns of two vicinal molecules.
Regarding the solid-state fluorescence, the translational shift of neigh-
boring molecules in the stack is an essential factor to understanding
the fluorescence behavior. In 1a–d, distinct stacking patterns are
observed. Undoubtedly, there is no spatial overlap of anthracene moi-
eties on adjacent molecules. The slip distances between neighboring
˚
molecules are 5.05, 5.18, 6.56, and 4.74 A for 1a, 1b, 1c, and 1d,
respectively. Compound 1c has the longest slip distance, and 1d has
the shortest distance. As shown in Fig. 3, the interplanar distances
˚
between anthracene moieties are 3.64, 3.54, 3.53, and 3.47 A for 1a,
1
b, 1c, and 1d, respectively. The shortest distance for overlapping
˚
nonbonded atoms in anthracene for 1a is 3.67–3.68 A, which is longer
˚
˚
compared with other anthracenes (3.58 A for 1b, 3.59 A for 1c, and
˚
.56–3.59 A for 1d). As a result, the displacements between neighbor-
3
ing anthracene moieties are the largest in 1c and the smallest in 1d. It
can be recognized that the hexyl groups in 1d greatly affect the dense

1
30
C. Kitamura et al.
FIGURE 4 Molecular structures of (a) 1b, (b) 1c, and (c) 1d. Upper, middle,
and lower structures show top view, side view, and view along the long
molecular axis, respectively.
packing arrangement, suggesting that molecular fasteners are effec-
tive in crystallization.
In a series of these observations, the length of the alkyl side chains
strongly affects not only the molecular geometry but also the crystal
packing arrangement. The interactions related to the specific arrange-
ments can be roughly classified into two types: (1) interactions
between aromatic units and (2) interactions between alkyl chains.
The former may be more favorable than the latter in 1a and 1b where
the herringbone structure, frequently found in oligoacenes [8], can be

Solid-State Packing and Fluorescence
131
FIGURE 5 Stacking patterns of two vicinal molecules of (a) 1a, (b) 1b, (c), 1d,
and (d) 1d. Hydrogen atoms are omitted for clarity.
seen. On the other hand, the elongation of alkyl chains over 2 carbon
atoms seems to enhance the latter. In 1c and 1d, the molecules are
arranged in stacks showing a parallel orientation of the anthracene
moiety, and two columnar structures consisting of aromatic and alkyl
regions can be observed. In particular, the hexyl groups in 1d promote
the lamellar structure and lead to the dense packing of anthracene
moieties.
Fluorescence in the Solid State
To examine the influence of crystal packing on the solid-state fluores-
cence properties, the fluorescence spectra of crystals of 1a–d and the
absolute quantum yields were measured. The fluorescence spectra of
crystals of 1a–d are shown in Fig. 6, and their spectral data are

1
32
C. Kitamura et al.
FIGURE 6 Fluorescence spectra of 1a (red), 1b (blue), 1c (green), and 1d
black) in the solid state, excited at 325 nm.
(
summarized in Table 2. The fluorescence properties in the solid state
are dependent on packing pattern. All of the fluorescence spectra in
the solid state are red-shifted with respect to those in solution, indicat-
ing the presence of intermolecular interactions in the crystals. The
crystals of 1a–c illustrate similar sharp fluorescence spectra, which
can be attributed to free exciton luminescence. Fluorescence maxima
were observed at 436–439 nm. On the other hand, the crystal of 1d
illustrates a slightly red-shifted and broader fluorescence spectrum
with a fluorescence maximum at 447 nm and a shoulder peak around
TABLE 2 Structural and Fluorescence Properties of 1a–d in the Solid State
Structural data
Fluorescence spectral data
a
b
Compound
Molecular structure
Packing mode
k
em =nm
U
f
1
1
1
1
a
b
c
Plane form
Plane form
Semi-chair form
Chair form
herringbone
herringbone
slipped-parallel
slipped-parallel
438, 462
439, 471
436, 462
447
0.39
0.68
0.85
0.36
d
a
Excited at 325 nm.
b
Absolute quantum yield in the solid state.

Solid-State Packing and Fluorescence
133
4
80 nm. It is well known that in pyrene crystal, the pyrene excimer
forms a more stable excited state compared with the free exciton state
because of large lattice-relaxation energy, and red-shifted and broad
fluorescence derived from the excimer is dominant [18]. Another fea-
ture of excimer emission is its low quantum yield, which results from
enhancement of nonradiative decay of excited states because of lattice
relaxation to form the excimer. In the crystal of 1d, the molecular
packing is not expected to be rigid because of an abundance of flexible
long alkyl groups; the slip distance between two adjacent anthracenes
˚
in one column (4.73 A) is the shortest among 1a–d (slip distances: 1a,
˚ ˚ ˚
5
.05 A; 1b, 5.18 A; 1c, 6.56 A), probably resulting in anthracene exci-
mer formation and relatively low quantum yield (U ¼ 0.36). On the
f
other hand, the quantum yields of crystals of 1a (U ¼ 0.39), 1b
f
(
U ¼ 0.68), and 1c (U ¼ 0.85) can be understood in terms of their
f
f
packing patterns. According to free exciton theory, all of the molecules
within the coherent size of a free exciton contribute to enhance the
macroscopic transition dipole moment between the lowest excited
state and the ground state. However, when the crystal adopts a 2-D
arrangement as well as has two molecules in the unit cell, the excited
state of the molecule in the crystal splits into two energy levels (Davy-
dov splitting), and then the transition dipole moment from the lower
split level to the ground state (this transition dipole moment is an
essential factor in the fluorescence quantum yield) can be zero,
depending on the packing patterns [19]. In such a case, the magnitude
of the solid-state fluorescence can be roughly estimated from linear
combination of the transition dipole moments of the two molecules
that have different orientations in the unit cell. Because, in the crystal
of 1a, the transition dipole moments of the two anthracenes (two dis-
tinguishable directions along the short molecular axes of anthracene
ꢀ
moieties in two adjacent columns) are perpendicular (90.0 ) to each
other, only 50% of the possible transition dipole moment is distributed
to the transition from the lower split state in the crystal. In the crystal
of 1b, the angle between directions along the short molecular axes of
ꢀ
the two anthracene moieties in the unit cell (117.2 ) is larger than
9
ꢀ
0 , and consequently more than 50% of the possible transition dipole
moment contributes to the transition from the lower split state to the
ground state. In contrast to the crystals of 1a and 1b, the crystal of 1c
has a parallel 1-D arrangement and has one molecule in the unit cell,
and so all possible transition dipole moments concentrate on the tran-
sition from the lowest excited state to the ground state. Therefore,
under assumptions that the crystals of 1a–c have similar nonradiative
decay rates and that the decrease in crystal rigidity is not serious
within the side-chain length of the n-propyl group, a higher quantum

1
34
C. Kitamura et al.
yield is expected as the transition dipole moment increases, i.e.,
c > 1b > 1a. In some organic materials, p–p stacking results in sup-
1
pression of the fluorescence quantum yield [5a,20]. In these materials,
the transition dipole moments of the molecules in the unit cell are
almost parallel to each other, and such molecular arrangements do
not distribute any transition dipole moments to the transition between
the lowest excited state and the ground state. Thus, the crystal of 1c
demonstrates that the conditions of a 1-D packing pattern, especially
the same direction of transition dipole moments and not too-soft crys-
tal rigidity, would play a significant role in the high fluorescence quan-
tum yield in the solid state.
CONCLUSIONS
1
,4,5,8-Tetra(alkyl)anthracenes (1a–d) were prepared from the
Diels–Alder reaction of bis(aryne) and 2,5-dialkylfurans. The structur-
al features in the crystals were characterized by X-ray analysis. The
change in alkyl side chain length of anthracene led to a drastic differ-
ence in molecular and crystal structures. In the packing patterns, the
methyl and ethyl derivatives (1a and 1b) have 2-D arrangements; on
the other hand, the n-propyl and n-hexyl derivatives (1c and 1d) adopt
1
-D arrangements. The packing mode affected the fluorescence
properties in the solid state. The fluorescence spectra of 1a–c exhib-
ited similarity in shape and wavelength, whereas 1d was slightly
red-shifted and had a broader spectrum. Among 1a–d, the crystal of
1
c exhibited the most intense fluorescence in the solid state. The fluor-
escence quantum yields in the solid state correlate with the transition
dipole moments derived from arrangements of the anthracene ring in
the unit cell as well as the crystal rigidity.
ACKNOWLEDGMENTS
We thank the Instrument Center, the Institute for Molecular Science,
for assistance in obtaining X-ray data. This work was supported by a
grant-in-aid (No. 18750123) from the Ministry of Education, Culture,
Sports, Science, and Technology, Japan. The Hyogo Science and Tech-
nology Association is gratefully acknowledged for financial support.
REFERENCES
[
[
1] Curtis, M. D., Cao, J., & Kampf, J. W. (2004). J. Am. Chem. Soc., 126, 4318.
2] Langhals, H., Potrawa, T., N o¨ th, H., & Linti, G. (1989). Angew. Chem., Int. Ed., 28, 478.

Solid-State Packing and Fluorescence
135
[
[
3] Cotrait, M., Marsau, P., Kessab, L., Grelier, S., Nourmanode, A., & Castellan, A.
1994). Aust. J. Chem., 47, 423.
4] (a) Yoshida, K., Ooyama, Y., Miyazaki, H., & Watanabe, S. (2002). J. Chem. Soc.,
Perkin Trans., 2, 700; (b) Yoshida, K., Ooyama, Y., Tanikawa, S., & Watanabe, S.
(
(2002). J. Chem. Soc., Perkin Trans., 2, 708; (c) Ooyama, Y., Nakamura, T., &
Yoshida, K. (2005). New J. Chem., 29, 439; (d) Ooyama, Y. & Yoshida, K. (2005).
New J. Chem., 29, 1204.
[
[
5] (a) Mizobe, Y., Tohnai, N., Miyata, M., & Hasegawa, Y. (2005). Chem. Commun.,
1
(
&
839; (b) Mizobe, Y., Ito, H., Hisaki, I., Miyata, M., Hasegawa, Y., & Tohnai, N.
2006). Chem. Commun., 2126; (c) Mizobe, Y., Miyata, M., Hisaki, I., Hasegawa, Y.,
Tohnai, N. (2006). Org. Lett., 8, 4295.
6] (a) Matsumoto, S., Tokunaga, W., Miura, H., & Mizuguchi, J. (2001). Bull. Chem.
Soc. Jpn., 74, 471; (b) Horiguchi, E., Matsumoto, S., Funabiki, K., & Matsui, M.
(2005). Bull. Chem. Soc. Jpn., 78, 1167; (c) Matsumoto, S., Uchida, Y., & Yanagita,
M. (2006). Chem. Lett., 35, 654; (d) Horiguchi, E., Matsumoto, S., Funabiki, K., &
Matsui, M. (2006). Bull. Chem. Soc. Jpn., 79, 799.
[
7] Dreuw, A., Pl o¨ tner, J., Lorenz, L., Wachtveitl, J., Djanhan, J. E., Br u¨ ning, J.,
Metz, T., Bolte, M., & Schmidt, M. U. (2005). Angew. Chem., Int. Ed., 44, 7783.
8] Bendikov, M., Wudl, F., & Perepichka, D. F. (2004). Chem. Rev., 104, 4891.
9] (a) Inokuchi, H., Saito, G., Wu, P., Seki, K., Tang, T. B., Mori, T., Imaeda, K., &
Yasuoka, N. (1986). Chem. Lett., 15, 1263; (b) Imaeda, K., Enoki, T., Shi, Z., Wu,
P., Okada, N., Yamochi, H., Saito, G., & Inokuchi, H. (1987). Bull. Chem. Soc.
Jpn., 60, 3163.
[
[
[
[
[
10] Hart, H., Bashir-Hashemi, A., Luo, J., & Meador, M. A. (1986). Tetrahedron, 42,
641.
11] McKeown, N. B., Chambrier, I., & Cook, M. J. (1990). J. Chem. Soc., Perkin Trans.,
, 1169.
12] Hart, H., Lai, C., Nwokogu, G., Shamouilian, S., Teuerstein, A., & Zlotogorski, C.
1980). J. Am. Chem. Soc., 102, 6651.
1
1
(
[
[
13] Hart, H., Raju, N., Meador, M. A., & Ward, D. L. (1981). J. Org. Chem., 46, 1251.
14] Altmare, A., Burla, M. C., Camalli, M., Cascarano, M., Giacovazzo, C., Guagliardi,
A., & Polidori, G. (1994). J. Appl. Cryst., 27, 435.
[15] Crystal Structure Analysis Package, Molecular Structure Corporation (1985, 1999).
The Woodlands, Texas.
[16] Kitamura, C., Abe, Y., Ouchi, M., & Yoneda, A. (2004). Anal. Sci., 20, x27.
[17] Brock, C. P. & Dunitz, J. D. (1990). Acta Cryst., B46, 795.
[18] Song, K. S. & Williams, R. T. (1993). Self-Trapped Excitons, Springer-Verlag:
Berlin, Chap. 8, p. 300.
[
[
19] Davydov, A. S. (1971). Theory of Molecular Excitons, Plenum Press: New York.
20] Spano, F. C. (2005). J. Chem. Phys., 122, 234701.