Thioindigo precursor: Control of polymorph of thioindigo

Article abstract of DOI:10.1016/j.tetlet.2004.10.028

The precursor was converted to thioindigo with P2₁/c structure by heating. 2,2′-Bi[-3(2H)-thiophenon]ylidene with two bicyclo[2.2.2]-octadiene moieties was quantitatively converted to thioindigo by the retro-Diels-Alder reaction. The thioindigo solid obtained from the precursor had the P2₁/c structure which was different from the commercial thioindigo samples (P2₁/n).

Full text of DOI:10.1016/j.tetlet.2004.10.028

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 9083–9086
Thioindigo precursor: control of polymorph of thioindigo
b
a,
Hidemitsu Uno, Kana Moriyama, Takayuki Ishikawa,
a
*
c
Noboru Ono and Hidenori Yahiro
d
a
Division of Synthesis and Analysis, Department of Molecular Science, Integrated Center for Sciences, Ehime University,
Bunkyo-cho 2-5, Matsuyama 790-8577, Japan
b
Inkjet Media Design Dept. 2, Inkjet Media Design Div., Inkjet Supply Material Development Center, Canon Inc.,
1
Department of Chemistry, Faculty of Science, Ehime University, Bunkyo-cho 2-5, Matsuyama 790-8577, Japan
6-1 Shimonoge 3-chome, Takatsu-ku, Kawasaki 213-8512, Japan
c
d
Department of Applied Chemistry, Faculty of Engeneering, Ehime University, Bunkyo-cho 3, Matsuyama 790-8577, Japan
Received 30 April 2004; revised 30 September 2004; accepted 6 October 2004
0
Abstract—2,2 -Bi[-3(2H)-thiophenon]ylidene with two bicyclo[2.2.2]-octadiene moieties was quantitatively converted to thioindigo
1
by the retro-Diels–Alder reaction. The thioindigo solid obtained from the precursor had the P2 /c structure which was different from
the commercial thioindigo samples (P2 /n).
1
Ó 2004 Elsevier Ltd. All rights reserved.
Indigo and thioindigo are known as vat dyestuffs and
have been used for coloring textiles since ancient times.
Their reduced precursors are called leuco forms and well
soluble in alkaline media. When the leuco precursors are
exposed to air, they are oxidized to the corresponding
insoluble dyes. Although their properties as the vat dye-
stuffs are very good, the low solubility of the dyestuffs
ð1Þ
themselves in common solvents limits their use just as
coloring material. In order to employ these pigments
as functional materials such as materials for memory
storage, equivalents of the pigments which can be quan-
titatively converted to the pigments without any by-
product, are required. Morphological control of such
The targeted equivalent 1 was prepared from commer-
cially available ethyl propiolate in six steps (Scheme 1).
In the literature, ethyl phenylsulfonylpropiolate (3)
1
dyestuffs is also one of the important problems. We
have already succeeded in the conversion of a precursor
to tetrabenzoporphyrins for the application of field
effect transistors by the thermal retro-Diels–Alder reac-
2
tion. We planned to utilize the methodology for
indigo dyes. In this communication, we discuss about
synthesis, thermal behaviors, and crystal structures of
the thioindigo equivalent 1 as well as thioindigo itself
(
Eq. 1).
Scheme 1. Reagents, Conditions, and Yields: (i) LiHMDS, PhSO
SPh, THF,ꢀ78°C; 95%. (ii) Dimethyldioxirane, acetone, rt. (iii) 1,3-
Cyclohexadiene, CHCl , rt; 84% (two steps). (iv) Ethyl thioglycolate,
2
Keywords: Thioindigo; Polymorphism; Retro-Diels–Alder reaction;
Soluble precursor.
3
NaH, THF, 0°C; 5a: 96%; thioglycolic acid, NaH, DMF, rt; 5b: 88%.
(v) NaOEt, toluene, 80°C; 6a: 79%; LDA, THF,ꢀ78°C; H SO (cat),
*
Corresponding author. Tel.: +81 89 927 9660; fax: +81 89 927
670; e-mail: uno@dpc.ehime-u.ac.jp
2
4
9
2 2
CH ClCH Cl, rt; 7: 74%.
0
040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.10.028

9084
H. Uno et al. / Tetrahedron Letters 45 (2004) 9083–9086
was obtained in good yield from ethyl propiolate by
reaction with LDA and S-phenyl benzenethiosulfate
3
followed by oxidation with mCPBA. In our hands,
however, we encountered low yields and poor
reproducibility. In order to suppress the by-product for-
mation derived from nucleophilic addition of diisoprop-
ylamine in the first step, LiHMDS was used instead
of LDA and ethyl phenylsulfanylpropiolate (2) was
obtained in 95% yield. Oxidation of 2 with mCPBA gave
3, but we could isolate 3 in low yields (less than 30%)
probably due to highly sensitive nature of 3 toward
nucleophiles. We changed the oxidation reagent to
4
dimethyldioxirane and used 3 in the next step without
3
purification. Thus, the Diels–Alder adduct 4 was
obtained in good yields (84%). Formation of thiophene
ring was achieved by the Diekmann condensation.
The Diels–Alder adduct 4 was treated with ethyl
thioglycolate and NaH in THF to give 5a in 96% yield.
Treatment of 5a with NaOEt gave 3-hydroxythiophene-
Figure 1. TG–DTA experiment of 1 to thioindigo. The precursor 1
(2.94mg) was heated at a rate of 20°C/min under a nitrogen
atmosphere.
5
The XRD experiments of 1, its thermally decomposed
sample (200°C, 5min), and commercial thioindigo were
carried out and results are shown in Fig. 2. The powder
pattern of thioindigo (Fig. 2a) obtained by the thermal
decomposition of 1 was quite different from that of com-
2
-carboxylate 6a in 79% yield. In spite of all our efforts,
removal of 2-ethoxycarbonyl group failed due to the
thermal instability of 6a (80°C, decomp.) and the route
was slightly changed. Addition of a disodium salt of
thioglycolic acid to 4 in DMF gave acid 5b in 88% yield.
The intramolecular condensation of 5b directly afforded
the 2H-thiophen-3-one 7 in 74% yield by an acidic work-
up of the initially formed product 6b.
9
mercially available thioindigo (Fig. 2b), and the former
and the latter were proven to be the known I (P2 /c) and
1
II (P2 /n) crystal packing types of thioindigo by simula-
tion,
1
10,11
respectively. Even when the commercial thioin-
digo sample was heated at 200°C, the reflection pattern
did not change. Thioindigo structure I which is formed
Oxidative dimerization of 7 to the targeted 1 was next
examined under a variety of conditions as shown in
Table 1. Two phase oxidation with potassium ferricyanide
10a
by sublimation at 280°C with partial decomposition
is P2 /c and the molecules stack along the b axis. There-
fore, the thioindigo molecules in this crystal system look
1
6
gave the best result and the targeted compound 1 was
obtained as a 1:1 mixture of diastereomers in 75% iso-
lated yield. In this oxidation, only an E-isomer of the
newly formed double bond was obtained.
Thermal behavior of the thioindigo precursor 1 was
examined. The UV spectra were measured during 5min
heating of an orange solution of 1 in diphenyl ether.
From this experiment, conversion of 1 to thioindigo
started at around 160°C and completed at 200°C. The
final spectra were completely identical with those of com-
mercially available thioindigo. Heating the solid
precursor 1 at 200°C for 10min gave analytically pure
thioindigo in quantitative yield. From the TG–DTA
experiment, almost the same decomposition temperature
as in solution was observed; the precursor lost 21.6% of
its weight corresponding to two molecules of ethylene be-
tween 140and 195 °C (Fig. 1). Careful inspection of the
TG curve (more obvious in DTG curve) revealed that
the earlier half of weight loss was sharp (144–160°C)
compared to the latter half (165–195°C).
Table 1. Oxidative dimerization of 7 to 1
Conditions
Yield/%
LDA, I
LDA, I
2
(0.5equiv), THF, rt
(1.0equiv), THF, rt
23
19
11
11
39
5
2
t-BuOK, I
N-Methylmorpholine, I
CAN (2.5equiv), aq MeCN, rt
2
(0.6equiv), THF, rt
Figure 2. XRD spectra of thioindigo samples. (a) Thioindigo obtained
by the pyrolysis (200°C, 5min) of 1 (upper) and the simulation pattern
2
(2.0equiv), CH Cl , rt
2 2
of the P2
upper), commercial thioindigo after 5min heating at 200°C (middle),
and the simulation pattern of the P2 /n crystal of thioindigo (lower).
1
/c crystal of thioindigo (lower); (b) commercial thioindigo
7
FeCl
K
3
–SiO
2
, CH
], H
2
Cl
O–EtOAc, rt
2
, rt
(
8
3
[Fe(CN)
6
2
75
1

H. Uno et al. / Tetrahedron Letters 45 (2004) 9083–9086
9085
to lie in parallel lines in the c–a plane (Fig. 3a). On the
other hand, thioindigo structure II which is formed by
1
0c
sublimation at 190°C is P2 /n and the thioindigo mole-
1
cules stacks along the a axis. Therefore, the thioindigo
molecules in this crystal system look to lie in zigzag lines
in the b–c plane (Fig. 3b).
Fortunately, single crystals for the X-ray analysis were
obtained by slow diffusion of hexane in a CH Cl solu-
2
2
1
2
tion of 1. The XRD pattern of the powder sample of 1
was proven to be identical with that obtained by simula-
tion. The chromophores stack along the a axis, the mean
plane distance is 4.62A, and the molecules look to lie in
parallel lines in the b–c plane (Fig. 4). This may explain
˚
the formation of thioindigo crystal type I (P2 /c) when 1
1
is heated to decompose to thioindigo.
Figure 5. DSC–XRD experiment of 1 to thioindigo. The precursor 1
was heated at a rate of 5°C/min under a nitrogen atmosphere and the
XRDs were measured (CuKa, 2h 5.00–44.99°, 30°/min) during the
interval of ca 8.1°C. (a) 123.0–131.2; (b) 131.2–139.4; (c) 139.4–147.6;
In order to examine the transformation mechanism in
detail, we measured the DSC–XRD and the results are
shown in Figure 5. Each XRD was recorded during
the temperature range of ca 8.1°C. The DSC curve
shows the existence of two decomposition periods corre-
sponding to the exothermic peaks at ca 160and 177 °C
during the whole decomposition. The XRD pattern of
the precursor 1 remained in the first (lower) decomposi-
(
d) 147.6–155.7; (e) 155.7–163.8; (f) 163.8–171.9; (g) 171.9–180.0; (h)
1
80.0–188.1; (i) 188.1–196.2; (j) 196.2–204.4; (k) 204.4–212.5°C.
of thioindigo appeared in the second (higher) decompo-
sition period (Fig. 5f–k). In the XRD of the intermediate
point (Fig. 5f, 163.8–171.9°C), a new diffraction pattern
tion period (Fig. 5a–e), while that of structure I(P2 /c)
1
(asterisks) was clearly observed as well as that of struc-
ture I. This intermediate XRD pattern was observed
during almost the whole range of decomposition
(
Fig. 5c–h; ca 140–188°C). Taking the result of the
TG–DTA experiment (Fig. 1) into an account, the inter-
mediate XRD pattern strongly suggested the existence
of an intermediate compound such as mono-ethylene-
bridged thioindigo, although further experiments were
required for determination. No XRD pattern of struc-
ture II or amorphous phase was observed during the
decomposition.
In conclusion, we succeeded in preparation of the thio-
indigo equivalent which quantitatively gives thioindigo
by heating. The crystal packing type of formed thioin-
digo is only obtained by high temperature sublimation
with partial decomposition and is well expected by the
crystal packing of the equivalent.
Figure 3. Crystal packing views of thioindigo. (a) Structure I (P2
1
/c)
along b axis. (b) Structure II (P2 /n) along a axis.
1
Acknowledgements
We appreciate the partial financial support of Grant-In-
Aid for Scientific Research (No. 15655013) from the
Ministry of Education, Culture, Sports, Science, and
Technology. T.I. thanks Canon Inc. for permission of
researching at Ehime University. We thank Mr. Akira
Kishi and Ms. Yayoi Taniguchi (Application Labora-
tory, Rigaku Co.) for their DSC–XRD measurements
[
RINT-Ultima II (XRD-DSC II)].
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.tetlet.
2004.10.028.
Figure 4. Packing diagram of 1. The mean plane distances between the
˚
chromophores of A, B, and C are 4.62 (A–B) and 2.63A (A–C).

9086
H. Uno et al. / Tetrahedron Letters 45 (2004) 9083–9086
10. (a) von Eller, H. Bull. Chim. Soc. Fr. 1955, 106, 1426; (b)
References and notes
von Eller, H. Bull. Chim. Soc. Fr. 1955, 106, 1438; (c)
Haase-Wessel, W.; Ohmasa, M.; S u¨ sse, P. Naturwissens-
1
2
. (a) Fukushima, K.; Nakatsu, K.; Takahashi, R.; Yama-
moto, H.; Gohda, H.; Homma, S. J. Phys. Chem. B 1998,
chaften 1977, 64, 435, Structure I: monoclinic, P2
1
/c,
Z = 2, a = 7.91(3), b = 3.97(2), c = 20.41(3)A, b = 93(1)°:
Structure II: monoclinic, P2 /n, Z = 2, a = 3.981(3),
˚
102, 5986; (b) Mizuguchi, J. J. Cryst. Soc. Jpn. 2002, 44, 50 .
. (a) Ito, S.; Murashima, T.; Uno, H.; Ono, N. Chem.
Commun. 1998, 1661; (b) Uno, H.; Ishikawa, T.; Hoshi,
T.; Ono, N. Tetrahedron Lett. 2003, 44, 5163.
1
˚
b = 20.65(2), c = 7.930(7)A, b = 98.84(5)°.
1. Powder simulation and single cystal analysis were per-
formed by CrystalStructure Ver. 3.6.0(Rigaku, 3-9-12
Akishima, Tokyo, Japan).
1
1
3. Shen, M.; Schultz, A. G. Tetrahedron Lett. 1981, 35, 3347.
4. Murray, R. W.; Singh, M. Org. Synth. Coll. 1998, 9, 228.
5. Campaigne, E. Thiophenes and their Benzo Derivatives:
16 2 2
2. Crystallographic data for 1: C20H O S , orange crystal,
triclinic, P–1, Z = 1 in a cell of dimensions a = 5.891(1),
˚
(iii) Synthesis and Applications; In Comprehensive Hetero-
cyclic Chemistry; Bird, C. W., Cheeseman, G. W. H.,
Katritzky, A. R., Rees, C. W., Eds.; Pergamon: Oxford,
b = 6.474(2), c = 10.710(2)A, a = 86.51(2)°, b = 79.61(1)°,
˚
3
ꢀ3
= 1.466gcm .
835 unique reflections (1996 measured at rt), 1327
c = 84.09(2)°, V = 399.26(16) A ,
D
calc
1
1
984; Vol. 4, Chapter 3.15.
. Spectroscopic data for 1: d
2H, m), 4.15 (2H, m), 4.01 (2H, m), and 1.63–1.44 (8H,
m); d (CDCl ) 184.5, 179.0, 179.0, 136.2, 136.0, 131.7,
0.4, 33.9, 25.8, and 25.4; mmax 2965, 2935, 1632, and
2
2
6
H 3
(CDCl ) 6.47 (2H, m), 6.36
1
with F > 2r(F ). The final R = 0.0586, wR (all) = 0.1629,
goodness-of-fit = 1.017 for 110 parameters. The molecule
occupies a special position. The crystal is a mixture of two
stereoisomers in 1:1 ratio, and the double and single bonds
in the bicyclo[2.2.2]-octadiene skeleton could not be
distinguished. In the refined structure, hydrogen atoms
are formally placed as if the double bonds occupy the trans
position. The crystallographic data are deposited at
Cambridge Crystallographic Data Centre (CCDC
(
C
3
4
1
ꢀ
1
602cm : kmax (CH Cl , loge) 483 (2.77)nm.
2
2
7
. Jempty, T. C.; Gogins, K. A. A.; Mazur, Y.; Miller, L. L.
J. Org. Chem. 1981, 46, 4545.
. Feringa, B.; Wynberg, H. J. Org. Chem. 1981, 46, 2547.
. Purchased from Tokyo Kasei Co. Ltd http://www.
tokyokasei.co.jp.
8
9
2
33280).