Improved synthesis of the π-electron donor bis(ethylenethio)tetrathiafulvalene (BET-TTF)

Article abstract of DOI:10.1055/s-1999-3451

Following a three step route, 5,6-dihydrothieno[2,3-d]1,3-dithiol-2-one (5) was synthesized in gram quantifies starting from commercially available reagents. Coupling of 5 with trimethyl phosphite gave the π-donor bis(ethylenethio)tetrathiafulvalene 1 in 80% yield.

Full text of DOI:10.1055/s-1999-3451

SHORT PAPER
577
Improved Synthesis of the p-Electron Donor
Bis(ethylenethio)tetrathiafulvalene (BET-TTF)
Aarón Pérez-Benítez, Judit Tarrés, Elisabet Ribera, Jaume Veciana, Concepció Rovira*
Institut de Ciència de Materials de Barcelona (CSIC), Campus Universitari, E-08193-Bellaterra, Spain
Fax +34(93)5805729; E-mail: c.rovira@icmab.es
Received 7 August 1998; revised 18 November 1998
TTF donor is obtained in good yields and gram quantities.
The first step was the diazotization-bromination of DL-
homocysteine thiolactone hydrochloride (2) to produce
the a-bromo-g-thiobutyrolactone (3), as reported by Mill-
er and Heindel.16 The nucleophilic displacement of the
bromine atom in 3 by the O-isopropylxanthic group was
achieved in 45 minutes and gave the crude product 4 pure
enough to be used in further reactions.17
Abstract: Following a three step route, 5,6-dihydrothieno[2,3-d]-
1,3-dithiol-2-one (5) was synthesized in gram quantities starting
from commercially available reagents. Coupling of 5 with trimethyl
phosphite gave the p-donor bis(ethylenethio)tetrathiafulvalene 1 in
80% yield.
Key words: tetrathiafulvalene analog, 1,3-dithiol-2-one, trimethyl
phosphite, coupling reactions
Since the discovery of the metallic charge transfer com-
plex TTF-TCNQ,1,2 a large number of TTF analogs have
been designed as electron-donating components to pro-
duce metallic or even superconducting cation radical
salts.3,4 The physical properties of these salts strongly de-
pend on the electronic and structural features of the TTF
derivative used, i.e. on its substituent pattern; those with
sulfur atoms in cyclic substituents being the p-donors
which gave rise to a larger number of superconductors.3,4
We have been involved in the synthesis of new p-donors
of the TTF family having a five-membered ring contain-
ing a sulfur atom as substituent of the TTF core.5 One of
them, BET-TTF (1), is an excellent donor which forms
highly conducting cation radical salts with both
diamagnetic6–9 and paramagnetic counterions.10 However,
the previously reported syntheses of the BET-TTF
donor5, 11 produce only small amounts (~130 milligrams)
of pure donor in each synthesis due to the poor yields of
the steps involved and to the difficult purification of the
intermediate compounds. Moreover, it takes more than
one month to perform the complete synthesis. These facts
have prompted us to improve the synthesis of the BET-
TTF donor by changing the synthetic pathway towards
one which gives the desired donor in gram quantities, thus
allowing the preparation of different charge-transfer salts.
NH3+Cl-
O
S
2
NaNO2, HBr,
27 %
Br
0 °C, 1.5 h
K
+ -SCSOiPr,
K+ -SCSOEt,
acetone, rt,
45 min.
O
S
acetone, rt,
45 min.
3
100 %
100 %
SCSOiPr
SCSOEt
O
O
S
S
H2SO4,
0 °C, 1h.
H2SO4,
0 °C, 1h.
4
6
60 %
S
O
S
S
5
P(OCH3)3,
reflux, 7 h.
81 %
S
S
S
S
S
S
1
Scheme 1
Since it is well known that to obtain TTF derivatives hav-
ing sulfur atoms directly attached to the TTF core, the
coupling of ketones gives better results than the coupling
of thiones,12–15 we have now developed a synthesis based
on the coupling of the 1,3-dithiol-2-one 5. The previous
thiadiazole route5 gave the thione precursors in very poor
yields and the coupling of the thione gave the donor in
only 23% yield.
The cyclization of the isopropylxanthic derivative 4 pro-
ceeds via its dehydration to give the corresponding 1,3-
dithiol-2-one 5. In contrast, the ethylxanthic derivative 6
(Scheme 1) does not give the ketone 5 but a dark red oil
composed by a complex mixture of compounds that was
not characterized.
Finally, the coupling of the ketone 5 to give 1 was accom-
plished in 7 hours by refluxing it with P(OMe)3 in 81%
yield. This yield is, as expected, much higher than that ob-
tained by coupling of the corresponding thione (23%).5
By recrystallization from chlorobenzene, high quality sin-
gle crystals of the donor 1 were obtained and by X-ray dif-
First we made an adaptation of the thiolactone route de-
veloped by Larsen and Lenoir for the synthesis of bis(eth-
ylenedithio)tetrathiafulvalene
(BEDT-TTF).13
This
synthesis consists of four steps (see Scheme 1) involving
commercially available starting materials and the BET-
Synthesis 1999, No. 4, 577–579 ISSN 0039-7881 © Thieme Stuttgart · New York

578
A. Pérez-Benítez et al.
SHORT PAPER
Melting points were measured in an electrothermal apparatus and
were uncorrected.
fraction it was concluded that 1 exists in the crystals
principally as the E-isomer as was also observed when the
donor was obtained by coupling of the thione.5
Tetrahydro-3-(i-propoxythiocarbonylthio)-2-oxothiophene (4)
a-Bromo-g-thiobutyrolactone (3; 7.17 g, 0.04 mol) was added with
stirring to a suspension of potassium O-(i-propyl)dithiocarbonate
(7.28 g, 0.04 mol) in anhyd acetone (250 mL) at r.t. in the dark. Af-
ter stirring for 45 min, the precipitate (KBr) formed was filtered off
and the solvent was removed under reduced pressure to give a green
oil.20 The oil was washed with H2O and extracted with Et2O. The or-
ganic phase was dried (MgSO4) and the solvent was evaporated to
yield 9.3 g (~100%) of 4 as a stable orange oil. In order to obtain an
analytically pure sample, the oil was purified by flash chromatogra-
phy on silica gel, eluting with EtOAc/hexanes (1:3) to give 4 in an
almost quantitative yield.
Attempts to prepare the oxygen-containing analogue of 1
(bisethyleneoxo-tetrathiafulvalene) following the same
procedure used in the synthesis of 1 failed because the
precursor ketone 10 is not obtained, but instead, the disul-
fide 9 is formed (Scheme 2). Even though the acid-cata-
lyzed ring closure of 8 was attempted by changing sulfuric
18
acid for HClO4 or CF3CO2H/AcOH/p-toluenesulfonic
acid mixture,19 the ketone 10 was never detected. The ring
closure of 8 was also attempted with P2O5 in refluxing
decalin but surprisingly the former oxygen atom of the
lactone was replaced by sulfur giving 5 instead of the an-
alogue 10. A similar exchange was described by Engler et
al. for the cyclization of the ethyl analogue of 8 with
P4S10.11 This procedure for the synthesis of 5 directly from
8 is very interesting since it starts from a commercial
product, α-bromo-γ-butyrolactone (7), and avoids the te-
dious synthesis of the α-bromo-γ-thiobutyrolactone (3).
Moreover, the total yield for the synthesis of the ketone 5
following the route 2 → 3 Æ 4 Æ 5 (Method A, 15%) is
similar to the yield obtained following the route 7 Æ 8 Æ
5 (Method B, 10%).
IR (film): ν = 1706 cm–1.
1H NMR (CDCl3): d = 1.40 [dd, 6 H, J = 6.3 Hz, CH(CH3)2], 2.45
(m, 1 H, SCHCH2), 2.85 (m, 1 H, SCHCH2), 3.51 (m, 2 H, SCH2),
4.55 (m, 1 H, SCH), 5.74 [sept, 1 H, J = 6.3 Hz, CH(CH3)2].
13C NMR (CDCl3): d = 21.1, 30.0, 31.9, 57.6, 79.2, 202.4, 210.3
Anal. calcd for C8H12O2S3 (236.4): C, 40.65; H, 5.12; S, 40.69.
Found C, 40.56; H, 5.11; S, 40.60.
5,6-Dihydrothieno[2, 3-d]-1, 3-dithiol-2-one (5)
Method A: Compound 4 (9.3 g, 0.039 mol) was added dropwise to
a stirred ice-cooled solution of concd H2SO4 (300 mL). After the ad-
dition was complete, the stirring was maintained for 1 h at r.t. and
then the mixture was poured slowly onto ice (1 L) and extracted
with toluene (4 î 300 mL). The combined organic layers were
washed with H2O (3 î 100 mL), dried (Na2SO4) and evaporated un-
der reduced pressure. The crude product was purified by flash chro-
matography on silica gel eluting with a mixture of CH2Cl2/hexanes
(1:10). The ketone 5 was eluted pure in the second fraction or with
the little less polar impurity which could be eliminated easily by re-
crystallization from hexanes to yield 4.15 g (60%) of 5 as white nee-
dles; mp 85–86 °C (Lit.11 mp 84–85 °C).
In conclusion, we have presented an improved synthesis
of the p-electron donor bis(ethylenethio)tetrathiaful-
valene (BET-TTF) (1) based on the coupling of the 1,3-
dithiol-2-one 5. With the current methodology it is possi-
ble to produce gram amounts of the donor in 2–3 days
(Method A) starting from commercially available prod-
ucts, since not only the overall yield has been improved
but also the intermediate compounds are readily purified;
one of them, compound 5, should be valuable in the syn-
thesis of new disymmetric TTF derivatives.
1H NMR (CDCl3): d = 3.10 (t, 2 H, J = 8.1 Hz, SCH2CH2), 3.55 (t,
2 H, J = 8.1 Hz, SCH2).
13C NMR (CDCl3): d = 33.0, 34.6, 118.0, 122.9, 195.1
DL-Homocysteine thiolactone hydrochloride (2), a-bromo-g-buty-
rolactone (7), potassium O-ethyldithiocarbonate and potassium O-
(isopropyl)dithiocarbonate were purchased from Aldrich and used
as received. a-Bromo-g-thiobutyrolactone (3) was obtained follow-
ing the procedure reported by Miller and Heindel.16
Anal. calcd for C5H4S3O (176.3): C, 34.09; H, 2.27; S, 54.54. Found
C, 34.32; H, 2.20; S, 54.35.
Method B: A solution of the compound 8 (6.85 g, 0.031 mol) in
decalin (70 mL) was refluxed with P2O5 (4.37 g, 0.031 mol) for 4 h.
The mixture was then stirred for 2 d at r.t. and purified by flash chro-
matography on silica gel eluting with a mixture of CH2Cl2/hexanes
(1:10). Recrystallization from hexanes gave 0.6 g of white needles
(11%) of 5; mp 85–86 °C.
1
FT-IR spectra were recorded on a Nicolet 710 spectrometer, H
NMR and 13C NMR on a Bruker ARX 300, and the mass spectra on
a Hewlett-Packard 5989X. Elemental analyses were performed us-
ing the Elemental Analyser EA 1108 (Carlo Erba Instruments).
K
+ -SCSOiPr,
S
acetone, rt,
45 min.
P2O5, decaline,
reflux, 5 h.
P(OCH3)3,
reflux, 7 h.
SCOiPr
Br
O
S
S
S
S
S
S
S
O
93 %
11 %
81 %
S
S
O
O
O
H2SO4,
0 °C, 1 h.
H2SO4,
0 °C, 1 h.
7
8
5
1
13 %
S
O
S
S
S
O
O
O
O
O
9
10
Scheme 2
Synthesis 1999, No. 4, 577–579 ISSN 0039-7881 © Thieme Stuttgart · New York

SHORT PAPER
Improved Synthesis of the p-Electron Donor Bis(ethylenethio)tetrathiafulvalene
579
5,5’, 6,6’-Tetrahydro-D2,2’-bithieno[2,3-d]-1,3-dithiol (1)
Ketone 5 (2.05 g, 0.0116 mol) was dissolved in freshly distilled tri-
methyl phosphite (26 mL) and refluxed for 7 h under an argon at-
mosphere, during which time red crystals of 1 precipitated. After
cooling to r.t., the product was filtered off, washed with Et2O and
dried; yield: 1.5 g (81%). Recrystallization from chlorobenzene af-
forded 1.40 g (75 %) of 1 as scarlet crystals. All spectroscopic data
were in accordance with those previously reported;5,11 mp 194–
197 °C (dec) (Lit.11 mp 195–196°C).
Acknowledgement
This work was supported by the Ministerio de Educación y Cultura
(DGCYT), Grant PB 96-0872-C02-01) and the Generalitat de Cata-
lunya (CIRIT) SGR 96-00106. A. P-B. thanks the CONACyT for a
predoctoral fellowship and the Universidad Autónoma de Puebla
for on live permission.
References
2-Oxo-3-(i-propoxythiocarbonylthio)tetrahydrofuran (8)
a-Bromo-g-butyrolactone (7; 15.84 g, 0.096 mol) was added with
stirring to a suspension of potassium O-(i-propyl)dithiocarbonate
(16.74 g, 0.096 mol) in anhyd acetone (170 mL) at r.t.in the dark.
After stirring for 45 min, the precipitate (KBr) formed was filtered
off and the solvent was eliminated under reduced pressure to give a
pink oil.20 The crude product was washed with H2O and extracted
with Et2O. The organic phase was dried (MgSO4) and the solvent
was evaporated to yield 19.7 g (93%) of 8 as a stable yellow oil.
(1) Coleman, L. B.; Cohen, M. J.; Sandman, D. J.; Yamagishi, F.
G.; Garito, A. F.; Heeger, A. J. Solid State Commun. 1973, 12,
1125.
(2) Ferraris, J. P.; Cowan, D. O.; Walatka, V.; Perlstein, J. H.
J. Am. Chem. Soc. 1973, 95, 948.
(3) Williams, J. M; Ferraro, J. R.; Thorn, R. J.; Carlson, K. D.;
Geiser, U.; Wang, H. H.; Kini, A. M.; Whangbo, M.-H.
Organic Superconductors; Prentice-Hall: Englewood Clifs,
New Jersey, 1992.
(4) Ishiguro, T.; Yamaji, K. Organic Superconductors; Springer-
Verlag: Heidelberg, 1990.
(5) Rovira, C.; Veciana, J.; Santaló, N.; Tarrés, J.; Cirujeda, J.;
Molins, E.; Llorca, J.; Espinosa, E. J. Org. Chem. 1994, 59,
3307.
(6) Rovira, C.; Santalo, N.; Veciana, J. Tetrahedron Lett. 1989,
30, 7249.
(7) Tarrés, J.; Santaló, N.; Más, M.; Molins, E.; Veciana, J.;
Rovira, C.; Yang, S.; Lee, H.; Cowan, D. O.; Doublet, M.-L.;
Canadell, E. Chem. Mater. 1995, 7, 1558.
IR (film): ν = 1776 cm–1.
1H NMR (CDCl3): d = 1.21 [dd, 6 H, J = 6.2 Hz, CH(CH3)2], 2.28
(m, 1 H, SCHCH2), 2.70 (m, 1 H, SCHCH2), 4.21 (m, 3 H, SCH,
OCH2), 5.55 [sept, 1 H, J = 6.2 Hz, CH(CH3)2].
13C NMR (CDCl3): d = 21.1, 29.7, 45.6, 66.6, 79.4, 173.4, 209.5
Anal. calcd for C8H12O3S2. (220.3): C, 43.63; H, 5.50; S, 29.06.
Found C, 43.21; H, 5.44; S, 28.82.
3, 3’-Dithiobisdihydro-2-furanone (9)
(8) Rovira, C.; Santaló, N.; Veciana, J.; Molins, E.; Miravitlles, C.
Synth. Met. 1991, 41, 2199.
Compound 8 (8 g, 0.0363 mol) was added dropwise to a magneti-
cally stirred ice-cooled solution of concd H2SO4 (300 mL). After the
addition was complete, the stirring was maintained for 1 h at r.t. and
then the solution was poured onto ice (1 L) and extracted with tolu-
ene (4 î 300 mL). The solvent was removed at reduced pressure to
give a yellow oil. Et2O and acetone were added to the oil until a ho-
mogeneous solution was formed. The solution was chilled at 5 °C
for 2 d to give a white solid that was purified by flash chromatogra-
phy on silica gel eluting with CH2Cl2/acetone (1:1). White needles
of 9 were obtained (560 mg, 13%); mp 115–116 °C (Lit.21 mp 111–
113 °C).
(9) Tarrés, J.; Veciana, J.; Rovira, C. Synth. Met. 1995, 70, 1167.
(10) Coronado, E.; Falvello, L. R.; Galán-Mascarós, J. R.;
Giménez-Saiz, C.; Gómez-García, C. J.; Lauhkin, V. N.;
Pérez-Benítez, A.; Rovira, C.; Veciana, J. Adv. Mater. 1997,
9, 984.
(11) Engler, E. M.; Patel, V. V.; Andersen, J. R.; Schumaker, R. R.;
Fukishima, A. A. J. Am. Chem. Soc. 1978, 100, 3769.
(12) Yamashita, Y.; Tomura, M.; Tanaka, S. J. Chem. Soc., Perkin
Trans. 1 1990, 3358.
(13) Larsen, J.; Lenoir, C. Synthesis 1989, 134.
(14) Krief, A. Tetrahedron 1986, 42, 1209.
IR (KBr) ν = 1747, 1774, 3472, 3520 cm–1.
(15) Schukat, G.; Fanghänel, E. Sulphur Rep. 1990, 14, 245.
(16) Miller, G. A.; Heindel, N. D. J. Org. Chem. 1981, 46, 4751.
(17) Operating in the dark precludes the formation of undesired
radical products that destabilise the crude product 4.
(18) Bhattacharya, A. K.; Hortmann, G. J. Org. Chem. 1974, 39,
95.
(19) Haley, N. F.; Fichtner, W. J. Org. Chem. 1980, 45, 175.
(20) The crude product can be used immediately in the following
step without a noticeable decrease of the yield.
(21) Reppe, W. Liebigs Ann. Chem. 1955, 596, 187.
1H NMR (CDCl3): δ = 2.53 (m, 1 H, SCH2CH2), 2.79 (m, 1 H,
SCH2CH2), 3.92 (m, 1 H, SCH2), 4.35 (m, 1 H, SCH2), 4.50 (m, 1
H, SCH).
13C NMR (CDCl3): δ = 29.3, 46.2, 66.7, 174.6.
Anal. calcd for C8H10O4S2 (170.3): C, 41.01; H, 4.30; S, 27.37.
Found C, 40.96; H, 4.23; S, 27.39.
MS: m/z (%) = 86 (26), 118 (100), 149 (4), 234 (M+, 9).
Synthesis 1999, No. 4, 577–579 ISSN 0039-7881 © Thieme Stuttgart · New York