New, optimized preparation of 1,2-dichlorocyclobuten-3,4-dione (C4O2Cl2) from squaric acid and oxalyl chloride

Article abstract of DOI:10.1016/j.tetlet.2007.03.091

1,2-Dichlorocyclobuten-3,4-dione, a key intermediate in the chemistry of cyclobutenes and their derivatives is obtained by the dimethylformamide catalyzed reaction of squaric acid and oxalyl chloride. The process was optimized by using guidelines from thermodynamics, transport properties, and experimental observations, and gives nearly quantitatively a product which can be easily brought to very high purity.

Full text of DOI:10.1016/j.tetlet.2007.03.091

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Tetrahedron Letters 48 (2007) 3595–3597
New, optimized preparation of 1,2-dichlorocyclobuten-
3,4-dione (C4O2Cl2) from squaric acid and oxalyl chloride
Bruno Lunelli*
Istituto per lo Studio dei Materiali Nanostrutturati (ISMN) CNR, Sezione di Bologna, 101 via P.Gobetti, I-40129 Bologna, Italy
Received 2 October 2006; revised 14 March 2007; accepted 16 March 2007
Available online 21 March 2007
Abstract—1,2-Dichlorocyclobuten-3,4-dione, a key intermediate in the chemistry of cyclobutenes and their derivatives is obtained
by the dimethylformamide catalyzed reaction of squaric acid and oxalyl chloride. The process was optimized by using guidelines
from thermodynamics, transport properties, and experimental observations, and gives nearly quantitatively a product which can
be easily brought to very high purity.
Ó 2007 Elsevier Ltd. All rights reserved.
1,2-Dichlorocyclobuten-3,4-dione (squaric acid dichlo-
ride, 1) is one of the key substances in the rich chemistry
of the cyclobutenes, and the products which start from
them.1 A number of its molecular properties in the solid,
solution, and gas phase have been measured and ana-
lyzed.2 CAUTION: dangerous by inhalation or penetra-
tion through the skin.3
Indeed, such reaction channel could be opened by the
basic catalyst dimethylformamide (henceforth DMF)7
and thionyl chloride SOCl2, allowing an easy, higher
yield preparation of 1,8 but the product still contains a
substantial fraction of 2: To get pure 1 using this route
it is necessary to follow the cumbersome purification
indicated above. Overall, the procedure is very time con-
suming. Furthermore, a recent publication9 reports the
formation of tarry material, which we did not find in
the present procedure.
It was first prepared by heating 1,2,3,3-tetrachloro-
cyclobuten-4-one (lachrymator, liquid at room tempera-
ture, mp 20 °C, nbp 176 °C, 2) with oleum. The
procedure had a rather low yield (40%) and the product
contained unreacted 2, which could be largely elimi-
nated by extraction and crystallization from petroleum
ether.4 We found that to get 1 with a purity of at least
99.9% from the product of this reaction the best proce-
dure consisted of several crystallizations from cyclohex-
ane, and then sublimation at reduced pressure, because 2
is less volatile than 1.2 This awkward process was used
because the most logical precursor to 1, the commer-
cially available 1,2-dihydroxycyclobuten-3,4-dione or
squaric acid 3, does not react with the common reagents
used for the preparation of acid chlorides in the absence
of a catalyst.5 A detailed study on catalysts for the prep-
aration of acid chlorides using SOCl2 was overlooked.6
A long known10 alternative to SOCl2 for the preparation
of acid chlorides is oxalyl chloride (mp ꢀ16 °C, nbp
63.5 °C, d20 1.478, vapor pressure at 20 °C ꢁ 150 Torr;11
CAUTION: dangerous similarly to 1, see above, easily
and safely handled using a new container);12 a cursory
indication of reaction of 3 with oxalyl chloride is given
in a patent.13
Having the need for gram quantities of pure 1 for an
investigation of possible new oxocarbons, we decided
to try the preparation of squaric acid chloride from
squaric acid and oxalyl chloride, as an alternative to
the optimization of the procedure using SOCl2, which
comprises at least two stoichiometrically independent
reactions. Below some underpinnings and the results
of the new process are given.
Squaric acid 3 is a monoclinic solid,14 mp >300 °C,
vapor pressure at room temperature below 10ꢀ6 mbar,15
soluble only in more or less basic, hydrogen-bond form-
ing solvents. If literature results concerning the reaction
Keywords: 1,2-Dichlorocyclobuten-3,4-dione; Oxalyl chloride; Cyclo-
butenes; Catalysis by dimethylformamide; Acyl halides; Reactive
intermediates.
*
Tel.: +39 051 639 9593; e-mail: b.lunelli@bo.ismn.cnr.it
0040-4039/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2007.03.091

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
3596
B. Lunelli / Tetrahedron Letters 48 (2007) 3595–3597
of oxalyl chloride with phenols and/or monobasic car-
boxylic acids were applicable to the DMF catalyzed
reaction of the peculiar, bibasic squaric acid, several
products could be formed, among which are (1) squaric
acid chloride 1; (2) squaric acid monochloride; (3) squa-
ric acid anhydrides; (4) mixed anhydrides of squaric and
oxalic acid; (5) 1,2,3,3-tetrachlorocyclobuten-4-one 2;
(6) phosgene by decomposition of oxalyl chloride.16
But at temperatures slightly higher than room tempera-
ture, atmospheric pressure and in the presence of cata-
lytic quantities of DMF preliminary experiments gave
good yields of apparently only 1. To try optimization
of the procedure we then analyzed the gaseous, liquid
and solid phases of several samples during and at the
end of the reaction carried out under different conditions
by means of IR spectrometry.
be given by the ratio of the mole fraction of 1 and the
squared mole fraction of oxalyl chloride, Kx ꢁ x(1)/x2
(oxalyl chloride).
The lack of oxalyl chloride in the final liquid phase fur-
ther reveals that (1) is a substantially quantitative reac-
tion. The reported yield of the preparation is lower than
quantitative with respect to oxalyl chloride because part
of it is carried away by the gases produced, especially
initially, when its mole fraction in the liquid phase is
high.
Temperature of the reaction: The reaction is slow at
room temperature; a temperature of 50 °C allows a suf-
ficient speed, is lower than the boiling point of oxalyl
chloride and higher than the freezing point of 1 in the
reacting system. This temperature reduces the quantity
of oxalyl chloride in the unreactive gas phase and avoids
hampering the contact between 3 and oxalyl chloride by
possible separation of solid 1.
In the gas phase we detected only HCl, CO, CO2, the
solvent (when used), traces of 1 and oxalyl chloride;
the latter disappeared in the last gases evolved. No phos-
gene was ever detected.
Solvent: CCl4, CH2Cl2, and the reagent oxalyl chloride
In the liquid phase after evolution of gas had ended only
1 and the solvent were found; no 2, phosgene nor other
species.
were used.
CCl4 (CAUTION: acutely and chronically toxic and
possibly a carcinogen)20 permits easy separation of the
compounds originating from DMF and subsequently
of 1 by evaporation. Room temperature evaporation
of the solution under vacuum causes some transfer of
1 to the distillate, which can be minimized by first sepa-
rating solid 1 from the solution by cooling21 and then
evaporating under vacuum the cool solution.
The solid phase at the end of the procedure consisted of
3 wetted by a weight slightly higher than that of the
DMF of a yellow oil (soluble in CH2Cl2, water and ace-
tonitrile) where squaric acid monochloride was not de-
tected.17 These results indicated a very new convenient
method to prepare pure squaric acid chloride, because
under the above conditions the chemical reaction of 3
and oxalyl chloride is described to an approximation
over 99% by the single stoichiometric equation
CH2Cl2 keeps in solution the substances originating
from DMF, hindering their separation and gives very
ð1Þ
which shows that oxalyl chloride might be more effi-
cient18 than SOCl2, because each mole forms three
rather than two moles of gaseous products and hence
brings about a higher entropy increase. The necessity
of heating to maintain the chosen, fixed reaction temper-
ature is compatible with an endothermic or slightly exo-
thermic process that did not show tendency to get out of
control.
concentrated solutions of 1 from which the latter crys-
tallizes with difficulty.
Oxalyl chloride is now solvent and reagent, whose
concentration in this case is the highest possible, with
potential advantages for the reaction speed and yield,
which however were not observed. Using this variant,
there is a larger removal of oxalyl chloride by the
evolved gases and sublimations become necessary to
remove the products presumably formed from
DMF.
Thermodynamically, reaction (1) can be considered an
heterogeneous reaction,19 where the gas and solid phases
have substantially and, respectively, rigorously fixed
composition from the beginning to the end of the reac-
tion. Then an approximate equilibrium constant can
Analysis of the products by IR spectrometry: The spectra
were recorded by means of a Bruker IFS 66v/S Fourier

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
B. Lunelli / Tetrahedron Letters 48 (2007) 3595–3597
3597
transform vacuum spectrometer. Resolution 2 cmꢀ1, 16
scans, apodization function three-term Blackman–
Harris. Using a demountable, variable thickness vac-
uum-tigh cell with ZnSe windows for liquids and mulls
of solids,22 and a 200 mm long cell with NaCl windows
for gases. Concentrations and partial pressures were
evaluated by comparison with archived or published
spectra of the pure substances run under the same
conditions.
References and notes
1. Mukkanti, A.; Periasami, M. Arkivoc 2005, 48–77; Online:
http://www.arkat-usa.org/ark/journal/2005/I11_Swami-
nathan/1223/SS-1223R.asp; Seitz, G.; Imming, P. Chem.
Rev. 1992, 92, 1227–1260.
2. (a) Mattes, R.; Schroebler, S. Chem. Ber. 1972, 105, 3761–
3764; (b) Hagen, K.; Hedberg, K.; Lunelli, B.; Giorgini,
M. G. J. Phys. Chem. 1988, 92, 4313–4315; (c) Lunelli, B.;
Giorgini, M. G. Spectrochim. Acta A 1987, 43, 829–835;
(d) Lunelli, B.; Orlandi, G.; Zerbetto, F.; Giorgini, M. G.
J. Mol. Struct. (Theochem) 1989, 201, 307–317; (e)
Caminati, W.; Fantoni, A. C.; Lunelli, B.; Scappini, F.
J. Mol. Spectrosc. 1988, 131, 154–159; (f) Cerioni, G.;
Janoschek, R.; Rappoport, Z.; Tidwell, T. T. J. Org.
Chem. 1996, 61, 6212–6217.
3. Cohen, S. Biological reactions of carbonyl halides. In The
Chemistry of Acyl Halides; Patai, S., Ed.; Interscience:
London, 1972; p 313; Lunelli, B. Chim. Ind. 2005, 87, 99.
4. Maahs, G. Liebigs Ann. Chem. 1965, 686, 55–63.
5. Maahs, G.; Hegenberg, P. Angew. Chem., Int. Ed. Engl.
1966, 5, 888–893.
6. Bosshard, H. H.; Mory, R.; Schmid, M.; Zollinger, Hch.
Helv. Chim. Acta 1959, 42, 1653–1658.
7. Colvin, E. W. In Comprehensive Organic Chemistry;
Barton, D., Ollis, W. D., Eds.; Pergamon: London,
1972; p 634.
8. De Selms, R. C.; Fox, C. J.; Riordan, R. C. Tetrahedron
Lett. 1970, 11, 781–782.
9. Ohno, M.; Yamamoto, Y.; Shirasaki, Y.; Eguki, S.
J. Chem. Soc., Perkin Trans. 1 1993, 263–271.
10. Adams, R.; Ulich, L. H. J. Am. Chem. Soc. 1920, 42, 599–
611, and references cited therein.
Preparation of 1: Finely powdered 3 (1.140 g, 10 mmol),
carbon tetrachloride (5 mL), DMF (40 lL), two
magnetic stirring bars and four Pyrex glass balls of
about 8 mm diameter were introduced into a 50 mL
round-bottomed flask previously filled with nitrogen
and connected to a reflux condenser23 with exit-only
gas valve. The mixture was frozen by immersion in
liquid nitrogen vapors and oxalyl chloride (2.540 g =
1.720 mL, 20 mmol) was introduced at once through
the condenser’s outlet. Then the flask was allowed to
warm to room temperature, agitation started at
300 rpm, and the flask was then lowered into an oil bath
thermostatted to 50 °C. Regular gas evolution started
immediately and was no more apparent after 50 min.
After further 10 min, heating and agitation were
discontinued, allowing cooling and decantation of the
mixture.
The flask content was filtered through a sintered glass
disk and washed two times with 1 mL of CCl4. The fil-
trate was cooled until it partially solidified and the liquid
part evaporated in vacuum to a container cooled by
liquid nitrogen. Then the pale yellow crystals were sub-
limed under vacuum, giving 1.300 g of product (mp
53 °C). The purity of this product was checked by com-
paring its IR spectrum in CCl4 with that of the four-
times sublimed squaric acid chloride used for the gas
phase electron diffraction:2b No differences were
apparent.
11. Aldrich 2005–2006 Catalog, p 1809.
12. Lunelli, B. Italian Patent BO99A 000702 of 22.12.1999.
13. U.S.A. Patent n.6495576.
14. Semmingsen, D.; Tun, Z.; Nelmes, R. J.; McMullan, R.
K.; Koetzle, T. F. Z. Kristallogr. 1995, 210, 934–947.
15. Evalutated from logP(Atm) ꢁ 11.8 ꢀ 8000/T with the
coefficients determined in the 442–504 K range. Private
communication from Dr. Daniela Ferro, ISMN-CNR
Roma, 1993.
16. Beilstein 2, II Suppl., p 509; phosgene is also a normal
component of the reagent-grade commercial product, see,
for example, Ref. 10.
17. By comparison of the IR spectrum in acetonitrile of the oil
with that of squaric acid monochloride (prepared accord-
ing to Schmidt, A. H.; Maibaum, H. Synthesis 1987, 134–
137) in the same solvent.
The residue washed with methanol to remove substances
presumably formed from DMF weighed about 20 mg
and its IR spectrum as a paraffin oil mull showed it to
be pure 3.
18. Sanderson, R. T. J. Chem. Educ. 1964, 41, 13–22;
Matthews, G. W. J. J. Chem. Educ. 1966, 43, 476.
19. Denbigh, K. The Principles of Chemical Equilibrium, 2nd
ed.; Cambridge University Press: Cambridge, 1966; p 159.
20. Daley, J. M.; Landolt, R. G. J. Chem. Educ. 2005, 82, 120–
121.
21. Lunelli, B. Inorg. Chim. Acta 2007, 360, 1217–1220.
22. Lunelli, B.; Comelli, F. Rev. Sci. Instrum. 1987, 58, 305–
307.
Acknowledgments
The author gratefully acknowledges the contribution of
two referees, one indicating redundant parts and the sec-
ond proposing many language improvements. Financial
support came from the ISMN of CNR, Sezione di Bolo-
gna; most of the IR measurements were carried out at
the Laboratory of Infrared Spectrometry ‘Fabio Rover-
si-Monaco’ of the University of Bologna.
23. The use of ice-cold water returns to the flask most of the
oxalyl chloride carried by the evolved gases, decreasing its
unreacted fraction.