Squaric acid difluoride

Article abstract of DOI:10.1016/S0022-1139(99)00138-4

Hitherto unknown squaric acid difluoride (3,4-difluoro-3-cyclobutene-1,2-dione) (1a) has been synthesized in 71% yield by gas phase fluorination of squaric acid dichloride (1c) with KF at 250°C. (1a) has been characterized by multinuclear NMR, mass, IR and Raman spectra. The IR spectrum is in excellent agreement with its ab initio prediction (MP2, 6-311 + G(2d, p)). The vacuum thermolysis of (1a) at 700°C does not yield the expected decarbonylation product FCCF (2a) although the precursor of the latter, F₂CCCO, and three C₃F₄ isomers (propyne, allene, cyclopropene) which are also found as decomposition products of (2a) have been identified.

Full text of DOI:10.1016/S0022-1139(99)00138-4

Journal of Fluorine Chemistry 99 (1999) 99±104
Squaric acid di¯uoride
Michael Senzlobera, Hans BuÈrgera,*, Reint Eujena,
Achim Gelessusb,1, Walter Thielb
È
Anorganische Chemie, FB9, GH-Universitat, 42097 Wuppertal, Germany
b
È È
Organisch-Chemisches Institut, Universitat, 8057 Zurich, Switzerland
a
Received 19 March 1999; accepted 30 March 1999
Abstract
Hitherto unknown squaric acid di¯uoride (3,4-di¯uoro-3-cyclobutene-1,2-dione) (1a) has been synthesized in 71% yield by gas phase
¯uorination of squaric acid dichloride (1c) with KF at 2508C. (1a) has been characterized by multinuclear NMR, mass, IR and Raman
spectra. The IR spectrum is in excellent agreement with its ab initio prediction (MP2, 6-311 + G(2d, p)). The vacuum thermolysis of (1a) at
7008C does not yield the expected decarbonylation product FCCF (2a) although the precursor of the latter, F2CCCO, and three C3F4
isomers (propyne, allene, cyclopropene) which are also found as decomposition products of (2a) have been identi®ed. # 1999 Elsevier
Science S.A. All rights reserved.
Keywords: Squaric acid di¯uoride; Fluorination; Thermolysis; Ab initio calculations
1. Introduction
nor any properties have yet been described. The structure
and the vibrational spectrum of (1a) have been predicted by
MNDO calculations [7], and fruitless attempts to obtain (1a)
by the ¯uorination of (1c) and squaric acid dibromide (1e)
have been reported. Fluorinating agents that were tested
comprise (a) KF or CsF and 18-crown-6 in benzene; (b)
anhydrous HF; (c) KF, NH4F, AgF, CsF and SbF5 in anhy-
drous HF; (d) polymer supported ¯uoride ion in pentane; (e)
KHF2 in CF3COOH; (f) wet Et3NÁ3 HF and (g) diethyla-
mino-sulfurtri¯uoride (DAST). Moreover, the ring opening
reaction of (1a) to give cis- and trans- (3a), X = F, has been
studied with theoretical methods [8] and the ab initio
structure has been optimized using MP2 calculations.
The present contribution reports on an easy and produc-
tive synthesis of (1a), and for the ®rst time, on some of its
chemical and spectroscopic properties. Particular attention
will be given to the thermal decomposition and the question
whether it is useful for the synthesis of (2a).
Dihalogeno derivatives of squaric acid (1), X1 = X2 = F,
Cl, Br, I, are potential precursors for dihalogenoethynes
XC>CX (2) and bis(ketenes) O==C==C(X)±C(X)==C==O
(3) which are dif®cult to prepare otherwise. Thus, dichlor-
oethyne (2c) has been observed by photoelectron spectro-
scopy when squaric acid dichloride (1c) was pyrolyzed at
6008C [1]. On the other hand, photochemically induced ring
opening occurred when (1c) in a noble gas matrix at 10 K
was irradiated (ꢀ > 335 nm) to give (3c), X = Cl as the major
product, the latter decomposing to CO, 2,3-dichlorocyclo-
propen-1-one, 3,3-dichloro-1,2-propadienone, and (2c)
upon further irradiation [2].
Our interest in di¯uoroethyne, FC>CF (2a) [3] and its
photoisomerization product di¯uorovinylidene F2C==C:
[4,5] stimulated us to investigate the thermal decomposition
of squaric acid di¯uoride (1a) as a possible pathway to the
very unstable and highly reactive C2F2 isomers. We soon
realized that (1a), the envisaged precursor of C2F2, had
never been described before. There is a patent [6] in which
(1a) is claimed, but to our knowledge, neither the synthesis
2. Synthesis and thermolysis of (1a)
Since the ¯uorination of (1c) in solution had failed to
yield any (1a) [7], we have attempted gas phase ¯uorination
in order to remove supposedly unstable (1a) as rapidly as
possible from the reactor. For this purpose, freshly sublimed
(1c) [9] was passed at 10 3 mbar over carefully dried KF
*Corresponding author. Tel.: +49-202-439-2498; fax: +49-202-439-2901
E-mail address: buerger1@uni-wuppertal.de (H. BuÈrger)
1Permanent address: Max-Planck-Institut fuÈr Polymerforschung, Ack-
ermannweg 10, 55128 Mainz, Germany.
0022-1139/99/$ ± see front matter # 1999 Elsevier Science S.A. All rights reserved.
PII: S 0 0 2 2 - 1 1 3 9 ( 9 9 ) 0 0 1 3 8 - 4

100
M. Senzlober et al. / Journal of Fluorine Chemistry 99 (1999) 99±104
heated to 2508C. After puri®cation by repeated sublimation
in vacuo, a 71% yield of (1a) was obtained as colourless,
strongly light-refracting crystals of m.p. 408C. When the
¯uorination was not brought to completion, some squaric
The contents of trap 3 could be separated into a fraction
consisting mostly of (5) and (6), while a second fraction was
enriched in (7). All products were identi®ed by their IR
spectra: (5) [16], (6) [17], (7) [18].
acid ¯uoridechloride (1b) could be detected by 19F and 13
C
The C3F4 isomers (5), (6) and (7) have previously been
observed as products of the decomposition of (2a) in the gas
phase at a pressure of several mbar [3]. Moreover, (6) has
been detected as a secondary product formed by reactions of
:CF2 with F2C==C: [19]. It has been argued that F2C==C:
may be readily formed by thermal rearrangement of (2a).
Thus, the formation of (6) by thermolysis of (1a) may well
involve the formation of (2a) as a primary product, into
which, after isomerization to di¯uorovinylidene, :CF2
inserts. Alternatively, rapid attack of :CF2 on di¯uoroviny-
lidene formed by thermolysis of (4) [20] (which was
de®nitely identi®ed as the major component in trap 2) also
may lead to (6).
Compound (7) has been observed as one of the products
in the photochemical decomposition of both di¯uoromaleic
anhydride [21] and CFI==CFI [22], both reactions also
yielding (2a). Indeed, (7) may be the common precursor
of (5) and (6), both products being formed in the CO2 laser
pyrolysis of (7) [23].
NMR spectroscopy (see below). As a more volatile by-
product, 1,4-di¯uoro-2-butyne-1,4-dione [10,11] was
obtained in the 1268C trap and identi®ed by mass, IR
and NMR spectra. Some CF2O was found in the LN2 trap.
(1a) was identi®ed by mass spectra, 13C and 19F NMR
spectra, gas phase infrared spectra, and Raman spectra
obtained of the solid and a solution in CDCl3. The assign-
ment of the vibrational spectra was supported by ab initio
calculations (see below).
In order to collect information and experience on the
thermolysis of dihalogeno derivatives of (1), in general, we
have studied, beforehand, the thermolysis of (1c), squaric
acid chloridebromide (1d), X1 = Cl, X2 = Br, and (1e),
X1 = X2 = Br. The latter two are accessible by the reaction
of (1c) with trimethylsilyl bromide [12], pure (1e) being
obtained when an excess of Me3SiBr was employed, while a
1 : 2 : 1 mixture of (1c), (1d) and (1e) was produced with a
1 : 1 ratio of the reactants.
The dihalogeno derivatives of (1) were pyrolyzed at
7008C in the gas phase at a pressure of 10 3 mbar by
passing them slowly through a quartz tube of 50 cm length
and 2.5 cm inner diameter. The reaction products were
pumped through a series of traps held at 78, 126 and
1968C. The different condensates were analyzed by low
resolution IR spectroscopy. The dihaloalkynes (2c), C2ClBr
(2d) and C2Br2 (2e) were obtained from (1c), (1d) and (1e),
respectively, in yields between 40 and 50% and identi®ed by
comparing their IR spectra with those reported in the
literature [13,14].
To our surprise, the pyrolysis of (1a) under the same
conditions did not yield any (2a) at all. Work-up in vacuo of
the volatile reaction products by fractional condensation at
1268C (trap 1), 1608C (trap 2) and 1968C (trap 3), and
investigation of the trap contents by low resolution IR
spectroscopy, and in part, by mass spectroscopy (see below)
gave the following results:
At pressures ꢁ0.1 mbar, none of the C3F4 isomers has
been detected among the decomposition products of (2a)
[21,22]. On the other hand, decomposition of (2a) in solu-
tion, and as a solid at 1968C, yields C2F4, most likely by
dimerization of :CF2 [3]. Concomitantly, a solid `carbon
polymer' built up of sp3 carbon centers (ꢁ (CC), Raman,
1
1326 cm [24]) deposits. The presumed decomposition
pathway of (1a) is shown in Scheme 2.
From the observed pyrolysis products of (1a), we con-
clude that, in spite of our inability to produce any (2a) owing
to secondary reactions, C2F2 is formed. Although (1a) is
obviously not suitable as a precursor for (2a), we are
planning to further investigate the decomposition of (1a)
under rigorously different conditions to support this con-
clusion.
3. Spectroscopic properties of squaric acid difluoride
ꢀ Trap 1 contained an involatile, yellow polymeric material
that changes in colour to black when exposed to air. We
presume that this material emerges from the polymeriza-
tion of difluoropropadienone (4) in analogy to observa-
tions by Brahms and Dailey [15].
We have studied (1a) and (1b) by 13C and 19F NMR
spectroscopy, and furthermore, (1a) by mass, IR and Raman
spectroscopy.
3.1. NMR spectra
ꢀ Trap 2. The material obtained in this trap was mostly (4),
which was identified by its IR spectrum [15]. In addition,
a second volatile, yet unidentified product exerting IR
absorptions at 2258(s), 1735(s), 1278(s), 1197(w) and
980(w) cm 1, and an involatile polymer formed by sec-
ondary reactions, were detected by IR spectroscopy.
ꢀ Trap 3 contained, along with some SiF4 and the uni-
dentified volatile product also found in trap 2, the three
C3F4 isomers CF3C>CF (5), CF2CCF2 (6) and cyclo-
C3F4 (7).
The NMR data of (1a) and (1b) are listed in Table 1, while
Fig. 1 displays the 13C and 19F NMR spectra of (1a). In
principle, both ole®nic and carbonyl resonances represent
the X parts of ABX spin systems. Due to the large differ-
ences in coupling constants J(AX), 364.0 Hz, and J(BX),
2.7 Hz, as well as the very small AB (3J(FF)) coupling of
1.0 Hz, the signal of the 13CFA == 12CFB fragment reveals a
®rst-order doublet of doublets pattern. In contrast, high-

M. Senzlober et al. / Journal of Fluorine Chemistry 99 (1999) 99±104
101
couplings in di¯uorocyclopropenone (2J(CF) 7 Hz) [13],
and perhaps also with the unusual isotope shift of the carbon
in the trans position to ¯uorine.
The relative low-®eld and high-®eld shifts of the 13C(1)
and 13C(2) signals of (1b) follow the usual F/Cl substitution
patterns observed for ole®nic carbon atoms, while those of
the carbonyl groups are hardly changed. As for (1a), the
assignment of the latter ones is questionable.
Table 1
13C NMR data of (1a) and (1b)a
(1a)
(1b)
71.3
ꢂ
ꢂ
19F
79.3
186.8/364.0
0.123
13C(1)/1J(CF)
197.7/370.5
0.122
Áꢂb
13C(2)/2J(CF)
ꢂ
Áꢂb
ꢂ
Áꢂb
186.8/2.7
0.031
174.6/5.7
0.026
13C(3)/3J(CF)
186.0/37.2
+0.005
186.0/8.1
0.033
187.1/41.9
+0.007
188.6/10.3
0.030
ꢂ
13C(4)/2J(CF)
3.2. Mass spectra
Áꢂb
3J(FF)
1.0
±
The mass spectrum of (1a) is characterized by the suc-
cessive loss of 2CO entities from the M+ ion. This even-
+
tually yields the C2F2 ion, which is the base peak. Mass
a In C6D6, chemical shifts in ppm relative to C6D6 at 128.0 ppm (13C)
and external CFCl3 (
19F). C(1) and C(2) refer to the olefinic, C(3) and
C(4) to the carbonyl carbons, respectively; (13C) shifts for (1c) are 187.2
and 189.8 ppm, respectively.
peaks corresponding to loss of COF and thereafter CO occur
with lower intensity. We note, however, that a strong peak at
m/e 43 (C2F+) has also been observed in the mass spectrum
of (2a) [3]. The mass spectrum of (1a) is signi®cantly
different from that of 1,4-di¯uoro-2-butyne-1,4-dione
[10,11], but it does not allow to distinguish between a
squaric acid (1a) and bis(ketene) structure (3a) since, to
the best of our knowledge, data on the latter are not yet
available for comparison.
b
(
13C) isotope shifts (ppm) of the (19F) resonances, ꢂ 19F(13C) ± ꢂ
19F(12C).
order effects with the expected six lines are clearly visible
for the carbonyl carbons when recorded with a 400 MHz
spectrometer (see insert of Fig. 1). This is even more clearly
evident from the appearance of the 13C satellites in the 19
F
spectrum (Fig. 1). All 13C nuclei cause a detectable isotope
shift of the 19F resonance, one of them being positive (Table
1). Due to an accidental coincidence of the low-®eld parts of
the 2J and 3J doublets of the 19F±13C(O) spectra, one of the
AB subsystems degenerates to a singlet, which proves equal
(presumably positive) signs for 2J(CF) and 3J(CF). A
conclusive assignment of C(3) and C(4) is not possible.
The proposed assignment contradicts the usual assumption
2J(CF) > 3J(CF). It is, however, in accord with the observed
Peaks were observed at m/e (relative intensity) 118 (52),
+
C4F2O2 (M+); 90 (78), C3F2O+, (M+ ± CO); 71 (16),
+
C3FO+, (M+ ± COF); 62 (100), C2F2 (M+ ± 2CO); 43
(13), C2F+.
3.3. Vibrational spectrum
We have recorded a gas phase infrared spectrum of (1a) in
the 400±4000 cm 1 region and Raman spectra of (1a) both
Fig. 1. Left hand side: 13C NMR spectrum of (1a), Bruker AC250, 62.90 MHz, solvent CDCl3 at 77.0 ppm; the insert shows the C(O) signals recorded with a
Bruker ARX400, 100.61 MHz, solvent C6D6 at 128.0 ppm. Right hand side: 19F NMR spectrum of (1a) with 13C satellites, Bruker ARX400, 376.50 MHz.

102
M. Senzlober et al. / Journal of Fluorine Chemistry 99 (1999) 99±104
Table 2
1
Infrared and Raman spectra of (1a) (cm
)
IR
Raman
Solid
a,b
Gasa
in CDCl3
1893 (8)
1873 (9)
1836 (59)
1694 (100)
1389 (86)
1139 (3)
1018 (9)
941 (13)
910 (11)
1910 w
1850 w
1810 s
1675 s
1885 (9), 0
1855 (12), 0
1815 (42), 0.7
1685 (100), 0.1
1157 w
1035 w
970 w
940 w
789 w
655 s
1152 (10), 0.4
1028 (3)
960 (5), 0.5
940 (5), 0.5
785 (7), 0.7
655
644 (7)
598 m
565 m
307 w
297 w
245 s
598 (16), 0.7
565 (9), 0.7
296 (12), 0.5
230 (38), 0.7
Scheme 1.
a Relative intensities in parentheses.
b Estimated depolarization ratio ꢃ = I /III.
?
The calculated harmonic wavenumbers and infrared
intensities are listed in Table 4. Those for (1a) are in
excellent agreement with the experimental data.
as a solid and in CDCl3 solution. Wavenumbers and relative
intensities are gathered in Table 2. Assuming planarity with
C2v symmetry, the 18 vibrational fundamentals ꢁ1±ꢁ18
transform as follows:
In order to rule out a bis(ketene) structure (3a) and a
bicyclobutane structure (8) for the carrier of the vibrational
spectrum (Table 2), the theoretical spectra of (3a) and (8)
were also calculated. These calculations were carried out at
the (lower) Hartree±Fock level with the 6±31G* basis set
[28,29] using analytical derivatives. Again, the structures
were gradient optimized applying appropriate symmetry
constraints. In order to reduce systematic errors in the
harmonic wavenumbers, the obtained values were uni-
formly scaled using the factor (1.13) 1 [30]. For (3a), three
conformations (i.e. cisoid C2v, transoid C2h and twisted C2)
were investigated. Only the twisted C2 structure was found
to be a minimum on the energy surface, which is in perfect
agreement with calculations carried out earlier by McAll-
ister [8].
Gvib ˆ 7A1 ‡ 3A2 ‡ 2B1 ‡ 6B2
The A2 species are infrared forbidden but Raman active.
The assignment is made on the basis of the ab initio
calculations displayed in Table 4 of both the fundamental
wavenumbers and infrared intensities. The agreement is
1
convincing. The two doublets at 1893/1873 cm
1
and
941/910 cm are ascribed to Fermi resonance. This is
not considered in the ab initio calculations.
4. Ab initio calculations
4.1. Computational details
Table 3
Calculated bond distances (Angstrom) and angles (degree) of (1a)
The quantum-chemical calculations for (1a) and its iso-
mers (3a) and 1,3-di¯uorobicyclobutane-2,4-dione (8)
(Scheme 1) were carried out using the Gaussian 94 series
of programs [25]. All calculations were performed on IBM
RS/6000 workstations at the computing center of the Uni-
versity of ZuÈrich and on the NEC-SX4 vector computer of
the Swiss Center for Scienti®c Computing (SCSC).
For (1a), the geometry was gradient optimized using the
6-311 + G(2d,p) basis set [26] at the MP2 level of theory
with a C2v symmetry constraint. The theoretical harmonic
wavenumbers were determined at the same theoretical level
using analytical second derivatives. The frozen core approx-
imation was not used. The calculated bond lengths and
angles are listed in column 2 of Table 3 and are compared to
the results obtained earlier by Zhou et al. [27].
[27]a
1.334
This workb
C1C2
1.366
1.499
1.577
1.305
1.200
94.1
C1C3
1.488
1.548
1.291
1.174
C3C4
C1F5
C3O7
C2C1C3
C1C3C4
C3C1F5
C1C3O7
94.1
85.9
132.1
136.7
85.9
133.1
136.4
a SCF/6-31G*.
b MP2/6-311 + G(2d, p).

M. Senzlober et al. / Journal of Fluorine Chemistry 99 (1999) 99±104
103
Table 4
Theoretical wavenumbers (cm 1) of (1a), (3a) and (8)
Symmetry species
(1a)a
Symmetry species
(3a)b
Symmetry species
(8)b
A1
B2
A1
B2
A1
B2
A1
B2
A2
A1
B2
A2
B1
A1
B2
B1
A1
A2
1855 (60)
1820 (222)
1697 (456)
1409 (318)
1150 (9)
1023 (30)
923 (71)
795 (2)
A
B
A
B
A
B
A
B
A
A
B
A
B
B
A
B
A
A
2140 (464)
2105 (1591)
1383 (8)
1315 (105)
1267 (286)
1011 (70)
700 (12)
676 (22)
568 (1)
A1
B1
A1
B2
B1
A2
B2
A2
A1
A1
B1
B2
A1
B2
B1
A2
A1
A1
1874 (2)
1732 (1883)
1511 (9)
1345 (1229)
1151 (0)
1145 (0)
932 (14)
781 (0)
784 (0)
682 (2)
639 (31)
590 (4)
507 (15)
501 (43)
447 (2)
652 (3)
616 (1)
570 (0)
569 (1)
601 (22)
575 (2)
446 (11)
254 (13)
216 (0)
285 (8)
300 (14)
273 (10)
227 (0)
270 (8)
223 (15)
222 (0)
130 (2)
118 (2)
216 (38)
38 (1)
113 (0)
55 (0)
a MP2, 6-311 + G(2d,p), intensities I (km mol 1).
b SCF, 6-31G*, scaled wavenumbers (see text), intensities I (km mol 1).
5. Discussion and conclusion
ism. Secondary reactions involving :CF2 predominate. We
believe that under collision-free conditions, i.e., lower
We have synthesized, for the ®rst time, (1a) by ¯uorina-
tion of (1c) in good yield. The main difference from
previous unsuccessful attempts is that, rather than in solu-
tion, we performed the ¯uorination in the gas phase at a
temperature as low as 2508C using the common ¯uorinating
agent KF.
We have unambiguously established by means of NMR,
mass and vibrational spectra the identity of the product
obtained. In order to rule out that any isomerization had
occurred during ¯uorination, which may not be evident from
NMR and mass spectra, we have performed ab initio
calculations of the infrared spectra of (1a), (3a) and (8).
The infrared stick spectra of (1a), (3a) and (8) which are
compared to the observed spectrum in Fig. 2 unambiguously
prove that the obtained product is (1a) and neither (3a) nor
(8). The calculated structure of (1a) (Table 3) is not too
different from that obtained previously [27]. While the
angles are quite consistent, all bond lengths are found to
be systematically longer when the calculations are per-
formed at a higher level. This ab initio structure can be
compared to that of (1c) determined by gas phase electron
Ê
diffraction [31]. The dimensions (in A and degree) of the
O2C4 unit (this work, Table 3/[31], 2ꢄ) are C1C2 1.366/
1.358(6), C2C3 1.499/1.516(6), C3C4 1.577/1.574(9),
C3O7 1.200/1.191(2) and C2C1C3 94.1/94.1(2). The agree-
ment between these structural parameters is surprisingly
good. On the other hand, this means that the structures of
and the bonding in (1a) and (1c) are closely related.
Although the thermolysis of (1a) has not yielded any
C2F2, it is likely from the characterized decomposition
products which are framed in Scheme 2 that either (2a)
or :C == CF2 are intermediates in the thermolysis mechan-
Fig. 2. Experimental(1aexp)andabinitiostickspectrafor(1a), (3a)and(8).
Note the agreement between the two upper spectra and the disagreement
of the experimental spectra with those computed for (3a) and (8).

104
M. Senzlober et al. / Journal of Fluorine Chemistry 99 (1999) 99±104
[6] Nippon Electric Co., Ltd. Jpn. Kokai Tokkyo Koho 80 161 377
(1980).
[7] B. Lunelli, G. Orlandi, F. Zerbetto, M.G. Giorgini, J. Mol. Struct.
(Theochem.) 201 (1989) 307.
[8] M.A. McAllister, T.T. Tidwell, J. Am. Chem. Soc. 116 (1994) 7233.
[9] R.C. De Selms, C.J. Fox, R.C. Riordan, Tetrahedron Lett. (1970)
781.
[10] F.E. Herkes, H.E. Simmons, J. Org. Chem. 40 (1975) 420.
[11] F.E. Herkes, H.E. Simmons, Synthesis (1973) 166.
[12] A.H. Schmidt, M. Russ, D. Grosse, Synthesis (1981) 216.
[13] P. Klaboe, E. Kloster-Jensen, D.H. Christensen, I. Johnsen, Spectro-
chim. Acta 26A (1970) 1567.
[14] D.H. Christensen, T. Stroyer-Hansen, P. Klaboe, E. Kloster-Jensen,
E.E. Tucker, Spectrochim. Acta 28A (1972) 939.
[15] J.C. Brahms, W.P. Dailey, J. Am. Chem. Soc. 111 (1989) 3071.
[16] H.B. Friedrich, D.J. Burton, P.A. Schemmer, Spectrochim. Acta 45A
(1989) 181.
[17] J.R. Durig, Y.S. Li, J.D. Witt, A.P. Zens, P.D. Ellis, Spectrochim.
Acta 33A (1977) 529.
[18] N.C. Craig, G.F. Fleming, J. Pranata, J. Phys. Chem. 89 (1985)
100.
È
È
[19] C. Kotting, W. Sander, M. Senzlober, H. Burger, Chem. Eur. J. 4
(1998) 1611.
[20] J.C. Brahms, W.P. Dailey, J. Am. Chem. Soc. 112 (1990) 4046.
[21] D. McNaughton, P. Elmes, Spectrochim. Acta 48A (1992) 605.
[22] H.B. Friedrich, D.C. Tardy, D.J. Burton, J. Fluorine Chem. 54 (1993)
53.
Scheme 2.
[23] H.B. Friedrich, D.J. Burton, D.C. Tardy, J. Phys. Chem. 91 (1987)
6334.
È
[24] M. Senzlober, Thesis, Universitat GH Wuppertal, 1998.
pressure and shorter pyrolysis time, it may be possible to
generate C2F2 from (1a).
[25] M.J. Frisch, G.W. Trucks, H.B. Schlegel, P.M.W. Gill, B.G. Johnson,
M.A. Robb, J.R. Cheeseman, T. Keith, G.A. Petersson, J.A.
Montgomery, K. Raghavachari, M.A. Al-Laham, V.G. Zakrzewski,
J.V. Ortiz, J.B. Foresman, J. Cioslowski, B.B. Stefanov, A.
Nanayakkara, M. Challacombe, C.Y. Peng, P.Y. Ayala, W. Chen,
M.W. Wong, J.L. Andres, E.S. Replogle, R. Gomperts, R.L. Martin,
D.J. Fox, J.S. Binkley, D.J. Defrees, J. Baker, J.P. Stewart, M. Head-
Gordon, C. Gonzalez, J.A. Pople, Gaussian 94, Gaussian, Inc.,
Pittsburgh, PA, 1995.
Acknowledgements
We thank the Deutsche Forschungsgemeinschaft and the
Fonds der Chemie for ®nancial support.
[26] R. Krishnan, J.S. Binkley, R. Seeger, J.A. Pople, J. Chem. Phys. 72
(1980) 650.
References
[27] L. Zhou, J. Gu, A. Tian, J. Mol. Struct. (Theochem.) 363 (1996)
333.
[1] H. Bock, W. Ried, U. Stein, Chem. Ber. 114 (1989) 673.
[2] I. Mincu, M. Hillebrand, A. Allouche, M. Cossu, P. Verlaque, J.P.
Aycard, J. Pourcin, J. Phys. Chem. 100 (1996) 16045.
[28] W.J. Hehre, R. Ditchfield, J.A. Pople, J. Chem. Phys. 56 (1972)
2257.
[29] P.C. Hariharan, J.A. Pople, Theoret. Chim. Acta 28 (1973) 213.
[30] J.A. Pople, H.B. Schlegel, R. Krishnan, D.J. DeFrees, J.S. Binkley,
M.J. Frisch, R.A. Whiteside, Int. J. Quant. Chem.: Quant. Chem.
Symp. 15 (1981) 269.
[3] H. BuÈrger, S. Sommer, J. Chem. Soc., Chem. Commun. (1991) 456.
È
[4] J. Breidung, H. BuÈrger, C. Kotting, R. Kopitzky, W. Sander, M.
Senzlober, W. Thiel, H. Willner, Angew. Chem. 109 (1997) 2072.
È
[5] J. Breidung, H. BuÈrger, C. Kotting, R. Kopitzky, W. Sander, M.
Senzlober, W. Thiel, H. Willner, Angew. Chem., Int. Ed. Engl. 36
(1997) 1983.
[31] K. Hagen, K. Hedberg, B. Lunelli, M.G. Giorgini, J. Phys. Chem. 92
(1988) 4313.