Gas phase structures and conformations of trifluoromethanesulfonyl fluoride, CF3SO2F, difluoromethanesulfonyl fluoride, CHF2SO2F, and difluoromethanesulfonyl chloride, CHF2SO2Cl

Article abstract of DOI:10.1016/S0022-2860(00)00382-3

The gas phase structures of CF₃SO₂F, CHF₂SO₂F, and CHF₂SO₂Cl have been studied by gas electron diffraction (GED) and quantum chemical calculations (HF/6-31G and B3LYP/6-31G). The two compounds CHF₂SO₂X (X = F or Cl) exist in the gas phase as mixtures of trans and gauche conformers (C-H bond trans or gauche to S-X bond). In the case of CHF₂SO₂F the gauche conformer prevails (84(17)%), whereas for the chlorine derivative the trans form is the major component (69(9)%). These compositions are reasonably well reproduced by both computational methods. The S-C bond lengths (1.835(5) ? in CF₃SO₂F, 1.822(5) ? in CHF₂SO₂ F and 1.846(5) ? in CH₂FSO₂Cl) are compared to those in other sulfonyl derivatives. (C) 2000 Elsevier Science B.V.

Full text of DOI:10.1016/S0022-2860(00)00382-3

Journal of Molecular Structure 550–551 (2000) 59–65
www.elsevier.nl/locate/molstruc
Gas phase structures and conformations of
trifluoromethanesulfonyl fluoride, CF SO F,
3
2
difluoromethanesulfonyl fluoride, CHF SO F, and
2
2
ଝ
difluoromethanesulfonyl chloride, CHF SO Cl
2
2
a
a
b
b
b
a,*
R. Haist , F. Trautner , J. Mohtasham , R. Winter , G.L. Gard , H. Oberhammer
a
Institut f u¨ r Physikalische und Theoretische Chemie, Universit a¨ t T u¨ bingen, 72076 T u¨ bingen, Germany
b
Department of Chemistry, Portland State University, Portland, OR 97207, USA
Received 7 June 1999; received in revised form 2 July 1999; accepted 2 July 1999
Abstract
The gas phase structures of CF
quantum chemical calculations (HF/6-31G and B3LYP/6-31G ). The two compounds CHF
phase as mixtures of trans and gauche conformers (C–H bond trans or gauche to S–X bond). In the case of CHF
gauche conformer prevails (84(17)%), whereas for the chlorine derivative the trans form is the major component (69(9)%).
3
SO
2
F, CHF
2
SO
2
F, and CHF
2
SO
2
Cl have been studied by gas electron diffraction (GED) and
SO
X (X ˆ F or Cl) exist in the gas
SO F the
ء
ء
2
2
2
2
˚
These compositions are reasonably well reproduced by both computational methods. The S–C bond lengths (1.835(5) A in
˚ ˚
CF
Elsevier Science B.V. All rights reserved.
3 2 2 2 2 2
SO F, 1.822(5) A in CHF SO F and 1.846(5) A in CH FSO Cl) are compared to those in other sulfonyl derivatives. ᭧ 2000
Keywords: Gasphase structures; Conformational properties; Gas electron diffraction; Quantum chemical calculations
˚
(CF ) SO (1.858(5) A [4]) these bond lengths are
3 2 2
1
. Introduction
˚
longer by about 0.08 A than those in the analogous
˚
non-fluorinated compounds (CH ) SO (1.808(4) A
3 2
Whereas S–C bond lengths in S(II) compounds
˚
[5]) and (CH ) SO (1.771(4) A [6]). This strong
3 2 2
change very little upon CH /CF substitution, these
3
3
bonds lengthen drastically with such substitutions in
S(IV) and S(VI) derivatives. The S–C bond length in
CH /CF substitution effect has been rationalized by
3 3
polar effects [7]. The net charge on sulfur atoms in
S(II) compounds is expected to be small and polar
effects play a minor role in the parent and in the
fluorinated derivative. In S(IV) and S(VI) compounds,
however, sulfur carries a high positive net charge. In
the methyl derivatives the carbon atom possesses a
small, probably negative net charge and the polar
interactions are negligible or slightly attractive. In
the fluorinated compounds, however, the carbon
atom carries a large positive charge and this leads to
˚
CF ) S (1.819(3) A [1]) is only slightly longer than
3 2
(
˚
that in (CH ) S (1.807(2) A [2]). However, in the sulf-
oxide (CF ) SO (1.885(4) A [3]) and in the sulfone
3
2
˚
3
2
ଝ
Dedicated to Professor James R. Durig on the occasion of his
5th birthday.
Corresponding author. Tel.: ϩ 49-7071-296907; fax: ϩ 49-
6
7
*
071-296910.
E-mail address: heinz.oberhammer@uni-tuebingen.de
(H. Oberhammer).
0
022-2860/00/$ - see front matter ᭧ 2000 Elsevier Science B.V. All rights reserved.
PII: S0022-2860(00)00382-3

6
0
R. Haist et al. / Journal of Molecular Structure 550–551 (2000) 59–65
Scheme 1.
The compound CHF SO Cl was prepared according
2
2
to the literature [8]. CHF SO F was prepared
2
2
according to the following: Into a 100 ml round
bottom flask, equipped with a condenser which was
connected to a trap cooled to Ϫ78ЊC, 6.20 g
3 2
Fig. 1. CF SO F. Experimental (dots) and calculated (full line)
molecular intensities for long (above) and short (below) nozzle-
to-plate distances and residuals.
(
41.2 mmol) of CF HSO Cl, 3.25 g (56.0 mmol) of
2 2
ϩ
ϩ
a strong S –C electrostatic repulsion and thus to
lengthening of this bond.
dry KF, 62.2 ml of CH COOH and 0.375 g of
(CH CO) O were added. The mixture was heated to
3
3
2
In the present study we are primarily interested
whether this simple electrostatic model applies also
to S(VI) compounds with other substituents. For this
purpose we determined the gas phase structures of two
compounds with partially fluorinated methyl groups,
CHF SO F and CHF SO Cl and that of CF SO F,
90ЊC for 10 h and distilled giving 2.74 g (20.4 mmol)
1
9
of CHF SO F, bp 51 ^ 1ЊC: The yield was 50%. F-
2
2
NMR (neat F-11 ext.): SF, d ˆ 36:7 ppm (t,d); CF ,
2
d ˆ Ϫ127:3 ppm (d,d). Integration ratio of CF to SF
2
2.0:1.0. H-NMR (neat, TMS ext.): d ˆ 6:6 ppm (d,t).
The infrared spectrum also confirmed the structure:
2
2
2
2
3
2
Ϫ1
Ϫ1
where one CF group of bis(trifluoromethyl)sulfone
nꢀCH† ˆ 3015 cm
(w), n ꢀSO † ˆ 1447 cm
3
as
2
Ϫ1
is replaced by a fluorine atom. The structures were
determined by gas electron diffraction (GED) and by
quantum chemical calculations.
(vs), n ꢀSO † ˆ 1236 cm (vs), nꢀC–F† ˆ 1160 and
s
2
Ϫ1 Ϫ1
1140 cm (vs) and nꢀS–F† ˆ 835 cm (s).
The electron diffraction intensities were recorded
with a Gasdiffraktograph KD-G2 [9] at nozzle-to-
plate distances of 25 and 50 cm and with an acceler-
ating voltage of about 60 kV. The samples were
cooled to Ϫ70ЊC (CF SO F), Ϫ28ЊC (CHF SO F)
2. Experimental section
3
2
2
2
and 4ЊC (CHF SO Cl), respectively. The inlet system
A commercial sample of CF SO F (99%) was used.
2
2
3
2
and the gas nozzle were at room temperature. The
photographic plates were analyzed with the usual
methods [1] and averaged molecular intensities for
CF SO F and CHF SO Cl in the s-ranges 2–18 and
3
2
˚
2
2
Ϫ1
8
–35 A are presented in Figs. 1 and 2. The inten-
sities for CHF SO F are very similar to those for
2
2
CF SO F and are not shown.
3
2
3
. Quantum chemical calculations
The geometric structures of the three title
ء
compounds were optimized with the HF/6-31G
method and with density functional theory (B3LYP/
2 2
Fig. 2. CHF SO Cl. Experimental (dots) and calculated (full line)
molecular intensities for long (above) and short (below) nozzle-to-
plate distances and residuals.
ء
6-31G ) using the Gaussian 94 program [10]. In the

R. Haist et al. / Journal of Molecular Structure 550–551 (2000) 59–65
61
The molecular intensities were modified with a diag-
onal weight matrix.
4
.1. CF SO F
3 2
The RDF is shown in Fig. 3. C overall symmetry
S
and C3v local symmetry for the CF group were
3
assumed in the least squares refinement. Eight
geometric parameters and eight vibrational ampli-
tudes were refined simultaneously. The following
correlation coefficients had values larger than ͉0.7͉:
CF=FCF ˆ Ϫ0:80; CSO=OSO ˆ Ϫ0:80; CSO=CSF ˆ
Ϫ0:72; SO=l1 ˆ Ϫ0:82; CSF=l9 ˆ 0:70; CSF=l11 ˆ
0
0
:88; CSO=l9 ˆ Ϫ0:76; CSO=l11 ˆ Ϫ0:76; l9=l11 ˆ
:78: The final parameters are collected in Table 1
3 2
Fig. 3. CF SO F. Experimental RDF and difference curve
RDFꢀexp† Ϫ RDFꢀcalc†:
(geometric parameters) and Table 2 (vibrational
amplitudes). The agreement factors for the intensities
of the long (50 cm) and short (25 cm) nozzle-to-plate
distances are R50 ˆ 2:2% and R25 ˆ 7:0%:
case of the two partially fluorinated sulfonyl deriva-
tives CHF SO X (X ˆ F or Cl) two possible confor-
2
2
mers exist, with the hydrogen atom trans or gauche
with respect to X (see Scheme 1). Vibrational ampli-
tudes were derived from calculated Cartesian force
4
.2. CHF SO F
2 2
ء
constants (B3LYP/6-31G ) with the program
The calculated RDFs for trans and gauche isomers
˚
ASYM40 [11].
(Fig. 4) differ mainly in the region around 2.9 A which
corresponds to (F···F)-gauche and (O···F)-gauche
distances. Comparison with the experimental curve
demonstrates that the gauche conformer is predomi-
nant. In the least squares refinements local CS
4
. Structure analyses
symmetry was assumed for the SO F group. The
2
Radial distribution functions (RDF) were calcu-
difference
between
the
S–C–F
angles
lated by Fourier transformations of the molecular
intensities, applying an artificial damping function
ꢀDSCF ˆ ꢀS–C–F2† Ϫ ꢀS–C–F1† and the S–C–H
2
2
angle were set to the calculated values (B3LYP).
Six vibrational amplitudes and 10 geometric para-
meters of the gauche conformer were refined simulta-
neously. The differences between geometric
parameters of the trans and gauche conformers and
the vibrational amplitudes of the trans form were set
to calculated values (B3LYP). The following correla-
tion coefficients had values larger than ͉0.7͉: SO=CF ˆ
Ϫ0:73; CF=FCF ˆ Ϫ0:82; CSO=OSO ˆ Ϫ0:74;
CSF=F ˆ Ϫ0:85; CF=l2 ˆ 0:89; FCF=l2 ˆ Ϫ0:72;
CSO=l8 ˆ Ϫ0:86: Refinements for various conforma-
tional compositions were performed and the quality of
the fit was measured by the agreement factor. The best
fit was obtained for 84(17)% gauche and 16(17)%
ꢀ
expꢀϪgs † with g ˆ 0:0019 A : Preliminary struc-
tural models were refined by least squares analyses.
Table 1
3 2
Experimental and calculated geometric parameters of CF SO F
a
ء
ء
GED
HF/6-31G
B3LYP/6-31G
S–C
SyO
S–F
1.835(5)
1.410(3)
1.543(3)
1.325(2)
95.4(7)
1.818
1.405
1.549
1.303
96.2
108.9
123.9
109.9
180
1.873
1.445
1.612
1.327
95.7
109.0
124.1
110.2
180
(
C–F)mean
C–S–F
C–SyO
OySyO
F–C–F
108.5(6)
124.1(10)
109.8(3)
b
F (F–S–C–F1) 180
trans ꢀR50 ˆ 3:9% and R ˆ 8:7%†: The uncertainty
25
a
was derived by Hamilton’s test [12] at the 1% signifi-
cance level. The final results are given in Table 3
(geometric parameters) and Table 2 (vibrational
˚
a
The r —values in A and Њ. Uncertainties are 3s values. Mole-
cular model is shown in Fig. 3.
b
Not refined.

6
2
R. Haist et al. / Journal of Molecular Structure 550–551 (2000) 59–65
Table 2
Experimental and calculated vibrational amplitudes for CF
3
SO
2
F, CH
2
FSO
2
F and CH
2
FSO
2
Cl
3 2
CF SO F
CF
2
HSO
2
F
CF
2
2
HSO Cl
a
b
a
b
a
b
GED
B3LYP
GED
B3LYP
GED
B3LYP
C–F
SyO
S–X
l1
l2
l3
l4
l5
l6
l7
l8
0.045(4)
0.036
0.045
0.036
0.045
0.055
0.058
0.069
0.058
0.093
0.075
0.085
0.226
0.193
0.078
0.045
0.044(3)
0.045
0.049(6)
0.060
0.070(7)
0.058
0.063(9)
0.063(9)
0.063(9)
0.096(17)
0.096(17)
0.068(13)
0.045
0.036
0.045
0.057
0.060
0.070
0.058
0.095
0.077
0.087
0.213
0.190
0.077
0.045
0.045
0.037
0.057
0.058
0.059
0.084
0.059
0.101
0.075
0.091
0.219
0.167
0.076
0.038(2)
0.053(2)
0.056(5)
0.059
0.069(4)
0.059
0.101
0.063(4)
0.063(4)
0.045(4)
0.047(7)
0.060(3)
0.060(7)
0.058
0.093
0.068(5)
0.085
S–C
F···F
O···X
O···O
C···X
S···F
C···O
l9
l10
l11
l12
l13
(X···F)gauche
(O···F)gauche
(Y···F)trans
0.116(22)
0.116(22)
0.075(7)
0.229(22)
0.144(28)
0.071(6)
a
˚
Values in A, uncertainties are 3s values. Amplitudes without uncertainties are not refined. X ˆ F or Cl and Y ˆ O, F or Cl.
b
Mean values are given for amplitudes which are not unique.
˚
amplitudes). Molecular models of both conformers
are shown in Fig. 5.
trans distance at 4.13 A. Comparison between calcu-
lated and experimental curves demonstrates that for
this compound the trans form is predominant. The
least squares refinements were performed analogous
4
.3. CHF SO Cl
2 2
to those described above. C overall symmetry was
S
assumed for this conformer. Nine geometric para-
meters and eight vibrational amplitudes of the trans
conformer were refined simultaneously. Differences
between geometric parameters of the gauche and
trans forms and vibrational amplitudes of the gauche
conformer were constrained to calculated values
The calculated RDFs for the two possible confor-
mers are shown in Fig. 6. Whereas the trans form
possesses two (Cl···F)-gauche distances around
˚
.17 A, the gauche conformer possesses one
3
(
˚
Cl···F)-gauche distance at 3.10 A and one (Cl···F)-
(B3LYP). The following correlation coefficients had
values larger than ͉0.7͉: CSO=OSO ˆ Ϫ0:83; CF=l2 ˆ
0
:71; CSO=l9 ˆ Ϫ0:84; OSO=l9 ˆ 0:72: The best fit
of the intensities was obtained for a mixture of 69(9)%
trans and 31(9)% gauche conformer ꢀR ˆ 3:7% and
5
0
R
ˆ 4:9%†: The final results are collected in Table 4
25
(geometric parameters) and Table 2 (vibrational
amplitudes).
5. Discussion
The GED intensities for CHF SO X (X ˆ F or Cl)
2
2
are reproduced best with mixtures of trans and gauche
2 2
Fig. 4. CHF SO F. Calculated RDFs for trans and gauche confor-
conformers. In the case of sulfonyl fluoride,
mers, experimental curve and difference curve for mixture. Impor-
tant interatomic distances for the gauche form are indicated by
vertical bars. For molecular models see Fig. 5.
CHF SO F, the gauche conformer is favored, but the
2
2
composition is not well determined in the GED

R. Haist et al. / Journal of Molecular Structure 550–551 (2000) 59–65
63
Table 3
Experimental and calculated geometric parameters for gauche conformer of CHF
2
2
SO F
a
ء
ء
GED
HF/6-31G
B3LYP/6-31G
S–C
SyO
S–F
1.822(5)
1.400(2)
1.542(3)
1.344(7)
1.807
1.408
1.556
1.320
1.077
95.9
109.6
123.5
106.7
1.5
1.862
1.447
1.619
1.340
1.094
95.1
109.7
123.9
108.1
1.6
(C–F)mean
b
C–H
C–S–F
1.100
95.2(14)
109.9(8)
123.8(17)
107.2(6)
(C–SyO)mean
OySyO
(
S–C–F)mean
b
SD –S CC F– Hˆ ꢀS–C–F2† Ϫ ꢀS–C–F1†
F–C–F
1.6
109.7
b
110.1
109.8
61.4
109.3
110.2
57.5
107.4(10)
65.7(35)
84(17)
c
F(gauche)
d
d
%
gauche
61
70
0
Ϫ1 e
DH /DE (kcal mol
)
Ϫ0.58(66)
0.14
Ϫ0.09
a
˚
a
The r —values in A and Њ. Uncertainties are 3s values. Molecular models are shown in Fig. 5.
Not refined.
b
c
d
e
Dihedral angle F(F–S–C–H).
Estimated from DE.
0
0
0
DH ˆ H ꢀgauche† Ϫ H ꢀtrans†:
experiment. If we neglect differences in the rotational
and vibrational contributions to the entropy difference
0
DS and take into account only the different multipli-
cities of trans and gauche conformers, we obtain from
0
the ratio of the mixture an enthalpy difference DH ˆ
0
0
Ϫ1
H ꢀgauche† Ϫ H ꢀtrans† ˆ Ϫ0:58ꢀ66† kcal mol
:
0
Considering the large uncertainty for DH , the
2 2
Fig. 6. CH FSO Cl. Calculated RDFs for gauche and trans confor-
mers, experimental curve and difference curve for mixture. Impor-
tant interatomic distances for the trans form are indicated by
vertical bars. For molecular models see Fig. 5.
Fig. 5. Molecular models of trans (above) and gauche (below)
conformers of CHF SO X compounds.
2
2

6
4
R. Haist et al. / Journal of Molecular Structure 550–551 (2000) 59–65
Table 4
Experimental and calculated geometric parameters for trans conformer of CHF
2
2
SO Cl
a
ء
ء
GED
HF/6-31G
B3LYP/6-31G
S–C
SyO
S–Cl
1.846(5)
1.417(1)
2.013(2)
1.330(3)
1.100
102.2(6)
1.817
1.417
2.005
1.318
1.079
101.3
107.2
122.2
108.9
107.1
109.7
180
1.882
1.456
2.104
1.336
1.095
99.8
107.5
122.7
108.4
106.4
110.5
180
(C–F)mean
b
C–H
C–S–Cl
(C–SyO)mean
107.9(6)
120.4(9)
107.8(5)
OySyO
S–C–F
S–C–H
F–C–F
F(trans)
b
106.7
110.5(6)
c
b
180
c
b
F(gauche)
trans
59.0
60.7
57.1
d
d
%
69(9)
0.90(19)
81
57
0
Ϫ1 e
DH /DE (kcal mol
)
1.27
0.59
a
˚
a
The r —values in A and Њ. Uncertainties are 3s values. Molecular models are shown in Fig. 5.
Not refined.
b
c
d
e
Dihedral angle F(Cl–S–C–H).
Estimated from DE.
0
0
0
DH ˆ H ꢀgauche† Ϫ H ꢀtrans†:
˚
calculated energy differences ꢀDE ˆ 0:14 from HF
lengths satisfactorily, i.e. to within ^0.03 A. The
Ϫ1
ء
and Ϫ0.09 kcal mol
from B3LYP, respectively)
B3LYP/6-31G method however, results in bond
lengths to sulfur which are too long by 0.04–0.09 A
˚
agree reasonably well with the experiment. In sulfonyl
chloride, CHF SO Cl, the trans form prevails and an
(see Tables 1, 3 and 4). Similar discrepancies between
S–X bond lengths predicted by the B3LYP method
and experimental values occur also for other S(IV)
and S(VI) compounds, e.g. for FSO NySF and
2
2
0
0
0
enthalpy difference DH ˆ H ꢀgauche† Ϫ H ꢀtrans†
Ϫ1
of 0.90(19) kcal mol
is derived from the GED
experiment. This value is intermediate between the
2
2
energy differences calculated by the theoretical
FSO NyS(O)F [13]. For these two molecules
2 2
Ϫ1
methods ꢀDE ˆ 1:27 kcal mol
from HF andء
B3LYP and MP2 calculations predict bond lengths
Ϫ1
˚
0.59 kcal mol
from B3LYP). The HF/6-31G
too long by 0.03–0.06 A. The bond angles obtained
approximation reproduces the experimental bond
by both methods are in close agreement with the
Table 5
S–C bond lengths in various four-valent S(VI) compounds
Compound
S–C
Compound
S–C
Compound
S–C
a
b
CH
CH
CH
3
3
3
SO
SO
SO
2
2
2
CH
3
1.771(4)
1.763(5)
1.759(6)
CF
3
CF
3
CF
3
SO
SO
SO
2
2
2
CF
3
1.858(5)
1.856(6)
1.835(5)
c
d
e
Cl
CHF
CHF
2
SO
SO
2
Cl
F
1.846(5)
1.822(5)
Cl
f
d
d
F
2
2
F
a
b
c
d
e
f
Ref. [6].
Ref. [4].
Ref. [14].
This work.
Ref. [16].
Ref. [15].

R. Haist et al. / Journal of Molecular Structure 550–551 (2000) 59–65
65
experimental values. The vibrational amplitudes which
have been refined in the GED analyses are similar to
those derived from the B3LYP force field except for
those corresponding to (X···F)-gauche and (O···F)-
gauche distances. These experimental amplitudes are
consistently smaller than the calculated values (see
Table 2). Since these amplitudes depend primarily on
the torsional vibration around the S–C bond, we
Research Grant, which made these joint investigations
possible. We express our appreciation to Dr Fred
Behr, 3M Research Center, for a sample of
CF SO F. H.O. thanks the Fonds der Chemischen
3
2
Industrie for financial support.
References
conclude that the calculated frequencies (B3LYP) for
Ϫ1
[1] H. Oberhammer, W. Gombler, H. Willner, J. Mol. Struct. 70
(1981) 273.
this vibration are too low (39 cm
in CF SO F,
3 2
Ϫ1
Ϫ1
4
9 cm in CHF SO F and 53 cm in CHF SO Cl).
2 2 2 2
[2] T. Iiyima, S. Tsuchiya, M. Kimura, Bull. Chem. Soc. Jpn 50
As mentioned in Section 1 the primary interest in the
(1977) 2564.
present structural studies is the dependence of the S–C
bond length on the degree of fluorination of the methyl
group and on the sulfur substituents. Table 5 compares
our values with results of previous gas phase studies.
Two trends can be observed: (1) lengthening of the S–C
bond with increasing fluorination of the methyl group
[3] H. Oberhammer, R.C. Kumar, G.D. Knerr, J.M. Shreeve,
Inorg. Chem. 20 (1981) 3871.
[4] H. Oberhammer, G.D. Knerr, J.M. Shreeve, J. Mol. Struct. 82
(1982) 143.
[5] V. Typke, Z. Naturforsch Teil A 33 (1978) 842.
[6] M. Hargittai, I. Hargittai, J. Mol. Struct. 20 (1974) 283.
[7] H. Oberhammer, J. Fluorine Chem. 23 (1983) 147.
[8] G.G.I. Moore, J. Org. Chem. 44 (1979) 1708.
(from left to right in the table); and (2) shortening of this
[9] H. Oberhammer, Molecular Structure by Diffraction Methods,
vol. 4, The Chemical Society, London, 1976 p. 24.
bond with increasing electronegativity of the sulfur
substituent (from top to bottom in the table). The first
trend can be rationalized by the simple electrostatic
model presented in Section 1. Increasing fluorination
at carbon from CH to CHF and to CF leads to
[
10] M.J. Frisch, G.W. Trucks, H.B. Schlegel, P.M.W. Gill, B.G.
Johnson, M.A. Robb, J.R. Cheeseman, T.A. Keith, G.A.
Peterson, 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 (Revision
B.1) Gaussian Inc. Pittsburgh, PA, 1955.
3
2
3
increased positive net charge on carbon and thus to elec-
trostatic repulsion and lengthening of the bond. The
second trend cannot be explained by a simple electro-
static model. Contraction of the sulfur valence shell,
which depends on the electronegativity of the sulfur
substituent has to be assumed. This contraction leads
to shortening of all bonds to sulfur.
[
[
[
11] L. Hedberg, I.M. Mills, J. Mol. Spectrosc. 160 (1993) 117.
12] W. Hamilton, Acta Crystallogr. 18 (1965) 502.
13] R. Haist, R.M.S. Alvarez, E.H. Cutin, C.O. Della Vedova, H.
Oberhammer, J. Mol. Struct. 484 (1999) 249.
[14] M. Hargittai, I. Hargittai, J. Chem. Phys. 59 (1973) 2513.
[15] I. Hargittai, M. Hargittai, J. Mol. Struct. 15 (1973) 399.
[16] J. Brunvoll, I. Hargittai, M. Kolonits, Z. Naturforsch. Teil A
33 (1978) 1236.
Acknowledgements
The authors are grateful to NATO for a Collaborative