Synthesis, electrochemistry, and gas-phase photoelectron spectroscopic and theoretical studies of 3,6-bis(perfluoroalkyl)-1,2-dithiins

Article abstract of DOI:10.1021/jo034683n

3,6-Bis(trifluoromethyl)- and 3,6-bis(pentafluoroethyl)-1,2-dithiin (la,b), the first known perfluoroalkyl-substituted 1,2-dithiins, were synthesized from (Z,Z)-1,4-bis(tert-butylthio)-1,3-butadiene (2) to evaluate the effects of electron-withdrawing groups on the ionization and oxidation potentials of 1,2-dithiins. Analysis of the photoelectron spectra of la and lb provided a basis for assigning orbital compositions. Ab initio calculations on these compounds showed that they adopt a twist geometry as does 1,2-dithiin (1c) itself. Cyclic voltammetric studies on la and lb revealed a reversible oxidation followed by an irreversible oxidation at much more positive potentials than for 1,2-dithiin and 3,6-dimethyl-1,2-dithiin (1d). The oxidation potentials determined electrochemically do not correlate with the ionization potentials determined by photoelectron spectroscopy. This result supports the previously advanced hypothesis that there is a geometry change on electrochemical oxidation leading to a planar radical cation.

Full text of DOI:10.1021/jo034683n

Syn th esis, Electr och em istr y, a n d Ga s-P h a se P h otoelectr on
Sp ectr oscop ic a n d Th eor etica l Stu d ies of
3,6-Bis(p er flu or oa lk yl)-1,2-d ith iin s
Edward D. Lorance,† Richard S. Glass,*,† Eric Block,*,‡ and Xiaojie Li‡
Departments of Chemistry, The University of Arizona, Tucson, Arizona 85721, and
State University of New York at Albany, Albany, New York 12222
rglass@u.arizona.edu
Received May 21, 2003
3,6-Bis(trifluoromethyl)- and 3,6-bis(pentafluoroethyl)-1,2-dithiin (1a ,b), the first known perfluo-
roalkyl-substituted 1,2-dithiins, were synthesized from (Z,Z)-1,4-bis(tert-butylthio)-1,3-butadiene
(2) to evaluate the effects of electron-withdrawing groups on the ionization and oxidation potentials
of 1,2-dithiins. Analysis of the photoelectron spectra of 1a and 1b provided a basis for assigning
orbital compositions. Ab initio calculations on these compounds showed that they adopt a twist
geometry as does 1,2-dithiin (1c) itself. Cyclic voltammetric studies on 1a and 1b revealed a
reversible oxidation followed by an irreversible oxidation at much more positive potentials than
for 1,2-dithiin and 3,6-dimethyl-1,2-dithiin (1d ). The oxidation potentials determined electrochemi-
cally do not correlate with the ionization potentials determined by photoelectron spectroscopy. This
result supports the previously advanced hypothesis that there is a geometry change on electro-
chemical oxidation leading to a planar radical cation.
In two previous papers,1,2 we reported the synthesis,
electrochemistry, and gas-phase photoelectron spectro-
scopic and theoretical studies of 1,2-dithiins and their
derivatives containing alkyl groups at the 3,6-positions
(1), as well as related 1,2-dichalcogenins with selenium
in place of sulfur. Analysis of the photoelectron spectra
substituent effects. Methyl substituents, in particular,
showed a notable effect. Electrochemical studies1 on these
materials revealed an apparently reversible one-electron
oxidation. However, an EC mechanism, in which a
chemical step follows the electron-transfer step, was
required to satisfactorily fit the electrochemical data. In
addition, the electrochemical oxidation potentials did not
correlate with the lowest ionization potentials obtained
from the photoelectron spectroscopic measurements.
Theoretical calculations suggested that the chemical step
on one-electron oxidation is a change in geometry from
the twisted 1,2-dichalcogenin to a planar or flattened
cation radical. The extent of the geometry change de-
pended on substituents. Since these previously reported
studies involved electron-donating alkyl groups, it was
deemed important to determine the effects of electron-
withdrawing substituents. This paper presents the syn-
thesis, electrochemistry, and gas-phase photoelectron
spectroscopic and theoretical studies on 3,6-bis(trifluo-
romethyl)- and 3,6-bis(pentafluoroethyl)-1,2-dithiin (1a
and 1b, respectively), the first known perfluoroalkyl-
substituted 1,2-dithiins.
in combination with the theoretical studies provided
insight into the compositions of the highest occupied
molecular orbitals.2 The first four filled frontier molecular
orbitals of these compounds are associated with orbitals
that are primarily carbon π and chalcogen lone pair in
character derived from the 1,3-diene and dichalcogen
fragments. Orbital assignments of the observed ionization
were made on the basis of changes in ionization cross-
sections as a function of ionization photon energy3 and
Resu lts a n d Discu ssion
Syn th esis. The strongly electron-withdrawing per-
fluoroalkyl or fluorine groups have not been previously
introduced into the 1,2-dithiin ring system. We have
succeeded in incorporating the trifluoromethyl and pen-
tafluoroethyl groups into the known4 1,2-dithiin precur-
† The University of Arizona.
‡ State University of New York at Albany.
(1) Block, E.; Birringer, M.; DeOrazio, R.; Fabian, J .; Glass, R. S.;
Guo, C.; He, C.; Lorance, E.; Qian, Q.; Schroeder, T. B.; Shan, Z.;
Thiruvazhi, M.; Wilson G. S.; Zhang, X. J . Am. Chem. Soc. 2000, 122,
5052-5064.
(2) Glass, R. S.; Gruhn, N. E.; Lichtenberger, D. L.; Lorance, E.;
Pollard, J . R.; Birringer, M.; Block, E.; DeOrazio, R.; He, C.; Shan, Z.;
Zhang, X. J . Am. Chem. Soc. 2000, 122, 5065-5074.
(3) Glass, R. S.; Broeker, J . L.; J atcko, M. E. Tetrahedron 1989, 45,
1263-1272.
(4) Koreeda, M.; Wang, Y. J . Org. Chem. 1997, 62, 446-447.
10.1021/jo034683n CCC: $25.00 © 2003 American Chemical Society
Published on Web 09/19/2003
8110
J . Org. Chem. 2003, 68, 8110-8114

3,6-Bis(perfluoroalkyl)-1,2-dithiins
TABLE 1. An a lytica l Rep r esen ta tion s of P h otoelectr on
Da ta
SCHEME 1
relative area
label
positiona (eV)
width (eV)a
He Ib
He IIc
3,6-Bis(trifluoromethyl)-1,2-dithiin (1a )
1
2
3
4
9.10
10.57
10.87
12.44
0.47
0.46
0.48
0.55
1
1
1.35
0.78
0.98
1.12
1.22
1.05
3,6-Bis(pentafluoroethyl)-1,2-dithiin (1b)
1
2
3
4
9.06
10.44
10.67
12.10
0.46
0.36
0.49
-
1
1
0.70
0.93
0.77
0.84
1.48
1.35
1,2-Dithiind (1c)
1
2
3
4
8.16
9.82
10.06
11.51
0.46
0.40
0.50
0.53
1
1
0.96
0.72
0.91
0.86
1.05
1.26
3,6-Dimethyl-1,2-dithiind (1d )
1
2
3
4
7.78
9.31
9.63
0.46
0.42
0.36
0.52
1
1
1.02
0.64
1.06
1.05
0.50
1.06
10.93
a
b
Electron volts (23 kcal/mol). Ionization using monochromatic
He I UV radiation (21.22 eV). c Ionization using monochromatic
d
He II UV radiation (40.81 eV). Data taken from ref 2.
1d to compounds with larger alkyl groups at C3/6. The
olefinic protons of 1a and 1b (1H NMR δ 6.95-6.96) are
significantly deshielded compared to the analogous pro-
tons in 1d (δ 6.05). Irradiation of 1a with visible light in
an NMR tube led to the gradual disappearance of the
signal at δ 6.95 ppm while sulfur precipitated on the
inner wall of NMR tube. New absorptions at δ 7.43 ppm
were seen which belong to 2,5-bis(trifluoromethyl)-
thiophene5b (5a ; reported: δ 7.48 ppm; presence con-
firmed by GC/MS), formed by extrusion of sulfur from
sor (E,E)-1,4-bis(tert-butylthio)-1,4-diiodo-1,3-butadiene
(3) through reaction with the corresponding perfluoro-
alkylcopper complex. Compound 3 was prepared in 47%
yield (along with 8% of the mono-iodo compound) by
treatment of (Z,Z)-1,4-bis(tert-butylthio)-1,3-butadiene (2)
with 4 equiv of superbase (1:1 t-BuOK/n-BuLi in THF)
at -110 °C followed by slow addition of I2/THF (Scheme
1). Wiemers and Burton5a reported the preparation of
solutions of CF3Cu and its thermal transformation to
CF3CF2Cu, which can be monitored by 19F NMR spec-
troscopy. Using this methodology, either reagent can be
prepared at will and coupled with 3. In this way, (Z,Z)-
1,4-bis(tert-butylthio)-1,4-bis(trifluoromethyl)-1,3-butadi-
ene (4a ) and (Z,Z)-1,4-bis(tert-butylthio)-1,4-bis(pentaflu-
oroethyl)-1,3-butadiene (4b) were formed from in situ
generated CF3Cu and CF3CF2Cu in 51 and 61% yield,
respectively, after column chromatography. Oxidative
deprotection and cyclization of (Z,Z)-1,4-disubstituted-
1,4-bis(tert-butylthio)-1,3-butadienes to the corresponding
3,6-disubstituted-1,2-dithiins has been reported.4 Using
this methodology, treatment of 4a and 4b with N-
bromosuccinimide (2 equiv) in CH3CN afforded 3,6-bis-
(trifluoromethyl)- and 3,6-bis(pentafluoroethyl)-1,2-dithi-
in (1a and 1b) as orange, volatile liquids in 68 and 65%
yield, respectively, after column chromatography.
1
1a .1 Weak H NMR signals (δ 7.20, 7.04 ppm (dd)) were
also seen corresponding to 2,4-bis(trifluoromethyl)thio-
phene, from photoisomerism of 5a .5c Irradiation of 1b
gave similar results, although a much longer irradiation
time was required to afford 2,5-bis(pentafluoroethyl)-
thiophene (5b).
Attempts to prepare 3,6-difluoro-1,2-dithiin (6) were
unsuccessful. Treatment of 2 with superbase at -110 °C
followed by addition of N-fluorobenzenesulfonimide5d
gave (Z,Z)-1,4-bis(tert-butylthio)-1,4-difluoro-1,3-butadi-
ene (7a ) in 41% yield (Scheme 1). However all attempts
to convert 7a into 6 (NCS, NBS, NIS, I2, or SmI2) led
only to decomposition, with formation of a black precipi-
tate. In another attempt to prepare 6, (E,E)-1,4-bis-
(benzylthio)-1,4-diiodo-1,3-butadiene (8)1 was treated in
THF at -110 °C sequentially with tert-butyllithium
followed by N-fluorobenzenesulfonimide, giving (Z,Z)-1,4-
bis(benzylthio)-1,4-difluoro-1,3-butadiene (7b). Reductive
debenzylation of (7b) with lithium 1-(N,N-dimethylami-
no)naphthalenide (LDMAN) at -45 °C followed by quench-
ing with acetyl chloride gave (Z,Z)-1,4-difluoro-1,4-bis-
(acetylthio)-1,3-butadiene (7c). Unfortunately, oxidation
(I2) of the hydrolysis product of 7c failed to give 6.
P h otoelectr on Sp ectr oscop y a n d Th eor etica l Ca l-
cu la tion s. The analytical representations of the photo-
electron spectroscopic data are shown in Table 1 with
those for 1,2-dithiin (1c) and 3,6-dimethyl-1,2-dithiin (1d )
Compound 1a shows a single 19F NMR peak at δ -67.3
ppm while 1b shows two peaks at δ -84.2 and -111.3
ppm. Compounds 1a and 1b show UV λmax at 448 and
452 nm, respectively, which is a bathochromic shift
compared to dimethyl derivative 1d (UV λmax at 422 nm).
Furthermore, the bathochromic shift observed on going
from 1a to 1b is opposite to the trend seen on going from
J . Org. Chem, Vol. 68, No. 21, 2003 8111

Lorance et al.
TABLE 2. Ca lcu la ted Geom etr ic P a r a m eter s a n d
for comparison. The spectra themselves are in the Sup-
porting Information. The electron-withdrawing perfluo-
roalkyl groups raise the ionization energies of the four
highest occupied molecular orbitals. The effects of the
trifluoromethyl and pentafluoroethyl groups on the high-
est occupied MO are almost the same, resulting in nearly
identical lowest ionization energies for 1a and 1b. This
result is consistent with the similar inductive substituent
parameters for the two groups.6-8 The lowest-energy
valence ionizations of 1,2-dithiins are associated with
orbitals derived from the butadiene and disulfide frag-
ments.2 The composition of the MOs can be derived by
the changes in relative intensity of the ionizations
associated with them because carbon-based orbitals have
a higher cross-section than sulfur-based orbitals on going
from lower energy monochromatic UV irradiation (He I)
to higher energy (He II).3 Because of the apparent
increase of Gaussian 3 relative to Gaussians 1 and 2 in
the photoelectron spectrum of 1c on going from He I to
He II as the ionization source, the third highest MO was
deduced to have more carbon character than the highest
and second highest MOs. The same pattern is seen with
both perfluoroalkyl compounds. This is important be-
cause in 3,6-dimethyl-1,2-dithiin, Gaussian 2 increases
in relative area compared with Gaussians 1 and 3 on
going from He I to He II as the ionization source. Since
the primary cause of the destabilization of ionization of
3,6-dimethyl-1,2-dithiin is hyperconjugative interaction
between the methyl E-symmetry C-H σ-bond orbitals
and π-orbitals of the 1,2-dithiin ring, this shows the
larger orbital contribution at C3 and C6 for the MOs
corresponding to ionizations 1 and 3 than that for
ionization 2 (which has a high degree of sulfur character).
The observed results with the perfluoroalkyl substitu-
ents, in which this effect is not seen, supports the
previously proposed suggestions.
Ab initio calculations using MP2/6-31+G* showed that
1a adopts a twist geometry after energy minimization
which is similar to that of 1c. The geometric parameters
for 1a are given in Table 2. These parameters are very
similar to those of 1c. As examples, the twist in 1c9
and a is revealed by the CSSC and CdC-CdC torsion
angles and they are 53.9°, 29.0° and 57.4°, 29.7°, respec-
tively. The calculated ionization potential for 1a (8.91 eV)
is in reasonable agreement with the experimental value
(9.10 eV).
Electr och em istr y. The electrochemistry of 1a ,b was
studied in CH3CN using the technique of cyclic voltam-
metry. Satisfactory results were obtained with a glassy
carbon electrode but not with a Pt electrode. As with 1c,
a reversible one-electron oxidation followed by an ir-
reversible oxidation at more positive potentials was ob-
served for 1a and b. The peak potential for the reversible
Ion iza tion P oten tia l for
3,6-Bis(tr iflu or om eth yl)-1,2-d ith iin a (1a )
parameter
S-S bond, Å
2.07
1.77
1.36
S-C bond, Å
CdC bond, Å
C-C bond (in ring), Å
C-C bond (CF3), Å
C-F bond, Å
1.45
1.49
1.36
96.5
SSC angle, deg
SCC angle, deg
CCC angle, deg
CCC angle, (CF3), deg
CCF angle, deg
CSSC torsion, deg
SSCC torsion (in ring), deg
SSCC torsion (CF3), deg
SCCC torsion, deg
CdC-CdC torsion, deg
CCCC torsion (CF3), deg
ionization potential, eV
122.1
123.1
122.0
110.4
57.4
-45.0
138.5
1.4
29.7
177.7
8.91 (9.10)b
a
b
Calculations were done using MP2/6-31+G*. Experimental
value.
oxidation was 1.25 and 1.48 V for 1a and b, respectively,
and 1.40 and 1.67 V, respectively, for the irreversible
oxidation. All of the peak potentials were measured at a
scan rate of 100 mV/s versus a Ag/0.1 M AgNO3 in
CH3CN reference electrode. As expected, it is more
difficult to oxidize 1a and 1b than 1c, whose oxidation
potentials are 0.65 and 1.21 V,1 because the lowest
ionization potentials for 1a ,b are almost 1 eV higher than
that for 1c. On this basis, one would have expected 1a ,b
to have similar oxidation potentials because their ioniza-
tion potentials are almost the same. However, the oxida-
tion potential of 1b is over 200 mV more positive than
that of 1a . Similar discrepancies were reported before and
explained in the following way. The chemical step in the
EC mechanism for oxidation of 1,2-dichalcogenins was
ascribed to a change in geometry from the twist confor-
mation to a planar or planarized radical cation. Because
the time scales for electrochemical and photoelectron
spectroscopic measurements are different, there is suf-
ficient time for a change in geometry in the electrochemi-
cal experiment but not in the adiabatic photoelectron
spectroscopic measurement. Therefore, ionization poten-
tials were not found to correlate with oxidation potentials
in cases where the oxidized and reduced species possess
radically different geometries. This is well-illustrated in
the comparison of 1a and 1b. These results may also
provide clearer insight into the geometric effect because
the substituent effect of perfluoroalkyl groups is pre-
dominantly inductive in contrast to alkyl groups whose
effects are due to hyperconjugation as well as induction.
Since inductive effects are geometry independent, they
are the same in twist 1a ,b as in planar 1a ,b radical
cations. The electrochemical results show that [1b]•+ is
less stable relative to 1b than is [1a ]•+ relative to 1a . The
relative destabilization of [1b]•+ is due to greater steric
bulk of the C2F5 group relative to the CF3 group which
destabilizes the planar form.10,11
(5) (a) Wiemers, D. M.; Burton, D. J . J . Am. Chem. Soc. 1986, 108,
832-834. (b) Nishida, M.; Fujii, S.; Aoki, T.; Hayakawa, Y. J . Fluorine
Chem. 1990, 46, 445-459. (c) Kobayashi, Y.; Kumadaki, I.; Ohsawa,
A.; Sekine Y. Tetrahedron Lett. 1974, 15, 2841-2844. (d) Davis, F. A.;
Han, W.; Murphy, C. K. J . Org. Chem. 1995, 60, 4730-4737.
(6) Allen, A. D.; Tidwell, T. T. Adv. Carbocation Chem. 1989, 1,
1-44.
(7) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165-195.
(8) Thompson, M.; Hewitt, P. A.; Wooliscroff, D. S. In Handbook of
X-ray and Ultraviolet Photoelectron Spectroscopy; Briggs, D., Ed.;
Heyden: London, 1977; pp 341-379.
(9) Gillies, J . Z.; Gillies, C. W.; Cotter, E. A.; Block, E.; DeOrazio,
R. J . Mol. Spectrosc. 1996, 180, 139-144.
Con clu sion s
Our studies with 1a ,b demonstrate the effects of
electron-withdrawing groups. They further validate the
8112 J . Org. Chem., Vol. 68, No. 21, 2003

3,6-Bis(perfluoroalkyl)-1,2-dithiins
J ) 279, 43 Hz), 117.0 (tq, J ) 276, 38 Hz), 132.5 (t, J )
2.8 Hz), 144.8 (s); 19F NMR (CDCl3/CFCl3) δ -78.9, -107.9;
GC-MS m/z 466 (M+), 409, 353, 320, 282, 251, 233, 182, 165,
146, 114, 57; IR (CCl4) ν 2965, 2925, 1367, 1322, 1209, 1190,
1158, 1060, 1050, 985, 923, 878 cm-1. Anal. Calcd for
key conclusions of previous work, namely that the ener-
gies of the second and third highest occupied molecular
orbitals are inverted in 3,6-dimethyl-1,2-dithiin (1d )
compared with 1c. In addition, the studies demonstrate
that ionization potentials and electrochemical oxidation
potentials for 1,2-dithiins do not correlate because of
geometry changes in electrochemical but not in spectro-
scopic measurements.
C
16H20F10S2: C, 41.20; H, 4.32. Found: C, 41.16; H, 4.31.
3,6-Bis(tr iflu or om eth yl)-1,2-d ith iin (1a ). This procedure
was conducted under red light. In a round-bottomed flask,
4a (123 mg, 0.33 mmol) was dissolved in CH2Cl2 (3 mL) and
CH3CN (3 mL). A solution of NBS (60 mg, 0.33 mmol) dissolved
in CH3CN (3 mL) and CH2Cl2 (3 mL) was then added dropwise
at room temperature. The resulting mixture was stirred for 1
h and then diluted with pentane (10 mL), washed sequentially
with water and brine, dried (Na2SO4), and concentrated in
vacuo. Chromatography (pentane) afforded 1a (22 mg, 68%
based on recovered 4a : 39 mg) as a volatile red-orange liquid
not easily freed, when working on a small scale, from pen-
tane: 1H NMR (pentane/(CH3)3CBr/CDCl3) δ 6.96 (s); 13C NMR
(pentane/CDCl3) δ 120.2 (q, J ) 280 Hz), 130.1 (q, J ) 6.0 Hz),
131.9 (q, J ) 6.0 Hz); 19F NMR (pentane/CDCl3/CFCl3) δ -67.3;
UV (CHCl3) λmax (ꢀ) 448 (890), 264 (1330), 242 (1589) nm;
GC-MS m/z 252 (M+), 233, 232, 213, 183, 139, 114, 69; HRMS
m/z 251.9489 (calcd for C6H2F6S2 251.9503). Repetition of this
synthesis using 30 mg of 4a and 2 equiv of NBS gave a higher
yield of 1a with little recovered 4a . A pentane solution of 1a
was used for measurement of photoelectron (PE) spectra; in
the PE spectrometer, the pentane was pumped off first at low
temperatures (see below).
Exp er im en ta l Section
SAF ETY WARNING: HMPA is highly toxic and a cancer
suspect agent.
Tr iflu or om eth ylca d m iu m .12 A three-necked round-bot-
tomed flask equipped with magnetic stirrer, thermometer, and
dry ice condenser under Ar was charged with anhydrous DMF
(4 mL) and activated Cd (acid washed) (0.45 g, 4 mmol). Then
CF2Br2 (0.42 g, 2 mmol) was added slowly to the suspension,
giving an exothermal reaction with a dark brown solution and
precipitate. The temperature of the reaction was kept below
90 °C by controlling the rate of addition of CF2Br2. After
completion of the addition of CF2Br2, the reaction mixture was
stirred for 1.5 h at room temperature. The cadmium reagent
was filtered through a medium-frit Schlenk funnel under argon
into a three-necked flask. The precipitate was washed with
DMF. The resulting dark brown solution of CF3Cd (19F NMR
(DMF/CFCl3) δ -34.7 (lit.12 δ -35.7); 0.8-0.95 mmol, esti-
mated 80-95% yield) was utilized immediately in subsequent
reactions.
Exposure of 1a to visible light in an NMR tube led to
formation of sulfur and appearance of 1H NMR signals
corresponding to 2,5-bis(trifluoromethyl)thiophene (5a ; lit.5b
δ 7.48 ppm, found δ 7.43 ppm; GC-MS m/z 220) and minor
signals corresponding to 2,4-bis(trifluoromethyl)thiophene (δ
7.20, 7.04 (dd)).5c
(Z,Z)-1,4-Bis(ter t-b u t ylt h io)-1,4-b is(t r iflu or om et h yl)-
1,3-bu ta d ien e (4a ). Following the reported4 procedure, except
that the deprotonation and iodination was carried out at -110
°C instead of -78 °C, with slow addition of I2 in THF, (E,E)-
1,4-bis(tert-butylthio)-1,4-diiodo-1,3-butadiene (3) was pre-
pared in 47% yield from (Z,Z)-1,4-bis(tert-butylthio)-1,3-
3,6-Bis(p en ta flu or oeth yl)-1,2-d ith iin (1b). This proce-
dure was conducted under red light. A solution of NBS (0.36
g, 2.0 mmol) dissolved in CH3CN (5 mL) was added dropwise
at room temperature to a solution of 4b (0.47 g, 1.0 mmol)
dissolved in CH3CN (10 mL) in a round-bottomed flask. After
the mixture was stirred overnight, pentane (25 mL) was added
and the resulting mixture washed sequentially with water and
brine. The organic layer was dried (Na2SO4), concentrated in
vacuo, and chromatographed (pentane) to give 1b (179 mg,
65% based on recovered 4b: 163 mg) as a volatile red-orange
liquid: 1H NMR (pentane/CDCl3) δ 6.95 (s); 13C NMR (pentane/
CDCl3) δ 114.0 (tq, J ) 276, 36 Hz), 118.4 (tq, J ) 272, 36
Hz), 127.6 (t, J ) 18 Hz), 132.8 (t, J ) 7.5 Hz); 19F NMR
(pentane/CDCl3/CFCl3) δ -84.2, -111.3; UV (CHCl3) λmax (ꢀ)
452 (960), 290 (1580), 246 (1105) nm; GC-MS m/z 352 (M+),
283, 233, 214, 189, 164, 114, 69; HRMS m/z 351.9424 (calcd
for C8H2F10S2 351.9439).
butadiene (2). In
a three-necked flask equipped with a
medium-frit Schlenk funnel and argon inlet, CF3Cd (0.8-0.95
mmol) in anhydrous DMF (4 mL) was mixed with an equal
volume of anhydrous HMPA (4 mL) and cooled to -20 °C. CuI
(0.19 g, 2.0 mmol) was quickly added followed by the addition
of 3 all at once. The reaction mixture was stirred for 1.5 h
and then quenched with water. The aqueous solution was
extracted with Et2O, and the organic extracts were combined,
washed sequentially with brine and water, dried (Na2SO4), and
concentrated in vacuo. Chromatography (hexanes) gave 4a (74
mg, 51%) as a colorless oil: 1H NMR (CDCl3) δ 1.35 (s, 9H),
8.04 (s,2H); 13C NMR (CDCl3) δ 31.8 (s), 51.4 (s), 122.8 (q, J )
272 Hz), 132.0 (q, J ) 18.5 Hz), 141.6 (q, J ) 5.3 Hz); 19F NMR
(CDCl3/CFCl3) δ -64.4; IR (neat) ν 2968, 2926, 2901, 2867
1562, 1460, 1368, 1237, 1173, 1134, 931, 743, 658, 595 cm-1
;
GC-MS m/z 366 (M+), 309, 254, 57; HRMS m/z 366.0898 (calcd
for C14H20F6S2 366.0912).
P h ot oelect r on Sp ect r oscop y. Gas-phase photoelectron
spectra were recorded using an instrument that features a 36-
cm, 8-cm gap hemispherical analyzer (McPherson) and custom
designed sample cells, discharge source, and detection and
control electronics13 that have been described in more detail
elsewhere.14 The excitation source was a quartz lamp with the
ability, depending on operating conditions, to produce He I
(21.21 eV), or He II (40.8 eV) photons. For the He I and He II
experiments, the ionization energy scale was calibrated using
(Z,Z)-1,4-Bis(ter t-bu tylth io)-1,4-bis(p en ta flu or oeth yl)-
1,3-bu ta d ien e (4b). In a three-necked flask equipped with a
medium-frit Schlenk funnel and argon inlet was added
CF3Cd in DMF-HMPA (8 mL) prepared as described above.
To this solution was added CuI (0.38 g, 2.0 mmol) at room
temperature and the mixture stirred for 4 h. Compound 3 (0.19
g, 0.40 mmol) was added all at once, and the reaction mixture
was stirred overnight. Water was added to quench the reaction,
and the aqueous solution was extracted with hexanes. The
extracts were washed with brine, dried (Na2SO4), and concen-
trated in vacuo. Chromatography (hexanes) yielded 4b (66 mg,
61%) as a white solid: mp 70-71 °C; 1H NMR (CDCl3) δ 1.34
(s, 9H), 8.11 (s, 1H); 13C NMR (CDCl3) δ 31.8, 51.9, 114.4 (tq,
2
the E1/2 ionization of methyl iodide (9.538 eV), with the Ar
2P3/2 ionization (15.759 eV) used as an internal energy scale
lock during data collection. During He I and He II data
collection the instrument resolution, measured using the full-
2
width-at-half-maximum of the Ar P3/2 ionization, was 0.015-
0.025 eV. All data are intensity-corrected with an experimen-
tally determined instrument analyzer sensitivity function that
(10) Smart, B. E. In Organofluorine Chemistry, Banks, R. E., Smart,
B. E., Tatlow, J . C., Eds.; Plenum: New York, 1994; pp 57-88.
(11) Be´gue´, J .-P.; Benayoud, F.; Bonnet-Delpon, D.; Allen, A. D.; Cox,
R. A.; Tidwell, T. T. Gazz. Chim. Ital. 1995, 125, 399-402.
(12) Burton, D. J .; Wiemers, D. M. J . Am. Chem. Soc. 1985, 107, 7,
5014-5015.
(13) Lichtenberger, D. L.; Kellogg, G. E.; Kristofzski, J . G.; Page,
D.; Turner, S.; Klinger, G.; Lorenzen, J . Rev. Sci. Instrum. 1986, 57,
2366.
(14) Westcott, B. L.; Gruhn, N. E.; Enemark, J . H. J . Am. Chem.
Soc. 1988, 120, 3382-3386.
J . Org. Chem, Vol. 68, No. 21, 2003 8113

Lorance et al.
assumes a linear dependence of analyzer transmission (inten-
sity) to the kinetic energy of the electrons within the energy
range of these experiments.
All of the spectra were corrected for the presence of
ionizations from secondary photons in the discharge sources.15
The He I spectra were corrected for ionizations from He Iâ
photons (1.9 eV higher in energy, and 3% of the intensity of
the He IR photons), the He II spectra were corrected for the
He IIâ photons (7.568 eV higher in energy, and 12% the
intensity of the He IIR photons).
The compounds studied are light and temperature sensitive
and were handled at low temperature to limit self-polymeri-
zation. Liquid samples were frozen and loaded into a sample
cell that had been cooled in a -40 °C freezer. The cell was
then placed in the instrument and allowed to slowly warm,
and the spectra were collected at the temperature (monitored
using a “K”-type thermocouple passed through a vacuum
feedthrough and attached directly to the sample cell) where
sufficient sample vapor pressure was achieved.
The valence ionization bands are represented analytically
with the best fit of asymmetric Gaussian peaks.16 The bands
are defined with the position, amplitude, halfwidth for the high
binding energy side of the peak, and the halfwidth for the low
binding energy side of the peak. The peak positions and
halfwidths are reproducible to about (0.02 eV (∼3F level).
Confidence limits for the relative integrated peak areas are
about 5%, with the primary source of uncertainty being the
determination of the baseline, which is caused by electron
scattering and taken to be linear over the small energy range
of these spectra. The total area under a series of overlapping
peaks is known with the same confidence, but the individual
peak areas are more uncertain. The fitting procedures used
are described in more detail elsewhere.13
Cyclic Volta m m etr y. Voltammograms were measured on
degassed solutions approximately 10-3 M in substrate and 0.1
M in tetra-n-butylammonium hexafluorophosphate, which
served as supporting electrolyte, in CH3CN as solvent versus
a Ag/0.10 M AgNO3 in CH3CN reference electrode. A 3 mm
diameter glassy planar carbon electrode served as the working
electrode, and the scan rate was 100 mV/s. A 0.5 mm diameter
platinum wire was used as the counter electrode. All electro-
chemical experiments were performed under rigorously anaer-
obic conditions in a Vacuum Atmospheres model HE-493
drybox or sealed electrochemical cell under red light. A Cypress
Systems electrochemical data acquisition system model CYSY-1
was used to acquire and process the data.
Ack n ow led gm en t. We gratefully acknowledge sup-
port of this work by the National Science Foundation
(R.S.G., CHE-9422200; E.B., CHE-9906566; E.D.L., for
a Graduate Fellowship), the Petroleum Research Fund,
administered by the American Chemical Society (E.B.),
and Dr. Nadine E. Gruhn at the Center for Gas-Phase
Electron Spectroscopy, The University of Arizona, for
helpful discussions and measuring the photoelectron
spectra.
Su p p or tin g In for m a tion Ava ila ble: General experimen-
tal and procedures for the synthesis of 7a -c and attempted
preparation of 6; table of atomic coordinates for optimized
geometry of 1a calculated by using MP2/6-31+G*; photoelec-
1
tron spectra of 1a and 1b; H NMR spectra of 1a ,b and 4a ;
13C NMR spectra of 1a and 4a . This material is available free
of charge via the Internet at http://pubs.acs.org.
J O034683N
Com p u ta tion a l Meth od ology. Computations were carried
out on an SGI Origin 2000 32-processor mainframe using a
single processor under the IRIX64 v6.4 operating system using
Gaussian 94 (rev. E.2).17 Calculations on unsubstituted 1,2-
dithiin were performed using the Hartree-Fock method, with
second-order Møller-Plesset perturbation corrections. The
6-31G basis set with heavy atom polarization functions (*) and
heavy atoms diffuse functions (+) was used.
The neutral species were geometry optimized and single-
point calculations were made for the cation radical at the
neutral species optimized geometry to obtain ionization po-
tentials (∆SCF).
(15) Turner, D. W.; Baker, C.; Baker, A. D.; Brundle, C. R. Molecular
Photoelectron Spectroscopy; Wiley-Interscience: London, 1970.
(16) Lichtenberger, D. L.; Copenhaver, A. J . J . Electron Spectrosc.
Relat. Phenom. 1990, 50, 335-352.
(17) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
J ohnson, B. G.; Robb, M. A.; Cheeseman, J . R.; Keith, T.; Petersson,
G. A.; Montgomery, J . A.; Raghavachari, K.; Al-Laham, M. A.;
Zakrewski, V. G.; Ortiz, J . V.; Foresman, J . B.; Cioslowski, B. B.;
Stefanov, A.; Nanayakkara, M.; Challacombe, M.; Peng, C. Y.; Ayala,
P. Y.; Chen, W.; Wong, M. W.; Andres, J . L.; Replogle, E. S.; Gomperts,
R.; Martin, R. L.; Fox, D. J .; Binkley, J . S.; Defrees, D. J .; Baker, J .;
Stewart, J . P.; Head-Gordon, M.; Gonzalez, C.; Pople, J . A. Gaussian
94, revision E.2.; Gaussian Inc.: Pittsburgh, PA, 1995.
8114 J . Org. Chem., Vol. 68, No. 21, 2003