A Theoretical and Structural Investigation of Thiocarbon Anions

Article abstract of DOI:10.1021/jo035144l

Density functional theory energies, geometries, and population analyses as well as nucleus-independent chemical shifts (NICS) have been used to investigate the structural and magnetic evidence for cyclic C_nS _n²- and C_nS_n (n = 3-6) electron delocalization. Localized molecular orbital contributions to NICS, computed by the individual gauge for localized orbitals method, dissect π effects from the σ single bonds and lone pair influences. C_nS _n^2- (n = 3-5) structures in D_nh symmetry are minima. Their aromaticity decreases with increasing ring size. C ₃S₃^2- is both σ and π aromatic, while C₄S₄^2- and C₅S₅ ^2- are much less aromatic. NICS(0)π, the C-C(π) contribution to NICS(0) (i.e., at the ring center), decreases gradually with ring size. In contrast, cyclic C₆S₆^2- prefers D_2h symmetry due to the balance between aromaticity, strain energy, and the S-S bond energies and is as aromatic as benzene. The theoretical prediction that C₆S₆^6- has D_6h minima was confirmed by X-ray structure analysis. Comparisons between thiocarbons and oxocarbons based on dissected NICS analysis show that C_nS_n ^2- (n = 3-5) and C₆S₆^6- are less aromatic in D_nh symmetry than their oxocarbon analogues.

Full text of DOI:10.1021/jo035144l

A Th eor etica l a n d Str u ctu r a l In vestiga tion of Th ioca r bon An ion s
Zhongfang Chen,*,†,‡ Liam R. Sutton,‡ Damian Moran,†,‡ Andreas Hirsch,*,‡ Walter Thiel,§ and
Paul von Rague´ Schleyer*,†,‡
Computational Chemistry Annex, The University of Georgia, Athens, Georgia 30602-2525, Institut fu¨r
Organische Chemie, Universita¨t Erlangen-Nu¨rnberg, Henkestrasse 42, 91054 Erlangen, Germany, and
Max-Planck-Institut fu¨r Kohlenforschung, 45466 Mu¨lheim an der Ruhr, Germany
zhongfang.chen@chemie.uni-erlangen.de; Hirsch@organik.uni-erlangen.de; schleyer@chem.uga.edu
Received August 3, 2003
Density functional theory energies, geometries, and population analyses as well as nucleus-
independent chemical shifts (NICS) have been used to investigate the structural and magnetic
2-
evidence for cyclic CnSn and CnSn (n ) 3-6) electron delocalization. Localized molecular orbital
contributions to NICS, computed by the individual gauge for localized orbitals method, dissect π
effects from the σ single bonds and lone pair influences. CnSn2- (n ) 3-5) structures in Dnh symmetry
are minima. Their aromaticity decreases with increasing ring size. C3S32- is both σ and π aromatic,
2-
2-
while C4S4 and C5S5 are much less aromatic. NICS(0)π, the C-C(π) contribution to NICS(0)
2-
(i.e., at the ring center), decreases gradually with ring size. In contrast, cyclic C6S6 prefers D2h
symmetry due to the balance between aromaticity, strain energy, and the S-S bond energies and
6-
is as aromatic as benzene. The theoretical prediction that C6S6 has D6h minima was confirmed
by X-ray structure analysis. Comparisons between thiocarbons and oxocarbons based on dissected
2-
6-
NICS analysis show that CnSn (n ) 3-5) and C6S6 are less aromatic in Dnh symmetry than
their oxocarbon analogues.
In tr od u ction
Since Fangha¨nel’s smooth and simple synthesis (Scheme
6-
1),6 C6S6 hexaanion (1) has been used as an electro-
active dendrimer core unit7 and as a component of a
variety of electron-rich organic charge-transfer salts 8 and
transition metal complexes9 that sometimes exhibit redox
communication between metal centers.10
Density-functional computations were used here to
investigate the electron delocalization in CnSn2- dianions
(n ) 3-6), in their neutral counterparts, and in the more
highly reduced C6S6 species. Thiocarbons and oxocarbons
are compared. The X-ray crystal structure of the C6S6 core
unit in hydrated Na6[C6S6] was determined.
Thiocarbon anions are the heavier analogues of the
well-investigated nonbenzenoid aromatic oxocarbons.1
The continuing interest in their structure and bonding
is due partly to the aromaticity2 of the monocyclic
oxocarbon anions CnOn2-, which are stabilized by the
delocalization of the π electrons.3 There has also been
significant interest in thiocarbons and their applications.
However, the aromaticity of thiocarbons has not been
2-
2-
studied. The thiocarbon anions C3S3 and C4S4 have
been synthesized,4 and C4S42-, tetrathiosquarate, is a
well-known and versatile ligand in metal complexes.4b,5
2-
2-
We found no literature pertaining to C5S5 and C6S6
.
Resu lts a n d Discu ssion
Th ioca r bon Dia n ion s Cn Sn 2- (n ) 3-5). Optimized
thiocarbon dianion geometries are shown in Figure 1,
while computed total energies and zero point energies
(ZPE) are summarized in Table 1. As for their oxygen
† University of Georgia.
‡ Universita¨t Erlangen-Nu¨rnberg.
§ Max-Planck-Institut fu¨r Kohlenforschung.
(1) (a) Oxocarbons; West, R., Ed.; Academic Press: New York, 1980.
(b) Serratosa, F. Acc. Chem. Res. 1983, 16, 170. (c) Seitz, G.; Imming,
P. Chem. Rev. 1992, 92, 1227.
2-
analogues,3b CnSn dianions (n ) 3-5) favor Dnh sym-
(2) See reviews: (a) Aromaticity and Antiaromaticity. Electronic and
Structural Aspects; Minkin, V. I., Glukhovtsev, M. N., Simkin, B. Y.,
Eds.; J ohn Wiley & Sons: New York, 1994. (b) Schleyer, P. v. R.; J iao,
H. Pure Appl. Chem. 1996, 68, 209. (c) Lloyd, D. J . Chem. Inf. Comput.
Sci. 1996, 36, 442. (d) Krygowski, T. M.; Cyranski, M. K.; Czarnocki,
Z.; Hafelinger, G.; Katritzky, A. R. Tetrahedron 2000, 56, 1783. (e)
Schleyer, P. v. R. (guest editor) Chem. Rev. Special issue on aromaticity,
May, 2001.
(3) (a) West, R.; Niu, N.-Y.; Powell, D. L.; Evans, M. V. J . Am. Chem.
Soc. 1960, 82, 6204. (b) Schleyer, P. v. R.; Najafian, K.; Kiran, B.; J iao,
H. J . Org. Chem. 2000, 65, 426. (c) Quinonero, D.; Frontera, A.;
Ballester, P.; Deya, P. M. Tetrahedron Lett. 2000, 41, 2001. (d)
Quinonero, D.; Garau, C.; Frontera, A.; Ballester, P.; Costa, A.; Deya,
P. M. Chem.-Eur. J . 2002, 8, 433.
(4) (a) Allmann, R.; Debaerdemaeker, T.; Mann, K.; Matusch, R.;
Schmiedel, R.; Seitz, G. Chem. Ber. 1976, 109, 2208. (b) Seitz, G.
Phosphorus, Sulfur Silicon Relat. Elem. 1989, 43, 311.
(5) (a) Go¨tzfried, F.; Grenz, R.; Urban, G.; Beck, W. Chem. Ber. 1985,
118, 4179. (b) Grenz, R.; Go¨tzfried, F.; Nagel, U.; Beck, W. Chem. Ber.
1986, 119, 1217.
(6) Richter, A. M.; Engels, V.; Beye, N.; Fangha¨nel, E. Z. Chem.
1989, 29, 444.
(7) Inoue, S.; Nishiguchi, S.; Murakami, S.; Aso, Y.; Otsubo, T.; Vill,
V.; Mori, A.; Ujiie, S. J . Chem. Res. (S) 1999, 596.
(8) (a) Frangha¨nel, E.; Alsleben, I.; Gebler, B.; Herrmann, A.;
Herrmann, R.; Palmer, T.; Strunk, K.; Ullrich, A.; Luders, K.; Mahd-
jour, H. Phosphorus Sulfur Silicon 1997, 120, 121. (b) Frangha¨nel, E.;
Herrmann, R.; Naarmann, H. Tetrahedron 1995, 51, 2533.
(9) (a) Yip, H. K.; Schier, A.; Riede, J .; Schmidbaur, H. J . Chem.
Soc., Dalton Trans. 1994, 2333. (b) Harnisch, J . A.; Angelici, R. J . Inorg.
Chim. Acta 2000, 300-302, 273.
(10) Nishihara, M.; Okuno, M.; Akimoto, N.; Kogawa, N.; Aramaki,
K. J . Chem. Soc., Dalton Trans. 1998, 2651.
10.1021/jo035144l CCC: $25.00 © 2003 American Chemical Society
Published on Web 10/15/2003
8808
J . Org. Chem. 2003, 68, 8808-8814

Investigation of Thiocarbon Anions
F IGURE 1. B3LYP/6-311+G* bond lengths (in Å) of CnSn2- (n ) 3-6) compared with the X-ray values (in parentheses), Wiberg
bond indices (WBI, in italic), and natural charges (q, underlined).
SCHEME 1. Syn th esis of Na 6 1: (i) Na SCH2P h (6 equ iv), DMF ; (ii) Na , NH3 (liqu id )
2-
metry (Figure 1). The computed bond lengths agree well
short C-C bond lengths in the strained C3S3 suggest
aromatic stabilization, as is found in cyclopropane.2a,12
The C-S WBI, which are 1.592 (C3S32-), 1.480 (C4S42-),
and 1.532 (C5S52-), also indicate delocalization in the
2- 11
with the available experimental data for C3S3
and
2- 4a
2-
C4S4
.
C3S3 has the shortest C-C (1.403 Å) and the
longest C-S (1.710 Å) bond length of the set, and the
C-C bonds are even shorter than those in C3O32- (1.433
carbon rings. Both the carbon and sulfur atoms in CnSn
2-
Å at the same level).3b The calculated CC Wiberg bond
have the negative charges.
2-
indices (WBI) are 1.108, 1.095, and 1.114 for C3S3
,
Among the classical criteria for aromaticity (i.e., struc-
tural, energetic (stability), reactivity, and magnetic),
nucleus-independent chemical shift (NICS), a magnetic
C4S42-, and C5S52-, respectively, and indicate that the
C-C bonds have very little double bond character.
However, compared with its higher analogues, the very
(12) (a) Dewar, M. J . S. J . Am. Chem. Soc., 1984, 106, 669. (b)
Cremer, D.; Gauss, J . J . Am. Chem. Soc., 1986, 108, 7467. (c) Exner,
K.; Schleyer, P. v. R. J . Phys. Chem. A 2001, 105, 3407.
(11) Baum, G.; Kaiser, F. J .; Massa, W.; Seitz, G. Angew. Chem.
1987, 99, 1199; Angew. Chem., Int. Ed. Engl. 1987, 26, 1163.
J . Org. Chem, Vol. 68, No. 23, 2003 8809

Chen et al.
2-
2-
TABLE 1. Cn Sn (n ) 3-6) B3LYP /6-311+G* Zer o-P oin t
aromaticity decreases with increasing CnSn ring size
of n ) 3-5, the same as for the CnOn2- series.3b The NICS
results above correlate well with the thiolate dianion
structures.
En er gies (ZP E, k ca l/m ol), Nu m ber of Im a gin a r y
F r equ en cies (NIm a g), Tota l En er gies (a u ), a n d Rela tive
En er gies (k ca l/m ol)
E
rel (Etot +
ZPE)
Total NICS(1) values indicate the strong diatropic ring
molecule symmetry ZPE NImag
Etot
current for C3S32- but weak diatropic currents for C4S4
2-
2-
and C5S52-. Generally, the degree decreases with increas-
ing ring size. Dissection shows that NICS(1)σ is negative
C3S3
C4S4
C5S5
C6S6
C6S6
C6S6
C6S6
D3h
D4h
D5h
D6h
D3d
D2h
11.5
16.6
21.4
25.7
25.4
26.1
25.3
0
0
-1308.96902
-1745.34100
-2181.69003
a
-2617.98951
-2617.99496
-2617.99606
2-
2-
2-
2-
2-
2-
0
(-7.0 ppm) for C3S32-, confirming its σ aromaticity,
3a
0
2- 3b
analogous to C3O3
.
The σ contributions in C4S42- and
2.7
0.0
-1.5
C5S52- are very small, even negligible, but the π contribu-
0
0
b
2-
C2v
tions are of a similar magnitude for CnSn (n ) 3-5).
Th ioca r bon Dia n ion s C6S62-. The lowest-energy
a
Geometry optimization at B3LYP/6-311+G* failed to converge,
ZPE given at B3LYP/6-31G(d). Bicyclic structure, see Figure 1.
b
2-
isomer of cyclic C6S6 has D2h symmetry, the same as
2-
its C6S6 neutral analogue.15 In D6h symmetry, C6S6 is
a third-order saddle point (NImag ) 3); following the first
imaginary mode led to a D3d minimum, 2.7 kcal/mol less
criterion, has proven to be a simple and efficient aroma-
ticity probe.13 Thus, it was found that “Oxocarbon acids
and their anions are examples where the criteria of
aromaticity that use reference systems are unsuccessful,
only the NICS criterion gives satisfactory results.”3d
Hence, NICS both at the ring centers, NICS(0), and at 1
Å above ring centers, NICS(1), were computed to analyze
the degree of aromaticity (Table 2). Here “dissected
NICS”, based on Kutzelnigg’s individual gauge for local-
ized orbitals (IGLO) method, which reveals the individual
contributions to the total shielding of localized orbitals
associated with bonds, lone pairs, and core electrons, is
employed.
stable than the D2h C6S62-. A C2 C6S62- isomer, which is
2- 3b
the most favorable structure for C6O6
function instability.
,
has wave-
As for C6S6,15,16 the stability of C6S6 is determined
by the balance between aromatic stabilization, ring
strain, and C-S and S-S bond strength affects. The most
stable cyclic C6S62- isomer, like C6S6, has D2h symmetry
and a strongly delocalized hexagonal carbon ring with
C-C bond lengths of approximately the benzene value,
1.40 Å. Hence, the aromatic stabilization adds into the
already favorable stability balance of the neutral ana-
logue. In contrast with the nonplanar D3d symmetric
2-
2-
CnSn (n ) 3-5) dissected NICS and 13C shifts are
2-
minima, the D2h C6S6 isomer has approximately equi-
summarized in Table 2. The computed CnSn2- carbon-13
chemical shift values agree well with the available
distant C-C bond lengths and meets the geometric
criteria of aromaticity (also see below).
2- 11
2- 4a
NICS(0)π analysis shows that the D2h hexathiolate
diaanion, the lowest energy cyclic isomer of C6S62-, is
much more aromatic than its D3d isomer, with a signifi-
cant shielding π-contribution (-22.4 ppm) , similar to
experimental values for C3S3
and C4S4
.
The σ
bonds contribute paramagnetically to NICS(0) for all
thiocarbon dianions. The C3S32- dianion C-C(σ) deshield-
ing contribution (6.0 ppm) is exceptionally small com-
2-
+
those in C3S3 (-24.9 ppm), while NICS(0)σ values are
pared to the corresponding value in C3H3 (10.2 ppm),
similar, 7.8 and 9.5 ppm, respectively, for D3d and D2h
isomers.
in both species carbon atoms are negatively charged,
while the C-C(σ) contribution in its oxygen analogue
2-
+
The σ contributions to NICS(1) are negligible for both
D2h and D3d isomers. However, π contributions are still
significant. For C6S62-(D3d), the NICS(1)π value is sim-
C3O3 and C3F3 are shielding (-11.7 and -9.3 ppm,
respectively). This may indicate a hidden σ aromaticity
2-
in C3S3 (see below). The relatively small deltathiolate
2-
ilar to those of CnSn (n ) 3-5) (-5.9 vs -5.6 to -6.6
dianion NICS(0)σ may be related to the negative charges
ppm), while C6S62-(D2h) exhibits the largest π contribu-
tions (-11.0 ppm) in the thiocarbon dianions investi-
gated.
2-
2-
+
on the C3S3 carbon atoms, since, in C3O3 and C3F3
,
the carbon atoms are positively charged.3b The four-
membered ring in C4S4 is σ antiaromatic, with abnor-
2-
2-
In general, substantial ring currents exist for CnSn
mally pronounced C-C(σ) contribution (30.4 ppm) com-
pared with its analogues, and much more so than in C4H8
(15.2 ppm).14
NICS(0)π values decrease with increasing CnSn2- (n )
3-5) ring size but are sufficiently diamagnetic that the
dianions are classified as π aromatic. For example, the
most pronounced C-C(π) shielding in C3S32- (-24.9 ppm)
is even stronger than that of benzene (-20.7 ppm)13c and
just a little less than that of C3O32- (-27.9 ppm).3b Thus,
(n ) 3-6), all thiocarbon dianions investigated, due to
the significant π contribution to both NICS(0)π and
NICS(1)π.
Neu tr a l Th ioca r bon s CnSn (n ) 3-6). For compari-
son, Dnh symmetric CnSn (n ) 3-6) thiocarbons, the
2-
neutral analogues of CnSn dianions, have been com-
puted. However, the wave functions of all the singlet
states are not stable at B3LYP/6-311+G*. While the wave
functions of the triplet states are stable for CnSn (n )
3-5), both singlet and triplet D6h C6S6 states have wave-
function instabilities. Moreover, the triplet states of C3S3
(D3h) and C4S4 (D4h) are local minimum while the C5S5
(13) (a) Schleyer, P. v. R.; Maerker, C.; Dransfeld, A.; J iao, H.;
Hommes, N. J . R. v. E. J . Am. Chem. Soc. 1996, 118, 6317. (b) Schleyer,
P. v. R.; J iao, H.; Hommes, N. J . R. v. E.; Malkin, V. G.; Malkina, O.
L. J . Am. Chem. Soc. 1997, 119, 12669-12670. (c) Schleyer, P. v. R.;
Manoharan, M.; Wang, Z.-X.; Kiran, B.; J iao, H.; Puchta, R.; Hommes,
N. J . R. v. E. Org. Lett. 2001, 3, 2465. (d) Cyran˜ski, M. K.; Krygowski,
T. M.; Katritzky, A. R.; Schleyer, P. v. R. J . Org.Chem. 2002, 67, 1333.
(e) Patchkovskii, S.; Thiel, W. J . Mol. Model. 2000, 6, 67.
(14) Moran, D.; Manoharan, M.; Heine, T.; Schleyer, P. v. R. Org.
Lett. 2003, 5, 23.
(15) (a) Frenking, G. Angew. Chem. 1990, 102, 1516; Angew. Chem.,
Int. Ed. Engl. 1990, 29, 1410. (b) Maier, G.; Schrot, J .; Reisenauer, H.
P.; Frenking, G.; J onas, V. Chem. Ber. 1992, 125, 265.
(16) Su¨lzle, D.; Beye, N.; Fanghaenel, E.; Schwarz, H. Chem. Ber.
1989, 122, 2411.
8810 J . Org. Chem., Vol. 68, No. 23, 2003

Investigation of Thiocarbon Anions
2-
TABLE 2. Cn Sn (n ) 3-6), C6O6Lim a n d C6S6Lim (m ) 4, 6) NICS(tot), NICS(σ), NICS(π), NICS(C-S or C-O), a n d Lon e
P a ir (LP ) Con tr ibu tion s a t, a n d 1.0 Å a bove, Rin g Cen ter sa
NICS(x)
NICS(tot)
C-C(σ)
C-C(π)
C-S(O)
LP (S/O)
core
13C NMR
2-
2-
2-
2-
2-
C3S3
C4S4
C5S5
C6S6
C6S6
D3h
D4h
D5h
D3d
D2h
D2h
D6h
C2h
D6h
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
-20.9
-10.3
15.5
-3.1
15.4
-0.9
1.0
-4.8
-5.9
-5.3
13.3
6.0
-7.0
30.4
-1.2
23.6
1.0
7.8
-1.2
9.5
1.1
9.4
1.5
-4.6
-2.3
11.0
2.0
10.9
4.1
-24.9
-5.6
-19.3
-6.6
-13.9
-6.4
-9.8
-5.9
-22.4
-11.0
-0.2
6.7
-18.7
-11.3
-0.8
4.2
-17.2
-8.8
0.0
4.2
5.6
6.8
6.0
5.0
5.8
5.1
7.1
4.5
3.4
1.6
10.2
3.5
3.0
1.7
-0.6
-2.8
-0.6
-1.5
-0.4
-1.6
0.0
-1.2
-0.3
-0.4
0.0
0.1
0.0
0.0
0.0
-0.3
0.3
1.4
1.8
-2.0
-1.3
-0.2
0.0
176.9
(176.5b)
234.2
(229.1c)
207.0
0.0
-2.4
-2.4
1.0
0.2
0.4
-0.4
-1.9
-0.4
0.4
-0.4
-1.9
-2.0
197.4
131.1[4];136.9[2]
C6O6Li4
C6O6Li6
C6S6Li4
C6S6Li6
10.3
d
-17.0
-12.1
13.3
7.4
-10.0
-9.6
-0.8
-0.4
Also summarized, CnSn2- (n ) 3-6) 13C chemical shifts (relative to TMS) (experimental values given in the parenthesws, and relative
a
intensities in brackets). GIAO-B3LYP/6-31+G*//B3LYP/6-311+G*. MePh3P salt, cited from ref 12. c Cited from ref 4a. The dissected
NICS assignment for C6S6Li6 is not clear-cut; however, the contributions from the six electrons are assigned to the CdC bonds.
a
b
d
TABLE 3. Cn Sn (n ) 3-6) B3LYP /6-311+G* Zer o-P oin t
En er gies (ZP E; k ca l/m ol), Nu m ber of Im a gin a r y
F r equ en cies (NIm a g), Tota l En er gies (a u ), a n d Rela tive
En er gies (k ca l/m ol)
bonding and the destabilization from the strain energy
in the four-membered rings.
Mor e High ly Red u ced F or m s: Tetr a a n ion a n d
Hexa a n ion . The existence of the tetraanion C6O64- and
hexaanion C6O66- is well established. Experimentally, the
E
rel (Etot +
ZPE)
molecule symmetry ZPE NImag
Etot
4-
tetrapotassium salt of the C6O6 tetraanion has been
6-
isolated;18 C6O6 was synthesized by thermal cyclo-
C3S3
C4S4
C5S5
C6S6
C6S6
C6S6
C6S6
C6S6
D
D
D
D2h
D3h
D3d
C2
3h (T)
4h (T)
5h (T)
14.2
15.8
20.1
26.9
27.1
25.5
25.6
26.4
0
0
2
0
0
0
0
0
-1308.86578
-1745.21333
-2181.54695
-2617.90670
-2617.89423
-2617.84489
-2617.83907
-2617.91960
trimerization of acetylene diolate salts, which are the first
species that can be isolated when the reaction of certain
metals with CO is carried out at high temperatures.19
0.0
7.7
37.4
41.1
-8.6
6-
C6S6 was synthesized by dealkylation of a thioether
precursor and has been explored in many experimental
studies.6-10 However, C6S64- has not yet been addressed.
There are formally four and six π electrons, respec-
tively, in C6O64-/C6S64- and C6O66-/C6S6,6- which should,
respectively, be antiaromatic and aromatic according to
the 4n + 2 Hu¨ckel rule. Is the tetraanion antiaromatic,
with a nonplanar structure? Conversely, is the hexaanion
aromatic, with a planar structure? To investigate the
structure of the tetraanions and hexaanions of C6O6 and
C6S6, avoiding the complication of the Coulomb repulsion,
lithium was used as a countercation.
a
C2v
a
Bicyclic structure, see Figure 2.
(D5h) triplet is a second-order saddle point (NIMAG )
2). D2h C6S6,15b,16 the most stable hexathiocarbon, and its
D3h, D3d, and C2 isomers are minima. The C6S6 isomer
energy order is in agreement with literature reports15
(Table 3); the D2h structure is lowest in energy, followed
by D3h (relative energy 7.7 kcal/mol), D3d (37.4 kcal/mol),
and C2 (41.1 kcal/mol).
Planar C6O6Li4 (D2h) is a true minimum at the B3LYP/
6-31G* level. C6S6Li4 (D2h), however, is a higher saddle
point (NImag ) 3); mode following of the first imaginary
frequency led to a nonplanar C2h minimum (Figure 3).
The most stable C6S6 isomer is not monocyclic but has
C
2v symmetry and two fused five-membered rings (Figure
2), since the highly strained four-membered sulfur rings
are avoided, as in its carbon analogue.17 This C2v struc-
ture is also favored by 1.5 kcal/mol for the C6S6 dianion
(Table 1 and Figure 1).
4-
The hexagonal carbon rings in C6O6 (0.115 Å) and
C6S64- (0.09 Å) have strong bond length alternation and
the tetraanions are antiaromatic; this is confirmed by the
NICS values (Table 2).
Figure 2 presents the B3LYP/6-311+G* structures of
the neutral thiocarbons; their total NICS values and the
specific contributions are presented in Table 4. Due to
the wave function instability of the CnSn (n ) 3-5) singlet
states, no reliable NICS values can be computed, only
the singlet states of the C6S6 cyclic isomers are discussed.
The wave function stable C6S6 (D2h) is nonaromatic, but
C6S6 (D3h) is highly aromatic. The nonaromatic D2h isomer
is lower in energy than the aromatic D3h isomer. The
greater stability of the D2h form is a compromise between
the stabilization due to aromaticity and to the strong S-S
In contrast, C6O6Li6 and C6S6Li6 have highly symmetric
D6h minima at the B3LYP/6-31G* level. The CC bond
lengths (1.411 and 1.421 Å in C6O6Li6 and C6S6Li6,
respectively) and WBI (1.311 and 1.340, respectively)
(Figure 3) indicate their strong electron delocalization in
the central carbon rings. According to the NICS analysis,
the total NICS(0) and NICS(1) values are -17.0 and
-12.1 ppm, respectively, for C6O6Li6, and -10.0 and -9.6
(18) West, R.; Niu, H. Y. J . Am. Chem. Soc. 1962, 84, 1324.
(19) (a) Liebig, J . Ann. Chem. 1834, 11, 182. (b) Heller, J . F. J ustus
Liebigs Ann. Chem. 1837, 24, 1.
(17) J iao H. J .; Wu H. S. J . Org. Chem. 2003, 68, 1475.
J . Org. Chem, Vol. 68, No. 23, 2003 8811

Chen et al.
F IGURE 2. B3LYP/6-311+G* bond lengths (in Å), Wiberg bond indices (WBI, in italic), and natural charges (q, underlined) of
2-
CnSn (n ) 3-6).
TABLE 4. Neu tr a l Th ioca r bon C6S6 Isom er s NICS(tot),
NICS(σ), NICS(π), NICS(C-S), a n d Lon e P a ir (LP )
Con tr ibu tion s a t, a n d 1.0 Å a bove, Rin g Cen ter s
taking pKa considerations into account. (Full details have
been submitted to the Crystallographic Structure Data-
base at Cambridge, CCDC 221343.) Figure 4 shows how
C6S66- anions form polymeric linear tapes through coor-
dination of para-related sulfur atoms by two η2-bridging
Na+ ions; the remaining Na+ ions are contained in cis-
edge-sharing octahedra, forming linear polymers of for-
mula [NaOH(H2O)3]∞. The C6S6-containing tapes run
parallel to the crystallographic a axis, the inorganic
polymers along the b axis. Bond lengths lie in typical
ranges (see Supporting Information); the benzenoid C
atoms are separated by 1.415(3)-1.421(3) Å; C-S dis-
tances are 1.770(2)-1.776(2) Å, which agree well with
the theoretical predictions. These minimal variations
indicate that the low symmetry of the anion’s environ-
ment has little effect on its structure.
NICS(x) NICS(tot) C-C(σ) C-C(π) C-S LP (S) core
C6S6 D2h
C6S6 D3h
0
1
0
1
11.0
5.4
-12.0
-9.2
12.2
2.0
6.9
-4.0
2.6
-22.5
-11.2
2.2
0.6
2.6
0.5
0.6 -0.4
-0.2
0.0
1.3 -0.6
1.7 -0.6
0.4
ppm, respectively, for C6S6Li6, thus confirming that
6-
6-
C6O6 and C6S6 are truly aromatics. The high NICS
values are mainly due to the π-electron contributions. All
6-
the analyses suggest that C6O6 is more aromatic than
6-
C6S6
.
6-
X-r a y Cr ysta l Str u ctu r e of C6S6 (a s Na 2C6S6H4‚
4Na OH‚12H2O). Encouraged by the prediction of a
6-
highly delocalized C6S6 structure, the aesthetically
attractive molecule was prepared and its X-ray crystal
structure determined.
Crystals grown from aqueous solutions of Na6C6S6
(Na 61) adopted the triclinic space group P1h, with a
formula best described as [Na2H4C6S6]‚4NaOH‚12H2O,
Con clu d in g Rem a r k s
2-
Cyclic Dnh CnSn (n ) 3-5) structures are minima,
whose aromaticity decreases with increasing ring size.
While C3S32- is both σ and π aromatic, C4S42- and C5S5
2-
8812 J . Org. Chem., Vol. 68, No. 23, 2003

Investigation of Thiocarbon Anions
F IGURE 3. B3LYP/6-311+G* bond lengths (in Å), Wiberg bond indices (WBI, in italic), and natural charges (q, underlined) of
C6O6Li4 (D2h), C6S6Li4 (C2h), and C6S6Li6 (D2h).
F IGURE 4. Crystal structure of Na2C6S6H4‚4NaOH‚12H2O.
are much less aromatic. Detailed dissected IGLO NICS
analysis reveals the localized orbital contributions to the
NICS shieldings. While NICS(0)π obtained from dissected
NICS computations shows a gradual decrease with ring
size, NICS(1)π is nearly constant for these three thio-
carbon dianion systems. Cyclic C6S62-, which prefers D2h
symmetry, exhibits pronounced aromatic character com-
parable to benzene and represents a nice balance between
aromaticity, strain energy, and strong S-S bonding.
Computations predict C6S66- (D6h) to be a minimum. Its
J . Org. Chem, Vol. 68, No. 23, 2003 8813

Chen et al.
crystal structure recorded for the first time is in accord
with the theoretical calculations. We hope to prepare
crystals containing the C6O66- core and to investigate the
electrochemical oxidation of the C6S66- system in future.
shielding. The IGLO calculations employed the Perdew-
Wang-91 functional in conjunction with the IGLO-III
TZ2P basis set. B3LYP/6-311+G** optimized dianion and
neutral species geometries were used, while B3LYP/6-
31G* geometries were used for higher reduced oxocarbons
and thiocarbons.
Com p u ta tion s
2-
Exp er im en ta l Section
The geometries of CnSn and CnSn (n ) 3-6) were
optimized first at B3LYP/6-31G* and then at B3LYP/6-
311+G* levels of density functional theory as imple-
mented in the Gaussian98 program.20 Frequency calcu-
lations, carried out at B3LYP/6-311+G*, determined the
nature of the stationary points and gave the ZPE.21
Optimized and frequency analysis have been performed
at B3LYP/6-31G* level for C6O6Li4 and C6O6Li6 and
C6S6Li4 and C6S6Li6. Wave-function stability was checked
at the B3LYP/6-311+G* level of theory for dianion and
neutral species and at the B3LYP/6-31G* for higher
reduced oxocarbons and thiocarbons.
NICS and their dissected localized molecular orbital
contributions, so-called dissected NICS, were computed
using the IGLO method and Pipek-Mezey localization
procedure,22 as implemented in the deMon NMR pro-
gram.23 Dissected NICS reveal the individual contribu-
tions of bond, lone pairs, and core electrons to the total
Syn th esis of Na 61. The procedure described by Fangha¨nel
was employed (see Scheme).6 Bz61 was synthesized at 0 °C in
DMF containing freshly prepared sodium benzylthiolate, to
which hexachlorobenzene was added. The colorless suspension
turned yellow very quickly and the product precipitated as a
yellow powder in less than 1 h. This was recrystallized from
3:1 v/v MeOH/CHCl3 to give lustrous yellow fibers. Bz61 was
then dealkylated in refluxing liquid NH3 by the action of
sodium metal for 30 min. The reaction was quenched with
MeOH, and the NH3 was allowed to boil off. Following washing
with N2-purged Et2O, the white residue was taken up in the
minimum quantity of N2-purged water at room temperature.
The resulting yellowish solution cooled to 0 °C and yielded
large, slightly gray crystals, which were amenable to X-ray
crystallographic structure solution. Both the intermediate
Bz61 and Na 61 gave analytical data in agreement with those
previously reported.
Ack n ow led gm en t . This work was supported by
the Alexander von Humboldt Stiftung (Z.C., L.R.S.),
Deutsche Forschungsgemeinschaft, the University of
Georgia, and the U.S. National Science Foundation
(Grant CHE-0209857). We thank Dr. Wolfgang A.
Donaubauer and Dr. Frank Hampel for crystal structure
measurement.
(20) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J . R.; Zakrzewski, V. G.; Montgomery, J . A.,
J r.; Stratmann, R. E.; Burant, J . C.; Dapprich, S.; Millam, J . M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J .;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J .; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J . B.; Cioslowski, J .; Ortiz, J . V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J .; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; J ohnson, B. G.; Chen,
W.; Wong, M. W.; Andres, J . L.; Head-Gordon, M.; Replogle, E. S.;
Pople, J . A. Gaussian 98, revision A.11; Gaussian, Inc.: Pittsburgh,
PA, 1998.
Su p p or tin g In for m a tion Ava ila ble: Gaussian98 archive
entries for the optimized structures and crystal data for
Na2C6S6H4‚4NaOH‚12H2O. This material is available free of
charge via the Internet at http://pubs.acs.org.
J O035144L
(21) Ab initio Molecular Orbital Theory; Hehre, W., Radom, L.,
Schleyer, P. v. R., Pople, J . A. Eds.; Wiley: New York, 1986.
(22) Pipek, J .; Mezey, P. G. Chem. Phys. 1989, 90, 4916.
(23) (a) Malkin, V. G.; Malkina, O. L.; Casida, M. E.; Salahub, D.
R. J . Am. Chem. Soc. 1994, 116, 5898. (b) Fleischer, U.; Kutzelnigg,
W.; Lazzeretti, P.; Mu¨hlenkamp, V. J . Am. Chem. Soc. 1994, 116, 5298.
8814 J . Org. Chem., Vol. 68, No. 23, 2003