Extended hypervalent E′...E-E...E′ 4c-6e (E, E′ = Se, S) interactions: Structure, stability and reactivity of 1-(8-PhE′C10H6)EE(C10H 6E′Ph-8′)-1′

Article abstract of DOI:10.1039/b805678a

The structure, stability and reactivity of E′...E- E...E′ (E₂E′₂) 4c-6e are examined employing naphthalene 1,8-positions in 1-(8-PhE′C₁₀H ₆)EE(C₁₀H₆E′Ph-8′)-1′ [1 (E = Se, E′ = S), 2 (E = S, E′ = Se), 3 (E = E′ = Se) and 4 (E = E′ = S)], together with 1-C₁₀H₇EEC ₁₀H₇-1′ [5 (E = Se) and 6 (E = S)]. Linear alignments of four Se₂S₂ atoms in 1 and 2 are confirmed by the X-ray analysis. 1 was not reduced by sodium borohydride, whereas 2 was, contrary to the expectation. Similarly, E-E in 3, 5 and 6 were cleaved, whereas that in 4 was not, when allowed to react with NaBH₄ in aqueous THF. The reactivity of the E-E bonds in E₂E′₂ in 1-4 is not controlled by the central E atoms but by the outside E′ atoms. Quantum chemical (QC) calculations are performed on 1-(8-MeE′C₁₀H ₆)EE(C₁₀H₆E′Me-8′)-1′ (13-16: models of 1-4, respectively), 5 and 6, together with the related species. Conformers having E₂E′₂ 4c-6e (abbreviated by BA) are the global minima for 13, 15 and 16, which reproduces the observed structures: BA must be stabilized by the formation of E₂E′ ₂ 4c-6e. 14 (AB) with double n(S)...σ*(Se-C) 3c-4e interactions is the global minimum, which shows that σ*(Se-C) plays a crucial role to stabilize the 3c-4e in AB. The novel reactivity of E-E is considered based on the QC calculations. The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008.

Full text of DOI:10.1039/b805678a

New Journal of Chemistry
An international journal of the chemical sciences
Volume 32 | Number 11 | November 2008 | Pages 1809–2052
www.rsc.org/njc
ISSN 1144-0546
PAPER
Nakanishi et al.
…
…
Extended hypervalent E‘ E–E E‘
4c–6e (E, E‘ = Se, S) interactions:
structure, stability and reactivity of
1-(8-PhE‘C10H6)EE(C10H6E‘Ph-8‘)-1‘

PAPER
www.rsc.org/njc | New Journal of Chemistry
Extended hypervalent E0ꢀ ꢀ ꢀE–Eꢀ ꢀ ꢀE0 4c–6e (E, E0 = Se, S)
interactions: structure, stability and reactivity of
1-(8-PhE0C10H6)EE(C10H6E0Ph-80)-10w
Waro Nakanishi,*a Satoko Hayashi,a Sayuri Morinaka,a Takahiro Sasamorib and
Norihiro Tokitohb
Received (in Montpellier, France) 3rd April 2008, Accepted 23rd May 2008
First published as an Advance Article on the web 17th July 2008
DOI: 10.1039/b805678a
The structure, stability and reactivity of E0ꢀ ꢀ ꢀE–Eꢀ ꢀ ꢀE0 (E2E02) 4c–6e are examined employing
naphthalene 1,8-positions in 1-(8-PhE0C10H6)EE(C10H6E0Ph-80)-10 [1 (E = Se, E0 = S), 2
(E = S, E0 = Se), 3 (E = E0 = Se) and 4 (E = E0 = S)], together with 1-C10H7EEC10H7-10
[5 (E = Se) and 6 (E = S)]. Linear alignments of four Se2S2 atoms in 1 and 2 are confirmed by
the X-ray analysis. 1 was not reduced by sodium borohydride, whereas 2 was, contrary to the
expectation. Similarly, E–E in 3, 5 and 6 were cleaved, whereas that in 4 was not, when allowed
to react with NaBH4 in aqueous THF. The reactivity of the E–E bonds in E2E02 in 1–4 is not
controlled by the central E atoms but by the outside E0 atoms. Quantum chemical (QC)
calculations are performed on 1-(8-MeE0C10H6)EE(C10H6E0Me-80)-10 (13–16: models of 1–4,
respectively), 5 and 6, together with the related species. Conformers having E2E02 4c–6e
(abbreviated by BA) are the global minima for 13, 15 and 16, which reproduces the observed
structures: BA must be stabilized by the formation of E2E02 4c–6e. 14 (AB) with double
n(S)ꢀ ꢀ ꢀs*(Se–C) 3c–4e interactions is the global minimum, which shows that s*(Se–C) plays a
crucial role to stabilize the 3c–4e in AB. The novel reactivity of E–E is considered based on the
QC calculations.
demonstrated to be very different from that of 3c–4e.10a,c,11,12
Introduction
However, details in the reactivity of extended hypervalent
bonds are still unclear, even recently.
Extended hypervalent bonds [m center–n electron bonds
(mc–ne: m Z 4)]1–6 higher than 3c–4e are of current interest.
They play an important role in physical, chemical and biologi-
cal properties in the compounds containing the bonds.1–5 Our
strategy to construct the extended hypervalent bonds are to
employ the interactions caused by direct orbital overlaps
between nonbonded atoms.3,4,6 The bonds are also shown to
play an important role in the pharmacological activity7 and
development of electronic materials.8,9 It is inevitable to control
weak interactions to design materials of high functionalities,
since they control fine structures and create delicate properties
of materials.10 The first member of the extended hypervalent
bonds is 4c–6e. Compounds containing 4c–6e with linear four
atoms have been gradually increased by the preparation
and characterization of the compounds,5 although they are
sometimes not recognized as 4c–6e. The nature of 4c–6e3 is
The driving force for the linear alignment in 4c–6e is CT
from two p-type lone pair orbitals of outside E0 atoms (np(E0))
to the central s*(E–E) orbital. The interaction is analyzed by
the 4c–6e model. Scheme 1 depicts the approximate molecular
orbital model for E4 4c–6e of the np(E)ꢀ ꢀ ꢀs*(E–E)ꢀ ꢀ ꢀnp(E)
type. The CT in 4c–6e may correspond to c2 of E4 4c–6e.
Naphthalene 1,8-positions supply an excellent system to
study the interactions between nonbonded atoms, since the
nonbonded distances are close to the sum of van der Waals
radii minus 1 A for those of main group elements. Indeed, four
atoms of the same kind construct extended hypervalent E4
4c–6e, but those of the different kinds also form E2E02 4c–6e.
Lots of possibilities would be derived from E2E02 4c–6e, where
E0 = N,5b,13 O,14 S,3d,15 Se,3a,b and halogens,5f,16 even if E is
limited to S and Se.
a Department of Material Science and Chemistry, Faculty of Systems
Engineering, Wakayama University, 930 Sakaedani, Wakayama,
640-8510, Japan. E-mail: nakanisi@sys.wakayama-u.ac.jp;
Fax: +81 73 457 8253; Tel: +81 73 457 8253
b Institute for Chemical Research, Kyoto University, Gokasho, Uji,
Kyoto, 611-0011, Japan
w Electronic supplementary information (ESI) available: Cartesian
coordinates for optimized structures of 5, 6 and 13–20. CCDC
reference numbers 626066 for 1 and 626067 for 2. For ESI and
crystallographic data in CIF or other electronic format see DOI:
10.1039/b805678a
Scheme 1 Approximate MO model for E4 4c–6e.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008 New J. Chem., 2008, 32, 1881–1889 | 1881

Table 1 Reduction of 1–6 with NaBH4, followed by excess MeIa
Entry
Compound
Product
Reactivity
Yieldb (%)
1
2
3
4
5
6
a
1 (SSeSeS)
2 (SeSSSe)
3 (SeSeSeSe)
4 (SSSS)
5 (SeSe)
1c
7
—
98c
96
97
95c
499
499
++
++
—
+++
+++
9
4c
11
12
6 (SS)
Chart 1 Naphthalene system containing E2E02 4c–6e (E, E0 = Se, S).
Dichalcogenide (25 mmol) was solved in aqueous THF, then added
sodium borohydride (200 mmol). After stirring for 20 min at 30 1C,
excess methyl iodide (100 mmol) was added. Isolated yield.
b
We investigated the structure, stability and reactivity of
E2E02 4c–6e in bis[8-(phenylthio)naphthyl] 1,10-diselenide (1)
and bis[8-(phenylselanyl)naphthyl] 1,10-disulfide (2), together
with bis[8-(phenylselanyl)naphthyl] 1,10-diselenide (3)3a–c
c
Recovered.
and
bis[8-(phenylthio)naphthyl]
1,10-disulfide
(4)3d
(Chart 1).17 Those of 1,10-dinaphthyl diselenide (5)18,19 and
1,10-dinaphthyl disulfide (6)19 are also discussed for conveni-
ence of comparison.
ð1Þ
During the course in the preparation of 1 and 2, we found
the novel reactivity of 1, relative to 2, toward the reduction
with sodium borohydride: 1 was not reduced by sodium
borohydride, whereas 2 was reduced, although 2 is expected
to be less reactive. The observation is the first typical example
caused by E2E02 4c–6e (E, E0 = Se, S), to the best of our
knowledge. The reactivity must arise from the unique bonding
scheme of Se2S2 4c–6e in 1 and 2. Compilation of the results, it
is summarized that the reactivity of the E–E bonds in
E0ꢀ ꢀ ꢀE–Eꢀ ꢀ ꢀE0 4c–6e toward the reduction with sodium boro-
hydride is not controlled by the central E atoms but by the
outside E0 atoms in 1-(8-PhE0C10H6)EE(C10H6E0Ph-80)-10.
Here, we report the properties of 1 and 2, together with
those of 3–6. The structures are determined for 1 and 2. The
stability is also clarified, which must be closely related to the
novel reactivity of 1–4. Quantum chemical (QC) calculations
are employed to understand the unique properties of
1-(8-PhE0C10H6)EE(C10H6E0Ph-80)-10.
ð2Þ
ð3Þ
The disulfide 2 was independently prepared by the reduction
of 1-(8-PhSeC10H6)SS2S(C10H6SePh-80)-10 with NaBH4 in
ethanol, since 2 was not obtained in the reaction shown by
eqn (1). Eqn (4) exhibits the reaction. The tetrasulfide was
obtained in the reaction of 1-Br-8-(PhSe)C10H6 with magne-
sium in diethyl ether, followed by addition of excess sulfur,20
although the tetrasulfide was contaminated by 2 to some
extent. 2 was confirmed to give 7 in the reduction with NaBH4
in aqueous THF, followed by MeI.
Results and discussion
Survey of the E–E bond reduction
An almost equimolar mixture of 1 and 2 was obtained in
the reaction of the naphtho[1,8-c,d]-1,2-selenathiolate
dianion with excess benzenediazonium chloride in aqueous
THF at 5 1C. 1-(Methylthio)-8-(phenylselanyl)naphthalene
(7) was isolated, together with 1, after treatment of the
mixture with excess sodium borohydride, followed by excess
methyl iodide. 1-(Methylselanyl)-8-(phenylthio)naphthalene
(8) and 2 were not detected in the reaction mixture, contrary
to the expectation. 1-(Methylselanyl)-8-(phenylselanyl)-
naphthalene (9)3b,c was isolated in the reduction of 3
under the conditions, whereas 1-(methylthio)-8-(phenylthio)-
naphthalene (10) was not obtained in the reduction of 4;
eqns (1) and (2) show the reactions.3 As shown in eqn (3),
ð4Þ
The reactivity of E–E in 1–6 is summarized as follows: (1)
The E–E bonds (E = S and Se) in 5 and 6 are cleaved by
sodium borohydride in aqueous THF. (2) The reactivity of the
E–E bonds (E2 s(2c–2e)) dramatically changes if the bonds are
incorporated in s(4c–6e) of the np(E0)ꢀ ꢀ ꢀs*(E–E)ꢀ ꢀ ꢀnp(E)
type. (3) The reactivity of the E–E bonds is not controlled
by the central E atoms but by the outside E0 atoms. The E–E
bonds are cleaved by NaBH4, when E0 = Se but not if E0 = S.
Why such a unique reactivity is observed in E–E of 1–6? The
reactivity of the compounds is elucidated, together with the
structures and the stability.
5
and
6
gave 1-(methylselanyl)naphthalene (11) and
1-(methylthio)naphthalene (12), respectively, under similar
conditions of the reduction. Table 1 summarizes the reactivity
of 1–6 toward sodium borohydride.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008
1882 | New J. Chem., 2008, 32, 1881–1889

Scheme 2 Types A, B and C in naphthalene system.
structures for
1-(8-MeSC10H6)SeSe(C10H6SMe-80)-10
(13: Chart 2) at the B3LYP level are employed for the
illustration, although the HMe atoms are removed. The
observed structures are all BA for 1–4. The results strongly
suggest that BA with E2E02 4c–6e (E, E0 = S, Se) are more
stable than AB constructed by double Eꢀ ꢀ ꢀE0–C 3c–4e and AA
with double Eꢀ ꢀ ꢀE0 2c–4e, although the crystal packing effect
must also play an important role to determine the structures.
Why are BA observed in 1–4? Why is the reactivity of E–E in
1–4 affected so much by E0? The reason is considered based on
the QC calculations, next.
Fig. 1 Structure of 1 with displacement ellipsoids shown at the 50%
probability level. Selected bond lengths (A) and angles (1): Se1–Se2
2.3561(6), S1–Se1 3.0122(9), S2–Se2 2.9495(9); S1–Se1–Se2
169.622(13), Se1–Se2–S2 179.225(12).
QC Calculations
Structures of 1 and 2
QC calculations are performed on 1-(8-MeE0C10H6)EE-
(C10H6E0Me-80)-10 (E, E0 = Se, S) (13–16: models of 1–4,
respectively), together with 5, 6 and 1-(8-MeE0C10H6)Eꢂ
(17–20). 17–20 are the anions formed from 13–16, respectively,
by the E–E bond reduction (Chart 2). Calculations are per-
formed using the Gaussian 03 program22 at the DFT
(B3LYP)23,24 and MP225 levels.
The X-ray crystallographic analyses are carried out for suita-
ble single crystals of 1 and 2, obtained via slow evaporation of
hexane–dichloromethane solutions. Figs. 1 and 2 show the
structures of 1 and 2, respectively. The Seꢀ ꢀ ꢀS distances in 1 are
3.0122(9) and 2.9495(9) A, which are shorter than the sum of
the van der Waals radii of Se and S21 by 0.69–0.75 A. The
structure of 2 is essentially the same as that of 1, although the
positions of S and Se are exchanged. The Sꢀ ꢀ ꢀSe distances in 2
are 3.0458(7) and 3.0662(7) A, which are shorter than the sum
of the van der Waals radii by 0.63–0.65 A. The distances are
slightly shorter in 1, relative to the case of 2 (by 0.075 A in the
average).
Table 2 collects selected bond lengths, angles and torsional
angles of the optimized structures for BA of 13–16, together
with the observed values of 1–4, necessary for the discussion.
Table 3 shows the results for 13–16, 5 and 6: Energies on the
energy potential surface and the sum of electronic and thermal
free energies at 273.16 K are given for the values at the B3LYP
level and those only on the energy potential surface at the MP2
level. The sum of electronic and thermal free energies corre-
sponds to the Gibbs free energy (G = H–TS), which contains
the zero-point correction. Natural bond orbital (NBO) analy-
sis26,27 are performed on AB and BA of 13–16 at the MP2
level. Table 4 collects the results. Table 5 shows the energies
for 17–20 on the energy potential surface at the both levels.
Structures of the naphthalene system are well classified using
type A (A), B and C, where the Se–CAr bond is placed almost per-
pendicular to the naphthyl plane in A, the bond is located on the
plane in B and C is the intermediate between A and B. Scheme 2
shows the notation.3c,10a,c,12 Plausible conformers of 1–4 are
A(1)A(10)A(8)A(80) (abbreviated by AA), A(1)A(10)B(8)B(80)
(AB) and B(1)B(10)A(8)A(80) (BA). Scheme 3 illustrates the AA,
AB and BA structures. The frameworks of the optimized
Fig. 2 Structure of 2 with displacement ellipsoids shown at the 50%
probability level. Selected bond lengths (A) and angles (1): S1–S2
2.0706(9), S1–Se1 3.0458(7), S2–Se2 3.0662(7); Se1–S1–S2 167.98(3),
S1–S2–Se2 168.12(3).
Scheme 3 AA, AB and BA structures in 1–4 and 13–16. Frameworks
for optimized structures of 13 being employed with R in replace of Me.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008 New J. Chem., 2008, 32, 1881–1889 | 1883

Table 3 Energies for AA, AB and BA of 13–16, together with those
for A and B of 5 and 6a
Compound
Calculated at the B3LYP levelg
DE(AA)b,c E(AB)d
DE(AB)b,e DE(BA)b,f
13 (SSeSeS)
14 (SeSSSe)
5.8
8.1h
21.0
ꢂ6448.8883
0.0
ꢂ6448.6170h 0.0h
ꢂ26.8
ꢂ28.6h
10.2j
ꢂ6448.8768
0.0i
18.4h
15 (SeSeSeSe) 22.3
ꢂ6448.6048h 0.0hk
3.4hl
ꢂ10455.5558
0.0
ꢂ11.0
ꢂ21.0h
ꢂ4.2
19.7h
7.1
ꢂ10455.2880h 0.0h
16 (SSSS)
5 (SeSe)m
6 (SS)m
ꢂ2442.2098
0.0
7.1h
ꢂ2441.9337h 0.0h
ꢂ4.5h
11.8
ꢂ5573.8510
0.0
ꢂ5573.6273h 0.0h
13.9h
10.2
ꢂ1567.1702
0.0
ꢂ1566.9420h 0.0h
11.8h
Calculated at the MP2 leveln
13 (SSeSeS)
14 (SeSSSe)
15 (SeSeSeSe)
16 (SSSS)
5 (SeSe)m
6 (SS)m
ꢂ6439.7877 0.0
ꢂ6439.7788 0.0o
ꢂ10444.5897 0.0
ꢂ2434.9772 0.0
ꢂ5566.3505 0.0
ꢂ1561.5384 0.0
ꢂ21.0
17.9p
ꢂ0.3
ꢂ0.2
25.3
Chart 2 Calculated species.
22.5
Table 5 also contains the relative energies assuming that of AB
a
Energies on the energy potential surface are given unless otherwise
as the standard for each.
noted. In kJ molꢂ1
as the standard for each. DE(BA)
6-311+G(d) basis sets being employed for Se and
.
DE(AA) = E(AA) ꢂ E(AB). In au. Taken
b
c
d
e
f
g
=
E(BA)
ꢂ
E(AB). The
and the
6-311G(d) basis sets for C and H. Sum of electronic and thermal free
Optimized structures around E2E0 4c–6e (E, E0 = Se, S)
2
S
h
Compounds 13–16 are the models of 1–4, respectively. The
selected bond lengths, angles and torsional angles of the
optimized structures for BA of 13–16 are collected in Table 2,
which correspond to the observed values for BA of 1–4. The
values optimized at the MP2 level are very close to those of
observed values, while 13–16 have two E0–Me groups for each
whereas 1–4 have two E0–Ph groups. Although the r(E–E) and
f(CEEC) values optimized at the B3LYP level are longer and
smaller than those of observed values, structures optimized at
the B3LYP level also reproduced the observed ones as a whole.
energies at 273.16 K. 30.2 kJ molꢂ1 from 13 (AB). 67.2 kJ molꢂ1
i
j
from 13 (BA). 32.0 kJ molꢂ1 from 13 (AB). 64.0 kJ molꢂ1 from 13
k
l
m
(BA). Data for 5 (A) and 6 (A) are given in the column of AB and
n
those for 5 (B) and 6 (B) are in that of BA. The 6-311+G(d) basis sets
being employed for Se and S and the 4-31G(d) method for C and
o
H. 23.4 kJ molꢂ1 from 13 (AB). 62.1 kJ molꢂ1 from 13 (BA).
p
The optimized structures of AA, AB and BA of 13 at the B3LYP
level are employed in Scheme 3, although the methyl protons
Table 2 Selected bond lengths, angles and torsional angles of the optimized structures for BA of 13–16, together with those of 1–4
Compound
(E, E0)
r(E–E)/A
r(Eꢀ ꢀ ꢀE0)/A
+(E0EE)/1
f(CEEC)/1
Calculated at the B3LYP levela
13 (BA)b
14 (BA)b
15 (BA)
16 (BA)
(Se, S)
2.4264
2.1614
2.4476
2.1461
3.0125c
3.0587c
3.0849
2.9764
174.83
167.31
172.68
169.30
85.68
85.56
85.94
84.68
(S, Se)
(Se, Se)
(S, S)
Calculated at the MP2 leveld
13 (BA)e
14 (BA)e
15 (BA)
16 (BA)
(Se, S)
2.3680
2.0899
2.3806
2.0814
3.0134f
3.0792f
3.0887
3.0043
175.68
171.14
174.54
172.61
87.09
88.74
89.31
86.33
(S, Se)
(Se, Se)
(S, S)
Determined by X-ray analysisg
1 (BA)
2 (BA)
3 (BA)i
4 (BA)j
a
(Se, S)
2.356
2.071
2.365
2.055
2.981h
3.056h
3.053
2.988
174.43
168.06
173.78
167.3
b
ꢂ91.50
ꢂ81.34
91.4
ꢂ89.0
(S, Se)
(Se, Se)
(S, S)
The 6-311+G(d) basis sets being employed for Se and S and the 6-311G(d) basis sets for C and H. E(14 (BA)) ꢂ E(13 (BA)) = 67.2
kJ molꢂ1
.
r(Eꢀ ꢀ ꢀE0: 14 (BA)) ꢂ r(Eꢀ ꢀ ꢀE0: 13 (BA)) = 0.046 A. The 6-311+G(d) basis sets being employed for Se and S and the 4-31G(d) basis
c
d
sets for C and H. E(14 (BA)) ꢂ E(13 (BA)) = 62.1 kJ molꢂ1
.
r(Eꢀ ꢀ ꢀE0: 14 (BA)) ꢂ r(Eꢀ ꢀ ꢀE0: 13 (BA)) = 0.066 A. Averaged values are shown,
e
f
g
h
i
j
if necessary. r(Eꢀ ꢀ ꢀE0: 2 (BA)) ꢂ r(Eꢀ ꢀ ꢀE0: 1 (BA)) = 0.075. Ref. 3a,b. Ref. 3c.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008
1884 | New J. Chem., 2008, 32, 1881–1889

Table 4 Results of NBO analysis for AB and BA of 13–16, at the MP2 levela,b
Compound
NBO(i)c
NBO(j)d
E(2)/kcal molꢂ1
[E(j) ꢂ E(i)]/au
F(i,j)/au
13 (BA)e
14 (BA)f
15 (BA)g
16 (BA)h
13 (AB)
14 (AB)i
15 (AB)j
16 (AB)
a
np(S)
np(Se)
np(Se)
np(S)
np(Se)
np(S)
s*(Se–Se)
s*(S–S)
s*(Se–Se)
s*(S–S)
s*(S–C)
s*(Se–C)
s*(Se–C)
s*(S–C)
13.87
9.22
15.89
8.17
5.27
9.32
9.19
5.77
b
0.54
0.56
0.51
0.58
0.69
0.67
0.64
0.73
0.077
0.064
0.080
0.062
0.059
0.071
0.069
0.059
np(Se)
np(S)
Second-order perturbation of Fock matrix (threshold being 0.50 kcal molꢂ1). The 6-311+G(d) basis sets being employed for Se and S and the
4-31G(d) basis sets for C and H. Donor orbitals. Acceptor orbitals. E(2: ns(S)-s*(Se–Se)) = 1.16 kcal molꢂ1
.
E(2: ns(Se)-s*(S–S)) =
c
d
e
f
0.60 kcal molꢂ1
.
E(2: ns(Se)-s*(Se–Se)) = 1.11 kcal molꢂ1
.
E(2: ns(S)-s*(S–S)) = 0.62 kcal molꢂ1
.
E(2: ns(S)-s*(Se–C)) = 0.83 kcal molꢂ1
.
g
h
i
j
E(2: ns(Se)-s*(Se–C)) = 0.67 kcal molꢂ1
.
Table 5 Energies of A and B in 17–20a
E(A)b
DE(B)c,d
E(A)b
DE(B)c,d
Anion
B3LYPe
MP2f
17
18
19
20
a
ꢂ3224.4944
ꢂ3224.4911
ꢂ5227.8258
ꢂ1221.1589
ꢂ9.5
ꢂ32.0
ꢂ30.5
ꢂ11.6
ꢂ3219.9179
ꢂ3219.9095
ꢂ5222.3162
ꢂ1217.5105
b
1.4
ꢂ17.4
ꢂ17.2
0.7
c
Energies on the energy potential surface are given. In au. DE(B) =
E(B) ꢂ E(A). In kJ molꢂ1
d
e
. The 6-311+G(d) basis sets being em-
f
ployed for Se and S and the 6-311G(d) basis sets for C and H. The 6-
311+G(d) basis sets being employed for Se and S and the 4-31G(d)
basis sets for C and H.
are removed to show R in place of Me. Optimized structures at
the MP2 levels are also employed in Schemes 4 and 5.
The nonbonded r(Seꢀ ꢀ ꢀS) value for 14 (BA) is predicted to
be shorter than that for 13 (BA) at the B3LYP and MP2 levels
by 0.046 and 0.066 A, respectively. The values must corre-
spond to the observed difference of 0.075 A between 2 (BA)
and 1 (BA).28 The smaller length must arise from the stronger
the np(S)ꢀ ꢀ ꢀs*(Se–Se)ꢀ ꢀ ꢀnp(S) 4c–6e interaction relative to the
np(Se)ꢀ ꢀ ꢀs*(S–S)ꢀ ꢀ ꢀnp(Se) 4c–6e interaction. The energy dif-
ferences are evaluated to be 67 and 62 kJ molꢂ1 at the B3LYP
and MP2 levels, respectively (see also Table 3 and Scheme 4).28
Scheme 4 Relative stability of np(E0)ꢀ ꢀ ꢀs*(E–E)ꢀ ꢀ ꢀnp(E0) 4c–6e vs.
double np(E)ꢀ ꢀ ꢀs*(E0–C) 3c–4e in 13 and 14, predicted at the MP2
level.
The observed structure of 1 is BA in crystals, which corres-
pond to 13 (BA) in the calculations. The calculated results well
reproduce the observed structure of 1, since 13 (BA) is predicted
to be more stable than 13 (AB) by 21 kJ molꢂ1 at the MP2 level.
(The optimized structure of 13 (BA) employed in Scheme 4 is
very similar to the observed structure of 1, although Ph in 1 are
replaced by Me in 13.) 14 is the model of 2. The optimized
structure of 14 (BA) also reproduces the observed structure of 2
(see, Fig. 2 and Scheme 4). However, 14 (BA) is predicted to be
less stable than 14 (AB) by 18 kJ molꢂ1 at the MP2 level or
10 kJ molꢂ1 at the B3LYP level. The energy difference becomes
to smaller if the thermal effect is considered as predicted at the
B3LYP level (3 kJ molꢂ1). Nevertheless, 14 (BA) is predicted to
be still less stable than 14 (AB) containing the thermal effect at
the B3LYP level.
Stability of E2E0 4c–6e (E, E0 = Se, S)
2
As mentioned above, BA, AB and AA are optimized to be
stable for 13–16 and A and B for 5 and 6, when calculated at
the B3LYP level (Tables 2 and 3). DFT calculations reveal
that BA and AB are more stable than AA in 13–16. BA are the
global minimum for 13, 15 and 16, although the energy
difference is small in 16. AB is predicted to be the global
minimum in 14. Thermal effect operates to stabilize BA more
than AB by up to 10 kJ molꢂ1. 5 (A) is predicted to be more
stable than the 5 (B) by ca. 10 kJ molꢂ1, so is 6 (A) than 6
(B).29 The MP2 calculations are performed on the key con-
formers (AB and BA) in 13–16, together with A and B for 5
and 6. Results at the MP2 level are essentially the same as
those at the B3LYP level, although BA are less stabilized than
AB at the MP2 level for 13–16. B are less stabilized than A by
11–14 kJ molꢂ1 for 5 and 6 at the MP2 level.
The crystal packing effect is expected to play an important role
to determine the structure of 2. The effect must operate to stabilize
14 (BA) more than 14 (AB) by over the calculated energy
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008 New J. Chem., 2008, 32, 1881–1889 | 1885

difference. Similar discrepancy is also observed in the structure of
6. 5 (A) and 6 (A) are predicted to be more stable than 5 (B) and 6
(B), respectively, by 23–25 kJ molꢂ1 at the MP2 level or
10–14 kJ molꢂ1 at the B3LYP level. Indeed, 5 (A) is detected in
crystals, but 6 is observed as B, irrespective of the predictions.30
15 and 16 are the models of 3 and 4, respectively: 3 (BA) and
4 (BA) are observed in crystals. 15 (BA) is predicted to be
slightly more stable than 15 (AB) at the MP2 level whereas the
former is more stable than the latter by 21 kJ molꢂ1 if thermal
effect is considered at the B3LYP level. The observed structure
of 4 (BA) seems to be reproduced by the calculations: 16 (BA) is
predicted to be slightly more stable than 16 (AB), although the
predicted energy difference is less than 5 kJ molꢂ1 at the both
level even if the thermal effect is considered. 4 (BA) must be
more stable than 4 (AB) under the experimental conditions,
since the BA structure is observed in 40 which have the methoxy
or nitro groups at the phenyl p,p0-positions in 4 (Chart 3). The
extended form such as B in 6 and BA in 1–4 would be much
stabilized by the crystal packing effect than expected. The
phenyl groups may also operate to stabilize the BA form.
Scheme 4 illustrates the energy differences between BA and
AB (DE(BA) = E(BA) ꢂ E(AB)), which corresponds to the
formation of np(E0)ꢀ ꢀ ꢀs*(E–E)ꢀ ꢀ ꢀnp(E0) 4c–6e from double
np(E)ꢀ ꢀ ꢀs*(E0–C) 3c–4e.31 13 (BA) with Sꢀ ꢀ ꢀSe–Seꢀ ꢀ ꢀS 4c–6e is
predicted to be more stable than 13 (AB) by 21 kJ molꢂ1 at the
MP2 level. It is worthwhile to comment that 14 (AB) with double
np(S)ꢀ ꢀ ꢀs*(Se–C) 3c–4e is predicted to be more stable than 14
(BA) with Seꢀ ꢀ ꢀS–Sꢀ ꢀ ꢀSe 4c–6e by 18 kJ molꢂ1 if calculated at
the same level. The observed nonbonded Seꢀ ꢀ ꢀS distances in 1
(BA) are slightly shorter than those in 2 (BA) (by 0.075 A), which
would be the reflection of the energy profiles in 13 and 14.
Whereas the stability of the np(E0)ꢀ ꢀ ꢀs*(E–E)ꢀ ꢀ ꢀnp(E0) 4c–6e in
BA is mainly controlled by the central s*(E–E), the stability of
the np(E)ꢀ ꢀ ꢀs*(E0–C) 3c–4e in AB must be determined by
s*(E0–C) (Table 3). Namely, the stability of BA are controlled
by s*(E–E) but that of AB by s*(E0–C) in 13–16.
donor NBO (i) and acceptor NBO (j), the stabilization energy
E(2) associated with delocalization is estimated as eqn (5),
where qi is the donor orbital occupancy, ej and ei are diagonal
elements (orbital energies) and F(i,j) is the off-diagonal NBO
Fock matrix element.
E(2) = DEij = qiF(i,j)2/(ej ꢂ ei)
(5)
The energy lowering effect by the CT interactions of the
np(E0)ꢀ ꢀ ꢀs*(E–E) type in BA and the np(E)ꢀ ꢀ ꢀs*(E0–C) type
in AB are collected in Table 4. The np(E0)ꢀ ꢀ ꢀs*(E–E) interac-
tion corresponds to half of the E2E02 4c–6e interaction in BA
and there are two np(E)ꢀ ꢀ ꢀs*(E0–C) interactions in AB. The
contributions from ns are given in the footnotes of Table 4.
The effect of the CT from np is discussed, since that from ns is
small and would not perturb the discussion.
The np(E0)ꢀ ꢀ ꢀs*(E–E) interactions in 13 (BA: E = Se, E0 = S)
and 15 (BA: E = E0 = Se) stabilize the system by 13.9 and
15.9 kcal molꢂ1, respectively, but the interactions in 14 (BA: E =
S, E0 = Se) and 16 (BA: E = E0 = S) do only by 9.2 and
8.2 kcal molꢂ1, respectively. The results show that the s*(Se–Se)
bonds in 13 (BA) and 15 (BA) operate to form and stabilize the
4c–6e bonds. On the other hand, the np(E)ꢀ ꢀ ꢀs*(E0–C) interac-
tions in 14 (AB: E = S, E0 = Se) and 15 (AB: E = E0 = Se)
stabilize the system by 9.3 and 9.2 kcal molꢂ1, respectively, but
the interactions in 13 (AB: E = Se, E0 = S) and 16 (AB: E =
E0 = S) do only by 5.3 and 5.8 kcal molꢂ1, respectively. The
results show that the s*(Se–C) bonds in 14 (AB) and 15 (AB)
operate to stabilize the 3c–4e bonds.
The AB structure was not detected even in 14. However, AB
would be detected if a more suitable system is investigated,
since the effect of the np(E0)ꢀ ꢀ ꢀs*(E–E) interaction is very close
to that of the np(E)ꢀ ꢀ ꢀs*(E0–C) interaction in 14.32
How is the unique reactivity of E2E02 4c–6e in 1–4 explained
as a whole? The reactivity is considered based on QC
calculations, next.
13 and 14 are isomers with each other. Therefore, the
energies of the conformers in 13 and 14 can be directly
compared. As shown in Scheme 4, 13 (BA) is the global
minimum in 13 and 14: 13 (BA) is more stable than 14 (BA)
by 62 kJ molꢂ1 at the MP2 level (67 kJ molꢂ1 at the DFT level)
(Table 2). The lesser stability of 14 (BA) versus higher stability
of 13 (BA) explains the high reactivity of 2, relative to 1,
toward the reduction under the conditions.
Reactivity of E2E02 4c–6e (E, E0 = Se, S)
The reactivity of 13–16 derived from 1–4 could be related to
the stability of the intermediate anions [1-(8-MeE0C10H6)Eꢂ]
(17–20), produced by the reduction of 13–16. As shown in
Table 5, 18 (B) and 19 (B) are more stable than 18 (A) and 19
(A), respectively, by 17 kJ molꢂ1, whereas the stability of A
After elucidation of the structures and the stability of the
E2E02 4c–6e (E, E0 = Se, S), the driving force for the formation
of the 4c–6e is examined based on the NBO analysis.
Driving force for the formation of E2E02 4c–6e (E, E0 = Se, S)
The CT interactions can be evaluated as the second order
perturbation of Fock matrix as defined by eqn (5). For each
Scheme 5 Relative stability of anions (17–20) produced by the
reduction of the E–E bonds in 13–16: The B forms are much more
stable than A for 18 and 19 whereas the stability is almost the same for
17 and 20. Values are predicted at the MP2 level.
Chart 3 40 Having methoxy or nitro groups at phenyl p,p0-positions
in 4.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008
1886 | New J. Chem., 2008, 32, 1881–1889

and B is very similar for 17 and 20 if calculated at the MP2
level. Scheme 5 summarizes the results.
points were determined on a Yanako MP-S3 melting point
apparatus and are uncorrected. NMR spectra were recorded at
25 1C on a JEOL AL-300 spectrometer (1H, 300 MHz; 13C,
If the reduction of 13–16 is started from the observed
structures of BA, the A conformers are expected to form as
the produced anions, 17–20. However, the produced anions
must be B for 18 and 19, since they are more stable than A. This
means that the B character will grow also in the transition states
of 18 and 19, which will stabilize the transition states of the
reduction. Namely, the E–E bonds in 14 and 15 must be
reduced more easily than the case of 13 and 16. QC calculations
explain well the observed results, since 14 and 15 corresponds to
2 and 3, respectively, of which E–E bonds reduced easily.
1
75.5 MHz; 77Se, 57.3 MHz). The H, 13C and 77Se chemical
shifts are given in ppm relative to those of TMS for 1H and 13
C
NMR spectra and relative to reference compound MeSeMe
for 77Se NMR spectra. Column chromatography was per-
formed with 300–400 mesh silica gel and basic alumina. Flash
column chromatography was performed with 300–400 mesh
silica gel and basic alumina and analytical thin layer chroma-
tography was performed on precoated silica gel plates
(60F-254) with the systems (v/v) indicated.
Naphtho[1,8-c,d]-1,2-selenathiole. This compound was
prepared according to the literature.33 45% yield, Mp
Conclusion
1
122.5–123.5 1C, H NMR (300 MHz, CDCl3, d, ppm, TMS):
Extended hypervalent bonds (mc–ne: m Z 4) play an impor-
tant role in physical, chemical and biological properties in the
compounds containing the bonds. As the first step to elucidate
the properties of E2E02 4c–6e, the structure, stability
and reactivity of E0ꢀ ꢀ ꢀE–Eꢀ ꢀ ꢀE0 4c–6e are examined for
1-(8-PhE0C10H6)EE(C10H6E0Ph-80)-10 [1 (E = Se, E0 = S), 2
(E = S, E0 = Se), 3 (E = E0 = Se) and 4 (E = E0 = S)],
together with 1-C10H7EEC10H7-10 [5 (E = Se) and 6 (E = S)].
Linear alignments of four Se2S2 atoms in 1 and 2 are
confirmed by the X-ray crystallographic analysis. The S–S
bond in 2 is cleaved in the reduction with NaBH4 in aqueous
THF, whereas the Se–Se bond in 1 is not, contrary to the
expectation. On the other hand, the Se–Se bond in 3 reacts
with NaBH4 under the same conditions, whereas the S–S bond
in 4 is not. The reactivity of the E–E bond in 1–4 is not
controlled by the central E atoms but by the outside E0 atoms.
QC calculations are performed on 1-(8-MeE0C10H6)EE-
(C10H6E0Me-80)-10 (13–16: models of 1–4, respectively), 5
and 6, together with the related species. 13 (BA), 15 (BA)
and 16 (BA) containing E2E02 4c–6e are shown to be the global
minima, which explain the observed structures. However, 14
(AB) containing double n(S)ꢀ ꢀ ꢀs*(Se–C) 3c–4e interactions is
optimized to be the global minimum, although the observed
structure in 2 correspond to 14 (BA). The important role of
s*(Se–C) to stabilize n(S)ꢀ ꢀ ꢀs*(Se–C) 3c–4e is well demon-
strated by the QC calculations, containing the NBO analysis.
The reactivity of 13 and 14 can be directly examined since
they are isomers with each other. 13 (BA) is the global minimum
and 14 (BA) is most unstable in 13 and 14: 13 (BA) is more
stable than 14 (BA) by 62 kJ molꢂ1 at the MP2 level (Table 3).
Therefore, the higher reactivity of 2 toward the reduction,
relative to the negligible reactivity of 1, is well explained by
the lesser stability of 14 (BA) vs. higher stability of 13 (BA). The
reactivity toward the reduction in 1–4 is also explained based on
the stability of the produced anions (1-(8-MeE0C10H6)Eꢂ).
7.18 (dd, 3J(H,H) = 7.5 Hz, 4J(H,H) = 1.1 Hz, 1H), 7.24
(t, 3J(H,H) = 7.7 Hz, 1H), 7.24–7.32 (m, 2H), 7.35 (dd,
3J(H,H) = 7.7 Hz, 4J(H,H) = 1.1 Hz, 1H), 7.45 (dd,
3J(H,H) = 6.8 Hz, 4J(H,H) = 2.2 Hz, 1H); 77Se NMR
(57.3 MHz, CDCl3, d, ppm, MeSeMe): 561.6.
An almost equimolar mixture of 1 and 2 is obtained in the
reaction of the naphtho[1,8-c,d]-1,2-selenathiolate dianion
with excess benzenediazonium chloride in aqueous THF, at
5 1C. 1, together with 7, was isolated after the reduction of a
mixture of 1 and 2 with NaBH4 in ethanol.
Bis[8-(phenylthio)naphthyl] 1,10-diselenide (1). 29% yield, mp
1
178.6–179.6 1C, H NMR (300 MHz, CDCl3, d, ppm, TMS):
7.00–7.06 (m, 4H), 7.07–7.24 (m, 6H), 7.14 (t, 3J(H,H) =
7.9 Hz, 2H), 7.48 (t, 3J(H,H) = 7.7 Hz, 2H), 7.69 (dd, 3J(H,H) =
8.1 and 4J(H,H) = 0.9 Hz, 2H), 7.91 (dd, 3J(H,H) = 8.1 and
4J(H,H) = 1.3 Hz, 2H), 7.93 (dd, 3J(H,H) = 7.5 and 4J(H,H) =
1.1 Hz, 2H), 7.94 (dd, 3J(H,H) = 8.1 and 4J(H,H) = 1.1 Hz,
2H); 77Se NMR (57.3 MHz, CDCl3, d, ppm, MeSeMe): 510.5.
Anal. Calc. for 1 (C32H22S2Se2): C, 61.15; H, 3.53%. Found:
C, 61.19; H, 3.60%.
1-(Methylthio)-8-(phenylselanyl)naphthalene (7). 64% yield,
mp 85.5–86.5 1C, 1H NMR (300 MHz, CDCl3, d, ppm, TMS):
2.56 (s, 3H), 7.12 (t, 3J(H,H) = 7.7 Hz, 1H), 7.24 (dd, 3J(H,H) =
4
7.6 and J(H,H) = 1.4 Hz, 1H), 7.31–7.41 (m, 4H), 7.59–7.65
(m, 3H), 7.74 (dd, 3J(H,H) = 8.6 and 4J(H,H) = 1.5 Hz, 1H),
7.77 (dd, 3J(H,H) = 7.3 and 4J(H,H) = 1.4 Hz, 1H); 13C NMR
(75.0 MHz, CDCl3, d, ppm, TMS): 32.64, 125.55, 125.99,
127.37, 128.35, 129.54, 129.62, 130.74, 133.46, 133.61, 133.90,
133.97, 135.40, 135.96, 136.19; 77Se NMR (57.3 MHz, CDCl3,
d, ppm, MeSeMe): 457.2. Anal. Calc. for 7 (C17H14SSe): C,
62.00; H, 4.28%. Found: C, 61.74; H, 4.13%.
Bis[8-(phenylselanyl)naphthyl] 1,10-disulfide (2). This
compound
was
prepared
by
the
reduction
of
1-(8-PhSeC10H6)S4(C10H6SePh-80)-10 (23) with NaBH4
in ethanol, 23 being prepared in the reaction of
Experimental
10b
1-Br-8-(PhSe)C10H6
with magnesium in diethyl ether,
followed by addition of sulfur. Recrystallization from CHCl3
General considerations
1
Manipulations were performed under the nitrogen or argon
atmosphere with standard vacuum-line techniques. Glassware
was dried at 130 1C overnight. Solvents and reagents were
purified by standard procedures as necessary. The melting
gave pure 2 in 22% yield. 2: Mp 178.6–179.6 1C, H NMR
(300 MHz, CDCl3, d, ppm, TMS): 7.19 (t, 3J(H,H) = 7.8 Hz,
2H), 7.27 (t, J(H,H) = 7.7 Hz, 2H), 7.28–7.33 (m, 6H), 7.38
3
(dd, 3J(H,H) = 7.5 and 4J(H,H) = 1.3 Hz, 2H), 7.44
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008 New J. Chem., 2008, 32, 1881–1889 | 1887

3
4
(dd, J(H,H) = 7.4 and J(H,H) = 1.3 Hz, 2H), 7.51–7.57 (m,
4H), 7.68 (dd, 3J(H,H) = 8.1 and 4J(H,H) = 1.1 Hz, 2H), 7.73
4 W. Nakanishi, S. Hayashi and N. Itoh, Chem. Commun., 2003,
124–125; W. Nakanishi, S. Hayashi and N. Itoh, J. Org. Chem.,
2004, 69, 1676–1684; W. Nakanishi, S. Hayashi, T. Furuta, N.
Itoh, Y. Nishina, M. Yamashita and Y. Yamamoto, Phosphorus,
Sulfur Silicon Relat. Elem., 2005, 180, 1351–1355; W. Nakanishi, S.
Hayashi, S. Yamaguchi and K. Tamao, Chem. Commun., 2004,
140–141.
3
4
(dd, J(H,H) = 8.1 and J(H,H) = 1.3 Hz, 2H); 77Se NMR
(57.3 MHz, CDCl3, d, ppm, MeSeMe): 455.1. Anal. Calc. for 2
(C32H22S2Se2): C, 61.15; H, 3.53%. Found: C, 61.22; H, 3.62%.
5 (a) G. Mugesh, A. Panda, H. B. Singh, N. S. Punekar and R. J.
Butcher, Chem. Commun., 1998, 2227–2228; (b) G. Mugesh, A.
Panda, H. B. Singh, N. S. Punekar and R. J. Butcher, J. Am. Chem.
Soc., 2001, 123, 839–850; (c) J. E. Drake, M. B. Hursthouse, M.
Kulcsar, M. E. Light and A. Silvestru, Phosphorus, Sulfur Silicon
Relat. Elem., 2001, 169, 293–296; (d) J. E. Drake, M. B. Hurst-
house, M. Kulcsar, M. E. Light and A. Silvestru, J. Organomet.
Chem., 2001, 623, 153–160; (e) G. Mugesh, A. Panda, S. Kumar, S.
Apte, H. B. Singh and R. J. Butcher, Organometallics, 2002, 21,
884–892; (f) J. R. Anacona, J. Gomez and D. Lorono, Acta
Crystallogr., Sect. C, 2003, 59, o277–o280; (g) G. Mugesh, H. B.
Singh and R. J. Butcher, Eur. J. Inorg. Chem., 1999, 1229–1236; (h)
G. Mugesh, H. B. Singh and R. J. Butcher, J. Organomet. Chem.,
1999, 577, 243–248; (i) D. Shimizu, N. Takeda and N. Tokitoh,
Chem. Commun., 2006, 177–178.
X-Ray crystal structure determinations. Single crystals of 1
and 2 were obtained by slow evaporation of hexane at room
temperature. X-Ray diffraction data for 1 and 2 were collected
on a Rigaku/MSC Mercury CCD diffractometer equipped with
a graphite-monochromated Mo-Ka radiation source (l =
0.71070 A). For 1, the structure analysis is based on 4329
observed reflections with I 4 2.00s(I) and 413 variable para-
ꢀ
meters; orange needles, 103 K, triclinic, space group P1 (no. 2),
a = 9.579(3), b = 11.979(4), c = 12.141(4) A, a = 91.773(4), b =
102.912(4), g = 110.758(2)1, V = 1260.5(7) A3, Z = 2, R =
0.020, Rw = 0.053, GOF = 1.000. For 2 the structure analysis
is based on 4812 observed reflections with I 4 2.00s(I) and
413 variable parameters; colorless needles, 103 K, triclinic,
space group C2/c (no. 15), a = 25.3639(5), b = 11.9664(4),
c = 19.4789(6) A, b = 118.9946(12)1, V = 5171.1(3) A3,
Z = 8, R = 0.026, Rw = 0.057, GOF = 1.054. The structures
were solved by direct method (SIR97),34 and refined by full-matrix
least-square method on F2 for all reflections (SHELXL-97).35
CCDC reference numbers 626066 for 1 and 626067 for 2.
For crystallographic data in CIF or other electronic format
see DOI: 10.1039/b805678a
6 Molecular Interactions. From van der Waals to Strongly Bound
Complexes, ed. S. Scheiner, Wiley, New York, 1997; K. D. Asmus,
Acc. Chem. Res., 1979, 12, 436–442; W. K. Musker, Acc. Chem.
Res., 1980, 13, 200–206.
7 F. T. Burling and B. M. Goldstein, J. Am. Chem. Soc., 1992, 114,
2313–2320; Y. Nagao, T. Hirata, S. Goto, S. Sano, A. Kakehi, K.
Iizuka and M. Shiro, J. Am. Chem. Soc., 1998, 120, 3104–3110; S.
Wu and A. Greer, J. Org. Chem., 2000, 65, 4883–4887; Y. Nagao,
H. Iimori, S. Goto, T. Hirata, S. Sano, H. Chuman and M. Shiro,
Tetrahedron Lett., 2002, 43, 1709–1712; E. Meyer, A. C. Joussef,
H. Gallardo, A. J. Bortoluzzi and R. L. Longo, Tetrahedron, 2003,
59, 10187–10193; Y. Nagao, T. Honjo, H. Iimori, S. Goto, S. Sano,
M. Shiro, K. Yamaguchi and Y. Sei, Tetrahedron Lett., 2004, 45,
8757–8761; J. C. Taylor and G. D. Markham, J. Biol. Chem., 1999,
274, 32909–32914; W. Brandt, A. Golbraikh, M. Tager and U.
Lendeckel, Eur. J. Biochem., 1999, 261, 89–97.
8 V. Lippolis and F. Isaia, Charge-Transfer (C.-T.) Adducts and
Related Compounds, in Handbook of Chalcogen Chemistry: New
Perspectives in Sulfur, Selenium and Tellurium, ed. F. A.
Devillanova, Royal Society of Chemistry, Cambridge, 2006, ch.
8.2, pp. 477–499C. Rethore, M. Fourmigue and N. Avarvari,
Chem. Commun., 2004, 1384–1385; C. Rethore, M. Fourmigue
and N. Avarvari, Tetrahedron, 2005, 61, 10935–10942; C. Rethore,
N. Avarvari, E. Canadell, P. Auban-Senzier and M. Fourmigue, J.
Am. Chem. Soc., 2005, 127, 5748–5749.
QC Calculations
Calculations are performed with the Gaussian 03 program.22
The density functional theory (DFT) level of the Becke three
parameter hybrid functionals with the Lee–Yang–Parr corre-
lation functional (B3LYP)23,24 are applied with the
6-311+G(d) basis sets being employed for Se and S and the
6-311G(d) basis sets for C and H. Frequency analysis is
performed for each conformer at the B3LYP level. The
Møller–Plesset second-order energy correlation (MP2)
method25 is also applied with the 6-311+G(d) basis sets for
Se and S and the 4-31G(d) basis sets for C and H.
9 C. Rethore, A. Madalan, M. Fourmigue, E. Canadell, E. B. Lopes,
M. Almeida, R. Clerac and N. Avarvari, New J. Chem., 2007, 31,
1468–1483.
10 (a) W. Nakanishi, S. Hayashi and T. Uehara, J. Phys. Chem. A, 1999,
103, 9906–9912; (b) W. Nakanishi, S. Hayashi and T. Uehara, Eur. J.
Org. Chem., 2001, 3933–3943; (c) W. Nakanishi and S. Hayashi, J.
Org. Chem., 2002, 67, 38–48; (d) S. Hayashi, H. Wada, T. Ueno and
W. Nakanishi, J. Org. Chem., 2006, 71, 5574–5585.
11 G. C. Pimentel, J. Chem. Phys., 1951, 19, 446–448; J. I. Musher,
Angew. Chem., Int. Ed. Engl., 1969, 8, 54–68; R. A. Hayes and J. C.
Martin, in Sulfurane Chemistry in Organic Sulfur Chemistry:
Theoretical and Experimental Advances, eds. F. Bernardi, I. G.
Csizmadia and A. Mangini, Elsevier, Amsterdam, 1985Chemistry
of Hypervalent Compounds, ed. K.-Y. Akiba, Wiley-VCH,
New York, 1999.
Acknowledgements
This work was partially supported by a Grant-in-Aid for
Scientific Research (Nos. 16550038, 19550041 and 20550042)
from the Ministry of Education, Culture, Sports, Science and
Technology, Japan.
References
12 W. Nakanishi, S. Hayashi, A. Sakaue, G. Ono and Y. Kawada, J.
Am. Chem. Soc., 1998, 120, 3635–3640.
13 S. Kumar, K. Kandasamy, H. B. Singh and R. J. Butcher, New J.
Chem., 2004, 28, 640–645.
1 W. Nakanishi, Hypervalent Chalcogen Compounds, in Handbook of
Chalcogen Chemistry: New Perspectives in Sulfur, Selenium and
Tellurium, ed. F. A. Devillanova, Royal Society of Chemistry,
Cambridge, 2007, ch. 10.3, pp. 644–668.
14 R.-F. Hu, Y.-H. Wen, J. Zhang, Z.-J. Li and Y.-G. Yao, Acta
Crystallogr., Sect. E, 2004, 60, o2029–o2031; S. Wang, H. Mao-Lin
and F. Chen, Acta Crystallogr., Sect. E, 2004, 60, m413–m415;
X.-J. Wang, Z.-F. Chen, B.-S. Kang, H. Liang, H.-Q. Liu, K.-B.
Yu, C.-Y. Su and Z.-N. Chen, Polyhedron, 1999, 18, 647–655.
15 J. C. Barnes, J. D. Paton, W. Schroth and L. Moegel, Acta
Crystallogr., Sect. B, 1982, 38, 1330–1332; D. J. Dahm, F. L.
May, J. J. D’Amico, C. C. Tung and R. W. Fuhrhop, Cryst. Struct.
Commun., 1977, 6, 393.
2 S. Alvarez, F. Mota and J. Novoa, J. Am. Chem. Soc., 1987, 109,
6586–6591; W. B. Farnham, D. A. Dixon and J. C. Calabrese, J.
Am. Chem. Soc., 1988, 110, 8453–8461, see also J. C. Martin and
M. M. Chau, J. Am. Chem. Soc., 1974, 96, 3319–3321.
3 (a) W. Nakanishi, S. Hayashi and S. Toyota, Chem. Commun.,
1996, 371–372; (b) W. Nakanishi, S. Hayashi and S. Toyota, J.
Org. Chem., 1998, 63, 8790–8800; (c) S. Hayashi and W.
Nakanishi, J. Org. Chem., 1999, 64, 6688–6696; (d) W. Nakanishi,
S. Hayashi and T. Arai, Chem. Commun., 2002, 2416–2417.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008
1888 | New J. Chem., 2008, 32, 1881–1889

16 T. C. W. Mak, W.-H. Yip, W.-H. Chan, G. Smith and C. H. L.
Kennard, Aust. J. Chem., 1989, 42, 1403–1406.
27 NBO Ver. 3.1, E. D. Glendening, A. E. Reed, J. E. Carpenter and
F. Weinhold.
17 The structures of 3 (E = E0 = Se) and 4 (E = E0 = S) were
determined by the X-ray crystallographic analysis:3a,b,d Four E
atoms in 3 and 4 align linearly among a lot of possibilities with the
free rotation around the C–Se and C–S bonds, respectively.
18 P. G. Jones, C. Kienitz and C. Thone, Z. Kristallogr., 1994, 209,
83–84.
28 The accuracy of the calculated bond lengths for the classical bonds
and the nonbonded distances with the method employed in this work
would be around 0.01 and 0.1 A, respectively. However, the accuracy
for the differences between conformers must be higher. Therefore,
the predicted r(Seꢀ ꢀ ꢀS) differences of 0.046 (B3LYP) and 0.066 A
(MP2) would explain the observed values of 0.075 A, although the
Ph groups in 1–4 are replaced by the Me groups in 13–16 (Table 2).
The discussion on the accuracy of the energies seems more delicate.
Therefore, we usually discuss the differences between those of the
conformational isomers.36 In our cases, the differences make some
sense if they are larger than several kJ molꢂ1, although they depend
on the calculation method.37 It is better to compare a series of
differences (not between only two), as shown in Tables 3–5. The
crystal packing effect and/or solvent effect must also be considered
when the values are compared with the experimental results.
29 The results suggest that 2 (BA) and 2 (AB) may be in equilibrium in
solutions.
19 Details will be reported elsewhere.
20 The reactivity of 2 is examined carefully, using the compound
independently prepared.
21 A. Bondi, J. Phys. Chem., 1964, 68, 441–451.
22 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, J. A. Montgomery, Jr, T. Vreven, K. N.
Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V.
Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A.
Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R.
Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J.
B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J.
J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M.
C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghava-
chari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S.
Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P.
Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A.
Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M.
W. Gill, B. Johnson, W. Chen, M. W. Wong, C. Gonzalez and J.
A. Pople, GAUSSIAN 03 (Revision B.05), Gaussian, Inc.,
Pittsburgh, PA, 2003.
23 C. Lee, W. Yang and R. G. Parr, Phys. Rev. B, 1988, 37, 785–789;
B. Miehlich, A. Savin, H. Stoll and H. Preuss, Chem. Phys. Lett.,
1989, 157, 200–2006.
24 A. D. Becke, Phys. Rev. A, 1988, 38, 3098–3100; A. D. Becke, J.
Chem. Phys., 1993, 98, 5648–5652.
25 C. Mfller and M. S. Plesset, Phys. Rev., 1934, 46, 618–622; J.
Gauss, J. Chem. Phys., 1993, 99, 3629–3643; J. Gauss, Ber.
Bunsenges, Phys. Chem., 1995, 99, 1001–1008.
26 A. E. Reed, R. B. Weinstock and F. Weinhold, J. Chem. Phys.,
1985, 83, 735–746; J. E. Carpenter and F. Weinhold, J. Mol.
Struct. (THEOCHEM), 1988, 169, 41–62.
30 For the structures of
CCDC-611301 (6)19
5 and 6, see CCDC-611300 (5) and
31 The process also contains the change from double p(E0)–p(Nap) to
double p(E)–p(Nap) interactions.
32 The charges at E and E0 are also very important to understand the
formation of np(E0)ꢀ ꢀ ꢀs*(E–E)ꢀ ꢀ ꢀnp(E0) 4c–6e in BA and the
np(E)ꢀ ꢀ ꢀs*(E0–C) 3c–4e in AB.3 However, it is difficult to discuss
directly based on the values, since they are mainly determined by
the conformers, A or B. Details will be discussed elsewhere.
33 J. Meinwald and D. Dauplaise, J. Am. Chem. Soc., 1977, 99, 7743.
34 A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C.
Giacovazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori and
R. Spagna, J. Appl. Crystallogr., 1999, 32, 115–119.
35 G. M. Sheldrick, SHELXL-97, Program for the Refinement of
Crystal Structures, University of Gottingen, Gottingen, 1997.
36 This idea is well supported by the high accuracy in the calculations
for the isodesmic reaction. See, J. B. Foresman and A. E. Frisch,
Exploring Chemistry with Electronic Structure Methods, Gaussian
Inc, Pittsburg, PA, 2nd edn, 1996, ch. 8.
37 1 (BA) and 2 (BA) are optimized employing the 6-311G(d) basis
sets for S and Se and the 4-31G basis sets for C and H. The former
is predicted to be more stable than the latter by 70.9 kJ molꢂ1 with
r(Sꢀ ꢀ ꢀSe) of 1 (BA) being shorter than that of 2 (BA) by 0.052 A.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008 New J. Chem., 2008, 32, 1881–1889 | 1889