Photochemistry of Bis(sulfonyl)diazomethanes

Article abstract of DOI:10.1002/1099-0690(200110)2001:20<3771::AID-EJOC3771>3.0.CO;2-0

The photochemistry of a series of bis(sulfonyl)diazomethanes was investigated in solution and in low temperature matrices. These diazo compounds are of interest as photoacid generators in photoresists for deep UV lithography. Triplet carbenes could be trapped in solution, but attempts to isolate the carbenes in argon matrices at 10 K were unsuccessful, the corresponding sulfenes and other rearranged products being observed under these conditions instead. This is in line with DFT and ab initio calculations, which predict the singlet carbenes to be transition states on the pathway to oxathiirene oxides, which can be looked upon as intramolecularly stabilized carbenes. The triplet carbenes lie energetically above the singlet transition states and so are not expected to have a prolonged lifetime even in low-temperature matrices.

Full text of DOI:10.1002/1099-0690(200110)2001:20<3771::AID-EJOC3771>3.0.CO;2-0

FULL PAPER
Photochemistry of Bis(sulfonyl)diazomethanes
[
a]
[a]
[a]
Wolfram Sander,* Anja Strehl, and Michael Winkler
Keywords: Matrix isolation / Photochemistry / Carbenes / Rearrangements / Sulfur heterocycles
The photochemistry of a series of bis(sulfonyl)diazomethanes
observed under these conditions instead. This is in line with
was investigated in solution and in low temperature matrices. DFT and ab initio calculations, which predict the singlet
These diazo compounds are of interest as photoacid gener- carbenes to be transition states on the pathway to oxathiirene
ators in photoresists for deep UV lithography. Triplet carb- oxides, which can be looked upon as intramolecularly stabil-
enes could be trapped in solution, but attempts to isolate the
carbenes in argon matrices at 10 K were unsuccessful, the
ized carbenes. The triplet carbenes lie energetically above
the singlet transition states and so are not expected to have
corresponding sulfenes and other rearranged products being a prolonged lifetime even in low-temperature matrices.
Introduction
wavelengths results in rearrangement to diphenylsulfene 5
hetero Wolff rearrangement), which subsequently yields di-
(
The development of new photoresists suitable for deep phenyl-α-sultine 6 (Scheme 2). The photochemistry of 3 is
UV lithography is a prerequisite for the development of thus similar to that of α-diazo ketones. A major difference
electronic devices of ever higher density.[
1Ϫ6]
The resolution is the higher stability of carbene 4 (with respect to re-
of the photolithographic process is limiting for the size of arrangement) in comparison to that of keto carbenes under
the circuit elements. Bis(sulfonyl)diazomethanes 1 have the matrix isolation conditions.
been proposed as constituents of photoresists for photoli-
thographic applications using deep UV (248 nm) expo-
[
7,8]
sure.
carbene-trapping products, together with sulfonic acids
Scheme 1). The sulfonic acids are capable of crosslinking
The photolysis of 1 in solution produces typical
(
acid-sensitive compounds in negative tone materials. Des-
pite the interest in technical applications of 1, only a few
mechanistic studies on the photochemistry of 1 and the role
Scheme 2
Laser flash photolysis of bis(phenylsulfonyl)diazome-
thane (1a) in acetonitrile was investigated by Scaiano et
al.[ These authors postulate the formation of a sulfene by
hetero Wolff rearrangement of carbene 2a (which could not
be detected directly) as the principal photochemistry of 1a.
10]
Scheme 1
of carbene 2 in the formation of acids have been published.
In solution, the photochemistry of 1 is fairly complex,
The photochemistry of phenyl(phenylsulfonyl)diazome- resulting in the formation of trapping products of the carb-
thane (3) has been studied in our laboratory previously.[
9]
enes 2 and also of rearranged products (Scheme 4). A study
Use of the matrix isolation technique allowed us to charac- addressing the question of photoacid generation by UV
terize carbene 4 as the primary photochemical product on photolysis of 1 in solution was published by Pawlowski et
irradiation at λ Ͼ 475 nm. Further irradiation at shorter al.[ It was shown that organic sulfonic acids were gener-
ated, and a complex sequence of reaction steps was postu-
8]
[
a]
Lehrstuhl für Organische Chemie II der Ruhr-Universität,
lated. Here we describe a matrix isolation investigation of
the photochemistry of a series of bis(sulfonyl)diazome-
thanes 1aϪ1e (Scheme 3).
4
4780 Bochum, Germany
Fax: (internat.) ϩ 49-(0)234/321-4353
E-mail: sander@xenon.orch.ruhr-uni-bochum.de
Eur. J. Org. Chem. 2001, 3771Ϫ3778

WILEY-VCH Verlag GmbH, 69451 Weinheim, 2001
1434Ϫ193X/01/1020Ϫ3771 $ 17.50ϩ.50/0
3771

W. Sander, A. Strehl, M. Winkler
FULL PAPER
Traces of bicyclohexyl clearly signify a radical mechanism.
In cyclohexene 7b is formed in 25% yield and the major
product is the expected addition product 9b (59% yield). In
Freon-113, polymeric material was mainly formed, together
with only 5% of S-p-tolyl toluene-4-thiosulfonate (10b). In
acetonitrile, 10b was formed in 48% yield, with S-p-tolyl 4-
methylthiobenzoate (11b) as a minor product in 5% yield.
These results clearly demonstrate that the triplet carbene 2b
is a primary product of the photolysis of 1b, and that it can
be trapped by hydrogen donors or olefins. In Freon and
acetonitrile, thioester 10b is formed by formal loss of CO2.
Obviously, this reaction requires a major rearrangement,
with migration of two oxygen atoms from the sulfur atom
to the carbene carbon atom.
Irradiation of 1b in an EPA glass (isopentane/diethyl
ether/ethanol, 4:4:2) at 77 K produces 10b as the major
nonpolymeric material, while 7b Ϫ the main product in eth-
anol at room temperature Ϫ is not found. Thus, the mech-
anism and product distribution of the photodecomposition
of 1b not only depend on the solvent, but also change dras-
tically on going from a solution at room temperature to a
glass at 77 K. To gain insight into these complex processes,
the photochemistry of the series 1aϪe was investigated un-
der matrix isolation conditions.
Scheme 3
Results and Discussion
Photochemistry of 1b in Solution
The photochemistry of bis(tolylsulfonyl)diazomethane
1b) as a prototypic derivative of 1 was investigated in a
Photochemistry of 1 in Argon Matrices
(
variety of solvents by irradiation with λ Ͼ 260 nm at room
The IR spectra of the diazo compounds 1a؊e, matrix-
temperature. In ethanol or ethanol/water (3:2 mixture), bis- isolated in argon at 10 K, exhibit strong IR absorptions in
Ϫ1
(
tolylsulfonyl)methane (7b) was formed in 90 and 85% the 2100Ϫ2135 cm range, attributable to the asym. CNN
yields, respectively (Scheme 4). Insertion products were not stretching vibration, and at 1350Ϫ1375 and 1150Ϫ1165
Ϫ1
formed in this solvent.
cm , attributable to the asym. and the sym. SO stretching
2
The photolysis of 1b in cyclohexane produces both the vibrations, respectively. Irradiation with λ Ͼ 395 nm (1aϪd)
insertion product 8b (in 63% yield) and 7b (in 23% yield). or λ Ͼ 350 nm (1e) results in the rapid disappearance of
Scheme 4
3772
Eur. J. Org. Chem. 2001, 3771Ϫ3778

Photochemistry of Bis(sulfonyl)diazomethanes
FULL PAPER
the IR absorptions attributable to 1 and formation of two and is identical within several wavenumbers to the IR spec-
major new products (Figure 1). These new products are trum of one of the photoproducts of 1a (Table 1). The small
photolabile, and irradiation at shorter wavelength (λ Ͼ differences in band positions are due to the different envir-
3
05 nm) yields CO, CO , and COS as common products. onments in argon and in argon/SO matrices. Comparison
2
2
The other products formed during the photolyses depend of these IR data with those of diphenylsulfene (5) shows
on the substituents R.
the similarities of the major IR bands attributed to the Cϭ
SO moiety (Table 2). Thus, it is clear that sulfene 12a is
2
one of the primary photoproducts of 1a.
Scheme 5
Table 1. IR spectroscopic data of matrix-isolated sulfene 12a, gen-
erated by trapping of carbene 4 with SO
2
or photolysis of diazo
compound 1a
1
2a
12a
ν˜ /cm
Ϫ1 [a]
Ϫ1 [b]
Assignment[c]
ν˜ /cm
I
rel
I
rel
5
6
7
9
86
93
64
34
100
64
48
9
83
29
26
39
588
692
763
100
81
54
15
78
26
21
46
δ (SO
phenyl
2
)
934
νsym (CS)
νsym (SO
νasym (CS)
νasym (SO
1
1
168
201
1170
1204
1385
1451
2
)
Figure 1. IR spectra of the photoproducts of diazo compounds
1aϪe
1388
452
2
)
1
phenyl
[
a]
Irradiation of diazo compounds 1 is expected to result in
the loss of N and formation of carbenes 2. Since carbenes
Formed by the reaction between matrix-isolated carbene 4 and
b]
2 2
SO (1% SO
-doped argon matrix). Ϫ [ Formed by irradiation
(λ Ͼ 395 nm) of diazo compound 1a in argon at 10 K. Ϫ Tentat-
ive assignment based on comparison with the IR data of di-
phenylsulfene.
2
[
c]
are easily trapped by molecular oxygen,[
11Ϫ18]
the
photolyses of 1 were repeated in 1% O -doped argon matri-
2
ces. Subsequent annealing for several hours at 38 K did not
result in any detectable new products, which suggests that
carbenes 2 Ϫ at least as triplet ground state carbenes Ϫ are
The other diazo compounds exhibit a similar photo-
not trapped in the matrices. A similar negative result was chemical behavior, resulting in the formation of sulfenes 12
found if CO was used as the carbene trapping reagent.[
In this case, no formation of ketenes, which should be easily moiety are almost independent of the nature of the sub-
detectable in the IR spectra, was observed. stituents R. An exception is 12e, with several combination
α-Sulfonylcarbenes are known to undergo photochemical vibrations of the CϭSO with the CH group. Since 12e is
19,20]
(Table 3). Most of the vibrations attributable to the CϭSO2
2
3
rearrangements to sulfenes (hetero Wolff rearrange- the smallest molecule in this series, its vibrational spectrum
[
9,21]
ment).
Carbenes 2 might thus rearrange to sulfenes 12. was calculated at the MP2/6Ϫ31G(d,p) level of theory. The
To provide evidence for the formation of sulfenes as one of calculated band positions are in reasonable agreement with
the products of the photolysis of 1, phenyl(phenylsulfonyl)- the experimental values.
sulfene (12a) was independently synthesized by SO trap-
Several possible structures have to be considered for the
2
ping of phenyl(phenylsulfonyl)carbene (4) (Scheme 5). This second major photoproduct of 1. By comparison with inde-
carbene is easily synthesized by irradiation of 3 in an argon pendently matrix-isolated compounds it was possible to rule
[
9]
matrix. In an SO -doped argon matrix (1Ϫ5% SO in ar- out the structures 10, 13, 14, and 15. Diazirines exhibit
2
2
Ϫ1
gon), a new set of IR absorptions is observed. On annealing characteristic absorptions around 1600 cm , and these are
at 38 K this new set of signals slightly increases in intensity, not found in the IR spectra. Diphenylsulfene (5) rearranges
Eur. J. Org. Chem. 2001, 3771Ϫ3778
3773

W. Sander, A. Strehl, M. Winkler
FULL PAPER
Table 2. Characteristic vibrations of diphenylsulfene (5) and
phenyl(phenylsulfonyl)sulfene (12a) associated with the CϭSO
moiety
2
(
Table 4). The trans isomer trans-16e is predicted to be 5
kcal/mol more stable than the cis isomer by both computa-
tional methods.
The IR spectrum of the photolysis products of 1e shows
several bands that nicely match the calculated absorptions
for cis-16e, while in the case of trans-16e there is less agree-
ment (Table 5). Because of overlapping absorptions (espe-
cially in the region characteristic of SO vibrations), how-
2
ever, the assignment of absorptions to 16e is difficult and
only tentative. Photolysis of the other diazo compounds
1
aϪd yields products that can be identified as 16aϪd by
comparison with 16e. We thus conclude that the two major
primary photoproducts of compounds 1 are the sulfenes 12
and the oxathiirene oxides 16.
Table 3. IR-spectroscopic data of the sulfenes 12 [ ν˜ /cmϪ1 (rel. intensity)]
1
2a[a]
12b[a]
12c[a]
12d[a]
12e[a]
12e[b]
Assignment
δ (SO
5
6
7
9
88 (100)
566 (100)
685 (89)
Ϫ
933 (12)
Ϫ
623 (100)
Ϫ
762 (64)
932 (11)
Ϫ
608 (89)
663 (14)
759 (18)
933 (14)
Ϫ
537 (100)
Ϫ
499 (41)
Ϫ
2
)
92 (81)
63 (54)
34 (15)
756 (38)
920 (6)
1106 (12)
793 (31)
907 (31)
1131 (54)
1167 (27)
1187 (71)
1321 (48)
1375 (76)
1395 (100)
Ϫ
ν
ν
ν
sym (CϭS) [ϩν (CH
3
)]
Ϫ
sym (SO
sym (SO
2
) ϩ ν (CH
)
3
)
[
c]
1
170 (78)
Ϫ
1168 (81)
Ϫ
1171 (78)
Ϫ
1152 (100)
Ϫ
1161
2
1166[
c]
combination
1
1
1
1
204 (26)
1205 (5)
1208 (6)
1206 (8)
1321 (41)
ν
ν
ν
asym (CϭS) [ϩν (CH
3
)]
345[
c]
1339
[c]
1349
[c]
1361
[c]
1352
1368
[c]
asym (SO
asym (SO
)
2
1367[
c]
[c]
385 (19)
451 (46)
1372 (32)
1453 (39)
1377 (26)
1479 (49)
2
) CϭSO
2
group
1468 (6)
Ϫ
[
a]
Argon, 10 K. Ϫ [b] Calculated at the MP2/6Ϫ31G(d,p) level of theory. Ϫ [c] Overlapping vibrations, only tentatively assigned to 12.
photochemically to α-sultine 6 (Scheme 2);[9] however, there Mechanistic Considerations
is no evidence for the corresponding cyclization of sulf-
enes 12.
The electron affinities (EAs) and the ionization potentials
A viable candidate for the second major photoproduct (IPs) of carbenes correlate with their nucleophilicities or
was found in the course of attempts to optimize the struc- electrophilicities.[ A high electron affinity corresponds to
22]
ture of the singlet carbene S-2e (R ϭ CH ) using ab initio a strongly electrophilic character, whilst nucleophilic carb-
3
[
MP2/6Ϫ31G(d,p)] or DFT [B3LYP/6Ϫ31G(d,p)] methods. enes are characterized by low ionization potentials. Only a
At both levels of theory, S-2e is a transition state rather few experimental data are available for carbenes; however,
than a minimum on the potential energy surface and the EA and IP are much less dependent than LUMO and
system converges to the structure of 3-methanesulfonyl-2- HOMO energies on the computational method applied, and
4
methyl-2λ -oxathiirene 2-oxide (16e). The cis-oxathiirene so can be calculated with some confidence (Table 6). For
oxide cis-16e is calculated to be 6 and 1 kcal/mol more (HSO ) C:, these calculations predict an unusually large EA
2
2
stable than S-2e at the MP2 and B3LYP levels, respectively and thus an extremely high electrophilicity both in its sing-
3774
Eur. J. Org. Chem. 2001, 3771Ϫ3778

Photochemistry of Bis(sulfonyl)diazomethanes
FULL PAPER
Table 4. Relative energies calculated at the MP2/6Ϫ31G(d,p) and double bonds (Figure 2). The SϪO bond is shorter than
B3LYP/6Ϫ31G(d,p) levels of theory, including zero point energy typical SϪO single bonds, indicating a partial double bond
(ZPE) correction
˚
character. The CϪO bond lengths of 1.728 and 1.655 A for
cis and trans are much longer than expected for typical
CϪO single bonds. The bonding situation in 16 can be ex-
plained by the two structures A and B, with A being a clas-
sical three-membered ring and B an intramolecular complex
of carbene 2 with one of the oxygen lone pairs donating
electron density to the vacant p-orbital at the carbene
center. The higher stability of the trans isomer coincides
with the shorter CϪO bond length and a smaller dipole
moment. Thus, 16 can be viewed as the intramolecularly
stabilized singlet carbene 2, which is in agreement with the
extreme electrophilicity predicted for these carbenes.
The oxathiirene oxides 16 are only a few kcal/mol more
stable than the much more reactive triplet carbenes T-2. At
room temperature, T-2 and 16 are in an equilibrium, and
so typical triplet carbene reactivity (H abstraction from the
solvent etc.) is observed in solution. If the temperature is
lowered (organic glass at 77 K or argon matrix at 10 K),
only products derived from 16 (mainly secondary
photoproducts and products of the reaction with the solv-
ent) are formed. The major photoproducts at low temper-
ature are the sulfenes 12. At the MP2 level, 12e (R ϭ CH3)
and 12f (R ϭ H) are predicted to be 70 kcal/mol and 68
kcal/mol more stable than the corresponding triplet carb-
enes T-2e and T-2f, respectively (Table 4).
The mechanism resulting in 10, which is a major
photoproduct of 1 in acetonitrile at room temperature, is
less straightforward. The formation of 10 requires the mi-
gration of two oxygen atoms from one sulfur atom to the
[
a]
Transition state.
Table 5. IR-spectroscopic data of the oxathiirene oxides 16 [ ν˜ /cmϪ1 (rel. intensity)]
1
6a[a]
16b[a]
16c[a]
16d[a]
16e[a]
cis-16e[b]
trans-16e[b]
Assignment
δ (SO
δ (COS)
ν (CO)
ν (SO)
5
8
9
63 (100)
45 (8)
86 (28)
557 (100)
847 (3)
562 (100)
848 (24)
969 (15)
1017 (55)
1095[
541 (100)
861 (32)
988 (18)
997 (14)
1079 (4)
511 (100)
853[
432 (65)
910 (20)
998 (20)
1013 (27)
1017 (16)
1158 (63)
1345 (87)
1364 (100)
501 (29)
803 (21)
894 (21)
1021 (17)
1056 (12)
1149 (43)
1338 (55)
1351 (100)
2
)
c]
974 (13)
1038 (14)
1074 (32)
961 (32)
986 (24)
993 (36)
1
1
1
022 (15)
087 (36)
c]
[
c]
[c]
[c]
[c]
[c]
181
Ϫ
1165
1189
1127[
1172
1341
1356
ν
ν
ν
sym (SO
asym (SO
asym (SO
2
)
2
c]
[c]
Ϫ
1363
Ϫ
1362
1335
)
)
[
c]
[c]
[c]
[c]
[c]
1
371
1333
2
[
a]
Argon, 10 K. Ϫ [b] Calculated at the MP2/6Ϫ31G(d,p) level of theory. Ϫ [c] Overlapping vibrations, only tentatively assigned to 16.
let and in its triplet states. From this, a very high reactivity carbene carbon atom, followed by decarboxylation.
is expected for the carbenes 2. Pawlowski et al. postulated that the 2,4-dioxa-1-thiabicyclo-
The oxathiirene oxides 16 are an interesting new class of [1.1.0]butane 17 might be an intermediate in this reaction
highly strained heterocycles. The CϪS bond lengths in cis- sequence (Scheme 6).[
8]
MP2/6Ϫ31G(d,p) calculations
and trans-16e are calculated [MP2/6Ϫ31G(d,p)] as 1.706 found that a structure related to 17 is indeed a minimum
˚
and 1.700 A, respectively, halfway between CϪS single and on the potential energy surface (Figure 3, Table 4).
Eur. J. Org. Chem. 2001, 3771Ϫ3778
3775

W. Sander, A. Strehl, M. Winkler
FULL PAPER
Table 6. Electron affinities EA and ionization potentials IP in eV for a series of carbenes
Carbene
EA(calcd.,vert.) [a]
IP(calcd.,vert.) [a]
EA(exp) [b]
IP(exp) [b]
30]
[32]
Singlet
Singlet
Singlet
Triplet
Singlet
H
Cl
2
C:
2
CϭC:
1.2
1.5
2.2
2.1
0.4
10.5
10.2
11.4
11.2
10.2
0.63[
1.6[
Ϫ
Ϫ
Ϫ
10.4
9.3
Ϫ
Ϫ
Ϫ
31]
[33]
C:
F
2
(HSO
(HSO
2
)HC:
)HC:
2
(transition state)
Triplet
Singlet
(HSO
(HSO
2
)
)
2
2
C:
C:
4.5
3.2
9.7
11.3
Ϫ
Ϫ
Ϫ
Ϫ
2
[
a]
Vertical EA and IP calculated at the B3LYP/6Ϫ311ϩϩG(d,p) level of theory. Ϫ [b] Adiabatic electron affinities.
Figure 3. Some geometric data of 17e calculated at the MP2/
6
Ϫ31G(d,p) level of theory
Figure 2. Some geometric data of cis- and trans-16e calculated at
the MP2/6Ϫ31G(d,p) level of theory
˚
At 1.620 A, the ring system CϪO bond is still quite long,
but considerably shorter than those in cis- or trans-16e
˚
(
1.728 and 1.655 A, respectively). The SϪO bond, on the
other hand, is longer (Figure 3). A bicyclic structure I, a
dipolar structure II, or a diradicaloid structure III can be
formulated as representing the electronic structure of 17e.
The carbon atom in 17e is highly pyramidalized, resulting
in a folded, four-membered ring system with a dihedral
angle of 23.2°, and a considerable amount of charge is
transferred from the ring sulfur atom to the ring carbon
atom (Figure 4). A topological analysis of the electron den-
sity distribution[ reveals that no maximum electron den-
sity path connects the two nuclei, and therefore no (3,Ϫ1)
bond critical point (BCP) exists between them; instead, a
26]
Scheme 6
(
3,ϩ1) ring critical point is found. Since the existence of a
BCP is usually taken as a necessary condition for the pres-
ence of a chemical bond, we reject description I for 17e. No
significant diradical character can be assigned to 17e, since
the RHF/6-31G(d,p) wavefunction is internally and extern-
ally stable[ and the vertical singlet triplet energy gap is
substantial [49.1 kcal/mol at B3LYP/cc-pVTZ//MP2/
27]
3776
Eur. J. Org. Chem. 2001, 3771Ϫ3778

Photochemistry of Bis(sulfonyl)diazomethanes
FULL PAPER
6
Ϫ31G(d,p)]. This indicates that structure II is the most (9), 1355.1 (39), 1350.9 (17), 1221.6 (3), 1197.8 (2), 1163.0 (37),
1
154.1 (28), 1088.0 (5), 994.2 (4), 811.9 (8), 586.4 (9), 572.1 (100)
satisfactory representation of 17e (or at least dominates a
Ϫ1
cm
.
resonance hybrid of several nonequivalent resonance struc-
[
28]
tures).
Bis(p-chlorophenylsulfonyl)diazomethane (1c): IR (Ar, 10 K): ν˜ (rel.
int.) ϭ 3776.1 (3), 3756.5 (4), 3711.0 (5), 2116.1 (59), 1612.2 (2),
1
588.1 (13), 1479.2 (21), 1399.8 (7), 1375.5 (24), 1360.6 (11), 1282.3
(
3), 1225.0 (11), 1208.2 (4), 1178.5 (4), 1171.8 (2), 1166.9 (39),
1
7
5
094.3 (10), 1086.1 (25), 1018.0 (5), 993.8 (26), 828.1 (8), 763.1 (5),
58.0 (48), 708.6 (7), 660.3 (17), 622.4 (100), 603.6 (3), 589.4 (2),
Ϫ1
66.8 (34), 484.2 (21) cm
.
Bis(cyclohexylsulfonyl)diazomethane (1d): IR (Ar, 10 K): ν˜ (rel.
int.) ϭ 2944.887 (9), 2866.4 (3), 2121.9 (81), 2091.4 (15), 1600.1
(
3), 1455.8 (7), 1365.4 (29), 1341.5 (43), 1327.6 (11), 1271.8 (3),
1
6
cm
232.3 (3), 1147.6 (61), 986.1 (9), 851.4 (3), 821.2 (2), 762.4 (4),
62.9 (7), 638.9 (2), 628.5 (3), 609.4 (100), 590.8 (3), 553.7 (18)
Ϫ1
.
Bis(methylsulfonyl)diazomethane (1e): IR (Ar, 10 K): ν˜ (rel. int.) ϭ
018.0 (1), 2935.9 (1), 2135.2 (15), 2131.7 (36), 1363.0 (26), 1361.2
100), 1355.2 (17), 1348.3 (6), 1322.8 (21), 1242.6 (36), 1157.9 (89),
3
(
1
5
008.1 (10), 1002.9 (31), 966.7 (27), 770.5(9), 758.7 (42), 587.4 (15),
Ϫ1
48.3 (35), 534.5 (61), 508.2 (17) cm
.
Calculations: Calculations were performed with the Gaussian 98
[23]
program package. Geometries and vibrations were calculated at
the B3LYP/6Ϫ31G(d,p)[ or MP2/6Ϫ31G(d,p) levels of theory.
24]
[25]
Figure 4. Important topological features of the electron density
distribution within the OCO plane of the four-membered ring of
[
26]
Electron densities for analysis within the AIM framework were
1
7e and NPA charges for some atoms of this molecule from calculated at the B3LYP/cc-pVTZ//MP2/6Ϫ31G(d,p) level of the-
B3LYP/cc-pVTZ//MP2/6Ϫ31G(d,p) calculations
ory, using Cartesian d and f functions. The AIM 2000 program by
BieglerϪKönig et al. was used for topological analysis of the elec-
tron density distribution.[
29]
MP2 predicts that 17e is about 3 kcal/mol below the
transition state S-2e and 3 kcal/mol above cis-16e. Thus, 17
is readily thermally accessible from 16 at room temperature,
providing a plausible route to 10. However, since neither 17
nor other intermediates could be directly observed in our
experiments, this pathway remains speculative.
Acknowledgments
This work was financially supported by the Deutsche Forschungs-
gemeinschaft and the Fonds der Chemischen Industrie. We thank
Hoechst AG for the donation of samples of diazo compounds 1.
Experimental Section
[1]
S. A. Macdonald, C. G. Willson, J. M. J. Frechet, Acc. Chem.
Res. 1994, 27, 151.
K. Harada, J. Polym. Mater. 1995, 12, 91.
E. Reichmanis, L. F. Thompson, in Materials Chemistry, vol.
245, Am. Chem. Soc., Washington, 1995, pp. 85.
H. Ito, Solid State Technol. 1996, 39, 164.
E. Hasegawa, K. Maeda, S. Iwasa, Polym. Adv. Technol. 2000,
Matrix Spectroscopy: Matrix isolation experiments were performed
by standard techniques with an APD CSW-202 Displex closed-
cycle helium cryostat. Matrices were produced by deposition of ar-
gon (Linde, 99.9999%) at 30 K on top of a CsI (IR) or sapphire
[2]
[3]
[
[
4]
5]
(UV/Vis) window at a rate of approximately 0.15 mmol/min. Irradi-
1
1, 560.
ation was carried out using Osram HBO 500-W high-pressure mer-
cury arc lamps in Oriel housings equipped with quartz optics. IR
irradiation from the lamps was absorbed by a 10-cm path of water
and a Schott KG1 filter. For broadband irradiation, Schott cut-off
filters were used (50% transmission at the wavelength specified),
and for narrowband irradiation interference filters in combination
with dichroic mirrors and cut-off filters were employed.
[6]
M. D. Stewart, K. Patterson, M. H. Somervell, C. G. Willson,
J. Phys. Org. Chem. 2000, 13, 767.
[7]
[8]
W. Spiess, H. Roschert, H. Wengenroth, G. Pawlowski, Microe-
lectron. Eng. 1993, 21, 267.
H. J. Merrem, R. Dammel, G. Pawlowski, J. Inf. Rec. 1996,
22, 481.
[9]
W. Sander, A. Kirschfeld, M. Halupka, J. Am. Chem. Soc.
1997, 119, 981.
Bis(phenylsulfonyl)diazomethane (1a): IR (Ar, 10 K): ν˜ (rel. int.) ϭ
[10]
F. Ortica, G. Pohlers, C. Coenjarts, E. V. Bejan, J. F. Cameron,
A. Zampini, M. Haigh, J. C. Scaiano, Org. Lett. 2000, 2, 3591.
W. Kirmse, L. Horner, H. Hoffmann, Justus Liebigs Ann.
Chem. 1958, 614, 19.
T. Sugawara, H. Iwamura, H. Hayashi, A. Sekiguchi, W. Ando,
M. T. H. Liu, Chem. Lett. 1983, 1261.
2
339.5 (3), 2115.8 (4), 2111.9 (54), 1450.8 (7), 1374.1 (29), 1359.2
[
[
[
11]
12]
13]
(18), 1225.8 (7), 1165.5 (46), 1087.6 (10), 1003.1 (5), 990.3 (11),
7
5
53.3 (5), 730.3 (4), 723.5 (17), 685.3 (10), 641.9 (11), 590.3 (100),
78.4 (3), 567.5 (18) cm
Ϫ1
.
Bis(p-tolylsulfonyl)diazomethane (1b): IR (Ar, 10 K): ν˜ (rel. int.) ϭ
951.4 (9), 2868.7 (6), 2116.2 (51), 1493.3 (2), 1454.8 (4), 1370.8
G. A. Bell, I. R. Dunkin, C. J. Shields, Spectrochim. Acta, Part
A 1985, 41, 1221.
2
Eur. J. Org. Chem. 2001, 3771Ϫ3778
3777

W. Sander, A. Strehl, M. Winkler
FULL PAPER
[
14]
W. Sander, Angew. Chem. 1985, 97, 964; Angew. Chem. Int. Ed.
Engl. 1985, 24, 988.
G. A. Ganzer, R. S. Sheridan, M. T. H. Liu, J. Am. Chem. Soc.
M. Head-Gordon, E. S. Replogle, J. A. Pople, Gaussian 98,
Revision A.3, Pittsburgh PA, 1998.
[
15]
[24]
A. D. Becke, Phys. Rev. A: Gen. Phys. 1988, 38, 3098.
R. Krishnan, J. S. Binkley, R. Seeger, J. A. Pople, J. Chem.
Phys. 1980, 72, 650.
[25]
1
986, 108, 1517.
[
[
16]
17]
W. W. Sander, J. Org. Chem. 1989, 54, 333.
W. Sander, Angew. Chem. 1990, 102, 362; Angew. Chem. Int.
Ed. Engl. 1990, 29, 344.
[26] [26a]
R. F. W. Bader, Atoms in molecules: a quantum theory, Clar-
endon Press, Oxford, New York, 1990. Ϫ [
26b]
P. Popelier,
[
[
18]
19]
Atoms in Molecules, Prentice Hall, Harlow, 2000.
R. Seeger, J. A. Pople, J. Chem. Phys. 1977, 66, 3045.
W. Sander, G. Bucher, S. Wierlacher, Chem. Rev. 1993, 93, 1583.
G. A. Bell, I. R. Dunkin, J. Chem. Soc., Faraday Trans. 2 1985,
[27]
[28]
Similar questions concerning the electronic structures of four-
membered ring systems containing heteroatoms have been dis-
cussed in the literature. See, for example, the interesting work
of Niecke et al.: E. Niecke, A. Fuchs, M. Nieger, Angew. Chem.
999, 111, 3213; Angew. Chem. Int. Ed. 1999, 38, 3028.
F. Biegler-König, J. Schönbohm, D. Bayles, J. Comp. Chem.
2001, 22, 545.
81, 725.
[
[
[
[
20]
21]
22]
23]
R. J. McMahon, O. L. Chapman, J. Am. Chem. Soc. 1986,
08, 1713.
B. E. Sarver, M. J. Jones, A. M. Van Leusen, J. Am. Chem.
Soc. 1975, 97, 4771.
1
1
[29]
W. Sander, C. Kötting, R. Hübert, J. Phys. Org. Chem. 2000,
[
[
30]
13, 561.
D. G. Leopold, K. K. Murray, A. E. S. Müller, W. C. Line-
berger, J. Chem. Phys. 1985, 83, 4849.
M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M.
A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Montgom-
ery, Jr., R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Mil-
lam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J.
Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Po-
melli, C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P.
Y. Ayala, Q. Cui, K. Morokuma, D. K. Malick, A. D. Rabuck,
K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz, B.
B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.
Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham,
C. Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe,
P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, J. L. Andres,
31] [31a]
K. K. Murray, D. G. Leopold, T. M. Miller, W. C. Line-
berger, J. Chem. Phys. 1988, 89, 5442. Ϫ [
31b]
R. L. Schwartz,
G. E. Davico, T. M. Ramond, W. C. Lineberger, J. Phys. Chem.
A 1999, 103, 8213
[
[
32]
33]
W. Reineke, K. Strein, Ber. Bunsen-Ges. Phys. Chem. 1976,
80, 343.
D. W. Kohn, E. S. J. Robles, C. F. Logan, J. Phys. Chem. 1993,
97, 4936.
Received April 9, 2001
[O01167]
3778
Eur. J. Org. Chem. 2001, 3771Ϫ3778