The displacement of RSO2 groups from tetrahedral carbon

Article abstract of DOI:10.1071/CH07237

The first known examples of direct displacement of RSO₂ groups from C(sp³) are described. The successful displacements are rationalized in terms of related experimental results and Coulombic stabilization in transition states, antici

Full text of DOI:10.1071/CH07237

Full Paper
CSIRO PUBLISHING
Aust. J. Chem. 2008, 61, 466–471
www.publish.csiro.au/journals/ajc
The Displacement of RSO2 Groups from Tetrahedral Carbon
Stephen DuffyA,B and Richard F. LanglerA,B
ADepartment of Chemistry, Mount Allison University, Sackville, New Brunswick, E4 L 1G8, Canada.
BCorresponding authors. Email: sduffy@mta.ca (analytical chemistry);
rlangler@mta.ca (organic chemistry)
The first known examples of direct displacement of RSO2 groups from C(sp3) are described.The successful displacements
are rationalized in terms of related experimental results and Coulombic stabilization in transition states, anticipated from
modern MO computations on ground state structures.
Manuscript received: 13 July 2007.
Final version: 27 March 2008.
Introduction
in Fig. 1 suggest that attack by simple nucleophiles at tetrahe-
dral carbon bearing an SO2 group should suffer from a powerful
Coulombic disincentive.
It has long been recognized that a wide range of common nucleo-
philes have great difficulty in displacing leaving groups α to
sulfone sulfonyl groups.[1–4] The common exception, nucleo-
philic displacement of an α-halogen by a mercaptide anion,[5–8]
is illustrated in Scheme 1.
No satisfactory explanation has been provided for the suc-
cess of mercaptide anions in these reactions. A closely related
and even more intriguing question centres on the sulfonyl group
itself. Why is there no known example in which a sulfonyl
group is displaced from tetrahedral carbon by a nucleophile (see
Scheme 2 for an obvious possibility)?
In accord with the foregoing discussion of C(sp3)S bond rup-
ture for sulfones and sulfonates, E2 eliminations work well for
sulfones in specially designed substrates (see ref. [9], p. 39).
Presumably, β-eliminations can exploit the natural exophilic-
ity (nucleofugality) of the sulfonyl groups because they do not
require the anionic centre to approach the carbon bearing the sul-
fonyl group. Successful eliminations of this type usually feature
incipient resonance in the transition state (see Fig. 2).
The present report arises from an effort to identify the first
example of sulfonyl group displacement from C(sp3).
A similar process, CS bond rupture, in nucleophilic sub-
stitutions of sulfonic acid esters has been discussed in some
detail.[9] Appropriate sulfonates, like their sulfone counterparts,
are known to be displaced with CS bond rupture by means of
nucleophilic aromatic substitutions but have not been displaced
by C(sp3)S bond rupture.[9] Semi-empirical molecular orbital
computations[9] suggest that C(sp3)S bond rupture in reactions
of sulfonates should be superior to C(sp3)O bond rupture (nor-
mally observed) on the bases of bond strengths and/or lowest
unoccupied molecular orbital (LUMO) coefficients. Based on
pKa values for the conjugate acids of the two leaving groups,
C(sp3)S and C(sp3)O bond rupture should be competitive.[9]
Both semi-empirical and ab initio computational results antic-
ipate the observed behaviour for sulfones and sulfonates based
on the net partial charges associated with the framework atoms
in/near the sulfonyl group (see Fig. 1).The computational results
Results and Discussion
As a key part of our synthetic program to prepare biologically
active α-substituted disulfides, we have developed chemistry that
ꢁ0.84
ꢁ0.84
(ꢁ0.71) (ꢁ0.71)
ꢁ0.85 ꢁ0.85
O
O
O
O
2.20 (1.54)
S
2.33
S
C
C
ꢁ0.57
(ꢁ0.75)
C
0.13
O
C
ꢁ0.55
ꢁ0.57
(ꢁ0.75)
ꢁ0.60
Fig. 1. PM3 and (in brackets) 6-31G∗ calculated skeletal atomic charges
for dimethyl sulfone and methyl methanesulfonate.
CH3SO2CH2Cl ꢀ PhCH2SNa
CH3SO2CH2SCH2Ph
C2H5OH
Scheme 1.
CH3SO2CH3 ꢀ PhCH2SNa
CH3SO2Na ꢀ PhCH2SCH3
C2H5OH
Scheme 2.
© CSIRO 2008
10.1071/CH07237
0004-9425/08/060466

Displacement of RSO2 Groups from C(sp3)
467
permits the one-pot conversion of dimethyl disulfide into an
α-acetate disulfide,[10,11] an α-propionate disulfide,[10,11] or an
α-benzoate disulfide.[12] Our initial report[13] describing the
application of these electrophiles to the synthesis of α-sulfone
disulfides provided the results shown in Scheme 3.
The failure of the α-propionate sulfide to do the apparently
simple substitution that the α-propionate disulfide does read-
ily, may suggest that the disulfide linkage assists in lowering
the transition-state energy in a manner that is not available to
the sulfide moiety. Application of Valence Bond Theory to the
transition state for the α-ester disulfide is shown in Fig. 3.
In Fig. 3, the latter two resonance forms provide an explicit
proposed role for the disulfide functionality in dispersing nega-
tivecharge and stabilizingthetransition state.The corresponding
forms for the reaction of an α-ester sulfide would be much less
favourable (see Fig. 4).
Fig. 5, would present very unfavourable Coulombics for nucleo-
philic attack by simple hard bases. But it is the hard bases that
present the best hope for successful attack at carbon because
they have little inclination to attack at sulfenyl sulfur. Because,
in analogy to computational results on sulfoxides and sulfones,
sulfinate anions are expected to have substantial partial positive
charge on sulfur, we have selected the sulfur atoms in sulfinate
anions as the nucleophiles to react with electrophilic carbon (see
Fig. 5) in α-sulfone disulfides. In this way, one might capitalize
ꢁ
OC(O)Rꢂ
OC(O)Rꢂ
RS
S
C
H
H
RS
S
C
H
H
ꢁ
Nu
Nu
Given the need for skeletal assistance in the E2 eliminations
that remove HSO2R (see Fig. 2) and the absence of a prece-
dent for bimolecular substitutions that eject sulfinate anions,
it seemed wise to seek examples of sulfonyl group displace-
ments using a helpful framework.This work began by examining
α-sulfonedisulfidesassuitablesulfonyl-containingelectrophiles
in the search for a first example of direct substitution with
C(sp3)S bond rupture.
OC(O)Rꢂ
OC(O)Rꢂ
ꢁ
RS
S
C
H
H
ꢁ
RS
S
C
H
H
Even simple α-sulfone disulfides are complex enough to offer
several electrophilic sites (see Fig. 5).
Nu
Nu
In the base-catalyzed reaction of 1 with benzenethiol, it is
the disulfide sulfur atom, indicated in Fig. 5, that is attacked
(see Scheme 4).[13]
Fig. 3. Proposed bond no-bond resonance forms for the transition state for
nucleophilic displacement at the α-carbon of an α-ester disulfide.
Thus the disulfide sulfur marked in Fig. 5 serves as the pre-
ferred soft acid site in 1. In accord with the introductory section
of this report, shifting nucleophilic attack to the C(sp3) marked in
OC(O)Rꢂ
OC(O)Rꢂ
ꢁ
R
S
C
H
H
ꢁ
R
S
C
H
H
ꢁ
O
O
Nu
Nu
Rꢃ
S
Fig. 4. Proposed bond no-bond resonance forms showing why α-ester sul-
fides are less prone to do bimolecular substitutions than α-ester disulfides
(see Fig. 3).
.........
RCH
CH
HC
CH
H
CH
CHRꢂ
CH3
SO2
CH2S
SCH3
B
Fig. 5. Two electrophilic sites in the α-sulfone disulfide 1.
Fig. 2. The transition state of a β-elimination of a sulfonyl group.
SO2Na
CH3
H2O
CH3SSCH2SO2
CH3
Acetone
ꢀ CH3SSCH2OC(O)C2H5
1
SO2Na
CH3
H2O
Acetone
ꢀ CH3SCH2OC(O)C2H5
Scheme 3.

468
S. Duffy and R. F. Langler
SO2CH2SH
CH3
Pyridine
CH2Cl2
SO2CH2SSCH3 ꢀ PhSH
CH3
ꢀ PhSSCH3
1
Scheme 4.
The presence of the sulfone disulfide substitution products 3
SO2SCH3
CH3
(in the first Scheme 6 equation) and 1 (in the second Scheme 6
equation) was established by 270 MHz 1H NMR and by GCMS.
The results in Scheme 6 establish that sulfonyl substitutions on
α-sulfone disulfides are reversible under our reaction conditions
and that neither reaction had reached equilibrium in 2 h. Fur-
thermore, the observation that 2 formed in the second reaction
indicates that at least some sulfinate-anion nucleophiles attack
at sulfenyl sulfur.
2
Fig. 6. Structure of methyl p-tolylthiosulfonate.
CH3
To confirm that sulfinate anions will react with other acti-
vated disulfide electrophiles under our reaction conditions, the
sodium salt of p-toluenesulfinic acid was reacted with phenyl
methyl disulfide (see Scheme 7).
Because thiosulfonates react with mercaptide anions to fur-
nish disulfides, it is not surprising that only disulfides were
isolated from the Scheme 7 reaction. It is, however, very likely
that 2 served as an intermediate in that reaction.
ꢀ
ꢁ
ꢀ
S
O
SO2SCH3 ꢀ CH2S
CH3
CH2
S
O
ꢁ
2
S
In accord with the results on the esters shown in Scheme 3,
sulfone sulfonyl groups were smoothly displaced from
α-sulfone disulfides by sulfinate anions (see Scheme 6) but
the corresponding reaction with an α-sulfone sulfide failed (see
Scheme 8).
In support of the view that the chemistry in Scheme 6 is
bimolecular, the sulfone disulfide 1 was recovered unchanged
after warming in aqueous acetone (see Scheme 9).
CH3
1
Scheme 5. A proposed concerted mechanism for the pyrolysis of 1 in the
GC system of a HP 5988A GLC/MS system.
on the greater reactivity of α-substituted disulfide frameworks
(see Fig. 3) and turn the Coulombic interactions to advantage.
It was our plan to initiate this work by screening crude
reaction mixtures by gas chromatography mass spectrometry
(GCMS) because we had authentic samples of the α-sulfone
disulfides that might form as products. In our earlier paper,[13]
a Hewlett–Packard 5988A instrument was employed to obtain
the mass spectra of several α-sulfone disulfides, including 1.
Although a variety of unambiguous criteria established that
1 was of good purity,[13] the HP instrument provided a GC
trace that showed a major contaminant (∼25%). The mass
spectrum/retention time of the contaminant matched the mass
spectrum/retention time of methyl p-tolylthiosulfonate 2 (see
Fig. 6).
When the reaction times for the first reaction in Scheme 6
were extended from 2 h → 24 h → 48 h → 96 h, equilibrium was
established (see Scheme 10).
It is now clear that α-sulfone disulfides have exophilic
(nucleofugal) sulfonyl groups that can be displaced by direct
nucleophilic attack at C(sp3). In an effort to extrapolate these
findings to possible sulfinate anion methylation by phenyl
methanesulfonate, we have established that no reaction occurs
under our reaction conditions (aqueous acetone), after a 24 h
reflux in aqueous ethanol or after a 24 h reflux in aqueous
dimethylformamide (DMF) (see Scheme 11).
Clearly, both favourable Coulombics and assistance from
the disulfide functionality were essential in bringing about the
successful substitutions shown in Schemes 6 and 10.
Hence, we concluded that the sulfone disulfide 1 undergoes
pyrolysis in the GC system that produced the thiosulfonate 2 (see
Scheme 5 for a proposed mechanism).
Not only does such a degradative process complicate the
examination of mixtures obtained from reactions on α-sulfone
disulfides such as 1 but the thiosulfonate 2 itself was a pos-
sible solution-phase reaction product, should sulfinate anions
carry out any nucleophilic attack at the disulfide linkage in anal-
ogy to that shown for the benzenethiol-derived anion implicit
in Scheme 4. When 1 was examined using the Varian CP-3800
GC equipped with a CP-8410 autoinjector connected to a Saturn
2000 mass selective detector, described earlier,[14] the thiosul-
fonate 2 was not produced as an artifact. Hence, the Varian
equipment was employed for this work.
A Possible Alternative Mechanism
A referee has suggested an alternative to the simple substitutions
at carbon entertained (see Fig. 3) for the displacement of ester
(Scheme 3) and sulfonyl groups (see Scheme 6) α to the disulfide
moiety. The referee’s proposal is depicted in Scheme 12.
Like the direct substitution at carbon, the Scheme 12 proposal
rationalizes the observation of some SS bond rupture (Schemes 6
and 7) in reactions that involve disulfides and p-toluenesulfinate
anions. Scheme 12 rationalizes the failure of the reactions in
Schemes 8 and 9. Unlike the SN2 proposal, Scheme 12 offers
no basis for anticipating the well-known success of elimination
reactions, in which sulfone sulfonyl groups are expelled and
The first successful displacements of alkyl/aryl sulfonyl
groups are presented in Scheme 6.

Displacement of RSO2 Groups from C(sp3)
469
2h
SO2CH2SSCH3 ꢀ CH3SO2Na
CH3
CH3SO2CH2SSCH3
Water
3 (6%)
Acetone
1
SO2CH2SSCH3
CH3
2h
1 (16%)
CH3SO2CH2SSCH3 ꢀ CH3
SO2Na
Water
Acetone
3
SO2SCH3
CH3
2 (3%)
Scheme 6.
24h
SO2Na
Sulfinate anions (hard bases) have little incentive to attack
PhSSCH3 ꢀ CH3
PhSSPh ꢀ PhSSCH3
Water
disulfide sulfenyl sulfur (soft acid). If the aryl substituent serves
as a modest electron releaser, it would soften p-toluenesulfinate
sulfur and facilitate attack at sulfenyl sulfur somewhat. Hence,
the success of toluenesulfinate anions and the apparent failure
of methanesulfinate anions to attack at sulfenyl sulfur leading
to formation of the thiosulfonte 2 and the absence of methyl
methanesulfonate (see Scheme 13). Nonetheless, in support
of a bimolecular reaction at carbon, methanesulfinate anions
are viable nucleophiles in the sulfonyl substitution shown in
Scheme 6.
(16%)
(84%)
Acetone
Scheme 7. Relative yields are shown.
24h
CH3SO2CH2S
SO2Na
CH3 ꢀ CH3
Water
Acetone
Scheme 8.
If the Scheme 12 mechanism did lead to significant amounts
of the α-sulfone disulfide products formed in these reactions,
it seems likely that some chain-extended products (see Fig. 7)
24h
SO2CH2SSCH3
CH3
1
would be observed. Neither high-field H NMR spectroscopy
Water
nor GCMS detected such products although both 4 and 5 are
Acetone
known to us.[14,16]
1
Thus, available evidence supports the view that the novel sul-
fonyl displacements, described herein, proceed, in the main, by
bimolecular substitution at carbon.
Scheme 9.
the puzzling dearth of examples of bimolecular displacement of
sulfone sulfonyl groups.
Consider the displacement of an ester group (propionate or
benzoate) α to a disulfide moiety. Reaction with sulfinate anions
(alkyl or aryl) is complete in 2 h at 50◦C (Scheme 3).[12,13]
In contrast, identical reaction times/conditions lead to only
∼10% displacement of a sulfonyl group α to a disulfide moiety
(Scheme6).ThefirsttwostepsarereminiscentofanE1cb mecha-
nism, in which proton abstraction from carbon has been replaced
by nucleophilic removal of sulfenyl sulfur from sulfenyl sulfur.
One would expect similar rates of nucleophilic attack at sulfenyl
sulfur in both esters and sulfones. It has been observed[15] that
elimination (probably E1cb) reactions of β-substituted phenones
proceed 118 times faster when phenylsulfonyl leaves than when
acetate leaves. Scheme 12 does not rationalize the much slower
displacements of sulfonyl groups that are expected for SN2
substitution at carbon.
When this work began, a good supply of the α-tosyl disulfide
1 was on hand. Only one experiment employed starting α-mesyl
disulfide 3 (Scheme 6). In contrast, four experiments (2 h, 24 h,
48 h, 96 h) employed starting α-tosyl disulfide 1. Despite the
strong bias in favour of its formation/observation in these reac-
tions between 1 and methanesulfinate anions, no CH3SO2SCH3
was observed (GCMS) in any reactions (see Scheme 13 for
example).
Conclusions
The first examples of nucleophilic attack that displace a sulfonyl
group from C(sp3) have been described. The deduction that such
reactions should be observable rests on (i) related experimental
results in which sulfone sulfonyl groups are expelled during E2
eliminations and (ii) modern molecular orbital computations that
describe the sulfonyl group as strongly polarized with very sub-
stantial partial charges on the heteroatoms and on the carbon
atoms attached to sulfonyl sulfur. Indeed, one could view our
results as an experimental test of the MO prediction that sub-
stantial negative charge is present at the C(sp3) atoms attached
to sulfonyl sulfur.The widespread practice of describing sulfonyl
groups in terms of dπ–pπ backbonding obscures the fundamen-
tal underlying polarization, inter alia, in sulfonyl groups. Earlier
computational studies,[17–19] that warn against heavy reliance
on d-orbitals in describing such groups, were most helpful in
exploring this chemistry.
Experimental
1H NMR (270 MHz) spectra were obtained on a JEOL JNM-
GSX 270 Fourier-transform NMR system.

470
S. Duffy and R. F. Langler
Acetone
CH3
SO2CH2SSCH3 ꢀ CH3SO2Na
SO2Na
CH3SO2CH2SSCH3 ꢀ CH3
Water
50°
96h
1
3
2:3
Scheme 10. Equilibrium position for reversible sulfonyl group displacements as determined by 270 MHz 1H NMR spectroscopy.
42 min. Helium gas was used as the carrier and the column flow
PhSO2CH3
PhSO2Na ꢀ CH3SO2OPh
rate was held constant at 1.0 mL min−1. On occasion, the GC
run was terminated before the final time was reached provided
that the analyte had eluted from the column. The mass selective
detector was operated in both electron impact (EI) and chemi-
cal (CH4) ionization (CI) modes. The mass range selected was,
typically, 50 to 500 amu. A minimum of three microscans were
averaged to produce each scan, which resulted in a data accu-
mulation rate of 0.5 s per scan for EI and 0.62 s per scan for
CI. The emission current was set at 30 µA. Percent yields were
calculated based on response factors for each of the identified
compounds, by comparing the integrated peak areas of standards
to the peak areas obtained from reaction products through the
use of the Saturn GC/MS workstation software.
Scheme 11.
XCH2SSCH3 ꢀ ArSO2Na
X ꢄ RSO2, RC(O)O
XCH2SNa ꢀ ArSO2SCH3
ꢁArSO2Na
ꢀArSO2Na
ArSO2CH2SNa
NaX ꢀ CH2S
ArSO2SCH3
Substitution Trials
Reactions were carried out as detailed for the representative
case below (first reaction, Scheme 6). The sulfone disulfide 1
(0.391 g, 1.6 mmol) and the sodium salt of methanesulfinic acid
(0.180 g, 1.8 mol) were covered with 1/4 water/acetone (5 mL)
and the reaction mixture was heated at 50◦C for 2 h.
Water (50 mL) was added and the resultant mixture extracted
with chloroform (3 × 50 mL aliquots). The organic layers were
combined, dried (MgSO4), filtered, and the solvent evaporated
furnishing unchanged sulfone disulfide 1 (1.2 mmol) and sul-
fone disulfide 3 (0.09 mmol, 6%). Quantitative analysis was
carried out by GCMS as described under the heading Yield
Determinations.
ArSO2CH2SSCH3 ꢀ ArSO2Na
Scheme 12. An alternative mechanism for sulfonyl substitutions adjacent
to an SS linkage.
24h
p-CH3(C6H4)SO2CH2
SSCH3 ꢀ CH3SO2Na
CH3SO2CH2SSCH3
Water
Acetone
3 (14%)
1
ꢀ CH3SO2SCH3
(0%)
The preparations and properties have been described earlier
ꢀ p-CH3(C6H4)SO2SCH3
2 (4%)
for key compounds 1,[13,20] 2,[21] and 3.[13]
Scheme 13.
Acknowledgements
The authors acknowledge financial support from Mount Allison University.
High-field NMR spectra were obtained by D. Durant.
RSO2CH2SCH2SSCH3
4 R ꢄ CH3
5 R ꢄ p-CH3(C6H4)SO2
References
[1] F. G. Bordwell, W. T. Brannen, J. Am. Chem. Soc. 1964, 86, 4645.
doi:10.1021/JA01075A025
Fig. 7. Structures for possible chain-extended products which were not
[2] L. A. Paquette, J. Am. Chem. Soc. 1964, 86, 4085. doi:10.1021/
JA01073A035
observed.
[3] F. G. Bordwell, B. Jarvis, J. Org. Chem. 1968, 33, 1182. doi:10.1021/
JO01267A054
Yield Determinations
Samples were weighed analytically and diluted to the low ppm
range before analysis.
[4] M. Cinquini, D. Landini, A. Maia, J. Chem. Soc. Chem. Commun.
1972, 734. doi:10.1039/C39720000734
[5] P. Robson, P. Speakman, D. Stewart, J. Chem. Soc. C 1968, 2180.
doi:10.1039/J39680002180
[6] W. G. Phillips, K. W. Ratts, J. Org. Chem. 1971, 36, 3145.
doi:10.1021/JO00820A016
[7] W. R. Hardstaff, R. F. Langler, J. Leahy, M. J. Newman, Can. J. Chem.
1975, 53, 2664. doi:10.1139/V75-377
[8] R. F. Langler, J. A. Pincock, Can. J. Chem. 1977, 55, 2316.
doi:10.1139/V77-318
[9] R. F. Langler, Sulfur Rep. 1996, 19, 1. doi:10.1080/
01961779608047904
Samples (1 µL) were analyzed using a Varian CP-3800
gas chromatograph (GC) equipped with a CP-8410 autoinjec-
tor, connected to a Saturn 2000 mass selective detector. The
GC oven contained a Varian 30 m × 0.25 mm i.d. × 0.25 µm
film thickness CP-Sil 8 CB low bleed/MS column. The GC
temperature program was as follows, injection temperature
250◦C, initial temperature 50◦C, hold for 2 min, ramp to
220◦C at +20◦C min−1, and hold for 2 min, ramp to 260◦C at
+5◦C min−1, and hold for 21.5 min for a total analysis time of

Displacement of RSO2 Groups from C(sp3)
471
[10] N. M. Georges, M. D. Johnson, R. F. Langler, S. D. Verma, Sulfur Lett.
1999, 22, 141.
[11] R. F. Langler, D. A. Ryan, S. D. Verma, Sulfur Lett. 2000, 24, 51.
[12] F. J. Baerlocher, M. O. Baerlocher, C. L. Chaulk, R. F. Langler,
E. M. O’Brien, Sulfur Lett. 2000, 24, 101.
[13] R. F. Langler, S. L. MacQuarrie, R. A. McNamara, P. E. O’Connor,
Aust. J. Chem. 1999, 52, 1119. doi:10.1071/CH99086
[14] S. A. Bewick, S. Duffy, S. P. Fletcher, R. F. Langler, H. G. Morrison,
E. M. O’Brien, C. Ross, V. C. Stephenson, Aust. J. Chem. 2005, 58,
218. doi:10.1071/CH03253
[16] C. G. Andrews, R. F. Langler, unpublished results.
[17] P. V. R. Schleyer, T. Clark, A. J. Kos, G. W. Spitznagel, C. Rohde,
D. Arad, K. N. Houk, N. G. Rondan, J. Am. Chem. Soc. 1984, 106,
6467. doi:10.1021/JA00334A001
[18] A. E. Reed, P. V. R. Schleyer, J. Am. Chem. Soc. 1990, 112, 1434.
doi:10.1021/JA00160A022
[19] E. Magnusson, J. Am. Chem. Soc. 1990, 112, 7940. doi:10.1021/
JA00178A014
[20] S. M. H. Kabir, R. F. Langler, Aust. J. Chem. 2005, 58, 362.
doi:10.1071/CH04170
[15] C. J. M. Stirling, Acc. Chem. Res. 1979, 12, 198. doi:10.1021/
AR50138A002
[21] R. J. Goodridge, T. W. Hambley, R. K. Haynes, J. Org. Chem. 1988,
53, 2881. doi:10.1021/JO00248A001