Date: Wed, 29 Sep 1999 21:36:24 -0500
Precedence: bulk
From: "susan lopez" 
Subject: POTW-Susan How Stable are Thiosulfoxides?
MIME-Version: 1.0
X-Priority: 3
X-MSMail-Priority: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300

How Unstable are Thiosulfoxides? An ab Initio MO study of  Various
Disulfanes RSSR (R= H, Me, Pr, All), Their Branched Isomers  R2SS and the
Related Transition States
Ralf Steudel*, Yana Drozdova, Karol Miaskiewicz, Roland H.  Hertwig, &
Wolfran Koch
J. Am. Chem. Soc. 1997, 119, 1990.
Goals of Paper: To study organic thiosulfoxides  R2SS, which play an
important role as reaction intermediates, yet  cannot be isolated or
directly observed. Using ab initio MO calculations, the  structures and
energies of these are obtained as well as the activation energies
necessary to form them from disulfanes (RSSR). The structures and energies
of  the transition states for the interconversion (RSSR <-> R2SS)  are
also discussed.
Methods: Geometries were optimized using Hartree-Fock and  MP2 level
(except Pr2S2 &  All2S2) and 6-31G* and 6-311G** basis sets. Vibrational
spectra info. was derived from force constant matricies and harmonic
frequencies  computed analytically for each stationary point. Additional
calculations on  H2S2 were done using much higher levels of theory to make
sure that the structures and relative energies were accurate. Briefly this
included: CASSCF level with contracted basis set ANO-S (atomic natural
orbital);  for energy calc. ANO-L basis set (larger) was used. Energies
were obtained via  CASPT2, MP4sdqt, and CCSD(T) [these are perturbation
thory levels?].
Results: HSSH<->H2SS and TS: Using the  higher level calc. the energy
difference between the two isomers was found to be  113-117 kJ/mol, yet
using 6-31G* or 6-311G** basis sets with MP2 or MP4 level  the difference
was found to be 141-145kJ/mol (in each of these the HSSH is the  more
stable species). This is an artifact of basis set choice, and the system
is  said to be "well-behaved"; enough that one can use the
smaller basis sets at MP2.  [30kJ/mol seems like a big enough E difference
to me- I' m not sure why the 143kJ/mol estimate is ok in this case?
Isn' t this a compelling argument to in fact use the higher level
rather that dismiss it?] The SS bond lengths suggest  double bond
character in H2SS. The saddle point (TS) is 210/67 kJ/mol above  these
isomers respectively, this suggests that at low enough T it should be
possible to detect H2SS using vibrational spectroscopy (Table 3).
MeSSMe<->Me2SS and TS: The MeSSMe isomer is only 84kJ/mol more  stable in
this case. The reason is that SS bonds are "semipolar"and the
(+)  charge on S1 in H2SS weakens the adjacent SH bonds (enlongating it),
yet in Me2SS the CS is shorter because C has (-0.4e) charge vs. H
(+0.1e); making the isomerization less E costly. The TS in this case is
340kJ/mol higher than MeSSMe due to the SS and CS(C in transit) bond
lengths  being longer and therefore weaker. For this reason a different
isomerization  pathway is proposed, instead of intramolecular CH3 shift,
ther is  homolysis of CS bond, migration of CH3 radical to the other
sulfur.  This is supported by previous bond dissociation work (Benson).
Pr2S2 and All2S2: The  Pr2SS system was studied to see if the stabilty
trend continues if  you lengthen the alkyl chain, and it does 74kJ/mol
less stable.  All2SS is 78 kJ/mol less stable than disulfane. The later
undergoes a  sigmatropic rearrangement (eg.3) leading to a very different
TS for which a  five-membered C3S2 ring structure is proposed that is
stabalized via solvent interactions. The E of the TS is 111/33 kJ/mol
higher  than the respective isomers. MeAllS2 and TS: The MeAllSS is
83kJ/mol  less stable than MeSSAll. Two TS were found; TS1 forms a S-S-C
triangle (like  TS(Me)) with an activation E of 298kJ/mol compared to
MeSSAll, and TS2 forms a  five membered ring TS (like TS of All2S2) that
is  110kJ/mol less stable than MeSSAll. Reactions: Me2SS ->  Me2S +S2 is
exothermic and spin forbidden fot triplet  S2, yet allowed for singlet S2
but endothermic.  Me2SS +Sx -> Me2S and Sx+1  (x>2) is exothermic.
Bonding: SS bond in R2SS is considered as either double or
"semipolar" depending on the electronegativity of the R. FSSF
for  example has very short SS bond length, due not only to the
electronegativity of  the F but also because hyperconjugation of the 3p
sulfur lone pairs are  partially deloccalized into the s* MO of the RS
bonds in same plane (gauche effect), this way  two pi bonds are created in
planes perpendicular to one another. The more EN the  R the lower the s*
energy and the stronger the p-bond formed. 
I chose this paper because it was relatively straight forward,  had quite
a bit of theory and dealt with relatively simple molecular species.  Best
of all it contained transition states for all of these molecules!