Kinetic and thermodynamic control in the synthesis of tetrahydro-pyrans and -furans from 1,4-diols by stereospecific phenylsulfanyl (PhS) migration: Competition between exo and endo transition states and between [1,2] and [1,4] sulfanyl participation

Article abstract of DOI:10.1055/s-1999-2814

The factors controlling the cyclisation of 1,4-diols with PhS migration to give THPs rather than THFs are reassessed. Furthermore we provide evidence for a competition between [1,2] and [1,4] sulfanyl participation and cyclisation onto a five membered cyclic sulfonium salt.

Full text of DOI:10.1055/s-1999-2814

LETTER
1211
Kinetic and Thermodynamic Control in the Synthesis of Tetrahydro-Pyrans
and -Furans from 1,4-Diols by Stereospecific Phenylsulfanyl (PhS) Migration:
Competition Between exo and endo Transition States and between [1,2] and
[1,4] Sulfanyl Participation
1
2
Jason Eames, Nikolai Kuhnert, Francis H. Sansbury, Stuart Warren*
University Chemical Laboratory, Lensfield Road, Cambridge, CB2 1EW, England
Fax (01223) 336913; e-mail sw134@cam.ac.uk
Received 3 June 1999
We have shown that the cyclisation of the 1,3-diols under
Abstract: The factors controlling the cyclisation of 1,4-diols with
PhS migration to give THPs rather than THFs are reassessed. Fur-
thermore we provide evidence for a competition between [1,2] and
[1,4] sulfanyl participation and cyclisation onto a five membered
cyclic sulfonium salt.
these conditions is under thermodynamic control and that
in the absence of acid the sulfonium ion 5, made from cy-
clic sulfites, gives 20-30% of oxetanes 6 as kinetic prod-
ucts by the 4-exo-tet cyclisation.5
Key words: cyclisations, heterocycles, thermodynamic and kinetic
control, sulfur chemistry, rearrangements
Competition between the 4-exo-tet cyclisation and the ki-
netically less favoured hybrid 6-endo/5-exo-tet cyclisa-
tion to give the THF 2 is probably biased because the
oxetane anti-6, and the transition state leading to it, are de-
stabilised by ring strain. A more equitable competition
arises in the cyclisation of 1,4-diols 3 where neither prod-
uct is strained and we might expect the pure 5-exo-tet cy-
clisation to be more kinetically favoured. This paper
concerns thermodynamic versus kinetic control in the cy-
clisation of 1,4-diols such as 3. Additionally we provide
evidence for a competition between [1,2] and [1,4] sulfa-
nyl participation when a second sulfide is present in the
molecule undergoing cyclisation.
The acid-catalysed cyclisation of 1,3-diols3 such as 1 and
1,4-diols4 such as 3 occurs by attack of the primary hy-
droxy group at the more substituted end of the intermedi-
ate sulfonium ion 5 or 7 leading to THFs 2 or THPs 4
respectively in spite of the partial endo nature (by Bald-
win’s rules5,6) of this mode of attack. Nevertheless there is
no doubt that the reaction takes place by a concerted sub-
stitution pathway since clean inversion at a chiral migra-
tion origin is always observed.3,4
In order to assess the relative thermodynamic stabilities of
the cyclisation products of 1,4-diols, THF and THP re-
spectively, we had to synthesise both compounds inde-
pendently and study their interconversion under the
rearrangement conditions. The potentially kinetic product
THF 8a which would result from a 5-exo-tet cyclisation of
7 has never been observed under acidic conditions, but
can be made in base (TsCl, pyridine) from the same diol.4a
We chose to study three series of carbocyclic and hetero-
cyclic six-membered rings based on 9a–c; X = CH2, O,
and S. The synthesis of the required 1,4-diols was
straightforward. Yields were generally excellent and giv-
en in Table 1.
PhS
OH
PhS
TsOH
OH
CH2Cl2
O
anti-1
anti-2, 100%
PhS
O
PhS
OH
X
syn-5
oxetane anti-6
from 4-exo-tet cyclisation
PhS
OH
PhS
PhS
TsOH
OH
OH
OH
Li
CHO
O
OEt
CH2Cl2
O
PhS
PhS
3
O
OEt
3
3a
4a, 100%
X
X
9 a-c
HCl
EtOH
10 a-c
PhS
PhS
OH
O
a: X = CH2
b: X = O
c: X = S
X
OH
X
7
THF 8a from
3 a-c
5-exo-tet cyclisation
Scheme 1
Scheme 2
Synlett 1999, No. 8, 1211–1214 ISSN 0936-5214 © Thieme Stuttgart · New York

1212
J. Eames et al.
LETTER
Acid catalysed rearrangements of the diols 3a–c gave ex-
clusively the THPs 4a–c. Compound 3c needed consider-
ably longer reaction times (11 hours compared with 5 min
for 3a and 3b). The THFs 8a–c (the alternative cyclisation
products) were obtained by direct ether formation (treat-
ment of diols 3a–c with TsCl/pyr via the primary toluene-
p-sulfonates 11a–c). Subjection of these THFs 8a-c, to
acid catalysis gave the THPs 4a–c again in almost quanti-
tative yields.
O
1. Et3N
CH2Cl2
S
PhS
O
–SO2
O
PhS
3 a-c
2. SOCl2
O
X
X
13 a-c
12 a-c
PhS
PhS
X = CH2 4a:8a
33:67
2:98
+
X = O
X = S
4b:8b 26:74
4c:8c
O
O
X
X
PhS
PhS
TsOH
TsOH
3 a-c
O
Scheme 4
CH2Cl2
CH2Cl2
O
X
X
4 a-c
8 a-c
(favoured by Baldwin’s rules) to give the THFs 8a, b as
the major kinetic products. We suggest that the ratios of
67:33 and 74:26 (THF:THP) give the approximate ratios
of the kinetic products of cyclisation of diols 3a, b.
OTs
PhS
TsCl
pyr
OH
11 a-c
3 a-c
X
Table 1: Yields of starting materials, rearranged THPs and unrearran-
Scheme 3
ged THFs
It is clear that the THPs 4a–c are the thermodynamic prod-
ucts from the acid catalysed rearrangements. The question
still remained whether the THP or the THF would be the
kinetic product of the reaction. We therefore carried out
the rearrangement in the absence of acid to prevent rear-
rangement of the THFs.5 The required cyclic sulfites 12a–
c were synthesised from the diols 13a–c with SOCl2 in ba-
sic solution. They decomposed on heating with loss of
SO2 to give the THPs 4a–c and THFs 8a–c in ratios of (X
= CH2) 33:67, (X = O) 26:74 and (X = S) 2:98. Presum-
ably both sulfites 12a, b decompose via the episulfonium
ion 13a–c which is captured by a 5-exo-tet cyclisation
ayields and reaction conditions are identical for either synthesis from
1,4-diols 3a–c or THFs 8a–c.
We can summarise the regioselectivities in cyclisation
onto the two classes of episulfonium ions in the diagrams
below for X = CH2. All four classes of heterocycles (2, 4,
6 and 8) can be made in good yield either by acid-cataly-
sed cyclisation for the spirocyclic compounds or by spe-
cial methods for the others. The exo:endo selectivity is
Ziram (Me2NCS2)2Zn, Ph3P, DEAD, see ref 3
O
PhS
oxetane
anti-6
20%
80%
OH
4-exo-tet
OH
PhS
PhS
TsOH
TsOH,
OH
CH2Cl2
CH2Cl2
PhS
PhS
anti-1
6-endo-tet/5-exo-tet
THF
anti-2,
100%
O
TsOH, CH2Cl2, 100%
TsCl, pyridine, 89%
THF
8a
O
67%
5-exo-tet
OH
PhS
PhS
OH
TsOH,
CH2Cl2
TsOH
OH
CH2C2
33%
PhS
3a
7-endo-tet/6-exo-tet
THP
4a
O
100%
TsOH, CH2Cl2, 100%
Scheme 5
Synlett 1999, No. 8, 1211–1214 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
Kinetic and Thermodynamic Control in the Synthesis of Tetrahydro-Pyrans and -Furans
1213
greater with the larger ring both because of the relaxation
of Baldwin’s rules and because the THF 8 is less strained
than the oxetane 6. Series b with X = O shows the same
tendency.7
PhS
PhS
PhS
TsCl
pyr
OH
a
OTs
[1,2] SPh
Cl
S
S
S
18
16
PhS
17
The very high exo selectivity in the cyclisation of the sul-
fide 11c X=S may be explained by [1,4] SPh participation.
Eliel8 has shown that [1,2] and [1,4] SR participation are
about equally efficient, but since a rather stable bicyclic
sulfonium ion 15 is formed,9 [1,4] participation 14 may be
favoured in this case. Capture by the tethered alkoxide in
15 can lead only to THF 8c. The THP can come only from
competing [1,2]-SPh participation via episulfonium ion
13c. Apparently the direct interconversion between the bi-
cyclic sulfonium ion 15 and the episulfonium ion 13
would involve a very strained transition state and does not
occur. By increasing the time of reflux for the acid catal-
ysed rearrangement of THF 8c (TsOH in CH2Cl2 gives
THF:THP ratios of 90:10 after 10 min; 64:36 after 15 min;
38:62 after 30 min; 10:90 after 5h and 0:100 after 11h) we
realise thermodynamic control. It is worth noting that
THF 8c is the kinetic product of both the [1,2]-SPh partic-
ipation and the [1,4]-SR participation.
PhS
OTs
Cl [1,4] SR
S
S
b
17a
Cl
20
b
a
PhS
PhS
Cl
Cl
S
S
21
19
Scheme 7
OH
PhS
PhS
TsOH
SH
SH
CH2Cl2
23
22
5-exo-tet
O
PhS
O
O
S
O
O
O
S
PhS
PhS
PhS
S
S
S
12 c
14
-SO2
O
24; 99% yield
PhS
Scheme 8
O
S
S
8 c
In conclusion we have shown that:
15
1) The rearranged THPs from the hybrid 7-endo/6-exo-tet
cyclisations of 1,4-diols 3a–c leading to 4a–c, are the
thermodynamic products of the acid catalysed rearrange-
ment.
Scheme 6
2) The pure 5-exo-tet cyclisation to give THFs, favoured
by Baldwin’s rules, is responsible for 60-100% of the ki-
netic product depending on the substitution.
Further evidence for competitive [1,2] and [1,4]-SR par-
ticipation came from rearrangement of alcohol 16 (ob-
tained from aldehyde 9c by LiAlH4 reduction) with TsCl
in pyridine to give an inseparable mixture of isomeric
chlorides 19 (by path a) and 21 (by path b) in a ratio of
52:48. The bicyclic sulfonium ion 20 from [1,4]-SR par-
ticipation can be captured by chloride anion to give either
19 or 21, whereas the spirocyclic sulfonium ion 18 from
[1,2] participation can give only the unrearranged chloride
19.
3) The THFs 8 rearrange rapidly under the reaction condi-
tions to give the thermodynamic products 4.
4) [1,4]-SR participation can compete with the usual
[1,2]-SR participation.
Acknowledgement
We thank EPSRC for a grant (to J. E.) and RTL and DFG (Deutsche
Forschungsgemeinschaft) for grants (to N. K.).
Related thiols 22 and sulfonamides with a 1,4-relationship
between OH and SH or NHTs rearrange under acid catal-
ysis to give only products, e.g. 24, from 5-exo-tet opening,
e.g. 23, of the sulfonium ion.10 All new compounds gave
satisfactory 1H and 13C NMR spectra and HRMS data.11
References and Notes
(1) New address: Department of Chemistry, Queen Mary and
Westfield College, University of London, London E1 4NS.
(2) New address: Department of Chemistry, University of Surrey,
Guildford GU2 5XH, England.
Synlett 1999, No. 8, 1211–1214 ISSN 0936-5214 © Thieme Stuttgart · New York

1214
J. Eames et al.
LETTER
(3) (a) Aggarwal, V. K.; Coldham, I.; McIntyre, S.; Sansbury, F.
H.; Villa, M.-J.; Warren, S. Tetrahedron Lett. 1988, 29, 4885;
(b) Aggarwal, V. K.; Coldham, I.; McIntyre, S.; Warren, S. J.
Chem. Soc., Perkin Trans. 1 1991, 451; (c) Chibale, K.;
Warren, S. J. Chem. Soc., Perkin Trans. 1 1995, 2411.
(4) (a) Djakovitch, L.; Eames, J.; Jones, R. V. H.; McIntyre, S.;
Warren, S. Tetrahedron Lett. 1995, 36, 1723; (b) Bird, P.;
Eames, J.; Jones, R. V. H.; Roddis, M.; Sturino, C. F.;
O’Sullivan, S.; Warren, S.; Westwell, M. S.; Worrall, J.
Tetrahedron Lett. 1995, 36, 1909; (c) Sansbury F.H. and
Warren S., Tetrahedron Lett. 1991, 32, 3425.
(5) Eames, J.; Warren, S. Tetrahedron Lett. 1996, 37, 3525.
(6) Baldwin, J. E. J. Chem. Soc. Chem. Commun, 1976, 734.
(7) The slightly higher THF:THP ratio in the rearrangement of 3b
(X=O) in comparison with 3a (X=CH2) might suggest that
there is some [1,4]-OR participation but this slight difference
is not conclusive. For [1,4]-OR participation see: (a) Allred, E.
L.; Winstein, S. J. Am. Chem. Soc. 1967, 89, 4008; (b) Martin,
O. R.; Yang, F.; Xie, F. Tetrahedron Lett. 1995, 36, 47.
(8) Eliel, E. L.; Pearson, W. H.; Jewell, L. M.; Abatjoglou, A. G.;
Kenan, W. R. Tetrahedron Lett. 1980, 21, 331.
TsCl (52 mg, 0.25 mmol) was added to a solution of diol 3a
(70 mg, 0.25 mmol) in pyridine (1 mL). The solution was
stirred overnight. Ether (20 mL) was added, the solution was
washed with HCl (10 ml, 3 M) and evaporated under reduced
pressure. The residue was purified by column
chromatography on silica gel eluting with light petroleum(40-
60 oC)-ether (9:1) to give the tetrahydrofuran 8a (65 mg, 98
%) as an oil; Rf [light petroleum (40-60 oC)-ether (9:1)] 0.2;
vmax (film, CDCl3)/cm-1 1600 (SPh); dH(400MHz CDCl3)
7.56-7.24 (5 H, m, SPh), 3.89-3.82 (1 H, dt, J 6.67 and 6.68,
OCHAHB), 3.72-3.70 (2 H, m, OCHBHA and CHO) and 2.08-
1.18 (14 H, m, 7 x CH2); dC(100MHz CDCl3) 137.3 (m-SPh),
131.8 (i-SPh), 128.4 (o- and p-SPh), 84.4 (CHO), 68.7
(CH2O), 56.6 (CSPh), 32.1, 30.2, 26.6, 26.3, 26.0, 21.8 and
21.7 (7 x CH2) (Found M+, 262.1389. C16H22OS requires M,
262.1391); m/z 262.1 (30%, M), 191.1 (100, C6H10SPh),
153.1 (90, M — SPh), 123.0 (20, CH2SPh), 81.1 (65, C6H9)
and 71.1 (65, M — C6H10SPh).
1-[(1’-Phenylsulfanyl)cyclohexyl]-butane-1,4-sulfite 12a.
Thionyl chloride (21 mg, 13.2 ml, 0.17 mmol) was added to a
solution of diol 3a (50 mg, 0.17 mmol) and Et3N (36 mg, 48
ml, 0.35 mmol) in CH2Cl2 (1 mL). The reaction was stirred for
5 min. Saturated NH4Cl (1 mL) was added and the solution
was extracted with ether (3 x 10 ml). The combined organic
extracts were dried (MgSO4) and evaporated under reduced
pressure. The residue was purified by column
(9) Similar sulfonium ions have been isolated some time ago:
Prelog, V.; Cerkovnikov E. Justus Liebigs Ann. Chem., 1939,
537, 214.
(10) (a) Eames, J.; Jones, R. V. H.; Warren, S. Tetrahedron Lett.,
1996, 37, 707; (b) Coldham, I.; Warren, S. J. Chem. Soc.,
Perkin Trans. 1, 1993, 1637.
chromatography on silica gel eluting with light petroleum (40-
60 oC)-ether (9:1) to give the sulfite 12a; (41.2 mg, 96 %) as
an oil; Rf [light petroleum(40-60 oC)-ether (9:1)] 0.2; vmax
(film, CDCl3)/cm-1 1550 (SPh); dH(400MHz, CDCl3) 7.56-
7.28 (5 H, m, SPh), 4.78 (1 H, d, J 10.56, CHO), 4.18-4.02 (2
H, m, CH2O), 2.51 (1 H, dt, J 15.14 and 3.63, CHAHBCH2)
and 1.98-1.21 (13 H, m, CHAHBCH2 and 5 x CH2);
(11) Selected experimental and spectroscopic details.
5-(Phenylsulfanyl)-1-oxaspiro[5.5]undecane 4a;
TsOH (3 mg, 17 mmol) was added to a solution of diol 3a (25
mg, 89 mmol) in CH2Cl2 (2 ml). The solution was refluxed for
5 min, cooled to r.t. and filtered through a silica plug using
CH2Cl2 (5 ml). The solvent was removed under reduced
pressure. The residue was purified by column
dC(100MHz, CDCl3) 137.7 (m-SPh), 130.6 (i-SPh), 129.2 (p-
SPh), 128.9 (o-SPh), 79.5 (CHO), 65.5 (CH2O), 50.8 (CSPh),
30.8, 30.7, 29.6, 27.5, 26.3, 21.8 and 21.7 (7 x CH2) (Found
M+, 326.1010. C16H22O3S2 requires M, 326.1010); m/z 326.1
(80%, M), 262.1 (70, M — SO2), 191.1 (100, C6H10SPh) and
109.0 (50, PhS).
chromatography on silica gel eluting with light petroleum (40-
60 oC)-ether (9:1) to give the tetrahydropyran 4a (23.1 mg, 99
%) as an oil; Rf [light petroleum (40-60 oC)-ether (9:1)] 0.4;
v
max (film, CDCl3)/cm-1 1600 (SPh); dH(400MHz, CDCl3)
7.49-7.17 (5 H, m, SPh), 3.66-3.56 (2 H, m, CH2O), 3.03 (1 H,
dd, J 11.17 and 4.26, CHSPh) and 2.22-1.05 (14 H, m, 7 x
CH2); dC(62.5MHz, CDCl3) 136.1 (i-SPh), 13.5 (m-SPh),
128.9 (p-SPh), 126.6 (o-SPh), 75.4 (CO), 59.9 (CH2O), 55.4
(CHSPh), 36.2, 27.1, 26.4, 25.9, 21.2 and 20.5 (6 x CH2)
(Found M+, 262.1395. C16H22OS requires M, 262.1391); m/z
262.1 (25%, M), 165.1 (100, C4H8SPh), 136.0 (80, C2H3SPh)
and 109.0 (5, PhS).
By 1H NMR the sulfite 12a; X=CH2 decomposed over 7 days
to give the tetrahydropyran 4a and tetrahydrofuran 8a (67:33)
as an inseparable mixture, identical spectroscopically to that
obtained previously.
Article Identifier:
1437-2096,E;1999,0,08,1211,1214,ftx,en;L14599ST.pdf
2-[(1’-Phenylsulfanyl)cyclohexyl]tetrahydrofuran 8a;
Synlett 1999, No. 8, 1211–1214 ISSN 0936-5214 © Thieme Stuttgart · New York