C-glycosylation reactions of sulfur-substituted glycosyl donors: Evidence against the role of neighboring-group participation

Article abstract of DOI:10.1021/ja0767783

Nucleophilic substitution reactions of C-4 sulfur-substituted tetrahydropyran acetals revealed that neighboring-group participation does not control product formation. Spectroscopic evidence for the formation of an intermediate sulfonium ion is provided,

Full text of DOI:10.1021/ja0767783

Published on Web 01/24/2008
C-Glycosylation Reactions of Sulfur-Substituted Glycosyl
Donors: Evidence against the Role of Neighboring-Group
Participation
Matthew G. Beaver, Susan B. Billings, and K. A. Woerpel*
Department of Chemistry, UniVersity of California, IrVine, California 92697-2025
Received September 7, 2007; E-mail: kwoerpel@uci.edu
Abstract: Nucleophilic substitution reactions of C-4 sulfur-substituted tetrahydropyran acetals revealed that
neighboring-group participation does not control product formation. Spectroscopic evidence for the formation
of an intermediate sulfonium ion is provided, as are data from nucleophilic substitution reactions
demonstrating that products are formed from oxocarbenium ion intermediates. The selectivity was not
sensitive to solvent or to which Lewis acid was employed. The identity of the heteroatom at the C-4 position
also did not significantly impact diastereoselectivity. Consequently, neighboring-group participation was
not responsible for the formation of either the major or the minor products. These studies implicate a Curtin-
Hammett kinetic scenario in which the formation of a low-energy intermediate does not necessitate its
involvement in the product-forming pathway.
Scheme 1. SN2 Pathway Postulated to Account for Trans Product
Formation
The use of neighboring-group participation to control stere-
ochemistry has proven to be an effective strategy for a number
of transformations.1-6 Carbohydrate chemistry, in particular, has
benefited from the application of anchimeric assistance in
glycosylation reactions.7,8 Sulfur-,9,10 iodine-,11,12 and acetoxy-
substituted5,13 glycosyl donors undergo highly trans-selective
reactions, which have been postulated to involve nucleophilic
ring-opening reactions (SN2) of onium ion intermediates (Scheme
1).14 For example, Boons and co-workers recently demonstrated
the utility of neighboring-group participation in their elegant
method for the synthesis of R-glycosides.15 A protecting group
containing a sulfur moiety was installed at the C-2 hydroxyl
group, and upon activation, the derived sulfonium ion interme-
diate was observed spectroscopically.15 The stereochemical
outcome of subsequent nucleophilic substitution reactions was
consistent with an SN2 pathway involving this intermediate.15,16
In this paper, we provide spectroscopic evidence for the
formation of an analogous sulfonium ion intermediate in a
tetrahydropyran system and demonstrate that the products are
not formed by an SN2 pathway. Instead, selectivity is the result
of a Curtin-Hammett kinetic scenario17 in which nucleophilic
additions to high-energy oxocarbenium ion intermediates occur
through overall lower energy pathways.18 This analysis is
consistent with previous experimental19-21 and computational22-24
studies demonstrating that the reactions of acetals bearing
heteroatoms capable of neighboring-group participation might
react through open cations. As was clearly demonstrated by
Halpern’s mechanistic studies of the asymmetric hydrogenation
(1) Rajender, A.; Gais, H.-J. Org. Lett. 2007, 9, 579-582.
(2) Liu, L.-X.; Huang, P.-Q. Tetrahedron: Asymmetry 2006, 17, 3265-3272.
(3) Judd, W. R.; Ban, S.; Aube´, J. J. Am. Chem. Soc. 2006, 128, 13736-
13741.
(4) Smoot, J. T.; Pornsuriyasak, P.; Demchenko, A. V. Angew. Chem., Int.
Ed. 2005, 44, 7123-7126.
(5) Kim, J.-H.; Yang, H.; Boons, G.-J. Angew. Chem., Int. Ed. 2005, 44, 947-
(17) Seeman, J. I. Chem. ReV. 1983, 83, 83-134.
949.
(18) Substitution reactions with inversion of stereochemistry at anomeric centers
have transition-state structures with significant oxocarbenium ion charac-
ter: (a) Crich, D.; Chandrasekera, N. S. Angew. Chem., Int. Ed. 2004, 43,
5386-5389. (b) El-Badri, M. H.; Willenbring, D.; Tantillo, D. J.; Gervay-
Hague, J. J. Org. Chem. 2007, 72, 4663-4672.
(6) Converso, A.; Saaidi, P.-L.; Sharpless, K. B.; Finn, M. G. J. Org. Chem.
2004, 69, 7336-7339.
(7) Marzabadi, C. H.; Franck, R. W. Tetrahedron 2000, 56, 8385-8417.
(8) Toshima, K.; Tatsuta, K. Chem. ReV. 1993, 93, 1503-1531.
(9) Roush, W. R.; Sebesta, D. P.; Bennett, C. E. Tetrahedron 1997, 53, 8825-
8836.
(19) White, J. M.; Lambert, J. B.; Spiniello, M.; Jones, S. A.; Gable, R. W.
Chem. Eur. J. 2002, 8, 2799-2811.
(10) Nicolaou, K. C.; Rodr´ıguez, R. M.; Mitchell, H. J.; Suzuki, H.; Fylaktakidou,
K. C.; Baudoin, O.; van Delft, F. L. Chem. Eur. J. 2000, 6, 3095-3115.
(11) Roush, W. R.; Bennett, C. E. J. Am. Chem. Soc. 1999, 121, 3541-3542.
(12) Chong, P. Y.; Roush, W. R. Org. Lett. 2002, 4, 4523-4526.
(13) Fan, E.; Shi, W.; Lowary, T. L. J. Org. Chem. 2007, 72, 2917-2928.
(14) Smoliakova, I. P. Curr. Org. Chem. 2000, 4, 589-608.
(15) Kim, J.-H.; Yang, H.; Park, J.; Boons, G.-J. J. Am. Chem. Soc. 2005, 127,
12090-12097.
(20) Billings, S. B.; Woerpel, K. A. J. Org. Chem. 2006, 71, 5171-5178.
(21) Lazareva, M. I.; Kryschenko, Y. K.; Caple, R.; Wakefield, D.; Hayford,
A.; Smit, W. A.; Shashkov, A. S. Tetrahedron Lett. 1998, 39, 8787-8790.
(22) Dudley, T. J.; Smoliakova, I. P.; Hoffmann, M. R. J. Org. Chem. 1999,
64, 1247-1253.
(23) (a) Bravo, F.; Viso, A.; Alca´zar, E.; Molas, P.; Bo, C.; Castillo´n, S. J.
Org. Chem. 2003, 68, 686-691. (b) For related experimental studies, see:
Boutureira, O.; Rodr´ıguez, M. A.; Benito, D.; Matheu, M. I.; D´ıaz, Y.;
Castillo´n, S. Eur. J. Org. Chem. 2007, 3564-3572.
(16) Exogenous sulfide additives also formed observable sulfonium ion inter-
mediates that led to products consistent with an SN2 pathway: Park, J.;
Kawatkar, S.; Kim, J.-H.; Boons, G.-J. Org. Lett. 2007, 9, 1959-1962.
(24) Nukada, T.; Berces, A.; Zgierski, M. Z.; Whitfield, D. M. J. Am. Chem.
Soc. 1998, 120, 13291-13295.
9
2082
J. AM. CHEM. SOC. 2008, 130, 2082-2086
10.1021/ja0767783 CCC: $40.75 © 2008 American Chemical Society

Reactions of Sulfur-Substituted Glycosyl Donors
A R T I C L E S
Scheme 2. Model System to Determine Mode of Stereocontrol
allyltrimethylsilane in the presence of BF3‚OEt2 provided 1,4-
cis products 10 preferentially (eq 1). The moderate cis-selectivity
is consistent with previously published data on 4-iodine-
substituted acetals, which indicated that iodonium ion 8 (X )
I, Scheme 2) was not a reactive conformer.26 The formation of
cis products 10 demonstrated that sulfonium ions 8 (X ) SR)
could not be the reactive conformers in these reactions.31 Instead,
the major products arose from addition to equatorially substituted
oxocarbenium ions 6 (X ) SR).
of olefins,25 the fact that a low-energy intermediate can be
structurally characterized does not establish that the intermediate
is involved in product formation. For C-glycosylation reactions,
it is clear that onium ion intermediates can be formed, but they
need not be responsible for product formation.
compound
X
cis:trans
yield (%)
9a
SEt
SPh
I
84:16
78:22
72:28
75
87
90
9b
To understand the role of neighboring-group participation
involving sulfur-substituted glycosyl donors, a model system
was designed to differentiate between stereoselective addition
to an oxocarbenium ion and stereospecific ring-opening of a
bridged intermediate. Nucleophilic substitution reactions of the
4-heteroatom-substituted acetal 3 would illuminate which of
these pathways was preferred.26 Activation of acetal 3 with a
Lewis acid would give rise to a cation that could adopt three
forms: the equatorial half-chair oxocarbenium ion 6, the axial
half-chair conformer 7, and the bridged intermediate 8 resulting
from heteroatom assistance. Nucleophilic addition to the ste-
reoelectronically preferred face27 of equatorial conformer 6
would afford the cis product 4. Alternatively, addition to axial
conformer 7 or ring-opening of sulfonium ion 8 with inversion
would provide the trans product 5 (Scheme 2). Selective
formation of the cis product 4, therefore, would provide evidence
against heteroatom assistance in the stereochemistry-determining
step. Formation of the trans product 5 would require additional
experiments to differentiate between the remaining two reaction
pathways.
9c26
The reactions of acetals 9a,b could lead to erroneous
conclusions regarding the plausibility of nucleophilic substitution
reactions via bridged intermediates. The proposed sulfonium
ion arising from acetal 9 in the presence of a Lewis acid may
be destabilized by the ring strain inherent to the [2.2.1] bridged
bicyclic ring system.32 Computational studies (B3LYP/6-31G*),
however, indicate that sulfonium ion 14 is g5.5 kcal/mol lower
in energy than oxocarbenium ions 12 and 13 (Figure 1).33 This
result is consistent with data regarding the greater stability of
sulfonium ion intermediates as compared to their corresponding
oxonium ions, although ring strain appears to attenuate this
preference.34-36 The calculated energies of the oxocarbenium
ion intermediates (12 and 13), however, are not consistent with
the diastereoselectivities observed in the reactions involving the
addition of allyltrimethylsilane (eq 1). Either the sulfonium ion
14 was not formed, or it was not a direct precursor to the
products.
Details of the experimental design deserve mention prior to
discussing the results of nucleophilic substitution reactions. In
all cases, anomeric acetates were employed as oxocarbenium
ion precursors.28 Although the methyl acetals provided com-
parable diastereomeric ratios upon nucleophilic substitution, their
reactions did not proceed to completion. Unless otherwise noted,
allyltrimethylsilane was employed as the nucleophile to mini-
mize steric destabilization upon nucleophile approach and
because nucleophilic attack is irreversible.29,30 Diastereomer
ratios were obtained from unpurified reaction mixtures, and
product stereochemistry was determined by analysis of both 1H
NMR coupling constant data and NOE measurements of the
purified products.
Figure 1. Calculated (B3LYP/6-31G*) enthalpies of 4-SMe-substituted
oxocarbenium ion conformers.
Relieving the ring strain of the [2.2.1] ring system should
provide access to a lower energy sulfonium ion intermediate,
(31) Attempts to characterize an intermediate [2.2.1]-bridged bicyclic sulfonium
ion 8 (X ) SEt) by NMR spectroscopy were inconclusive.
(32) The analogous [2.2.1]-bridged carbocyclic ring system has a strain value
of 14.4 kcal/mol: Wiberg, K. B. Angew. Chem., Int. Ed. Engl. 1986, 25,
312-322.
Initial studies demonstrated that the major products of
nucleophilic addition reactions to sulfur-substituted acetals 9a,b
(3, X ) SEt, SPh) did not arise from an SN2 pathway involving
a sulfonium ion intermediate. Treatment of acetals 9 with
(33) Both stereoisomers of the bridged sulfonium ions 14 and 17 were found to
be minima, but only the lowest energy structure is shown. The details of
computational studies are provided as Supporting Information.
(34) Smith, B. J. J. Phys. Chem. A 1998, 102, 4728-4733.
(35) Smith (ref 34) demonstrated in a 1-SMe substituted pyran ring system that
the protonated sulfonium ion form is favored enthalpically (∆H°) by 22
kcal/mol as compared to the dissociated oxocarbenium ion and methanethiol.
Taking into account 14 kcal/mol ring strain for the [2.2.1] bicyclic system,
the sulfonium ion 14 should be favored by 8 kcal/mol, in good agreement
with computational results (Figure 1).
(36) Other strained sulfonium ions have previously been observed. See, for
example: (a) Aggarwal, V. K.; Charmant, J.; Dudin, L.; Porcelloni, M.;
Richardson, J. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 5467-5471. (b)
Huang, X.; Batchelor, R. J.; Einstein, F. W. B.; Bennet, A. J. J. Org. Chem.
1994, 59, 7108-7116. (c) Wilder, P., Jr.; Feliu-Otero, L. A. J. Org. Chem.
1966, 31, 4264-4265.
(25) Halpern, J. Science 1982, 217, 401-407.
(26) Ayala, L.; Lucero, C. G.; Romero, J. A. C.; Tabacco, S. A.; Woerpel, K.
A. J. Am. Chem. Soc. 2003, 125, 15521-15528.
(27) Deslongchamps, P. Stereoelectronic Effects in Organic Chemistry; Perga-
mon: New York, 1983; pp 209-221.
(28) Acetals 9, 18, and 24 were prepared from 3,4-dihydro-2-methoxypyran.
Details of these syntheses are provided as Supporting Information.
(29) Bear, T. J.; Shaw, J. T.; Woerpel, K. A. J. Org. Chem. 2002, 67, 2056-
2064.
(30) Hagen, G.; Mayr, H. J. Am. Chem. Soc. 1991, 113, 4954-4961.
9
J. AM. CHEM. SOC. VOL. 130, NO. 6, 2008 2083

A R T I C L E S
Beaver et al.
Scheme 3. Spectroscopic Evidence for the Formation of
Sulfonium Ion 21
Figure 2. Calculated (B3LYP/6-31G*) enthalpies of 4-CH2SR-substituted
oxocarbenium ion conformers.
and, consequently, more product derived from this intermediate
should be obtained. Extension of the sulfonium ion bridge of
14 by one methylene unit should alleviate ring strain. On the
basis of the analysis of ring strain values of analogous
carbocyclic systems, the [2.2.2]-bridged bicyclic sulfonium ion
17a should be less strained by 7 kcal/mol as compared to the
[2.2.1] system.32 Calculations (B3LYP/6-31G*) support this
hypothesis: the [2.2.2] bridged sulfonium ion 17a was calculated
to be 12.4 kcal/mol lower in energy than the equatorially
substituted oxocarbenium ion half-chair conformer 15a (Figure
2), and thus approximately 7 kcal/mol less strained than the
[2.2.1]-bridged bicyclic sulfonium ion 14 (Figure 1).33 Data for
the thiophenyl-substituted cation show a similar preference for
the bridged sulfonium ion 17b.
Nucleophilic substitution reactions of acetate 18 indicate that
the [2.2.2]-bicyclic sulfonium ion analogous to 17 is not the
reactive conformer. Treatment of 4-CH2SEt-substituted acetate
18 with allyltrimethylsilane and BF3‚OEt2 provided 1,4-cis
product 19 as the major diastereomer upon warming to room
temperature (eq 2). The cis product 19 could not arise from
stereospecific ring-opening of sulfonium ion 17. Instead, addi-
tion to the equatorially substituted oxocarbenium ion half-chair
conformer 15 (R ) Et) would provide the observed product.
treatment of acetate 18 with Lewis acid. An anomeric mixture
of acetate 18 was treated with SnBr4 in CD2Cl2 at -40 °C
(Scheme 3).37 As described by Boons for an analogous system,
an upfield shift of the anomeric protons (H1 to H1′) was
observed, consistent with the formation of sulfonium ion 21.15
In addition, characteristic downfield shifts were observed for
the protons adjacent to the sulfur atom, indicating the presence
of an electron-withdrawing functionality on sulfur. Furthermore,
irradiation of the bridgehead proton H1′ provided NOE cor-
relations to the protons on the ethyl substituent (H7′ and H8′),
demonstrating the proximity of the SEt group to H1′. An HMBC
correlation between bridgehead carbon C1′ and H7′ indicated
that the SEt group was connected to C1′. Although the bicyclic
cation 21 was formed quantitatively at -40 °C, it did not appear
to be a reactive intermediate, because nucleophilic substitution
reactions with allyltrimethylsilane did not occur at this temper-
ature (eq 3).
The higher temperature required to effect substitution of the
4-CH2SEt-substituted acetate 18 as compared to the 4-SR-
substituted acetals 9a,b (eq 1) provides insight into the stability
of sulfonium ion 21. Allylation of acetate 18 did not proceed
under the standard conditions employed for acetals 9a,b (-45
°C); only hydrolysis products were isolated upon low-temper-
ature quenching and aqueous workup (eq 3). We hypothesized
that a stable sulfonium ion such as 21 was formed upon addition
of BF3‚OEt2 to the reaction mixture. It was possible that
sulfonium ion 21 would be unreactive toward allyltrimethylsi-
lane.21 Instead, nucleophilic addition could occur only upon
opening of the stable sulfonium ion through the high-energy
oxocarbenium ion conformers 15 and 16 (R ) Et, Figure 2).21
NMR spectroscopic data of the thiophenyl-substituted acetal
are also consistent with the formation of a bridged sulfonium
ion (17b, eq 4).38 In contrast to the spectroscopic data obtained
for the thioalkyl-substituted acetal 18, a significant and reversible
temperature dependence was observed upon the addition of
SnBr4 to chloroacetate 23 at -50 °C.39 While nearly complete
(37) Employing PhSMe as the substrate under identical conditions indicated
that the observed spectroscopic changes were not the result of complexation
to the Lewis acid.
(38) In analogy to the characterization of sulfonium ion 21, characteristic
chemical shifts and NOE correlations indicated the formation of sulfonium
ion 17b. An HMBC correlation was not obtained in this system, however,
presumably due to the fluctionality of the intermediate.
(39) Chloroacetate 23 was employed in spectroscopic studies because of its
greater ability to ionize under Lewis acidic conditions. The acetate 24 also
formed sulfonium ion 17b, but the equilibrium significantly favored the
starting acetate. Both chloroacetate 23 and acetate 24 afforded comparable
product ratios upon subjection to standard allylation conditions.
Low-temperature NMR spectroscopy confirmed that the
[2.2.2]-bridged bicyclic sulfonium ion 21 was formed upon
9
2084 J. AM. CHEM. SOC. VOL. 130, NO. 6, 2008

Reactions of Sulfur-Substituted Glycosyl Donors
A R T I C L E S
Table 1. Effect of Solvent and Lewis Acid on Allylation
conversion to sulfonium ion 17b was observed at -80 °C, only
a trace quantity of this species was observed at -20 °C. As
depicted in eq 4, generation of sulfonium ion 17b involves the
formation of an ion pair40,41 and the loss of one degree of
rotation;42 these changes would result in a negative entropy for
the reaction (∆S° < 0).43 An increase in temperature, therefore,
would force the equilibrium toward the starting acetal.
entry
Lewis acid
solvent
25:26a
yield (%)b
1
2
3
4
5
6
7
8
BF3‚OEt2
BF3‚OEt2
BF3‚OEt2
BF3‚OEt2
SnBr4
Me3SiOTf
EtAlCl2
TiCl4
CH2Cl2
CHCl3
Et2O
C6H5CH3
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
87:13
86:14
85:15
84:16
87:13
89:11
87:13
88:12
89
77
-
74
79
75
80
75
c
Nucleophilic additions to 4-CH2SPh-substituted acetate 24
also afforded the major diastereomer arising from addition to
the equatorially substituted oxocarbenium ion 15b. Treatment
of 4-CH2SPh-substituted acetate 24 with allyltrimethylsilane in
the presence of BF3‚OEt2 at -45 °C provided 1,4-cis 25 as the
major product (eq 5). The selectivity of this reaction was not
sensitive to solvent or to which Lewis acid was employed (Table
1). In all cases, the major product arose from nucleophilic
addition to the stereoelectronically preferred face of the equa-
torially substituted oxocarbenium ion half-chair conformer 15b,
not the labile bicyclic cation 17b that was observed spectro-
scopically (eq 4).
The difference in stability and reactivity between the 4-CH2-
SEt- and 4-CH2SPh-substituted cations is informative. Low-
temperature NMR spectroscopy indicated that the thioalkyl-
bridged sulfonium ion 21 was stable at all temperatures
examined. Allylation, however, did not occur at -45 °C; instead,
reactivity only occurred when the higher energy oxocarbenium
ion conformers were formed rapidly upon warming to room
temperature. In contrast, the thiophenyl-substituted acetal 24
formed the more labile sulfonium ion 17b, as confirmed by
variable-temperature NMR spectroscopy. Allylation occurred
readily at -45 °C, albeit not through the intermediate sulfonium
ion. These results suggest that the sulfonium ions observed in
these systems are not reactive intermediates, but rather resting
states that are transformed into oxocarbenium ion conformers
and subsequently trapped by allyltrimethylsilane to afford the
major products.
While the bicyclic sulfonium ion 17b does not contribute to
formation of the major product, an experiment was designed to
explore whether this intermediate was responsible for the
formation of the minor product. The 1,4-trans product 26 could
be formed either by addition to the axially substituted oxocar-
benium ion conformer 16b or by ring-opening of sulfonium ion
17b (Figure 2). To determine which of these processes was
occurring, heteroatoms of varying donor ability were incorpo-
rated at the C-4 position. If the minor product arose from ring-
opening of sulfonium ion 17b, heteroatoms with greater donor
ability would provide more of the 1,4-trans product 26.7 Because
donor ability increases with nucleophilicity,44 selenium-, sulfur-,
a Determined by GC and 1H NMR spectroscopy of the unpurified reaction
mixture. b Isolated yield. c GC analysis of the unpurified reaction mixture
indicated minimal conversion to product.
Table 2. Effect of Heteroatom on Allylation
entry
compound
X
cis:transa
yield (%)b
1
2
3
4
5
6
7
8
18
SEt
SPh
SePh
OPh
Cl
Br
I
Bn
89:11
87:13
91:9
91:9
92:8
92:8
93:7
93:7
87
91
92
89
90
85
66
77
24a
24b
24c
24d
24e
24f
24gc
a Determined by GC and 1H NMR spectroscopy of the unpurified reaction
mixture. b Isolated yield. c See ref 26.
and iodine-substituted acetals should provide the highest levels
of 1,4-trans product 26,45 while substrates containing more
electronegative heteroatoms (bromine, chlorine, or oxygen)
would be less likely to form onium ions and should provide
less trans product 26.7
Products arising from neighboring-group participation were
not the major products for any of the heteroatom-substituted
acetals examined (eq 6, Table 2). Regardless of the type of
heteroatom present, the allylated product 1,4-cis 25 predomi-
nated (eq 6, Table 2). The selectivities for strong donors (entries
1-3, 7) were comparable to those in which no heteroatom was
present (entry 8).26 Because no consistent correlation connects
donor ability and diastereoselectivity, it can be concluded that
neighboring-group participation does not significantly, if at all,
contribute to formation of the minor product. Therefore, the
product ratios obtained upon addition of allyltrimethylsilane are
consistent with a scenario in which product formation arose from
nucleophilic addition to the oxocarbenium ion half-chair
conformers.
It is possible that a strong nucleophile is necessary to observe
substitution through the sulfonium ion conformer. In glycosy-
lation reactions where heteroatom assistance is utilized, alcohols
are employed as the nucleophiles. According to Mayr’s nucleo-
(40) Hayes, P. G.; Piers, W. E.; Parvez, M. Organometallics 2005, 24, 1173-
1183.
(41) Bowyer, P. M.; Ledwith, A.; Sherrington, D. C. J. Chem. Soc. B 1971,
1511-1514.
(42) Page, M. I.; Jencks, W. P. Proc. Natl. Acad. Sci. U.S.A. 1971, 68, 1678-
1683.
(43) A van’t Hoff plot revealed the following thermodynamic parameters: ∆H°
) -9.5 kcal/mol and ∆S° ) -43 cal/mol‚K.
(44) Capon, B. Quart. ReV. 1964, 18, 45-111.
(45) Pearson, R. G.; Sobel, H.; Songstad, J. J. Am. Chem. Soc. 1968, 90, 319-
326.
9
J. AM. CHEM. SOC. VOL. 130, NO. 6, 2008 2085

A R T I C L E S
Beaver et al.
Table 3. Effects of Nucleophile on Selectivity
trend shown in Table 3 can be explained by considering
oxocarbenium ions to be the reactive intermediates, it cannot
be readily explained by consideration of stereospecific ring-
opening of sulfonium ion 17b.
The data presented in this paper demonstrate that sulfonium
ions arising from 4-sulfur substituted acetals can be stable
intermediates in nucleophilic substitution reactions to simple
tetrahydropyran systems. Spectroscopic evidence confirmed the
presence of these intermediates with varying stability depending
on the nucleophilicity of the sulfur atom. The reactivity
difference between thioalkyl- and thiophenyl-substituted acetals
indicated that the sulfonium ion intermediates are not reactive
intermediates, but that these intermediates open to form higher
energy oxocarbenium ions. Nucleophilic addition to the pre-
ferred equatorially substituted oxocarbenium ion conformer is
responsible for the stereoselectivity of the reaction.
The characterization of low-energy intermediates is a common
practice for determining the source of stereochemical induction
in a variety of reactions. The observation of stabilized inter-
mediates, however, does not necessitate that the formation of
the major products occur through these structures. It is important
to consider all possible reactive intermediates before concluding
which one must be responsible for product formation. This study
affirms the importance of the Curtin-Hammett principle, which
has been used successfully to explain the counterintuitive
stereochemical outcomes for a number of reactions.17
nucleophilicity
yield
(%)c
entry
Nu
product
parametera
cis:transb
1
25a/26a
1.8
87:13
89
82
2
27a/28a
6.22
80:20
NC-SiMe3
3
4
27b/28b
27c/28c
not defined
10.21
63:37
48:52
87
83
a See ref 46. b Determined by GC and 1H NMR spectroscopy of the
unpurified reaction mixture. c Isolated yield.
philicity parameters,46 alcohols have a nucleophilicity value of
N ≈ 6.47 Allyltrimethylsilane (N ) 1.8) is less nucleophilic than
an alcohol, so it may not be able to react by an SN2 pathway
through the bridged intermediate.48
Upon increasing nucleophilicity, selective formation of the
1,4-trans product 28 was not observed. Instead, stereoselectivity
diminished as nucleophilicity increased (eq 7, Table 3).49 Only
a moderate loss of stereocontrol was observed for the addition
of the acetophenone-derived silyl enol ether (entry 2), which
has comparable nucleophilicity to an alcohol.47 The stereocontrol
was eroded further as the nucleophilicity was increased (entries
3 and 4). A similar trend has been reported for the nucleophilic
substitution reactions of Me3SiCN to oxocarbenium ions.50 This
loss of stereocontrol may be attributed to reactivity near the
diffusion limit, in which both faces of each oxocarbenium ion
conformer are available for nucleophilic addition. While the
Acknowledgment. This research was supported by the
National Institutes of Health, National Institute of General
Medical Science (Grant GM-61066). S.B.B. thanks Novartis for
a graduate fellowship. K.A.W. thanks Amgen and Lilly for
generous support for research. Professor James Nowick (UCI)
is acknowledged for helpful discussions. We thank Deepthi
Sudhakar (UCI) for her contributions to this project. We thank
Dr. John Greaves and Ms. Shirin Sorooshian (UCI) for mass
spectrometric data, and Dr. Phil Dennison (UCI) for assistance
with NMR studies.
(46) Mayr, H.; Kempf, B.; Ofial, A. R. Acc. Chem. Res. 2003, 36, 66-77.
(47) Minegishi, S.; Kobayashi, S.; Mayr, H. J. Am. Chem. Soc. 2004, 126, 5174-
5181.
(48) Denmark has proposed that increasing nucleophilicity leads to stereospecific
ring-opening reactions of chiral dioxanes: Denmark, S. E.; Almstead, N.
G. J. Org. Chem. 1991, 56, 6485-6487.
(49) A similar loss of selectivity was observed for the nucleophilic substitution
reactions of alkyl sulfide 18.
Supporting Information Available: Complete experimental
procedures, product characterization, stereochemical proofs, and
GC and spectral data for selected compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
(50) Shenoy, S. R.; Smith, D. M.; Woerpel, K. A. J. Am. Chem. Soc. 2006,
128, 8671-8677.
JA0767783
9
2086 J. AM. CHEM. SOC. VOL. 130, NO. 6, 2008