Stereochemistry of 1,2-elimination reactions at the E2-E1cB interface - Tert-butyl 3-tosyloxybutanoate and its thioester

Article abstract of DOI:10.1039/b801592a

Experimental data on the stereoselectivity of base-catalyzed 1,2-elimination reactions that produce conjugated carbonyl compounds are scarce in spite of the importance of these reactions in organic and biochemistry. As part of a comprehensive study in this area, we have synthesized stereospecifically-deuterated β-tosyloxybutanoate esters and thioesters and studied the stereoselectivity of their elimination reactions under non-ion pairing conditions. With the availability of both the (2R*,3R*) and (2R*,3S*) diastereomers the innate stereoselectivity could be determined unambiguously. ¹H and ²H NMR data show that these substrates produce 5-6% syn elimination, the usual amount for acyclic substrates undergoing E2 reactions. Contrary to earlier suggestions, activation by a carbonyl group has virtually no influence upon the stereoselectivity. Elimination of the (2R*,3R*) diastereomer of the β-tosyloxyester and thioester produces 21-25% of the (Z)-alkene, much more than observed with a poorer β-nucleofuge. A relatively large amount of (Z)-alkene product seems to be a good marker for an E2 pathway, in which the transition state is E1cB-like, rather than an E1cB_irrev mechanism. Syn KIE values were higher than those for anti elimination for the esters as well as the thioesters. Experimental challenges to the synthesis of stereospecifically-deuterated β-tosyloxyesters are discussed. The Royal Society of Chemistry 2008.

Full text of DOI:10.1039/b801592a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Stereochemistry of 1,2-elimination reactions at the E2–E1cB
interface—tert-butyl 3-tosyloxybutanoate and its thioester†
Jerry R. Mohrig,* David G. Alberg, Craig H. Cartwright, Mary Kay H. Pflum, Jeffrey S. Aldrich,
J. Kyle Anderson, Shelby R. Anderson, Ryan L. Fimmen and Amy K. Snover
Received 29th January 2008, Accepted 19th February 2008
First published as an Advance Article on the web 14th March 2008
DOI: 10.1039/b801592a
Experimental data on the stereoselectivity of base-catalyzed 1,2-elimination reactions that produce
conjugated carbonyl compounds are scarce in spite of the importance of these reactions in organic and
biochemistry. As part of a comprehensive study in this area, we have synthesized
stereospecifically-deuterated b-tosyloxybutanoate esters and thioesters and studied the stereoselectivity
of their elimination reactions under non-ion pairing conditions. With the availability of both the
(2R*,3R*) and (2R*,3S*) diastereomers the innate stereoselectivity could be determined
unambiguously. 1H and 2H NMR data show that these substrates produce 5–6% syn elimination, the
usual amount for acyclic substrates undergoing E2 reactions. Contrary to earlier suggestions, activation
by a carbonyl group has virtually no influence upon the stereoselectivity. Elimination of the (2R*,3R*)
diastereomer of the b-tosyloxyester and thioester produces 21–25% of the (Z)-alkene, much more than
observed with a poorer b-nucleofuge. A relatively large amount of (Z)-alkene product seems to be a
good marker for an E2 pathway, in which the transition state is E1cB-like, rather than an E1cBirrev
mechanism. Syn KIE values were higher than those for anti elimination for the esters as well as the
thioesters. Experimental challenges to the synthesis of stereospecifically-deuterated b-tosyloxyesters are
discussed.
chosen because it provides an appropriate model for the substrate
Introduction
of enoyl-CoA hydratase (EC 4.2.1.17), which catalyzes the syn
elimination–addition of water in the S-b-hydroxybutyryl CoA/S-
crotonyl CoA reaction.6
There has been substantial interest in the stereochemistry of base-
catalyzed E2 and E1cB-like 1,2-elimination reactions over the
past 40 years. It is generally recognized that anti elimination
will dominate over syn elimination under normal conditions,
where ion pairing or the complex conformational factors of cyclic
compounds do not play a major role.1–2 In anti eliminations
orbital overlap is maximized and torsional strain is minimized.
However, it has been suggested that E1cB-transition states, such
as those expected for relatively acidic thioesters, may favor syn
stereochemistry.1,3–4 If the E2 transition state becomes more E1cB-
like, in the More O’Ferrall–Jencks notation, the normal confor-
mational, electronic and orbital symmetry factors favoring anti
elimination could become less important, allowing syn elimination
to compete more effectively; however, the experimental base on
which this proposal depends is quite small.
The activating influence of a carbonyl group on the stereo-
chemistry of 1,2-elimination reactions is an important piece of the
puzzle. Before our research began, very little was known about
how the pKa of a carbonyl substrate affects the stereochemistry
of a base-catalyzed 1,2-elimination reaction to form a conjugated
alkene. Our research has focused on simple, acyclic b-substituted
butanoate esters and thioesters, using conditions where the effects
of aggregation phenomena do not dominate.5 This system was
In addition to the acidity of the proton, the nature of the leaving
group may also play an important role in the stereochemistry
of 1,2-elimination reactions.1–2,4 This paper reports the reaction
stereoselectivity where the p-toluenesulfonyloxy or tosyloxy group
is the nucleofuge. Tosylate is an excellent leaving group and is a
common reference nucleofuge for stereochemical studies of 1,2-
elimination reactions.2,4,7 The two substrates that form our data
set are stereospecifically-labelled tert-butyl (2R*,3R*)-3-tosyloxy-
2,3-2H2-butanoate (1) and its (2R*,3S*) diastereomer (2) and S-
tert-butyl (2R*,3R*)-3-tosyloxy-2,3-2H2-butanethioate (3) and its
(2R*,3S*) diastereomer (4). In every case these substrates were
converted cleanly into tert-butyl (E)-2-butenoate (5) and S-tert-
butyl (E)-2-butenethioate (6), respectively, plus a small percentage
of the (Z)-isomers.
Results and discussion
The synthesis of stereospecifically-labelled 1 and 2 presented
a challenge. We had expected to model the synthesis on that
of tert-butyl (2R*,3R*)-3-acetoxy-2,3-2H2-butanoate (7) and its
(2R*,3S*) diastereomer (8),5 which involved the rigorous syn
deuterogenation of the (E)- and (Z)-isomers of tert-butyl 3-
acetoxy-2-butenoate by Wilkinson’s catalyst,8 Fig. 1.
Using this kind of methodology for the synthesis of 1 and 2
depended on the successful synthesis of the (E)- and (Z)-enol
tosylates and their stereoselective deuterogenation. Reaction of the
Department of Chemistry, Carleton College, Northfield, MN 55057, USA.
E-mail: jmohrig@carleton.edu
† Electronic supplementary information (ESI) available: Experimental
data on silyl enol ethers 13–14 and their deuterogenation to give 15 and
16. See DOI: 10.1039/b801592a
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 1641–1646 | 1641
©

View Article Online
both 1 and 2 to obtain the necessary kinetic isotope effects (KIEs),
we abandoned the silyl enol ether synthetic methodology.
We ultimately resorted to the synthesis of 1 and 2 by hydrolysis of
the acetoxy esters 7 and 8 followed by tosylation of the resulting b-
hydroxy esters, even though a fair amount of the stereospecifically-
deuterated acetoxy ester was lost to competing base-catalyzed
elimination reactions, Fig. 3. Use of a more polar solvent mixture
produced a lower elimination/hydrolysis ratio, making the loss due
to elimination (∼30%) acceptable.5 Isotopic exchange at C-2 of the
b-hydroxy esters was minimized (<2%) by carefully monitoring the
hydrolysis reactions. Substrates 1 and 2 contained 5–9% of the C-2
diprotonated esters.
The thioesters 3 and 4 were synthesized by deblocking the
tert-butyl esters 1 and 2 with TFA and esterification with 2-
methyl-2-propanethiol,9 Fig. 3. As long as excess TFA was not
present, no H/D exchange or rearrangement was observed in the
transesterification reaction.
In our elimination reactions of ester 1 with KOH in 3 : 1 v/v
EtOH–H2O a deuteron is removed by anti elimination in the
formation of the deuterated (E)-alkene 5, whereas a proton is
removed from ester 2. Reactions of 1 produced 79% of 5 plus 21%
of its deuterated isomer, tert-butyl (Z)-2-butenoate. With ester 2
only 2.2% of the deuterated (Z)-alkene was produced. Elimination
of TsOH from nondeuterated tert-butyl 3-tosyloxybutanoate 11
produced 6–7% tert-butyl (Z)-2-butenoate.
Fig. 1 Preparation of acetoxy esters 7 and 8.
enolate of tert-butyl acetoacetate with p-toluenesulfonyl fluoride
led to a 3 : 1 mixture of tert-butyl (E)- and (Z)-3-tosyloxy-2-
butenoate (9a and 9b), which could readily be separated by flash
chromatography on SiO2. However, exhaustive efforts to find
conditions under which 9a could be deuterogenated to 2 were
unsuccessful. In fact, deuterogenation of 9a led solely to cleavage
of the tert-butyl ester, producing (E)-3-tosyloxy-2-butenoic acid
(10), Fig. 2.
Fig. 2 Synthesis and attempted deuterogenation of 9a.
(E)-Alkenes 5 and 6 from the base-catalyzed elimination of
TsOH from substrates 1–4 were purified by preparative GC before
multiple 2H NMR integrations were used to determine the amount
of anti and syn elimination from the isotopically-labeled diastere-
omers, as shown in Table 1. The (2R*,3R*) diastereomers produce
much more syn elimination than the (2R*,3S*) diastereomers due
to the adverse primary KIE for anti elimination of the (2R*,3R*)
compounds.
Hydrogenation of 9a with 5% Rh/C under more vigorous
conditions in 1 : 1 v/v EtOD–C6H6 produced tert-butyl ethyl
ether, p-toluenesulfonic acid (TsOH), butanoic acid, and ethyl
butanoate. This experiment suggests that 10 results from acid-
catalyzed cleavage of the tert-butyl ester through an SN1 pathway,
presumably because a small amount of TsOH is produced under
=
the hydrogenation conditions. No addition of H2 to the C C
The elimination of TsOH from thioesters 3 and 4 with KOH
in 3 : 1 EtOH–H2O produced mainly deuterated S-tert-butyl
(E)-2-butenethioate (6) plus a smaller amount of deuterated S-
tert-butyl (Z)-2-butenethioate.5 Thioester 3 produced 25% of the
(Z)-isomer whereas 4 produced 1–2%. Elimination of TsOH from
nondeuterated 12 gave 7% S-tert-butyl (Z)-2-butenethioate. Our
kinetic studies show that thioester 12 reacts 18 times faster than
was observed in the presence of 5% Pt/C, 5% Rh/Al2O3, or 5%
Pd/BaSO4. 9a is remarkably resistant to simple hydrogenation.
Another possible approach for the synthesis of stereo-
specifically-labelled 1 and 2, which was ultimately unsuccess-
ful, was the deuterogenation of tert-butyl (E)- and (Z)-3-
trimethylsilyloxy-2-butenoate, followed by cleavage of the silyl
ether and subsequent tosylation of the b-hydroxy esters. Deutero-
genation of the silyl enol ether with Wilkinson’s catalyst under
the usual conditions was successful and stereospecific. Since the
trimethylsilyl enol ether was unstable on SiO2, even in the presence
◦
ester 11 at 30.0 C with KOH in 3 : 1 EtOH–H2O, reflecting the
greater activating influence of the thioester group.10
In order to ensure the validity of the results shown in Table 1,
we have carried out three sets of control experiments on 1 and
11. Virtually no isomerization of tert-butyl (Z)-2-butenoate11 to
its (E)-isomer 5 was observed under the reaction conditions.
GC analysis showed that only 0.33% 0.01% of the (Z)-isomer
rearranged to 5 under elimination conditions; such a small amount
of isomerization would have no effect on our results or conclusions.
When the reaction was carried out with only 50% of the KOH
R
of Et3N, chromatography was attempted using Florisilꢀ. However,
separation of the (E)- and (Z)-isomers was poor and recovered
yields were small. Chromatography of the tert-butyldimethylsilyl
R
enol ether on Florisilꢀ was successful in producing pure tert-butyl
(E)-3-(tert-butyldimethylsilyloxy)-2-butenoate, but only vanish-
ingly small amounts of pure (Z)-isomer could be isolated. Since
the success of our elimination studies required the availability of
Fig. 3 Synthesis of tosyloxyesters 1 and 2 and thioesters 3 and 4.
1642 | Org. Biomol. Chem., 2008, 6, 1641–1646 This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Table 1 Stereoselectivity data and KIEs for esters 1 and 2 and thioesters 3 and 4
a
b
%synR*R*
%synR*S*
kR*S*/kR*R*
(kH/kD)syn
(kH/kD)anti
1,2: Y = OC(CH3)3
3,4: Y = SC(CH3)3
a (kH/kD)syn = %synR*R*/%synR*S* × kR*R*/kR*S*
13.6 0.2
12.3 0.2
1.0 0.2
1.0 0.2
2.31 0.07
1.98 0.06
5.9
6.2
2.6
2.2
.
b (kH/kD)anti = %antiR*S*/%antiR*R* × kR*S*/kR*R*
.
necessary for complete elimination of 1, the deuterium content
of 5 was virtually identical to that observed at complete reaction;
in addition, the recovered 1 showed no loss of stereochemical
integrity. Finally, carrying out the elimination reaction using
nondeuterated 11 with KOD and EtOD–D2O showed <1% H/D
exchange in the recovered 5.
The kH/kD KIEs in Table 1 are consistent with the KIE values
for other thioester and ester substrates,1,5,12 and for KIE values in
the elimination reactions of b-tosyloxyethyl phenyl sulfone.13 In
all cases the KIEs for syn elimination were higher than those for
anti elimination.
Using the data in Table 1, the innate stereoselectivities of the
1,2-elimination reactions, those which would be expected in the
absence of deuterium labels, can be calculated in a straightforward
manner. The results are shown in Table 2. Although (kH/kD)syn is
subject to a substantial error, this error is not propagated to the
innate stereoselectivities shown in Table 2. Any error in (kH/kD)syn
is offset by the compensating error in the percentage of syn
elimination from the (2R*,3S*) diastereomer by which it must be
multiplied to obtain the innate stereoselectivity. As an additional
check, the calculated stereoselectivity percentages shown in Table 2
were the same when calculated from both the (2R*,3R*) and
(2R*,3S*) diastereomers. Secondary deuterium KIEs are unlikely
to be greater than 1.03 and would have a negligible effect on the
results.12
The stereochemical results from our b-tosyloxy substrates
show the usual amount of syn elimination for acyclic substrates
undergoing E2 reactions; 4–6% is common under non-ion pairing
conditions.2 Table 2 shows that the carbonyl group has virtually no
influence upon the stereoselectivity of these elimination reactions;
this was also the case that we found with b-trimethylacetoxy
esters and thioesters.5 Our experimental results indicate that base-
catalyzed 1,2-elimination reactions that produce conjugated esters
and thioesters follow the same stereochemical pattern as those
found with unactivated substrates. This pattern suggests that
contrary to earlier suggestions electronic factors do not induce
a preference for syn elimination in esters or thioesters with good
or modest leaving groups. We will report our results on systems
with a poor leaving group, which clearly react by the E1cBirrev
mechanism, in the near future.14
Thibblin has pointed out that there can be a hyperconjugative
interaction between the b-leaving group and the proton being
abstracted. This suggests that proton transfer and cleavage of
the leaving group are to some extent coupled both in the E2 and
E1cB pathways. Accordingly, a periplanar positioning between the
base and the tosyloxy group is associated with some double-bond
character of the Ca–Cb bond.15 Calculations have corroborated
this idea.16
A classic problem in diagnosing reaction mechanisms has
been using kinetic studies to distinguish between the E1cBirrev
mechanism and the E2 mechanism in which the transition state
is E1cB-like.1 Perhaps the surest sign that 11 and 12 undergo
elimination by the E2 mechanism is the relative percentages of
their (E)- and (Z)-alkene products. Base-catalyzed elimination
of TsOH from simple secondary alkyl tosylates under non-ion
pairing conditions normally produces 15–35% (Z)-alkene by the
E2 pathway, with the range reflecting steric effects as well as
product stability.17,18
Steric effects have been implicated because other excellent
leaving groups in E2 reactions, in particular iodide and bromide,
produce only 67% as much of the (Z)-alkene as found with
tosylate substrates.17 This differential is less pronounced in aprotic
solvents (80%) than in protic solvents (56%), due to H-bonding
of the tosyloxy group, which effectively increases its size. The
data include the 2-pentyl, 2-hexyl, 3-hexyl, 2-decyl, and 5-decyl
systems, all studied under non-ion pairing conditions. On average,
the secondary alkyl iodides and bromides produce 17–18% 3%
(Z)-alkene, whereas in protic solvents the tosylates produce 31%
5% (Z)-alkene. Nevertheless, all of these secondary acyclic iodides,
bromides, and tosylates produce fairly large amounts of (Z)-alkene
products, making this phenomenon a hallmark of the E2 pathway.
Whereas simple acyclic (E)-alkenes are only about 4.2 kJ mol−1
lower in energy than the (Z)-isomers, calculations have indi-
cated that the (E)-isomers of tert-butyl 2-butenoate and S-tert-
butyl 2-butenethioate are 8.8 kJ mol−1 more stable than their
(Z)-isomers.5 Nonetheless, 7% of the (Z)-isomers are formed
in the KOH-catalyzed eliminations of TsOH from 11 and 12
in EtOH–H2O. This is a 4–5-fold increase in the percentage
of (Z)-alkene compared to the elimination reactions of b-
trimethylacetoxybutanoate esters and thioesters, which likely
Table 2 Innate stereoselectivity of base-catalyzed p-toluenesulfonic acid
elimination
Y
%syn elimination
OC(CH3)3
SC(CH3)3
5.6 1.2
5.9 1.2
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 1641–1646 | 1643
©

View Article Online
proceed through E1cBirrev pathways.5 Furthermore, the alkene
products from the stereospecifically-deuterated substrates 1 and
3 contained 21% and 25% of the deuterated (Z)-alkene, where
an anti elimination leading to the (E)-alkene involves removal of a
deuteron. By contrast, the analogous stereospecifically-deuterated
b-trimethylacetoxybutanoate esters and thioesters produced 1.3–
1.5% of the deuterated (Z)-alkene. In addition, we have observed
very small amounts of (Z)-alkene products in the elimination of b-
(m-trifluoromethylphenoxy)butanoate esters and thioesters, which
are definitely at the E1cB interface.14
A relatively large amount of (Z)-alkene from elimination
reactions producing conjugated carbonyl compounds seems to
be a good marker for an E2 pathway, in which the transition
state is E1cB-like, rather than an E1cBirrev mechanism. It is not
unlikely that it may also be an additional mechanistic marker
for E2 pathways in elimination reactions of substrates with b-
activating groups such as cyano, nitro or sulfonyl, which produce
1,2-disubstituted alkenes, especially when the energy difference
between (E)- and (Z)-alkene products is reasonably large.
Attempted reduction of 9a with 2H2
Reaction of 9a (3.0 g, 9.5 mmol) for 4 days at 40 ◦C with 2H2 (150
psi) and Rh(PPh3)3Cl (0.30 g, 0.32 mmol) in C6H6 or 5% Rh/C
in cyclohexane led to (E)-3-tosyloxy-2-butenoic acid 10 (0.81 g,
34%), mp 123.5–124.5 ◦C (from EtOH–H2O). 10: found: C, 51.1;
H, 5.0; S, 12.8. Calc. for C11H12O5S: C, 51.5; H, 4.7; S, 12.5%; dH
(300 MHz; CDCl3; Me4Si) 2.31 (3 H, d), 2.47 (3 H, s), 5.70 (1
H, q), 7.37 (2 H, d), 7.82 (2 H, d). Reaction of 9a with H2 and
5% Rh/C in 1 : 1 v/v EtOD–C6H6 produced p-toluenesulfonic
acid, ethyl butanoate, butanoic acid, and tert-butyl ethyl
ether.
tert-Butyl 3-tosyloxybutanoate 11
Reaction of tert-butyl 3-hydroxybutanoate with p-tosyl chloride
in the presence of DMAP in CH2Cl2 produced 11: found: C,
57.3; H, 6.75; S, 10.3. Calc. for C15H22O5S: C, 57.3; H, 7.05; S,
10.2%; dH (300 MHz; CDCl3; Me4Si) 1.32 (3 H, d), 1.40 (9 H, s),
2.42 (3 H, s), 2.46 (2 H, d), 4.97 (1 H, m), 7.33 (2 H, d), 7.79
(2 H, d).
Experimental
S-tert-Butyl 3-tosyloxybutanethioate 12
General methods
1H NMR spectra were run on a 200 MHz Bruker/IBM or
300 MHz Nicollet FT NMR spectrometer. The solvent for 76 MHz
2H NMR spectroscopy was C6H6 with a small amount of C6D6
as an internal reference (dD 7.15). Multiple 1H and 2H NMR
integrations were used to determine the isotopic purities of
deuterated substrates, and elimination and competition results
were corrected for the presence of C-2 diprotonated substrates
and diastereomeric impurities. Capillary GC was run with a 5%
phenylmethyl silicone (SPB-5) column. Preparatory GC was done
on a modified Varian Aerograph Model 90 with an 8 ft × 3/8
in 5% Carbowax 20 M or 15% methylsilicone column. Flash
chromatography was carried out using silica gel (Aldrich, 7–230
Reaction of 11 with TFA, TFAA and (CH3)3CSH gave 12, mp
42–43.5 ◦C (from hexane). 12: found: C, 54.7; H, 6.7. Calc. for
C15H22O4S2: C, 54.5; H, 6.7%; dH (60 MHz; CDCl3; Me4Si) 1.3
(3 H, d), 1.4 (9 H, s), 2.4 (3 H, s), 2.7 (2 H, d), 4.9 (1 H, m),
7.3 (2 H, d), 7.8 (2 H, d); m/z (ESIMS) 353.0858 (M+; calc. for
C15H22O4S2Na: 353.0852).
tert-Butyl (2R*,3R*)- and (2R*,3S*)-3-hydroxy-
2,3-2H2-butanoate
Hydrolyses of tert-butyl (2R*,3R*)- and (2R*,3S*)-3-acetoxy-2,3-
2H2-butanoate5 were carried out in stirred solutions of 1 : 1
v/v EtOH–H2O at 22 ◦C for 45–60 min, using 2.0 mL solvent
per 1.0 g substrate and 10% molar excess KOH. Reactions
were quenched with 1–2 drops of acetic acid and after standard
workup the crude product mixtures were used in the syntheses of
1 and 2.
˚
mesh, 60 A). For elimination reactions of deuterated substrates, all
glassware was soaked in NaHCO3 solution, rinsed with distilled
water and oven-dried. Melting points were calibrated against a
benzoic acid standard.
tert-Butyl (E)- and (Z)-3-tosyloxy-2-butenoate 9a and 9b
tert-Butyl (2R*,3R*)- and (2R*,3S*)-3-tosyloxy-
tert-Butyl acetoacetate (36 g, 0.22 mol) was added dropwise
to a stirred slurry of KH (0.23 mol) in THF (250 mL, N2).
After 1 h a solution of tosyl fluoride (38 g, 0.22 mol) in N,Nꢀ-
dimethylpropyleneurea (DMPU, 60 mL) was added over a 20 min
period and the reaction was stirred for 84 h. Aqueous workup, with
acidification to pH 7, and Et2O extractions gave a 3 : 1 mixture
of 9a and 9b. Flash chromatography (15 : 1 SiO2 : crude product,
Et2O–hexane) gave 9a (20.6 g, 30%), mp 35.5–37 ◦C (from hexane),
9b (6.9 g, 10%), mp 64.5–65.5 ◦C (from hexane), and tert-butyl
(E)- and (Z)-3-fluoro-2-butenoate (5%). 9a: found: C, 57.5; H,
6.6. Calc. for C15H20O5S: C, 57.7; H, 6.45%; dH (300 MHz; CDCl3;
Me4Si) 1.45 (9 H, s, Me), 2.3 (3 H, d, Me), 2.46 (3 H, s, Me), 5.64
(1 H, q, =CH), 7.4 (2 H, d), 7.8 (2 H, d). 9b: found: C, 57.7; H,
6.4; S, 10.55. Calc. for C15H20O5S: C, 57.7; H, 6.45; S, 10.3%; dH
(300 MHz; CDCl3; Me4Si) 1.41 (9 H, s, Me), 2.0 (3 H, d, Me), 2.43
(3 H, s, Me), 5.42 (1 H, q, =CH), 7.33 (2 H, d), 7.87 (2 H, d).
2,3-2H2-butanoate 1 and 2
A solution of tosyl chloride (17.6 g, 90.5 mmol) in pyridine (68 mL)
was added dropwise to tert-butyl (2R*,3R*)- and (2R*,3S*)-3-
hydroxy-2,3-2H2-butanoate (3.0 g, 18.5 mmol) over 30 min while
stirring, and the reaction was continued for 14 h. After filtration,
a few mL of ice were added and the mixture was stirred for 10 min.
Addition of hexane, washing with H2O, HCl and NaHCO3, and
purification by flash chromatography (60 : 1 SiO2 : compd, 10–20%
Et2O–hexane) produced 1 (4.7 g, 77%) and 2 (3.7 g, 63%). 1: dD
(76 MHz; C6H6; C6D6) 2.07 (s, 2CD), 5.06 (s, 3CD); dH (200 MHz;
CDCl3; Me4Si) 1.31 (3 H, s), 1.44 (9 H, s), 2.44 (3 H, s), 2.63 (1 H,
br s), 7.31 (2 H, d), 7.81 (2 H, d). 2: dD (76 MHz; C6H6; C6D6) 2.45
(s, 2CD), 5.06 (s, 3CD); dH (200 MHz; CDCl3; Me4Si) 1.31 (3 H,
s), 1.44 (9 H, s), 2.44 (3 H, s), 2.40 (1 H, br s), 7.31 (2 H, d), 7.81
(2 H, d).
1644 | Org. Biomol. Chem., 2008, 6, 1641–1646
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
S-tert-Butyl (2R*,3R*)- and (2R*,3S*)-3-tosyloxy-2,3-2H2-
butanethioate 3 and 4
Values for (kH/kD)syn are subject to much greater relative error
than (kH/kD)anti, because the values of the syn KIE depend
directly upon the very small percentage of syn elimination from
the (2R*,3S*) diastereomers. A worst-case analysis showed that
whereas (kH/kD)anti could have as much as 5% relative error, the
relative error for (kH/kD)syn could be as high as 25%.
◦
To 3.0 g (9.5 mmol) of esters 1 and 2 at 0 C (N2, stirring) were
added 3.0 molar equiv. TFA, and the mixture was allowed to return
to rt. After 13 h 1.2 molar equiv. of TFAA were added at 0 ◦C. After
1 hr 1.2 molar equiv. of 2-methyl-2-propanethiol were added and
the reaction was continued for 22 h, followed by an aqueous
workup (Et2O, H2O, NaHCO3, evaporation). Recrystallization
from hexane produced 3 (2.05 g, 65%), mp 46–47 ◦C and 4 (2.4 g,
Conclusion
◦
76%), mp 46–48 C. 3: dD (76 MHz; C6H6; C6D6) 2.19 (s, 2CD),
In conclusion, we have shown that contrary to earlier suggestions,
activation by a carbonyl group has virtually no influence upon
the stereoselectivity of base-catalyzed, 1,2-elimination reactions
of b-tosyloxybutanoate esters and thioesters, even though these
reactions take place at the E2–E1cB interface. Under conditions
in which aggregation phenomena and the complex conformational
effects of cyclic compounds play no role, electronic effects by
themselves do not seem to be a major determining factor leading
to syn elimination.
5.00 (s, 3CD); dH (200 MHz; CDCl3; Me4Si) 1.29 (3 H, s), 1.40
(9 H, s), 2.44 (3 H, s), 2.85 (1 H, br s), 7.31 (2 H, d), 7.81 (2 H,
d). 4: dD (76 MHz; C6H6; C6D6) 2.61 (s, 2CD), 5.00 (s, 3CD); dH
(200 MHz; CDCl3; Me4Si) 1.29 (3 H, s), 1.40 (9 H, s), 2.44 (3 H,
s), 2.59 (1 H, br s), 7.31 (2 H, d), 7.81 (2 H, d).
General method for elimination reactions of deuterated substrates
Stereospecifically deuterated tosyloxyester and thioester sub-
strates (250–800 mg) were stirred in 3 : 1 v/v EtOH–H2O in a
22–25 ◦C water bath with 10% molar excess KOH. Concentrations
were 1.6 M for 1 and 2 and 1.13 M for 3 and 4. Reaction times for
esters 1 and 2 were 7–8 min and for thioesters 3 and 4 were 30 s.
Reactions were quenched with 2–4 drops of acetic acid and then
neutralized with NaHCO3. Flash chromatography (SiO2/hexane–
Et2O), careful evaporation at rt or short path distillation, analysis
by capillary GC, and separation by preparatory GC led to the
recovery of deuterated 5 from ester substrates and 6 from thioester
substrates. The (Z)-alkene product from 3 was also recovered in
one experiment and was ∼99% deuterated at C-2. Alkenes 5 and 6
were analyzed by multiple 2H NMR integrations (C6H6) of samples
from two or more separate experiments. 5: dD (76 MHz; C6H6;
C6D6) 5.73 (s, 2CD), 6.82 (s, 3CD); dH (200 MHz; C6D6; C6H6)
1.34 (3 H, s), 1.41 (9 H, s), 5.75 (s); dH (200 MHz; CDCl3; Me4Si)
1.45 (9 H, s), 1.85 (3 H, s), 5.75 (s). 6: dD (76 MHz; C6H6; C6D6)
5.91 (s, 2CD), 6.69 (s, 3CD); dH (200 MHz; CDCl3; Me4Si) 1.46 (9
H, s), 1.81 (3 H, s), 5.95 (s).
Acknowledgements
We acknowledge the generous support of the National Science
Foundation (NSF Grant CHE-8505408) and the National In-
stitutes of Health (NIH Grant GM40018). Acknowledgement is
made to the donors of the Petroleum Research Fund, administered
by the American Chemical Society, for partial support of this
research. Student stipends were also provided by the Howard
Hughes Medical Institute and the 3M Foundation. We are grateful
to the National Science Foundation (CHE-8409822) and the
3M Foundation for funding the purchase of a 200 MHz NMR
spectrometer. We thank Dr Stephen B. Philson and the University
of Minnesota for making possible the acquisition of 76 MHz 2H
NMR spectra.
References
1 J. R. Gandler, in The Chemistry of Doubly-Bonded Functional Groups,
Suppl. A, ed. S. Patai, Wiley & Sons, New York, 1989, vol. 2, ch. 12,
pp. 733–797.
2 W. H. Saunders Jr. and A. F. Cockerill, Mechanisms of Elimination
Reactions, Wiley & Sons, New York, 1973, ch. 3, pp. 105–163.
3 S. Gronert, J. Am. Chem. Soc., 1992, 114, 2349–2354; W. H. Saunders,
Jr., J. Org. Chem., 2000, 65, 681–684.
kH/kD kinetic isotope effects
KIEs were determined from the percentages of syn and anti
elimination from substrates 1–4 and the relative rates of the
diastereomeric pairs by a series of competition reactions using
approximately a 1 : 1 ratio of the (2R*,3R*) and (2R*,3S*)
diastereomers and 50–60% of the KOH necessary for complete
elimination. The relative rates were corrected for the percentages
of (Z)-alkenes from the two diastereomers in the calculation of
kR*S*/kR*R*. For each pair of substrates 2–3 competition reactions
were run. After SiO2/hexane–ether flash chromatography and
careful rotary evaporation at <30 ◦C, the extent of the reactions
4 B. R. Dohner and W. H. Saunders, Jr., J. Am. Chem. Soc., 1986, 108,
245–247.
5 J. R. Mohrig, H. K. Carlson, J. M. Coughlin, G. E. Hofmeister, L. A.
McMartin, E. G. Rowley, E. E. Trimmer, A. J. Wild and S. C. Schultz,
J. Org. Chem., 2007, 72, 793–798.
6 D. J. Creighton and N. S. R. K. Murthy, in The Enzymes, ed.
D. S. Sigman and P. D. Boyer, Academic Press, New York, 3rd edn,
1990, vol. 19, ch. 7; R. L. D’Ordine, B. J. Bahnson, P. J. Tonge and V. E.
Anderson, Biochemistry, 1994, 33, 14733–14742.
7 A. F. Cockerill and R. G. Harrison, in The Chemistry of Doubly-Bonded
Functional Groups, Suppl. A, ed. S. Patai, Wiley & Sons, New York,
1977, part 1, ch. 4, pp. 149–329; C. J. M. Stirling, Acc. Chem. Res.,
1979, 12, 198–203; D. B. Boyd, J. Org. Chem., 1985, 50, 885–886.
8 J. R. Mohrig, S. L. Dabora, T. F. Foster and S. C. Schultz, J. Org.
Chem., 1984, 49, 5179–5182.
1
of 3 and 4 was ascertained by 200 MHz H NMR integrations
of the tosyloxy CH3 peak, as well as the allyl CH3 peaks of 6
and its (Z)-isomer, using carefully determined sensitivity factors
(CDCl3, 2.0 s delay). The extent of the reactions of 1 and 2 and
the diastereomeric composition in reactions of 1/2 and 3/4 were
9 J. R. Mohrig, P. J. Vreede, S. C. Schultz and C. A. Fierke, J. Org. Chem.,
1981, 46, 4655–4658.
10 T. L. Amyes and J. P. Richard, J. Am. Chem. Soc., 1992, 114, 10297–
10302; T. L. Amyes and J. P. Richard, J. Am. Chem. Soc., 1996, 118,
3129–3141.
2
determined directly by multiple 76 MHz H integrations (C6H6)
of the C3 alkene and C3 substrate signals and of the C2 signals of
the (2R*,3R*) and (2R*,3S*) substrates, respectively.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 1641–1646 | 1645
©

View Article Online
11 E. V. Dehmlow and J. Wilkenloh, Chem. Ber., 1990, 123, 583–587.
12 P. Ryberg and O. Matsson, J. Am. Chem. Soc., 2001, 123, 2712–
2718.
13 D. R. Marshall, P. J. Thomas and C. J. M. Stirling, J. Chem. Soc., Perkin
Trans. 2, 1977, 1898–1909.
14 J. R. Mohrig, unpublished work.
16 W. H. Saunders, Jr., J. Org. Chem., 1999, 64, 861–865; I. V. Alabugin
and T. A. Zeidan, J. Am. Chem. Soc., 2002, 124, 3175–3184.
17 R. A. Bartsch and J. Za´vada, Chem. Rev., 1980, 80, 453–494.
18 W. H. Saunders, Jr. and A. F. Cockerill, Mechanisms of Elimination
Reactions, Wiley & Sons, New York, 1973, ch. 4, pp. 165–203; J. Sicher,
J. Za´vada and M. Pa´nkova´, Collect. Czech. Chem. Commun., 1971,
36, 3140–3161; J. Za´vada and M. Pa´nkova´, Collect. Czech. Chem.
Commun., 1977, 42, 3421–3438.
15 A. Thibblin, Chem. Scr., 1980, 121–127; Q. Meng and A. Thibblin,
J. Chem. Soc., Perkin Trans. 2, 1998, 583–589.
1646 | Org. Biomol. Chem., 2008, 6, 1641–1646
This journal is
The Royal Society of Chemistry 2008
©