Catalytic asymmetric hetero diels-alder reactions of N-sulfinyl dienophiles with chiral bis(oxazoline)copper(II) and -zinc(II) triflates

Article abstract of DOI:10.1002/ejoc.200600083

Asymmetric hetero Diels-Alder (HDA) reactions of N-sulfinyl dienophiles with bis(oxazoline)copper(II) and -zinc(II) triflates are described. The cycloaddition with cyclic and acyclic 1,3-dienes has been studied. The copper catalyst was found to be more ef

Full text of DOI:10.1002/ejoc.200600083

FULL PAPER
DOI: 10.1002/ejoc.200600083
Catalytic Asymmetric Hetero Diels–Alder Reactions of N-Sulfinyl Dienophiles
with Chiral Bis(oxazoline)copper(II) and -zinc(II) Triflates
Molla Mellese Endeshaw,[a] Annette Bayer,[b] Lars K. Hansen,[b] and Odd R. Gautun*[a]
Keywords: Asymmetric catalysis / Lewis acids / Cycloaddition / N-Sulfinyl dienophiles
Asymmetric hetero Diels–Alder (HDA) reactions of N-sulfinyl
assumed to arise from the formation of less reactive hetero-
dienophiles with bis(oxazoline)copper(II) and -zinc(II) triflates chiral complexes. The copper catalyst showed a nearly linear
are described. The cycloaddition with cyclic and acyclic 1,3-
dienes has been studied. The copper catalyst was found to
be more efficient. With 10 mol-% of catalyst loading, a pro-
nounced enhancement in turnover, diastereoselectivity (54–
Ͼ90% de), and enantioselectivity (30–98% ee) was obtained
by the addition of 100 mol-% of TMSOTf. The role of the
additive is unclear, but we hypothesize that it is involved in
the breakdown of catalyst-sulfine aggregates assists in the
release of catalyst from the HDA adducts. Mechanistic stud-
ies revealed a positive nonlinear effect with the zinc catalyst,
relationship between the enantioselectivity of the ligand
tested and the HDA product formed. The relative configura-
tions and, in one case, the absolute configuration of the HDA
products were established by X-ray analysis. Finally, an
enantioselective route to (3aR,6aS)-3,3a,4,6a-tetrahydrocy-
clopenta[d][1,3]oxazol-2-one, a precursor of the C-ring in ag-
elastatin A, applying the asymmetric HDA reaction is de-
scribed.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2006)
shown in Scheme 1. TMSOTf was assumed to assist the re-
lease of the chiral catalyst from the HDA adducts and
thereby, improve the catalytic turnover. Herein, we present
further studies of this catalytic reaction including the effect
of catalyst and TMSOTf loading, the use of other additives,
non linear effects, and cycloaddition with cyclic and acyclic
1,3-dienes. Furthermore, an enantioselective route to
(3aR,6aS)-3,3a,4,6a-tetrahydrocyclopenta[d][1,3]oxazol-2-
one [(3aR,6aS)-20], a precursor of the C-ring in agelastatin
A,[8] applying the asymmetric HDA reaction is presented.
Introduction
Hetero Diels–Alder (HDA) reactions of N-sulfinylaniline
to conjugated dienes were first described by Wichterle and
Rocek in 1953.[1] Since then, a number of Diels–Alder (DA)
reactions for various types of N-sulfinyl compounds have
been reported.[2] The resulting 1,2-thiazine 1-oxide adducts
can be further transformed into synthetically useful deriva-
tives including homoallylic amines and vicinal amino
alcohols by well-established techniques.[2f] Only a few exam-
ples of asymmetric HDA reactions with either chiral N-sulf-
inyl dienophiles[3] or chiral dienes[4] have been reported to
proceed with high diastereoselectivities (Ͼ97% de). The ap-
plication of chiral Lewis-acid complexes as catalysts for DA
and HDA reactions has been demonstrated to give excellent
chiral induction for many different reaction systems.[5]
Recently, we described our preliminary findings with
10 mol-% of bis(oxazoline)copper(II) or zinc(II)[6] triflates
in combination with trimethylsilyl trifluoromethanesulfon-
ate (TMSOTf, 100 mol-%) to catalyze asymmetric HDA re-
actions of N-sulfinyl dienophiles.[7a] The use of TMSOTf as
an additive was crucial to obtain high yields (68–86%) and
enantioselectivities (97–98% ee) for the applied test system
Scheme 1. The test system: HDA reactions of N-sulfinyl dieno-
philes 1a and 1b with 1,3-cyclohexadiene (2).
Results and Discussion
Reaction Optimization
[a] Department of Chemistry, Norwegian University of Science
and Technology (NTNU),
A series of chiral Lewis acids have previously been
screened as promoters for the asymmetric HDA test reac-
tion of N-sulfinyl dienophiles 1a[9] and 1b[2b] with 1,3-cyclo-
hexadiene (2) shown in Scheme 1.[7] The best results were
7491 Trondheim, Norway
Fax: +47-73594256
E-mail: odd.gautun@chem.ntnu.no
[b] Department of Chemistry, University of Tromsø,
9037 Tromsø, Norway
obtained with stoichiometric amounts of 4a-Cu(OTf)2 or
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
[10]
4a-Zn(OTf)2
(Figure 1), providing the endo adducts 3a
Eur. J. Org. Chem. 2006, 5249–5259
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5249

M. M. Endeshaw, A. Bayer, L. K. Hansen, O. R. Gautun
FULL PAPER
(X = Cbz) or 3b (X = Ts) in good enantioselectivities (90–
98% ee), diastereoselectivities (Ͼ90% de), and yields (63–
85%).[7a,7b] Attempts to perform the test reactions with
catalytic amounts of the Lewis acids gave rather disappoint-
ing results. A distinct drop in de and ee was observed in
reactions with 10 mol-% loading of ent-4a-Cu(OTf)2 (ent =
enantiomer), as shown in Table 1, Entries 1 and 2 for N-
sulfinyl 1a and 1b, respectively. Better results were obtained
with ent-4a-Zn(OTf)2 (Table 1, Entries 11 and 12), but the
yields and stereoselectivities decreased compared to those
of the reactions with stoichiometric amounts of the catalyst.
The results indicate that the catalysts are not released from
the HDA adducts, thus inhibiting the catalyst turnover. The
high yield observed in the reaction with the Ts-containing
dienophile 1b (93%, Table 1, Entry 2) may be explained by
a pronounced, competitive, uncatalyzed, HDA reaction.
The uncatalyzed reaction at –85 °C (reaction time 17 h)
Figure 1. Chiral ligands tested.
gave a 55% yield of product 3b (endo/exo, 64:36). For the tioselective aldol additions of enol silanes to pyruvate es-
Cbz-substituted dienophile 1a, the uncatalyzed reaction ters.[12] Silyl crossover experiments identified TMSOTf as
gave a 17% yield of 3a (endo/exo, 14:86) at –55 °C after an additive to accelerate these reactions. TMSOTf is known
24 h.
to be an extremely powerful silylating agent.[13] The pres-
Zhang and Flann recently suggested a six-membered ence of oxygen atoms in the HDA adducts 3 leads to a
(O,O) chelate for the adduct of N-sulfinylphosphoramidates working hypothesis involving cleavage of the M-(O,O) che-
[1, X = R2O(O)P] and SnCl4, based on NMR and elemental late (M = Cu or Zn) by addition of TMSOTf. An increased
analysis of the complex.[11] A stereochemical model involv- catalytic turnover of the test system (Scheme 1) is possible
ing a bidentate coordination of N-sulfinyl dienophile 1a or according to this hypothesis. Indeed, a significant improve-
1b to the chiral Lewis acid with the sulfinyl oxygen and the ment of the turnover, as well as the de and ee, was observed
carbonyl oxygen or one of the sulfonyl oxygens, respectively, when TMSOTf (100 mol-%) was used as an additive with
might explain the excellent stereochemical outcome in the 10 mol-% of catalyst 4a-Cu(OTf)2 or 4a-Zn(OTf)2. The re-
stoichiometric catalyst reactions.[7a] Similarly, a strong sults are shown in Table 1. The best results were obtained
(O,O) chelate for the HDA adducts 3a or 3b and the chiral for 1a with 4a-Cu(OTf)2 (10 mol-%) giving a 68% yield of
Lewis acid will also explain an inhibited catalyst turnover endo-3a (98% ee, Ͼ90% de; Table 1, Entry 4), and for 1b
in the catalytic reactions. The use of TMSOTf as an addi- with 4a-Zn(OTf)2 (10 mol-%), giving an 86% yield of endo-
tive in combination with 4c-Cu(OTf)2 (Figure 1) has pre- 3b (97% ee, Ͼ90% de; Table 1, Entry 14). The de’s
viously been reported by Evans et al. in the catalytic enan- (endo/exo ratios) of the products were determined by 1H
Table 1. The effect of achiral additives (100 mol-%) on the HDA reactions of 1a or 1b with 1,3-cyclohexadiene with 10 mol-% of 4a-
Cu(OTf)2 or 4a-Zn(OTf)2.
Entry N-Sulfine
Additive
Chiral Lewis acid
T [°C]
Time [h] Yield (%)[a] Config. of endo-3
endo/exo[b]
% ee[c]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
1a
1b
1a
1a
1a
1b
1b
1a
1b
1a
1a
1b
1a
1b
1b
1b
1a
1b
1b
–
–
ent-4a-Cu(OTf)2
ent-4a-Cu(OTf)2
4a-Cu(OTf)2
4a-Cu(OTf)2
4a-Cu(OTf)2
4a-Cu(OTf)2
4a-Cu(OTf)2
4a-Cu(OTf)2
4a-Cu(OTf)2
4a-Cu(OTf)2
ent-4a-Zn(OTf)2
ent-4a-Zn(OTf)2
4a-Zn(OTf)2
4a-Zn(OTf)2
4a-Zn(OTf)2
4a-Zn(OTf)2
4a-Zn(OTf)2
4a-Zn(OTf)2
4a-Zn(OTf)2
–55
–85
–75
–75
–75
–75
–75
–75
–75
–75
–75
–75
–75
–75
–75
–75
–75
–75
–75
22
24
22
4
1
22
4
4
4
4
22
22
22
22
4
1
4
25
93
85
68
54
56
60
86
74
15
30
39
70
86
83
82
62
66
25
(1R,2S,4S)
(1R,2S,4S)
(1S,2R,4R)
(1S,2R,4R)
(1S,2R,4R)
(1S,2R,4R)
(1S,2R,4R)
(1S,2R,4R)
(1S,2R,4R)
–
(1R,2S,4S)
(1R,2S,4S)
(1S,2R,4R)
(1S,2R,4R)
(1S,2R,4R)
(1S,2R,4R)
(1S,2R,4R)
(1S,2R,4R)
(1S,2R,4R)
38:62
82:18
Ͼ95:Ͻ5
Ͼ95:Ͻ5
90:10
92:8
90:10
Ͼ95:Ͻ5
92:8
55:45
75:25
92:8
15
36
89
98
94
80
75
88
80
–
61
78
86
97
96
94
90
82
47
TMSOTf
TMSOTf
TMSOTf
TMSOTf
TMSOTf
TIPSOTf
TIPSOTf
DME
–
–
TMSOTf
TMSOTf
TMSOTf
TMSOTf
TIPSOTf
TIPSOTf
DME
92:8
Ͼ95:Ͻ5
Ͼ95:Ͻ5
Ͼ95:Ͻ5
Ͼ95:Ͻ5
83:17
4
4
84:16
1
[a] Isolated yield. [b] Determined by H NMR (400 MHz) of the crude product. [c] Determined by chiral HPLC.
5250
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 5249–5259

Hetero Diels–Alder Reactions of N-Sulfinyl Dienophiles
FULL PAPER
NMR (400 MHz) of the crude products, and a de Ͼ90%
means that only one product was observed. The ee in the
reaction of 1a and 2, catalyzed by 4a-Cu(OTf)2, was de-
pendent on the reaction time. It appears from the results in
Table 1 that the optimum reaction time is 4 h (98% ee,
Table 1, Entry 4) and that longer reaction times erode the
ee of endo-3a (22 h, 89% ee, Table 1, Entry 3). For 1b, cata-
lyzed by 4a-Zn(OTf)2, the ee was unaffected by a long reac-
tion time (22 h, 97% ee, Table 1, Entry 14). Therefore, a
reaction time of 4 h was chosen for both test reactions.
Evaluation of triisopropylsilyl trifluoromethanesulfonate
(TIPSOTf) and 1,2-dimethoxyethane (DME) as additives in
the HDA test reactions revealed that TIPSOTf has an effect
(for 1a: 88% ee, Table 1, Entry 8; for 1b: 82% ee, Table 1,
Entry 18) but generally gives less enantioselectivity com-
pared to TMSOTf. DME, expected to coordinate competi-
tively with the Lewis acid and to help the release of the
catalyst from the HDA adduct, improved neither the cata-
lyst turnover nor the ee of the HDA products (1a: Table 1,
Entry 10; 1b: Table 1, Entry 19).
Figure 2. The effect of TMSOTf loading on the ee value (%) of the
reaction of 1a or 1b with 2 in the presence of 10 mol-% of 4a-
Zn(OTf)2.
The effect of catalyst loading on the yield and % ee of
The relationship between the TMSOTf loading and the the HDA reactions of 1a or 1b and 2 with TMSOTf
ee of the HDA products endo-3a and endo-3b was investi- (100 mol-%) in CH2Cl2 at –75 °C (4 h) were tested. The re-
gated for the test reactions (Scheme 1) catalyzed by 10 mol- lationship between catalyst loading and ee is shown in Fig-
% of 4a-Zn(OTf)2, as shown in Figure 2. For both reac- ure 3. Unfortunately, an attempt to decrease the loading of
tions, the optimum additive loading was 100 mol-%. Since 4a-Zn(OTf)2 to 5 mol-% (73% yield and 83% ee) and
TMSOTf is a Lewis acid,[14] and loadings over 100 mol-% 2 mol-% (69% yield and 63% ee) reduced both the yield
eroded the ee of the products, a non-stereoselective, cata- and ee of endo-3b. This was also the case with the 4a-Cu-
lyzed, background reaction may explain the lowered (OTf)2-catalyzed reaction of 1a and 2 when catalyst load-
enantioslectivities.
ings of 10 mol-% (85% yield and 96% ee), 5 mol-% (72%
Several solvents were screened with respect to stereo- yield and 91% ee), 2 mol-% (62% yield and 89% ee), and
chemical outcome of the HDA test reactions (Scheme 1 and 1 mol-% (44% yield and 85% ee) were tested. From this
Table 2). Solutions of the catalysts 4a-Cu(OTf)2 and 4a- analysis, a catalyst loading of 10 mol-% for both catalysts
Zn(OTf)2 in CH2Cl2, toluene, THF, and a 1:1 mixture of was used in further studies.
CH2Cl2 and toluene were generated by the standard pro-
The relationship between the enantiomeric purity of the
cedure. The solvent survey (Table 2) revealed that CH2Cl2 chiral ligand 4a and the ee of the product was investigated
was the optimal solvent with respect to yields and stereose- for the HDA test reactions shown in Scheme 1. In this
lectivity (1a: 85% yield, 89% ee, Table 2, Entry 1; 1b: 86% study, 10 mol-% of the catalyst and 100 mol-% of TMSOTf
yield, 97% ee, Table 2, Entry 5). Toluene gave moderate ee were used. The results are plotted in Figure 4. A positive
(1a: 80% ee, Table 2, Entry 2; 1b: 56% ee, Table 2, Entry nonlinear effect (NLE)[15] was observed for the 4a-Zn-
6), but the chiral Lewis acids were only slightly soluble in (OTf)2-catalyzed reaction of 1b and 2. The deviation from
this system, which was reflected in low product yields (28% linearity indicates that the active catalyst is a dimer or
and 31%). However, when toluene was mixed with CH2Cl2, higher-order aggregate, and the homochiral species giving
the yield doubled (Table 2, Entries 3 and 7). THF was not enantiomeric enrichment was more active than the hetero-
a suitable solvent. In fact, storing the N-sulfines in THF in chiral species giving racemic products. For the 4a-Cu-
the freezer decomposed these dienophiles.
(OTf)2-catalyzed reaction of 1a and 2, a linear relationship
Table 2. The effect of solvents on the HDA reactions of 1a or 1b and 1,3-cyclohexadiene (2) with 10 mol-% of 4a-Cu(OTf)2 or 4a-Zn(OTf)2
and 100 mol-% of TMSOTf at –75 °C for 22 h.
Entry
N-Sulfine
Solvent
Chiral Lewis acid
Yield (%)[a]
endo/exo[b]
% ee[c]
1
2
3
4
5
6
7
8
1a
1a
1a
1a
1b
1b
1b
1b
CH2Cl2
toluene
toluene/CH2Cl2, 1:1
THF
CH2Cl2
toluene
toluene/CH2Cl2, 1:1
THF
4a-Cu(OTf)2
4a-Cu(OTf)2
4a-Cu(OTf)2
4a-Cu(OTf)2
4a-Zn(OTf)2
4a-Zn(OTf)2
4a-Zn(OTf)2
4a-Zn(OTf)2
85
28
61
10
86
31
60
15
Ͼ95:Ͻ5
Ͼ95:Ͻ5
Ͼ95:Ͻ5
89:11
Ͼ95:Ͻ5
90:10
89
80
86
–
97
56
92
–
Ͼ95:Ͻ5
55:45
1
[a] Isolated yield. [b] Determined by H NMR (400 MHz) of the crude product. [c] Determined by chiral HPLC.
Eur. J. Org. Chem. 2006, 5249–5259
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5251

M. M. Endeshaw, A. Bayer, L. K. Hansen, O. R. Gautun
FULL PAPER
favored 3-substituted 1,2-thiazine 1-oxides at low tempera-
ture (5 °C) and the less crowded, thermodynamically
favored, 6-substituted products at higher temperature
(80 °C).[2e,18]
Figure 3. The effect of 4a-Cu(OTf)2 or 4a-Zn(OTf)2 loading on the
% ee of the HDA reactions of 1a or 1b, respectively, and 1,3-cyclo-
hexadiene (2) with TMSOTf (100 mol-%) as an achiral additive.
Scheme 2. HDA reactions of 1a or 1b with acyclic dienes 7–10.
or a slightly positive NLE was observed. The result is in
accordance with the linear relationship found by Evans et
al. for 4a-Cu(SbF6)2-catalysed HDA and ene reactions.[16]
However, the result obtained herein deviates from our find-
ings for the same reaction with a stoichiometric amount of
catalyst, where a positive NLE was observed.[7a]
The uncatalyzed HDA reaction of 10[17,19] with N-sulfine
1a or 1b at room temperature for 22 h afforded mixtures of
stereoisomers and regioisomers. For 1a, a mixture of cis-
14a, trans-14a, cis-15a, and trans-15a was formed in a ratio
of 1:3.1:1:7.1 and in 86% total yield. For 1b, a mixture of
cis-14b, cis-15b, and trans-15b was formed in a ratio of
3.3:1.7:1 and in 90% total yield. The effect of adding stoi-
chiometric amounts of 4a-Cu(OTf)2 to these reactions was
amazing. For both 1a and 1b, only cis-14a (31% yield and
42% ee, Table 3, Entry 17) and cis-14b (50% yield and 71%
ee, Table 3, Entry 19), respectively, were observed by 1H
NMR of the crude products.
In general, stoichiometric amounts of 4a-Cu(OTf)2 pro-
vided cycloadducts 11–14 in better stereoselectivities (40–
97% ee and 66% to Ͼ90% de) and yields (29–68%) than
the 4a-Zn(OTf)2-promoted reactions (26–58% ee, 74% to
Ͼ90% de, and 19–40% yield). Furthermore, higher ee’s
were observed for N-sulfine 1b with dienes 7 (97% ee,
Table 3, Entry 4), 8 (87% ee, Table 3, Entry 9), 9 (92% ee,
Table 3, Entry 15), and 10 (71% ee, Table 3, Entry 19) than
Figure 4. Investigation of nonlinear effects for the HDA reactions for the reaction of 1a (7: 70% ee, 8: 77% ee, 9: 40% ee, 10:
of 1a or 1b and 2, catalyzed by 4a-Cu(OTf)2 (10 mol-%), or 4a-
42% ee, Table 3, Entries 1, 6, 12, and 17, respectively). The
Zn(OTf)2 (10 mol-%), respectively, in the presence of TMSOTf
direction of the stereoselectivity was the same in all reac-
(100 mol-%).
tions where the absolute configuration was determined and
the catalysts were working. The ee was, in general, deter-
mined by chiral HPLC analysis (with either a Chiralpak
AD or Chiracel OD-H column) of the initial cycloadducts.
Dienes
The results of the 4a-Cu(OTf)2-catalyzed reaction of N- For the HDA products cis-13a and cis-13b, a ring opening
sulfine 1a or 1b with the acyclic dienes 7–10 (see Scheme 2) to 16a and 16b, respectively, with phenylmagnesium bro-
are presented in Table 3. Under the influence of the Lewis mide was necessary in order to determine the % ee by chiral
acids, reactions with the dienes 7–10 provided the cis iso- HPLC analysis (Scheme 3).
mers as the major products. For the unsymmetrical dienes
Attempts to perform the HDA reaction of 1a and 8 with
8 and 9,[17] only the 3-substituted regioisomers 12 and 13, catalytic amounts (10 mol-%) of either 4a-Cu(OTf)2 or 4a-
respectively, were formed. Similar results have been pub- Zn(OTf) at –45 °C for 22 h, without additive, afforded race-
lished for the uncatalyzed HDA reactions of N-sulfines 1 mic trans-12a (Ͼ90% de) in Ϸ 20% yield (Scheme 2). The
(X = Ts, COTol) with 1-substituted dienes (RЈ = Me, Ph, p- effect of adding TMSOTf (100 mol-%) to the 4a-Cu(OTf)2-
NO2Ph, p-MeOPh), which usually provided the kinetically catalyzed reaction was remarkable. A switch in diastereo-
5252
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 5249–5259

Hetero Diels–Alder Reactions of N-Sulfinyl Dienophiles
FULL PAPER
Table 3. Selected catalyzed HDA reactions of 1a or 1b with acyclic dienes 7–10.
Entry N-Sulfine
Lewis acid (mol-%)
T [°C]
time [h] HDA adduct Yield (%)[a] Config. of cis isomer cis/trans[b]
% ee[c]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
1a
1a
1a
1b
1b
1a
1a
1a
1b
1b
1b
1a
1a
1a
1b
1b
1a
1a
1b
1b
4a-Cu(OTf)2 (100)
4a-Cu(OTf)2 (10)[d]
4b-Cu(OTf)2 (10)[d]
ent-4a-Cu(OTf)2 (100)
4a-Cu(OTf)2 (10)[d]
ent-4a-Cu(OTf)2 (100)
4a-Cu(OTf)2 (10)[d]
4b-Cu(OTf)2 (10)[d]
ent-4a-Cu(OTf)2 (100)
4a-Cu(OTf)2 (10)[d]
4a-Zn(OTf)2 (10)[d]
4a-Cu(OTf) (100)
–45
–45
–45
–85
–75
–45
–45
–45
–85
–75
–75
–40
–40
–40
–45
–45
–40
–45
–45
–45
22
4
4
25
4
22
4
4
22
2
4
22
4
6
22
4
22
4
22
4
11a
11a
11a
11b
11b
12a
12a
12a
12b
12b
12b
13a
13a
13a
13b
–
72
39
42
68
67
60
47
43
68
15
69
43
84
38
68
0
(1R,3S,6R)
(1R,3S,6R)
(1R,3S,6R)
(1S,3R,6S)
–
(1S,3R)
(1R,3S)
(1R,3S)
(1S,3R)
(1S,3R)
Ͼ95:Ͻ5
Ͼ95:Ͻ5
Ͼ95:Ͻ5
Ͼ95:Ͻ5
Ͼ95:Ͻ5
Ͼ95:Ͻ5
Ͼ95:Ͻ5
Ͼ95:Ͻ5
73:27
Ͼ95:Ͻ5
75:25
83:17
77:23
94:6
70
60
66
97
0
77
81
74
87
3
(1S,3R)
8
[e]
40
43
18
92
–
42
30
71
–
4a-Cu(OTf)2 (10)[d]
6-Cu(OTf)2 (10)[d]
4a-Cu(OTf)2 (100)
4a-Cu(OTf)2 (10)[d]
4a-Cu(OTf)2 (100)
4a-Cu(OTf)2 (10)[d]
4a-Cu(OTf)2 (100)
4a-Cu(OTf)2 (10)[d]
[e]
[e]
(1R,3R)
Ͼ95:Ͻ5
–
–
[e]
14a
14a
14b
–
31
46
50
0
Ͼ95:Ͻ5[f]
Ͼ95:Ͻ5[f]
Ͼ95:Ͻ5[f]
–
[e]
[e]
–
[a] Isolated yield. [b] Determined by 1H NMR (400 MHz) of the crude product. [c] Determined by chiral HPLC. [d] Reaction with
1
100 mol-% of TMSOTf. [e] The absolute configuration was not determined. [f] Only the cis-14 HDA product was observed by H NMR
(400 MHz) of the crude product.
favorably with 4a in either stereoselectivity or yield. Bolm
and Simic´’s new and promising C2-symmetric bis(sulfox-
imine) 6 did not improve the ee’s in the tested reactions (e.g.
Table 3, Entry 14).[21] In Cu(OTf)2-catalyzed reactions with
ligands 5a[22] and 5b,[23] no products were obtained in either
case.
Not surprisingly, attempts to react either sulfinyl com-
Scheme 3. Ring opening of cis-13 to 16, in order to determine the
pound 1a or 1b with (Z,Z)-1,3-cyclooctadiene under uncat-
alyzed conditions or in the presence of 4a-M(OTf)2 (M =
Cu or Zn) gave no HDA products at all. Little or no reac-
tivity of (Z,Z)-1,3-cyclooctadiene in HDA reactions with
other hetero dienophiles has also been observed by
others.[24]
% ee by chiral HPLC analysis. a) PhMgBr, –60 °C, THF, 16a (65%
yield), 16b (85% yield).
selectivity to cis-12a (Ͼ90% de) in 81% ee and 47% yield
(Table 3, Entry 7) in a shorter reaction time (4 h) showed
that the catalyst was working. Similarly, reactions with the
The general rate enhancement observed in the reactions
dienes 7, 9, and 10 afforded the cis adducts in 39–84% yield, of 1a with catalytic amounts of 4a-Cu(OTf)2 (10 mol-%) in
54% to Ͼ90% de, and 30–60% ee (Table 3, Entries 2, 13, combination with 100 mol-% of TMSOTf (4 h reaction
and 18, respectively). These results are comparable to those time) as compared to reactions with stoichiometric
of the reactions promoted by stoichiometric amounts of 4a- amounts of 4a-Cu(OTf)2 (22 h reaction time), indicates that
Cu(OTf)2 (Table 3, Entries 1, 12, and 17). On the other the role of TMSOTf is more complicated than first ex-
hand, adding TMSOTf to similar experiments with 1b gave pected. In addition to the assumed assistance of the release
rather disappointing results. Either full racemization or no of the chiral catalyst from the HDA adducts, TMSOTf may
cycloaddition at all was observed (Table 3, Entries 5, 10, 16, also dissolve or break down aggregates composed of the
and 20). A competitive, uncatalyzed, HDA reaction may catalyst, N-sulfine, and dienophile. Support for the latter
account for the racemic products.
comes from the nearly linear relationship observed between
Since TMSOTf is a Lewis acid,[14] test reactions were run the enantiomeric purity of the chiral ligand 4a and the ee
with TMSOTf (100 mol-%) as the only promoter. Neither 7 of the product for the HDA test reactions (see Scheme 1
nor 8 provided cycloadducts with 1a at –45 °C in 4 h of and Figure 3) containing 10 mol-% of 4a-Cu(OTf)2 and
reaction time.
100 mol-% of TMSOTf, versus the positive nonlinear effect
The moderate enantioselectivities observed for the reac- found in the stoichiometric reaction.[7a] Furthermore, HDA
tion of acyclic dienes and 1a, catalyzed by 4a-Cu(OTf)2, reactions of 1a and 8, loaded with 100 mol-% of both 4a-
prompted us to screen other chiral ligands (shown in Fig- Cu(OTf)2 and TMSOTf, showed that the reaction was com-
ure 1). The most promising results are shown in Table 3. plete in less than 1 h (cis-12a, 40% yield, Ͼ90% de, and
However, except for the new bis(oxazoline) analogue 4b[20] 83% ee). Prolonged reaction times up to 22 h neither in-
(Table 3, Entries 3 and 8), none of the ligands compared creased the yield nor changed the stereoselectivity.
Eur. J. Org. Chem. 2006, 5249–5259
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5253

M. M. Endeshaw, A. Bayer, L. K. Hansen, O. R. Gautun
FULL PAPER
Application
Recently, Weinreb and co-workers reported the first race-
mic total synthesis of the antitumor marine sponge alkaloid
agelastatin A (Scheme 4).[8] The synthesis started with an
uncatalyzed HDA reaction between cyclopentadiene (17)
and N-sulfinyl methyl carbamate (1c) at 0 °C in benzene to
afford the HDA adduct in high yield (Scheme 4). Since this
compound was prone to a retro Diels–Alder reaction at
room temperature, it was immediately treated with phenyl-
magnesium bromide to produce the allylic sulfoxide 19c in
86% overall yield. The latter compound was further re-
arranged to the olefinic oxazolidinone 20, which, in turn,
controls the construction of the relative configuration of the
chiral part in agelastatin A (C-ring).
Scheme 5. An enantioselective route to oxazolidinone 20: a) 4a-Cu-
(OTf)2 (10 mol-%), TMSOTf (100 mol-%), CH2Cl2, –75 °C, 3 h;
b) MeOH, –75 °C Ǟ room temp., 4 h; c) PhMgBr (2 equiv.),
THF, –60 °C, a)–c) one pot 50% yield; d) HMPT, EtOH, 80 °C,
(3aR,6aS)-20 (64%), 21b (19%).
Upon heating with HMPT in ethanol, the sulfoxide mix-
ture 19b underwent a [2,3]-sigmatropic rearrangement to af-
ford the olefinic oxazolidinone (3aR,6aS)-20[25] in 64%
yield and the uncyclized hydroxy benzyl carbamate 21b in
19% yield. According to Weinreb and co-workers, the ethyl
carbamate relative of the latter compound (21a, Scheme 4)
can be cyclized to 20 with potassium tert-butoxide in high
yield.[8]
Stereochemical Model
We have earlier pointed out that a tetrahedral metal cen-
ter or a stereochemically equivalent geometry may explain
the stereochemical outcome of the HDA reactions of N-
sulfinyl dieneophiles 1a and 1b catalyzed by either 4a-
Zn(OTf)2 or 4a-Cu(OTf)2 when the catalysts were work-
ing.[7a,7b] All thiazine oxides had the R configuration at sul-
fur (see below for the determination of the absolute config-
uration of the HDA adducts), implying that all dienes ap-
proach dienophiles 1a and 1b from the same side. The pos-
tulated model shown in Figure 5 assumes that N-sulfinyl
dienophiles 1a and 1b (last compound is not shown) have a
bidentate coordination to the chiral Lewis acid with the
sulfinyl oxygen and the carbonyl oxygen (for 1a) or one
of the sulfonyl oxygens (for 1b). While the possibility of
tetrahedral geometry for CuII complexes has been the sub-
ject of some debate,[16,26] tetrahedral metal centers are well
known for ZnII complexes.[27]
Scheme 4. Weinreb’s racemic synthesis of the olefinic oxazolidinone
20, a precursor of the C-ring in agelastatin A.[8] a) PhH, 0 °C;
b) PhMgBr, THF, –60 °C, 86% (two steps); c) HMPT, EtOH,
80 °C, 20 (44%), 21a (40%); d) KOtBu, THF, 83%.
Herein, an asymmetric route to oxazolidinone 20, follow-
ing Weinreb’s protocol,[8] is presented (Scheme 5). The
asymmetric HDA reaction between 1a and cyclopentadiene
(17), catalyzed by 10 mol-% of 4a-Cu(OTf)2 in CH2Cl2 at
–75 °C for 3 h, was quenched by the addition of methanol
to produce the allylic sulfoxide 19a (X = OMe). A change
of ring-opening reagents from the original Weinreb proto-
col[8] was necessary due to the incompatibility of CH2Cl2
and phenylmagnesium bromide. Attempts to rearrange 19a
to 20 with hexamethylphosphorous triamide (HMPT) in re-
fluxing ethanol did not work and, therefore, 19a (X = OMe)
was further transformed into 19b (X = Ph). This three-step
procedure, (a–c in one pot) starting from 1a and 17, af-
forded an inseparable 4:1 mixture of two sulfoxide epimers
of 19b in 50% yield. The enantiomeric purity of the major
and minor epimers was determined to be 80% and 16% ee,
respectively, by chiral HPLC analysis.
Figure 5. Postulated intermediate with a tetrahedral arrangement.
Configuration of the Diels–Alder Products
The determination of the absolute configuration of the
HDA adducts endo-3[7c–7d] and cis-11–cis-12[7a] have been
5254
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 5249–5259

Hetero Diels–Alder Reactions of N-Sulfinyl Dienophiles
FULL PAPER
IMS and HRCIMS) were obtained with perfluorokerozene (PFK)
as a standard to provide the reference masses. The elemental analy-
ses were performed at the Institute of Chemical Technology, Prague
Central Laboratories, Czech Republic. HPLC was performed with
a 4.6 mm×25 cm Daicel Chiralpak AD, Chiralcel OD-H, or Chi-
ralcel OJ column. Melting points are reported uncorrected.
described elsewhere. For the thiazine oxides trans-13a,
trans-13b, trans-14a, cis-14b, cis/trans-15a, and cis-15b, the
relative configurations were determined by X-ray analy-
ses.[28] For cis-13b, X-ray analysis revealed the absolute con-
figuration.[28] The absolute configuration of the unstable
thiazine oxide 18 was established by chemical correlation
with the known cyclic carbamate (3aR,6aS)-20[25]
(Scheme 5). The 4:1 mixture of sulfur epimers 19b was re-
arranged and cyclized to (3aR,6aS)-20,[25] and thus, the
major thiazine oxide 18 was found to have a (1S,4R) config-
uration. A major endo-18 product with a (1S,2R,4R) config-
uration fits with the stereochemical model described above.
General Procedure for the Uncatalyzed HDA Reaction: To a solu-
tion of the N-sulfinyl compound 1 (X = Cbz, 0.6 , X = Ts, 0.5 ,
1.8 mmol) in dry CH2Cl2 was added a solution of the diene (1 ,
1.5–2.5 equiv.) in dry CH2Cl2. The reaction was stirred at room
temperature for 24 h under a N2 atmosphere. The solvent was evap-
1
orated in vacuo, and the crude product was analyzed by H NMR
to determine the diastereomeric ratio and, on some occasions, also
the regioisomeric ratio. The crude product was purified by FC.
General Catalyst Preparation. Preparation of the Bis(oxazoline)cop-
per(II) Catalyst [4a-Cu(OTf)2] or the Bis(oxazoline)zinc(II) Catalyst
[4a-Zn(OTf)2]:[10] An oven-dried round-bottomed flask was
charged with copper(II) triflate or zinc(II) triflate (0.025 mmol) un-
der an argon atmosphere. To this mixture, dry CH2Cl2 (2 mL) and
a solution of the phenyl bis(oxazoline) ligand (4a, 0.5 , 52 µL,
0.026 mmol) in CH2Cl2 were added, and the resulting suspension
was stirred for 2 h, at which time, most of the solids had dissolved.
A light green solution was observed in the generation of the cop-
per(II) catalyst, while no color was observed in the preparation of
the zinc catalyst.
Conclusions
In conclusion, catalytic asymmetric HDA reactions of N-
sulfinyl dienophiles with bis(oxazoline)copper(II) and
-zinc(II) triflates have been presented. The presence of the
additive TMSOTf was crucial to achieving catalytic turn-
over. The copper catalyst was found to be more efficient.
Two N-sulfinyl dienophiles, 1a and 1b, were tested through-
out this study. In general, the Cbz-substituted sulfine 1a
was more suitable under catalytic conditions than the Ts-
substituted sulfine 1b. For 1b, a pronounced, competitive,
uncatalyzed, HDA reaction reduced the impact of the cata-
lyst. The present study included both cyclic and acyclic
dienes. For the cyclic dienes, good yields and up to 98% ee
was obtained. The acyclic dienes, however, were less suc-
cessful under the catalytic conditions. The configurations
of several cycloadducts have been determined and can be
explained by a stereochemical model proposing a tetrahe-
dral metal center. The synthetic potential of the reaction
has been demonstrated in the enantioselective synthesis of
(3aR,6aS)-3,3a,4,6a-tetrahydrocyclopenta[d][1,3]oxazol-2-
one, a precursor of the C-ring in agelastatin A.
General Procedure for the Asymmetric HDA Reaction: A solution
of the N-sulfinyl compound 1 (620 µL, 0.4 , 0.248 mmol) in
CH2Cl2 was added to a precooled solution of the catalyst at –75 °C.
A precooled solution of the diene (2 equiv.) in CH2Cl2 (500 µL)
was added to the reaction mixture, followed by the addition of
TMSOTf (1 equiv.). The diene solution was added slowly along the
wall of the round-bottomed flask. The reaction mixture was stirred
at –75 °C for 1–22 h and quenched by the addition of phosphate
buffer (pH 7, 4 mL). The reaction was warmed to room tempera-
ture and was extracted with CH2Cl2 (3×4 mL). The combined or-
ganic layers were dried (MgSO4), and the solvent was removed in
1
vacuo. H NMR of the crude product revealed the diastereomeric
ratio. The crude was purified by FC, and the ee was determined by
chiral HPLC.
Experimental Section
HDA Reactions for the Investigation of the Nonlinear Effect (10 mol-
% of Catalyst): Catalysts of different enantiomeric compositions
were prepared according to the procedure described above. The
scalemic mixtures of the bis(oxazoline) ligand were obtained by
mixing 4a (0.5  in CH2Cl2) and ent-4a (ent = enantiomer, 0.5 
in CH2Cl2) under an argon atmosphere before addition to the reac-
tion. The HDA reactions were performed at –75 °C for 4 h, as de-
scribed in the general procedure given above.
General: Benzyl N-sulfinylcarbamate (1a)[9] and N-sulfinyl-4-tol-
uenesulfonamide (1b)[2b] were prepared according to the liter-
ature procedure and stored in the freezer as solutions in dry
CH2Cl2. (S,S)-2,2Ј-Isopropylidenebis(4-phenyl-2-oxazoline) (4a),[10]
[S(S),SЈ(S)]-N,NЈ-(2,6-pyridinediyldimethylidyne)bis[p-toluenesulf-
inamide] (5b),[23] (S,S)-1,2-bis(S-methyl-S-phenylsulfonimidoyl)-
benzene (6),[21] (E)-1-phenyl-1,3-butadiene (9),[17] and (1E,3E)-1-
methyl-4-phenylbutadiene (10),[17,19] were prepared according to
the literature procedure. Analytical data for endo-3, exo-3,[7d] cis-
11, trans-11, cis-12, and trans-12[7a] are described elsewhere. Sol-
vents were dried according to standard procedures.[29] TLC was
performed on silica gel plates and visualized with UV light and a
phosphomolybdic acid in ethanol solution. Reaction products were
purified by flash chromatography (FC) with silica gel (35–70 µm).
1H NMR (300 MHz, 400 MHz, and 600 MHz) and 13C NMR
(75 MHz, 100 MHz, and 150 MHz) spectra were obtained from
solutions of CDCl3, and chemical shifts were assigned by 2D corre-
lation techniques. The electron-impact mass spectra (EIMS) were
recorded at 70 eV with a direct inlet and the chemical ionozation
mass spectra (CIMS) were obtained with methane (ionized at
200 eV) as the carrier gas. The high resolution mass spectra (HRE-
Benzyl 3,6-Dihydro-3-phenyl-1λ4,2-thiazine-2-carboxylate (13a): An
asymmetric HDA reaction between 1a and (E)-1-phenyl-1,3-buta-
diene (9),[17] catalyzed by 4a-Cu(OTf)2 (10 mol-%) according to the
general procedure, afforded a mixture of cis-13a and trans-13a
(3.4:1). FC (EtOAc/pentane, 50:50) of the crude product yielded
105.4 mg (65% yield) of cis-13a as a colorless, viscous oil, and
31.0 mg (19% yield) of trans-13a as a white solid. Analytical data
for cis-13a: Rf (EtOAc/pentane, 50:50) = 0.28. [α]2D0 = +7.9 (c = 1,
1
CH2Cl2). H NMR (600 MHz, CDCl3): δ = 7.50 (d, J = 7.2 Hz, 2
H, Ph), 7.34–7.30 (m, 4 H, Ph), 7.28–7.25 (m, 4 H, Ph), 6.05 (dt, J
= 10.2, 3 Hz, 1 H, 4-H), 5.82–5.79 (m, 1 H, 5-H), 5.46 (br. s, 1 H,
3-H), 5.08 (AB, J = 12.5 Hz, 2 H, Bn), 3.57 (ddd, J = 8.4, 7.2,
1.2 Hz, 1 H, 6-H), 3.46–3.42 (m, 1 H, 6-H) ppm. 13C NMR
(150 MHz, CDCl3): δ = 154.2 (C=O), 140.2, 135.2, 130.3 (4-C),
Eur. J. Org. Chem. 2006, 5249–5259
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5255

M. M. Endeshaw, A. Bayer, L. K. Hansen, O. R. Gautun
FULL PAPER
129.3, 128.8, 128.7, 128.6, 128.4, 127.9, 112.4 (5-C), 68.9 (Bn), 58.6
(s, 3 H, Me) ppm. 13C NMR (150 MHz, CDCl3): δ = 144.0, 137.3,
(3-C), 48.7 (6-C) ppm. IR (KBr): ν = 3059 (w), 1708 (s, C=O), 1494 136.8, 130.3 (4-C), 129.4, 129.38, 128.5, 127.9, 127.1, 112.1 (5-C),
˜
(m), 1452 (m), 1439 (m), 1380 (m), 1299 (s), 1199 (m), 1147 (m), 59.8 (3-C), 49.8 (6-C), 21.3 (Me) ppm. IR (KBr): ν = 3056 (w),
˜
1093 (m) cm–1. EIMS: m/z (%) = 328 (2), 327 (9) [M]+, 279 (6), 235 1598 (m), 1494 (m), 1453 (m), 1355 (s), 1303 (m), 1183 (m), 1166
(9), 131 (11), 130 (100), 129 (77), 128 (43), 115 (29), 107 (21), 91
(91). C18H17NO3S (327.09): calcd. C 66.03, H 5.23, N 4.28; found
C 65.79, H 5.35, N 4.21. The optical purity of the product was
determined to be 43% ee by transformation to 16a (experimental
details are shown below). Analytical data for trans-13a: Rf (EtOAc/
pentane, 50:50) = 0.18. 1H NMR (600 MHz, CDCl3): δ = 7.37–
7.27 (m, 6 H, Ph); 7.22–7.15 (m, 4 H, Ph), 6.33 (ddd, J = 5.4, 4,
(s), 1101 (m), 1067 (m) cm–1. EIMS: m/z (%) = 347 (0.1) [M]+, 260
(7), 172 (5), 171 (50), 155 (45), 130 (14), 129 (14), 108 (13), 107
(16), 91 (100), 65 (20). C17H17NO3S2 (347.06): C 58.77, H 4.93, N
4.03; found C 58.99, H 5.21, N 3.73. The absolute configuration of
cis-13b (1R,3R) was corroborated by X-ray crystallographic analy-
sis.[28] The optical purity of the product was determined to be 92%
ee by transformation to 16b (experimental details are shown be-
2 Hz, 1 H, 4-H), 5.81–5.78 (m, 1 H, 5-H); 5.53 (d, J = 5.4 Hz, 1 low). A racemic sample of cis-13b melted at 116–118 °C (from
H, 3-H), 5.15 (AB, J = 9.6 Hz, 2 H, Bn), 3.61 (m, 2 H, 6-H) ppm.
13C NMR (150 MHz, CDCl3): δ = 154.9 (C=O), 140.3, 135.0, 132.5
(4-C), 129.1, 128.7, 128.6, 128.2, 127.8, 126.1, 112.5 (5-C), 69.3
CH2Cl2/pentane/EtOAc). The uncatalyzed HDA reaction between
1b and 9 afforded a mixture of cis-13b and trans-13b (3:1). FC
(EtOAc/pentane, 50:50) of the crude product yielded 375.0 mg
(60% yield) of cis-13b as a white solid and 125.0 mg (20% yield)
(Bn), 57.5 (3-C), 48.1 (6-C) ppm. IR (KBr): ν = 3059 (w), 1708 (s,
˜
C=O), 1494 (m), 1452 (m), 1439 (m), 1380 (m), 1299 (s), 1199 (m), of trans-13b as a white solid. Analytical data for trans-13b
1147 (m), 1093 (m) cm–1. EIMS: m/z (%) = 327 (6) [M]+, 279 (13),
235 (10), 130 (90), 129 (63), 128 (32), 127 (11), 115 (28), 107 (18),
91 (100). C18H17NO3S (327.09): calcd. C 66.03, H 5.23 N 4.28;
found C 65.75, H 5.52, N 4.41. The relative configuration of trans-
13a (1R*,3S*) was corroborated by X-ray crystallographic analy-
sis.[28] A racemic sample melted at 132–133 °C (from CH2Cl2/pen-
tane/EtOAc).
(1R*,3S*): Rf (EtOAc/pentane, 50:50) = 0.17. M.p. 121–123 °C
(from CH2Cl2/pentane/EtOAc). H NMR (600 MHz, CDCl3): δ =
1
7.37 (app. d, J = 9 Hz, 2 H, Ts), 7.22 (app. tt, J = 7.2, 2.4 Hz, 1
H, Ph), 7.18–7.15 (m, 2 H, Ph), 7.13 (app. d, J = 9 Hz, 2 H, Ts),
7.09–7.07 (m, 2 H, Ph), 6.04 (ddd, J = 7.2, 4.8, 3.0 Hz, 1 H, 4-H),
5.71–5.66 (m, 1 H, 5-H), 5.21–5.19 (m, 1 H, 3-H), 3.74 (app. dq, J
= 16, 4.2 Hz, 1 H, 6-H), 3.62 (ddd, J = 15.9, 8.4, 1.8 Hz, 1 H, 6-
H), 2.4 (s, 3 H, Me) ppm. 13C NMR (150 MHz, CDCl3): δ = 144.5,
137.2, 135.9, 132.4 (4-C), 129.6, 128.7, 128.5, 128.4, 128.0, 111.2
Ring Opening of cis-13a to 16a in Order to Determine the ee Value
by Chiral HPLC: A solution of PhMgBr (1 , 360 µL, 0.36 mmol)
in THF was added to a stirred solution of cis-13a {α]2D0 = +7.9 (c
= 1, CH2Cl2), 119 mg, 0.36 mmol} in dry THF (5 mL) at –60 °C.
The resultant mixture was stirred for 30 min and then hydrolyzed
with saturated aqueous NH4Cl (3 mL). The layers were separated,
and the aqueous layer was extracted with diethyl ether (3×3 mL).
The combined organic layers were washed with brine (3 mL), dried
(MgSO4), and concentrated in vacuo. Purification of the crude
product by FC (EtOAc/pentane, 33:67 to 50:50) yielded the allylic
sulfoxide 16a (95.4 mg, 65%) as a colorless oil. Analytical data for
16a: [α]2D0 = –46.8 (c = 1.0, CH2Cl2). HPLC (Chiralpak AD, iPrOH/
(5-C), 58.7 (3-C), 50.2 (6-C), 21.8 (Me) ppm. IR (KBr): ν = 3060
˜
(w), 1596 (m), 1492 (m), 1453 (s), 1339 (m), 1183 (m), 1113 (m),
1080 (m), 1062 (m), 974 (m), 928 (m) cm–1. CIMS: m/z (%) = 348
(7) [M+1]+, 219 (11), 218 (86), 216 (16), 172 (13), 155 (100), 131
(34), 130 (71), 129 (47), 115 (21), 91 (48). C17H17NO3S2 (347.06):
calcd. C 58.77, H 4.93, N 4.03; found C 58.68, H 4.92, N 4.17. The
relative configuration of trans-13b (1R*,3S*) was corroborated by
X-ray crystallographic analysis.[28]
Ring Opening of cis-13b to 16b in Order to Determine the ee Value
by Chiral HPLC: cis-13b (1R,3R), [α]2D0 = –49.6 (c = 1.0, CH2Cl2),
hexane, 20:80, 1 mL min–1, 230 nm): 43% ee, tR 15.1 min (major) was ring opened by PhMgBr according to the procedure described
1
and 27.0 min. H NMR (400 MHz, CDCl3): δ = 7.59 (br. s, 2 H,
Ph), 7.49–7.40 (m, 3 H, Ph), 7.36–7.28 (m, 8 H, Ph), 7.22–7.17 (m,
2 H, Ph), 5.91 (app. t, J = 8.8 Hz, 1 H, 2-H), 5.67–5.54 (m, 1 H,
above for the ring opening of cis-13a. This afforded 16b (173.6 mg,
85% yield) as a white solid. Analytical data for 16b: [α]2D0 = –222
(c = 1.0, CH2Cl2). HPLC (Chiralcel OJ, iPrOH/hexane, 30:70,
3-H), 5.42 (br. s, 2 H, 1-H/NH), 5.11 (s, 2 H, Bn), 3.88–3.73 (m, 1 0.5 mL min–1, 230 nm): 92% ee, tR 27.9 min [1S,S(R) isomer] and
H, 4-H), 3.72–3.65 (m, 1 H, 4-H) ppm. 13C NMR (100 MHz,
CDCl3): δ = 155.8 (C=O), 143.2, 140.7, 137.4 (2-C), 136.5, 131.3,
129.4, 129.0, 128.8, 128.4, 128.3, 127.9, 126.7, 124.4, 119.8 (3-C),
33.5 min [1R,S(S) isomer]. H NMR (400 MHz, CDCl3): δ = 7.60
(app. d, J = 8.4 Hz, 2 H, Ts), 7.58–7.53 (m, 2 H, Ph), 7.52–7.45
(m, 3 H, Ph), 7.22–7.12 (m, 7 H, Ph/Ts), 6.00 (d, J = 5.2 Hz, 1 H,
1
76.8 (1-C), 67.2 (Bn), 55.1 (4-C) ppm. IR (neat): ν = 3285 (m), NH), 5.85 (dd, J = 10.8, 7.6 Hz, 1 H, 2-H), 5.22 (dq, J = 8.4,
˜
3059 (w), 3030 (w), 2923 (w), 1716 (s), 1533 (m), 1252 (m), 1085
(m), 1036 (s) cm–1. CIMS: m/z (%) = 406 (4) [M+1]+, 281 (11), 280
(35), 279 (37), 255 (17), 240 (10), 220 (13), 218 (16), 188 (18), 172
(17), 171 (14), 130 (12), 129 (34), 126 (11), 125 (19), 109 (11), 108
1.2 Hz, 1 H, 3-H), 4.99 (t, J = 6 Hz, 1 H, 1-H), 3.78 (ddd, J = 8.4,
4.8, 1.2 Hz, 1 H, 4-H), 3.41 (dd, J = 8.0, 5.6 Hz, 1 H, 4-H), 2.38
(s, 3 H, Me) ppm. 13C NMR (100 MHz, CDCl3): δ = 143.3, 142.0,
139.9, 138.7 (2-C), 137.8, 131.6, 129.5, 129.4, 128.8, 127.9, 127.5,
(23), 107 (13), 91 (100). C24H23NO3S (405.14): calcd. C 71.08, H 127.3, 124.4, 118.6 (3-C), 55.4 (1-C), 53.7 (4-C), 21.7 (Me) ppm.
5.72, N 3.45 S 7.91; found C 70.88, H 5.71, N 3.45, S 7.62.
IR (KBr): ν = 3250 (m), 3060 (m), 2920 (w), 1598 (s), 1493 (s),
˜
1443 (s), 1327 (s), 1161 (s) cm–1. CIMS: m/z (%) = 426 (0.6)
[M+1]+, 425 (2) [M]+, 300 (20), 299 (100), 297 (27), 259 (61), 254
(19), 155 (49), 145 (17), 144 (86), 143 (32), 125 (28), 91 (53).
C23H23NO3S2 (425.11): calcd. C 64.91, H 5.45, N 3.29; found C
64.84, H 5.39, N 3.01. A racemic sample of 16b melted at
58–59 °C.
3,6-Dihydro-3-phenyl-2-tosyl-1λ4,2-thiazine 1-Oxide (13b): The
asymmetric HDA reaction between N-sulfinyl 1b and (E)-1-phenyl-
1,3-butadiene (9),[17] catalyzed by 4a-Cu(OTf)2 (100 mol-%) ac-
cording to the general procedure, afforded exclusively cis-13b. FC
(EtOAc/pentane, 50:50) of the crude product yielded 86.2 mg (68%
yield) of cis-13b as a white solid. Analytical data for cis-13b
(1R,3R): Rf (EtOAc/pentane, 50:50) = 0.33. [α]2D0 = –49.6 (c = 1.0,
CH2Cl2). 1H NMR (600 MHz, CDCl3): δ = 7.28–7.26 (m, 2 H, Ph),
Benzyl 3,6-Dihydro-6-methyl-3-phenyl-1λ4,2-thiazine-2-carboxylate
(14a) and Benzyl 3,6-Dihydro-3-methyl-6-phenyl-1λ4,2-thiazine-2-
7.25 (app. d, J = 10.8 Hz, 2 H, Ts), 7.17–7.09 (m, 3 H, Ph), 6.98 carboxylate (15a): The uncatalyzed reaction between 1a and
(app. d, J = 10.8 Hz, 2 H, Ts), 5.99 (dt, J = 10.8, 2.8 Hz, 1 H, 4-
H), 5.81–5.77 (m, 1 H, 5-H), 5.54 (dd, J = 12, 4.2 Hz, 1 H, 3-H),
3.65 (dd, J = 15.6, 7.2 Hz, 1 H, 6-H), 3.53–3.47 (m, 1 H, 6-H), 2.30
(1E,3E)-1-methyl-4-phenylbutadiene (10),[17,19] according to the ge-
neral procedure, afforded a mixture of cis-14a, trans-14a, cis-15a,
and trans-15a in a ratio of 1:3.1:1:7.1. FC (EtOAc/pentane, 33:67
5256
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 5249–5259

Hetero Diels–Alder Reactions of N-Sulfinyl Dienophiles
FULL PAPER
to 50:50) of the crude product yielded 134 mg (22% yield) of trans-
14aas a white solid, 309 mg (50% yield) of trans-15a as white solid,
and a third fraction containing a mixture of cis-14a and cis-15a.
FC (Et2O/CH2Cl2, 5:95) of the latter fraction afforded 42.6 mg (7%
yield) of cis-14a and 42.6 mg (7% yield) of cis-15a, both as white
solids. Analytical data for cis-14a (1R*,3R*,6R*): Rf (EtOAc/pen-
J = 10.8, 3.6 Hz, 1 H, 4-H), 5.96 (ddd, J = 6.8, 4, 1.2 Hz, 1 H, 5-
H), 5.17 (AB, J = 12.4 Hz, 1 H, Bn), 5.12 (AB, J = 12.4 Hz, 1 H,
Bn), 4.76 (dd, J = 6.8, 1.2 Hz, 1 H, 6-H), 4.48–4.42 (m, 1 H, 3-H),
1.50 (d, J = 6.4 Hz, 3 H, Me) ppm. 13C NMR (150 MHz, CDCl3):
δ = 155.6 (C=O), 135.4, 134.4, 132.7, 129.1, 129.0, 128.8, 128.7,
128.4, 127.8, 115.2 (5-C), 68.8 (Bn), 65.1 (6-C), 48.6 (3-C), 20.7
tane, 50:50) = 0.35. M.p. 121–122 °C (from CH2Cl2/pentane). (Me) ppm. IR (KBr): ν = 3033 (w), 1723 (s, C=O), 1496 (m), 1455
˜
HPLC (Chiralpak AD, iPrOH/hexane, 40:60, 0.7 mL min–1,
230 nm): tR 9.5 min and 11.5 min. H NMR (600 MHz, CDCl3): δ
= 7.48 (d, J = 7.2 Hz, 2 H, Ph), 7.36–7.30 (m, 5 H, Ph), 7.25 (tt, J
(m), 1438 (m), 1381 (m), 1264 (s), 1184 (m), 1089 (m), 1058 (m)
cm–1. EIMS: m/z (%) = 341 (0.7) [M]+, 293 (2), 144 (25), 129 (37),
128 (17), 91 (100). HRCIMS: calcd. for C19H19NO3S [M+1]+
1
= 8.4, 1.2 Hz, 3 H, Ph), 5.97 (dt, J = 8, 2.8 Hz, 1 H, 4-H), 5.48 (dt, 342.1164; found 342.1158. C19H19NO3S (341.11): calcd. C 66.84,
J = 9.6, 3.2 Hz, 1 H, 5-H), 5.40 (dd, J = 6.4, 3.2 Hz, 1 H, 3-H),
5.18 (AB, J = 12 Hz, 1 H, Bn), 5.08 (AB, J = 12 Hz, 1 H, Bn),
H 5.61, N 4.10, S 9.39; found C 66.52, H 5.60, N 4.11, S 9.46. The
relative configuration of trans-15a (1R*,3R*,6R*) was corrobo-
3.43–3.36 (m, 1 H, 6-H), 1.56 (d, J = 7.6 Hz, 3 H, Me) ppm. 13C rated by X-ray crystallographic analysis.[28] The asymmetric HDA
NMR (150 MHz, CDCl3): δ = 154.0 (C=O), 140.2, 135.2, 130.0 (4-
C), 128.8, 128.7, 128.6, 128.4, 127.8, 119.6 (5-C), 68.9 (Bn), 58.5 (100 mol-%) according to the general procedure, afforded exclu-
reaction between 1a and 10,[17,19] promoted by 4a-Cu(OTf)2
(3-C), 52.3 (6-C), 16.1 (Me) ppm. IR (KBr): ν = 3032 (w), 1702 (s, sively cis-14a. FC (EtOAc/pentane, 33:67) of the crude product
˜
C=O), 1493 (m), 1455 (s), 1385 (m), 1375 (m), 1287 (s), 1113 (m), yielded 22.5 mg (31% yield) of cis-14a as white solid. Analytical
1077 (s) cm–1. EIMS: m/z (%) = 341 (4) [M]+, 144 (41), 129 (11), 91 data: [α]2D0 = +2.6 (c = 1.0, CH2Cl2). HPLC (Chiralpak AD, iPrOH/
(100), 84 (18). HRCIMS: calcd. for C19H19NO3S [M+1]+ 342.1164; hexane, 40:60, 0.7 mL min–1, 230 nm): 42% ee, tR 9.5 min (major)
found 342.1166. C19H19NO3S (341.11): calcd. C 66.84, H 5.61, N
4.10, S 9.39; found C 66.58, H 5.65, N 4.05, S 9.38. Analytical data
for trans-14a (1R*,3S*,6S*): Rf (EtOAc/pentane, 50:50) = 0.11.
M.p. 110–111 °C (from CH2Cl2/pentane). HPLC (Chiralpak AD,
iPrOH/hexane, 40:60, 0.7 mL min–1, 230 nm): tR 11.6 min and
13.7 min. 1H NMR (600 MHz, CDCl3): δ = 7.39–7.24 (m, 8 H,
Ph), 7.14–7.12 (m, 2 H, Ph), 5.95 (dd, J = 10.8, 3.6 Hz, 1 H, 4-H);
5.77 (ddd, J = 6.8, 4, 1.6 Hz, 1 H, 5-H), 5.35–5.34 (m, 1 H, 3-H),
5.15 (AB, J = 12 Hz, 1 H, Bn), 5.06 (AB, J = 12 Hz, 1 H, Bn),
3.56–3.49 (m, 1 H, 6-H), 1.56 (d, J = 7.2 Hz, 3 H, Me) ppm. 13C
NMR (150 MHz, CDCl3): δ = 156.0 (C=O), 140.1, 135.0, 130.0 (4-
C), 129.1, 128.7, 128.6, 128.2, 127.6, 126.6, 117.6 (5-C), 69.3 (Bn),
and 11.5 min.
3,6-Dihydro-6-methyl-3-phenyl-2-tosyl-1λ4,2-thiazine 1-Oxide (14b)
and 3,6-Dihydro-3-methyl-6-phenyl-2-tosyl-1λ4,2-thiazine 1-Oxide
(15b): The uncatalyzed reaction between N-sulfinyl compound 1b
and 10,[17,19] according to the general procedure, afforded a mixture
of cis-14b, cis-15b, and trans-15b. FC (EtOAc/pentane, 33:67 to
50:50) of the crude product yielded 108.9 mg (15% yield) of trans-
15b as a white solid, and a fraction containing a mixture of cis-
14b and cis-15b. FC (EtOAc/pentane, 20:80) of the latter fraction
afforded 363 mg (50% yield) of cis-14b and 181.5 mg (25% yield)
of cis-15b, both as white solids. Analytical data for cis-14b
(1R*,3R*,6R*): Rf (EtOAc/pentane, 50:50) = 0.29. M.p. 134–
136 °C (from CH2Cl2/pentane/EtOAc). HPLC (Chiralcel OD-H,
iPrOH/hexane, 20:80, 0.7 mL min–1, 230 nm): tR 11.18 min and
15.06 min. 1H NMR (600 MHz, CDCl3): δ = 7.42–7.30 (m, 2 H,
Ph), 7.24 (app. d, J = 7.2 Hz, 2 H, Ts), 7.16–7.08 (m, 3 H, Ph),
6.97 (app. d, J = 7.2 Hz, 2 H, Ts), 5.93 (dt, J = 10.8, 2.8 Hz, 1 H,
4-H), 5.51–5.45 (m, 2 H, 3-H/5-H), 3.55–3.49 (m, 1 H, 6-H), 2.30
(s, 3 H, TsMe), 1.57 (d, J = 7.2 Hz, 3 H, Me) ppm. 13C NMR
(150 MHz, CDCl3): δ = 144.0, 137.4, 136.9, 130.1 (4-C), 129.6,
129.4, 128.6, 127.9, 127.2, 119.6 (5-C), 60.0 (3-C), 54.4 (6-C), 21.7
56.7 (3-C), 54.2 (6-C), 16.7 (Me) ppm. IR (KBr): ν = 3062 (w),
˜
1732 (s, C=O), 1495 (m), 1449 (m), 1384 (m), 1281 (s), 1247 (s),
1097 (s), 1051 (s) cm–1. EIMS: m/z (%) = 341 (1) [M]+, 151 (6), 144
(55), 129 (100), 128 (52), 115 (20), 107 (37), 91 (95). HRCIMS:
calcd. for C19H19NO3S [M+1]+ 342.1164; found 342.1164.
C19H19NO3S (341.11): calcd. C 66.84, H 5.61, N 4.10, S 9.39; found
C 66.82, H 5.52, N 4.13, S 9.60. The relative configuration of trans-
14a (1R*,3S*,6S*) was corroborated by X-ray crystallographic
analysis.[28] Analytical data for cis-15a (1R*,3S*,6S*): Rf (EtOAc/
pentane, 50:50) = 0.48. M.p. 123–124 °C (from CH2Cl2/pentane).
HPLC (Chiralpak AD, iPrOH/hexane, 40:60, 0.7 mL min–1,
(TsMe), 16.3 (Me) ppm. IR (KBr): ν = 3041 (w), 1597 (m), 1493
˜
(m), 1455 (m), 1355 (s), 1164 (s), 1113 (m), 1049 (m), 947 (s) cm–1.
EIMS: m/z (%) = 361 (0.2) [M]+, 219 (2), 217 (25), 129 (100), 91
(100), 65 (31). C18H19NO3S2 (361.08): calcd. C 59.81, H 5.30, N
3.87; found C 59.76, H 5.24, N 3.89. The relative configuration of
cis-14b (1R*,3R*,6R*) was corroborated by X-ray crystallographic
analysis.[28] Analytical data for cis-15b (1R*,3S*,6S*): Rf (EtOAc/
pentane, 50:50) = 0.37. M.p. 137–138 °C (from CH2Cl2/pentane/
EtOAc). HPLC (Chiralpak AD, iPrOH/hexane, 10:90,
1.0 mL min–1, 230 nm): tR 17.3 min and 19.4 min. 1H NMR
(600 MHz, CDCl3): δ = 7.86 (app. d, J = 8.4 Hz, 2 H, Ts), 7.43–
7.31 (m, 7 H, Ph/Ts), 6.09 (dt, J = 10.8, 3.2 Hz, 1 H, 4-H), 5.75
(dt, J = 11.2, 2.0 Hz, 1 H, 5-H), 4.68–4.62 (m, 1 H, 3-H), 4.42 (dd,
J = 5.2, 2.8 Hz, 1 H, 6-H), 2.45 (s, 3 H, TsMe), 1.38 (d, J = 7.2 Hz,
3 H, Me) ppm. 13C NMR (150 MHz, CDCl3): δ = 144.0, 137.4,
136.9, 130.1 (4-C), 129.6, 129.4, 128.6, 127.9, 127.2, 119.6, 60.0 (3-
1
230 nm): tR 7.0 min and 8.1 min. H NMR (600 MHz, CDCl3): δ
= 7.43–7.34 (m, 10 H, Ph), 6.14 (dt, J = 11.4, 3.0 Hz, 1 H, 4-H),
5.76 (dt, J = 11.1, 2.4 Hz, 1 H, 5-H), 5.31 (AB, J = 12.6 Hz, 1 H,
Bn), 5.28 (AB, J = 12 Hz, 1 H, Bn), 4.64–4.60 (m, 1 H, 3-H), 4.52
(dd, J = 6, 3.0 Hz, 1 H, 6-H), 1.54 (d, J = 7.2 Hz, 3 H, Me) ppm.
13C NMR (150 MHz, CDCl3): δ = 153.8 (C=O), 135.4, 134.7, 131.8
(4-C), 129.9, 129.2, 129.1, 128.9, 128.7, 128.4, 118.1 (5-C), 68.8
(Bn), 63.8 (6-C), 49.8 (3-C), 22.4 (Me) ppm. IR (KBr): ν = 3032
˜
(w), 1702 (s, C=O), 1493 (m), 1455 (s), 1385 (m), 1375 (m), 1287
(s), 1113 (m), 1077 (s) cm–1. EIMS: m/z (%) = 341 (2) [M]+, 144
(43), 129 (71), 107 (21), 91 (100), 79 (17). HRCIMS: calcd. for
C19H19NO3S [M+1]+ 342.1164; found 342.1168. C19H19NO3S
(341.11): calcd. C 66.84, H 5.61, N 4.10, S 9.39; found C 66.82, H
5.54, N 4.12, S 9.31. The relative configuration of cis-15a
(1R*,3S*,6S*) was corroborated by X-ray crystallographic analy-
sis.[28] Analytical data for trans-15a (1R*,3R*,6R*): Rf (EtOAc/pen-
tane, 50:50) = 0.19. M.p. 114–115 °C (from CH2Cl2/pentane).
HPLC (Chiralpak AD, iPrOH/hexane, 40:60, 0.7 mL min–1,
C), 54.4 (6-C), 21.7 (TsMe), 16.3 (Me) ppm. IR (KBr): ν = 3012
˜
(w), 1597 (s), 1493 (m), 1453 (m), 1355 (s), 1166 (s), 1112 (m), 1049
(m), 994 (s), 923 (s) cm–1. EIMS: m/z (%) = 361 (0.01) [M]+, 219
(4), 217 (53), 171 (12), 155 (95), 144 (93), 143 (44), 129 (100), 128
(89), 91 (99), 65 (50). C18H19NO3S2 (361.08): calcd. C 59.81, H
5.30, N 3.87; found C 59.70, H 5.24, N 3.89. The relative configura-
1
230 nm): tR 12.6 min and 14.7 min. H NMR (600 MHz, CDCl3):
δ = 7.40–7.28 (m, 8 H, Ph), 7.23 (d, J = 6.6 Hz, 2 H, Ph), 6.35 (dd,
Eur. J. Org. Chem. 2006, 5249–5259
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5257

M. M. Endeshaw, A. Bayer, L. K. Hansen, O. R. Gautun
FULL PAPER
tion of cis-15b (1R*,3S*,6S*) was corroborated by X-ray crystallo-
graphic analysis.[28] Analytical data for trans-15b (1R*,3R*,6R*):
Rf (EtOAc/pentane, 50:50) = 0.16. M.p. 131–132 °C (CH2Cl2/
pentane/EtOAc). HPLC (Chiralpak AD, iPrOH/hexane, 40:60,
0.7 mL min–1, 230 nm): tR 16.4 min and 37.1 min. 1H NMR
(600 MHz, CDCl3): δ = 7.44–7.32 (m, 5 H, Ph), 7.23 (app. d, J =
8 Hz, 2 H, Ts), 7.10 (d, J = 8 Hz, 2 H, Ts), 6.15 (dd, J = 10.8,
2.8 Hz, 1 H, 4-H), 5.88 (dd, J = 10.4, 6.4 Hz, 1 H, 5-H), 4.80 (d, J
= 6.4 Hz, 1 H, 6-H), 4.34–4.31 (m, 1 H, 3-H), 2.39 (s, 3 H, TsMe),
1.40 (d, J = 6.8 Hz, 3 H, Me) ppm. 13C NMR (150 MHz, CDCl3):
δ = 144.4, 136.9, 134.2, 132.5, 129.9, 129.6, 129.3, 129.0, 127.5,
115.4 (5-C), 67.0 (6-C), 49.4 (3-C), 21.8 (TsMe), 19.9 (Me) ppm.
128.3, 128.2, 128.1 (3-C), 124.2, 71.4 (4-C), 66.5 (Bn), 54.0 (1-C),
29.3 (5-C) ppm.
(3aR,6aS)-3,3a,4,6a-Tetrahydrocyclopenta[d][1,3]oxazol-2-one
[(3aR,6aS)-20]: The title compound was prepared from the mixture
of sulfur epimers 19b (103.0 mg, 0.302 mmol) according to the pro-
cedure of Weinreb and co-workers.[8] The crude product was recrys-
tallized from diethyl ether affording (3aR,6aS)-20 (24.2 mg, 64%
yield) as a white solid. Analytical data for (3aR,6aS)-20: m.p. 116–
117 °C (ref.[25] m.p. 117–118 °C). [α]2D0 = +5.8 (c = 0.32, CHCl3)
{ref.[25] [α]D20 = +7.6 (c = 0.32, CHCl3}. 1H NMR (400 MHz, CDCl3):
δ = 6.59 (br. s, 1 H, NH), 6.08–6.06 (m, 1 H, 5-H), 5.87–5.85 (m,
1 H, 6-H), 5.56 (app. d, J = 7.6 Hz, 1 H, 6a-H), 4.45 (app. t, J =
7.3 Hz, 1 H, 3a-H), 2.72–2.64 (m, 1 H, 4-H), 2.47 (m, 1 H, 4-H)
ppm. 13C NMR (100 MHz, CDCl3): δ = 159.7 (C=O), 135.7 (5-C),
128.3 (6-C), 86.3 (6a-C), 53.6 (3a-C), 40.7 (4-C) ppm.
IR (KBr): ν = 3012 (w), 1596 (s), 1493 (s), 1451 (s), 1343 (s), 1188
˜
(s), 1162 (s), 1102 (s), 1087 (s), 1042 (m), 922 (s) cm–1. EIMS: m/z
(%) = 219 (1), 217 (13), 155 (50), 144 (61), 129 (100), 91 (81), 65
(14). C18H19NO3S2 (361.08): calcd. C 59.81, H 5.30, N 3.87; found
C 59.73, H 5.13, N 3.88. The asymmetric HDA reaction between
1b and 10,[17,19] promoted by 4a-Cu(OTf)2 (100 mol-%) according
to the general procedure, afforded exclusively cis-14b. FC (EtOAc/
pentane, 20:80) of the crude product yielded 90 mg (50% yield) of
cis-14b as white solid. Analytical data: [α]2D0 = –36 (c = 1.0,
CH2Cl2). HPLC (Chiralcel OD-H, iPrOH/hexane, 20:80,
0.7 mL min–1, 230 nm): 71% ee, tR 11.18 min and 15.06 min
(major).
Supporting Information (see footnote on the first page of this arti-
cle): Experimental procedures for the synthesis of the chiral ligands
4b and 5a and their analytical data.
Acknowledgments
We thank the Norwegian Research Council (Grant 140593/431 to
M. M. E. and Grant 122792/432 to A. B.) for generous financial
support. We also thank Associate Professor Helge Kjøsen and Mrs.
Julie Jackson (NTNU) for performing the mass spectrometric mea-
surements.
One-Pot Preparation of Benzyl [(1R,4S)-4-(Phenylsulfinyl)cy-
clopent-2-enyl]carbamate (19b) from Cyclopentadiene (17): A solu-
tion of the N-sulfinyl carbamate 1a (2.5 mL, 0.496 mmol, 0.2 ) in
CH2Cl2 was added to the precooled solution of 4a-Cu(OTf)2
(10 mol-%) at –75 °C.
A precooled solution of 17 (100 µL,
0.992 mmol) in CH2Cl2 (100 µL) was added to the reaction mix-
ture, followed by TMSOTf (90 µL, 0.496 mmol). The reaction mix-
ture was stirred at –75 °C for 3 h and then added dry methanol
(3 mL). The resultant mixture was warmed to room temperature
and stirred at this temperature for 4 h. Water (2 mL) was poured
into the reaction mixture, and the layers separated. The aqueous
layer was extracted with CH2Cl2 (3×2 mL). The combined organic
layers were dried (MgSO4), passed through a short silica gel col-
umn, and the solvent was removed in vacuo. The residue was dis-
solved in dry THF (5 mL) and cooled to –60 °C. A solution of
PhMgBr (1.0 , 1 mL, 0.992 mmol) in THF was added by syringe,
and the resultant mixture was warmed to room temperature over-
night. The reaction mixture was hydrolyzed with saturated aqueous
NH4Cl (5 mL) and water (1 mL). The layers were separated, and
the aqueous layer was extracted with diethyl ether (3×3 mL). The
combined organic layers were dried (MgSO4), and the solvent was
removed in vacuo. FC (EtOAc/pentane, 67:33 to 75:25) of the crude
provided an inseparable mixture of sulfur epimers 19b in a ratio of
4:1 (84.6 mg, 50% total yield) as a pale yellow oil. Analytical data
of the mixture 19b: [α]2D0 = +201.3 (c = 1.0, CH2Cl2). HPLC (Chi-
ralpak AD, iPrOH/hexane, 10:90, 1 mL min–1, 230 nm) major epi-
mer: 80% ee, tR 26.4 min (major) and 44.6 min, minor epimer: 16%
[1] O. Wichterle, J. Rocek, Chem. Listy 1953, 47, 1768–1780.
[2] a) G. Kresze, A. Maschke, R. Albrecht, K. Bederke, H. P.
Patzchke, H. Smalla, A. Trede, Angew. Chem. Int. Ed. Engl.
1962, 1, 89–98; b) G. Kresze, W. Wucherpfenning, Angew.
Chem. Int. Ed. Engl. 1967, 6, 149–167; c) S. M. Weinreb, R. R.
Staib, Tetrahedron 1982, 38, 3087–3128; d) R. Bussas, G.
Kresze, H. Münsterer, A. Schwöbel, Sulfur Rep. 1983, 2, 215–
378; e) D. L. Boger, S. M. Weinreb, Hetero Diels–Alder Meth-
odology in Organic Synthesis, Academic Press, San Diego, 1987,
p. 1; f) S. M. Weinreb, Acc. Chem. Res. 1988, 21, 313–318; g)
H. Waldmann, Synthesis 1994, 535–551.
[3] a) J. K. Whitesell, D. James, J. F. Carpenter, J. Chem. Soc.,
Chem. Commun. 1985, 1449–1450; b) S. W. Remiszewski, J.
Yang, S. M. Weinreb, Tetrahedron Lett. 1986, 27, 1853–1856.
[4] T. Hamada, H. Sato, M. Hikota, O. Yonemitsu, Tetrahedron
Lett. 1989, 30, 6405–6408.
[5] a) K. Narasaka, Synthesis 1991, 1–11; b) H. B. Kagan, O. Ri-
ant, Chem. Rev. 1992, 92, 1007–1019; c) L. C. Dias, J. Braz.
Chem. Soc. 1997, 8, 289–332; d) K. A. Jørgensen, Angew.
Chem. Int. Ed. 2000, 39, 3558–3588.
[6] For reviews of C2-bisoxazoline Lewis-acid complexes as chiral
catalysts, see, for example: a) A. K. Ghosh, P. Mathivanan, J.
Cappiello, Tetrahedron: Asymmetry 1998, 9, 1–45; b) K. A.
Jørgensen, M. Johannesen, S. Yao, H. Audrain, J. Thorhauge,
Acc. Chem. Res. 1999, 32, 605–613; c) H. McManus, P. J. Guiry,
Chem. Rev. 2004, 104, 4151–4202; For a recent review of ZnII
catalysts, see; d) Y. Motoyama, H. Nishiyama, in: Lewis Acids
in Organic Synthesis, vol. 1 (Ed.: H. Yamamoto), Wiley-VCH,
Weinheim, 2000, p. 59; For recent reviews on CuII catalysts,
see; e) J. S. Johnson, D. A. Evans, Acc. Chem. Res. 2000, 33,
325–335; f) M. P. Sibi, G. R. Cook, in: Lewis Acids in Organic
Synthesis, vol. 2 (Ed.: H. Yamamoto), Wiley-VCH, Weinheim,
2000, p. 543.
ee tR 21.5 min and 23.7 min (major). IR (KBr): ν = 3317 (m, NH),
˜
3033 (w), 1716 (s, C=O), 1513 (s), 1443 (m), 1322 (s), 1234 (s), 1029
(s) cm–1. CIMS: m/z (%) = 342 (0.04) [M+1]+, 321 (0.1), 215 (16),
186 (12), 171 (15), 125 (27), 109 (14), 91 (100), 77(10). C19H19NO3S
(341.11): calcd. C 66.84, H 5.61, N 4.10; found C 66.62, H 5.70, N
4.10. Data for major sulfur epimer 19b: 1H NMR (400 MHz,
CDCl3): δ = 7.58–7.46 (m, 5 H, Ph), 7.40–7.28 (m, 5 H, Ph), 6.16
(m, 1 H, 2-H), 6.02 (d, J = 3.6 Hz, 1 H, 3-H), 5.84 (d, J = 9.6 Hz,
1 H, NH), 5.11 (AB, J = 12.8 Hz, 2 H, Bn), 4.79 (t, J = 8.8 Hz, 1
H, 1-H), 3.80 (m, 1 H, 4-H), 2.16 (dd, J = 15.2, 8.4 Hz, 1 H, 5-H),
1.75 (d, J = 15.2 Hz, 1 H, 5-H) ppm. 13C NMR (100 MHz, CDCl3):
δ = 155.9 (C=O), 142.2, 139.7 (2-C), 137.0, 131.1, 129.5, 128.6,
[7] a) A. Bayer, M. M. Endeshaw, O. R. Gautun, J. Org. Chem.
2004, 69, 7198–7205; b) A. Bayer, O. R. Gautun, Tetrahedron:
Asymmetry 2001, 12, 2937–2939; c) A. Bayer, O. R. Gautun,
Tetrahedron Lett. 2000, 41, 3743–3746; d) A. Bayer, L. K. Han-
5258
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 5249–5259

Hetero Diels–Alder Reactions of N-Sulfinyl Dienophiles
FULL PAPER
sen, O. R. Gautun, Tetrahedron: Asymmetry 2002, 13, 2407–
[25] An authentic sample of (3aR,6aS)-20 was prepared according
2415.
to: M. Lloyd, PCT Int. Appl. WO 2000050412, 2000:
.
[8]
a) D. Stien, G. T. Anderson, C. E. Chase, Y.-h. Koh, S. M.
Weinreb, J. Am. Chem. Soc. 1999, 121, 9574–9579; b) G. T. An-
derson, C. E. Chase, Y.-h. Koh, D. Stien, S. M. Weinreb, J. Org.
Chem. 1998, 63, 7594–7595.
[9]
R. S. Garigipati, A. J. Freyer, R. R. Whittle, S. M. Weinreb, J.
Am. Chem. Soc. 1984, 106, 7861–7867.
[10]
E. J. Corey, N. Imai, N. H.-Y. Zhang, J. Am. Chem. Soc. 1991,
113, 728–729.
[11]
[12]
Y. Zhang, C. J. Flann, J. Org. Chem. 1998, 63, 1372–1378.
D. A. Evans, C. S. Burgey, M. C. Kozlowski, S. W. Tregay, J.
Am. Chem. Soc. 1999, 121, 686–699.
[13]
H. Emde, D. Domsch, H. Feger, U. Frick, A. Götz, H. H. Her-
gott, K. Hofmann, W. Kober, K. Krägeloh, T. Oesterle, W.
Steppan, W. West, G. Simchen, Synthesis 1982, 1–26.
M. Oishi, in: Lewis Acids in Organic Synthesis, vol. 1 (Ed.: H.
Yamamoto), Wiley-VCH, Weinheim, 2000, pp. 355–393.
a) D. Guilaneux, S. H. Zhao, O. Samuel, D. Rainford, H. B.
Kagan, J. Am. Chem. Soc. 1994, 116, 9430–9439; b) C. Girard,
H. B. Kagan, Angew. Chem. Int. Ed. 1998, 37, 2922–2959.
[26] a) D. A. Evans, S. W. Tregay, C. S. Burgey, N. A. Paras, T.
Vojkovsky, J. Am. Chem. Soc. 2000, 122, 7936–7943 and refer-
ences cited therein; b) M. Johannesen, K. A. Jørgensen, J. Org.
Chem. 1995, 60, 5757–5762; c) M. Johannesen, S. Yao, K. A.
Jørgensen, Chem. Commun. 1997, 2169–2170; d) M. Johann-
esen, S. Yao, A. Graven, K. A. Jørgensen, Pure Appl. Chem.
1998, 70, 1117–1122.
[27] a) D. A. Evans, M. C. Kozolowski, J. S. Tedrow, Tetrahedron
Lett. 1996, 37, 7481–7484; b) S. Yao, M. Johannesen, K. A.
Jørgensen, J. Chem. Soc., Perkin Trans. 1 1997, 2345–2349; for
radical addition reactions, see: c) J. H. Wu, R. Radinov, N. A.
Porter, J. Am. Chem. Soc. 1995, 117, 11029–11030; d) M. Band-
ini, P. G. Cozzi, M. de Angelis, A. Umani-Ronchi, Tetrahedron
Lett. 2000, 41, 1601–1605.
[28] Inquiries concerning X-ray analysis should be addressed to
L. K. H. CCDC-256017 (for trans-13a), -256126 [for (1R,3R)-
13b], -256018 (for trans-13b), -256939 (for trans-14a), -256889
(for cis-14b), -256937 (for cis-15a), -256938 (for trans-15a),
and -256888 (for cis-15b) contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
[14]
[15]
[16] D. A. Evans, J. S. Johnson, C. S. Burgey, K. R. Campos, Tetra-
hedron Lett. 1999, 40, 2879–2882.
[17] T. Okamoto, K. Kobayashi, S. Oka, S. Tanimoto, J. Org. Chem.
1988, 53, 4897–4901.
[18] a) E. S. Levchenko, Y. G. BalЈon, A. A. Kisilenko, Zh. Org.
Khim. 1965, 1, 155–159; b) G. Kresze, U. Wagner, Justus Lie-
bigs Ann. Chem. 1972, 762, 93–105; c) C. Carpanelli, G. Galani,
Gazz. Chim. Ital. 1982, 112, 187–190.
[19] For 1H NMR spectroscopic data of (E,E)-10 and (E,Z)-10 see:
W. Kirmse, S. Kopannia, J. Org. Chem. 1998, 63, 1178–1184.
[20] See the supporting information for the preparation of 4b.
[21] C. Bolm, O. Simic´, J. Am. Chem. Soc. 2001, 123, 3830–3831.
[22] M. Adamczyk, R. E. Reddy, Tetrahedron: Asymmetry 1998, 9,
3919–3921.
[23] T. D. Owens, A. J. Souers, J. A. Ellman, J. Org. Chem. 2003,
[29] W. L. F. Armarego, D. D. Perrin, Purification of Laboratory
Chemicals, 4. ed., Butterworth Heinemann, Oxford, 1996.
Received: February 1, 2006
68, 3–10.
[24] E. R. Møller, K. A. Jørgensen, J. Org. Chem. 1996, 61, 5770–
5778.
Published Online: October 2, 2006
Eur. J. Org. Chem. 2006, 5249–5259
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5259