NH-Isoxazolo-bicycles; New molecular scaffolds for organocatalysis

Article abstract of DOI:10.1016/j.tet.2011.01.034

A new scaffold for organocatalysis of the Diels-Alder reaction is disclosed; an isoxazolidine ring forms the core of the catalyst and its activity is enhanced by judicious fusion of a second five-membered heterocycle to the c-edge. The catalyst performanc

Full text of DOI:10.1016/j.tet.2011.01.034

Tetrahedron 67 (2011) 2132e2138
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
NH-Isoxazolo-bicycles; new molecular scaffolds for organocatalysis
Linda Doyle, Frances Heaney *
Department of Chemistry, National University of Ireland, Maynooth, Co. Kildare, Ireland
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 27 August 2010
Received in revised form 20 December 2010
Accepted 11 January 2011
Available online 15 January 2011
A new scaffold for organocatalysis of the DielseAlder reaction is disclosed; an isoxazolidine ring forms
the core of the catalyst and its activity is enhanced by judicious fusion of a second five-membered
heterocycle to the c-edge. The catalyst performance is improved by the incorporation of an endo-cyclic
electron withdrawing group adjacent to the fusion point. The organic core is effective only in the
presence of an acid co-catalyst and whilst the two-component system shows potential as an enantio-
selective catalyst it is demonstrated that stereocontrol is a feature of the organic core and is fully in-
dependent of the choice of co-catalyst.
Keywords:
Organocatalysis
DielseAlder
Ó 2011 Elsevier Ltd. All rights reserved.
Isoxazolofused bicycle
Enantioselectivity
1. Introduction
Significantly, a recent report finds isoxazolidines and pyr-
azolidines poor catalytic scaffolds, and concludes that catalysts
Over the last decade organocatalysis has emerged as a vibrant
research area and a wide repertoire of synthetic transformations has
designed with an
ring and incorporate an exo-cyclic electron withdrawing group on
the b-position, e.g. 4. In contrast, in this paper we describe sig-
nificant organocatalytic activity of the 5,5-bicyclic isoxazolidines
5e7, Fig. 2.
a-heteroatom should be based on a six-membered
1
e3
15
been demonstrated.
ing modes of operation have been developed and whilst SOMO and
counter ion catalysis are at early stages of development many ap-
plications of enamine, iminium and hydrogen bonding catalysis
A range of catalyst architectures with vary-
4
5
6
7
8
H
3
H
H
have been reported. To-date the most successful catalysts are de-
rivatives of the five-membered, secondary amine heterocycles
O23
N
H
4
a
65O
9
10
O
NCH3
O
NR
1
6
a
proline and imidazolidinone. However, recent developments of
N
H
N
H
iminium ion catalysis indicate significant enhancement of activity
Ar
Ar
6
O
Ph
7
O
with catalysts bearing a second heteroatom
a-to the secondary
5
11,12
amine. Thus, conformationally rigid hydrazides, e.g. 1,
cyclic and
a. Ar = Ph
a Ar = Ph
a R = Me
b R = Ph
acyclic hydroxylamine or hydrazine derivatives, e.g. 2, 3,1
3,14
are
b. Ar = 2-furyl
b Ar = 2-furyl
c. Ar = 2-thienyl
c Ar = 2-thienyl
shown to be effective catalysts, Fig. 1.
Fig. 2. Isoxazolo-fused bicyclic organocatalysts.
Featuring a five-membered ring, a secondary amine and an
-heteroatom NH-isoxazolidines bear the critical architectural
elements of an organocatalyst designed to accelerate organic trans-
formations through LUMO-lowering iminium catalysis. Conforma-
tional rigidity was built into the catalyst design and the first
candidates were the bicyclic isoxazolidines 5 with an aryl substituent
a
CO2Bn
NH
H
N
N
O
CO Et
2
O
N
NH
N
H
N
H
Ph
O
2
1
3
4
Fig. 1.
a
-Heteroatom secondary amine organocatalysts.
at the ring junction and an endo-cyclic electron attracting carbonyl
0
group
b
-to the nucleophilic nitrogen atom. The organic core of the
16,17
catalyst was prepared as described elsewhere
and the Diel-
seAlder cycloaddition between cyclopentadiene and trans-cinna-
maldehyde, Scheme 1, was selected as the model reaction for testing
catalytic activity.
*
Corresponding author. Tel.: þ353 1 708 3802; fax: þ353 1 708 3815; e-mail
address: mary.f.heaney@nuim.ie (F. Heaney).
0
040-4020/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2011.01.034

L. Doyle, F. Heaney / Tetrahedron 67 (2011) 2132e2138
2133
quantitative results after 48 h (entries 9e12, Table 1). We identified
h reacting time as the most suitable for performance limitation
CHO
i
CHO
Ph
6
studies. Further, it is noted that the simple mixing of equimolar
amounts of the organic core with (þ)-10-CSA furnished the same
result as the preformed salt (entry 13, Table 1). Control runs in-
dicated neither the bicycle nor CSA alone promoted the reaction
Ph
CHO
8-endo
Scheme 1. Model reaction used in this study.
Ph
-exo
i. catalyst, rt
8
(entries 14, 15 Table 1). A comparison of reactivity with 1, 5 and
10 mol % catalyst loadings determined the limitation of catalytic
activity. With lower loadings after 6 h reaction time the extent of
conversion to the products was poor to moderate. After 24 h a sig-
nificant improvement in product yield was noted (entries 16e18,
Table 1), however, the performance of the catalyst was deemed
optimal at 10 mol % loading. The activity of 5a$(þ)-10-CSA was
subsequently studied in a variety of solvents (6 h, rt). Considering
2
. Results and discussions
Initial catalytic runs, analysed by 1H NMR spectroscopy,
employed 5a together with a range of acid co-catalysts including
0 0
2 4 4
SO , HClO , tartaric acid and 1,1 -binaphthyl-2,2 -diyl
HCl, TFA, H
hydrogen phosphate 9. No relationship was observed between
catalytic activity and acid pK and whilst perchloric acid was almost
as effective as (þ)-10-camphor sulfonic acid (CSA), with a future
view to counter ion and enantioselective organocatalysis, the latter
was selected as co-catalyst of choice for further study (entries 1e8,
Table 1). A time study revealed 71% conversion at 2 h rising to near
2
both yield and diastereoselectivity MeOH/H O (86%, 38:63) and
a
MeOH (88%, 39:61) were found to be superior to more bulky alco-
i
t
hols ( PrOH, BuOH), chlorinated solvents (CH
non-protic solvents (DMSO, dioxane) or hydrocarbon solvents
toluene). No cycloaddition products formed in pure H O [see
2 2 3
Cl , CHCl ), polar
(
2
Supplementary data]. In all cases diastereoselectivity was in favour
of the exo-adduct (w2:1) as expected for catalysis by the iminium
Table 1
14
route. Whilst no effort was made to re-isolate the catalyst at the
DielseAlder reaction between cyclopentadiene and trans-cinnamaldehyde catalysed
1
by 5a$HXa
end of the reaction H NMR spectral analysis of the crude reaction
products showed resonance signals attributable to the parent bi-
cycle 5a so confirming the stability of the catalyst to the reaction
conditions.
18
Entry Co-catalyst HX
Time Loading Conversion (%)b endo/exob HX pK
a
(h)
Yield (%)c
There is little precedent for the existence of iminium ions of
isoxazolidine frameworks; aside from sporadic mention as tran-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
TFA
SO
HCl
6
6
6
6
6
6
6
6
48
24
4
2
6
10 mol % 32
25 38:62
60 39:61
52 38:62
53 38:62
0.5
ꢀ3
ꢀ8
H
2
4
d
10 mol % 78
10 mol % 66
10 mol % 65
19,20
sient intermediates
to the best of our knowledge the benzi-
(ꢁ)-9
1.14
2.98
2.98
soxazolium trifluoroacetate 10, Fig. 3, is the only claimed stable
L
-(þ)-Tartaric acid
-(ꢀ)-Tartaric acid
HClO
10 mol %
10 mol %
0
0
0
0
d
21
cation. Thus, in support of an intermediary iminium species in the
D
d
catalytic cycle, interaction between equimolar amounts of the
4
10 mol % 88
10 mol % 93
10 mol % 97
10 mol % 94
10 mol % 84
10 mol % 71
10 mol % 93
82 40:60
86 40:60
96 39:61
85 40:60
75 36:64
61 36:64
86 38:62
ꢀ10
(þ)-10-CSA
(þ)-10-CSA
(þ)-10-CSA
(þ)-10-CSA
(þ)-10-CSA
(þ)-10-CSA
1.17
isoxazolidinium salt, 5a$(þ)-10-CSA, and trans-cinnamaldehyde
1
was followed by H NMR spectroscopy. In CD
3
OD/D
2
O (19:1) for-
0
1
2
3
4
5
6
7
8
mation of a single geometrical isomer of the iminium ion was vis-
ible within minutes. However, formation of the dimethyl acetal of
cinnamaldehyde, 11, indicated by minor signals at 6.65, 6.20 and
5.00 ppm, was competing with iminium formation. Major signals at
d
e
e
None
48
48
6
6
24
10 mol %
10 mol %
1 mol %
1
7
9
d
d
d
d
(þ)-10-CSA
(þ)-10-CSA
(þ)-10-CSA
(þ)-10-CSA
d
8
.62 (d, 10.0 Hz) and 8.15 ppm (d, 15.5 Hz) are, with reference to
43:57
22
1
3
reported structures, deemed characteristic of H and H of the
5 mol % 65
5 mol % 89
51 39:61
73 41:59
conjugated iminium ion 12, Fig. 4. 2D-COSY experiments indicate
2
the resonance of H lies within a multiplet signal at w7.6 ppm. To
a
Conditions: 3 equiv of diene, 1 equiv of dienophile, 5a (10 mol %), HX (10 mol %);
ꢂ
3
eliminate acetal formation, the experiment was repeated in CD CN,
MeOH/H
2
O (19:1), 25 C.
b
Estimated from the 1H NMR spectrum by relating the integrals of the aldehyde
iminium ion formation, characterized by the appearance of doublet
resonances at 8.26 and 8.01 ppm, was much slower and w55% of
the starting aldehyde remained after w2 h. Iminium ion geometry
is recognized as central to stereocontrol of catalysed reactions and
thus it was disappointing that nuclear Overhauser difference
spectroscopy experiments were inconclusive. Thus, Dreiding scale
protons of the starting material and the DielseAlder adducts.
c
Combined isolated yield.
Identical results were obtained in experiments conducted with pre-formation of
d
the salt rather than the addition of the two components of the catalytic system in
a 1:1 ratio.
e
Control runs employing either the organic core or the co-catalyst alone.
iPr
iPr
iPr
Et
O
O
O
OCD3
OCD3
O
O
O
P
P
N
OH
O
OH
H
NO2
Ph
iPr
CF3CO2
9
10
11
iPr
iPr
13
Fig. 3. Structures of phosphoric acids 9 and 13, iminium ion 10, and acetal 11.

2134
L. Doyle, F. Heaney / Tetrahedron 67 (2011) 2132e2138
CHO
H
H
Ph
O
O
O
O
O
N
N
H
Ph
H
Ph
H2
O
2 O
1
SO3
O
H
H1
O
O
Z-12
E-12
Ph
N
H3
H
Ph
H3
Ph
5a
.(+)-10-CSA
Fig. 4. Proposed iminium species from 5a and (þ)-10-CSA with trans-cinnamaldehyde.
models of Z- and E-12 were examined and these suggest E-12
suffers significant repulsive non-bonding interactions between the
substrate olefin and the oxo-group; Z-12 is thus most likely to
represent the adopted geometry, Fig. 4.
O
H
O
To explore the scope of the reaction electron rich and electron
poor dienophiles o-, and p-nitrocinnamaldehyde and o- and
p-methoxycinnamaldehyde were selected as reaction partners with
cyclopentadiene. As shown in entries 1e4, Table 2, product yields
and selectivities varied significantly. It is known that the kinetics of
iminium promoted reactions may depend both on iminium forma-
N
H
.
CSA
Ph
O
O
CH3
O
Fig. 5. Proposed interaction between 5a$CSA and trans-o-methoxycinnamaldehyde.
10b
tion and the carbonecarbon bond forming event or simply on the
2
3
latter. Thus, steric effects are believed to account for the poor yield
In order to test the tolerance of the catalytic core to structural
variation the bridgehead phenyl was replaced by furyl and thienyl
rings. However, neither 5b$CSA nor 5c$CSA were as effective as
of DielseAlder products from cycloaddition of cyclopentadiene to
Table 2
5a$CSA in promoting the DielseAlder reaction with cycloadducts
Effect of variation in substrate and catalyst structure on the DielseAlder reaction
a
between cyclopentadiene and
a
,b
-unsaturated aldehydes
furnished in just 35 and 60% yield, respectively. In each case, in-
volvement of an intermediate iminium species is supported by
a preference for the exo-adduct (w57:43), Table 2, entries 5e7. In
rationalising the reduced reactivity of the heteroaryl analogues of
Conversion (%)b endo/exob
Entry Catalyst Dienophile
Yield (%)c
1
2
3
4
5
6
7
8
9
1
1
1
1
1
5a$(þ)-10-CSA o-nitrocinnamaldehdye
5a$(þ)-10-CSA p-nitrocinnamaldehdye
61
88
52 55:45
86 40:60
83 40:60
16 38:62
86 38:62
35 43:57
60 43:57
86 35:65
48 39:61
46 41:59
5a we propose an intramolecular H-bond between the iso-
xazolidine NH and the pendant heterocycle may increase the acti-
vation energy of the iminium forming step. The particularly
sluggish reactivity of 5b may result from a furan-HX interaction
5a$(þ)-10-CSA o-methoxycinnamaldehyde 96
5a$(þ)-10-CSA p-methoxycinnamaldehyde 26
5a$(þ)-10-CSA trans-cinnamaldehyde
5b$(þ)-10-CSA trans-cinnamaldehyde
93
40
62
95
53
49
1
74
70
65
involving both the
p-electron system and the lone pair. The thio-
5c$(þ)-10-CSA
trans-cinnamaldehyde
6a$(þ)-10-CSA trans-cinnamaldehyde
phene-HX interaction is expected to involve only the
p
-electron
e
2
5
system and this may account for its more minimal impact on
catalytic activity. In a further probe of structure/activity relation-
ships catalytic systems based on the lactam fused isoxazolidines
6b$(þ)-10-CSA trans-cinnamaldehyde
0
1
2
3
4
6c$(þ)-10-CSA
trans-cinnamaldehyde
e
7a$(þ)-10-CSA trans-cinnamaldehyde
d
d
d,e
7a$(þ)-10-CSA trans-cinnamaldehyde
65 38:62
64 38:62
62 39:61
6aec were examined. Results for the DielseAlder reaction in the
e
7b$(þ)-10-CSA trans-cinnamaldehyde
d,e
presence of 6aec$(þ)-10-CSA, summarised in Table 2, entries 8e10
suggest no significant change in reactivity followed from re-
placement of the electron withdrawing endo-cyclic ester of 5 with
7b$(þ)-10-CSA trans-cinnamaldehyde
a
Conditions: 3 equiv of diene, 1 equiv of dienophile, Organic core 5e7 (10 mol %),
ꢂ
(
þ)-10-CSA (10 mol %), MeOH/H
2
O (19:1), 25 C, 6 h.
b
Estimated from the 1H NMR spectrum by relating the integrals of the aldehyde
the amide in 6.
0
protons of the starting material and the DielseAlder adducts.
To determine if the
b
-electron withdrawing lactone and lactam
c
Combined isolated yield.
groups of 5 and 6 were important for catalytic activity the pyrro-
lidine fused isoxazolidines 7 were designed and their CSA salts
prepared. The catalytic performance of diamine-protic acid orga-
d
7a/b (10 mol %), (þ)-10-CSA (20 mol %).
e
Identical results were obtained in experiments conducted with pre-formation of
the salt rather than the addition of the two components of the catalytic system in
a 1:1 ratio.
2
6e28
nocatalysts is known to be sensitive to the acid/amine ratio.
A
þ
18
comparison of the pKBH values, obtained from ACD labs, of the
parent heterocycles suggest that the N-methyl pyrroloisoxazole 7a
should preferentially protonate at the tertiary amine and will re-
quire a second equivalent of CSA to form the isoxazolidinium salt.
Indeed, under standard reaction conditions, 7a$CSA was an in-
effective catalyst, however, with 20 mol % (þ)-10-CSA catalytic
activity was observed and cycloadducts, obtained with exo-selec-
tivity, were isolated in 65% yield (Table 2, entries 11, 12). The
N-phenyl analogue, 7b proved equally active with the employment
of either 10 or 20 mol % (þ)-10-CSA and cycloadducts were fur-
nished in yields of w63% (Table 2, entries 13, 14). The significant
o- (52%) with respect to p-nitrocinnamaldehyde (86%). The reduced
electrophilicity of the aldehyde and the H-bonding potential are
likely responsible for the wide span in product yield from reaction of
p-(16%) and o-methoxycinnamaldehyde (83%) with the diene. In the
former case mesomeric donation may dominate causing the low
yield. It is acknowledged that subtle interplays of H-bonding net-
works can determine the overall reaction profile of iminium ion
formation24 and we postulate, as shown in Fig. 5, that a H-bonding
interaction involving the o-OMe substituent and the NH- of the
catalytic systemmayassistiminiumformationthrough anchorageof
the relevant functional groups as well as reducing mesomeric do-
nation to the aldehyde culminating in the high yield of DielseAlder
products from reaction of o-methoxycinnamaldehyde.
reduction in catalytic activity of the pyrrolidine fused bicycles,
0
7$CSA, is suggestive of a role for the
b
-endo-cyclic electron with-
drawing groups of 5 and 6 in the catalytic cycle.

L. Doyle, F. Heaney / Tetrahedron 67 (2011) 2132e2138
2135
The potential of a
performance of -heteroatom organocatalysts has previously been
noted. Ogilvie believes this feature accelerates iminium hydrolysis
whilst Tomkinson, following from an observation that the syner-
gistic effect of an -heteroatom and a -carbonyl group is more
prominent in anhydrous than in aq MeOH, believes the carbonyl
moiety functions as a proton shuttle. To understand better the
structure activity of relationships of the bicyclic isoxazolidine cat-
alysts 5e7 the DielseAlder reactions were repeated in anhydrous
MeOH. The pyrrolidine fused catalysts 7a,b$CSA performed more
poorly in dry solvents with yields dropping by 13e30% (Table 3,
entries 6e8) whilst increases in yields of similar magnitudes were
observed for the lactone/lactam fused isoxazolidines 5/6aec$CSA
b
-electron withdrawing group to enhance the
Tricyclic benzopyranoisoxazolidine derivatives have found suc-
2
9
a
cess as chiral auxiliaries for asymmetric alkylation and with a fu-
ture goal to develop an enantioselective organocatalyst 5a was
resolved employing either (þ)- or (ꢀ)-CSA as resolving agent. HPLC
analysis confirmed (ꢀ)-5a was obtained optically pure whilst (þ)-5a
was furnished with w90% ee. Prior to testing the enantiocontrol with
2
2
a
b
14
catalytic systems based on the resolved bicycles we explored the
possibility for counter ion directed organocatalysis.5
,30,31
Control
experiments confirmed that rac-5a, in combination with either (þ)-
or (ꢀ)-CSA, furnished products with essentially with no enantioen-
richment. Similarly, no enhancement was observed with (þ)- or
0
0
(ꢀ)-1,1 -binaphthyl-2,2 -diyl hydrogen phosphate 9 nor with the
more sterically demanding (R)-TRIP 13 as co-catalyst with rac-5a; in
the case of (R)-TRIP, limited solubility of the co-catalyst is reflected in
the poor chemical yield, results are summarised in Table 4, entries
(
Table 3, entries 1e5). These results clearly highlight the in-
0
volvement of the
b
-carbonyl in the catalytic cycle, and we conclude
Table 3
a
Effect of water in the reaction solvent on the model DielseAlder reaction
Entry
Catalystb
Solventc
Conversion (%)d
Yield (%)e
endo/exod
Solvent
Conversion (%)d
Yield (%)e
endo/exod
1
2
3
4
5
6
7
8
5a
5a
aq MeOH
aq MeOH
aq MeOH
aq MeOH
aq MeOH
aq MeOH
aq MeOH
aq MeOH
93
84
40
62
95
74
70
65
86
75
35
60
86
65
64
62
38:62
36:64
43:57
43:57
35:63
38:62
38:62
39:61
anh. MeOH
anh. MeOH
anh. MeOH
anh. MeOH
anh. MeOH
anh. MeOH
anh. MeOH
anh. MeOH
92
91
52
91
96
58
35
46
88
82
47
86
85
52
31
43
39:61
36:64
45:55
41:59
34:66
42:58
30:70
44:56
f
5b
5c
6a
g
7a
7b
7b
g
a
ꢂ
Conditions: 3 equiv of diene, 1 equiv of dienophile, 25 C, 6 h.
b
c
d
e
f
5
e7 (10 mol %), (þ)-10-CSA (10 mol %).
MeOH/H O (19:1).
2
Estimated from the 1H NMR spectrum by relating the integrals of the aldehyde protons of the starting material and the DielseAlder adducts.
Combined isolated yield.
Reaction repeated over 4 h as after 6 h (entry 1) the extent of conversion was deemed too great to allow influence of solvent to be judged.
g
7a/b (10 mol %), (þ)-10-CSA (20 mol %).
Table 4
Investigation of the potential of the salts of (þ)- and (ꢀ)-5a to effect stereoselective catalysis of the model DielseAlder reactionb
a
ꢂ
Conversion (%)c
endo/exod
endo eee,f
exo eee,f
Entry
Catalyst
Temp ( C)
Time (h)
Yield (%)d
1
2
3
4
5
6
7
8
9
(ꢁ)-5a$(þ)-10-CSA
(ꢁ)-5a$(ꢀ)-10-CSA
(ꢁ)-5a$(þ)-9
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
93
84
62
50
11
92
90
82
83
66
56
31
38
5
86
82
57
42
d
38:62
38:62
35:65
39:61
45:55
37:63
37:63
38:62
39:61
38:62
39:61
39:61
39:61
32:68
45:55
38:62
37:63
39:61
0.3
5
0.4
0.5
9
2
5
1.3
1.9
9
(ꢁ)-5a$(ꢀ)-9
(ꢁ)-5a$(R)-TRIP
(þ)-5a$(þ)-10-CSA
(þ)-5a$(ꢀ)-10-CSA
(ꢀ)-5a$(þ)-10-CSA
(ꢀ)-5a$(ꢀ)-10-CSA
(þ)-5a$(þ)-9
84
83
74
69
52
47
27
32
d
41
40
37
37
45
39
37
26
25
34
32
38
35
58
58
42
54
59
58
54
48
51
22
51
56
56
10
11
12
13
14
15
16
17
18
(þ)-5a$(ꢀ)-9
(ꢀ)-5a$(þ)-9
(ꢀ)-5a$(ꢀ)-9
(þ)-5a$(R)-TRIP
(ꢀ)-5a$(R)-TRIP
6
d
(þ)-5a$HClO
(þ)-5a$HCl
4
93
87
65
91
78
42
(þ)-5a$(ꢁ)-9
a
b
c
d
e
f
(
ꢀ)-5a Homochiral; (þ)-5a enantioenriched, w90% ee.
2
Conditions: 3 equiv of diene, 1 equiv of dienophile, MeOH/H O (19:1), 5a (10 mol %) co-catalyst (10 mol %).
Estimated from the 1H NMR spectrum by relating the integrals of the aldehyde protons of the starting material and the DielseAlder adducts.
Combined isolated yield.
Estimated by GLC.
Those reactions catalysed by salts of (þ)-5a provide products with the opposite sense of enrichment compared to those promoted by salts of (ꢀ)-5a.
that a carbonyl group located sufficiently intimately to the sec-
ondary amine can enhance catalytic performance in anhydrous
1e5. The failure to detect any enantioselection with the racemic bi-
cycle rules out the possibility for counter ion directed catalysis with
this system.
solvents. The apparent acceptance of the location of this group ei-
0
ther endo- or exo-cyclic, and either
b
- or
b
-to the secondary amine
Gratifingly, catalytic systems based on (þ)-5a furnished Dielse
broadens the possibilities for future catalyst design.
Alder products with a modest degree of enantioselectivity. For

2136
L. Doyle, F. Heaney / Tetrahedron 67 (2011) 2132e2138
ꢂ
example, the reaction promoted by (þ)-5a(þ)-CSA at 25 C furnished
the exo-adduct with 58% ee and the endo-adduct with 41% ee, similar
diastereomeric and enantiomeric selectivity was observed with
diethyl ether provided the exo-adducts in highest enantioselectivity
whilst in dioxane the endo-adducts were formed more selectively.
(
ꢀ)-CSA as co-catalyst (Table 4, entries 6,7). In reactions catalysed by
3
. Conclusion
(
ꢀ)-5a in combination with either (þ)- or (ꢀ)-CSA yields and selec-
tivities (opposite sense) were slightly lower (Table 4, entries 8,9).
Whilst the rate of the DielseAlder reactions were significantly com-
promised at reduced temperature there was only a marginal im-
We have discovered a new scaffold for organocatalysis of the
DielseAlder reaction. The active catalyst comprising an iso-
xazolidine ring with a second five-membered ring fused to the c-
edge, is successfully used with camphor sulfonic acid as co-catalyst.
The catalytic activity is influenced by the nature of the aryl group at
the fusion point and is sensitive to reaction solvent. Those catalysts
featuring a
fective in anhydrous MeOH whilst those lacking this functionality
perform better in wet MeOH. The conformational rigidity of the 5,5-
bicycle together with the judiciously positioned endo-cyclic car-
bonyl group combine to provide a valuable catalytic framework in
contrast to monocyclic five-membered
which are ineffective promotersof iminiumion mediated DielseAlder
reactions. H NMR studies confirm an intermediary iminium ion in
the catalytic cycle and although initial results are modest, the
potential for asymmetric induction, arising from the organic core of
the catalytic system, and independent of the counter ion, has been
demonstrated.
ꢂ
provement in enantioselectivity at ꢀ20 C [see Supplementary data].
The sense of asymmetric induction is consistent with selective for-
mation of the Z-iminium ion and the shielding of one face of the
catalyst framework by the bridgehead phenyl substituent as shown in
Fig. 6.
0
b
-endo-cyclic electron withdrawing group are most ef-
O
O
H2
N
a-heteroatom heterocycles,
H
Ph
2-Re,3-Si face
H1
O
Ph
15 1
H3
Fig. 6. Proposed alignment for the stereocontrolled DielseAlder reaction.
Catalytic activity was lower with the phosphinic acid co-catalyst
(
þ)- and (ꢀ)-9, however, in all cases the enantiocontrol paralleled
that seen with corresponding CSA salts, chemical yields were too
low with (R)-TRIP 13 as co-catalyst to permit comment on enan-
tiocontrol (Table 4, entries 10e15). The results in Table 4 do not
provide evidence for the operation of matched or mis-matched
pairs with the two component catalytic systems based on (þ)- or
4
4
. Experimental
.1. General
Solvents were dried and purified in accordance with established
(
ꢀ)-5a with either enantiomer of CSA or the phosphinic acid 9.
33
1
13
procedures. All H and C NMR spectra were recorded on
Thus, we hypothesised that enantiocontrol may rest entirely with
the bicyclic nucleus. To demonstrate independence from the
ꢂ
a Bruker Avance spectrometer at a probe temperature of 25 C,
1
13
operating at 300 MHz for the H nucleus and 75.5 MHz for the
nucleus. Infrared spectra were recorded on a PerkineElmer System
000 FT spectrometer as thin films or as Nujol mulls. Melting points
C
counter ion, catalytic systems based on (þ)-5a with HClO
4
, HCl and
rac-9 were examined. In all three cases, (Table 4, entries 16e18) the
reaction products were formed with almost the same selectivity as
observed with (þ)-5a$(þ)-CSA as catalyst.
2
were measured on a Stuart Scientific (Bibby) Melting Point SMPI
apparatus and are uncorrected. Microanalytical data were recorded
on an Exeter Analytical CE-440 elemental analyser in Cork and
Dublin. Flash column chromatography was performed using silica
gel 60 (Merck, 0.040e0.063 mm) on a Buchi Automated Flash
system. Analytical thin layer chromatography was carried out on
In an effort to optimise enantiocontrol the model reaction was
screened in a range of solvents. The results, summarised in Table 5
indicate, as previously observed for bipyrrolidine organocatalysts,
3
2
lower chemical yields and enantioselectivities in organic solvents
in comparison to aq methanol. In particular, reactions in toluene and
aluminium sheets pre-coated with Merck TLC Silica gel 60 F254
,
Table 5
a
Influence of solvent on enantioselectivity of the model DielseAlder reaction
Entry
1
Solvent
Catalystb
Conversion (%)c
Yield (%)d
endo/exoc
endo eee
exo eee
f
f
f
f
MeOH/H
Toluene
Dioxane
2
O
2
O
2
O
2
O
(þ)-5a$(þ)-10-CSA
92
16
46
28
90
28
48
41
82
29
35
36
83
18
33
34
84
14
45
25
83
19
30
36
74
25
28
32
69
12
27
29
37:63
45:55
47:53
46:54
37:63
46:54
49:51
47:53
38:62
48:52
52:48
43:57
39:61
43:57
51:49
41:59
41
11
32
8
40
3
30
4
37
0.8
13
9
58
21
4
2
Et O
30
58
31
0.4
35
42
19
4
21
54
16
5
2
3
4
MeOH/H
Toluene
Dioxane
(þ)-5a$(ꢀ)-10-CSA
(ꢀ)-5a$(þ)-10-CSA
(ꢀ)-5a$(ꢀ)-10-CSA
2
Et O
MeOH/H
Toluene
Dioxane
2
Et O
MeOH/H
Toluene
Dioxane
37
8
13
12
2
Et O
18
a
Conditions: 3 equiv of diene, 1 equiv of dienophile, MeOH/H
2
O (19:1).
b
c
d
e
f
5a (10 mol %) CSA (10 mol %).
Estimated from the 1H NMR spectrum by relating the integrals of the aldehyde protons of the starting material and the DielseAlder adducts.
Combined isolated yield.
Estimated on the product aldehydes by chiral GLC.
(
ꢀ)-5a Homochiral; (þ)-5a enantioenriched, w90% ee.

L. Doyle, F. Heaney / Tetrahedron 67 (2011) 2132e2138
2137
ꢂ
ꢀ1
developed sheets were visualised using a portable UVItec CV-006
lamp (
¼254). Analytical gas-phase chromatography (GC) was
performed on a PerkineElmer Clarus 500 using a SUPLECO Beta
Dex 110 fused silica capillary column (30 mꢃ0.25 mmꢃ0.25 m;
60 C then 2 C/min to 200 C). High Performance Liquid Chro-
4.48, N, 2.49%]; mp 110e119 C; vmax (KBr)/cm 3410, 1777, 1590,
1507, 1326, 1231; (300 MHz, CD
l
d
d
3
OD) 8.09 (2H, d J¼8.9) 7.99 (2H,
d J¼8.2) 7.60e7.55 (4H, m) 7.49e7.29 (5H, m) 7.27e7.22 (4H, m)
m
4.68e4.61 (1H, m) 4.47e4.44 (1H, m) 4.43e4.19 (2H, m) 3.75e3.68
ꢂ
ꢂ
ꢂ
1
C 3
(1H, m); d (75.5, CD OD) 176.2,147.6, 147.5, 132.3, 132.2, 142.1, 131.6,
matography (HPLC) was performed on a PerkineElmer Totalchrom
v.6.2.0.0.1 or Gilson analytical HPLC using a CHIRALCEL OD ana-
lytical column in both cases (250ꢃ4.6 mm) [flow rate 1 mL/min,
iso-propyl alcohol/hexane 1:1]. Optical rotations were measured on
an Optical Activity AA 100 polarimeter in a 2 dm polarimeter tube.
130.8, 129.3, 129.1, 128.9, 128.3, 126.5, 126.4, 125.3, 121.4, 120.5, 79.0,
75.6, 71.2, 48.0.
4.2.3. Preparation of the camphor sulfonic acid salt of rac-6a, 6a.(þ)-
10-CSA 5-methyl-6a-phenylhexahydro-1H-pyrrolo[3,4-c][1,2]oxazol-
1-ium (7,7-dimethyl-2-oxobicyclo[2.2.1]hept-1-yl)methane sulfonate.
rac-6a (200 mg, 9.169 mmol) and (þ)-10-CSA (213 mg, 9.169 mmol)
were dissolved in hot acetone (1.6 mL) and the solution was allowed
to cool to rt until precipitation of white solid had completed
ꢂ
Reactions at 25 C were carried out in a Grant W14 water bath
without stirring. Low-temperature reactions were carried out in an
ethanol bath and the temperature was controlled by a Julabo FT920
cooler.
(
384 mg, 93%); [found: C, 58.70; H, 6.71; N 6.11. C22
H
30
N
2
SO
6
re-
ꢂ
ꢀ1
4
.2. Typical experimental procedure for catalytic runs
quires C, 58.66; H, 6.72, N, 6.22%]; mp 180e186 C; vmax (KBr)/cm
d
3455, 2929, 2645, 1741, 1711, 1590, 1504, 1451, 1267; d (300 MHz,
To a solution of the organic catalyst (0.076 mmol, 10 mol %) and
DMSO) 7.48e7.39 (5H, m) 6.48 (2H, br s) 4.16e4.06 (2H, m) 3.80 (1H,
dd J¼10.5 and 7.5) 3.44e3.38 (2H, br m) 2.91 (1H, d J¼9.8) 2.78 (3H,
s) 2.67e2.43 (2H, m) 2.30e2.21 (1H, m) 1.97e1.78 (3H, m) 1.36e1.24
acid co-catalyst (0.076 mmol, 10 mol %) in MeOH/H
2
O 19:1
(
0.78 mL) was added trans-cinnamaldehyde (100 mg, 0.76 mmol)
and the resulting solution was stirred at rt for 5 min. To this solution
C
(2H, m) 1.04 (3H, s) 0.75 (3H, s); d (75.5, DMSO) 216.0, 170.3, 136.0,
was added freshly cracked cyclopentadiene (150 mg, 2.273 mmol,
128.4, 128.2, 126.6, 78.7, 76.6, 58.1, 51.5, 47.1, 46.8, 42.2, 40.3, 39.8,
29.8, 26.3, 24.1, 20.0, 19.5.
ꢂ
3
equiv) in a single aliquot and the solution was maintained at 25 C
for a further 6 h. The reaction solvent was evaporated and hydro-
lysis followed by stirring a CHCl (2 mL), TFA (1 mL) and H O (1 mL)
solution of the residue for 3 h at rt. The solution was neutralised
with satd aq NaHCO , the organic layer was extracted with DCM
2ꢃ10 mL), washed with H O (5 mL), dried (MgSO ), filtered and
evaporated to yield the crude product as a brown oil. Purification by
3
2
4.2.4. Preparation of the camphor sulfonic acid salt of rac-7a, 7a.(þ)-
10-CSA 5-methyl-6a-phenylhexahydro-1H-pyrrolo[3,4-c][1,2]oxazol-
5-ium (7,7-dimethyl-2-oxobicyclo[2.2.1]hept-1-yl)methane sulfonate.
rac-7a (250 mg, 1.230 mmol) and (þ)-10-CSA (290 mg, 1.230 mmol)
were dissolved in warm acetone (2 mL). The resulting solution was
allowed to cool to rt and stirring continued at rt until precipitation of
the title product as a white solid was complete (540 mg, 100%).
3
(
2
4
1
2
flash column chromatography followed (SiO , Hex/EtOAc 98:2). H
NMR spectral analysis of the crude reaction mixtures was used to
establish the extent of conversion of the starting material as well as
the product exo/endo ratio through the integration of the signals
[found: C, 60.55; H, 7.41; N 6.27. C22
7.39, N, 6.42%]; mp 170e176 C; vmax (KBr)/cm 3447, 3224, 2975,
H
32
N
2
SO
5
requires C, 60.49; H,
ꢂ
ꢀ1
1
representing the respective aldehyde protons. H NMR data for the
d 3
2497, 1735, 1457, 1240; d (300 MHz, CDCl ) 11.01 (1H, br m)
cycloaddition products agreed with previously reported literature
7.52e7.28 (5H, m) 4.44e3.32 (6H, br m) 3.14e2.92 (4H, d, br m) 2.75
(1H, d J¼14.5) 2.57e2.47 (1H, br m) 2.29e2.17 (1H, m) 2.04e1.93
(2H, m) 1.80e1.66 (2H, m) 1.37e1.29 (1H, m) 1.02 (3H, s) 0.78 (3H, s);
values.14
The enantiomeric ratios were determined and diastereomeric
ratios confirmed by GC analysis retention time; exo isomers: 10.31
and 10.52 min, endo isomers: 10.79 and 10.96 min.
d
C
(75.5, CDCl
64.4, 63.7, 59.4, 57.4, 53.0, 50.8, 50.3, 46.9, 46.4, 41.9, 41.6, 40.2, 39.8,
8.7, 26.0, 23.5, 18.8, 18.8.
3
) 216.1, 138.5,136.7, 128.1,127.7, 127.4, 125.2, 77.8, 77.2,
2
4.2.1. Preparation of the camphor sulfonic acid salt of racemic 5a, 5a$
(
þ)-10-CSA 6-oxo-6a-phenyltetrahydro-1H,3H-furo[3,4-c][1,2]oxaz
4.2.5. Preparation of the bis-salt from camphor sulfonic acid and rac-
7a, 7a.2(þ)-10-CSA 5-methyl-6a-phenylhexahydro-1H-pyrrolo[3,4-
c][1,2]oxazol-1,5-diium (7,7-dimethyl-2-oxobicyclo[2.2.1]hept-1-yl)
methane sulfonate. rac-7a (201 mg, 0.985 mmol) and (þ)-10-CSA
(460 mg, 1.980 mmol) were added to warm acetone (1.6 mL). Upon
cooling to rt the bis-salt precipitated as a yellow solid (660 mg,
ol-1-ium (7,7-dimethyl-2-oxobicyclo[2.2.1]hept-1-yl)methane sulfo-
nate. Hot acetone (1 mL) was added to a flask containing rac-5a
(
100 mg, 0.49 mmol) and (þ)-10-CSA (133 mg, 0.488 mmol). The
resulting solution was allowed to stir at rt until the title salt pre-
cipitated as a white solid (228 mg, 98%); [found: C, 57.58; H, 6.22; N
þ
þ
3
1
1
.23. C21
H
27NSO
7
requires C, 57.62; H, 6.21, N, 3.20%]; mp
100%); HRMS: MþNa found 691.2689. C32
H
48
N
2
S
2
O
9
þNa re-
ꢂ
ꢀ1
ꢂ
ꢀ1
53e158 C; vmax (KBr)/cm 3451, 2952, 2530, 1781, 1734, 1451,
quires 691.2693 (diff ꢀ0.65 ppm); mp 97e106 C; vmax (KBr)/cm
386, 1231;
d
d
(300 MHz, DMSO) 7.56e7.52 (2H, m), 7.50e7.37 (3H,
d 3
3451, 2960, 2571, 1741, 1646, 1455, 1168, 1039; d (300 MHz, CDCl )
m) 4.66 (1H, dd J¼9.7 and 7.2) 4.44 (1H, dd J¼9.7 and 1.8) 4.21 (1H,
10.93e10.51 (3H, br) 7.61e7.41 (5H, m) 4.89e3.70 (7H, br m)
0
dd J¼8.7 and 2.2) 4.04 (1H, dd J¼8.7 and 8.7) 3.66e3.59 (1H, m)
3.12e3.07 (5H, br m) 2.65 (2H, d J¼14.7) 2.24e2.22 (4H, m )
3
.01 (1H, d J¼14.8) 2.63e2.50 (2H, m) 2.31e2.23 (1H, m) 1.99e1.80
2.03e1.89 (4H, m) 1.77 (2H, m) 1.62e1.53 (2H, m) 1.36e1.26 (2H, m)
(
3H, m) 1.40e1.25 (2H, m) 1.04 (3H, s) 0.76 (3H, s);
d
C
(75.5, DMSO)
C 3
0.98 (6H, s) 0.76 (6H, s); d (75.5, CDCl ) 216.7, 207.0, 133.8, 133.5,
2
4
15.7, 176.1, 134.2, 128.7, 128.6, 126.7, 78.6, 75.3, 70.1, 58.0, 47.9, 47.0,
2.2, 42.1, 38.6, 26.3, 24.2, 19.9, 19.4.
133.3, 129.7, 129.6, 129.5, 129.3, 127.3, 127.2, 79.5, 79.4, 78.8, 76.7,
75.9, 63.7, 63.3, 60.9, 59.7, 58.3, 50.7, 50.2, 48.0, 47.5, 42.7, 42.6, 41.7,
40.9, 40.6, 30.9, 26.9, 24.5, 19.8, 19.7.
0
4
.2.2. Preparation of the isoxazolidinium salt from (S)-(þ)-1,1 -bina
0
phthyl-2,2 -diylhydrogen phosphate with rac-5a, 5a$(þ)-9 6-oxo-6a-
4.2.6. Preparation of the camphor sulfonic acid salt of rac-7b, 7b.(þ)-
10-CSA 5,6a-diphenylhexahydro-1H-pyrrolo[3,4-c][1,2]oxazol-1-ium
(7,7-dimethyl-2-oxobicyclo[2.2.1]hept-1-yl)methanesulfonate. rac-7b
(202 mg, 0.759 m mol) and (þ)-10-CSA (176 mg, 0.759 mmol) were
dissolved in hot acetone (1.2 mL) following cooling to rt stirring was
continued until precipitation was complete. The salt was filtered
and dried to yield the title product as a brown solid (365 mg, 97%);
0
phenyltetrahydro-1H,3H-furo[3,4-c][1,2]oxazol-1-ium (þ)-1,1 -binap
0
hthyl-2,2 -diylhydrogen phosphate. To a flask containing rac-5a
0
0
(
50 mg, 0.244 mmol) and (S)-(þ)-1,1 -binaphthyl-2,2 -diylhydrogen
phosphate (85 mg, 0.244 mmol) was added hot MeOH (8 mL). The
resulting solution was stirred at rt for 15 h prior to evaporation to
dryness to yield the product as a white solid (135 mg, 100%); [found:
7 2
C, 66.45; H, 4.61; N 2.23. C31H24NPO $1/2H O requires C, 66.19; H,
34 2 5
[found: C, 64.69; H, 6.75; N 5.46. C27H N SO requires C, 65.01; H,

2138
L. Doyle, F. Heaney / Tetrahedron 67 (2011) 2132e2138
ꢂ
ꢀ1
6
1
7
(
(
(
.88, N, 5.62%]; mp 110e118 C; vmax (KBr)/cm 3424, 2958, 2663,
741, 1600, 1501, 1197, 1038, 756; (300 MHz, CDCl ) 8.85 (2H, br s)
.61e7.56 (2H, m) 7.39e7.30 (3H, m) 7.27e7.22 (2H, m) 6.85e6.76
2 2
Following extraction with CH Cl optical purity was measured by
d
d
3
HPLC using a chiracel OD column, mobile phase Hexane/iso-propyl
alcohol (1:1), flow rate 1 mL/min; (þ)-5a 7.9 min; (ꢀ)-5a 9.1 min
(þ)-5b was found with 90% ee; (ꢀ)-5a was found enantiomerically
pure.
3H, m) 4.84 (1H, dd J¼8.3 and 8.3) 4.39 (1H, dd J¼7.4 and 3.9) 4.25
1H, dd J¼8.3 and 4.0) 3.92e3.85 (1H, m) 3.73e3.62 (3H, m) 3.03
1H, d J¼14.8) 2.51 (1H, d J¼14.8) 2.42e2.32 (1H, ddd J¼14.9, 12.0
and 3.8) 2.29e2.20 (1H, m) 1.98e1.95 (1H, m) 1.93e1.81 (1H, m)
Acknowledgements
1
1
1
1
.75 (1H, br d J¼18.3) 1.54e1.45 (1H, ddd J¼14.9, 9.3 and 4.7)
.28e1.2 (1H, m) 0.93 (3H, s) 0.70 (3H, s); (75.5, CDCl ) 216.7,
46.9, 146.8, 135.7, 135.6, 129.3, 129.3, 129.2, 129.2, 126.7, 126.7,
19.4, 119.2, 114.7, 114.6, 78.9, 78.8, 78.2, 59.0, 58.8, 58.3, 54.6, 54.4,
d
C
3
Financial support from the Embark Postgraduate Research
Scholarship Scheme (Grant reference number RS/2005/37) and
from the National University of Ireland, Maynooth for a John and
Pat Hume Scholarship to L.D. is gratefully acknowledged.
5
0.3, 47.9, 47.4, 42.8, 42.6, 26.9, 24.6, 19.8, 19.8.
4
7
.2.7. Preparation of the bis-salt from camphor sulfonic acid and rac-
b, 7b.2(þ)-10-CSA 5,6a-diphenylhexahydro-1H-pyrrolo[3,4-c][1,2]
Supplementary data
oxazole-1,5-diium (7,7-dimethyl-2-oxobicyclo[2.2.1]hept-1-yl)meth-
Supplementary data related to this article can be found online at
doi:10.1016/j.tet.2011.01.034. These data include MOL files and
InChIKeys of the most important compounds described in this
article.
anesulfonate. rac-7b (150 mg, 0.564 mmol) and (þ)-10-CSA
(
262 mg, 1.127 mmol) were dissolved in hot acetone (1.8 mL) and
following cooling to rt stirring continued until precipitation was
complete. The salt was filtered and dried to yield the title product as
ꢂ
ꢀ1
a dark green solid (390 mg, 95%); mp 84e96 C; vmax (KBr)/cm
415, 2960, 1744, 1600, 1415, 1154, 1038; (300 MHz, CDCl ) 12.32
3H, br) 7.63e6.91 (10H, m) 4.87 (1H, dd J¼8.3 and 8.3) 4.74 (1H, dd
J¼12.0 and 2.5) 4.45 (1H, br dd J¼8.3 and 2.6) 4.20e4.19 (1H, m)
.97 (1H, d J¼12.0) 3.94e3.74 (2H, m) 3.18 (2H, d J¼14.8) 3.71 (2H,
d J¼14.8) 2.40e2.25 (4H, m) 2.02e1.87 (4H, m) 1.77 (2H, d J¼18.4)
.62e1.53 (2H, m) 1.33e1.25 (2H, m) 0.97 (6H, s) 0.76 (6H, s);
75.5, CDCl ) 217.4, 207.8, 144.3, 144.0, 133.8, 133.6, 129.8, 129.7,
29.4, 127.2, 129.2, 117.1, 116.8, 79.4, 77.5, 60.0, 59.9, 58.4, 56.4, 56.2,
References and notes
3
(
d
d
3
1. Bertelsen, S.; Jørgensen, K. A. Chem. Soc. Rev. 2009, 38, 2178.
2
3
. MacMillan, D. W. C. Nature 2008, 455, 304.
. Gaunt, M. J.; Johansson, C. C. C.; McNally, A.; Vo, N. T. Drug Discovery Today 2007,
3
12, 8.
4. Jang, H.-Y.; Hong, J.-B.; MacMillan, D. W. C. J. Am. Chem. Soc. 2007, 129, 7004.
5
6
. Mayer, S.; List, B. Angew. Chem., Int. Ed. 2006, 45, 4193.
. List, B. Acc. Chem. Res. 2004, 37, 548.
1
(
1
4
d
C
3
7
. Erkkilae, A.; Majander, I.; Pihko, P. M. Chem. Rev. 2007, 107, 5416.
8. Schreiner, P. R. Chem. Soc. Rev. 2003, 32, 289.
. Notz, W.; Tanaka, F.; Barbas, C. F. I. Acc. Chem. Res. 2004, 37, 580.
0. (a) Mangion, I. K.; Northrup, A. B.; MacMillan, D. W. C. Angew. Chem., Int. Ed.
004, 43, 6722; (b) Austin, J. F.; MacMillan, D. W. C. J. Am. Chem. Soc. 2002, 124,
1172.
1. Lemay, M.; Ogilvie, W. W. J. Org. Chem. 2006, 71, 4663.
2. Lemay, M.; Aumand, L.; Ogilvie, W. W. Adv. Synth. Catal. 2007, 349, 441.
3. Cavill, J. L.; Peters, J.-U.; Tomkinson, N. C. O. Chem. Commun. 2003, 728.
9
1
9.8, 48.2, 47.8, 42.8, 42.6, 30.9, 26.9, 25.0, 19.8, 19.7.
2
4
.2.8. Classical resolution of 5a with (þ)-10-CSA. In a modification
34
1
1
1
of a literature procedure rac-5a (234 mg, 1.143 mmol) and (þ)-10-
CSA (266 mg, 1.143 mmol) were dissolved in warm acetone
(
13.5 mL). After cooling to rt stirring was continued for a further
14. Cavill, J. L.; Elliott, R. L.; Evans, G.; Jones, I. L.; Platts, J. A.; Ruda, A. M.; Tom-
2
5
kinson, N. C. O. Tetrahedron 2006, 62, 410.
2
6 h. A first crop of white crystals was collected (145 mg, 58%) [a]
D
15. Brazier, J. B.; Cavill, J. L.; Elliott, R. L.; Evans, G.; Gibbs, T. J. K.; Jones, I. L.; Platts,
þ76 (c 0.0023, MeOH). The mother liquor and washings were
J. A.; Tomkinson, N. C. O. Tetrahedron 2009, 65, 9961.
concentrated to x7 mL and allowed to stir at rt for 4 h after which
16. Heaney, F.; O’Mahony, C. J. Chem. Soc., Perkin Trans. 1 1998, 341; Heaney, F.;
2
5
a second crop was collected (10 mg, 4%) [
The mother liquor and washings were concentrated to x3.5 mL
and left to stir at rt for 26 h prior to collection of a third crop (11 mg,
5
a
]
D
þ74 (c 0.0072, MeOH).
Fenlon, J.; McArdle, P.; Cunningham, D. Org. Biomol. Chem. 2003, 1, 1122.
17. Heaney, F.; Doyle, L. Tetrahedron 2010, 66, 7041e7049.
1
8. ACD Labs, Calculated using Advanced Chemistry Development; Software V8.14 for
Solaris, 1994e2010. Estimated pKBHþ values; isoxazolidine (5.49þ/ꢀ0.2), N-
methylpyrrolidine (10.55þ/ꢀ0.2), N-phenylpyrrolidine (5.68þ/ꢀ0.4) and 3,5-
diphenylisoxazolidine (pKBH 4.20).
2
5
%) [
a
]
D
þ56 (c 0.0006, MeOH). Total yield of (þ)-5a$(þ)-10-CSA
based on the first two crops was 62%.
19. Katagiri, N.; Okada, M.; Morishita, Y.; Kaneko, C. Tetrahedron 1997, 53, 5725.
20. Ranganathan, D.; Ranganathan, S.; Rao, C. B.; Raman, K. Tetrahedron 1981, 37,
4
.2.9. Classical resolution of 5a with (ꢀ)-10-CSA. rac-5a (1500 mg,
629.
2
1. Fedotov, A. N.; Shishkina, I. N.; Mochalov, S. S.; Subbotin, O. A.; Shabarov, Y. S.
Zh. Org. Khim. 1987, 23, 112.
2. Lemay, M.; Ogilvie, W. W. Org. Lett. 2005, 7, 4141.
7.314 mmol) and (ꢀ)-10-CSA (3199 mg, 7.314 mmol) were dissolved
in warm acetone (58 mL). After cooling to rt stirring was continued
2
for a further 26 h after which a first crop of crystals was collected
23. Evans, G.; Gibbs, T. J. K.; Jenkins, R. L.; Coles, S. J.; Hursthouse, M. B.; Platts, J. A.;
Tomkinson, N. C. O. Angew. Chem., Int. Ed. 2008, 47, 2820.
25
(
321 mg, 20%) [
a
]
D
ꢀ74 (c 0.002, MeOH). The mother liquor and
24. Evans, G. J. S.; White, K.; Platts, J. A.; Tomkinson, N. C. O. Org. Biomol. Chem.
washings were concentrated to x49 mL and stirred continued at rt
for 69 h prior to collection of a second crop (144 mg, 9%) [
2006, 4, 2616.
2
5
a]
D
ꢀ78
25. Huanga, D.-M.; Wang, Y.-B.; Viscoc, L. M.; Taoc, F.-M. Chem. Phys. Lett. 2005, 407,
(c 0.0099, MeOH). The collection/concentration process was re-
222.
2
5
26. Nakadai, M.; Saito, S.; Yamamoto, H. Tetrahedron 2002, 58, 8167.
peated affording the following crops: third (93 mg, 6%) [
0
fifth (276 mg, 17%) [a]
D
a
]
D
þ21 (c
27. Kano, T.; Tanaka, Y.; Maruoka, K. Org. Lett. 2006, 8, 2687.
2
5
.008, MeOH), fourth (240 mg, 15%) [
a
]
D
ꢀ75 (c 0.0011, MeOH),
28. Gotoh, H.; Hayashi, Y. Org. Lett. 2007, 9, 2859; Kim, K. H.; Lee, S.; Lee, D.-W.; Ko,
25
þ36 (c 0.0014, MeOH). Total yield of
D.-H.; Ha, D.-C. Tetrahedron Lett. 2005, 46, 5991.
2
3
9. Abiko, A. Rev. Heteroat. Chem. 1997, 17, 51; Abiko, A.; Moriya, O.; Filla, S. A.;
Masamune, S. Angew. Chem., Int. Ed. Engl. 1995, 34, 793.
0. Bartoli, G.; Melchiorre, P. Synlett 2008, 1759.
(ꢀ)-5a$(ꢀ)-10-CSA (crops 1,2 and 4, all white solids) was 44%.
2
5
4
.2.10. Determination of optical purity of resolved 5a. (þ)-5a [
a
]
D
31. Lacour, J.; Linder, D. Science 2007, 317, 462.
32. Ma, Y.; Jin, S.; Kan, Y.; Zhang, Y. J.; Zhang, W. Tetrahedron 2010, 66, 3849.
2
5
þ112 (c 0.0010, MeOH) and (ꢀ)-5a [
a]
D
ꢀ110 (c 0.0064, MeOH)
3
3. Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory Chemicals, 3rd ed.;
were obtained quantitatively by treatment of the corresponding
Pergamon: Oxford, 1988.
salts, (þ)-5a$(þ)-10-CSA and (ꢀ)-5a$(ꢀ)-10-CSA with 1 M NaOH.
34. Abiko, A.; Davis, W. M.; Masamune, S. Tetrahedron: Asymmetry 1995, 6, 1295.