Diastereoselective synthesis of substituted prolines via 5-endo-trig cyclisations of aza-[2,3]-Wittig sigmatropic rearrangement products

Article abstract of DOI:10.1016/j.tet.2010.04.095

The major diastereoisomers of aza-[2,3]-Wittig sigmatropic rearrangement products from α-amino acid derivatives are susceptible to a rare nucleophilic 5-endo-trig cyclisations of an amine onto a non-conjugated vinylsilane in high yield and complete diastereocontrol. Five examples are presented, with cyclisation yields between 35 and 87%. A rationale for the stereoselectivity of the cyclisation is forwarded based upon the steric control factors that have been documented for the aza-[2,3]-Wittig sigmatropic rearrangement. A discussion of the mechanism in the context of the reaction conditions is also presented. Oxidation of the silyl group to a hydroxyl group and complete removal were demonstrated for synthetic utility.

Full text of DOI:10.1016/j.tet.2010.04.095

Tetrahedron 66 (2010) 6300e6308
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Diastereoselective synthesis of substituted prolines via 5-endo-trig cyclisations of
aza-[2,3]-Wittig sigmatropic rearrangement products
*
James C. Anderson , Elizabeth A. Davies
Department of Chemistry, University College London, 20 Gordon Street, London, WC1H 0AJ, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 6 March 2010
Received in revised form 15 April 2010
Accepted 22 April 2010
Available online 29 April 2010
The major diastereoisomers of aza-[2,3]-Wittig sigmatropic rearrangement products from a-amino acid
derivatives are susceptible to a rare nucleophilic 5-endo-trig cyclisations of an amine onto a non-con-
jugated vinylsilane in high yield and complete diastereocontrol. Five examples are presented, with
cyclisation yields between 35 and 87%. A rationale for the stereoselectivity of the cyclisation is forwarded
based upon the steric control factors that have been documented for the aza-[2,3]-Wittig sigmatropic
rearrangement. A discussion of the mechanism in the context of the reaction conditions is also presented.
Oxidation of the silyl group to a hydroxyl group and complete removal were demonstrated for synthetic
utility.
Keywords:
5-endo-trig
Cyclisation
Aza-[2,3]-Wittig
Proline
Ó 2010 Elsevier Ltd. All rights reserved.
Sigmatropic rearrangement
Vinyl silane
1. Introduction
intermediate a-silyl stabilized carbanion. We have summarised
some of our attempted strategies towards preparing enantio-
merically pure aza-[2,3]-Wittig rearrangement products and de-
tailed the use of (ꢀ)-8 phenylmenthol as a chiral auxiliary.9 We
have also shown that internal chirality transfer can also be ef-
fective for the generation of enantiomerically enriched products.10
In all of these studies the majority of reactions gave high yields of
rearrangement product with no evidence of cyclic products being
identified.
We have developed the aza-[2,3]-Wittig sigmatropic rear-
rangement as a synthetic method for the synthesis of unnatural
amino acids (Scheme 1).1 Our work differed from the work of
others in the field whose rearrangements relied upon the relief of
ring strain inherent to b
-lactams2 and aziridines3 that resulted in
ring expansion products. Despite the efficacy of the [2,3]-Wittig
sigmatropic rearrangement for acyclic allylic ethers,3,4 the pro-
motion of the analogous acyclic aza-[2,3]-Wittig sigmatropic
rearrangement proved to be considerably more demanding.5 The
calculated small build up of negative charge at the central vinyl
carbon atom in the transition states of [2,3]-Wittig rearrange-
ments6 led us to develop aza-[2,3]-Wittig rearrangement pre-
cursors that contained an anion stabilizing dimethylphenylsilyl
substituent (Scheme 1). We have defined the limits of activation
and, in addition, control of diastereoselectivity imparted by the
dimethylphenylsilyl group upon the aza-[2,3]-Wittig sigmatropic
rearrangements on acyclic allylic amines7 and have used this
strategy to good effect for the synthesis of naturally occurring
amino acids.8 Conceivably the incorporation of an anion stabiliz-
ing group could have rendered the rearrangement non concerted,
but in all our previous work we never identified any cyclised
products, which may have resulted from protonation of an
Scheme 1. R, R0, R00¼H, alkyl, Ar; G¼carboxylic acid derivative.
We were inspired by Kawabata and Fuji’s work, which in-
vestigated the asymmetric alkylation of
a-amino acids to produce
a,
a
-disubstituted -amino acid derivatives. The success of this work
a
relies upon the transfer of central chirality to an axially chiral
enolate, which is then regenerated as central chirality upon alkyl-
ation.11 An intramolecular example has been developed for the
synthesis of prolines (Scheme 2).11e
* Corresponding author. E-mail address: j.c.anderson@ucl.ac.uk (J.C. Anderson).
0040-4020/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2010.04.095

J.C. Anderson, E.A. Davies / Tetrahedron 66 (2010) 6300e6308
6301
i
Scheme 2. R¼CH3, Pr, Bn, CH2CH2SMe, 4-EtOC6H4CH2.
We had reported the synthesis of
a,a-disubstituted a-amino
acid derivatives from the aza-[2,3]-Wittig rearrangement of
a-
amino acids (Scheme 3).12 Attempts at also trying to access axially
chiral enolates in the initial deprotonation step of enantiomeri-
cally pure rearrangement precursors only resulted in racemic
products.
Scheme 4. Synthesis of amino ester rearrangement precursors.
Scheme 3. Synthesis of a,a-disubstituted a-amino acids.
We reinvestigated this research in order to try and find
conditions that would enable enantioselective rearrangements
via axially chiral enolates.13 Although this was unsuccessful we
were able to characterize an unusual cyclisation reaction that
gave densely substituted prolines in diastereomerically defined
form.
Scheme 5. Formation of proline 7 and postulated mechanism.
2. Results and discussion
We thought that treatment of 6 with KH and 18-C-6 would lead
to the aza-[2,3]-Wittig rearranged product,12 but for the very first
time we isolated the cyclised product 7 in 79% yield (Scheme 5). The
reaction was monitored closely by TLC and over 1 h only the dis-
appearance of starting material and formation of 7 was observed.
Maintaining the reaction at 0 ꢁC for 1 h or longer resulted only in
the isolation of starting material. Possibly the proline 7 could be
Our published route to the required rearrangement precursors
derived from enantiomerically pure
a-amino acids required the
use of neat DIBAL for the synthesis of Z-1-bromo-2-(phenyl-
dimethylsilyl)-but-3-ene. Problems with the acquisition of neat
DIBAL led us to develop an alternative route. Treatment of alcohol
17b with SOCl2, according to the method of Chan et al.,14 led to an
inseparable mixture of the three allylic chlorides 2e4 (20:1:3) in
98% yield (Scheme 4). Determination of the stereochemistry of
alkene 2 was by one dimensional NOE experiments. Irradiation of
formed from an incomplete rearrangement pathway, where the a-
silyl anion intermediate 8 was protonated on work up. However,
quenching the reaction with MeOD led to no deuterium in-
corporation. As KH was used as base the scenario of protonation of
8 from the reaction solvent led us to conduct the experiment in
-THF-d8. Again no deuteration was observed.
the vinylic proton (
d
6.57, 1H, q, J¼7.0 Hz) induced an enhance-
ment at CH2Cl ( 4.22, 2H, s) and CHCH3
d
(d
1.68, 3H, d, J¼7.1 Hz),
which strongly suggested Z-stereochemistry. This assignment was
supported by irradiation of the minor compound 3 at the vinylic
The 1H NMR spectrum of 7 showed a number of duplicated
peaks, which suggested the presence of diastereoisomers,
rotamers or impurities. Removal of the Boc group with 4 M HCl in
dioxane gave 9 a single diastereoisomer by crude 1H NMR and
was purified to give an oil in 86% yield. The diastereomer depicted
(Scheme 5) was verified by one dimensional NOE experiments.
proton (
solely at CHCH3
d
6.19, 1H, q, J¼6.7 Hz), which induced an enhancement
(d
1.87, 3H, d, J¼6.8 Hz), suggesting E-stereo-
chemistry. The structure of compound 4 was assigned based on
the remaining 1H NMR signals. Treatment of the N-Boc methyl
esters of glycine, alanine and phenylalanine with KH, followed by
the mixture of chlorides 2e4, in the presence of catalytic tetra-n-
butyl ammonium iodide, led to the rearrangement precursors
5aec in good yields (Scheme 4). The valine analogue 5d was
prepared from alkylation of valine methyl ester with isomeric
chlorides 2e4 and then Boc protection in refluxing Et3N. Rear-
rangement precursors 5aed were all isolated free from isomeric
impurities that could have conceivably arisen from reaction with
3 or 4. The alanine amide derivative 6 (Scheme 5) was prepared
as previously described.12
Key enhancements were observed upon irradiation of H-3 (
d
2.53,
1H, appt. quintet, J¼6.6 Hz) at H-4 (6.7%,
d
1.46, 1H, ddd, J¼11.8,
9.0, 5.9 Hz), which suggested that H-3 and H-4 are on the same
face of the proline ring. This interpretation was supported by
a 7.0% enhancement for the NOE experiment in the other di-
rection. Irradiation of H-3 also caused an enhancement of the CH3
groups of the ethyl substituents of the amide group (1.2%,
d 1.46,
6H, t, J¼7.0 Hz), suggesting that the amide group is also on the

6302
J.C. Anderson, E.A. Davies / Tetrahedron 66 (2010) 6300e6308
same face as H-3. This was corroborated by an enhancement of
the C-2 CH3 group (1.8%, 1.27, 3H, s) upon irradiation of the C-3
CH3 group (
0.93, 3H, d, J¼7.2 Hz). In addition previous studies
had determined that aza-[2,3]-Wittig rearrangement of 6 gave the
2S ,3R
diastereoisomer (dr>20:1, Scheme 1).12 It therefore
aza-[2,3]-Wittig rearrangement was unfavourable under the re-
action conditions. As before, the 1H NMR spectrum of 12b con-
tained multiple peaks. This was simplified by removing the Boc
group with 4 M HCl in dioxane and gave 13b as a single di-
astereoisomer by 1H NMR.
d
d
*
*
seemed probable that this relative stereochemistry was also
exhibited in proline 7 (and 8) and that the C-4 silyl substituent
As suitable crystals could not be obtained for X-ray structure
determination, one dimensional NOE experiments were conducted
in order to verify the stereochemistry at C-4. The relative stereo-
was thus 4R . The isolation of a cyclic product from an attempted
*
aza-[2,3]-Wittig rearrangement was an important discovery and
prompted us to further investigate the generality of this process
and the mechanism of a rare 5-endo-trig cyclisation.
The amide rearrangement precursor 6, derived from alanine,
was the only a-amino acid that we had found to undergo the aza-
chemistry between C-2 and C-3 had already been established as
12
2R
*
, 3R
*
.
Irradiation of H-3 (
d
2.81, 1H, app. quintet, J¼7.0 Hz)
induced an enhancement at H-4 (2.2%,
d
1.80, 1H, ddd, J¼11.2, 9.4,
6.4 Hz), which was also detected in the reverse experiment (1.9%).
These were similar results to some of the NOE enhancements we
had observed for 9(7) and led us to assign the C-4 stereochemistry
[2,3]-Wittig rearrangement in this class of substrate. We had in-
vestigated many other base systems to investigate the rearrange-
ment of amide precursors derived from phenyl glycine and valine,
and concluded that steric inhibition of resonance prevented
deprotonation.12 No cyclic products were observed in any of these
experiments. The analogous methyl ester derivatives rearranged
with opposite and varying degrees of diastereoselectivity (Scheme
1). We had shown that the best base systems for aza-[2,3]-Wittig
rearrangements were either nBuLi or LDA in THF/HMPA or DMPU
mixtures, or KH, 18-C-6 in THF.7 We quickly found that for pre-
cursors 5bed treatment with KHMDS in THF/DMPU (4:1), led to
efficient cyclisation reactions over prolonged stirring at rt (Scheme
6, and Table 1).
as 4R .
*
The phenylalanine derivative 5c had previously been shown
to undergo aza-[2,3]-Wittig rearrangement with KH/18-C-6 to
give a surprising dr of 1:1 (Scheme 1).12 Treatment of 5c with
KHMDS and stirring at rt for 14 h led to the isolation of aza-
[2,3]-Wittig rearranged product 11c (37%) and cyclised product
12c in 35% yield. The relative stereochemistry of 11c was con-
firmed by single crystal X-ray structure determination.13 The 1H
NMR of proline 12c was over complicated again so the Boc
group was removed as before (Scheme 7) to give 13c in 98%
yield. Neither of pyrrolidines 12c or 13c crystallized so one
dimensional NOE experiments were conducted on 13c to con-
firm the relative stereochemistry. As before, mutual enhance-
ments between H-3 (3.9%,
and H-4 (5.5%,
C-4 stereochemistry as 4R
hancement at CH2Ph (1.2%,
diation of H-3, there was no other NOE data to confirm the
stereochemistry of C-2. However, the X-ray data of 11c had
d
2.77, 1H, app. quintet, J¼6.7 Hz)
d
1.55, 1H, ddd, J¼11.2, 9.9, 6.2 Hz) confirmed the
*
. Although we observed an en-
d
3.12, 1H, d, J¼12.8 Hz) upon irra-
confirmed that the 2S
*
diastereisomer had not rearranged,
a result consistent with the isolation of 11b, and suggested that
the stereochemistry of 12c was the depicted 2R
*
,3R
*
,4R
*
diastereoisomer.
Scheme 6. Cyclisation of amino ester precursors 5bed.
L
Aza-[2,3]-Wittig rearrangement of 5b using KH and 18-C-6 had
led to the rearranged products 10b:11b (7:1) in 77% yield (Scheme
1).12 Treatment with KHMDS with prolonged stirring at rt for 14 h
led to the isolation of cyclised material 12b (41%) and 10b:11b in
a reduced diastereomeric ratio of 3:1 in 40% yield. Monitoring of
the reaction by TLC revealed the disappearance of starting material
to give a spot corresponding to rearranged compounds 10b:11b.
This spot then slowly became smaller as another spot corre-
sponding to the cyclised product slowly grew in. This suggested
that cyclisation occurs after rearrangement (vide infra). Allowing
the reaction to stir for 48 h at rt led to the isolation of the cyclised
product 12b in 68% yield and only 11% of the rearranged products
remained 10b:11b with a reversed diastereomeric ratio of 1:10. This
suggested that cyclisation of the minor diastereoisomer from the
Scheme 7. Boc deprotection.
Treatment of the valine derived precursor 5d under the cycli-
sation conditions led to rapid rearrangement by TLC. Prolonged
stirring for 48 h at rt led to the exclusive formation of pyrrolidine
12d. This is in line with the aza-[2,3]-Wittig rearrangement of this
substrate (Scheme 1), which is known to proceed with near
complete control of diastereoselectivity. The 1H NMR of 12d was
again complicated, but showed only one compound when the Boc
group was removed (Scheme 7, 99%). Fortunately suitable crystals
of 12d could be grown and an X-ray structure determination
confirmed the relative stereochemistry depicted (Scheme 8).13
This is consistent with the stereochemical assignment of the
rearranged product derived from 5d,12 and would again suggest
that the major diastereoisomer from the rearrangement of these
amino ester derived precursors are prone to cyclisation under
these particular reaction conditions. One dimensional NOE ex-
periments on 12d gave enhancements consistent with this
structure (Scheme 8) and corroborated the previous assignments
for 12b and 12c.
Table 1
aCyclisation and rearrangement of amino ester precursors 5bed
5
R
t (h)
Yield 10 (%)
Yield 11 (%)
b
c
d
Me
Bn
iPr
48
14
48
11
37
d
68
35
87
a
KHMDS (2.5 equiv), THF/DMPU (4:1), 0 ꢁC to rt.

J.C. Anderson, E.A. Davies / Tetrahedron 66 (2010) 6300e6308
6303
crystallise and no clearly defined single signals in the 1H NMR of 19
made it impossible for the relative stereochemistry between the
quaternary centre and the C-silyl centre to be determined. By
analogy with the previous cyclisations reported in this paper, we
tentatively assign the stereochemistry of the pyrrolidine 18 as that
depicted.
To show the utility of this cyclisation for the synthesis of
substituted proline derivatives we oxidized the phenyldimethylsilyl
group with mercury(II) acetate and peracetic acid. Under these
conditions alcohol 23 was isolated in 67% yield (Scheme 12).
Scheme 8. X-ray structure of 12d13
.
Conformation that cyclisation proceeded from the rearranged
product was provided by treatment of rearrangement product 14
with KHMDS (Scheme 9). This led to the isolation of the corre-
sponding pyrrolidine 12d in 92% isolated yield.
Scheme 12. (i) Hg(OAc)2, AcO2H (35 wt % in AcOH), 4 h. (ii) TBAF, wet THF, 80 ꢁC, 2 h.
The silicon group had been essential to facilitate and control the
diastereoselectivity of the aza-[2,3]-Wittig rearrangement and
subsequent cyclisation. Efficient protodesilylation could be ach-
ieved by using the conditions of Roush et al.21 to give 24 in 74%
yield.
The results suggest that the cyclisation occurs from the major
diastereoisomer of the aza-[2,3]-Wittig rearranged products 10, but
not 11 (Scheme 6). The stereochemical rationale for the aza-[2,3]-
Wittig rearrangement we believe is determined by the least pseudo
1,3-diaxial interaction between the phenyldimethylsilyl group and
the substituent adjacent to the migrating centre.7,8,10,12 In the ester
series 5bed, the alkyl group is considered the larger group based
upon A-value analysis and gives rise to predominantly isomer 10.12
The phenylalanine derivative is an anomalous result, which rear-
ranges with no diastereoselectivity. The major compound 10 can
then theoretically cyclise through one of two extreme transitions
state structures 22 and 23 (Scheme 13). Presumably the pseudo 1,3-
diaxial interaction with the silicon substituent is more demanding
than any 1,2 interaction, so transition state structure 22 is favoured,
it would seem exclusively, over 23, leading to intermediate silyl
anion 24 Protonation then occurs from the least hindered face of
this more reactive conformer leading to the observed stereo-
chemistry for the proline products 12bed. The rearrangement of
amide precursor 6 also fits this stereochemical model. In this sys-
tem the diethylamide group acts as the larger substituent with
respect to the methyl group and controls the diastereoselectivity of
the rearrangement and the cyclisation to give 7. In this rearrange-
ment and like valine precursor 5d no minor diastereosiomer was
detected. The cyclisation of the amide aza-[2,3]-Wittig rearranged
product would seem to be particularly fast, so that none of the
rearranged product could be detected by TLC. The rate of this par-
ticular cyclisation reaction could be enhanced by the amide group
increasing the Thorpe/Ingold effect with respect to the ester
derivatives.
Scheme 9. i) KH, 18-C-6, THF, 0 ꢁC to rt, 72%.12 (ii) KHMDS, THF/DMPU (4:1), 0 ꢁC to rt,
14 h, 92%.
To determine the importance of the a,a-disubstitution pattern of
the precursors with respect to cyclisation, treatment of 5a with
KHMDS under identical conditions led only to the normal rear-
ranged product 15 in 91% yield and dr>20:1 (Scheme 10).7d No
cyclisation product was observed and highlights the importance of
the geminal substitution pattern enhancing cyclisation, possibly
due to the Thorpe/Ingold effect.15
Scheme 10. Attempted cyclisation of glycine derivative 5a.
We were also interested to see whether the cyclisation would
proceed on a terminal alkene. Treatment of 16 under the cyclisation
conditions led to the isolation of the rearranged compound 17 (14%)
and the cyclised product 18 (67%) (Scheme 11).
Deuterium quench studies on the cyclisation of 6 had not
revealed any deuteration. Similarly when the cyclisation of 5d
(Scheme 6) was quenched with MeOD, no deuterium in-
corporation was detected in the pyrrolidine 12d. These results
would suggest that protonation of the developing cyclic in-
termediate had already occurred. In the experiments using
KHMDS an equivalent of hexamethyldisilazane is produced. In the
Scheme 11. Cyclisation of terminal alkene substrate.
route towards cyclisation the developing
a-silyl carbanion could
be stabilized and eventually protonated by hexamethyldisilazane
(Scheme 14).
Removal of the Boc group from 18 in quantitative yield gave
a single diastereoisomer 19 by 1H NMR. The reluctance of 18 or 19 to

6304
J.C. Anderson, E.A. Davies / Tetrahedron 66 (2010) 6300e6308
Scheme 13. Explanation for diastereoselective cyclisations.
pattern adjacent to the migrating carbanion is also an essential
structural motif for rearrangement (compare Scheme 10 with
Schemes 6 and 11).
3. Conclusion
The cyclisations 6 to 7 (Scheme 5), 5bed to 12bed (Scheme 6)
and 16 to 18 (Scheme 11) are rare examples of 5-endo-trig cyclisa-
tion processes. There is compelling evidence to believe that cycli-
sation occurs after aza-[2,3]-Wittig sigmatropic rearrangement and
that cyclisation occurs from the major, favoured diastereoisomer,
suggesting that factors, which have been detailed to control the
stereochemistry of the rearrangement are also strongly dominant
for the efficacy of the cyclisation reaction. They can be described as
nucleophilic 5-endo-trig cyclisations of an amine onto a non-con-
jugated alkene. To the best of our knowledge these represent the
first examples of this type of reaction involving a carbamate pro-
tected nitrogen and the first examples of any type of 5-endo-trig
cyclisation onto a vinylsilane.
Scheme 14. Possible cyclisation transition state model.
This would render the reaction irreversible if we assume that
the KHMDS is then not basic enough to deprotonate to silicon and
a
facilitate the ring opening process. This mechanism is consistent
with our previous observations that under the base systems, which
we had previously used for aza-[2,3]-Wittig rearrangement (nBuLi
or LDA in THF/HMPA or DMPU mixtures, or KH, 18-C-6 in THF) 7 no
cyclisation products had been isolated or detected. Presumably an
acidic enough proton is not available for protonation of a de-
veloping silicon stabilized carbanion. However, this still does not
explain the result of amide precursor 6, which was rearranged/
cyclised with KH and even in THF-d8 as reaction solvent still didn’t
give any detectable deuterium incorporation! This result remains
an anomaly for future investigation, but it may be possible that this
substrate is more prone to cyclisation because of the bulky amide
substituent.
4. Experimental
4.1. General
This mechanism would constitute a 5-endo-trig process, which
is normally geometrically disfavoured.16 However, there have been
numerous accounts of the employment of 5-endo-trig cyclisations
in the synthesis of five-membered rings, suggesting that the pro-
cess may be more useful than previously considered. Of particular
interest to this work is the formation of nitrogen heterocycles via
a 5-endo-trig cyclisation. This transformation allows for the con-
venient preparation of the prolific pyrrolidine motif and can be
divided into three classes: radical-initiated,17 electrophile-driven
and nucleophile-driven. The nucleophile-driven, anionic 5-endo-
trig cyclisation of homoallylic amines is less common than its cat-
ionic counterpart and usually necessitates the presence of an
electron-withdrawing group or a conjugated group in the central
vinylic position. For example, Padwa et al. have shown that 5-endo-
trig cyclisation of vinylsulfones occurs in high yields.18 Craig et al.
have also made use of the 5-endo-trig cyclisation of vinylsulfones in
their total synthesis of (þ)-monomorine I.19 Pyrrolidine formation
via nucleophilic 5-endo-trig cyclisation has also been realised with
the use of a trifluoromethyl group.20 The cyclisation was attributed
to the highly electrophilic nature of the double bond and the sta-
All non-aqueous reactions were carried out under an oxygen-
free atmosphere of nitrogen in flame-dried glassware with rigorous
exclusion of moisture. All reactions were monitored by thin layer
chromatography using Merck 5554 60F254 silica gel coated plates.
Visualisation was achieved using ultraviolet light and then either
potassium permanganate or anisaldehyde. Flash-column chroma-
tography was performed using Merck silica gel 60 as the stationary
phase and Fisher, certified or specified grade solvents. Solvents and
reagents were either used as supplied or, when necessary, purified
in accordance with standard procedures as described below. Ace-
tonitrile (bp 82 ꢁC), DMPU (bp 146 ꢁC at 44 mm Hg) and triethyl-
amine (bp 89 ꢁC) were distilled from calcium hydride and stored
over 4 Å molecular sieves. Anhydrous grade dichloromethane (bp
39 ꢁC) was distilled from calcium hydride immediately prior to use.
Anhydrous grade tetrahydrofuran (bp 66 ꢁC) was freshly distilled
from sodium/benzophenone ketyl immediately prior to use. 18-
crown-6 was recrystallised from acetonitrile. For work-up pur-
poses, standard laboratory grade solvents were used as supplied.
bilised a-CF3 carbanion intermediate, both of which are caused by
4.1.1. Allyl chlorides (2e4). To a stirred solution of alcohol 1 (85 mg,
0.41 mmol) in Et2O (2 mL) was added dropwise via cannula a so-
lution of thionyl chloride (59 mg, 0.50 mmol) in Et2O (1 mLþ0.5 mL
wash). The reaction was stirred for 20 h, after which time the crude
material was adsorbed onto silica. Gel filtration (eluting with pet.
ether) furnished chlorides 2, 3 and 4 (91 mg, 98%) as an inseparable
mixture in a ratio of 20:3:1; Rf¼0.51 (100% pet. ether); IR nmax (film)
the strong electron-withdrawing ability of the CF3 group. This
demonstrates the nucleophilic 5-endo-trig cyclisation of an amine
onto a non-conjugated alkene. Although the dimethylphenylsilyl
group is not electron-withdrawing, its ability to stabilise a-negative
charge must be a contributory factor that allows cyclisation for
precursors 6 and 5bed. In addition the geminal substitution

J.C. Anderson, E.A. Davies / Tetrahedron 66 (2010) 6300e6308
6305
3070e2914 (CeH), 1613, 1427, 1251, 1131, 1116, 1068 cmꢀ1; 1H NMR
2 (400 MHz, CDCl3)
purified by column chromatography using SCX powder. Before
loading the product, the column was flushed with AcOH (5% in
MeOH), then MeOH. The crude material was loaded in CH2Cl2 and
the column was once again flushed with MeOH. Methanolic am-
monia (2%) was then used as the eluent to furnish 9 (132 mg, 86%)
as a colourless oil; IR nmax (solution in CH2Cl2) 3019 (NeH), 2873
d
0.52 (6H, s, Si(CH3)2), 1.68 (3H, d, J¼7.1, C]
CHCH3), 4.22 (2H, s, CH2Cl), 6.57 (1H, q, J¼7.0, C]CHCH3), 7.38e7.41
(3H, m, ArH), 7.56e7.62 (2H, m, ArH); 1H NMR 4 (400 MHz, CDCl3)
d
0.43 (3H, s, Si(CH3)2), 0.52 (3H, s, Si(CH3)2), 1.56 (3H, d, J¼6.7,
CHCH3), 4.74 (1H, q, J¼6.7, CHCH3), 5.59 (1H, s, C]CH2), 6.12 (1H, s,
C]CH2), 7.38e7.41 (3H, m, ArH), 7.56e7.62 (2H, m, ArH); 1H NMR 3
(CeH), 1615 (C]O), 1360, 1215, 1109, 1064 cmꢀ1
;
1H NMR
(400 MHz, CDCl3)
d
0.52 (6H, s, Si(CH3)2), 1.87 (3H, d, J¼6.8, C]
(500 MHz, CDCl3) 0.27 (3H, s, Si(CH3)2), 0.35 (3H, s, Si(CH3)2), 0.93
d
CHCH3), 4.19 (2H, s, CH2Cl), 6.19 (1H, q, J¼6.7, C]CHCH3), 7.38e7.41
(3H, d, J¼7.2, CHCH3), 1.11 (6H, t, J¼7.0, N(CH2CH3)2), 1.27 (3H, s,
CCH3), 1.46 (1H, ddd, J¼11.8, 9.0, 5.9, CHSiMe2Ph), 2.53 (1H, app.
quintet, J¼6.6, CHCH3), 2.99 (1H, app. t, J¼11.1, NCH2), 3.13 (1H, app.
t, J¼9.6, NCH2), 3.25e3.50 (4H, br m, N(CH2CH3)2), 7.33e7.34 (3H,
(3H, m, ArH), 7.56e7.62 (2H, m, ArH); 13C NMR 2 (400 MHz, CDCl3)
d
ꢀ1.7 (Si(CH3)2), 18.2 (C]CHCH3), 52.3 (CH2Cl), 127.9 (CH), 129.7
(CH), 134.1 (CH), 137.6 (C), 138.5 (C), 144.7 (CH). Anal. Calcd for
C12H17ClSi: C 64.11, H 7.62, found: C 63.97, H 7.58 %.
m, ArH), 7.48e7.49 (2H, m, ArH); 13C NMR (500 MHz, CDCl3)
d
ꢀ3.0
(Si(CH3)2), ꢀ2.3 (Si(CH3)2), 12.6 (CH3), 13.6 (CH3), 13.9 (CH3), 21.9
(CH3), 32.3 (CH), 41.1 (CH2), 41.3 (CH), 42.7 (CH2), 45.7 (CH2), 70.5
(C), 127.8 (ArCH), 129.0 (ArCH), 133.8 (ArCH), 138.8 (ArC), 176.3 (C]
O); m/z (ESþ) 333 (100%, MHþ); HRMS C19H33N2OSi calcd 333.2357,
found 333.2357.
4.1.2. Glycine ester rearrangement precursor (5a). Prepared
according to the method previously reported for the allyl bromide,
but with the addition of tetra-n-butyl ammonium iodide.7d Data
was also identical to that reported earlier.7d
4.1.6. (2R
*
,3R
*
,4R )-4-(Dimethylphenylsilanyl)-2,3-dimethylpyrroli-
*
4.1.3. Alanine, phenylalanine and valine ester rearrangement pre-
cursors (5bed). Prepared according to the method previously
reported for the allyl bromide, but with the addition of tetra-n-
butyl ammonium iodide.12 Data was also identical to that reported
earlier.12
dine-1,2-dicarboxylic acid 1-tert-butyl ester 2-methyl ester (12b). To
a stirred solution of 5b (616 mg, 1.58 mmol), dried by azeotrope
from toluene, in THF/DMPU (4:1, 5 mL) at 0 ꢁC was added KMHDS
(2.5 equiv of 0.5 M solution in toluene) dropwise. The reaction
was stirred for 48 h, allowing to warm to rt. The reaction was
then quenched with saturated aq NH4Cl (5 mL) and Et2O (5 mL)
added. The layers were separated and the aqueous layer was re-
extracted with Et2O (2ꢂ5 mL). The combined organic layers were
washed with brine (2ꢂ15 mL), dried (MgSO4) and the solvent
removed in vacuo. The crude products were purified by flash-
column chromatography (25% Et2O/pet. ether) to furnish 12b
(420 mg, 41%) as a colourless oil; Rf¼0.22 (25% Et2O/pet. ether); IR
nmax (solution in CHCl3) 2954e2879 (CeH), 1732 (C]O), 1683 (C]
4.1.4. (2S
*
,3R
*
,4R )-2-Diethylcarbamoyl-4-(dimethylphenylsilanyl)-
*
2,3-dimethylpyrrolidine-1-carboxylic acid tert-butyl ester (7). To
a stirred suspension of KH (223 mg, 30% suspension in mineral oil,
washed twice with pet. ether, 5.56 mmol) in THF (16 mL) at 0 ꢁC
was added 6 (801 mg, 1.85 mmol) and 18-crown-6 (490 mg,
1.85 mmol) in THF (5 mLþ1 mL wash). The reaction was stirred for
1 h, then quenched with saturated aq NH4Cl (10 mL). Et2O (10 mL)
was added and the layers were separated. The aqueous layer was
re-extracted with Et2O (10 mL) and the combined organic layers
were washed with water (15 mL) then brine (15 mL), dried (MgSO4)
and concentrated in vacuo. Purification by flash-column chroma-
tography (50% Et2O/pet. ether) yielded 7 (602 mg, 75%) as a col-
ourless oil; Rf¼0.32 (50% Et2O/pet. ether); IR nmax (solution in
CHCl3) 2975 (CeH), 1681 (C]O), 1629 (C]O), 1455, 1374,
O), 1392, 1368, 1118 cmꢀ1
;
1H NMR (500 MHz, DMSO-d6)
d
0.30e0.35 (6H, m, Si(CH3)2), 0.89rot., 0.91rot. (3H, d, J¼7.3,
CHCH3), 1.30e1.43 (12H, m, C(CH3)3, CCH3), 1.77 (1H, ddd, J¼12.5,
8.1, 6.2, CHSiMe2Ph), 2.27rot., 2.36rot. (1H, app. quintet, J¼6.7,
CHCH3), 3.29e3.41 (1H, m, NCH2), 3.55e3.66 (4H, m, NCH2, OCH3),
7.38e7.41 (3H, m, ArH), 7.49e7.54 (2H, m, ArH); 13C NMR
(500 MHz, DMSO-d6)
d
ꢀ2.4 (Si(CH3)2), ꢀ2.2 (Si(CH3)2), 13.9 (CH3),
1114 cmꢀ1 1H NMR (500 MHz, DMSO-d6)
; d 0.26e0.47 (6H, m, Si
14.1 (CH3), 19.7 (CH3), 20.7 (CH3), 28.3 (CH), 28.5 (CH), 28.6 (CH),
44.7 (CH), 45.9 (CH), 48.5 (CH2), 48.7 (CH2), 52.6 (CH3), 52.7 (CH3),
69.1 (C), 69.5 (C), 79.1 (C), 79.3 (C), 128.4 (CH), 128.5 (CH), 129.7
(CH3)2), 0.88rot., 0.92rot. (3H, d, J¼6.8, CHCH3), 0.98e1.05 (6H, m, N
(CH2CH3)2), 1.32rot., 1.38rot. (9H, s, C(CH3)3), 1.50e1.53 (4H, m, CCH3,
CHSiMe2Ph), 2.29rot.
3.00e3.55 (6H, m, NCH2, N(CH2CH3)2), 7.37e7.40 (3H, m, ArH),
7.47e7.48 (2H, m, ArH); 13C NMR (500 MHz, DMSO-d6)
,
2.38rot. (1H, br quintet, J¼6.6, CHCH3),
(CH), 133.9 (CH), 134.1 (CH), 138.1 (C), 138.2 (C), 154.0 (C]O), 175.1
13
(C]O), 175.4 (C]O); C NMR (270 MHz, DMSO-d6, 80 ꢁC)
d
ꢀ2.6
d
ꢀ2.0 (Si
(Si(CH3)2), ꢀ2.1 (Si(CH3)2), 14.0, 28.7, 28.9, 48.8, 52.5, 69.2, 79.1,
128.4, 129.7, 133.9, 138.4, 153.5 (C]O), three peaks missing; m/z
(ESþ) 414 (100%, MNaþ), 358 (73%, MNaþꢀC4H8), 314 (93%,
MNaþꢀBoc); HRMS C21H33NNaO4Si calcd 414.2071, found
414.2075. Compound 10b (68 mg, 11%, dr 10) was also isolated
and had identical data to that reported.12
(CH3)2), 13.3 (CH3), 13.6 (CH3), 21.1 (CH3), 22.4 (CH3), 27.3 (CH3), 27.6
(CH3), 28.5 (CH), 28.8 (CH), 42.9 (CH), 43.9 (CH), 48.1 (CH2), 48.6
(CH2), 71.5 (C), 71.7 (C), 78.3 (C), 78.6 (C),128.4 (ArCH),129.7 (ArCH),
133.7 (ArCH), 138.3 (ArC), 155.7 (C]O), 171.4 (C]O), 171.9 (C]O);
13C NMR (270 MHz, DMSO-d6, 80 ꢁC)
d
ꢀ2.8 (Si(CH3)2), ꢀ2.1 (Si
(CH3)2), 13.5, 13.6, 21.5, 27.8, 28.8, 43.5, 48.6, 71.9, 78.7, 128.4, 129.6,
133.9, 138.5, 172.1 (C]O), two peaks missing; m/z (ESþ) 455 (100%,
MNaþ), 333 (60%, MHþ); HRMS C24H40N2NaO3Si calcd 455.2700,
found 455.2681.
4.1.7. (2R
*
,3R
*
,4R )-4-(Dimethylphenylsilanyl)-2,3-dimethylpyrroli-
*
dine-2-carboxylic acid methyl ester (13b). Boc-protected proline
12b (100 mg, 0.256 mmol) was deprotected according to the
method for 9 except that 13b (72 mg, 98%) was isolated as a col-
ourless oil with no need for further purification. IR nmax (solution in
CHCl3) 3168 (NeH), 2958 (CeH), 1710 (C]O), 1454, 1372,
4.1.5. (2S
*
,3R
*
,4R )-4-(Dimethylphenylsilanyl)-2,3-dimethylpyrroli-
*
dine-2-carboxylic acid diethylamide (9). To Boc-protected pyrroli-
dine 7 (200 mg, 0.463 mmol) was added, with stirring, HCl (3 equiv
of 4 M solution in dioxane). The reaction was stirred for 3 h, after
which time the solvent was removed in vacuo. The residue was
taken up in CH2Cl2 (3 mL) and saturated aq NaHCO3 (3 mL/mmol)
added. The layers were separated and the aqueous layer was re-
extracted with CH2Cl2 (5 mL/mmol). The combined organic layers
were washed with brine (2ꢂ10 mL), dried (MgSO4) and the solvent
removed in vacuo to yield the crude pyrrolidine 9, which was
1039 cmꢀ1 1H NMR (500 MHz, C6D6)
; d 0.35 (3H, s, Si(CH3)2), 0.42
(3H, s, Si(CH3)2), 0.99 (3H, d, J¼7.3, CHCH3), 1.37 (3H, s, CCH3), 1.80
(1H, ddd, J¼11.2, 9.4, 6.4, CHSiMe2Ph), 2.12 (1H, br s, NH), 2.81 (1H,
app. quintet, J¼7.0, CHCH3), 3.18 (1H, app. t, J¼10.5, NCH2), 3.28 (1H,
app. t, J¼9.6, NCH2), 3.39 (3H, s, OCH3), 7.29e7.32 (3H, m, ArH),
7.51e7.52 (2H, m, ArH); 13C NMR (400 MHz, CDCl3)
(CH3)2), 0.0 (Si(CH3)2), 16.1 (CH3), 23.8 (CH3), 34.3 (CH), 44.5 (CH),
48.6 (NCH2), 73.0 (NHC), 130.3 (ArCH), 131.6 (ArCH), 136.2 (ArCH),
d
ꢀ0.7 (Si

6306
J.C. Anderson, E.A. Davies / Tetrahedron 66 (2010) 6300e6308
141.1 (ArC), 181.5 (C]O); m/z (ESþ) 292 (100%, MHþ); HRMS
C16H26NO2Si calcd 292.1727, found 292.1730.
(CH), 31.9 (CH), 32.9 (CH), 38.8 (CH), 49.9 (CH2), 50.1 (CH2), 51.3
(CH3), 51.5 (CH3), 75.9 (C), 76.4 (C), 79.2 (C), 79.3 (C), 128.4 (ArCH),
129.6 (ArCH), 133.9 (ArCH), 138.4 (ArC), 153.5 (C]O), 154.1 (C]O),
13
4.1.8. (2R
*
,3R
*
,4R
*
)-2-Benzyl-4-(dimethylphenylsilanyl)-3-methyl-
173.7 (C]O), 174.0 (C]O); C NMR (270 MHz, DMSO-d6, 80 ꢁC)
pyrrolidine-1,2-dicarboxylic acid 1-tert-butyl ester 2-methyl ester
(12c). Cyclisation of 5c (103 mg, 0.221 mmol) was achieved via the
conditions for 12b, except stirring for 14 h. Purification by flash-
column chromatography (25% Et2O/pet. ether) gave 12c (36 mg,
35%) as a colourless oil; Rf¼0.19 (25% Et2O/pet. ether); IR nmax (so-
lution in CHCl3) 2978 (CeH), 1732, (C]O), 1683 (C]O), 1393, 1367,
d
ꢀ2.4 (Si(CH3)2), ꢀ1.8 (Si(CH3)2), 17.4, 19.8, 20.0, 28.6, 30.8, 32.6,
50.3, 51.2, 76.6, 79.4, 128.5, 129.6, 133.9, 138.7, 153.7 (C]O), 173.7
(C]O); m/z (ESþ) 442 (40%, MNaþ), 320 (100%, MHþꢀBoc); HRMS
C23H37NNaO4Si: calcd 422.2384, found 442.2387; Anal. Calcd for
C23H37NO4Si: C 65.83, H 8.89, N 3.34, found C 65.57, H 8.90, N 3.29%.
Suitable crystals for X-ray analysis were grown from Et2O/Petrol.
1114 cmꢀ1 1H NMR (500 MHz, CDCl3)
; d 0.05e0.32 (6H, m, Si
(CH3)2), 0.73e0.85 (3H, m, CHCH3), 1.13rot., 1.42rot., 1.46rot. (9H, s, C
(CH3)3), 1.82e1.98 (1H, m, CHSiMe2Ph), 2.56rot., 2.71rot. (1H, app.
quintet, J¼8.4, CHCH3), 2.97e3.85 (7H, m, NCH2, OCH3, CH2Ph),
7.13e7.23 (5H, m, ArH), 7.35e7.38 (3H, m, ArH), 7.51e7.53 (2H, m,
4.1.11. (2R
*
,3R
*
,4R )-4-(Dimethyphenylsilanyl)-2-isopropyl-2-me-
*
thoxy-carbonyl-3-methylpyrrolidium chloride (13d). Boc protected
proline 12d (156 mg, 0.372 mmol) was deprotected according to
the method for 9 except that 13d (131 mg, 99%) was isolated as
a colourless solid (mp 80.4e81.5 ꢁC) with no need for further pu-
rification. IR nmax (solution in CHCl3) 2926e2854 (CeH), 1746 (C]
ArH); 13C NMR (500 MHz, DMSO-d6)
d
ꢀ2.6 (Si(CH3)2), ꢀ2.1 (Si
(CH3)2), 16.0 (CH3), 28.0 (CH), 28.6 (CH), 28.8 (CH), 29.1 (CH), 36.6
(CH2), 37.2 (CH2), 45.1 (CH), 47.3 (CH), 48.8 (CH2), 52.6 (CH3), 72.0
(C), 73.0 (C), 79.5 (C), 79.6 (C), 126.3 (CH), 127.2 (CH), 128.3 (CH),
128.4 (CH),128.7 (CH),129.7 (CH),130.4 (CH),130.6 (CH),130.7 (CH),
133.6 (CH), 133.9 (CH), 134.0 (CH), 138.2 (C), 138.7 (C), 138.8 (C),
O), 1459, 1380, 1266, 1113 cmꢀ1 1H NMR (500 MHz, CDCl3)
; d 0.38
(3H, s, Si(CH3)2), 0.44 (3H, s, Si(CH3)2), 0.96 (3H, d, J¼7.3, CHCH3),
1.12 (3H, d, J¼6.9, CH(CH3)2), 1.21 (3H, d, J¼6.8, CH(CH3)2), 1.99 (1H,
ddd, J¼13.7, 8.4, 6.4, CHSiMe2Ph), 2.32 (1H, septet, J¼6.8, CH(CH3)2),
2.69 (1H, app. quintet, J¼6.8, CHCH3), 3.73 (1H, app. t, J¼12.5,
NCH2), 3.79 (3H, s, OCH3), 3.88 (1H, dd, J¼11.4, 8.5, NCH2), 7.36e7.40
(3H, m, ArH), 7.49e7.51 (2H, m, ArH), 9.55 (1H, br s, NH); 13C NMR
13
153.4 (C]O), 174.3 (C]O); C NMR (270 MHz, DMSO-d6, 80 ꢁC)
d
ꢀ2.5 (Si(CH3)2), ꢀ2.0 (Si(CH3)2), 15.8, 28.5, 29.3, 37.1, 46.2, 49.0,
52.2, 72.7, 79.6, 126.3, 128.3, 128.5, 129.7, 130.5, 130.8, 133.7, 134.0,
138.4, 138.9, 153.9 (C]O), 174.4 (C]O); m/z (ESþ) 490 (100%,
MNaþ), 434 (74%, MNaþꢀC4H8), 390 (61%, MNaþꢀBoc); HRMS
C27H37NNaO4Si calcd 490.2384, found 490.2407. Compound 10c
(38 mg, 37%) was also isolated and had identical data to that
reported.12
(270 MHz, DMSO-d6)
d
ꢀ3.1 (Si(CH3)2), ꢀ2.3 (Si(CH3)2), 15.2
(CHCH3), 17.8 (CH(CH3)2), 18.4 (CH(CH3)2), 29.8 (CHSiMe2Ph), 32.8
(CH(CH3)2), 41.6 (CHCH3), 47.1 (NCH2), 52.7 (OCH3), 81.6 (NC), 128.4
(ArCH), 129.8 (ArCH), 133.6 (ArCH), 136.6 (ArC), 169.2 (C]O); m/z
(ESþ) 320 (100%, MHþꢀHCl); HRMS C18H30NO2Si: calcd 320.2046,
found 320.2036.
4.1.9. (2R
*
,3R
*
,4R )-2-Benzyl-4-(dimethylphenylsilanyl)-3-methyl-
*
pyrrolidine-2-carboxylic acid methyl ester (13c). Boc protected
proline 12c (48 mg, 0.10 mmol) was deprotected according to the
method for 9 except that 13b (37 mg, 98%) was isolated as a col-
ourless oil with no need for further purification. IR nmax (solution in
CHCl3) 3341 (NeH), 2952e2873 (CeH), 1727 (C]O), 1455, 1375,
4.1.12. Terminal alkene rearrangement precursor (16). To a stirred
suspension of KH (44 mg, 1.1 mmol) in THF (3 mL) at 0 ꢁC was
added
a solution of Boc-(L)-valine methyl ester (231 mg,
1.00 mmol) in THF (3 mL). The reaction was stirred for 1 h at 0 ꢁC,
then 1-bromo-2-(phenyldimethylsilyl)-prop-2-ene7d (306 mg,
1.20 mmol) in THF (2 mL) was added. The reaction was stirred for
a further 14 h, then quenched with saturated aq NH4Cl (5 mL). Et2O
(4 mL) was added and the layers separated. The aqueous layer was
re-extracted with Et2O (5 mL) and the combined organic layers
were washed with water (10 mL), brine (10 mL), dried (MgSO4) and
concentrated in vacuo. Purification by flash-column chromatogra-
phy (25% Et2O/pet. ether) yielded 16 (320 mg, 79%) as a colourless
1110, 1049 cmꢀ1; 1H NMR (500 MHz, CDCl3)
d 0.38 (3H, s, Si(CH3)2),
0.44 (3H, s, Si(CH3)2), 1.16 (3H, d, J¼7.2, CHCH3), 1.55 (1H, ddd,
J¼11.2, 9.9, 6.2, CHSiMe2Ph), 2.45 (1H, br s, NH), 2.77 (1H, app.
quintet, J¼6.7, CHCH3), 2.84 (1H, d, J¼12.8, CH2Ph), 3.12 (1H, d,
J¼12.8, CH2Ph), 3.20 (1H, dd, J¼11.2, 10.2, NCH2), 3.27 (1H, app. t,
J¼9.8, NCH2), 3.63 (3H, s, OCH3), 7.19e7.20 (2H, m, ArH), 7.24e7.32
(3H, m, ArH), 7.41e7.44 (3H, m, ArH), 7.54e7.57 (2H, m, ArH); 13C
NMR (500 MHz, CDCl3)
d
ꢀ3.0 (Si(CH3)2), ꢀ2.6 (Si(CH3)2), 14.0
oil; Rf¼0.37 (25% Et2O/pet. ether); [
a
]
20 ꢀ26.5 (c, 2.1, CHCl3); IR nmax
D
(CHCH3), 31.3 (CHSiMe2Ph), 41.8 (CH2Ph), 43.1 (CHCH3), 45.8
(NCH2), 52.1 (OCH3), 75.6 (NHC), 126.6 (ArCH), 127.9 (ArCH), 128.3
(ArCH), 129.1 (ArCH), 129.5 (ArCH), 133.8 (ArCH), 138.0 (ArC), 138.7
(ArC), 177.5 (C]O); m/z (ESþ) 390 (94%, MNaþ), 368 (47%, MHþ),
308 (48%); HRMS C22H29NNaO2Si calcd 390.1860, found 390.1845.
(solution in CHCl3) 2965e2876 (CeH), 1737 (C]O), 1683 (C]O),
1455, 1391, 1368, 1329, 1138, 1110 cmꢀ1; 1H NMR (500 MHz, CDCl3)
d
0.39 (3H, s, Si(CH3)2), 0.40 (3H, s, Si(CH3)2), 0.86 (3H, d, J¼6.8, CH
(CH3)2), 0.90 (3H, d, J¼6.5, CH(CH3)2), 1.40e1.48 (9H, m, C(CH3)3),
2.08e2.22 (1H, br m, CH(CH3)2), 3.59 (3H, s, OCH3), 3.71e4.04 (3H,
m, NCH2, NCH), 5.35e5.42 (1H, br m, C]CH2), 5.55e5.65 (1H, br m,
C]CH2), 7.34e7.37 (3H, m, ArH), 7.52e7.54 (2H, m, ArH); 13C NMR
4.1.10. (2R
*
,3R
*
,4R )-4-(Dimethylphenylsilanyl)-2-isopropyl-3-
*
methyl-pyrrolidine-1,2-dicarboxylic acid 1-tert-butyl ester 2-methyl
ester (12d). Cyclisation of 5d (1.68 g, 4.01 mmol) was achieved via
the conditions for 12b and after purification by column chroma-
tography (20% Et2O/pet. ether) gave 12d (1.34 g, 87%) as a colourless
solid (mp 72.1e73.3 ꢁC); Rf¼0.21 (20% Et2O/pet. ether); IR nmax
(solution in CHCl3) 2967e2878 (CeH), 1733 (C]O), 1682 (C]O),
(270 MHz, DMSO-d6)
d
ꢀ3.1 (Si(CH3)2), 18.4, 18.9, 19.5, 20.3, 28.0,
47.2 (CH2), 49.0 (CH2), 51.8 (CH), 62.9, 64.4, 79.4, 79.7, 121.8 (CH2),
124.8 (CH2), 128.0 (CH), 129.1 (CH), 133.5 (CH), 137.1 (C), 145.3 (C),
13
155.4 (C]O), 170.6 (C]O); C NMR (270 MHz, DMSO-d6, 80 ꢁC)
d
ꢀ3.9 (Si(CH3)2), ꢀ3.8 (Si(CH3)2), 18.1, 19.1, 27.4, 47.8, 50.7, 63.3,
78.8, 122.7, 127.2, 128.6, 133.1, 136.7, 144.7, 154.5 (C]O), 170.0 (C]
O); m/z (ESþ) 428 (100%, MNaþ), 372 (27%, MNaþꢀC4H8), 306 (16%,
MHþꢀBoc); HRMS C22H35NNaO4Si calcd 428.2228, found
428.2222; Anal. Calcd for C22H35NO4Si: C 65.15, H 8.70, N 3.45,
found C 65.12, H 8.78, N 3.21%.
1392, 1367, 1350, 1113 cmꢀ1
;
1H NMR (400 MHz, DMSO-d6)
d
0.30rot., 0.32rot., 0.37rot. (6H, s, Si(CH3)2), 0.79 (3H, d, J¼7.3, CH
(CH3)2), 0.82rot., 0.84rot. (3H, d, J¼6.8, CH(CH3)2), 1.12 (3H, app. t,
J¼6.7, CHCH3), 1.37rot., 1.46rot. (9H, s, C(CH3)3), 1.70e1.84 (1H, m,
CHSiMe2Ph), 2.63 (1H, app. sextet, J¼6.7, CH(CH3)2), 2.74 (1H, app.
septet, J¼6.9, CHCH3), 3.34e3.58 (5H, m, NCH2, OCH3), 7.38e7.40
(3H, m, ArH), 7.52e7.54 (2H, m, ArH); 13C NMR (270 MHz, DMSO-
4.1.13. Rearrangement of 16 to give (2R,S)-2-tert-Butox-
ycarbonylamino-4-(dimethylphenylsilanyl)-2-isopropylpent-4-enoic
d6)
d
ꢀ2.3 (Si(CH3)2), ꢀ2.1 (Si(CH3)2), ꢀ1.8 (Si(CH3)2), 17.4 (CH3), 17.5
acid methyl ester (17) and (2S
*
,4R )-4-(dimethylphenylsilanyl)-2-
*
(CH3), 19.8 (CH3), 20.3 (CH3), 28.3 (CH3), 28.5 (CH3), 30.5 (CH), 31.4
isopropylpyrrolidine-1,2-dicarboxylic acid 1-tert-butyl ester 2-methyl

J.C. Anderson, E.A. Davies / Tetrahedron 66 (2010) 6300e6308
6307
ester (18). Treatment of 16 under the cyclisation conditions for 12b
gave a crude product that was purified by column chromatography
(20% Et2O/pet. ether) to yield 17 (36 mg, 14%) as a colourless oil and
18 (170 mg, 67%) as a colourless oil. Data for 17 Rf¼0.45 (25% Et2O/
pet. ether); IR nmax (solution in CHCl3) 3421 (NeH), 2954 (CeH),
1715 (C]O), 1456, 1391, 1367, 1110, 1069 cmꢀ1; 1H NMR (400 MHz,
2.72rot., 2.79rot. (1H, septet, J¼6.9, CH(CH3)2), 3.22rot., 3.30rot. (1H,
dd, J¼11.8, 4.0, NCH2), 3.55e3.65 (4H, m, OCH3, NCH2), 3.95 (1H,
app. dt, J¼8.4, 4.5, CHOH), 4.23rot., 4.27rot. (1H, d, J¼7.9, CHOH); 13
C
NMR (400 MHz, C6D6) d 11.8 (CH3), 11.9 (CH3), 16.9 (CH3), 17.2 (CH3),
18.9 (CH3), 19.0 (CH3), 28.3 (CH3), 28.4 (CH3), 30.8 (CH), 32.6 (CH),
41.1 (CH), 42.3 (CH), 51.5 (CH3), 52.1 (CH3), 57.1 (CH2), 57.7 (CH2),
72.8 (C), 72.9 (CH), 73.4 (C), 73.7 (CH), 79.6 (C), 79.7 (C), 153.1 (C]
O), 154.1 (C]O), 174.5 (C]O), 175.5 (C]O); m/z (ESþ) 324 (100%,
MNaþ); HRMS C15H27NNaO5 calcd 324.1781, found 324.1773.
CDCl3)
d 0.35 (3H, s, Si(CH3)3), 0.38 (3H, s, Si(CH3)3), 0.84 (3H, d,
J¼6.9, CH(CH3)2), 0.88 (3H, d, J¼6.9, CH(CH3)2), 1.43 (9H, s, C(CH3)3),
2.44 (1H, septet, J¼6.9, CH(CH3)2), 2.74 (1H, d, J¼16.1, CH2), 3.40
(1H, d, J¼16.1, CH2), 3.58 (3H, s, OCH3), 5.47 (1H, s, C]CH2 or NH),
5.55 (1H, s, C]CH2 or NH), 5.75 (1H, s, C]CH2 or NH), 7.35e7.36
(3H, m, ArH), 7.50e7.52 (2H, m, ArH); 13C NMR (500 MHz, CDCl3)
4.1.16. (2R
acid 1-tert-butyl ester 2-methyl ester (21). To a stirred solution of
12d (26 mg, 62 mol) in DMF (0.5 mL) was added TBAF (0.31 mL of
*
,3S
*
)-2-Isopropyl-3-methylpyrrolidine-1,2-dicarboxylic
d
ꢀ2.9 (Si(CH3)2), ꢀ2.4 (Si(CH3)2), 17.7 (CH(CH3)2), 17.8 (CH(CH3)2),
m
28.5 (C(CH3)3), 34.8 (CH(CH3)2), 36.3 (CH2), 52.1 (OCH3), 66.3 (C),
78.9 (C), 127.7 (ArCH), 128.2 (H2C]C), 128.9 (ArCH), 134.1 (ArCH),
138.4 (C), 146.0 (C), 153.8 (C]O), 173.5 (C]O); m/z (ESþ) 428 (68%,
MNaþ), 372 (92%, MNaþꢀC4H8), 328 (100%, MNaþꢀBoc); HRMS
C22H35NNaO4Si calcd 428.2228, found 428.2222. Data for 18
Rf¼0.28 (25% Et2O/pet. ether); IR nmax (solution in CHCl3) 2968
(CeH),1710 (C]O),1494,1392,1367,1110 cmꢀ1; 1H NMR (500 MHz,
1 M solution in THF, 0.31 mmol). The reaction was heated to 80 ꢁC
and stirred for 2 h. The reaction was then allowed to cool to rt and
diluted with EtOAc (1 mL) and pet. ether (1 mL). The solution was
washed sequentially with 1 M aq HCl (1 mL), saturated aq KHCO3
(1 mL) and brine (1 mL). The organic layer was then dried (MgSO4)
and concentrated in vacuo. Purification by flash-column chroma-
tography (50% Et2O/pet. ether) gave 21 (13 mg, 74%) as a colourless
oil; Rf¼0.44 (50% Et2O/pet. ether); IR nmax (solution in CH2Cl2) 2876
CDCl3)
d
0.30 (6H, s, Si(CH3)2), 0.90 (3H, d, J¼6.3, CH(CH3)2), 1.11
(3H, d, J¼6.6, CH(CH3)2), 1.40rot., 1.43rot. (9H, s, C(CH3)3), 1.57e1.65
(1H, m, CHSiMe2Ph), 1.92 (1H, app. t, J¼13.3, CCH2), 2.19rot., 2.26rot.
(1H, dd, J¼12.8, 7.8, CCH2), 2.68rot., 2.78rot. (1H, septet, J¼6.8CH
(CH3)2), 3.35rot., 3.45rot. (1H, app. t, J¼11.7, NCH2), 3.60 (3H, s, OCH3),
3.67rot., 3.75rot. (1H, app t, J¼9.5, NCH2), 7.36e7.37 (3H, m, ArH),
(CeH), 1740 (C]O), 1686 (C]O), 1398, 1367, 1216, 1118 cmꢀ1
;
1H
NMR (500 MHz, CDCl3)
d
0.79rot., 0.81rot. (3H, d, J¼6.8, CH(CH3)2),
0.98rot., 0.99rot. (3H, d, J¼6.5, CHCH3), 1.14 (3H, d, J¼7.0, CH(CH3)2),
1.39rot., 1.42rot. (9H, s, C(CH3)3), 1.65e1.86 (2H, m, NCH2CH2),
2.30e2.41 (1H, m, CHCH3), 2.81rot., 2.95rot. (1H, septet, J¼6.9, CH
(CH3)2), 3.08 (1H, ddd, J¼11.9, 10.8, 5.8, NCH2), 3.66 (3H, s, OCH3),
3.73rot., 3.85rot. (1H, app. dd, J¼10.8, 8.3, NCH2); 13C NMR (270 MHz,
7.49e7.50 (2H, m, ArH); 13C NMR (500 MHz, CDCl3)
d
ꢀ5.0 (Si
(CH3)2), ꢀ1.7 (Si(CH3)2), 19.2 (CH3), 19.6 (CH3), 24.0 (CH), 24.9 (CH),
28.1 (CH3), 28.3 (CH3), 30.9 (CH), 31.4 (CH), 34.2 (CH2), 35.2 (CH2),
51.1 (CH2), 51.4 (CH2), 51.5 (CH3), 52.1 (CH3), 71.5 (C), 72.0 (C), 79.3
(C), 80.0 (C), 127.8 (ArCH), 129.2 (ArCH), 133.6 (ArCH), 136.7 (ArC),
153.9 (C]O), 175.0 (C]O); m/z (ESþ) 428 (100%, MNaþ); HRMS
C22H35NNaO4Si calcd 428.2228, found 428.2225.
DMSO-d6)
d 17.0 (CH3), 17.1 (CH3), 17.2 (CH3), 17.4 (CH3), 18.9 (CH3),
19.1 (CH3), 28.3 (CH3), 28.5 (CH3), 31.0 (CH), 32.1 (CH2), 32.3 (CH2),
32.7 (CH2), 37.1 (CH), 38.2 (CH), 47.5 (CH2), 51.3 (CH3), 72.6 (C), 72.8
(C), 79.3 (C), 79.9 (C), 153.4 (C]O), 153.6 (C]O), 173.5 (C]O), 173.6
(C]O); 13C NMR (270 MHz, DMSO-d6, 80 ꢁC)
d 17.3, 17.5, 19.3, 28.6,
32.2, 32.4, 37.0, 38.0, 47.5, 51.4, 73.0, 79.3, 152.8 (C]O), 173.3 (C]
O); m/z (ESþ) 308 (100%, MNaþ); HRMS C15H27NNaO4 calcd
308.1832, found 308.1833.
4.1.14. (2S
*
,4R )-4-(Dimethylphenylsilanyl)-2-isopropylpyrrolidine-
*
1,2-dicarboxylic acid 2-methyl ester (19). Boc protected proline 18
(99 mg, 0.24 mmol) was deprotected according to the method for 9
except that 19 (73 mg, 96%) was isolated as a colourless oil with no
need for further purification. IR nmax (solution in CH2Cl2) 3020
Acknowledgements
(NeH), 2925 (CeH), 1722 (C]O), 1216, 845 cmꢀ1
;
1H NMR
We thank the University of Nottingham for funding, Prof A.J.
Blake for single crystal X-ray structure determination, Mr. T. Hol-
lingworth and Mr. D. Hooper for providing mass spectra and Dr. T.
Liu for microanalytical data.
(500 MHz, CDCl3)
d
0.30 (6H, s, Si(CH3)2), 0.87 (3H, d, J¼6.7, CH
(CH3)2), 0.94 (3H, d, J¼6.8, CH(CH3)2), 1.52e1.69 (2H, m, CHSiMe2Ph,
NH), 1.90e2.03 (3H, m, CCH2, CH(CH3)2), 2.69 (1H, dd, J¼12.5, 9.3,
NCH2), 3.07 (1H, dd, J¼9.2, 6.9, NCH2), 3.71 (3H, s, OCH3), 7.32e7.42
(3H, m, ArH), 7.53e7.54 (2H, m, ArH); 13C NMR (500 MHz, CDCl3)
References and notes
d
ꢀ2.8 ((SiCH3)2), ꢀ2.5 (Si(CH3)2), 18.0 (CH3), 25.7 (CH), 31.2 (CH),
1. Anderson, J. C.; Siddons, D. C.; Smith, S. C.; Swarbrick, M. E. J. Chem. Soc., Chem.
Commun. 1995, 1835.
35.5 (CH2), 53.2 (CH2), 62.1 (CH3), 71.4 (C), 127.7 (ArCH), 128.7
(ArCH), 133.9 (ArCH), 138.2 (ArC), 173.9 (C]O); m/z (ESþ) 306
(100%, MHþ); HRMS C17H28NO2Si calcd 306.1884, found 306.1885.
2. Durst, T.; Van den Elzen, R.; Lebelle, M. J. J. Am. Chem. Soc. 1972, 94, 9261.
3. (a) Åhman, J.; Somfai, P. J. Am. Chem. Soc. 1994, 116, 9781; (b) Åhman, J.; Somfai,
P. Tetrahedron Lett. 1995, 36, 303; (c) Åhman, J.; Somfai, P. Tetrahedron 1995, 51,
9747; (d) Åhman, J.; Somfai, P. Tetrahedron Lett. 1996, 37, 2495; (e) Åhman, J.;
Jarevång, T.; Somfai, P. J. Org. Chem. 1996, 61, 8148; (f) Coldham, I.; Collis, A. J.;
Mould, R. J.; Rathmell, R. E. J. Chem. Soc., Perkin Trans. 11995, 2739; (g) Coldham,
I.; Collis, A. J.; Mould, R. J.; Rathmell, R. E. Tetrahedron Lett. 1995, 36, 3557.
4. Nakai, T.; Mikami, K. Org. React. 1994, 46, 105.
5. (a) Broka, C. A.; Shen, T. J. Am. Chem. Soc. 1989, 111, 2981; (b) Coldham, I. J. Chem.
Soc., Perkin Trans. 1 1993, 1275; (c) Gawley, R. E.; Zhang, Q.; Campagna, S. J. Am.
Chem. Soc. 1995, 117, 11817; (d) Tomoyasu, T.; Tomooka, K.; Nakai, T. Tetrahedron
Lett. 2003, 44, 1239; (e) Coldham, I.; Middleton, M. L.; Taylor, P. L. J. Chem. Soc.,
Perkin Trans. 1 1998, 2817.
6. Wu, Y.-D.; Houk, K. N.; Marshall, J. A. J. Org. Chem. 1990, 55, 1421.
7. (a) Anderson, J. C.; Siddons, D. C.; Smith, S. C.; Swarbrick, M. E. J. Org. Chem.
1996, 61, 4820; (b) Anderson, J. C.; Smith, S. C.; Swarbrick, M. E. J. Chem. Soc.,
Perkin Trans. 1 1997, 1517; (c) Anderson, J. C.; Dupau, P.; Siddons, D. C.; Smith, S.
C.; Swarbrick, M. E. Tetrahedron Lett. 1998, 39, 2649; (d) Anderson, J. C.; Flaherty,
A.; Swarbrick, M. E. J. Org. Chem. 2000, 65, 9152.
4.1.15. (2R
*
,3R
*
,4R )-4-Hydroxy-2-isopropyl-3-methylpyrrolidine-
*
1,2-dicarboxylic acid 1-tert-butyl ester 2-methyl ester (20). To
a stirred solution of 12d (52 mg, 0.12 mmol) in peracetic acid so-
lution (0.13 mL of 36e40 wt % solution in AcOH, 0.62 mmol) was
added mercury (II) acetate (47 mg, 0.15 mmol). The reaction was
stirred for 4 h then Et2O (1 mL) was added. The solution was
washed with saturated aq sodium thiosulfate (1 mL), water (1 mL),
saturated aq NaHCO3 (1 mL) and brine (1 mL). The organic layer
was then dried (MgSO4) and concentrated in vacuo. The crude
product was purified by flash-column chromatography (40% Et2O/
pet. ether) to yield 20 (25 mg, 67%) as a colourless oil; Rf¼0.38 (40%
Et2O/pet. ether); IR nmax (solution in CH2Cl2) 3445 (OeH), 2876
(CeH), 1752 (C]O), 1690 (C]O), 1398, 1368, 1139, 1110, 1076 cmꢀ1
1H NMR (270 MHz, DMSO-d6)
(CH3)2), 0.91 (3H, d, J¼7.2, CHCH3), 1.09rot., 1.10rot. (3H, d, J¼7.0, CH
(CH3)2), 1.33rot., 1.39rot. (9H, s, C(CH3)3), 2.35e2.47 (1H, m, CHCH3),
;
8. (a) Anderson, J. C.; Flaherty, A. J. Chem. Soc., Perkin Trans. 1 2001, 267; (b) An-
derson, J. C.; Whiting, M. J. Org. Chem. 2003, 68, 6160.
9. Anderson, J. C.; O’Loughlin, J. M. A.; Tornos, J. A. Org. Biomol. Chem. 2005, 3,
2741.
d
0.73rot., 0.74rot. (3H, d, J¼7.0, CH
10. Anderson, J. C.; Ford, J. G.; Whiting, M. Org. Biomol. Chem. 2005, 3, 3734.

6308
J.C. Anderson, E.A. Davies / Tetrahedron 66 (2010) 6300e6308
11. (a) Kawabata, T.; Wirth, T.; Yahiro, K.; Suzuki, H.; Fuji, K. J. Am. Chem. Soc.1994, 116,
10809; (b) Fuji, K.; Kawabata, T. Chem.dEur. J.1998, 4, 373; (c) Kawabata, T.; Suzuki,
H.; Nagae, Y.; Fuji, K. Angew. Chem., Int. Ed. 2000, 39, 2155; (d)Kawabata, T.;Chen, J.;
Suzuki, H.; Nagae, Y.; Kinoshita, T.; Chancharunee, S.; Fuji, K. Org. Lett. 2000, 2,
3883;(e)Kawabata, T.;Matsuda, S.;Kawakami, S. J.Am.Chem. Soc. 2003,125,13012;
(f) Kawabata, T.; Chen, J.; Suzuki, H.; Fuji, K. Synthesis 2005,1368; (g) Kawabata, T.;
Majumdar, S.; Tsubaki, K.; Monguchi, D. Org. Biomol. Chem. 2005, 3, 1609.
12. Anderson, J. C.; Skerratt, S. J. Chem. Soc., Perkin Trans. 1 2002, 2871.
13. Crystallographic data (excluding structure factors) for 11c and 12d have been
deposited with the Cambridge Crystallographic Data Centre as supplementary
publication nos. CCDC 768583 and CCDC 768584 (awaiting), respectively.
Copies of the data can be obtained, free of charge, on application to CCDC, 12
Union Road, Cambridge CB2 1EZ, UK, (fax: þ44 (0)1223 336,033 or e-mail:
deposit@ccdc.cam.ac.Uk).
14. Chan, T. H.; Mychajlowskij, W.; Ong, B. S.; Harpp, D. N. J. Org. Chem. 1978, 43,
1526.
15. (a) Beesley, R. M.; Ingold, C. K.; Thorpe, J. F. J. Chem. Soc. 1915, 107, 1080; (b)
Bachrach, S. M. J. Org. Chem. 2008, 73, 2466 and references there in.
16. (a) Baldwin, J. E. J. Chem. Soc., Chem. Commun. 1976, 734; (b) Baldwin, J. E.;
Cutting, J.; Dupont, W.; Kruse, L.; Silberman, L.; Thomas, R. C. J. Chem. Soc.,
Chem. Commun. 1976, 736.
17. Chatgilialoglu, G.; Ferreri, C.; Guerra, M.; Timokhin, V.; Froudakis, G.; Gimisis, T.
J. Am. Chem. Soc. 2002, 124, 10765.
18. Padwa, A.; Norman, B. H. J. Org. Chem. 1990, 53, 4801.
19. Berry, M. B.; Craig, D.; Jones, P. S.; Rowlands, G. J. Chem. Commun. 1997, 2141.
20. Nadano, R.; Iwai, Y.; Mori, T.; Ichikawa, J. J. Org. Chem. 2006, 71, 8748.
21. Heitzman, C. L.; Lambert, W. T.; Mertz, E.; Shotwell, J. B.; Tinsley, J. M.; Va, P.;
Roush, W. R. Org. Lett. 2005, 7, 2405.