Chirality transfer in the aza-[2,3]-Wittig sigmatropic rearrangement

View Article Online / Journal Homepage / Table of Contents for this issue

A R T I C L E

Chirality transfer in the aza-[2,3]-Wittig sigmatropic rearrangement†

James C. Anderson,*^aJ. Gair Ford^band Matthew Whiting^a

^aSchool of Chemistry, University of Nottingham, Nottingham, UK NG7 2RD.

E-mail: j.anderson@nottingham.ac.uk; Fax: +44(0)115 9513564; Tel: +44(0)115 9514194

^bAstraZeneca Process R & D, Charter Way Silk Road Business Park, Macclesﬁeld, Cheshire,

UK SK10 2NA. E-mail: Gair.Ford@astrazeneca.com; Tel: +44(0)1625 513375

Received 29th July 2005, Accepted 22nd August 2005

First published as an Advance Article on the web 12th September 2005

The aza-[2,3]-Wittig sigmatropic rearrangements of substrates derived from enantiomerically pure alanine, valine and

serine with phenyl and ester anion stabilising groups were investigated for their efﬁciency in chirality transfer. It was

found that a methyl substituent at the stereogenic centre of the rearrangement precursors was inadequate to control

the alkene stereoselectivity and enantioselectivity of the rearrangement. Ester stabilised anions of valine and serine

derivatives were the most successful with up to 66% yield, 14 : 1 alkene (E)-stereoselection and 88% chirality transfer.

A limitation to the steric bulk of the stereogenic centre was noted in that the substituent has to be bulky enough to

dictate alkene stereoselection, but not too large to compromise the directing effect of the activating phenyldimethyl

silyl substituent on the anion stabilising group. Experimental evidence suggested a possible complimentary

coordinating effect of an O–MOM serine substituent, which may assist alkene stereoselectivity and enantioselectivity.

Introduction

We have detailed racemic studies into the aza-[2,3]-Wittig

sigmatropic rearrangement which have shown it to be a useful

method for the synthesis of certain unnatural amino acids¹

and that these can be used as building blocks in target

synthesis.²More recently we have summarised some of our

attempted strategies towards preparing enantiomerically pure

aza-[2,3]-rearrangement products and detailed the use of (−)-8

(2)

Results and discussion

phenylmenthol as a chiral auxiliary.³An (−)-8-phenylmenthol

ester of a glycine derived migrating group controlled the absolute

stereochemistry of aza-[2,3]-Wittig sigmatropic rearrangements

with diastereoselectivities of ∼3 : 1 with respect to the auxiliary

and ca. 50% isolated yields of enantiomerically pure products.

This chiral auxiliary approach was only moderately successful

and we have been investigating alternative strategies to control

stereoselectivity. This paper describes in detail our attempts

to use internal chirality transfer to control the absolute stere-

ochemistry of the aza-[2,3]-Wittig sigmatropic rearrangement

(eqn. (1)).

Our preliminary study investigated the rearrangement of aza-

[2,3]-Wittig precursors 1. These precursors allowed us to inves-

tigate the steric properties of the substituent at the chiral C1

centre by comparing a methyl substituent with an isopropyl

substituent. A terminal alkene was chosen so as to eliminate

analysis of vicinal diastereoselectivity in the product and be-

cause differentiation between potentially competing [1,2] and

the desired [2,3] rearrangement would be monitored by the

C1 substituent. The phenyl group was chosen as the anion

stabilising group as this was the most robust group available.

Retrosynthesis of 1 led to vinyl silane 2 which could be derived

by methylenation of acyl silane 3, itself derived from N-Boc

protected amino acid 4 (Scheme 1).

(1)

For the related oxy-[2,3]-Wittig rearrangement there are

three main ways by which absolute stereochemistry has been

controlled; asymmetric induction from a chiral auxiliary,⁴enan-

tioselective rearrangements⁵and internal chirality transfer.^6,7

The closest examples to the strategy described in this current

paper were in work carried out on 1,4 chirality transfer (eqn. (2)),

where they observed 81–98% chirality transfer with 93–98%

E selectivity, dependent upon the choice of anion stabilising

group.⁸As far as we are aware there have been no studies

investigating internal chirality transfer in the aza-[2,3]-Wittig

sigmatropic rearrangement.

Scheme 1 Retrosynthesis.

The synthesis of acyl silanes has received considerable

attention⁹and it has been shown that these can be prepared

from N-Pht and N-Ts a-amino acid chlorides on treatment

with a silylcuprate reagent without loss of enantiomeric purity.¹⁰

However under the same reaction conditions, N-Boc pro-

tected amino acid chlorides have been shown to form the

corresponding oxazoline through spontaneous intramolecular

cyclisation.¹¹This has been avoided in the synthesis of N-

Boc-phenylalanine and N-Boc-isoleucine based acyl silanes

by activation of the amino acids as their acyl imidazoles

followed by reaction with (Me₂PhSi)₂CuCNLi₂.¹²To adopt

this approach we treated N-Boc alanine and N-Boc valine

† This paper is dedicated to my friend and mentor Professor

Steven V. Ley on the occasion of his 60^thbirthday.

3 7 3 4

O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 3 7 3 4 – 3 7 4 8

T h i s j o u r n a l i s

T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5

©

View Article Online

Table 1 Aza-[2,3]-Wittig sigmatropic rearrangement of (S)-6b

with CDI (carbonyldiimidazole) to give the corresponding

acylimidazoles (S)-5a,b (Scheme 2). The crude products were

then treated with (Me₂PhSi)₂CuCNLi₂to give the acyl silanes

Temp/^◦C

Time/mins

Yield (%)

E : Z^a

Ee (E) (%)^b

13

(S)-3a,b in 48% and 76% yields respectively. In the case of

the lower yielding and sterically less demanding substrate (5a)

a number of byproducts were observed which, although they

could never be fully separated from themselves, appeared by

characteristic peaks in their ¹H NMR spectra to be products of

di-addition of the silyl nucleophile. Although the oleﬁnation

of acyl silanes has been investigated using Wittig^14,15and

Peterson oleﬁnation¹⁶methods, a-branching, enolisation and

competing Brook rearrangement severely reduced yields. We

could ﬁnd no precedent for the oleﬁnation of an acyl silane

containing an enantiomerically pure a-chiral centre. We decided

to use the non-basic Tebbe reagent¹⁷which had been shown

to be suitable for the oleﬁnation of numerous carboxylic acid

derivatives, including those containing readily enolisable a-

protons.¹⁸Careful optimisation of the reaction conditions led

to the use of 1.5 equiv. of Tebbe reagent in THF at −40 ^◦C for

1–1.5 h only. This led to reproducible yields of vinyl silanes (S)-

2a,b in reproducible isolated yields of 64% and 50% respectively

in addition to recovered starting materials. The lower yield of

the valine derived compound we attribute to increased steric

bulk at the a-position hindering attack of the bulky titanium

reagent. Prolonged reaction temperatures led to degradation of

material probably caused by the highly Lewis acidic Me₂AlCl

liberated during the Tebbe oleﬁnation. The use of the much

milder Petasis reagent,¹⁹which has been used for the oleﬁnation

of an acyl silane,²⁰unfortunately gave complete degradation of

material. Benzylation of the amines (S)-2a,b required the use

of KH to effect efﬁcient deprotonation which then gave (S)-

6a,b in 84% and 88% yield respectively. The synthetic sequence

(Scheme 2) was then repeated on the enantiomeric amino acid

series and chiral HPLC analysis conﬁrmed that no racemisation

had occurred in the synthesis.²¹

−78

−20

0

10

5

48

100^c

74

1 : 1

2 : 1

3 : 1

4 : 1

—

58

56

rt

5

63

^aDetermined by ¹H NMR. ^bDetermined by chiral HPLC.^{22 c}Conversion

by ¹H NMR.

rearrangement of the sterically more demanding valine derived

aza-[2,3]-Wittig rearrangement precursor (S)-6b.

(3)

Complete conversion of (S)-6b required 1.5 equiv. of n-BuLi,

as in the rearrangement of the alanine derived precursor (S)-6a.

Higher yields were obtained at near ambient reaction tempera-

tures, but a non intuitive temperature dependence of the alkene

stereoselection was noted (Table 1). The best diastereoselection

was achieved by rearrangement at room temperature (4 : 1 E :

Z) in comparison to −78 ^◦C (1 : 1 E : Z). Differentiation

of the alkene stereoisomers was again inferred from selected

one dimensional NOE experiments. The enantiomeric excess of

(E)-7b was determined to be 58% from the rearrangement at

0 ^◦C and 56% from the rearrangement at room temperature by

chiral HPLC analysis.²²The absolute stereochemistry was not

determined due to the moderate enantioselectivity.

(4)

The stereoselection from the rearrangements of (S)-6a,b

(eqns. (3,4)) can be rationalised by analysing the proposed

transition states for this pericyclic reaction (Fig. 1). Assuming

a ﬁve membered envelope type transition state then our chiral

rearrangement precursors [(S)-6a,b] can undergo rearrangement

through one of four possible transition states to either enan-

tiomer of both the E and Z alkene product. We expected, in

accord with a literature example in the oxy-[2,3]-Wittig series,²³

that the 1,2-interaction between R and SiMe₂Ph in TS3 and TS4

would favour TS1 and TS2 and hence the E-alkene geometry.

The degree of chirality transfer would be determined by whether

the anion stabilising phenyl substituent chose to adopt an endo

(TS2, TS3) or exo (TS1, TS4) orientation in the transition

state. Previous work of ours with achiral substrates has shown

that the avoidance of the pseudo-1,3-diaxial steric interaction

between Ph and SiMe₂Ph, such as that present in TS2 and TS3,

has dominated stereoselection.^1aIn the present chiral systems,

TS1 and TS3 also possess destabilising pseudo-1,3-diaxial steric

interactions between R and the anion stabilising group Ph. An

additional complicating factor that separates this methodology

from the oxy-[2,3]-Wittig rearrangement technology is the

conﬁgurational stability of benzylic anions adjacent to a N-Boc

dipole. Deprotonation of (S)-6a,b at temperatures below about

−40 ^◦C,²⁴assuming that the remote stereocentre at C1 does not

impart complete stereocontrol, would give two diastereomeric

Scheme

2

Reagents and conditions: i, CDI, THF, 0

^◦C; ii,

(Me₂PhSi)₂CuCNLi₂, THF, −78 ^◦C; iii, Cp₂Ti(CH₂)ClAlMe₂, THF,

−40 ^◦C; iv, KH, THF, rt, 30 mins, then BnBr, TBAI, rt, 16 h.

Aza-[2,3]-Wittig sigmatropic rearrangement of the alanine

derived precursor (S)-6a was achieved by using 1.5 equiv. n-

BuLi in THF–HMPA (4 : 1) at −78 C with quenching of the

◦

reaction after disappearance of starting material (TLC, 15 mins),

with MeOH. This gave the rearranged product 7a as a 1.8 : 1

mixture of inseparable E : Z diastereoisomers in 56% isolated

yield (eqn. (3)). Lower quantities of n-BuLi gave incomplete

conversion and higher reaction temperatures (−20 and 0 ^◦C)

gave rearrangement in less than 5 mins, but with a 1 : 1 mixture of

alkene diastereoisomers. The E and Z alkenes were distinguished

by selected one dimensional NOE experiments. Irradiation of the

vinylic proton of (E)-7a (d 6.07, 1H, q, J = 6.5 Hz) gave a 4.7%

enhancement of the SiMe₂signals (d 0.35 and 0.37, 2 × 3H, s),

whereas irradiation of the vinylic proton of (Z)-7a (d 6.20, q,

J = 6.8 Hz) gave a 6.1% enhancement of one of the CH₂protons

(d 2.18–2.23, 1H, m) and a 2.7% enhancement of the benzylic

CH (d 4.52–4.75, 2H, m, NH + PhCH). At this preliminary

stage we decided there was little to be gained from looking

into the enantioselectivity of this reaction (eqn. (3)) as the

alkene stereoselection was so poor. Instead we investigated the

O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 3 7 3 4 – 3 7 4 8

3 7 3 5

View Article Online

Fig. 1 Proposed possible transition state models for the rearrangement of (S)-6.

organolithium species (S,S) and (S,R)-8 (eqn. (5)). Accepting

that aza-[2,3]-Wittig sigmatropic rearrangements are stereospe-

ciﬁc, proceeding with complete inversion of conﬁguration at

the lithium-bearing carbanion centre,^25,26two extreme scenarios

are then possible. If rearrangement occurs much faster than

racemisation of the lithio carbanion each diastereoisomer will

rearrange through separate pairs of transition states that will

lead to enantiomeric pairs of products containing either oleﬁn

geometry. That is (S,S)-8 through TS2 or TS4 to give (R,E) or

(R,Z)-7 and (S,R)-8 through TS1 or TS3 to give (S,E) or (S,Z)-

7. For each product enantiomer the E : Z ratio will depend

on the relative destabilising steric interactions in the pairs of

transition states. Alternatively if racemisation is much faster

than rearrangement then the reaction will proceed through the

lowest energy transition state of TS1-4.

We have already shown that achiral methyl ester aza-[2,3]-

rearrangement precursors rearrange with good diastereo-

control.^1bWe hoped to access the desired precursors (S)-9a,b

directly from (S)-2a,b by alkylation with a suitable electrophile.

Direct alkylation of the potassium anion of (S)-2a,b with methyl

bromoacetate or its iodo analogue proved unsuccessful under

a variety of conditions. The failure of this reaction was most

probably due to deprotonation of the electrophile by the aza

anion and led us to investigate the sodium salt of bromo-

and iodoacetic acid in an attempt to reduce the acidity of

the methylene protons, but both were unsuccessful. A more

circuitous route involved removal of the Boc group from (S)-

2a,b to give the amine (S)-10a,b in good yield (Scheme 3).

Mono alkylation with methyl bromoacetate by a literature

procedure,²⁷with the essential modiﬁcation of performing the

reaction at 0 ^◦C, gave (S)-11a, and (S)-11b in 81% and 77% yields

respectively. Reprotection with Boc₂O proved very difﬁcult, but

under the forcing conditions of Kabalka et al.²⁸we were able to

synthesise (S)-9a in 95% yield. Surprisingly even under these and

even harsher conditions,²⁹no product formation was observed

for the valine derived secondary amine (S)-10b. We therefore

attempted to alkylate (S)-2b with commercially available 2-

bromoethoxy-t-butyldimethylsilane as a two carbon electrophile

that contained no acidic protons and appropriate functionality

for subsequent elaboration to the required ester. Alkylation

required considerable optimisation starting from the conditions

we had used to benzylate (S)-2b. Guided by deuterium quench

studies with D₂O we found complete deprotonation of (S)-

2b with KH in DMF required 80 mins at 0 ^◦C. Addition of

an electrophile and catalytic TBAI (10 mol%) and stirring for

30 mins at 0 ^◦C then no more than 1 h at room temperature led to

a reproducible 63% of (S)-12b. Selective removal of the silyloxy

protecting group was achieved using TBAF (92%) and the free

alcohol oxidised to the corresponding acid using Jones’ reagent

followed by immediate esteriﬁcation with diazomethane to give

(S)-9b in 57% yield over two steps (Scheme 3). The enantiomers

(R)-9a,b were synthesised in an identical fashion and HPLC

analysis conﬁrmed greater than 95% enantiomeric purity.³⁰

Rearrangement precursor (S)-9a was treated with KH/18-C-

6 or LDA to initiate aza-[2,3]-rearrangement as we knew from

previous studies that the ester function was incompatible with

n-BuLi. Optimisation of temperature and equivalents of base

led to a 65% yield of desired product 13a, but no improvement

in the alkene stereoselectivity of ∼1 : 1 by ¹H NMR (eqn. (6)).

(5)

The low alkene stereoselection observed in the rearrangement

of alanine derived precursor (S)-6a (eqn. (3)) indicates that the

methyl group is not large enough to differentiate efﬁciently

between the pairs of transition states TS1/2 and TS3/4.

Rearrangement of the valine derived precursor (S)-6b (eqn.

(4)) gave good alkene stereoselection only at near ambient

temperatures (Table 1). At these temperatures (speciﬁcally 0 ^◦C

and rt) racemisation of the lithio carbanion will be rapid and the

alkene stereoselection reﬂects the effectiveness of the isopropyl

group to favour the pair of transition states TS1/2. The 56–

58% enantiomeric excess indicates the similar transition state

energies of TS1 and TS2. In order to improve the stereoselection

of the process we decided to investigate planar carbanions

generated by using an ester as the anion stabilising group.

We assumed that the complications arising because of the

conﬁguration of the lithio carbanion at low temperatures would

be circumvented, such that stereoselection would be dictated

by access to any of the four possible transition states at any

temperature.

3 7 3 6

O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 3 7 3 4 – 3 7 4 8

View Article Online

Table 2 Aza-[2,3]-Wittig sigmatropic rearrangement of (S)-9b

Base

Temp/^◦C

Time/h Yield (%)^aE : Z^b

Ee (E) (%)^c

KH

0

−20

−40

−20

1

2

16

1.5

6

66

50

12

56

10 : 1

5 : 1

70

79

66

50

—

LDA

KHMDS −40

^aIsolated by ﬂash chromatography, yield refers to inseparable (E/Z)-13b

1

mixture. ^bDetermined by H NMR. ^cEe of major isomer, determined

by chiral HPLC. ^dReaction in the presence of 20% HMPA instead of

18-C-6.

Scheme 3 Reagents and conditions: i, 4 M HCl in dioxane, rt; ii,

BrCH₂CO₂Me, Et₃N, DMF, 0 ^◦C; iii, Boc₂O, Et₃N, 89 ^◦C; iv, KH,

DMF, 0 ^◦C, 80 mins; BrCH₂CH₂OTBDMS, TBAI, 0 ^◦C, 30 mins then

rt 1 h; v, TBAF, THF, rt, 92%; vi, Jones’ reagent, acetone, 0 ^◦C; CH₂N₂,

Et₂O, rt 62%.

Differentiation of the alkene stereoisomers was again inferred

from selected one dimensional NOE experiments.

Scheme 4 Reagents and conditions: i, O₃, MeOH–CH₂Cl₂, −78 ^◦C;

H₂O, rt, 14 h; ii, NaIO₄, MeOH–H₂O, rt; CH₂N₂, Et₂O, rt.

The large and positive value of the observed [a]_Dand the

literature value show that the major enantiomer of (E)-13b is

the S-enantiomer. The optical purity calculated by compar-

ison of the [a]_Dvalues was in good agreement with the ee

previously determined by chiral HPLC (67% op cf 70% ee).

From the determination of the absolute stereochemistry of the

major rearrangement product (E,S)-13b we can infer that the

rearrangement prefers to pass through transition state structure

TS5 (Fig. 2). In this structure (TS5) the bulky silicon group is

dictating all stereochemistry by forcing both the methyl ester

and isopropyl substituent to be exo in spite of the unfavourable

1,3-interaction between them. This controls the rearrangement

(6)

Investigation into the use of freshly prepared³¹and titrated³²

KHMDS with 18-C-6 at −40 ^◦C for 6 h did lead to a 1 :

7 E : Z stereoselectivity, but only in a poor 25% yield. This

alkene stereoselection is far greater than when using KH, even

though it is the same counter ion. Re-subjecting the product

(1 : 7 E : Z) to the reaction conditions veriﬁed that there was

no selective destruction of one diastereoisomer. The origin and

sense of this increased selectivity remains unclear. The degree

of chirality transfer in these reactions was not measured due to

either the poor alkene stereoselection and/or the poor yields

obtained. This substrate [(S)-9a] was behaving similarly to the

corresponding benzyl substrate [(S)-6a] and we could again

conclude that the stereogenic methyl substituent was, in the

majority of reactions, not of sufﬁcient steric bulk to energetically

differentiate diastereomeric transition states.

◦

at 0 C in 66% yield with a chirality transfer of 85% (70% ee)

and 91% E alkene selectivity.

Rearrangement of valine precursor (S)-9b using the KH/18-

C-6 conditions led to the best yield and alkene stereoselection

(eqn. (7), Table 2).

Fig. 2 Favoured transition state structure.

With the successful rearrangement of our valine based models

occurring with good transfer of chirality we moved on to look at

a more functionalised precursor containing an a-hydroxymethyl

substituent that could be derived from serine. We planned that

the judicious choice of hydroxyl protecting group would enable

the protected a-hydroxymethyl substituent to mimic the effect

of the isopropyl group through steric bulk or in some similar

way by coordination. The synthesis of serine derived precursor

16 required the use of the bulky TBDPS protecting group

for the a-hydroxymethyl substituent and followed the same

synthetic sequence as for the synthesis of the alanine precursor

9a (Scheme 5).³⁷The enantiomer (R)-16 was synthesised in an

identical fashion and HPLC analysis conﬁrmed greater than

95% enantiomeric purity.³⁸

Optimisation of the [2,3]-rearrangement of (S)-16 found the

use of an excess of LDA (1.8 equiv.) in THF–HMPA (4 : 1)

at 0 ^◦C for 10 mins gave 17 in 14 : 1 E : Z stereoselectivity

in 66% yield (Scheme 6). Use of KH (2 equiv.)/18-C-6 led to

complete E-alkene stereoselection, but in a lower 36% yield. Pro-

longed reaction times or lower temperatures led to degradation

(7)

The use of LDA or KHMDS gave inferior results. The (E)-alkene

was veriﬁed from selected one dimensional NOE experiments.

The highest enantiomeric excess of the major product (E)-

13b was determined to be 79% (50% yield) by chiral HPLC.³³

Determination of the absolute stereochemistry of (E)-13b was

achieved by degradation to a known aspartic acid derivative.

Treatment of (E)-13b (70% ee) with ozone and stirring with water

overnight gave the ketol 14 in 66% yield (Scheme 4).³⁴Cleavage

with NaIO₄³⁵and esteriﬁcation with diazomethane gave N-Boc-

L-aspartic acid dimethyl ester (15) in 47% yield, [a]_D+20.5

(c 0.4, CHCl₃), [lit.³⁶[a]_D+30.8 (c 2.1, CHCl₃)].

O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 3 7 3 4 – 3 7 4 8

3 7 3 7

View Article Online

(14 : 1 E : Z, 40% ee) compared to the valine analogue (E,S)-13b

(10 : 1 E : Z, 79% ee). Ignoring any subtle electronic effects and

assuming the OTBDPS group is incapable of coordination to

the metal counter ion, it would appear that there is an interplay

between the steric bulk of the substituent at the stereogenic

centre being large enough to dictate alkene stereoselection, but

not so large as to counter too severely the directing effect

of the activating phenyldimethylsilyl group, towards the anion

stabilising group (CO₂Me in this case).

We envisaged two possible solutions to this problem (Fig. 4).

Changing the protecting group of the pendant CH₂OP group

from a bulky silane to a more activated group, e.g. the MOM

group, could enable a positive chelating interaction with the

cation associated with the enolate that could only take place

when both groups were exo (TS8). Extrapolating this idea to its

extreme convinced us that if the hydroxymethyl group and the

ester group could be covalently linked to form a lactone then if

rearrangement could occur, the constraints of the system would

allow only one transition state structure to be accessed (TS9).

Scheme 5 Reagents and conditions: i, TBDPSCl, imidazole, DMF,

rt; K₂CO₃, MeOH, rt, 94%; ii, CDI, THF,

0

^◦C, 94%; iii,

(Me₂PhSi)₂CuCNLi₂, THF, −80 ^◦C, 25%; iv, Cp₂Ti(CH₂)ClAlMe₂,

THF, −40 ^◦C, 35%; v, TFA–CH₂Cl₂, 0 C, 81%; vi, BrCH₂CO₂Me, Et₃N,

DMF, 0 ^◦C, 80%; vii, Boc₂O, Et₃N, 89 ^◦C, 93%.

Scheme 6 Reagents and conditions: i, 1.8 equiv. LDA, THF–HMPA

(4 : 1), 0 ^◦C, 10 mins; ii, O₃, MeOH–CH₂Cl₂, −78 ^◦C; H₂O, rt, 14 h,

64%; ii, NaIO₄, MeOH–H₂O, rt; CH₂N₂, Et₂O, rt, 60%.

of product. The (E)-alkene was veriﬁed from selected one

dimensional NOE experiments (Scheme 6). The enantiomeric

excess of (E)-17 was determined to be 40% for the material

from the LDA rearrangement and 55% for the material from the

KH rearrangement by chiral HPLC analysis.³⁹Determination of

the absolute stereochemistry was achieved in a similar fashion

to before by degradation of (E)-17 to N-Boc-L-aspartic acid

dimethyl ester (15, Scheme 6) that was found to have [a]_D+9.3

(c 0.3, CHCl₃) [lit.³⁶[a]_D+30.8 (c 2.1, CHCl₃)]. The positive

[a]_Dproves that (E)-17 is the S-enantiomer. This result also

conﬁrms that the bulky silicon group is again dictating the alkene

geometry and the major S-stereocentre (TS6 R = CH₂OTBDPS,

Fig. 3). The increased preference for formation of the E-alkene,

but the reduced chirality transfer compared to the valine derived

ester (S)-9b is opposite to what would be expected based on the

perceived steric demands of a protected CH₂O–P group versus

the normally larger iso-Pr substituent. However, these results can

be adequately explained if we consider the CH₂OTBDPS group

to be sterically larger than the iso-Pr substituent. The larger

steric bulk of the substituent attached to the stereogenic centre

should increase the E-alkene stereoselection by avoiding the

destabilising 1,2-interaction with the phenyldimethylsilyl group.

Unfortunately the anion stabilising group then has to experience

a destabilising 1,3-interaction with the R substituent if exo (TS6

in Fig. 3) or a 1,3-interaction with the phenyldimethylsilyl group

if endo (TS7 in Fig. 3). This explains the increased alkene

stereoselectivity, but reduced enantioselectivity of (E,S)-17

Fig. 4 Transition state models for better enantioselectivity.

Following a synthetic sequence identical to that used for

the preparation of 9b, but starting from each enantiomer of

serine with MOM protection gave a low overall yield for the

synthesis of the vinyl silane and the ﬁnal step oxidation to the

ester. This route was left unoptimised as chiral HPLC revealed

each enantiomer to have an ee of only 43%.⁴⁰Although not

conﬁrmed at present, the most likely points for racemisation

are thought to be either the silylcuprate addition step or the

initial MOM protection. To rapidly assess the potential of

this approach we decided to derivatise the TBDPS protected

rearrangement precursor (R)-16 (eqn. (8)). Deprotection of (R)-

16 with a HF·pyridine complex gave the desired alcohol in

65% yield. Reprotection with MOMCl and Hunig’s base gave

the desired precursor (R)-18 in 47% yield. Analysis by chiral

HPLC unfortunately revealed the material to be in 87% ee.³⁹

Although previous batches of either enantiomer of 16 used in

the rearrangements (Scheme 6) had previously been shown to

be enatiomerically pure,³⁷this particular batch had obviously

suffered some racemisation during the synthetic sequence to

prepare it from N-Boc-L-serine (Scheme 5). We think that ageing

of the reagents or slight changes in reaction conditions could

have led to some epimerisation along the route. Although not

ideal, the potential of this approach at this stage could still be

assessed using the 87% ee material (R)-18.

(8)

The best aza-[2,3]-rearrangement conditions were found to be

◦

using KH/18-C-6 (2 equiv.) in THF at 0 C for 20 mins. This

gave 19 with 10 : 1 E : Z stereoselectivity in 65% yield (eqn. (9)).

Use of LDA (2 equiv.) in THF–HMPA (4 : 1) at 0 ^◦C for 20 mins

led to a reduced alkene stereoselection (5 : 1 E : Z) and a lower

52% yield. Prolonged reaction times or lower temperatures led to

product degradation. The (E)-alkene geometry was veriﬁed from

selected one dimensional NOE experiments (eqn. (9)). The sense

of enantioinduction was assumed to be identical to the previous

Fig.

3 Competing transition state structures that erode

enantioselectivity.

3 7 3 8

O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 3 7 3 4 – 3 7 4 8

View Article Online

examples (Schemes 4 and 6).⁴¹The enantiomeric excess of (E)-

19 was determined to be 65% by chiral HPLC analysis.⁴²As the

starting material was in 87% ee this translates to an ee of 75%

(corresponding to 88% chirality transfer) if enantiomerically

pure (R)-18 had been used. The alkene stereoselectivity is slightly

less than when the O-silyl protecting group was used, but is still

very good. We believe the increased enantioselectivity of this

rearrangement (eqn. (9)) supports the potential for chelation

to control the stereoselectivity of this type of chirality transfer

in the aza-[2,3]-rearrangement. Further studies and reﬁnements

are ongoing.

alkene stereoselection (14 : 1, E : Z), gave a reduced 40%

enantioselectivity. By considering transition state models, this

was rationalised as the interplay between the bulky O-TBDPS

group avoiding the vinyl dimethylphenylsilyl substituent, but

then causing a detrimental interaction with the anion stabilising

group (Fig. 3). Attempts to counteract this by encouraging a

positive chelating interaction between a coordinating O-MOM

serine substituent and the cation associated with the enolate

(Fig. 4) led to a 65% yield of rearranged material with 10 : 1 E :

Z alkene stereoselectivity and a chirality transfer equivalent to

88%. This result implies a complimentary coordinating effect

or a subtle balance of sterics in the serine derived series.

Extrapolation to a covalently bound cyclic precursor [(S)-20]

gave no rearrangement. Further investigations into the steric

and possible coordinating interactions of chirality transfer in the

aza-[2,3]-Wittig sigmatropic rearrangement and applications to

target synthesis will be reported in due course.

Experimental

Our general experimental details have been published.^1b

(9)

Synthesis of thecyclicprecursor (S)-20 was most easily accom-

plished by O-silyl deprotection of (S)-16 (87% ee) followed by

cyclisation. In the event treatment with a HF·pyridine complex

gave alcohol (S)-21 with 10% of cyclised (S)-20 (eqn. (10)).

After separation by chromatography, the primary alcohol (S)-21

could be treated with TsOH in toluene at 50 ^◦C in the presence

[(1S)-2-Imidazol-1-yl-1-methyl-2-oxo-ethyl]-carbamic acid

tert-butyl ester (S)-5a

A solution of (S)-4a (5.00 g, 26.4 mmol) in THF (25 mL) at

0

^◦C was treated portion wise, over a period of 5 mins with

CDI (4.70 g, 29.0 mmol, 1.1 equiv.). After stirring for 30 mins,

the reaction was diluted with Et₂O (50 mL) and cautiously

quenched with H₂O. The resultant mixture was washed twice

with H₂O followed by satd. aq. NaCl before being dried

(MgSO₄) and concentrated under reduced pressure to furnish the

acylimidazole (S)-5a as a viscous oil which solidiﬁed on standing

to give a white solid (6.18 g, 98%) and was used directly in the

next step. Mp 74–76 ^◦C; [a]_D−9.7 (c 2.0, CHCl₃); m_max(ﬁlm)/cm⁻¹

3437, 2978, 2935, 1728, 1714; d_H(400 MHz; CDCl₃) 1.45 (9H, s,

^tBu), 1.52 (3H, d, J 7.1, CHCH₃), 4.98 (1H, quin, J 7.2, CHCH₃),

˚

of 4 A molecular sieves for 5 days to give the desired cyclised

product (S)-20 in 90% yield. Upon subjection of (S)-20 to a

variety of strong bases (KH, LDA, LiTMP) in THF along with

18-C-6 or HMPA, no rearrangement product was observed.

The only isolable product which was formed after extended

reaction periods at room temperature was the hydroxy acid

(47%) caused from direct hydrolysis of (S)-20 by residual water

or a hydroxide ion. The hydroxy acid recyclised on standing at

room temperature to reform (S)-20. The reluctance to undergo

rearrangement must be attributed to the high energy of the

strained bicyclic transition state TS9 through which the reaction

would have to proceed. The high energy of this transition

state prevents rearrangement and allows alternative destructive

pathways to predominate.

=

5.23 (1H, bd, J 7.0, NH), 7.13 (1H, s, CONCH CH), 7.53 (1H,

t, J 1.5, CONCH CH), 8.27 (1H, s, NCH N); d_C(100 MHz;

=

CDCl₃) 21.3, 30.8, 51.9, 83.4, 118.8, 133.8, 139.0, 157.6, 172.8.

[(1R)-2-Imidazol-1-yl-1-methyl-2-oxo-ethyl]-carbamic acid

tert-butyl ester (R)-5a

N-Boc amino acid (R)-4a (5.00 g, 26.4 mmol) was converted by

the procedure above to the corresponding acylimidazole (R)-5a

(5.26 g, 84%) as a viscous oil which solidiﬁed on standing to give

a white solid. Mp 73–75 ^◦C; [a]_D+9.7 (c 2.1, CHCl₃). All other

data were identical to that for (S)-5a.

[(1S)-2-(Dimethylphenylsilanyl)-1-methyl-2-oxo-ethyl]-carbamic

acid tert-butyl ester (S)-3a

A solution of PhMe₂SiLi (∼1 M, 2 equiv.)¹³in THF at 0 ^◦C

was added, by syringe over several minutes, to a suspension of

CuCN (1.79 g, 20.0 mmol, 1 equiv. dried in situ by heating with

a heat gun under a steady stream of argon) in THF (60 mL) at

0 ^◦C. The resultant deep red solution was stirred at 0 ^◦C for 20

mins before being cooled to −78 ^◦C and transferred by cannula

into a solution of (S)-5a (4.78 g, 20.0 mmol) in THF (60 mL)

at −78 ^◦C. The reaction was stirred for 30 mins before being

quenched by the addition of satd. aq. NH₄Cl and warmed to rt.

The reaction was opened to the air and vigorously stirred. After

1 hour any solids were removed by ﬁltration and the ﬁltrate was

extracted twice with Et₂O. The combined organic extracts were

washed with H₂O, followed by satd. aq. NaCl and dried (MgSO₄)

before being concentrated under reduced pressure. The resulting

crude product was puriﬁed by ﬂash column chromatography

(12% EtOAc–pet. ether) to give the acyl silane (S)-3a (2.95 g,

48%) as a white solid. Mp 64–65 ^◦C; [a]_D+47.5 (c 1.9, CHCl₃);

m_max(ﬁlm)/cm⁻¹3424, 2937, 2932, 1704, 1643; d_H(400 MHz;

CDCl₃) 0.53 (3H, s, SiMe₂), 0.53 (3H, s, SiMe₂), 1.08 (3H, d,

J 7.2, CHCH₃), 1.43 (9H, s, ^tBu), 4.48 (1H, bq, J 7.0, CHCH₃),

(10)

Summary

The synthesis of enantiomerically pure aza-[2,3]-Wittig rear-

rangement precursors from amino acids alanine, valine and

serine was successful via Tebbe methylenation of the corre-

sponding acyl silanes. Potential racemisation of the chiral centre

is possible when adjacent to an activated carbonyl function,

but can be avoided by the use of fresh silyl cuprate and Tebbe

reagent. Rearrangement of phenyl and ester stabilised anions

of precursors derived from alanine and valine showed that

the activating phenyldimethylsilyl group was able to dictate

stereoselectivity effectively only with the larger i-Pr chiral

substituent. The ester stabilised anion from the (S)-i-Pr series

gave a 66% yield with 85% chirality transfer (S) and 91%

alkene stereoselectivity (E). The sense of selectivity could

be understood from consideration of transition state models.

Development of the more functionalised serine series showed

that the O-TBDPS serine derivative while efﬁcient at controlling

O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 3 7 3 4 – 3 7 4 8

3 7 3 9

View Article Online

5.26 (1H, bs, NH), 7.37–7.43 (3H, m, Ar–H), 7.54–7.57 (2H,

m, Ar–H); d_C(100 MHz; CDCl₃) −4.2, −4.2, 16.5, 28.4, 59.8,

79.6, 128.4, 130.2, 134.1, 155.1; m/z (ES⁺) 308.1681 (4%, MH⁺,

C₁₆H₂₆NO₃Si requires 308.1682), 174 (100%, MH₂− SiMe₂Ph).

153.5; m/z (CI⁺) 396.2356 (16%, MH⁺, C₂₄H₃₄NO₂Si requires

+

396.2359), 262 (100%, MH₂− SiMe₂Ph).

Enantiomer (R)-6a not synthesised as rearrangement of (S)-

6a was unsatisfactory.

+

[(1S)-1-(Imidazole-1-carbonyl)-2-methylpropyl]-carbamic acid

tert-butyl ester (S)-5b

[(1R)-2-(Dimethylphenylsilanyl)-1-methyl-2-oxo-ethyl]-carbamic

acid tert-butyl ester (R)-3a

In an identical procedure to the preparation of (S)-5a, N-Boc

amino acid (S)-4b (5.00 g, 23.0 mmol) was converted to its

corresponding acylimidazole (S)-5b (6.02 g, 98%) as a white

solid and was used directly in the next step. Mp 82–83 ^◦C; [a]_D

+38.4 (c 2.0, CHCl₃); m_max(ﬁlm)/cm⁻¹3445, 3438, 2970, 2934,

1714; d_H(400 MHz; CDCl₃) 0.95 (3H, d, J 6.8, CHCH₃), 1.04

Acylimidazole (R)-5a (2.00 g, 8.37 mmol) was converted by

the procedure above to the corresponding acyl silane (R)-3a

(2.95 g, 48%) as a white solid. Mp 64–66 ^◦C; [a]_D−38.4 (c 2.2,

CHCl₃); m/z (ES⁺) 308.1674 (30%, MH⁺, C₁₆H₂₆NO₃Si requires

308.1682). All other data were identical to that for (S)-3a.

t

(3H, d, J 6.8, CHCH₃), 1.45 (9H, s, Bu), 2.17 (1H, oct, J 6.6,

CH[CH₃]₂), 4.79 (1H, dd, J 9.0, 5.8, NHCH), 5.24 (1H, d, J 9.0,

[(1S)-2-(Dimethylphenylsilanyl)-1-methylallyl]-carbamic acid

tert-butyl ester (S)-2a

=

NH), 7.13 (1H, d, J 1.0, CONCH CH), 7.53 (1H, t, J 1.6,

CONCH CH), 8.26 (1H, s, NCH N); d_C(100 MHz; CDCl₃)

A solution of Tebbe’s reagent in PhMe (0.5 M, 50.0 mL,

25.0 mmol, 1.3 equiv.) was added dropwise to a solution of (S)-3a

(6.03 g, 19.6 mmol, 1 equiv.) in THF (100 mL) at −40 ^◦C. The

reaction was allowed to stir at −40 ^◦C for 1 h before being

diluted with Et₂O (100 mL) and cautiously quenched with 0.1 M

NaOH. After gas evolution ceased, the reaction was cautiously

allowed to warm to rt with further evolution of gas. Excess

solvent was removed under reduced pressure and the residue

was taken up in pet. ether. The remaining solid was ﬁltered off

and the ﬁltrate concentrated under reduced pressure to give the

crude product. Puriﬁcation by ﬂash column chromatography

(10% EtOAc–pet. ether) furnished the desired product (S)-2a

(3.6 g, 60%) as a white solid along with recovered starting

material (0.55 g, 9%). Mp 47–49 ^◦C; [a]_D−40.5 (c 3.4, CHCl₃);

m_max(ﬁlm)/cm⁻¹3444, 2972, 2931, 1704; d_H(400 MHz; CDCl₃)

=

17.2, 19.6, 31.5, 58.5, 80.7, 116.3, 131.4, 136.5, 155.7, 169.8.

[(1R)-1-(Imidazole-1-carbonyl)-2-methylpropyl]-carbamic acid

tert-butyl ester (R)-5b

In an identical procedure to the preparation of (S)-5a, N-Boc

amino acid (R)-4b (5.00 g, 23.0 mmol) was converted to the

corresponding acylimidazole (R)-5b (5.76 g, 94%) as a white

◦

solid. Mp 82–83 C; [a]_D−38.0 (c 2.0, CHCl₃). All other data

were identical to that for (S)-5b.

[(1S)-1-(Dimethylphenylsilanecarbonyl)-2-methylpropyl]-

carbamic acid tert-butyl ester (S)-3b

In an identical procedure to the preparation of (S)-3a, acylim-

idazole (S)-5b (2.00 g, 7.50 mmol) was converted to its corre-

sponding acyl silane (S)-3b (1.90 g, 76%) as a pale oil. [a]_D−93.7

(c 2.0, CHCl₃); m_max(ﬁlm)/cm⁻¹3433, 2966, 2932, 1712, 1641; d_H

(400 MHz; CDCl₃) 0.47 (3H, d, J 6.8, CHCH₃), 0.52 (3H, s,

SiMe₂), 0.54 (3H, s, SiMe₂), 0.88 (3H, d, J 6.8, CHCH₃), 1.43

(9H, s, ^tBu), 2.06–2.14 (1H, m, CH[CH₃]₂), 4.48 (1H, dd, J 8.6,

2.6, NHCH), 5.07 (1H, d, J 8.2, NH), 7.37–7.43 (3H, m, Ar–

H), 7.55–7.59 (2H, m, Ar–H); d_C(100 MHz; CDCl₃) −4.4, −4.3,

14.8, 20.3, 28.1, 28.4, 68.5, 79.5, 128.3, 130.1, 133.9, 134.1, 156.0;

m/z (ES⁺) 336.1999 (7%, MH⁺, C₁₈H₃₀NO₃Si requires 336.1995),

0.43 (6H, s, SiMe₂), 1.13 (3H, d, J 5.9, CHCH₃), 1.42 (9H, s,

t

=

Bu), 4.36 (2H, bs, CHCH₃, NH), 5.48 (1H, d, J 1.8, C CH₂),

=

5.84 (1H, dd, J 1.6, 0.8, C CH₂), 7.35–7.37 (3H, m, Ar–H),

7.52–7.55 (2H, m, Ar–H); d_C(100 MHz; CDCl₃) −2.4, 21.5,

28.5, 50.5, 79.1, 125.5, 127.9, 129.2, 133.9, 138.3, 152.4, 155.0;

m/z (ES⁺) 306.1897 (8%, MH⁺, C₁₇H₂₈NO₂Si requires 306.1889),

172 (100%, MH₂− SiMe₂Ph).

+

[(1R)-2-(Dimethylphenylsilanyl)-1-methylallyl]-carbamic acid

tert-butyl ester (R)-2a

+

202 (100%, MH₂− SiMe₂Ph).

Acyl silane (R)-3a (200 mg, 0.65 mmol, 1 equiv.) was converted

by the procedure above to the corresponding vinyl silane (R)-2a

(127 mg, 64%) as a white solid along with recovered starting ma-

terial (15 mg, 8%). Mp 46–48 ^◦C; [a]_D+37.9 (c 2.0, CHCl₃); m/z

(ES⁺) 306.1897 (100%, MH⁺, C₁₇H₂₈NO₂Si requires 306.1889).

All other data were identical to that for (S)-2a.

[(1R)-1-(Dimethylphenylsilanecarbonyl)-2-methylpropyl]-

carbamic acid tert-butyl ester (R)-3b

In an identical procedure to the preparation of (S)-3a, acylim-

idazole (R)-5b (6.15 g, 23.0 mmol) was converted to its corre-

sponding acyl silane (R)-3b (5.30 g, 69%) as a pale oil. [a]_D+93.8

(c 2.2, CHCl₃); m/z (ES⁺) 336.1989 (100%, MH⁺, C₁₈H₃₀NO₃Si

requires 336.1995). All other data were identical to that for (S)-

3b.

Benzyl-[(1S)-2-(dimethylphenylsilanyl)-1-methylallyl]-carbamic

acid tert-butyl ester (S)-6a

A solution of (S)-2a (1.40 g, 4.60 mmol) in THF (25 mL) was

transferred by cannula into a suspension of KH (1.5 equiv.

washed twice with pet. ether) in THF (35 mL) at rt. After stirring

for 30 mins, BnBr (0.61 mL, 5.1 mmol, 1.1 equiv.) and TBAI

(170 mg, 0.46 mmol, 0.10 equiv.) were added and the reaction

was left for a further 16 hours before being quenched by the

cautious addition of H₂O. The resultant mixture was extracted

twice with Et₂O, the combined organics washed with H₂O, satd.

aq. NaCl, dried (MgSO₄), concentrated under reduced pressure

and puriﬁed by ﬂash column chromatography (4% EtOAc–pet.

ether) to furnish rearrangement precursor (S)-6a (1.52 g, 84%)

as a clear oil. [a]_D−64.0 (c 3.0, CHCl₃); m_max(ﬁlm)/cm⁻¹2960,

2935, 1682; d_H(400 MHz; CDCl₃, 55 ^◦C) 0.44 (6H, s, SiMe₂), 1.11

(3H, d, J 6.9, CHCH₃), 1.41 (9H, s, ^tBu), 3.92 (1H, bd, J 16.2,

NCH₂), 4.19 (1H, bs, NCH₂), 5.02 (1H, bs, CHCH₃), 5.65 (1H, s,

[(1S)-2-(Dimethylphenylsilanyl)-1-isopropylallyl]-carbamic acid

tert-butyl ester (S)-2b

In an identical procedure to the preparation of (S)-2a, acyl silane

(S)-3b (1.2 g, 3.58 mmol) was converted to its corresponding

vinyl silane (S)-2b (546 mg, 46%) as a pale oil along with

recovered starting material (310 mg, 26%). [a]_D+36.4 (c 1.1,

CHCl₃); m_max(ﬁlm)/cm⁻¹3448, 2962, 2930, 1709; d_H(400 MHz;

CDCl₃) 0.41 (3H, s, SiMe₂), 0.43 (3H, s, SiMe₂), 0.72 (3H, d,

J 6.7, CHCH₃), 0.82 (3H, d, J 6.8, CHCH₃), 1.42 (9H, s, ^tBu),

1.70 (1H, oct, J 6.6, CH[CH₃]₂), 4.06 (1H, t, J 7.3, NHCH), 4.40

=

(1H, d, J 9.0, NH), 5.53 (1H, dd, J 2.0, 1.0, C CH₂), 5.77 (1H,

t, J 1.6, C CH₂), 7.34–7.37 (3H, m, Ar–H), 7.52–7.55 (2H, m,

=

Ar–H); d_C(100 MHz; CDCl₃) −2.4, −2.1, 17.1, 20.5, 28.5, 31.1,

60.9, 79.0, 126.6, 127.9, 129.2, 134.0, 138.1, 150.8, 155.5; m/z

(ES⁺) 334.2198 (12%, MH⁺, C₁₉H₃₂NO₂Si requires 334.2202),

=

C CH₂), 5.81 (1H, s, C CH₂), 7.14–7.55 (10H, m, Ar–H); d_C

(100 MHz; CDCl₃) −2.9, −2.7, 17.8, 28.4, 41.5, 46.7, 79.6, 126.4,

126.7, 127.4, 127.8, 128.1, 129.1, 138.1, 140.0, 140.6, 150.6,

+

200 (100%, MH₂− SiMe₂Ph).

3 7 4 0

O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 3 7 3 4 – 3 7 4 8

View Article Online

t

[(1R)-2-(Dimethylphenylsilanyl)-1-isopropylallyl]-carbamic acid

tert-butyl ester (R)-2b

396.2359), 262 (38%, MH⁺− Bu, -Ph), 135 (36%, SiMe₂Ph⁺),

57 (52%, ^tBu⁺).

In an identical procedure to the preparation of (S)-2a, acyl silane

(R)-3b (1.41 g, 4.21 mmol) was converted to its corresponding

vinyl silane (R)-2b (662 mg, 47%) as a pale oil along with

recovered starting material (300 mg, 21%). [a]_D−36.8 (c 1.1,

CHCl₃); m/z (ES⁺) 334.2209 (11%, MH⁺, C₁₉H₃₂NO₂Si requires

[3-(Dimethylphenylsilanyl)-5-methyl-1-phenylhex-3-enyl]-

carbamic acid tert-butyl ester 7b

Rearrangement precursor (S)-6b (100 mg, 0.24 mmol) was

dissolved in THF–HMPA (2 mL, 4 : 1) and cooled to 0 ^◦C before

being treated with n-BuLi (170 lL of a 2.1 M solution in hexanes,

0.35 mmol, 1.5 equiv.). After 5 mins, the bright red solution was

quenched by the addition of MeOH and allowed to warm to

rt. The reaction was diluted with water (15 mL), extracted with

Et₂O (2 × 15 mL), the combined organics were washed with

satd. aq. NaCl (15 mL), dried (MgSO₄), concentrated under

reduced pressure and puriﬁed by ﬂash column chromatography

(5% EtOAc–pet. ether) to give 7b (74 mg, 74%) as a clear oil and

an inseparable 3 : 1 mixture of E and Z alkene isomers (E alkene

ee 58%²²). [a]_D+3.8 (c 2.5, CHCl₃); m_max(ﬁlm)/cm⁻¹3446, 2959,

2932, 1710; d_H(500 MHz; CDCl₃) 0.47 (9H, s, SiMe₂)_E, 0.48

(9H, s, SiMe₂)_E, 0.53 (3H, s, SiMe₂)_Z, 0.54 (3H, s, SiMe₂)_Z, 0.85

(3H, d, J 6.5, CHCH₃)_Z, 0.92 (9H, d, J 6.3, CHCH₃)_E, 0.95 (3H,

d, J 6.6, CHCH₃)_Z, 1.10 (9H, d, J 6.5, CHCH₃)_E, 1.49 (36H, bs,

^tBu)_{E + Z}, 2.30 (1H, dd, J 13.5, 10.1, CHCH₂)_Z, 2.48–2.52 (4H, m,

CHCH_2E, CH[CH₃]_2,Z), 2.59–2.64 (4H, m, CHCH₂)_{E + Z}, 2.70–

2.80 (3H, m, CH[CH₃]₂)_E, 4.46 (4H, bs, NHCH)_{E + Z}, 4.74, (3H,

+

332.2202), 200 (100%, MH₂− SiMe₂Ph). All other data were

identical to that for (S)-2b.

Benzyl-[(1S)-2-(dimethylphenylsilanyl)-1-isopropylallyl]-

carbamic acid tert-butyl ester (S)-6b

In an identical procedure to the preparation of (S)-6a, vinyl

silane (S)-2b (355 mg, 1.02 mmol) was converted to rearrange-

ment precursor (S)-6b (397 mg, 88%) as a clear oil, in a 1 : 1

ratio of Boc rotamers (>95% ee²¹). [a]_D−107 (c 1.3, CHCl₃);

m_max(ﬁlm)/cm⁻¹2960, 2930, 1678; d_H(500 MHz; CDCl₃) 0.33

(1.5H, s, SiMe₂), 0.38 (1.5H, s, SiMe₂), 0.46 (1.5H, s, SiMe₂)_rot,

0.50 (1.5H, s, SiMe₂)_rot, 0.72–0.77 (4.5H, m, CHCH₃), 0.84

(1.5H, d, J 6.5, CHCH₃), 1.38 (4.5H, s, ^tBu)_rot, 1.51 (4.5H, s, ^tBu),

2.05–2.16 (1H, m, CH[CH₃]₂), 4.14 (0.5H, d, J 15.9, NCH₂)_rot,

4.17 (0.5H, d, J 15.7, NCH₂)_rot, 4.26 (0.5H, d, J 15.7, NCH₂),

4.30 (0.5H, d, J 15.8, NCH₂), 4.47 (0.5H, d, J 10.8, NCH), 4.73

=

(0.5H, d, J 11.0, NCH)_rot, 5.71 (0.5H, s, C CH₂)_rot, 5.78 (0.5H, s,

=

bs, NH)_E, 4.85, (1H, bs, NH)_Z, 5.85 (3H, d, J 9.6, C CH)_E,

5.90 (1H, d, J 10.5, C CH)_Z, 7.15–7.65 (40H, m, Ar–H)_{E + Z}; d_C

=

C CH₂), 5.90 (0.5H, s, C CH₂)_rot, 5.96 (0.5H, s, C CH₂), 7.20–

7.56 (10H, m, Ar–H); d_C(125 MHz; CDCl₃) −3.6, −3.4, −3.1,

19.3, 19.6, 20.5, 20.8, 27.8, 27.9, 28.4, 28.5, 45.5, 45.9, 61.8, 62.9,

79.6, 79.8, 126.5, 127.7, 127.8, 127.9, 128.1, 129.0, 129.1, 129.3,

129.9, 134.1, 134.2, 137.5, 137.7, 140.0, 140.2, 146.6, 147.5,

156.7, 156.4; m/z (CI⁺) 424.2672 (13%, MH⁺, C₂₆H₃₈NO₂Si

requires 424.2640), 324 (100%, MH⁺− Boc).

=

(125 MHz; CDCl₃) −2.2, −2.1, −0.9, −0.6, 22.3, 22.7, 22.8,

28.0, 28.4, 31.5, 38.0, 46.7, 54.8, 79.1, 126.0, 126.1, 126.9,

127.9, 128.4, 128.9, 129.0, 130.7, 131.8, 134.0, 134.2, 138.7,

139.4, 143.8, 153.0, 155.1, 155.8; m/z (FAB) 424.2669 (3%,

MH⁺, C₂₆H₃₇NO₂Si requires 424.2672), 135 (100%, SiMe₂Ph⁺),

57 (56%, ^tBu⁺).

In an identical fashion (R)-6b (94 mg, 0.22 mmol) was

rearranged to give 7b (63 mg, 67%) as an inseparable, mixture

of alkene isomers in an E : Z ratio of 3 : 1 (E alkene ee

58%²²). [a]_D+1.5 (c 2.5, CHCl₃); m/z (FAB) 424.2659 (7%, MH⁺,

C₂₆H₃₇NO₂Si requires 424.2672). All other data were identical

to that shown above.

Benzyl-[(1R)-2-(dimethylphenylsilanyl)-1-isopropylallyl]-

carbamic acid tert-butyl ester (R)-6b

In an identical procedure to the preparation of (S)-6a, vinyl

silane (R)-2b (1.12 g, 3.36 mmol) was converted to rearrange-

ment precursor (R)-6b (1.23 g, 87%) as a clear oil in a 1 : 1 ratio

of Boc rotamers (>95% ee²¹). [a]_D+105 (c 2.6, CHCl₃); m/z

(CI⁺) 424.2652 (18%, MH⁺, C₂₆H₃₈NO₂Si requires 424.2640).

All other data were identical to that for (S)-6b.

(1S)-2-(Dimethylphenylsilanyl)-1-methylallylamine (S)-10a

The Boc protected amine (S)-2a (1.20 g, 3.93 mmol) was treated

with a solution of HCl in dioxane (4 M, 3.93 mL, 15.7 mmol,

4 equiv.), and stirred at rt for 1.5 hours. The reaction was

diluted with H₂O (20 mL) and washed with Et₂O (20 mL),

the organic portion was back extracted with H₂O (20 mL)

and the combined aqueous phases were basiﬁed with satd. aq.

NaHCO₃solution. The basiﬁed aqueous phase was extracted

with Et₂O (2 × 25 mL), the organic phases combined, washed

with brine (50 mL), dried (MgSO₄) and concentrated under

reduced pressure to give (S)-10a (767 mg, 95%) as a pale oil,

with no further puriﬁcation necessary. [a]_D−2.6 (c 3.4, CHCl₃);

m_max(ﬁlm)/cm⁻¹3374, 2954; d_H(400 MHz; CDCl₃) 0.43 (3H, s,

SiMe₂), 0.44 (3H, s, SiMe₂), 1.11 (3H, d, J 6.7, CHCH₃), 2.06

(2H, bs, NH), 3.64 (1H, q, J 6.6, CHCH₃), 5.45 (1H, dd, J 2.3,

1(R,S)-[3-(Dimethylphenylsilanyl)-1-phenylpent-3-(Z,E)-enyl]-

carbamic acid tert-butyl ester 7a

Rearrangement precursor (S)-6a (100 mg, 0.25 mmol) was

dissolved in THF–HMPA (2 mL, 4 : 1) and cooled to −78 C

◦

before being treated with n-BuLi (180 lL of a 2.1 M solution

in hexanes, 0.38 mmol, 1.5 equiv.). After 15 mins, the bright

red solution was quenched by the addition of MeOH and

allowed to warm to rt. The reaction was diluted with H₂O

(15 mL) and extracted with Et₂O (2 × 15 mL). The combined

organics were washed with satd. aq. NaCl (15 mL), dried

(MgSO₄), concentrated under reduced pressure and puriﬁed by

ﬂash column chromatography (4% EtOAc–pet. ether) to give 7a

(56 mg, 56%) as a clear oil and an inseparable 1.8 : 1 mixture of E

and Z alkene isomers. [a]_D+0.8 (c 2.2, CHCl₃); m_max(ﬁlm)/cm⁻¹

3445, 2956, 2929, 1711; d_H(500 MHz; CDCl₃) 0.35 (5.4H, s,

SiMe₂)_E, 0.37 (5.4H, s, SiMe₂)_E, 0.46 (3H, s, SiMe₂)_Z, 0.46

(3H, s, SiMe₂)_Z, 1.39 (25.2H, bs, ^tBu)_{E + Z}, 1.66 (8.4H, ad, J 6.7,

CHCH₃)_{E + Z}, 2.18–2.23 (1H, m, CHCH₂)_Z, 2.41–2.44 (1.8H, m,

CHCH₂)_E, 2.54–2.58 (2.8H, m, CHCH₂)_{E + Z}, 4.52–4.75 (5.6H,

=

0.8, C CH₂), 5.91 (1H, dd, J 2.3, 1.5, C CH₂), 7.34–7.37 (3H,

m, Ar–H), 7.53–7.56 (2H, m, Ar–H); d_C(100 MHz; CDCl₃)

−2.1, −1.9, 24.6, 51.6, 123.9, 127.9, 129.1, 134.0, 138.7, 156.6.

m/z (ES⁺) 206.1369 (24%, MH⁺, C₁₃H₁₉NSi requires 206.1365),

128 (100%, M⁺− Ph).

(1R)-2-(Dimethylphenylsilanyl)-1-methylallylamine (R)-10a

=

m, NH, NHCH)_{E + Z}, 6.07 (1.8H, q, J 6.5, C CH)_E, 6.20 (1H,

q, J 6.8, C CH)_Z, 6.91–7.56 (28H, m, Ar–H)_{E + Z}; d_C(125 MHz;

The Boc protected amine (R)-2a (409 mg, 1.34 mmol) was

converted by the procedure above to its corresponding primary

amine (R)-10a (258 mg, 94%) as a pale oil, with no further

puriﬁcation necessary. [a]_D+2.8 (c 2.2, CHCl₃); m/z (ES⁺)

206.1362 (24%, MH⁺, C₁₃H₁₉NSi requires 206.1365). All other

data were identical to that for (S)-10a.

=

CDCl₃) −2.4, −2.3, −1.2, −0.9, 15.1, 18.4, 28.4, 37.5, 47.2, 54.8,

79.2, 126.0, 126.1, 126.8, 126.9, 127.9, 128.0, 128.4, 128.9, 129.1,

134.0, 134.2, 134.9, 136.4, 138.6, 139.1, 139.8, 142.2, 143.6,

155.2; m/z (CI⁺) 396.2340 (17%, MH⁺, C₂₄H₃₄NO₂Si requires

O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 3 7 3 4 – 3 7 4 8

3 7 4 1

View Article Online

[(1S)-2-(Dimethylphenylsilanyl)-1-methylallylamino]-acetic acid

methyl ester (S)-11a

m_max(ﬁlm)/cm⁻¹3378, 2956 (C–H); d_H(400 MHz; CDCl₃) 0.41

(3H, s, SiMe₂), 0.44 (3H, s, SiMe₂), 0.78 (3H, d, J 6.7, CHCH₃),

0.83 (3H, d, J 6.8, CHCH₃), 1.20 (2H, bs, NH₂), 1.67 (1H, oct,

J 6.6, CH[CH₃]₂), 3.21 (1H, d, J 6.2, NH₂CH), 5.52 (1H, dd,

J 2.5, 0.7, C CH₂), 5.84 (1H, dd, J 2.5, 1.3, C CH₂), 7.34–

7.37 (3H, m, Ar–H), 7.53–7.56 (2H, m, Ar–H); d_C(100 MHz;

CDCl₃) −2.1, −1.8, 16.8, 21.0, 32.1, 62.6, 125.9, 127.8, 129.0,

134.0, 138.7, 154.5; m/z (ES⁺) 234.1679 (28%, MH⁺, C₁₄H₂₄NSi

requires 234.1678), 156 (100%, M⁺− Ph).

Primary amine (S)-10a (767 mg, 3.74 mmol) was dissolved in

DMF (11 mL) and cooled to 0 ^◦C before being treated with Et₃N

(626 ll, 4.49 mmol, 1.2 equiv.) followed by methyl bromoacetate

(425 ll, 4.49 mmol, 1.2 equiv.). The reaction was allowed to

stir at 0 ^◦C for 75 mins before being diluted with EtOAc and

washed three times with H₂O, followed by satd. aq. NaCl.

The resulting solution was dried (MgSO₄), concentrated under

reduced pressure and puriﬁed by ﬂash column chromatography

(20% EtOAc–pet. ether) to give (S)-11a as a clear oil (834 mg,

81%). [a]_D−16.8 (c 3.0, CHCl₃); m_max(ﬁlm)/cm⁻¹3340, 2954,

1738; d_H(400 MHz; CDCl₃) 0.42 (3H, s, SiMe₂), 0.43 (3H, s,

SiMe₂), 1.10 (3H, d, J 6.6, CHCH₃), 3.12 (1H, d, J 17.5,

NHCH₂), 3.20 (1H, d, J 17.4, NHCH₂), 3.33 (1H, q, J 6.6,

=

(1R)-2-(Dimethylphenylsilanyl)-1-isopropylallylamine (R)-10b

In an identical procedure to the preparation of (S)-10b Boc

protected amine (R)-2b (537 mg, 1.61 mmol) was converted to

(R)-10b (312 mg, 83%) as a pale oil that required no further

puriﬁcation. [a]_D−14.4 (c 1.9, CHCl₃); m/z (ES⁺) 234.1685

(24%, MH⁺, C₁₄H₂₄NSi requires 234.1678). All other data were

identical to that for (S)-10b.

=

CHCH₃), 3.67 (3H, s, OMe), 5.50 (1H, d, J 2.6, C CH₂),

5.88 (1H, dd, J 2.6, 1.0, C CH₂), 7.33–7.36 (3H, m, Ar–H),

=

7.52–7.54 (2H, m, Ar–H); d_C(100 MHz; CDCl₃) −2.2, −2.1,

23.0, 48.6, 51.7, 58.5, 126.4, 127.8, 129.0, 134.0, 138.5, 153.6,

173.2; m/z (ES⁺) 278.1573 (62%, MH⁺, C₁₅H₂₄NO₂Si requires

278.1576).

[(1S)-2-(Dimethylphenylsilanyl)-1-isopropylallylamino]-acetic

acid methyl ester (S)-11b

In an identical procedure to the preparation of (S)-11a, except

the reaction was left to stir for 2 hours, primary amine (S)-

10b (275 mg, 1.18 mmol) was converted to (S)-11b (277 mg,

77%) as a clear oil that required no further puriﬁcation.

[a]_D−20.6 (c 3.3, CHCl₃); m_max(ﬁlm)/cm⁻¹3364, 2954, 1732;

d_H(400 MHz; CDCl₃) 0.41 (3H, s, SiMe₂), 0.43 (3H, s,

SiMe₂), 0.82 (3H, d, J 6.8, CHCH₃), 0.91 (3H, d, J 6.8,

CHCH₃), 1.62 (1H, oct, J 6.9, CH[CH₃]₂), 1.68 (1H, bs, NH),

2.06 (1H, d, J 6.5, NHCH), 3.04 (1H, d, J 17.6, NHCH₂), 3.15

(1H, d, J 17.5, NHCH₂), 3.65 (3H, s, OMe), 5.56 (1H, d, J 2.8,

C CH₂), 5.74 (1H, dd, J 2.8, 0.9, C CH₂), 7.32–7.35 (3H, m,

Ar–H), 7.52–7.55 (2H, m, Ar–H); d_C(100 MHz; CDCl₃) −2.0,

−1.7, 18.5, 21.0, 31.2, 48.8, 51.6, 71.8, 127.7, 128.9, 134.0, 138.9,

151.3, 173.5; m/z (ES⁺) 306.1888 (100%, MH⁺, C₁₇H₂₈NO₂Si

requires 306.1889).

[(1R)-2-(Dimethylphenylsilanyl)-1-methylallylamino]-acetic acid

methyl ester (R)-11a

The primary amine (R)-10a (223 mg, 1.09 mmol) was converted

by the procedure above to (R)-11a (277 mg, 92%) as a clear

oil, without need for further puriﬁcation. [a]_D+16.8 (c 3.2,

CHCl₃); m/z (ES⁺) 278.1574 (62%, MH⁺, C₁₅H₂₄NO₂Si requires

278.1576). All other data were identical to that for (S)-11a.

=

{tert-Butoxycarbonyl-[(1S)-2-(dimethylphenylsilanyl)-1-

methylallyl]-amino}-acetic acid methyl ester (S)-9a

A solution of (S)-11a (447 mg, 1.61 mmol) in Et₃N (8 mL) was

treated with Boc anhydride (423 mg, 1.94 mmol, 1.2 equiv.) and

heated to reﬂux for 4 h. The reaction was allowed to cool to rt, the

solvent removed under reduced pressure and the residue puriﬁed

by ﬂash column chromatography (10% EtOAc–pet. ether) to

give (S)-9a (579 mg, 95%) as a clear oil in a 2 : 1 mixture of Boc

rotamers (>95% ee³⁰). [a]_D−42.1 (c 1.7, CHCl₃); m_max(ﬁlm)/cm⁻¹

2954, 1753, 1691; d_H(400 MHz; CDCl₃) 0.39–0.42 (6H, m,

SiMe₂), 1.17–1.19 (3H, m, CHCH₃), 1.40 (6H, s, ^tBu), 1.44 (3H, s,

^tBu)_rot, 3.13 (0.66H, d, J 18.0, NCH₂), 3.19 (0.66H, d, J 18.0,

NCH₂), 3.31 (0.33H, d, J 17.6, NCH₂)_rot, 3.63 (0.33H, d, J 17.6,

NCH₂)_rot, 3.65 (2H, s, OMe), 3.67 (1H, s, OMe)_rot, 4.87 (0.33H,

q, J 6.5, CHCH₃)_rot, 5.11 (0.66H, q, J 6.8, CHCH₃), 5.61–5.64

[(1R)-2-(Dimethylphenylsilanyl)-1-isopropylallylamino]-acetic

acid methyl ester (R)-11b

Primary amine (R)-10b (282 mg, 1.21 mmol) was converted by

the procedure above to (R)-11b (277 mg, 77%) as a clear oil that

required no further puriﬁcation. [a]_D+20.1 (c 2.8, CHCl₃); m/z

(ES⁺) 306.1902 (100%, MH⁺, C₁₇H₂₈NO₂Si requires 306.1889).

All other data were identical to that for (S)-11b.

[2-(tert-Butyldimethylsilanyloxy)-ethyl]-[(1S)-2-(dimethyl-

phenylsilanyl)-1-isopropylallyl]-carbamic acid tert-butyl ester

(S)-12b

=

(1H, m, C CH₂), 5.77–5.79 (1H, m, C CH₂), 7.34–7.36 (3H,

m, Ar–H), 7.46–7.49 (2H, m, Ar–H); d_C(100 MHz; CDCl₃)

−3.6, −2.7, 16.1, 17.0, 28.3, 28.4, 44.1, 51.7, 51.8, 52.3, 54.0,

80.0, 80.4, 127.6, 127.8, 127.9, 129.0, 129.2, 133.8, 137.7, 138.0,

150.4, 150.8, 154.6, 155.0, 171.0, 171.3; m/z (ES⁺) 400.1931

(100%, MNa⁺, C₂₀H₃₁NO₄NaSi requires 400.1920), 378 (82%,

MH⁺).

A solution of the N-Boc protected amine (S)-2b (1.53 g,

4.60 mmol) in DMF (3 mL) was transferred by cannula into

a stirred suspension of KH (2 equiv. of a 30% w/w suspension

in m_◦ineral oil, washed twice with pet. ether) in DMF (6 mL)

at 0 C. After stirring for 80 mins, BrCH₂CH₂OTBS (506 lL,

2.36 mmol, 2 equiv.) and TBAI (4.4 mg, 0.12 mmol, 0.10 equiv.)

were_◦added and the reaction was stirred for a further 30 mins

at 0 C followed by 1 h at rt before being quenched with pH 7

phosphate buffer solution. The resulting mixture was diluted

with Et₂O and washed three times with H₂O, followed by satd.

aq. NaCl, dried (MgSO₄), concentrated under reduced pressure

and puriﬁed by ﬂash column chromatography (1.5% EtOAc in

pet. ether) to give (S)-12b (1.43 g, 63%) a clear oil as a 0.6 :

0.4 mixture of Boc rotamers with recovered starting material

(281 mg, 18%). [a]_D−82.0 (c 1.4, CHCl₃); m_max(ﬁlm)/cm⁻¹2957,

2929, 2885, 2856, 1679; d_H(500 MHz; CDCl₃) 0.03 (6H, s,

{tert-Butoxycarbonyl-[(1R)-2-(dimethylphenylsilanyl)-1-

methylallyl]-amino}-acetic acid methyl ester (R)-9a

Secondary amine (R)-11a (35 mg, 0.13 mmol) was converted

by the procedure above to (R)-9a (40 mg, 85%) as a clear oil

in a 2 : 1 mixture of Boc rotamers (>95% ee³⁰). [a]_D+41.5 (c

1.0, CHCl₃); m/z (ES⁺) 378.2113 (100%, MH⁺, C₂₀H₃₂NO₄Si

requires 378.2101). All other data were identical to that for (S)-

9a.

(1S)-2-(Dimethylphenylsilanyl)-1-isopropylallylamine (S)-10b

t

SiMe₂Bu), 0.38 (2.4H, s, SiMe₂Ph)_rot, 0.39 (1.8H, s, SiMe₂Ph),

In an identical procedure to the preparation of (S)-10a, except

stirring the reaction for 3 h, Boc protected amine (S)-2b (1.12 g,

3.36 mmol) was converted to (S)-10b (306 mg, 88%) as a pale oil

that required no further puriﬁcation. [a]_D+14.6 (c 2.0, CHCl₃);

0.39 (1.8H, s, SiMe₂Ph), 0.75 (1.2H, d, J 6.4, CHCH₃)_rot, 0.78

(1.8H, d, J 6.4, CHCH₃), 0.85 (3H, d, J 6.5, CHCH₃), 0.88

(3.6H, s, Si^tBu)_rot, 0.89 (4.4H, s, Si^tBu), 1.44 (3.6H, s, O^tBu)_rot,

1.46 (4.4H, s, O^tBu), 2.11–2.44 (1H, m, CH[CH₃]₂), 2.80 (0.6H,

3 7 4 2

O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 3 7 3 4 – 3 7 4 8

View Article Online

ddd, J 15.6, 10.5, 5.2, OCH₂), 2.90 (0.4H, ddd, J 14.1, 9.6, 5.2,

OCH₂)_rot, 3.15 (0.6H, ddd, J 15.6, 10.4, 5.4, OCH₂), 3.25 (0.4H,

ddd, J 13.6, 9.5, 5.7, OCH₂)_rot, 3.40 (0.6H, td, J 9.7, 5.2, NCH₂),

3.44 (0.4H, td, J 9.3, 5.2, NCH₂)_rot, 3.54 (0.6H, td, J 9.7, 5.5,

NCH₂), 3.67 (0.4H, td, J 9.3, 5.8, NCH₂)_rot, 4.31 (0.4H, d, J

10.9, NCH)_rot, 4.56 (0.6H, d, J 11.0, NCH), 5.69 (0.6H, d, J

column chromatography (5% EtOAc–pet. ether) and gave (S)-9b

(200 mg, 62%) as a clear oil in a 0.6 : 0.4 mixture of Boc rotamers

(>95% ee³⁰). [a]_D−98.0 (c 2.6, CHCl₃); m_max(ﬁlm)/cm⁻¹2956,

2931, 2875, 1785, 1693; d_H(500 MHz; CDCl₃) 0.41 (1.2H, s,

SiMe₂)_rot, 0.42 (1.8H, s, SiMe₂), 0.44 (1.8H, s, SiMe₂), 0.45

(1.2H, s, SiMe₂)_rot, 0.70 (1.2H, d, J 6.6, CHCH₃)_rot, 0.73 (1.8H,

d, J 6.5, CHCH₃), 0.95 (1.8H, d, J 6.4, CHCH₃), 0.96 (1.2H,

d, J 6.3, CHCH₃)_rot, 1.41 (5.4H, s, Bu), 1.48 (3.6H, s, Bu)_rot,

1.94–2.05 (1H, m, CH[CH₃]₂), 3.42 (0.6H, d, J 17.5, NCH₂),

3.52 (0.4H, d, J 17.3, NCH₂)_rot, 3.67 (3H, s, OMe), 3.73 (0.6H,

d, J 17.3, NCH₂), 3.87 (0.4H, d, J 17.1, NCH₂)_rot, 4.37 (0.4H,

d, J 11.0, NCH)_rot, 4.59 (0.6H, d, J 11.1, NCH), 5.63 (0.6H, d,

=

1.3, C CH₂), 5.74 (0.4H, d, J 1.2, C CH₂)_rot, 5.90 (0.6H, s,

t

=

C CH₂), 5.92 (0.4H, s, C H₂)_rot, 7.32–7.36 (3H, m, Ar–H),

7.49–7.53 (2H, m, Ar–H); d_C(125 MHz; CDCl₃) −5.2, −3.7,

−3.5, −3.0, 18.4, 19.1, 19.6, 20.4, 20.6, 26.0, 27.2, 27.5, 28.5,

28.6, 44.0, 44.1, 60.7, 60.8, 61.3, 62.0, 79.3, 79.6, 127.7, 127.8,

128.4, 129.0, 129.2, 134.1, 137.5, 137.7, 146.8, 147.4, 155.6,

156.0; m/z (FAB) 492.3347 (1%, MH⁺, C₂₇H₅₀NO₃Si₂requires

492.3329), 392 (8%, MH⁺− Boc), 135 (58%, PhMe₂Si⁺).

=

J 1.9, C CH₂), 5.69 (0.4H, d, J 1.5, C CH₂)_rot, 5.75 (0.6H, s,

=

C CH₂), 5.79 (0.4H, s, C CH₂)_rot, 7.33–7.36 (3H, m, Ar–H),

7.50–7.55 (2H, m, Ar–H); d_C(125 MHz; CDCl₃) −3.8, −3.8,

−3.6, −3.2, 19.3, 19.8, 20.2, 20.4, 27.6, 27.9, 28.3, 28.44, 43.7,

44.1, 51.6, 51.8, 60.8, 62.1, 80.1, 80.5, 127.7, 127.8, 128.2, 129.0,

129.2, 134.1, 134.2, 137.6, 137.8, 147.1, 147.8, 155.2, 155.5,

170.7, 170.9; m/z (FAB) 406.2417 (7%, MH⁺, C₂₂H₃₆NO₄Si

requires 406.2414), 135 (100%, PhMe₂Si⁺).

[2-(tert-Butyldimethylsilanyloxy)-ethyl]-[(1R)-2-(dimethyl-

phenylsilanyl)-1-isopropylallyl]-carbamic acid tert-butyl ester

(R)-12b

N-Boc protected amine (R)-2b (1.00 g, 3.00 mmol) was converted

by the procedure above to (R)-12b (970 mg, 66%) a clear oil in a

0.6 : 0.4 mixture of Boc rotamers along with recovered starting

material (189 mg, 19%). [a]_D+84.3 (c 1.3, CHCl₃); m/z (FAB⁺):

m/z 492.3329 (6%, MH⁺, C₂₇H₅₀NO₃Si₂requires 492.3329). All

other data were identical to that for (S)-12b.

{tert-Butoxycarbonyl-[(1R)-2-(dimethylphenylsilanyl)-1-isopro-

pylallyl]-amino}-acetic acid methyl ester (R)-9b

Silyl ether (R)-12b (945 mg, 1.93 mmol) was converted by the

procedure above to the primary alcohol (668 mg, 92%) as a clear

oil in a 0.7 : 0.3 mixture of Boc rotamers. [a]_D+92.4 (c 1.6,

CHCl₃); m/z (CI⁺) 378.2447 (12%, MH⁺, C₂₁H₃₆NO₃Si requires

378.2464). All other data were identical to that for the primary

alcohol derived from (S)-12b.

{tert-Butoxycarbonyl-[(1S)-2-(dimethylphenylsilanyl)-1-iso-

propylallyl]-amino}-acetic acid methyl ester (S)-9b

A solution of silyl ether (S)-12b (1.43 g, 2.91 mmol) in THF

(30 mL) was treated with TBAF (7.28 mL of a 1 M solution

in THF, 7.28 mmol, 2.5 equiv.). The reaction was stirred

at rt for 30 mins before being quenched by the addition of

pH 7 phosphate buffer solution. The reaction was diluted with

Et₂O (30 mL), washed with H₂O (30 mL), satd. aq. NaCl

(30 mL), dried (MgSO₄), concentrated under reduced pressure

and puriﬁed by ﬂash column chromatography (20% EtOAc–

pet. ether) to give the primary alcohol (1.00 g, 92%) as a clear

oil in a 0.7 : 0.3 mixture of Boc rotamers. [a]_D−90.0 (c 1.8,

CHCl₃); m_max(ﬁlm)/cm⁻¹3624, 3400, 2962, 2930, 1680, 1651; d_H

(400 MHz; CDCl₃) 0.40 (2.1H, s, SiMe₂), 0.41 (3.9H, s, SiMe₂),

0.77 (3H, d, J 6.4, CHCH₃), 0.87 (0.9H, d, J 6.6, CHCH₃)_rot,

0.89 (2.1H, d, J 6.6, CHCH₃), 1.44 (6.3H, s, ^tBu), 1.46 (2.7H, s,

O^tBu)_rot, 2.08–2.16 (1H, m, CH[CH₃]₂), 2.91 (0.3H, dt, J 14.4,

6.5, OCH₂)_rot, 3.06 (0.7H, ddd, J 14.9, 6.3, 3.6, OCH₂), 3.26

(0.3H, dt, J 14.5, 6.2, OCH₂)_rot, 3.34 (0.7H, ddd, J 14.9, 6.5, 3.6,

OCH₂), 3.51–3.65 (2H, m, NCH₂), 4.34 (0.7H, d, J 11.0, NCH),

The primary alcohol from above (643 mg, 1.71 mmol) was

converted by the procedure above to give (R)-9b (256 mg, 38%)

as a clear oil in a 0.6 : 0.4 mixture of Boc rotamers (>95% ee³⁰)

along with recovered starting material (100 mg, 16%). [a]_D+96.2

(c 1.2, CHCl₃); m/z (FAB) 406.2419 (7%, MH⁺, C₂₂H₃₆NO₄Si

requires 406.2414). All other data were identical to that for (S)-

9b.

2-tert-Butoxycarbonylamino-4-(dimethylphenylsilanyl)-hex-4-

enoic acid methyl ester 13a

A solution of (S)-9a (150 mg, 0.400 mmol, dried by azeotrope

with toluene) in THF (1 mL) was transferred by cannula into a

stirred suspension of KH (134 mg of a 30% w/w suspension in

mineral oil, washed twice with pet. ether, 1.00 mmol, 2.5 equiv.)

in THF (1 mL) at −40 ^◦C. 18-C-6 (106 mg, 0.400 mmol,

1.0 equiv.) was added and the reaction was allowed to stir for 2 h

before being allowed to warm to −20 ^◦C and stirred for a further

2.5 h. The reaction was quenched with MeOH, allowed to warm

to rt, diluted with H₂O (20 mL) and extracted with Et₂O (2 ×

20 mL). The organic extracts were washed with satd. aq. NaCl

(20 mL), dried (MgSO₄), concentrated under reduced pressure

and puriﬁed by ﬂash column chromatography (12% EtOAc–pet.

ether) to give 13a (98 mg, 65%) as a clear oil and an inseparable

1 : 0.9–1 mixture of E and Z alkene isomers. [a]_D+0.4 (c 2.3,

CHCl₃); m_max(ﬁlm)/cm⁻¹3443, 2954, 1740, 1713; d_H(400 MHz;

CDCl₃) 0.40 (3H, s, SiMe₂)_E, 0.41 (3H, s, SiMe₂)_E, 0.46 (3H, s,

=

4.59 (0.3H, d, J 11.0, NCH)_rot, 5.70 (0.3H, d, J 1.6, C CH₂)_rot,

=

5.74 (0.7H, d, J 1.7, C CH₂), 5.85 (0.7H, s, C CH₂), 5.86

(0.3H, s, C CH₂)_rot, 7.33–7.37 (3H, m, Ar–H), 7.48–7.51 (2H,

=

m, Ar–H); d_C(100 MHz; CDCl₃) −3.7, −3.6, −3.5, −3.0, 19.0,

20.3, 20.5, 21.1, 27.3, 27.6, 28.4, 28.5, 44.4, 44.7, 61.1, 62.3,

63.9, 80.0, 80.8, 127.8, 128.5, 129.0, 129.3, 129.3, 134.1, 137.3,

137.5, 146.9, 148.0, 155.9, 158.3; m/z (FAB⁺) 378.2429 (2%,

MH⁺, C₂₁H₃₆NO₃Si requires 378.2464), 135 (35%, PhMe₂Si⁺),

57 (100%, ^tBu⁺).

t

A solution of the primary alcohol from above (309 mg,

0.820 mmol, 1 equiv.) in acetone (9 mL) was cooled to 0 ^◦C, and

treated with Jones’ reagent (1.03 mL). The reaction was stirred

at 0 ^◦C for 10 mins before being diluted with H₂O (30 mL)

and extracted with Et₂O (2 × 30 mL). The combined organics

were washed with H₂O (2 × 20 mL) followed by satd. aq. NaCl

(20 mL) before being dried (MgSO₄) and concentrated under

reduced pressure. The crude acid thus obtained was dissolved in

Et₂O (10 mL) and CH₂N₂was bubbled through the solution

under a stream of N₂until a yellow colouration persisted.

Further N₂was bubbled through the reaction until the yellow

colouration dissipated and the excess solvent was removed under

reduced pressure. The crude product was puriﬁed by ﬂash

SiMe₂)_Z, 0.46 (3H, s, SiMe₂)_Z, 1.43 (9H, bs, Bu)_E, 1.44 (9H,

t

bs, Bu)_Z, 1.63 (3H, d, J 5.6, CHCH₃)_Z, 1.76 (3H, d, J 6.7,

CHCH₃)_E, 2.32 (1H, dd, J 13.4, 9.0, CHCH₂)_Z, 2.50 (1H, dd,

J 13.3, 9.2, CHCH₂)_E, 2.58–2.69 (2 H, m, CHCH₂)_{E + Z}

,

3.67 (3H, s, OMe)_E, 3.68 (3H, s, OMe)_Z, 4.25–4.30 (2H, m,

NHCH)_{E + Z}, 4.75 (1H, bd, J 7.9, NH)_E, 4.85 (1H, bd, J 7.8,

=

NH)_Z, 6.11 (1H, q, J 6.5, C CH)_E, 6.22 (1H, q, J 6.9, C CH)_Z,

7.33–7.36 (6H, m, Ar–H)_{E + Z}, 7.51–7.55 (4H, m, Ar–H)_{E + Z}; d_C

(100 MHz; CDCl₃) −2.6, −2.4, −1.2, 15.1, 18.5, 28.3, 32.2, 41.5,

51.9, 52.1, 53.5, 53.9, 79.6, 127.8, 127.9, 128.9, 129.0, 133.1,

133.8, 134.0, 134.9, 138.3, 138.9, 140.7, 142.9, 155.1, 173.2,

173.3; m/z (EI⁺) 377.2006 (3%, MH⁺, C₂₀H₃₁NO₄Si requires

377.2022), 135 (100%, PhMe₂Si⁺).

O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 3 7 3 4 – 3 7 4 8

3 7 4 3

View Article Online

(2S)-2-tert-Butoxycarbonylamino-4-(dimethylphenylsilanyl)-6-

methylhept-4-enoic acid methyl ester 13b

was diluted with H₂O (15 mL) and extracted with EtOAc (3 ×

15 mL). The organic extracts were then washed with satd. aq.

NaCl (15 mL), dried (MgSO₄) and concentrated under reduced

pressure. The crude acid thus obtained was dissolved in Et₂O

(5 mL) and CH₂N₂was bubbled through the solution under a

stream of N₂until a yellow colouration persisted. Further N₂

was bubbled through the reaction to remove the excess CH₂N₂,

and the solvent was removed under reduced pressure to furnish

the crude product. Puriﬁcation by ﬂash column chromatography

(25% EtOAc–pet. ether) gave 15 (8.1 mg, 47%) as a white solid.

A

solution of rearrangement precursor (S)-9b (50 mg,

0.13 mmol, dried by azeotrope with toluene) in THF (1 mL)

was transferred by cannula into a stirred suspension of KH

(34 mg of a 30% w/w suspension in mineral oil, washed twice

with pet. ether, 0.26 mmol, 2 equiv.) in THF (1 mL) at −78 ^◦C.

18-C-6 (110 mg, 0.40 mmol, 1 equiv.) was added and the reaction

was allowed to stir at −78 ^◦C for 10 mins, before being allowed

◦

to warm to −20 C and stirred for 2 hours and then warmed

◦

Mp 62–64 C (lit.³⁶60 ^◦C); [a]_D+20.5 (c 0.4, CHCl₃), lit.³⁶

to 0 ^◦C and stirred for a further 1 hour. The reaction was

quenched with MeOH and allowed to warm to rt before being

diluted with Et₂O (20 mL). The reaction mixture was then

washed with H₂O (2 × 20 mL), satd. aq. NaCl (20 mL), dried

(MgSO₄), concentrated under reduced pressure and puriﬁed by

ﬂash column chromatography (10% EtOAc–pet. ether) to give

13b (33 mg, 66%) as a clear oil and an inseparable >10 : 1 mixture

of E and Z alkene isomers (E alkene ee 70%³³). [a]_D+8.5 (c 1.3,

CHCl₃); m_max(ﬁlm)/cm⁻¹3444, 2959, 2931, 2868, 1741, 1712; d_H

(500 MHz; CDCl₃) for E alkene 0.40 (6H, s, SiMe₂), 0.95 (3H, d,

J 6.7, CHCH₃), 0.99 (3H, d, J 6.5, CHCH₃), 1.42 (9H, s, ^tBu),

248 (1H, dd, J 13.6, 9.6, CHCH₂), 2.55 (1H, dd, J 13.6, 5.4,

CHCH₂), 3.67 (3H, s, OMe), 4.19–4.25 (2H, m, NHCH), 4.70

+30.8^◦(c 2.1, CHCl₃), therefore op 67%; (found C, 50.93; H,

7.72; N, 5.31. C₁₁H₁₉NO₆requires C, 50.57; H, 7.33; N, 5.36%.);

t

d_H(500 MHz; CDCl₃) 1.46 (9H, s, Bu), 2.83 (1H, dd, J 16.9,

4.6, CH₂), 3.01 (1H, dd, J 17.0, 4.4, CH₂), 3.71 (3H, s, OMe),

3.77 (3H, s, OMe), 4.59 (1H, dt, J 8.4, 4.2, CH), 5.49 (1H, d, J

7.6, NH).

{tert-Butoxycarbonyl-[(1S)-1-(tert-butyldiphenylsilanyloxy-

methyl)-2-(dimethylphenylsilanyl)-allyl]-amino}-acetic acid

methyl ester (S)-16

A solution of N-Boc-L-serine (2.50 g, 12.2 mmol) in DMF

(25 mL) was treated with imidazole (2.16 g, 31.7 mmol, 2.6

equiv.) followed by TBDPS–Cl (7.92 mL, 30.5 mmol, 2.5 equiv.)

and stirred at rt for 16 h. The reaction mixture was diluted with

EtOAc (70 mL), washed with H₂O (3 × 70 mL), followed by

satd. aq. NaCl (70 mL), dried (MgSO₄) and concentrated under

reduced pressure. The viscous oil thus obtained was suspended

in MeOH–H₂O (43 mL, 70 : 30), treated with K₂CO₃(4.24 g,

30.5 mmol, 2.5 equiv.) and stirred at rt for 2 h. The MeOH was

removed under reduced pressure and the remaining aqueous

portion was acidiﬁed to pH 1 with 10% citric acid solution before

being extracted with CH₂Cl₂(3 × 40 mL). The organic phases

were dried (MgSO₄) and concentrated under reduced pressure

to give the crude product. The crude product was dissolved in

the minimum quantity of hot EtOAc, diluted to 70 mL with

pet. ether and left to recrystallise overnight at rt. Filtration gave

N–Boc–O–BDPS–(S)-serine (4.94 g, 94%) as a white crystalline

=

(1H, d, J 7.9, NH), 5.78 (1H, d, J 9.7, C CH), 7.33–7.36 (3H,

m, Ar–H), 7.50–7.53 (2H, m, Ar–H); d_C(100 MHz; CDCl₃) for

E alkene −2.3, −2.1, 22.7, 22.8, 27.9, 28.4, 32.6, 52.1, 53.6, 79.7,

127.9, 128.1, 129.0, 130.2, 134.1, 138.6, 154.0, 155.1, 173.4; m/z

(CI⁺) 406.2409 (7%, MH⁺, C₂₂H₃₆NO₄Si requires 406.2414), 272

(100%, MH₂− PhMe₂Si), 135 (12%, PhMe₂Si⁺).

+

In an identical fashion (R)-9b (50 mg, 0.13 mmol) was

rearranged to give 13b (25 mg, 50%) as a clear oil and an

inseparable >10 : 1 mixture of E and Z alkene isomers (E

alkene ee 70%³³). [a]_D−9.2 (c 1.2, CHCl₃); m/z (FAB) 406.2411

(10%, MH⁺, C₂₂H₃₆NO₄Si requires 406.2414). All other data

were identical to that detailed above.

(2S, 5R,S)-2-tert-Butoxycarbonylamino-5-hydroxy-6-methyl-4-

oxo-heptanoic acid methyl ester 14

◦

solid. Mp 157–158 C (lit.⁴³155–156 C); (found C, 64.87; H,

7.46; N, 3.05. C₂₄H₃₃NO₅Sirequires C, 64.98;H, 7.50; N, 3.16%.);

d_H(400 MHz; CDCl₃) 1.07 (9H, s, Si^tBu), 1.48 (9H, s, O^tBu), 3.91

(1H, dd, J 9.9, 3.3, CHCH₂), 4.13 (1H, dd, J 9.6, 1.8, CHCH₂),

4.43–4.46 (1H, bm, NHCH), 5.37 (1H, d, J 8.1, NH), 7.41–7.45

(6H, m, Ar–H), 7.63–7.66 (4H, m, Ar–H).

Ozone was bubbled through a solution of (E,S)-13b (formed by

rearrangement of (S)-9b) (45 mg, 0.11 mmol) in MeOH–CH₂Cl₂

(4.4 mL, 9 : 1) at −78 ^◦C until a blue colouration persisted; O₂

was then bubbled through the solution to remove the excess

ozone until it again became clear. After the reaction had been

allowed to warm to rt, H₂O (1 mL) was added. The reaction was

stirred at rt for 18 h before being diluted with EtOAc (20 mL) and

washed with H₂O (20 mL) followed by satd. aq. NaCl (20 mL).

The organic portion was dried (MgSO₄), concentrated under

reduced pressure and puriﬁed by ﬂash column chromatography

(30% EtOAc–pet. ether) to give 14 (22 mg, 66%) as a clear oil

and a 1 : 1.5 mixture of inseparable diastereoisomers. [a]_D+30.0

(c 2.0, CHCl₃); m_max(ﬁlm)/cm⁻¹3527, 3439, 2971, 2932, 2877,

1714; d_H(400 MHz; CDCl₃) 0.93–0.98 (15H, m, CH[CH₃]₂), 1.44

(22.5H, s, ^tBu), 1.97–2.06 (2.5H, m, CH[CH₃]₂), 3.03 (1H, dd, J

In an identical procedure to the preparation of (S)-5a, N–

Boc–O–TBDPS–(S)–serine (9.38 g, 21.7 mmol) was converted

to its corresponding acylimidazole (10.45 g, quant.) as a viscous

oil. [a]_D+9.3 (c 3.8, CHCl₃); m_max(ﬁlm)/cm⁻¹3446, 2932, 2859,

1704; d_H(400 MHz; CDCl₃) 0.96 (9H, s, Si^tBu), 1.45 (9H, s,

O^tBu), 3.91–3.99 (2H, m, CHCH₂), 5.08 (1H, dt, J 8.1, 4.0,

=

NHCH), 5.57 (1H, bs, NH), 7.09 (1H, s, CONCH CH), 7.33–

=

7.49 (11H, m, Ar–H, CONCH CH), 8.21 (1H, s, NCH N);

d_C(100 MHz; CDCl₃) 19.1, 26.6, 28.3, 52.5, 64.4, 80.8,

116.2, 127.9, 130.1, 132.0, 132.2, 135.3, 135.5, 136.5, 155.1,

168.4.

18.4, 4.4, CHCH₂)_min, 3.18 (1.5H, dd, J 18.2, 4.0, CHCH₂)_maj

,

3.37 (1.5H, dd, J 18.5, 4.6, CHCH₂)_maj, 3.46 (1H, dd, J 18.2, 4.0,

CHCH₂)_min, 3.75 (4.5H, s, OMe)_maj, 3.76 (3H, s, OMe)_min, 4.09

(1.5H, d, J 6.5, CHOH)_maj, 4.11 (1H, d, J 6.2, CHOH)_min, 4.57–

4.61 (2.5H, m, NHCH), 5.54–5.58 (2.5H, m, NH), 9.23 (1.5H, s,

OH)_maj, 9.34 (1H, s, OH)_min; d_C(100 MHz; CDCl₃) 19.3, 26.8,

28.4, 42.0, 42.1, 49.3, 52.8, 62.2, 80.3, 90.3, 90.3, 128.0, 130.1,

132.5, 132.6, 132.7, 135.6, 135.7, 155.7, 172.0, 172.2, 207.4,

207.9. m/z (CI⁺) 304.1747 (6%, MH⁺, C₁₄H₂₆NO₆Si requires

304.1760), 57 (35%, ^tBu⁺).

In an identical procedure to the preparation of (S)-3a, except

that the reaction was maintained at under −80 ^◦C using a CO₂–

Et₂O bath, the acylimidazole from above (980 mg, 1.99 mmol)

was converted to its corresponding acyl silane (277 mg, 25%) as

a pale oil. [a]_D+28.5 (c 2.9, CHCl₃); m_max(ﬁlm)/cm⁻¹3435, 2932,

2859, 1705, 1646; d_H(400 MHz; CDCl₃) 0.48 (3H, s, SiMe₂Ph),

0.48 (3H, s, SiMe₂Ph), 0.97 (9H, s, Si^tBu), 1.43 (9H, s, O^tBu),

3.77 (1H, dd, J 10.9, 3.2, CHCH₂), 3.94 (1H, dd, J 10.9, 2.9,

CHCH₂), 4.47 (1H, dt, J 7.9, 3.0, NHCH), 5.53 (1H, d, J 7.8,

NH), 7.33–7.52 (15H, m, Ar–H); d_C(100 MHz; CDCl₃) −4.5,

−4.1, 19.3, 26.8, 28.4, 62.3, 66.0, 79.5, 127.7, 128.3, 129.8, 130.1,

132.9, 133.1, 133.9, 134.1, 135.7, 155.2. m/z (CI⁺) 562.2836 (5%,

MH⁺, C₃₂H₄₄NO₄Si₂requires 562.2809), 135 (87%, PhMe₂Si⁺),

57 (100%, ^tBu⁺).

(2S)-2-tert-Butoxycarbonylamino-succinic acid dimethyl ester 15

A solution of ketol 14 (20 mg, 0.070 mmol) in MeOH (1 mL)

was treated with a solution of NaIO₄(140 mg, 0.66 mmol,

10 equiv.) in H₂O (0.5 mL) and stirred at rt for 42 h. The reaction

3 7 4 4

O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 3 7 3 4 – 3 7 4 8

View Article Online

In an identical procedure to the preparation of (S)-2a, the

acyl silane was converted to its corresponding vinyl silane (35%

plus 22% starting material) as a clear oil. [a]_D−2.5 (c 1.1,

CHCl₃); m_max(ﬁlm)/cm⁻¹3442, 2960, 2932, 2895, 2859, 1709; d_H

(400 MHz; CDCl₃) 0.34 (6H, s, SiMe₂), 1.02 (9H, s, Si^tBu),

1.43 (9H, s, O^tBu), 3.38 (1H, dd, J 10.2, 6.8, CHCH₂), 3.60

(1H, dd, J 10.5, 4.5, CHCH₂), 4.41 (1H, bs, NHCH), 4.71

{tert-Butoxycarbonyl-[(1R)-1-(tert-butyldiphenylsilanyloxy-

methyl)-2-(dimethylphenylsilanyl)-allyl]-amino}-acetic acid

methyl ester (R)-16

As the enantiomer of the rearrangement precursor was only

required for HPLC studies, it was prepared from the O-TBDMS

vinyl silane (from Tebbe methylenation of the acyl silane derived

from N-Boc-O-TBDMS-(R)-serine) by deprotection with TBAF

and reprotection with TBDPSCl. The remainder of the synthesis

from the O-TBDPS vinyl silane was completed as above.

In an identical procedure to the preparation of (S)-5a, N-

Boc-O-TBS-(R)-serine (8.10 g, 25.3 mmol) was converted to its

corresponding acylimidazole (7.96 g, 85%) as a viscous oil. [a]_D

−8.1 (c 1.8, CHCl₃); m_max(ﬁlm)/cm⁻¹3441, 2930, 2884, 2857 (C–

H), 1704; d_H(500 MHz; CDCl₃) −0.05 (3H, s, SiMe₂), −0.04

(3H, s, SiMe₂), 0.78 (9H, s, Si^tBu), 1.45 (9H, s, O^tBu), 3.87 (1H,

dd, J 9.9, 6.2, CHCH₂), 3.99 (1H, dd, J 9.9, 4.3, CHCH₂),

5.00–5.06 (1H, m, NHCH), 5.46 (1H, d, J 8.2, NH), 7.11

=

(1H, d, J 8.3, NH), 5.54 (1H, d, J 0.8, C CH₂), 5.86 (1H, s,

=

C CH₂), 7.25–7.59 (15H, m, Ar–H); d_C(100 MHz; CDCl₃)

−2.7, −2.5, 19.3, 26.9, 28.5, 56.4, 65.6, 79.0, 127.6, 127.8, 127.9,

129.1, 129.7, 129.8, 133.4, 133.9, 135.7, 137.9, 148.3, 155.3. m/z

(FAB): 560.3040 (19%, MH⁺, C₃₃H₄₆NO₃Si₂requires 560.3016),

460 (40%, MH⁺− Boc), 135 (100%, PhMe₂Si⁺).

The N-Boc protected amine from above (210 mg, 0.375 mmol)

◦

in CH₂Cl₂(1.5 mL) was cooled to 0 C and tre_◦ated with TFA

(0.5 mL). The reaction was allowed to stir at 0 C for 30 mins

before being quenched by the cautious addition of 10% aq.

K₂CO₃. The reaction mixture was then diluted with EtOAc

(20 mL), washed with H₂O (20 mL), followed by satd. aq. NaCl

(20 mL), dried (MgSO₄), concentrated under reduced pressure

and puriﬁed by ﬂash column chromatography (40% EtOAc–pet.

ether) to give the primary amine as a clear oil (140 mg, 81%).

[a]_D+3.4 (c 1.1, CHCl₃); m_max(ﬁlm)/cm⁻¹3389, 2932, 2858; d_H

(500 MHz; CDCl₃) 0.28 (3H, s, SiMe₂), 0.31 (3H, s, SiMe₂), 1.04

(9H, s, ^tBu), 1.72 (2H, bs, NH₂), 3.29 (1H, t, J 9.5, CHCH₂), 3.57

(1H, dd, J 10.0, 3.4, CHCH₂), 3.71 (1H, dd, J 8.7, 3.2, NH₂CH),

=

(1H, s, CONCH CH), 7.53 (1H, s, CONCH CH), 8.25 (1H, s,

NCH N); d_C(125 MHz; CDCl₃) −5.7, −5.4, 18.1, 25.6, 28.3,

=

55.4, 64.1, 80.9, 116.3, 131.2, 136.7, 155.1, 168.8.

In an identical procedure to the preparation of (S)-3a, the

acylimidazole from above (7.80 g, 21.1 mmol) was converted to

its corresponding acyl silane (2.4 g, 26%) as a pale oil. [a]_D−54.2

(c 1.7, CHCl₃); m_max(ﬁlm)/cm⁻¹3430, 2954, 2930, 2885, 2858,

t

1704, 1645; d_H(500 MHz; CDCl₃) −0.08 (3H, s, SiMe₂Bu),

t

=

−0.06 (3H, s, SiMe₂Bu), 0.50 (3H, s, SiMe₂Ph), 0.53 (3H, s,

5.54 (1H, d, J 2.2, C CH₂), 5.98 (1H, s, C CH₂), 7.29–7.63

(15H, m, Ar–H); d_C(125 MHz; CDCl₃) −2.5, −2.3, 19.3, 26.9,

57.3, 68.9, 127.3, 127.7, 127.8, 129.0, 129.7, 133.5, 133.7, 133.9,

135.6, 135.6, 138.2, 150.5; m/z (FAB): 460.2530 (28%, MH⁺,

C₂₈H₃₈NOSi₂requires 460.2492), 135 (100%, PhMe₂Si⁺).

SiMe₂Ph), 0.79 (9H, s, Si^tBu), 1.43 (9H, s, O^tBu), 3.77 (1H, dd,

J 10.8, 3.1, CHCH₂), 3.97 (1H, dd, J 10.8, 2.6, CHCH₂), 4.38

(1H, dt, J 7.3, 3.0, NHCH), 5.52 (1H, d, J 7.0, NH), 7.31–

7.43 (3H, m, Ar–H), 7.54–7.59 (2H, m, Ar–H); d_C(125 MHz;

CDCl₃) −5.6, −4.4, −4.0, 18.2, 25.8, 28.4, 61.6, 66.2, 79.5, 128.3,

130.1, 134.1, 155.3, 240.1; m/z (FAB) 438.2490 (16%, MH⁺,

C₂₂H₄₀NO₄Si₂requires 438.2496), 338 (32%, MH⁺− Boc), 135

(100%, PhMe₂Si⁺).

In an identical procedure to the preparation of (S)-11a,

the primary amine from above (796 mg, 1.73 mmol) was

alkylated with methyl bromoacetate to give the corresponding

ester (731 mg, 80%) as a clear oil. [a]_D−17.5 (c 1.2, CHCl₃);

m_max(ﬁlm)/cm⁻¹3338, 2954, 2931, 2858, 1738; d_H(500 MHz;

CDCl₃) 0.23 (3H, s, SiMe₂), 0.26 (3H, s, SiMe₂), 1.04 (9H, s,

^tBu), 2.45 (1H, bs, NH), 3.07 (1H, d, J 17.4, NHCH₂), 3.27 (1H,

d, J 17.2, NHCH₂), 3.37 (1H, t, J 10.6, CHCH₂), 3.44–3.47

(2H, m, CHCH₂, NHCH), 3.66 (3H, s, OMe), 5.55 (1H, d, J

In an identical procedure to the preparation of (S)-2a, the

acyl silane from above (2.03 g, 4.65 mmol) was converted to its

corresponding vinyl silane (810 mg, 40%) as a pale oil along

with recovered starting material (297 mg, 15%). [a]_D+9.5 (c

1.0, CHCl₃); m_max(ﬁlm)/cm⁻¹3445 (N–H), 2955, 2930, 2885,

t

=

2857, 1707; d_H(500 MHz; CDCl₃) −0.07 (3H, s, SiMe₂Bu),

2.9, C CH₂), 6.00 (1H, d, J 2.8, C CH₂), 7.25–7.63 (15H, m,

Ar–H); d_C(125 MHz; CDCl₃) −2.8, −2.8, 19.2, 26.9, 48.5, 51.6,

64.2, 67.6, 127.7, 129.0, 129.5, 129.7, 129.7, 133.4, 133.6, 133.9,

135.7, 137.9, 148.1, 172.9. m/z (FAB): 532.2741 (55%, MH⁺,

C₃₁H₄₂NO₃Si₂requires 532.2703), 135 (100%, PhMe₂Si⁺).

t

−0.05 (3H, s, SiMe₂Bu), 0.42 (3H, s, SiMe₂Ph), 0.43 (3H, s,

SiMe₂Ph), 0.84 (9H, s, Si^tBu), 1.43 (9H, s, O^tBu), 3.42–3.47 (1H,

m, CHCH₂), 3.55 (1H, dd, J 10.3, 4.4, CHCH₂), 4.31 (1H, bs,

=

NHCH), 4.74 (1H, bs, NH), 5.56 (1H, s, C CH₂), 5.89 (1H, s,

C CH₂), 7.33–7.37 (3H, m, Ar–H), 7.52–7.54 (2H, m, Ar–H);

=

A solution of the ester from above (715 mg, 1.35 mmol)

in Et₃N (1.35 mL) was treated with Boc anhydride (461 mg,

2.02 mmol, 1.5 equiv.) and heated to reﬂux. After 16 h, further

Boc anhydride was added (154 mg, 0.670 mmol, 0.5 equiv.) and

the reaction was reﬂuxed for a further 24 hours. After this time

the reaction was allowed to cool to rt and concentrated under

reduced pressure, to give the crude product that was puriﬁed

by ﬂash column chromatography (6% EtOAc–pet. ether) to give

the rearrangement precursor (S)-16 (791 mg, 93%) a clear oil

and a 0.6 : 0.4 mixture of Boc rotamers (>95% ee³⁸). [a]_D

−38.7 (c 0.5, CHCl₃); m_max(ﬁlm)/cm⁻¹2954, 2930, 2857, 1753,

1692; d_H(400 MHz; CDCl₃) 0.39–0.41 (6H, m, SiMe₂), 1.04

(5.4H, s, Si^tBu), 1.06 (3.6H, s, Si^tBu)_rot, 1.39 (3.6H, s, O^tBu)_rot,

1.44 (5.4H, s, O^tBu), 3.36 (0.6H, d, J 17.5, NCH₂), 3.57–3.65

(1.4H, m, NCH₂), 3.60 (1.2H, s, OMe)_rot, 3.61 (1.8H, s, OMe),

3.75–3.81 (2H, m, CHCH₂), 4.94 (0.4H, t, J 6.8, NCH)_rot,

d_C(125 MHz; CDCl₃) −5.4, −2.5, −2.3, 18.3, 25.9, 28.5, 56.2,

64.7, 78.2, 127.5, 127.9, 129.2, 134.1, 138.1, 148.4, 155; m/z

(FAB) 436.2688 (25%, MH⁺, C₂₃H₄₂NO₃Si₂requires 436.2703),

336 (40%, MH⁺− Boc), 135 (100%, PhMe₂Si⁺).

A solution of the vinyl silane above (705 mg, 1.62 mmol)

in THF (14 mL) was treated with TBAF (2.43 mL of a 1 M

solution in THF, 2.43 mmol, 1.5 equiv.). After stirring for 5 mins,

the reaction was quenched by the addition of pH 7 phosphate

buffer solution (15 mL). The reaction mixture was washed

with H₂O (20 mL), followed by satd. aq. NaCl (20 mL), dried

(MgSO₄), concentrated under reduced pressure and puriﬁed by

ﬂash column chromatography (30% EtOAc–pet. ether), which

gave the alcohol (515 mg, 99%) as a white solid. Mp 99–

101 ^◦C; [a]_D+27.2 (c 1.2, CHCl₃); (found C, 63.35; H, 8.31;

N, 4.43. C₁₇H₂₇NO₃Si requires C, 63.51; H, 8.47; N, 4.36%.);

m_max(ﬁlm)/cm⁻¹3580, 3440, 2962, 1707; d_H(500 MHz; CDCl₃)

0.46 (3H, s, SiMe₂), 0.46 (3H, s, SiMe₂), 1.43 (9H, s, ^tBu), 2.29

(1H, bs, OH), 3.42–3.56 (1H, bm, CHCH₂), 3.51–3.55 (1H, bm,

CHCH₂), 4.35–4.39 (1H, bm, NHCH), 4.92 (1H, bs, NH), 5.64

=

5.18 (0.6H, t, J 6.7, NCH), 5.64 (0.6H, s, C CH₂), 5.73

(0.4H, s, C CH₂)_rot, 5.86 (0.6H, s, C CH₂), 5.99 (0.4H, s,

=

C CH₂)_rot, 7.32–7.66 (15H, m, Ar–H); d_C(100 MHz; CDCl₃)

−3.3, −3.1, −2.8, 19.2, 19.3, 16.8, 26.9, 28.3, 44.7, 44.9, 51.6,

51.8, 57.5, 58.7, 62.5, 127.7, 127.8, 127.9, 129.0, 129.1, 129.2,

129.6, 129.7, 129.8, 134.9, 135.6, 135.7, 137.6, 137.9, 145.6,

146.4, 154.9, 155.2, 170.6, 171.0; m/z (FAB): 632.3234 (8%,

MH⁺, C₃₆H₅₀NO₅Si₂requires 632.3228), 532 (48%, MH⁺− Boc),

135 (100%, PhMe₂Si⁺).

=

(1H, s, C CH₂), 5.91 (1H, s, C CH₂), 7.34–7.38 (3H, m, Ar–

H), 7.52–7.54 (2H, m, Ar–H); d_C(125 MHz; CDCl₃) −2.9, −2.6,

28.4, 56.6, 65.1, 79.6, 127.7, 128.1, 129.4, 133.8, 137.4, 148.2,

156.0; m/z (FAB) 322.1847 (10%, MH⁺, C₁₇H₂₈NO₃Si₂requires

+

322.1838), 188 (100%, MH₂− PhMe₂Si).

O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 3 7 3 4 – 3 7 4 8

3 7 4 5

View Article Online

A solution of the alcohol above (487 mg, 1.52 mmol) in

DMF (5.0 mL) was treated with imidazole (109 mg, 1.60 mmol,

1.05 equiv.) followed by TBDPS–Cl (419 lL, 1.60 mmol,

1.05 equiv.) and allowed to stir at rt for 16 hours. The reaction

was then diluted with EtOAc (20 mL), washed with H₂O

(3 × 20 mL), followed by satd. aq. NaCl (20 mL), dried

(MgSO₄), concentrated under reduced pressure and puriﬁed by

ﬂash column chromatography (4% EtOAc–pet. ether) to give

O-TBDPS vinyl silane (755 mg, 89%) as a clear oil (1.25 g,

93%). [a]_D+2.5 (c 1.1, CHCl₃); m_max(ﬁlm)/cm⁻¹3442, 2960,

2932, 2895, 2859, 1709; d_H(400 MHz; CDCl₃) 0.34 (6H, s,

SiMe₂), 1.02 (9H, s, Si^tBu), 1.43 (9H, s, O^tBu), 3.38 (1H, dd,

J 10.2, 6.8, CHCH₂), 3.60 (1H, dd, J 10.5, 4.5, CHCH₂), 4.41

(1H, bs, NHCH), 4.71 (1H, d, J 8.3, NH), 5.54 (1H, d, J

0.8, C CH₂), 5.86 (1H, s, C CH₂), 7.25–7.59 (15H, m, Ar–

H); d_C(100 MHz; CDCl₃) −2.7, −2.5, 19.3, 26.9, 28.5, 56.4,

65.6, 79.0, 127.6, 127.8, 127.9, 129.1, 129.7, 129.8, 133.4, 133.9,

135.7, 137.9, 148.3, 155.3; m/z (FAB) 560.2989 (4%, MH⁺,

C₃₃H₄₆NO₃Si₂requires 560.3016), 460 (96%, MH⁺− Boc), 135

(100%, PhMe₂Si⁺).

(FAB) 632.3257 (4%, MH⁺, C₃₆H₅₀NO₅Si₂requires 632.3228),

532 (17%, MH⁺− Boc), 135 (100%, PhMe₂Si⁺).

In an identical fashion (R)-16 (50 mg, 0.080 mmol) was

rearranged to give 17 (24 mg, 58%) as a clear oil and an

inseparable 14 : 1 mixture of E and Z alkene isomers (E alkene,

ee 40%³⁹). [a]_D−6.8 (c 2.0, CHCl₃); m/z (FAB) 632.3248 (2%,

MH⁺, C₃₆H₅₀NO₅Si₂requires 632.3228), 532 (8%, MH⁺− Boc),

135 (100%, PhMe₂Si⁺). All other data identical to that described

above.

(2S)-2-tert-Butoxycarbonylamino-succinic acid dimethyl ester 15

In an identical procedure to the preparation of 14, vinyl

silane (E,S)-17 (formed by rearrangement of (S)-16) (41 mg,

0.070 mmol) was converted to its corresponding 1,2-ketol

(18 mg, 64%) as a clear oil and a 1 : 1 mixture of inseparable

diastereoisomers. [a]_D+4.3 (c 1.5, CHCl₃); m_max(ﬁlm)/cm⁻¹3441,

2932, 2860, 1747, 1714; d_H(400 MHz; CDCl₃) 1.04 (9H s,

Si^tBu), 1.05 (9H s, Si^tBu), 1.43 (18H, s, O^tBu), 3.10–3.53 (4H, m,

NHCHCH₂), 3.70 (3H, s, OMe), 3.72 (3H, s, OMe), 3.95–4.04

(4H, m, CH₂O), 4.43–4.47 (2H, m, CHOH), 4.61–4.65 (2H,

m, NHCH), 5.47–5.51 (2H, m, NH), 7.39–7.46 (6H, m, Ar–

H), 7.64–7.69 (4H, m, Ar–H), 9.10 (2H, s, OH); d_C(100 MHz;

CDCl₃) 19.3, 26.8, 28.4, 42.0, 42.1, 49.3, 52.8, 62.2, 80.3, 90.3,

128.0, 130.1, 132.5, 132.6, 132.7, 135.6, 135.7, 155.7, 172.0,

172.2, 207.4, 207.9; m/z (CI⁺) 530.2567 (5%, MH⁺, C₂₈H₄₀NO₇Si

requires 530.2574), 430 (26%, MH⁺− Boc), 57 (37%, ^tBu⁺).

In an identical procedure to the preparation of 15, the 1,2-

ketol from above (10 mg, 0.02 mmol) was converted to aspartic

=

Following the procedure used in the synthesis of (S)-16,

N-Boc protected O-TBDPS vinyl silane from above (730 mg,

1.30 mmol) was converted into its corresponding primary amine

(558 mg, 93%) as a clear oil. [a]_D−3.8 (c 1.5, CHCl₃); m/z

(CI⁺) 460.2498 (34%, MH⁺, C₂₈H₃₈NOSi₂requires 460.2492),

t

406 (83%, M⁺− Bu), 135 (92%, PhMe₂Si⁺). All other data were

identical to the enantiomer prepared in the synthesis of (S)-16.

Following the procedure used in the synthesis of (S)-16, the

primary amine from above (556 mg, 1.21 mmol) was alkylated

with methyl bromoacetate to give the ester (508 mg, 80%) as a

clear oil. [a]_D+16.0 (c 1.0, CHCl₃); m/z (FAB⁺) 532.2687 (88%,

MH⁺, C₃₁H₄₂NO₃Si₂requires 532.2703), 135 (100%, PhMe₂Si⁺).

All other data were identical to the enantiomer prepared in the

synthesis of (S)-16.

◦

acid derivative 15 (3 mg, 60%) as a white solid. Mp 58–60 C,

(lit.³⁶60 ^◦C); [a]_D+9.3 (c 0.3, CHCl₃), lit.³⁶+30.8 (c 2.1, CHCl₃),

therefore op 30%; all other data in agreement with previous

sample and the literature.³⁶

{tert-Butoxycarbonyl-[(1R)-2-(dimethylphenylsilanyl)-1-meth-

oxymethoxymethylallyl]-amino}-acetic acid methyl ester (R)-18

Following the procedure used in the synthesis of (S)-16, the

secondary amine from above (490 mg, 0.92 mmol) was Boc

protected to give the rearrangement precursor (R)-16 (547 mg,

94%) as a clear oil and a 0.6 : 0.4 mixture of Boc rotamers

(>95% ee³⁸). [a]_D+34.2 (c 1.5, CHCl₃). m/z (FAB) 632.3221

(7%, MH⁺, C₃₆H₅₀NO₅Si₂requires 632.3228), 532 (58%, MH⁺−

Boc), 135 (100%, PhMe₂Si⁺). All other data were identical to

that for (S)-16.

A solution of TBDPS ether (R)-16 (299 mg, 0.470 mmol) in

THF (2.0 mL) and pyridine (1.0 mL) was cooled to 0 ^◦C

and treated with HF·pyridine (250 lL of a 70% w/w solution,

◦

9.40 mmol, 20 equiv.). The reaction was stirred at 0 C for 15

mins before being allowed to warm to rt and stirred for an

additional 18 hours. The reaction was then diluted with H₂O

(10 mL) before being quenched by the addition of satd. aq.

NaHCO₃and extracted with EtOAc (2 × 15 mL). The combined

organics were washed with satd. aq. CuSO₄(15 mL), dried

(MgSO₄), concentrated under reduced pressure and puriﬁed by

ﬂash column chromatography (16% EtOAc–pet. ether), which

gave recovered starting material (4 mg, 2%), cyclised product

(R)-20 (30 mg, 18%) and the desired alcohol (R)-21 (120 mg,

65%) as a clear oil and a 1 : 1 mixture of Boc rotamers. [a]_D

+51.4 (c 1.4, CHCl₃); m_max(ﬁlm)/cm⁻¹3478, 2956, 1737, 1682; d_H

(500 MHz; CDCl₃) 0.40 (3H, s, SiMe₂), 0.41 (1.5H, s, SiMe₂),

(2S)-2-tert-Butoxycarbonylamino-6-(tert-butyldiphenylsilanyl-

oxy)-4-(dimethylphenylsilanyl)-hex-4-enoic acid methyl ester

(E,S)-17

A

solution of rearrangement precursor (S)-16 (50 mg,

0.080 mmol, dried by azeotrope with toluene) in THF–HMPA

(1.5 mL, 4 : 1) was cooled to 0 ^◦C before being treated dropwise

with LDA (148 lL of a 1 M solution in THF and hexane,

◦

t

0.144 mmol, 1.8 equiv.). After stirring for 10 mins at 0 C, the

0.44 (1.5H, s, SiMe₂),1.40 (4.5H, s, Bu), 1.44 (4.5H, s, Bu),

3.17 (0.5H, d, J 17.8, NCH₂), 3.22 (0.5H, d, J 18.0, NCH₂),

3.27 (0.5H, d, J 17.7, NCH₂), 3.52 (0.5H, d, J 17.7, NCH₂),

3.54–3.60 (1H, m, CHCH₂), 3.71 (1.5H, s, OMe), 3.73 (1.5H, s,

OMe), 3.73–3.81 (1.5H, m, CHCH₂, OH), 4.02 (0.5H, dd, J

11.0, 3.4, CHCH₂), 4.90–4.93 (0.5H, m, NCH), 5.07–5.10 (0.5H,

reaction was quenched by pouring into pH 7 phosphate buffer

solution (15 mL). The mixture thus obtained was extracted

with Et₂O (2 × 15 mL), the combined organics were washed

with H₂O (20 mL), followed by satd. aq. NaCl (20 mL), dried

(MgSO₄) concentrated under reduced pressure and puriﬁed by

ﬂash column chromatography (7.5% EtOAc–pet. ether) to give

(E,S)-17 (33 mg, 66%) as a clear oil and an inseparable 14 : 1

mixture of E and Z alkene isomers (E alkene ee 40%¹⁷). [a]_D

+7.2 (c 1.2, CHCl₃); m_max(ﬁlm)/cm⁻¹3436, 2955, 2932, 2859,

1741, 1711; d_H(400 MHz; CDCl₃) E-alkene 0.40 (6H, s, SiMe₂),

1.06 (9H, s, Si^tBu), 1.37 (9H, s, O^tBu), 2.32 (2H, bd, J 7.5,

NHCHCH₂), 3.57 (3H, s, OMe), 4.10 (1H, q, J 7.7, NHCH),

4.27 (1H, dd, J 13.9, 5.8, CH₂O), 4.32 (1H, dd, J 13.8, 5.6,

=

m, NCH), 5.64 (0.5H, t, J 1.3, C CH₂), 5.65 (0.5H, t, J 1.3,

C CH₂), 5.69 (0.5H, t, J 1.4, C CH₂), 5.70 (0.5H, t, J 1.4,

=

C CH₂), 7.34–7.38 (3H, m, Ar–H), 7.45–7.48 (2H, m, Ar–H);

d_C(125 MHz; CDCl₃) −3.8, −3.2, −3.0, 28.2, 28.2, 43.9, 44.4,

52.5, 52.7, 58.0, 59.3, 61.5, 80.6, 81.2, 127.9, 128.0, 128.3, 128.5,

129.2, 129.4, 133.8, 137.0, 137.5, 147.7, 148.0, 155.0, 155.7,

173.8, 173.8; m/z (CI⁺): 394.2038 (17%, MH⁺, C₂₀H₃₂NO₅Si

requires 394.2050), 294 (35%, MH⁺− Boc), 262 (100%, MH⁺−

Boc, -MeOH).

=

CH₂O), 4.81 (1H, d, J 7.9, NH), 6.12 (1H, t, J 5.6, C CH),

7.35–7.68 (15H, m, Ar–H); d_C(100 MHz; CDCl₃) E-alkene −2.7,

−2.6, 19.2, 26.9, 28.4, 32.8, 52.1, 53.4, 61.3, 79.7, 127.8, 127.9,

129.2, 129.8, 133.6, 134.1, 135.7, 137.4, 144.7, 155.1, 173.0; m/z

A solution of the primary alcohol (R)-21 from above (120 mg,

0.310 mmol) in CH₂Cl₂(0.9 mL) was treated with Hunig’s base

(211 lL, 1.22 mmol, 4 equiv.) followed by MOM–Cl (93 lL,

3 7 4 6

O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 3 7 3 4 – 3 7 4 8

View Article Online

1.2 mmol, 4 equiv.) The reaction was stirred at rt for 3 h before

the reaction was quenched with H₂O and extracted with CH₂Cl₂

(2 × 15 mL). The combined organics were washed with satd.

aq. NaCl, dried (MgSO₄), concentrated under reduced pressure

and puriﬁed by ﬂash column chromatography (15% EtOAc–

pet. ether) to give the rearrangement precursor (R)-18 (63 mg,

47%) as a clear oil and a 0.6 : 0.4 mixture of Boc rotamers

(87% ee⁴⁰). [a]_D+37.6 (c 2.0, CHCl₃); m_max(ﬁlm)/cm⁻¹2953, 1754,

1694; d_H(400 MHz; CDCl₃) 0.42 (3H, s, SiMe₂), 0.43 (3H, s,

SiMe₂), 1.41 (5.4H, s, ^tBu), 1.44 (3.6H, s, ^tBu)_rot, 3.22 (0.6H, d, J

17.7, NCH₂), 3.30 (1.2H, s, CHOMe)_rot, 3.32 (1.8H, s, CHOMe),

3.42 (0.6H, d, J 17.8, NCH₂), 3.54 (0.4H, d, J 17.4, NCH₂)_rot,

3.62 (0.4H, d, J 17.4, NCH₂)_rot, 3.64 (1.8H, s, CO₂Me), 3.66

(1.2H, s, CO₂Me)_rot, 3.70–3.78 (2H, m, CHCH₂), 4.48–4.53 (2H,

m, OCH₂O), 4.92 (0.4H, t, J 6.0, NCH)_rot, 5.13 (0.6H, t, J 5.6,

A solution of (S)-21 (115 mg, 0.290 mmol, 1 equiv.) in toluene

(3 mL) was treated with TsOH (6 mg, 0.03 mmol, 0.1 equiv.),

˚

powdered 4 A molecular sieves (0.5 g) and the reaction was

heated at 50 ^◦C for 64 h. Further TsOH (15 mg, 0.080 mmol, 0.25

˚

equiv.) and powdered 4 A molecular sieves (0.5 g) were added

and the reaction was heated at 50 ^◦C for a further 48 h. The

reaction was ﬁltered through a glass sinter and the ﬁltrate was

concentrated under reduced pressure to give the crude product

that was puriﬁed by ﬂash column chromatography (20% EtOAc–

pet. ether) to give the rearrangement precursor (S)-20 as a white

solid (95 mg, 90%, presumably 87% ee from (S)-16). Mp 86–

88 ^◦C; [a]_D−18.6 (c 1.8, CHCl₃); m_max(ﬁlm)/cm⁻¹2962, 2931,

1755, 1698; d_H(400 MHz; CDCl₃, 55 ^◦C) 0.47 (3H, s, SiMe₂), 0.47

(3H, s, SiMe₂), 1.47 (9H, s, ^tBu), 3.89 (1H, d, J 18.3, NCH₂), 4.13

(1H, dd, J 11.8, 3.7, OCH₂), 4.23 (1H, dd, J 11.9, 3.9, OCH₂),

4.31 (1H, d, J 18.3, NCH₂), 4.73 (1H, bs, NCH), 5.71 (1H, t, J

=

NCH), 5.75 (1H, s, C CH₂), 5.93 (0.4H, s, C CH₂)_rot, 5.97

=

(0.6H, s, C CH₂), 7.32–7.40 (3H, m, Ar–H), 7.48–7.52 (2H,

1.4, C CH₂), 5.83 (1H, t, J 1.5, C CH₂), 7.37–7.40 (3H, m, Ar–

H), 7.51–7.54 (2H, m, Ar–H); d_C(100 MHz; CDCl₃) −2.4, 28.4,

44.7, 68.8, 81.4, 128.3, 128.7, 129.7, 134.0, 136.8, 146.0, 153.6,

167.2. m/z (FAB) 362.1791 (7%, MH⁺, C₁₉H₂₈NO₄Si requires

362.1788), 135 (46%, PhMe₂Si⁺), 57 (100%, ^tBu⁺).

m, Ar–H); d_C(100 MHz; CDCl₃) −3.6, −2.9, 28.3, 28.3, 44.9,

45.4, 51.7, 51.9, 55.6, 55.7, 55.8, 57.2, 67.1, 67.4, 80.3, 80.7,

96.7, 97.0, 127.9, 128.0, 129.1, 129.3, 133.9, 137.4, 137.8, 146.6,

147.1, 154.8, 155.1, 170.8, 171.1; m/z (CI⁺) 438.2297 (6%, MH⁺,

C₂₂H₃₆NO₆Si requires 438.2312), 338 (13%, MH⁺− Boc), 135

(13%, SiMe₂Ph⁺), 57 (17%, ^tBu⁺).

{tert-Butoxycarbonyl-[(1S)-2-(dimethyl-phenyl-silanyl)-1-hydro-

xymethyl-allyl]-amino}-acetic acid formed from attempted

rearrangement of (S)-20

(2R)-2-tert-Butoxycarbonylamino-4-(dimethylphenylsilanyl)-6-

methoxymethoxyhex-4-enoic acid methyl ester (E,S)-19

A solution of lactone (S)-20 (30 mg, 0.080 mmol, 87% ee, dried

A

solution of rearrangement precursor (R)-18 (17 mg,

by azeotrope with toluene) and 18-C-6 (42 mg, 0.17 mmol,

0.040 mmol, 87% ee,⁴⁰dried by azeotrope with toluene) and

18-C-6 (23 mg, 0.090 mmol, 2 equiv.) in THF (0.4 mL) was

cooled to 0 ^◦C before being transferred by cannula into a stirred

suspension of KH (12 mg of a 30% w/w suspension in mineral

oil, washed twice with pet. ether, 0.090 mm_◦ol, 2 equiv.) in

THF (0.4 mL). After stirring for 20 mins at 0 C, the reaction

was quenched by pouring into pH 7 phosphate buffer solution

(15 mL). The mixture thus obtained was extracted with Et₂O

(2 × 15 mL), the combined organics were washed with satd.

aq. NaCl (20 mL), dried (MgSO₄), concentrated under reduced

pressure and puriﬁed by ﬂash column chromatography (10%

EtOAc–pet. ether) to give (E,S)-19 (10 mg, 59%) as a clear oil

and an inseparable 10 : 1 mixture of E and Z alkene isomers

(E alkene ee 65%⁴²). [a]_D−3.4 (c 2.1, CHCl₃); m_max(ﬁlm)/cm⁻¹

3440, 2932, 1743, 1713; d_H(400 MHz; CDCl₃) signals for major

E alkene: 0.44 (6H, s, SiMe₂), 1.42 (9H, s, ^tBu), 2.57 (2H, bd, J

8.0, NHCHCH₂), 3.39 (3H, s, CH₂OMe), 3.67 (3H, s, CO₂Me),

4.07–4.24 (3H, m, CHCH₂O, NHCH), 4.66 (2H, s, OCH₂O),

◦

2 equiv.) in THF (0.8 mL) at 0 C was transferred by cannula

into a stirred suspension of KH (22 mg of a 30% w/w suspension

in mineral oil, washed twice with pet. ether, 0.17 mmol, 2 equiv.)

at 0 ^◦C. After 30 mins, the reaction was allowed to warm to

rt and stirred for a further 1.5 h when TLC indicated only a

trace of starting material remaining. The reaction was quenched

by pouring into a solution of 10% citric acid (15 mL) and

the resultant solution was extracted with Et₂O (2 × 10 mL).

The combined organics were washed with satd. aq. NaCl

(15 mL), dried (MgSO₄), concentrated under reduced pressure

and puriﬁed by ﬂash column chromatography (EtOAc) to give

the corresponding hydroxy acid (15 mg, 48%) as a clear oil and

a 0.6 : 0.4 mixture of Boc rotamers which rapidly recyclised

on standing at rt to regenerate the starting material (S)-20. d_H

(400 MHz; CDCl₃) 0.39–0.46 (6H, m, SiMe₂), 1.40 (5.4H, s, ^tBu),

1.42 (3.6H, s, ^tBu)_rot, 3.22 (0.6H, d, J 14.6, NCH₂), 3.25 (0.6H,

d, J 17.8, NCH₂), 3.32 (0.4H, d, J 17.9, NCH₂)_rot, 3.58–3.64

(1H, m, CHCH₂), 3.59 (0.4H, d, J 17.9, NCH₂)_rot, 3.75 (0.4H,

dd, J 12.3, 4.0, CHCH₂)_rot, 3.83 (0.6H, dd, J 12.1, 4.1, CHCH₂),

4.89–4.92 (0.4H, m, NCH)_rot, 5.05–5.09 (0.6H, m, NCH), 5.65–

=

5.34 (1H, d, J 7.6, NH), 6.15 (1H, t, J 6.2, C CH), 7.35–7.38

(3H, m, Ar–H), 7.45–7.48 (2H, m, Ar–H). Signals for minor Z

t

alkene: 0.46 (6H, s, SiMe₂), 1.43 (9H, s, Bu), 2.37 (1H, dd, J

=

5.71 (2H, m, C CH₂), 7.35–7.46 (5H, m, Ar–H); d_C(100 MHz;

13.1, 8.9, NHCHCH₂), 2.72 (1H, dd, J 13.0, 4.4, NHCHCH₂),

3.23 (3H, s, CH₂OMe), 3.69 (3H, s, CO₂Me), 3.92 (2H, d, J

6.6, CHCH₂O), 4.26–4.32 (1H, m, NHCH), 4.43 (1H, d, J 6.6,

OCH₂O), 4.45 (1H, d, J 6.6, OCH₂O), 4.88 (1H, d, J 8.2, NH),

CDCl₃) −3.8, −3.1, −3.0, −2.5, 28.2, 28.3, 43.8, 44.6, 57.4, 59.1,

61.0, 61.2, 81.4, 81.6, 128.0, 128.3, 129.3, 129.4, 133.8, 133.9,

136.9, 137.2, 147.3, 147.5, 155.4, 156.0, 174.8, 175.3. No further

data was obtained due to rapid recyclisation to (S)-20.

=

6.21 (1H, t, J 6.6, C CH), 7.35–7.38 (3H, m, Ar–H), 7.45–7.48

(2H, m, Ar–H); d_C(100 MHz; CDCl₃) −2.7, −2.5, 28.4, 32.9,

52.2, 53.1, 55.6, 79.6, 96.1, 128.0, 129.3, 134.1, 137.4, 139.9,

140.8, 155.3, 173.2; m/z (FAB) 460 (15%, MNa⁺), 438.2337 (2%,

MH⁺, C₂₂H₃₆NO₆Si requires 438.2312), 328 (6%, MH⁺− Boc),

135 (31%, PhMe₂Si⁺), 57 (100%, ^tBu⁺).

Acknowledgements

This work is part of the PhD thesis of MPW. We thank

AstraZeneca and the University of Nottingham for funding,

Mr T. Spencer for combustion analyses and Mr G. Coxhill and

Mr T. Hollingworth for providing mass spectra.

(5S)-5-[1-(Dimethylphenylsilanyl)-vinyl]-2-oxo-morpholine-4-

carboxylic acid tert-butyl ester (S)-20

References

In an identical procedure to the preparation of the primary

alcohol (R)-21 derived from (R)-16 in the synthesis of (R)-18, a

solution of TBDPS ether (S)-16 (207 mg, 0.340 mmol, 87% ee³⁸)

gave recovered starting material (S)-16 (33 mg, 16%), cyclised

product (S)-20 (12 mg, 10%) and the alcohol (S)-21 (84 mg, 63%)

as a clear oil and a 1 : 1 mixture of Boc rotamers. [a]_D−50.1

(c 2.2, CHCl₃); m/z (CI⁺) 394.2038 (17%, MH⁺, C₂₀H₃₂NO₅Si

requires 394.2050). All other data were identical to (R)-21.

1 (a) J. C. Anderson, A. Flaherty and M. E. Swarbrick, J. Org. Chem.,

2000, 65, 9152 and references therein; (b) J. C. Anderson and S.

Skerratt, J. Chem. Soc., Perkin Trans. 1, 2002, 2871–9.

2 (a) J. C. Anderson and A. Flaherty, J. Chem. Soc., Perkin Trans. 1,

2001, 267; (b) J. C. Anderson and M. Whiting, J. Org. Chem., 2003,

68, 6160–3.

3 J. C. Anderson, J. M. A. O’Loughlin and J. A. Tornos, Org. Biomol.

Chem., 2005, 3, 2741.

O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 3 7 3 4 – 3 7 4 8

3 7 4 7

View Article Online

4 (a) K. Mikami, K. Fujimoto, T. Kasuga and T. Nakai, Tetrahedron

Lett., 1984, 25, 6011; (b) K. Mikami, O. Takahashi, T. Kasuga and

T. Nakai, Chem. Lett., 1985, 1729; (c) K. Mikami, K. Fujimoto,

T. Kasuga and T. Nakai, Tetrahedron Lett., 1984, 25, 6011; (d) K.

Mikami, T. Kasuga, K. Fujimoto and T. Nakai, Tetrahedron Lett.,

1986, 27, 4185; (e) O. Takahashi, K. Mikami and T. Nakai, Chem.

Lett., 1987, 69; (f) E. Nakai and T. Nakai, Tetrahedron Lett., 1988,

29, 4587; (g) R. Bru¨ckner, Chem. Ber., 1989, 122, 193; (h) D. Enders,

D. Backhaus and J. Runsink, Angew. Chem., Int. Ed. Engl., 1994, 33,

2098; (i) D. Enders and D. Backhaus, Synlett, 1995, 631.

5 (a) J. A. Marshall and J. Lebreton, J. Am. Chem. Soc., 1988, 110,

2925; (b) J. A. Marshall and X. Wang, J. Org. Chem., 1992, 57, 2747;

(c) K. Fujimoto and T. Nakai, Tetrahedron Lett., 1994, 35, 5019.

6 (a) N. Sayo, K. Azuma, K. Mikami and T. Nakai, Tetrahedron Lett.,

1984, 25, 565; (b) D. J.-S. Tsai and M. M. Midland, J. Org. Chem.,

1984, 49, 1843; (c) N. Sayo, E.-I. Kitahara and T. Nakai, Chem. Lett.,

1984, 259.

23 A. D. Kaye and G. Pattenden, Tetrahedron Lett., 1986, 27, 2033.

24 This temperature is from our observations on attempted enantiose-

lective aza-[2,3]-Wittig rearrangements with chiral benzylic anions.

See ref. 3.

25 R. E. Gawley, Q. Zhang and S. Campagna, J. Am. Chem. Soc., 1995,

117, 11817.

26 (a) E. K. Verner and T. Cohen, J. Am. Chem. Soc., 1992, 114,

375; (b) R. Hoffmann and R. Bru¨ckner, Angew. Chem., Int. Ed.

Engl., 1992, 31, 647; (c) K. Tomooka, T. Igarishi, M. Watanabe and

T. Nakai, Tetrahedron Lett., 1992, 33, 5795; (d) K. Tomooka, T.

Igarishi, N. Komine and T. Nakai, Synth. Org. Chem. Jpn., 1995, 53,

480.

27 K. Takenouchi, M. Tabe, K. Watanabe, A. Hazato, Y. Kato, M.

Shionoya, T. Koike and E. Kimura, J. Org. Chem., 1993, 58, 6895.

28 G. W. Kabalka, N.-S. Li and R. D. Pace, Synth. Commun., 1995, 25,

2135.

29 Reactions were attempted with longer reaction times, higher con-

centrations of Boc₂O, in the presence of DMAP and higher boiling

DMF.

7 T. Nakai and K. Mikami, Org. React., 1994, 46, 105.

8 N. Sayo, F. Shirai and T. Nakai, Chem. Lett., 1984, 255.

9 For a review see: A. Ricci and A. Degl’Innocenti, Synthesis, 1989,

647.

10 B. F. Bonini, M. Comes-Franchini, G. Mazzanti, U. Passamonti, A.

Ricci and P. Zani, J. Org. Chem., 1996, 61, 7242.

30 Daicel Chiralcel OD–R column, 250 × 4.6 mm, 70 : 30 MeCN–H₂O,

0.7 mL min⁻¹, retention times for (R)-9a 14.2 mins, (S)-9a 16.0 mins

and (S)-9b 11.6, (R)-9b 16.2.

˚

31 J. Ahman and P. Somfai, Synth. Commun., 1995, 25, 2301.

11 L. A. Carpino, M. Beyermann, H. Wenschuh and M. Beinert, Acc.

Chem. Res., 1996, 29, 268.

32 R. E. Ireland and R. S. Meissner, J. Org. Chem., 1991, 56, 4566.

33 Daicel Chiralcel OD–R column, 250 × 4.6 mm, 52.5 : 47.5 MeCN–

H₂O, 0.5 mL min⁻¹, retention times (E,S)-13b 59.7 mins and (E,R)-

13b 63.6 mins.

34 G. Buchi and H. Wuest, J. Am. Chem. Soc., 1978, 100, 294.

35 (a) G. Dryhurst, Periodate Oxidation of Diols and other Functional

Groups, Pergamon Press, New York, 1970; (b) D. N. Gupta, P. Hodge

and J. E. Davies, J. Chem. Soc., Perkin Trans. 1, 1981, 2970.

36 J. M. Padron, G. Kokotos, T. Martin, T. Markidid, W. A. Gibbons

and V. S. Martin, Tetrahedron: Asymmetry, 1998, 9, 3381.

37 The corresponding TBS protected serieswas synthesised, but this suf-

fered from a difﬁcult selective Boc deprotection step that proceeded

in only 39% yield.

12 B. F. Bonini, M. Comes-Franchini, M. Fochi, F. Laborio, G.

Mazzanti, A. Ricci and G. Varchi, J. Org. Chem., 1999, 64, 8008.

13 I. Fleming, T. W. Newton and F. Roessler, J. Chem. Soc., Perkin

Trans. 1, 1981, 2527.

14 A. G. Brook, Acc. Chem. Res., 1974, 7, 77.

15 (a) J. A. Soderquist and C. L. Anderson, Tetrahedron Lett., 1988,

29, 2425; (b) J. A. Soderquist and C. L. Anderson, Tetrahedron Lett.,

1988, 29, 2777.

16 A. Fu¨rstner, G. Kolleger and H. Weidmann, J. Organomet. Chem.,

1991, 414, 295.

17 (a) F. N. Tebbe, G. W. Parshall and G. S. Reddy, J. Am. Chem. Soc.,

1978, 100, 3611; (b) S. H. Pine, Org. React., 1993, 43, 1.

18 L. Clawson, S. L. Buchwald and R. H. Grubbs, Tetrahedron Lett.,

1984, 25, 5733.

38 Daicel Chiralcel OD–R column, 250 × 4.6 mm, 80 : 20 MeCN–H₂O,

0.5 mL min⁻¹, retention times for (S)-16 18.4, (R)-16 24.0.

39 Daicel Chiralcel AD column, 150 × 2.1 mm, 99.5 : 0.5 hexane–IPA,

0.2 mL min⁻¹, retention times (E,R)-17 8.4 mins, (E,S)-17 13.3 mins.

40 Daicel Chiralcel OD–H column, 150 × 4.6 mm, 99 : 1 hexane–IPA,

0.5 mL min⁻¹, retention times (S)-18 9.9 mins, (R)-18 19.0 mins.

41 This assumption is also supported by the sign of the optical rotations

(E,R)-19 -3.4 (c 2.1, CHCl₃), (E,S)-17 +7.2 (c 1.2, CHCl₃), (E,S)-13b

+8.5 (c 1.3, CHCl₃).

19 A. N. Petasis and E. I. Bzowej, J. Am. Chem. Soc., 1978, 100, 3611.

20 A. N. Petasis and S.-P. Lu, Tetrahedron Lett., 1995, 14, 2393.

21 Due to the unsatisfactory rearrangement of (S)-6a only the enan-

tiomeric purity of 6b was determined. Daicel Chiralcel OD–H

column, 150 × 4.6 mm, 99.75 : 0.25 hexane–IPA, 0.5 mL min⁻¹

,

retention times (R)-6b 7.5 mins, (S)-6b 8.4 mins. From optical

rotation data for the single enantiomers of 5a, 3a and 2a we assumed

no racemisation occurred as the route was identical to that for the

synthesis of 6b and this was veriﬁed by later analysis of similar series.

22 Daicel Chiralcel OJ-RH column, 150 × 4.6 mm, 55 : 45 MeCN–H₂O,

0.6 mL min⁻¹, retention times for (E)-7b 29.0 mins and 31.2 mins.

42 Daicel Chiralcel OD–H column, 150 × 4.6 mm, 99.25 : 0.75 hexane–

IPA, 0.5 mL min⁻¹, retention times (E,R)-19 14.6 mins, (E,S)-19 16.1

mins.

43 J. S. Davies, C. L. Higginbotham, E. J. Treemer, C. Brown and R. C.

Treadgold, J. Chem. Soc., Perkin Trans. 1, 1992, 10, 3043.

3 7 4 8

O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 3 7 3 4 – 3 7 4 8

Article Doi

Chirality transfer in the aza-[2,3]-Wittig sigmatropic rearrangement

DOI: 10.1039/b510756c

Source and publish data:

Authors:

Article abstract of DOI:10.1039/b510756c

Full text of DOI:10.1039/b510756c

Products guided by the article

R&D Labs maybe for 867326-19-0

Relevant to this article

Hot Product