On the origins of diastereoselectivity in the conjugate additions of the antipodes of lithium N-benzyl-(N-α-methylbenzyl)amide to enantiopure cis- and trans-dioxolane containing α,β-unsaturated esters

Article abstract of DOI:10.1039/c2ob25099c

"Matching" and "mismatching" effects in the doubly diastereoselective conjugate additions of the antipodes of lithium N-benzyl-(N-α-methylbenzyl)amide to enantiopure cis- and trans-dioxolane containing α,β-unsaturated esters have been investigated. High levels of substrate control were established first upon conjugate addition of achiral lithium N-benzyl-N-isopropylamide to both tert-butyl (S,S,E)-4,5-O- isopropylidene-4,5-dihydroxyhex-2-enoate and tert-butyl (4R,5S,E)-4,5-O- isopropylidene-4,5-dihydroxyhex-2-enoate. However, upon conjugate addition of lithium (R)-N-benzyl-(N-α-methylbenzyl)amide and lithium (S)-N-benzyl-(N-α-methylbenzyl)amide to these substrates, neither reaction pairing reinforced the apparent sense of substrate control. These reactions do not, therefore, conform to the classical doubly diastereoselective "matching" or "mismatching" pattern usually exhibited by this class of reaction. A comparison of these reactions with the previously reported doubly diastereoselective conjugate addition reactions of lithium amide reagents to analogous substrates is also discussed.

Full text of DOI:10.1039/c2ob25099c

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
This article is part of the
OBC 10th anniversary
themed issue
All articles in this issue will be gathered together
online at
www.rsc.org/OBC10

Dynamic ArtVicielweALritnicklesOnline
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 6186
www.rsc.org/obc
PAPER
On the origins of diastereoselectivity in the conjugate additions of the
antipodes of lithium N-benzyl-(N-α-methylbenzyl)amide to enantiopure
cis- and trans-dioxolane containing α,β-unsaturated esters†‡
Stephen G. Davies,* Emma M. Foster, Aileen B. Frost, James A. Lee, Paul M. Roberts and
James E. Thomson
Received 13th January 2012, Accepted 16th March 2012
DOI: 10.1039/c2ob25099c
“Matching” and “mismatching” effects in the doubly diastereoselective conjugate additions of the
antipodes of lithium N-benzyl-(N-α-methylbenzyl)amide to enantiopure cis- and trans-dioxolane
containing α,β-unsaturated esters have been investigated. High levels of substrate control were established
first upon conjugate addition of achiral lithium N-benzyl-N-isopropylamide to both tert-butyl (S,S,E)-4,5-
O-isopropylidene-4,5-dihydroxyhex-2-enoate and tert-butyl (4R,5S,E)-4,5-O-isopropylidene-4,5-
dihydroxyhex-2-enoate. However, upon conjugate addition of lithium (R)-N-benzyl-(N-α-methylbenzyl)-
amide and lithium (S)-N-benzyl-(N-α-methylbenzyl)amide to these substrates, neither reaction pairing
reinforced the apparent sense of substrate control. These reactions do not, therefore, conform to the
classical doubly diastereoselective “matching” or “mismatching” pattern usually exhibited by this class of
reaction. A comparison of these reactions with the previously reported doubly diastereoselective
conjugate addition reactions of lithium amide reagents to analogous substrates is also discussed.
The individual stereochemical preferences of the two chiral
species in a doubly diastereoselective reaction can either
Introduction
Double asymmetric induction occurs in a reaction when two
chiral species are involved. The action of one chiral species upon
another will result in diastereoisomeric transition states; the
lowest in energy dictating the major stereochemical outcome of
the reaction, and the difference in energy between the transition
states determining the degree of diastereoselectivity. Masamune
et al. were the first to define the concept of double asymmetric
induction in the context of “matched” and “mismatched” reac-
tion pairings,1 stating: “the degree of asymmetric induction is
approximated to be (a × b) for a matched pair and (a ÷ b) for a
mismatched pair, where a and b are the diastereofacial selectiv-
ities of a substrate and a reagent, respectively.” It is also stipu-
lated, however, that “the multiplicativity will be valid only in a
qualitative sense”, and that “many secondary interactions which
occur in the regions remote from the reaction site are entirely
ignored [in this model].”
reinforce or oppose one another. For the “matched” reaction
pairing both species favour the same stereochemical outcome
and very high levels of diastereoselectivity are often observed.1
When the two chiral agents favour opposite stereochemical out-
comes the pairing is termed “mismatched” and a mixture of dia-
stereoisomeric products is usually observed.1 In the latter case,
the chiral agent with the higher level of directing ability dictates
the predominant stereochemical outcome of the reaction, if any.
Double asymmetric induction can arise in a reaction as a result
of chirality in many forms. For example, the reaction may
involve a chiral substrate and a chiral reagent, a chiral substrate
and chiral catalyst (with a further achiral reaction partner),
rearrangement of a substrate with two stereogenic centres, or the
reaction of a chiral substrate in a chiral solvent. Double asym-
metric induction has shown increased utility in synthesis in
recent years,2 both as a means of improving reaction diastereo-
selectivity and also as a tool for mechanistic investigations.3 The
phenomenon has found application in many different classes of
reaction, for example dihydroxylations,4 epoxidations,5 catalytic
hydrogenation,6 aldol reactions,7 conjugate additions8 and peri-
cyclic reactions including Diels–Alder reactions,9 aza-Claisen
rearrangements10 and 1,3-dipolar cycloadditions.11 As part of a
long-term goal directed towards the ab initio asymmetric syn-
thesis of unnatural amino sugars we have previously investigated
the effects of double asymmetric induction in the doubly
Department of Chemistry, Chemistry Research Laboratory,
University of Oxford, Mansfield Road, Oxford OX1 3TA, UK.
E-mail: steve.davies@chem.ox.ac.uk
†This article is part of the Organic & Biomolecular Chemistry 10th
Anniversary issue.
‡Electronic supplementary information (ESI) available: Full experimen-
tal details, copies of 1H and 13C NMR spectra, details of molecular mod-
elling calculations, and crystallographic data (for structures CCDC
846621–846624); see DOI: 10.1039/c2ob25099c
6186 | Org. Biomol. Chem., 2012, 10, 6186–6200
This journal is © The Royal Society of Chemistry 2012

View Article Online
diastereoselective conjugate additions of lithium amides12,13
(R)-1 and (S)-1 to enantiopure α,β-unsaturated esters containing
cis- or trans-dioxolane units.14,15 In the case of α,β-unsaturated
ester 3,16 conjugate addition of lithium (S)-N-benzyl-N-
(α-methylbenzyl)amide (S)-1 was found to correspond to the
empirically “matched” reaction pairing giving 3,4-anti-6 as the
major diastereoisomer (93 : 7 dr). Conjugate addition of lithium
(R)-N-benzyl-N-(α-methylbenzyl)amide (R)-1 to α,β-unsaturated
ester 3 represented the empirically “mismatched” reaction
pairing, giving a 50 : 50 mixture of C(3)-epimers 3,4-anti-7 and
3,4-syn-13. Addition of achiral lithium N-benzyl-N-isopropyl-
amide 2 to 3 gave 3,4-anti-5 as the sole diastereoisomer (>99 : 1
dr), confirming that under purely substrate control attack of the
Si face at C(3) within 3 is preferred. The effect of having a C(6)-
silyloxy substituent within the substrate was then probed by
investigating the analogous α,β-unsaturated ester 4.17 The
doubly diastereoselective conjugate additions of (R)-1 and (S)-1
to 4 were found to follow the same trend: addition of (S)-1 pro-
ceeded to give 3,4-anti-9 in >99 : 1 dr and 90% yield, represent-
ing the empirically “matched” reaction pairing. Conjugate
addition of (R)-1 represented the empirically “mismatched” reac-
tion pairing, and was seen to operate under considerable sub-
strate control as 3,4-anti-10 was isolated as the major
diastereoisomer, consistent with 8 being formed exclusively
upon conjugate addition of 2 to 4 (Fig. 1).
“matched” reaction pairing was found to be the conjugate
addition of (S)-1 to 18, which gave 3,4-syn-30 in >99 : 1 dr. Con-
jugate addition of lithium amides (R)-1 and 2 to 18 resulted in
formation of 3,4-anti products 23 and 22 as the major diastereo-
isomers in 70 : 30 and 75 : 25 dr, respectively (Fig. 2).
Intrigued by this apparent anomaly we proposed to investigate
the effects of double asymmetric induction upon the conjugate
addition of lithium amides (R)-1 and (S)-1 to cis- and trans-di-
oxolane containing α,β-unsaturated esters 31 and 32 which
incorporate a C(6)-methyl group in each case (Fig. 3).
The corresponding conjugate additions of (R)-1, (S)-1 and 2 to
α,β-unsaturated esters 1718 and 18, containing trans-dioxolane
units, were also investigated. An empirically “matched” reaction
pairing was observed upon conjugate addition of (R)-1 to 17, as
3,4-anti-20 was obtained as the sole product of this reaction
(>99 : 1 dr). Reaction of (S)-1 with 17 gave a 35 : 65 mixture of
β-amino esters 3,4-anti-21 and 3,4-syn-27, representing the
empirically “mismatched” reaction pairing. As the 3,4-syn-
product 27 predominated in this system, this indicated that the
lithium amide reagent was exerting the dominant stereocontrol.
Conjugate addition of the achiral lithium amide 2 confirmed that
the 3,4-anti product 19 dominated when the reaction was operat-
ing under only substrate control. However, the corresponding
results for the analogous C(6)-silyloxy substituted α,β-unsatu-
rated ester 18 did not fit this pattern. Unusually, the empirically
Fig. 2 The doubly diastereoselective conjugate additions of (R)-1, (S)-
1 and 2 to trans-dioxolane containing α,β-unsaturated esters 17 and 18.
Fig. 3 The doubly diastereoselective conjugate additions of the anti-
podes of lithium amide 1 to cis- and trans-dioxolane containing α,β-
unsaturated esters 31 and 32.
Results and discussion
Syntheses of cis- and trans-dioxolane containing α,β-unsaturated
esters
We have previously reported the synthesis of cis-dioxolane con-
taining α,β-unsaturated ester 3 from D-ribose 35, which proceeds
via intermediate alcohol 37.21 It was anticipated that deoxygena-
tion of alcohol 37 followed by cross metathesis with tert-butyl
acrylate would give α,β-unsaturated ester 31. Thus, acetonide
protection of D-ribose 3519 and subsequent Appel reaction20
gave iodide 36 in 64% yield, and treatment of 36 with BuLi
Fig. 1 The doubly diastereoselective conjugate additions of (R)-1,
(S)-1 and 2 to cis-dioxolane containing α,β-unsaturated esters 3 and 4.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6186–6200 | 6187

View Article Online
followed by in situ reduction of the intermediate aldehyde with
DIBAL-H gave alcohol 37 in 91% yield. Alcohol 37 was then
derivatised to the corresponding mesylate 38, tosylate 39 and
iodide 40 in quantitative, 95 and 72% yield, respectively, in the
hope that displacement of the terminal leaving group with
hydride reagents could be achieved (Scheme 1).21
Scheme 2 Reagents and conditions: (i) tert-butyl acrylate, Hoveyda-
Grubbs II, CH2Cl2, reflux, 24 h; (ii) H2 (1 atm), Pd/C, Et3N, MeOH, rt,
24 h.
involving hydrogenolytic reduction of an iodide intermediate
was also adopted for the synthesis of trans-dioxolane containing
α,β-unsaturated ester 32. Acetonide protection of dimethyl L-tar-
trate 44, followed by reduction with LiAlH4 gave diol 45 in 87%
yield over the two steps. Attempted mono-iodination of 45 gave
poor mass return and, despite attempted optimisation, 46 was
isolated in only 16% yield. Subsequent hydrogenolysis27 of 46
gave alcohol 47 which was used immediately in a one-pot
Swern–Wittig reaction. This approach avoided potential pro-
blems associated with the isolation of aldehyde 48, and gave a
65 : 35 [(E) : (Z)] mixture of diastereoisomers from which the
major product (E)-32 was isolated as a single diastereoisomer
(>99 : 1 dr) in 29% overall yield from 46 (Scheme 3).
Unfortunately, the mono-iodination of 45 was not scalable as
the already poor yield of iodide 46 was observed to decrease
further with increasing reaction scale (up to 1.0 g). It was there-
fore envisaged that the existing route may be improved via the
introduction of a protecting group strategy. Thus, the mono-O-
TBDMS and mono-O-benzyl protected derivatives of 45 were
prepared in 50 and 55% yield, respectively, and in each case an
Appel reaction of either 49 or 50 gave the corresponding iodides
51 and 52 in 95 and 82% yield. Hydrogenolysis of O-TBDMS
protected iodide 51 gave 53 in 83% yield, and subsequent treat-
ment of 53 with TBAF effected O-TBDMS deprotection to give
alcohol 47. Treatment of the crude reaction mixture containing
47 under the one-pot Swern–Wittig olefination procedure gave a
Scheme 1 Reagents and conditions: (i) conc. HCl, acetone–MeOH
(v/v 1 : 1), 60 °C, 1 h; (ii) imidazole, PPh3, I2, PhMe–MeCN (v/v 5 : 1),
60 °C, 1 h; (iii) BuLi, THF, −78 °C, 2 h then DIBAL-H, −78 °C to rt,
16 h; (iv) MsCl, Et3N, DMAP, CH2Cl2, 0 °C to rt, 3 h; (v) TsCl, Et3N,
DMAP, CH2Cl2, 0 °C to rt, 48 h.
Numerous attempts to convert mesylate 38, tosylate 39 or
iodide 40 into 41 were conducted. Unfortunately, reduction with
NaBH4, LiAlH4, lithium aminoborohydrides, or Superhydride®,
using various procedures,22 gave either returned starting material
or complex mixtures of products from which 41 could not be
isolated. Similarly, attempted reduction of iodide 40 via radical
processes23 or deoxygenation of alcohol 37 using the Barton–
McCombie reaction24 (via the intermediacy of the corresponding
xanthate ester) also failed to produce 41. Moreover, conversion
of iodide 40 into the corresponding organometallic reagents (e.g.
organolithium or Grignard reagent) followed by treatment with a
proton source25 was also unsuccessful. It therefore became clear
that an alternative route to access α,β-unsaturated ester 31 would
be necessary and it was envisaged that cross-metathesis of iodide
40 with tert-butyl acrylate, followed by chemoselective
reduction26 of 42 could lead to α,β-unsaturated ester 31.
Hoveyda-Grubbs II catalysed cross-metathesis of iodide 40 with
tert-butyl acrylate proceeded to give 42 in 66% yield and >99 : 1
dr. A range of different conditions were then screened for the
chemoselective reduction of 42, which typically gave mixtures
of starting material 42 and the desired product 31 which were
found to be easily separable. However, ester 43 was also pro-
duced via over reduction of 31 and it was found that this product
could not be separated from 31 chromatographically. After exten-
sive optimisation varying the reaction concentration, time and
catalyst loading, a procedure was developed which gave ∼90%
conversion to 31 exclusively. Purification of this mixture gave 31
in 73% isolated yield, and recovered starting material 42 in 10%
yield, which was then recycled. Preparation of enantiopure
cis-dioxolane containing α,β-unsaturated ester 31 was therefore
achieved via a six step synthesis in 20% overall yield from
D-ribose 35 (Scheme 2).
Scheme 3 Reagents and conditions: (i) DMP, TsOH, PhMe, reflux,
16 h; (ii) LiAlH4, THF, reflux, 16 h; (iii) imidazole, PPh3, I2, PhMe–
MeCN (v/v 5 : 1), 60 °C, 1 h; (iv) H2 (1 atm), Pd(OH)2/C, Et3N, MeOH,
rt, 16 h; (v) (COCl)2, DMSO, CH2Cl2, −78 °C, 1 h, then Et3N, rt; (vi)
tert-butyl (triphenylphosphoranylidene)acetate, CH2Cl2, rt, 16 h.
In light of the success of this approach for the synthesis of
cis-dioxolane containing α,β-unsaturated ester 31, a strategy
6188 | Org. Biomol. Chem., 2012, 10, 6186–6200
This journal is © The Royal Society of Chemistry 2012

View Article Online
Scheme 4 Reagents and conditions: (i) NaH, THF, 0 °C, 45 min, then TBDMSCl, 0 °C to rt, 16 h; (ii) NaH, THF, 0 °C, 45 min, then BnBr, 0 °C to
rt, 16 h; (iii) imidazole, PPh3, I2, PhMe–MeCN (v/v 5 : 1), 60 °C, 1 h; (iv) H2 (1 atm), Pd(OH)2/C, Et3N, MeOH, rt, 24 h; (v) TBAF, THF, rt, 18 h;
(vi) H2 (1 atm), Pd/C, EtOAc–AcOH (v/v 8 : 1), rt, 24 h; (vii) (COCl)2, DMSO, CH2Cl2, −78 °C, 1 h, then Et3N; (viii) tert-butyl (triphenylphosphora-
nylidene)acetate, CH2Cl2, rt, 16 h.
Scheme 5 Reagents and conditions: (i) THF, −78 °C, 2 h; (ii) Et2O, −20 °C, 5 h.
69 : 31 [(E) : (Z)] mixture of diastereoisomers, from which the
major product (E)-32 was isolated in >99 : 1 dr and 36% overall
yield from 53. It was envisaged that the O-benzyl protected
iodide 52 could be reduced to alcohol 47 in a tandem hydroge-
nolysis procedure, and upon treatment of a solution of 52 in
MeOH with Pearlman’s catalyst [Pd(OH)2/C] and Et3N under an
atmosphere of hydrogen, cleavage of the C–I bond was achieved
readily. However, it was found that removal of the O-benzyl
group did not occur under these conditions, even at elevated
pressures. It was therefore found to be necessary to purify 54,
which was isolated in 92% yield, before hydrogenolytic removal
of the O-benzyl group (in the absence of Et3N) to give 47. A
one-pot Swern–Wittig reaction on this sample of 47 then gave a
68 : 32 [(E) : (Z)] mixture of diastereoisomers, from which the
major product (E)-32 was isolated in >99 : 1 dr and 38% overall
yield from 54.28 Despite the introduction of additional steps, the
yield of α,β-unsaturated ester 32 was greatly improved by
employing either a mono-O-TBDMS or a mono-O-benzyl pro-
tecting group strategy (giving 32 in 14 and 16% overall yield
from diol 45 respectively), relative to the route in which protect-
ing groups were not used, in which 32 was produced from diol
45 in 5% overall yield.30 Both procedures employing protecting
groups were readily amenable to scale-up and 32 was therefore
produced in multigram quantities (Scheme 4).
α,β-unsaturated esters under our standard reaction conditions.14
In particular, deprotonation at the γ-position of the substrate by
the basic lithium amide reagent has been observed as a common
side reaction when the reaction is conducted in THF at −78 °C.
For example, addition of (S)-1 to 3 gave an 82 : 18 mixture of
3,4-anti-6 (>99 : 1 dr) and β,γ-unsaturated ester (Z)-55 upon
reaction in THF at −78 °C. However, suppression of this
γ-deprotonation pathway can be achieved if the conjugate
addition reaction is carried out in Et2O at −20 °C.31 Under these
conditions no evidence of 55 was observed upon conjugate
addition of (S)-1 to 3, although the diastereoselectivity of the
reaction was slightly compromised giving 3,4-anti-6 in 70%
yield and 93 : 7 dr (Scheme 5).
In light of this, the conjugate additions of lithium amides
(R)-1, (S)-1 and 2 to cis-dioxolane containing α,β-unsaturated
ester 31 were carried out on a small scale in both THF at
−78 °C, and Et2O at −20 °C. This enabled us to examine the
product ratios in each case and to identify which solvent may be
most amenable to scale up in order to obtain enough material to
correlate the stereochemical outcomes of these reactions via
hydrogenolysis. The level of substrate control elicited by α,β-un-
saturated ester 31 was established first upon conjugate addition
of achiral lithium N-benzyl-N-isopropylamide 2. When the con-
jugate addition of 2 to 31 was carried out in THF at −78 °C, an
82 : 18 mixture of β-amino ester 3,4-anti-56 (>99 : 1 dr) and
1
β,γ-unsaturated ester 58 was observed in the H NMR spectrum
Lithium amide conjugate additions to cis-dioxolane containing
α,β-unsaturated ester 31
of the crude reaction mixture. Chromatographic purification
enabled isolation of 56 in 41% yield and >99 : 1 dr, in addition
to 58 which was isolated in 18% yield as a single diastereo-
isomer (>99 : 1 dr). The (Z)-configuration within 58 was then
We have previously encountered some problems upon conjugate
addition of lithium amides to cis-dioxolane containing
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6186–6200 | 6189

View Article Online
Scheme 6 Reagents and conditions: (i) THF, −78 °C, 2 h; (ii) Et2O, −20 °C, 5 h.
Scheme 7 Reagents and conditions: (i) H2, Pd(OH)2/C, MeOH, rt,
16 h.
established via 1H NMR NOE spectroscopic analysis of the
purified sample. When the conjugate addition reaction was
carried out in Et2O at −20 °C, complete suppression of the
γ-deprotonation pathway was achieved; however, this was
accompanied by a decrease in diastereoselectivity from >99 : 1
dr in THF to 88 : 12 dr in Et2O (Scheme 6).
The relative configuration within 3,4-anti-56 was unambigu-
ously established via hydrogenolysis to give a crystalline deriva-
tive. Thus, a solution of 3,4-anti-56 (>99 : 1 dr) in MeOH was
treated with Pd(OH)2/C under an atmosphere of hydrogen to
give 59 in 97% yield and >99 : 1 dr (Scheme 7). Subsequent
recrystallization and single crystal X-ray diffraction analysis‡
enabled the relative configuration within 3,4-anti-59 to be estab-
lished unambiguously;32 in both cases the absolute (3R,4R,5S)-
configurations within 56 and 59 were then assigned relative to
the known configurations of the (D-ribose 35 derived) C(4) and
C(5) stereogenic centres (Fig. 4).
From this established sense of substrate control it was then
predicted that the conjugate addition of lithium (S)-N-benzyl-N-
(α-methylbenzyl)amide (S)-1 to α,β-unsaturated ester 31 would
be the “matched” pairing of chiral reagents, and that the conju-
gate addition of lithium (R)-N-benzyl-N-(α-methylbenzyl)amide
(R)-1 to α,β-unsaturated ester 31 would be the “mismatched”
pairing. Experimentally, however, conjugate addition of (S)-1 to
31 in THF at −78 °C gave 3,4-anti-60 as the major diastereo-
isomer (in 84 : 16 dr), with 23% conversion to β,γ-unsaturated
ester (Z)-58 also being observed. On changing the conditions to
Fig. 4 X-ray crystal structure of 3,4-anti-59 (selected H atoms are
omitted for clarity).
Et2O at −20 °C, the γ-deprotonation pathway was completely
suppressed giving an 85 : 15 mixture of β-amino esters 3,4-anti-
60 and 3,4-syn-61, respectively. Upon purification, the major dia-
stereoisomer 60 was isolated in 39% yield and >99 : 1 dr, and
the minor diastereoisomer 61 was isolated in 12% yield and
>99 : 1 dr (Scheme 8). The diastereoselectivity of the conjugate
addition of (S)-1 to 31 (under either set of reaction conditions)
was therefore found to be inferior to that observed under sub-
strate control alone.
In order to cross correlate the stereochemical outcome from
the conjugate addition of (S)-1 to α,β-unsaturated ester 31 with
that observed upon conjugate addition of 2 to 31, β-amino esters
3,4-anti-60 and 3,4-syn-61 were subjected to a tandem hydroge-
nolysis–reductive alkylation procedure. Hydrogenolysis of 3,4-
anti-60 (>99 : 1 dr) in the presence of acetone gave 59 in 79%
yield and >99 : 1 dr and, hydrogenolysis of 3,4-syn-61 (>99 : 1
dr) under identical conditions gave 62 in 99% yield and >99 : 1
dr. The spectroscopic data, including specific rotation, of the
sample of 59 so formed were found to be identical to those for
the sample obtained by hydrogenolysis of 56, thus providing
unequivocal evidence for the sense of diastereoselectivity
observed upon conjugate addition of (S)-1 to α,β-unsaturated
ester 31 (Scheme 9).
Scheme 8 Reagents and conditions: (i) THF, −78 °C, 2 h; (ii) Et2O, −20 °C, 5 h.
6190 | Org. Biomol. Chem., 2012, 10, 6186–6200
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 5 X-ray crystal structures of 3,4-anti-63 [left] and 3,4-syn-64
[right] (selected H atoms are omitted for clarity).
rotation, to those obtained for the sample derived from hydro-
genolysis of 3,4-syn-61 (Scheme 11).
Scheme 9 Reagents and conditions: (i) H2, Pd(OH)2/C, MeOH–
acetone (v/v 9 : 1), rt, 16 h.
Conjugate addition of lithium (R)-N-benzyl-(N-α-methylben-
zyl)amide (R)-1 to α,β-unsaturated ester 31 in THF at −78 °C
gave an 11 : 29 : 60 mixture of the diastereoisomeric conjugate
addition products 3,4-anti-63 and 3,4-syn-64, and β,γ-unsatu-
rated ester (Z)-58, respectively. It was found that upon changing
the reaction conditions to Et2O at −20 °C, the γ-deprotonation
pathway was again completely suppressed giving 3,4-anti 63 as
the major diastereoisomer (60 : 40 dr). In this instance, the major
diastereoisomer 63 was isolated in 38% yield and >99 : 1 dr and
the minor diastereoisomer 64 was isolated in 22% yield and
>99 : 1 dr (Scheme 10). The diastereoselectivity of conjugate
addition of (R)-1 to 31 (under either set of reaction conditions)
was therefore also found to be inferior to that observed under
substrate control alone.
Both diastereoisomeric β-amino esters 3,4-anti-63 and 3,4-
syn-64 were found to be crystalline and so single crystal X-ray
diffraction analyses‡ enabled the relative configurations within
both 63 and 64 to be determined (Fig. 5).32 In both cases
the absolute configurations within (3R,4R,5S,αR)-63 and
(3S,4R,5S,αR)-64 were assigned unambiguously, relative to the
known configurations of the α-methylbenzyl stereogenic centre,
and the (D-ribose 35 derived) C(4) and C(5) stereogenic centres.
Tandem hydrogenolysis–reductive alkylation of 63 (>99 : 1 dr)
in the presence of acetone gave 59 in 87% yield and >99 : 1 dr,
and treatment of 64 (>99 : 1 dr) under identical conditions gave
62 in 84% yield and >99 : 1 dr. This sample of 59 was found to
have identical spectroscopic data, including specific rotation, to
those obtained for the samples of 59 derived from hydrogenoly-
sis of either 3,4-anti-56 or 3,4-anti-60, and the sample of 62 was
found to have identical spectroscopic data, including specific
Scheme 11 Reagents and conditions: (i) H2, Pd(OH)2/C, MeOH–
acetone (v/v 9 : 1), rt, 16 h.
In accordance with our previous observations concerning the
conjugate additions of lithium amides to cis-dioxolane contain-
ing α,β-unsaturated esters,14 the formation of β,γ-unsaturated
ester (Z)-58 was observed in all cases when the reactions were
conducted in THF at −78 °C, and this γ-deprotonation pathway
was completely suppressed when the reactions were performed
in Et2O at −20 °C. Under the latter set of conditions, conjugate
addition of lithium N-benzyl-N-isopropylamide 2 to cis-dioxo-
lane containing α,β-unsaturated ester 31 showed a clear diaster-
eofacial preference for formation of the 3,4-anti diastereoisomer,
although this apparent sense of substrate control was not
reinforced upon conjugate addition of either (R)-1 or (S)-1.
Scheme 10 Reagents and conditions: (i) THF, −78 °C, 2 h; (ii) Et2O, −20 °C, 5 h.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6186–6200 | 6191

View Article Online
Lithium amide conjugate additions to trans-dioxolane containing
α,β-unsaturated ester 32
As we have not previously observed the γ-deprotonation
pathway upon conjugate addition to trans-dioxolane containing
α,β-unsaturated esters, investigations into the conjugate addition
of lithium amides (R)-1, (S)-1 and 2 to trans-dioxolane contain-
ing α,β-unsaturated ester 32 were undertaken next under our
standard conditions (i.e., in THF at −78 °C). The extent of sub-
strate control upon conjugate addition to α,β-unsaturated ester 32
was established via reaction of achiral lithium N-benzyl-N-iso-
propylamide 2 with 32, which gave an 82 : 18 mixture of 3,4-
anti-65 and 3,4-syn-66, respectively. Upon purification, the
major diastereoisomer 65 was isolated in 55% yield and >99 : 1
dr, whilst the minor diastereoisomer 66 was isolated in 14%
yield and >99 : 1 dr (Scheme 12).
Fig. 6 X-ray crystal structure of 3,4-anti-67 (selected H atoms are
omitted for clarity).
(α-methylbenzyl)amide (R)-1 to α,β-unsaturated ester 32 would
be the “matched” pairing of chiral reagents, and that the conju-
gate addition of lithium (S)-N-benzyl-N-(α-methylbenzyl)amide
(S)-1 to α,β-unsaturated ester 32 would be the “mismatched”
pairing. Experimentally, however, conjugate addition of (R)-1 to
32 gave a 76 : 24 mixture of 3,4-anti-69 and 3,4-syn-70, respect-
ively, with chromatographic purification of the crude reaction
mixture giving 69 in 98 : 2 dr and 47% isolated yield, and 70 in
13% yield and >99 : 1 dr (Scheme 14).
Scheme 12 Reagents and conditions: (i) THF, −78 °C, 2 h.
The relative configurations within 3,4-anti-65 and 3,4-syn-66
were unambiguously established via hydrogenolysis which
gave a crystalline derivative: hydrogenolysis of 3,4-anti-65
(>99 : 1 dr) gave 67 which was isolated in 92% yield and >99 : 1
dr. Similarly, hydrogenolysis of 3,4-syn-66 (>99 : 1 dr) under
identical conditions gave 68 in 66% isolated yield and >99 : 1 dr
(Scheme 13). Subsequent single crystal X-ray diffraction
analysis‡ established the relative configuration within 67 unam-
biguously,32 with the absolute (S,S,S)-configuration within 67
being assigned relative to the known configurations of the
(dimethyl L-tartrate 44 derived) C(4) and C(5) stereogenic centres
(Fig. 6). This analysis therefore allowed the absolute (S,S,S)-
configuration within 65 [and the absolute (3R,4S,5S)-configur-
ations within 66 and 68] to also be assigned unambiguously.
From this established sense of substrate control it was then
predicted that the conjugate addition of lithium (R)-N-benzyl-N-
Scheme 14 Reagents and conditions: (i) THF, −78 °C, 2 h.
The stereochemical outcome of this reaction was then estab-
lished by hydrogenolytic chemical correlation to N-isopropyl
substituted β-amino esters 67 and 68. Hydrogenolysis of 3,4-
anti-69 (98 : 2 dr) in the presence of acetone gave 67 in 99%
yield and 98 : 2 dr, and hydrogenolysis of 3,4-syn-70 (>99 : 1 dr)
under identical conditions gave 68 in 98% yield and >99 : 1 dr
(Scheme 15). These samples of 67 and 68 were found to have
identical spectroscopic data, including specific rotations, to those
obtained previously, providing unequivocal evidence of the
Scheme 13 Reagents and conditions: (i) H2 (1 atm), Pd(OH)2/C,
Scheme 15 Reagents and conditions: (i) H2, Pd(OH)2/C, MeOH–
MeOH, rt, 16 h.
acetone (v/v 9 : 1), rt, 16 h.
6192 | Org. Biomol. Chem., 2012, 10, 6186–6200
This journal is © The Royal Society of Chemistry 2012

View Article Online
The origins of diastereoselectivity observed upon conjugate
addition
sense of stereoinduction upon conjugate addition of (R)-1 to
α,β-unsaturated ester 32.
Upon conjugate addition of (S)-1 to 32 the apparent sense of
substrate control was found to be overwhelmed by the lithium
amide reagent, giving a predominance of the 3,4-syn-diastereo-
isomer 72 (84 : 16 dr). After chromatographic purification of the
crude reaction mixture, the major diastereoisomeric product 72
was isolated in 57% yield and >99 : 1 dr, and the minor dia-
stereoisomer 71 was isolated in 10% yield and 82 : 18 dr
(Scheme 16).
The conjugate addition of enantiopure lithium amides [such as
(R)-1 and (S)-1] to achiral α,β-unsaturated esters has been
shown, by us and others, to be highly diastereoselective (>95 : 5
dr) for many substrates,13 with a highly predictable sense of dia-
stereoselectivity. We have previously reported a transition state
mnemonic which correctly and reliably rationalises the diaster-
eoselectivity of this class of conjugate addition reactions.34
However, given the known tendency of lithium amides to exist
as a variety of aggregates in a range of solvents,35 the true origin
of diastereoselectivity within these reactions is unclear and our
investigations in this area are ongoing. Our preliminary studies
have shown that (i) the Michael acceptor is required to adopt an
s-cis conformation upon conjugate addition;3,36 (ii) only second-
ary lithium amides undergo highly diastereoselective transforma-
tions;3a and (iii) two-fold stoichiometries of the lithium amide
reagent are required in a few cases.37 Considering the conjugate
additions of (R)-1, (S)-1 and 2 to cis-dioxolane containing
α,β-unsaturated esters 3,16 417 and 31, the formation of the corre-
sponding (Z)-β,γ-unsaturated esters 55, 58 and 74 was observed
in all cases when the reactions were conducted in THF at
−78 °C.14 We have previously postulated that formation of (Z)-
β,γ-unsaturated esters is consistent with γ-deprotonation in reac-
tive conformation 73A, and it may be assumed that conjugate
addition also proceeds via a similar reactive conformation of the
α,β-unsaturated ester.14,38 Upon inspection of the full set of data
for these three substrates we determined that the “facial selectiv-
ity” of both conjugate addition and γ-deprotonation (i.e., the
[A + C] : B ratio) should be considered. From these data it is
clear that cis-dioxolane containing α,β-unsaturated esters 3, 4
and 31 all exhibit exceptionally high levels of substrate control,
showing a clear diastereofacial preference for either γ-deprotona-
tion or conjugate addition to give the corresponding 3,4-anti
diastereoisomer. These experimentally determined reaction out-
comes are consistent with approach of the lithium amide reagent
on the lower face (as drawn) of the α,β-unsaturated ester in either
conformations 73A, 73B or 73C where the upper face is steri-
cally blocked in all accessible conformations by the C(5)-substi-
tuent (as illustrated for 73C, for example) and it is not possible
for the expected “mismatched” lithium amide reagent to over-
whelm this inherent substrate control; this trend was also
observed upon reaction in Et2O at −20 °C (Fig. 7).
Scheme 16 Reagents and conditions: (i) THF, −78 °C, 2 h.
In each case the configurations within 3,4-anti-71 and 3,4-syn-
7233 were again established by hydrogenolytic chemical corre-
lation to the corresponding N-isopropyl substituted β-amino
esters 67 and 68. Hydrogenolysis of 3,4-anti-71 (82 : 18 dr) in
the presence of acetone effected reductive alkylation of 71 to
give 67 in 97% yield and 82 : 18 dr, whereas hydrogenolysis of
the sample of 3,4-syn-72 (>99 : 1 dr) under the same conditions
gave 68 in 90% yield and >99 : 1 dr (Scheme 17). In both cases
1
the H and 13C NMR spectroscopic data for these samples of 67
and 68 were identical to those for the authentic samples prepared
from either 65 or 69, and either 66 or 70, respectively, providing
unequivocal evidence of the sense of stereoinduction upon con-
jugate addition of (S)-1 to α,β-unsaturated ester 32.
These data demonstrate that neither (R)-1 nor (S)-1 underwent
conjugate addition to trans-dioxolane containing α,β-unsaturated
The conjugate additions of achiral lithium N-benzyl-N-isopro-
pylamide 2 to trans-dioxolane containing α,β-unsaturated esters
17,18 18 and 32 also displayed a clear diastereofacial preference
for formation of the 3,4-anti diastereoisomers as the major pro-
ducts, although the levels of “facial selectivity” (which in this
case is equivalent to the dr of crude reaction mixture as γ-depro-
tonation was not observed for these substrates) under substrate
control alone were not as great as those observed for cis-dioxo-
lane containing α,β-unsaturated esters 3, 4 and 31. We have pre-
viously postulated a model for conjugate addition to trans-
dioxolane containing α,β-unsaturated esters which is consistent
with this stereochemical outcome [i.e., conjugate addition on the
upper face (as drawn) of 75 gives rise to the corresponding 3,4-
anti-diastereoisomeric product],14 although other possible reac-
tive conformations of the α,β-unsaturated ester should not be
Scheme 17 Reagents and conditions: (i) H2, Pd(OH)2/C, MeOH–
acetone (v/v 9 : 1), rt, 16 h.
ester 32 with the very high degree of diastereoselectivity that
would usually be expected from conjugate addition reactions of
these lithium amide reagents. In both cases the lithium amide
reagent was the dominant stereocontrolling factor. However,
neither conjugate addition reaction appeared to reinforce the
apparent sense of substrate control.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6186–6200 | 6193

View Article Online
Fig. 7 The doubly diastereoselective conjugate additions of (R)-1, (S)-1 and 2 to cis-dioxolane containing α,β-unsaturated esters 3, 4 and 31.
b
[aCombined yield of both diastereoisomeric conjugate addition products, each isolated in >99 : 1 dr; The combined yields in parentheses correspond
to the analogous reactions in Et2O].
Fig. 8 The doubly diastereoselective conjugate additions of (R)-1, (S)-1 and 2 to trans-dioxolane containing α,β-unsaturated esters 17, 18 and 32.
[aCombined yield of both diastereoisomeric conjugate addition products, each isolated in >99 : 1 dr].
discounted.38 In each case, the sense of diastereoselectivity
observed upon conjugate addition of either antipode of lithium
N-benzyl-N-(α-methylbenzyl)amide 1 to trans-dioxolane con-
taining α,β-unsaturated esters 17, 18 and 32 was found to corre-
late with reagent control, a finding which is also consistent with
a decreased level of substrate control for 17, 18 and 32 with
respect to cis-dioxolane containing α,β-unsaturated esters 3, 4
and 31. Since conjugate addition to the lower face of 75 is not
blocked by the C(5)-substituent, in this case the reagent control
is able to overwhelm the inherent substrate control in the “mis-
matched” reaction. The C(5)-vinyl substituted α,β-unsaturated
ester 17 was the only one of these substrates for which an
enhancement in diastereoselectivity was observed upon reaction
of the expected “matched” pairing of chiral reagents (Fig. 8). In
both cases, molecular modelling calculations (MM2 force field)
of α,β-unsaturated esters 31 and 32 produced minimum energy
conformations consistent with these proposed models to rational-
ise the diastereoselectivity observed upon the conjugate additions
of the antipodes of lithium N-benzyl-(N-α-methylbenzyl)amide
1 to α,β-unsaturated esters such as 31 and 32.‡
As the high diastereoselectivity usually synonymous with con-
jugate addition of either antipode of lithium amide 1 was not
observed upon conjugate addition to some of these enantiopure
dioxolane containing α,β-unsaturated esters, it seems that the
6194 | Org. Biomol. Chem., 2012, 10, 6186–6200
This journal is © The Royal Society of Chemistry 2012

View Article Online
lithium amide reagent may be prevented from partaking in its
normal mode of reactivity. In addition, the fact that the sense of
substrate control in some of these reactions is not augmented
upon the conjugate addition of either chiral lithium amide (R)-1
or (S)-1 suggests that the factors which influence the substrate
and reagent control in each case are not independent and thus the
reaction pairings cannot be simply classified according to the
classical “matched” and “mismatched” definitions of Masamune
and co-workers.1 These effects may be due to chelation between
the lithium amide reagent and any of the pendant oxygen atoms
within the α,β-unsaturated esters, giving rise to alternative reac-
tion pathways that compete with the normal mode of reactivity.
characteristic peaks are reported in cm−1. NMR spectra were
recorded on Bruker Avance spectrometers in the deuterated
solvent stated. Spectra were recorded at rt. The field was locked
by external referencing to the relevant deuteron resonance. When
the diastereotopic methyl groups of acetonide and isopropyl
functionalities could not be unambiguously assigned, the
descriptors MeCMe and MeCHMe were employed. Low-resol-
ution mass spectra were recorded on either a VG MassLab
20–250 or a Micromass Platform 1 spectrometer. Accurate mass
measurements were run on either a Bruker MicroTOF internally
calibrated with polyalanine, or a Micromass GCT instrument
fitted with a Scientific Glass Instruments BPX5 column (15 m ×
0.25 mm) using amyl acetate as a lock mass.
Conclusion
(2R,3R,4S,5S)-2-Methoxy-3,4-O-isopropylidene-3,4-dihydroxy-5-
iodomethyltetrahydrofuran 36
Classical “matching” and “mismatching” effects in the doubly
diastereoselective conjugate additions of the antipodes of lithium
N-benzyl-(N-α-methylbenzyl)amide to tert-butyl (S,S,E)-4,5-O-
isopropylidene-4,5-dihydroxyhex-2-enoate
and
tert-butyl
(4R,5S,E)-4,5-O-isopropylidene-4,5-dihydroxyhex-2-enoate were
not observed. The levels of substrate control were established in
each case upon conjugate addition of achiral lithium N-benzyl-
N-isopropylamide. However, upon conjugate addition of lithium
(R)-N-benzyl-(N-α-methylbenzyl)amide and lithium (S)-N-
benzyl-(N-α-methylbenzyl)amide to these substrates, neither
reaction pairing reinforced the apparent sense of substrate
control. These reactions do not, therefore, conform to the doubly
diastereoselective “matching” or “mismatching” pattern usually
exhibited by this class of reaction. Further investigations into the
origin of diastereoselectivity observed upon conjugate addition
of enantiopure lithium amides to dioxolane containing α,β-unsa-
turated esters are ongoing within our laboratories. The impli-
cation of the above finding is however clear: cyclic dioxolane
protection should be avoided in favour of acyclic protection.8a,c
Conc aq HCl (2.0 mL) was added to a solution of D-ribose 35
(50.0 g, 0.333 mol) in acetone–MeOH (v/v 1 : 1, 700 mL). The
resultant solution was heated at 60 °C for 1 h then allowed to
cool to rt and neutralised by the addition of Na2CO3 (∼10 g).
The resultant suspension was filtered through Celite® (eluent
EtOAc) and the filtrate was concentrated in vacuo. The residue
was dissolved in EtOAc (250 mL) and the resultant solution was
washed with H2O (250 mL). The aqueous layer was extracted
with EtOAc (2 × 250 mL) and the combined organic extracts
were dried and concentrated in vacuo. PPh3 (105 g, 0.40 mol)
and imidazole (34.0 g, 0.500 mol) were added to the residue and
the resultant mixture was dissolved in PhMe–MeCN (v/v 5 : 1,
1 L). I2 (101 g, 0.399 mol) was then added and the reaction
mixture was heated at 60 °C for 1 h, then allowed to cool to rt
and diluted with Et2O (250 mL). The resultant mixture was
washed sequentially with 10% aq Na2S2O3 (1 L), H2O (1 L) and
brine (1 L), then dried and concentrated in vacuo. Purification
via flash column chromatography (eluent 30–40 °C petrol–Et2O,
20 : 1) gave 36 as an orange oil (66.4 g, 64%, >99 : 1 dr);19,40
[α]2D4 − 72.3 (c 1.0 in CHCl3); {lit.41 [α]2D4 − 79.8 (c 1.0 in
CHCl3)}; δH (400 MHz, CDCl3) 1.34 (3H, s, MeCMe), 1.49
(3H, s, MeCMe), 3.17 (1H, app t, J 10.1, CHAHBI), 3.30 (1H,
dd, J 10.1, 5.8, CHAHBI), 3.38 (3H, s, OMe), 4.45 (1H, app dd,
J 10.1, 5.8, C(5)H), 4.64 (1H, app d, J 5.8, C(4)H), 4.78 (1H,
app d, J 5.8, C(3)H), 5.06 (1H, app s, C(2)H).
Experimental
General experimental details
Reactions involving organometallic or other moisture-sensitive
reagents were carried out under a nitrogen or argon atmosphere
using standard vacuum line techniques and glassware that was
flame dried and cooled under nitrogen before use. BuLi was pur-
chased from Sigma-Aldrich (as a solution in hexanes) and
titrated against diphenylacetic acid before use. Solvents were
dried according to the procedure outlined by Grubbs and co-
workers.39 Water was purified by an Elix® UV-10 system. All
other reagents were used as supplied without prior purification.
Organic layers were dried over MgSO4. Thin layer chromato-
graphy was performed on aluminium plates coated with 60 F254
silica. Plates were visualised using UV light (254 nm), iodine,
1% aq KMnO4, or 10% ethanolic phosphomolybdic acid. Flash
column chromatography was performed on Kieselgel 60 silica.
Melting points were recorded on a Gallenkamp Hot Stage
apparatus and are uncorrected. Optical rotations were recorded
on a Perkin-Elmer 241 polarimeter with a water-jacketed 10 cm
cell. Specific rotations are reported in 10−1 deg cm2 g−1 and con-
centrations in g/100 mL. IR spectra were recorded on a Bruker
Tensor 27 FT-IR spectrometer on an ATR module. Selected
(4S,5R)-2,2-Dimethyl-4-hydroxymethyl-5-vinyl-1,3-dioxolane 37
BuLi (2.1 M in hexanes, 31.7 mL, 66.7 mmol) was added to a
solution of 36 (20.9 g, 66.7 mmol, >99 : 1 dr) in THF (340 mL)
at −78 °C and the resultant solution was stirred at −78 °C for
2 h. DIBAL-H (1.0 M in THF, 100 mL, 100 mmol) was then
added via cannula and the resultant mixture was allowed to
warm to rt over 16 h. Acetone (500 mL) and satd aq sodium
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6186–6200 | 6195

View Article Online
potassium tartrate (500 mL) were added sequentially and stirring
was continued for 1 h at rt. The reaction mixture was then parti-
tioned between brine (300 mL) and EtOAc (300 mL), and the
aqueous layer was extracted with EtOAc (300 mL). The com-
bined organic extracts were then dried and concentrated in
vacuo. Purification via flash column chromatography (eluent
30–40 °C petrol–EtOAc, 2 : 1) gave 37 as a pale yellow oil
(9.60 g, 91%, >99 : 1 dr);42 [α]D24 − 45.7 (c 1.0 in CHCl3);
{lit.43 [α]D24 − 44.0 (c 4.9 in CHCl3)}; δH (400 MHz, CDCl3)
1.40 (3H, s, MeCMe), 1.52 (3H, s, MeCMe), 1.94 (1H, s, OH),
3.58 (2H, app t, J 5.5, CH2OH), 4.24–4.30 (1H, m, C(4)H), 4.65
(1H, app t, J 7.0, C(5)H), 5.29 (1H, dd, J 10.4, 1.0,
CHvCHAHB), 5.40 (1H, dd, J 17.4, 1.0, CHvCHAHB), 5.87
(1H, ddd, J 17.4, 10.4, 7.0, CHvCH2).
(3H, s, MeCMe), 1.50 (9H, s, CMe3), 1.53 (3H, s, MeCMe),
3.02 (1H, dd, J 10.4, 6.5, C(6)HA), 3.13 (1H, dd, J 10.4, 7.5,
C(6)HB), 4.52 (1H, app q, J 6.5, C(5)H), 4.79 (1H, app dt, J 6.5,
1.3, C(4)H), 6.09 (1H, dd, J 15.5, 1.3, C(2)H), 6.81 (1H, dd,
J 15.5, 6.5, C(3)H); δC (100 MHz, CDCl3) 3.1 (C(6)), 25.5, 28.0
(CMe2), 28.1 (CMe3), 76.9 (C(4)), 78.5 (C(5)), 80.8 (CMe3),
109.6 (CMe2), 125.8 (C(2)), 139.8 (C(3)), 165.0 (C(1)); m/z
(ESI+) 759 ([2M + Na]+, 100%), 391 ([M + Na]+, 53%); HRMS
+
(ESI+) C13H21INaO4 ([M + Na]+) requires 391.0377; found
391.0365.
tert-Butyl (4R,5S)-4,5-O-isopropylidene-4,5-dihydroxyhex-2-
enoate 31
(R,R)-2,2-Dimethyl-4-iodomethyl-5-vinyl-1,3-dioxolane 40
Pd/C (2.27% w/w of substrate, 23 mg) and Et3N (5.20 mL,
27.2 mmol) were added to a solution of 42 (1.00 g, 2.72 mmol,
>99 : 1 dr) in MeOH (179 mL) at rt. The resultant mixture was
degassed and saturated with H2, then left to stir under an atmos-
phere of H2 (1 atm) for 24 h. The reaction mixture was then
filtered through Celite® (eluent MeOH) and the filtrate was con-
centrated in vacuo to give a 90 : 10 mixture of 31 and 42. Purifi-
cation via flash column chromatography (eluent 30–40 °C
petrol–Et2O, 5 : 1) gave 42 as a yellow solid (110 mg, 10%,
>99 : 1 dr) and 31 as a pale yellow oil (480 mg, 73%, >99 : 1
dr); [α]2D4 + 3.0 (c 1.0 in CHCl3); νmax (ATR) 2982, 2937, 2905
(C–H), 1714 (CvO), 1659 (CvC); δH (400 MHz, CDCl3) 1.17
(3H, d, J 6.3, C(6)H3), 1.38 (3H, s, MeCMe), 1.49 (9H, s,
CMe3), 1.53 (3H, s, MeCMe), 4.42 (1H, app quintet, J 6.3,
C(5)H), 4.64 (1H, app t, J 6.3, C(4)H), 6.00 (1H, app d, J 15.7,
C(2)H), 6.73 (1H, dd, J 15.7, 6.3, C(3)H); δC (100 MHz,
CDCl3) 16.2 (C(6)), 25.4, 28.0 (CMe2), 28.1 (CMe3), 74.0
(C(5)), 77.7 (C(4)), 80.6 (CMe3), 108.6 (CMe2), 124.9 (C(2)),
142.3 (C(3)), 165.3 (C(1)); m/z (ESI+) 265 ([M + Na]+, 100%);
PPh3 (396 mg, 1.51 mmol) and imidazole (129 mg, 1.89 mmol)
were added to a solution of 37 (200 mg, 1.26 mmol, >99 : 1 dr)
in PhMe–MeCN (v/v 5 : 1, 4 mL). I2 (384 mg, 1.51 mmol) was
then added and the resultant mixture was heated at 60 °C for 1 h.
The reaction mixture was allowed to cool to rt, diluted with Et2O
(5 mL) and washed sequentially with 10% aq Na2S2O3 (5 mL),
H2O (5 mL) and brine (5 mL), then dried and concentrated
in vacuo. Purification via flash column chromatography (eluent
30–40 °C petrol–Et2O, 2 : 1) gave 40 as a pale yellow oil
(244 mg, 72%, >99 : 1 dr); [α]2D4 − 13.7 (c 1.0 in CHCl3); νmax
(ATR) 2986, 2935 (C–H); δH (400 MHz, CDCl3) 1.34 (3H, s,
MeCMe), 1.47 (3H, s, MeCMe), 3.02 (1H, dd, J 10.2, 6.3,
CHAHBI), 3.10 (1H, dd, J 10.2, 7.5, CHAHBI), 4.40 (1H, app dt,
J 7.5, 6.3, C(4)H), 4.59 (1H, app t, J 6.3, C(5)H), 5.29 (1H, d,
J 10.6, CHvCHAHB), 5.39 (1H, d, J 17.4, CHvCHAHB), 5.81
(1H, ddd, J 17.4, 10.6, 6.3, CHvCH2); δC (100 MHz, CDCl3)
4.0 (CH2I), 25.6, 28.2 (CMe2), 78.4 (C(4)), 79.1 (C(5)), 109.0
(CMe2), 119.3 (CHvCH2), 132.4 (CHvCH2); m/z (FI+) 268
+
HRMS (ESI+) C13H22NaO4 ([M + Na]+) requires 265.1410;
found 265.1403.
+
([M]+, 100%); HRMS (FI+) C8H13IO2 ([M]+) requires
267.9955; found 267.9966.
(S,S)-2,2-Dimethyl-4,5-bis(hydroxymethyl)-1,3-dioxolane 45
tert-Butyl (R,R,E)-4,5-O-isopropylidene-4,5-dihydroxy-6-
iodohex-2-enoate 42
Step 1: DMP (12.0 mL, 96.9 mmol) and TsOH (128 mg,
0.65 mmol) were added to a solution of dimethyl L-tartrate 44
(11.5 g, 64.6 mmol) in PhMe (75 mL). The resultant mixture
was fitted with a Dean–Stark apparatus and heated at reflux for
16 h. The reaction mixture was then cooled to rt and satd aq
NaHCO3 (50 mL) was added. The resultant mixture was stirred
at rt for 15 min then the aqueous layer was extracted with EtOAc
(2 × 30 mL). The combined organic extracts were sequentially
washed with H2O (40 mL) and brine (40 mL), then dried and
concentrated in vacuo to give dimethyl (R,R)-2,2-dimethyl-1,3-
dioxolane-4,5-dicarboxylate as a yellow oil (13.5 g, 96%, >99 : 1
dr);44 [α]D24 − 56.1 (c 1.0 in MeOH); {lit.45 [α]2D4 − 49.1 (c 1.0 in
Hoveyda-Grubbs II (1.55 g, 1.82 mmol) was added to a
degassed solution of 40 (4.89 g, 18.2 mmol, >99 : 1 dr) and tert-
butyl acrylate (7.01 mL, 54.7 mmol) in CH2Cl2 (94 mL) and the
resultant mixture was heated at reflux for 24 h. The reaction
mixture was then allowed to cool to rt and concentrated
in vacuo. Purification via flash column chromatography (eluent
30–40 °C petrol–EtOAc, 20 : 1) gave 42 as a pale yellow solid
(4.45 g, 66%, >99 : 1 dr); mp 39–43 °C; [α]2D4 − 0.7 (c 1.0 in
CHCl3); νmax (ATR) 3031, 2978, 2934, 2903, 2852, 2361, 2342
(C–H), 1696 (CvO), 1645 (CvC); δH (400 MHz, CDCl3) 1.39
6196 | Org. Biomol. Chem., 2012, 10, 6186–6200
This journal is © The Royal Society of Chemistry 2012

View Article Online
MeOH)}; δH (400 MHz, CDCl3) 1.50 (6H, s, CMe2), 3.83 (6H,
s, CO2Me), 4.82 (2H, s, C(4)H, C(5)H).
in vacuo. Purification via flash column chromatography (eluent
30–40 °C petrol–Et2O, 2 : 1) gave 52 as a pale yellow oil
Step 2: LiAlH4 (1.0 M in THF, 100 mL, 100 mmol) was
added dropwise to a stirred solution of dimethyl (R,R)-2,2-
dimethyl-1,3-dioxolane-4,5-dicarboxylate (9.32 g, 42.7 mmol,
>99 : 1 dr) in THF (160 mL) at 0 °C. The resultant mixture was
heated at reflux for 16 h then allowed to cool to rt. 10% aq
NaOH (150 mL), H2O (70 mL) and EtOAc (150 mL) were
added and the resultant mixture was stirred at rt for 1 h. The
reaction mixture was then filtered through Celite® (eluent
EtOAc) and the filtrate was dried and concentrated in vacuo to
give 45 as a pale yellow oil (6.55 g, 91%, >99 : 1 dr);46 δH
(400 MHz, CDCl3) 1.44 (6H, s, CMe2), 2.52 (2H, br s, OH),
3.67–3.74 (2H, m, CHAHBOH), 3.81–3.87 (2H, m, CHAHBOH),
4.02–4.05 (2H, m, C(4)H, C(5)H).
(939 mg, 82%, >99 : 1 dr);49 [α]D24 − 9.6 (c 1.0 in CHCl3);
49
{lit.
[α]2D3 − 10.1 (c 2.1 in CHCl3)}; δH (400 MHz, CDCl3)
1.43 (3H, s, MeCMe), 1.48 (3H, s, MeCMe), 3.29 (1H, dd,
J 10.6, 5.5, CHAHBI), 3.36 (1H, dd, J 10.6, 5.1, CHAHBI), 3.64
(1H, dd, J 10.2, 5.1, CHAHBOBn), 3.68 (1H, dd, J 10.2, 5.1,
CHAHBOBn), 3.87 (1H, app dt, J 7.5, 5.1, C(4)H), 3.98 (1H,
app dt, J 7.5, 5.1, C(5)H), 4.60 (2H, app s, CH2Ph), 7.28–7.40
(5H, m, Ph).
(S,S)-2,2-Dimethyl-4-benzyloxymethyl-5-methyl-1,3-dioxolane 54
(S,S)-2,2-Dimethyl-4-benzyloxymethyl-5-hydroxymethyl-1,3-
dioxolane 50
Pd(OH)2/C (50% w/w of substrate, 100 mg) and Et3N (0.20 mL,
1.66 mmol) were added to a solution of 52 (200 mg, 0.55 mmol,
>99 : 1 dr) in MeOH (5 mL) at rt. The resultant solution was
degassed and saturated with H2, then left to stir under an atmos-
phere of H2 (1 atm) for 24 h. The reaction mixture was then
filtered through Celite® (eluent MeOH) and the filtrate was con-
centrated in vacuo. Purification via flash column chromatography
(eluent 30–40 °C petrol–Et2O, 10 : 1) gave 54 as a yellow oil
(117 mg, 92%, >99 : 1 dr);50 [α]D24 + 10.4 (c 1.0 in CHCl3);
{lit.50 [α]D24 + 10.1 (c 1.4 in CHCl3)}; δH (400 MHz, CDCl3)
1.30 (3H, d, J 6.1, C(5)Me), 1.41 (3H, s, MeCMe), 1.44 (3H, s,
MeCMe), 3.56 (1H, dd, J 10.2, 4.4, CHAHBOBn), 3.61 (1H, dd,
J 10.2, 5.5, CHAHBOBn), 3.74–3.79 (1H, m, C(4)H), 3.94 (1H,
dq, J 8.2, 6.1, C(5)H), 4.58 (1H, d, J 12.2, CHAHBPh), 4.62
(1H, d, J 12.2, CHAHBPh), 7.28–7.38 (5H, m, Ph).
NaH (60% dispersion in mineral oil, 50 mg, 1.23 mmol) was
stirred in 30–40 °C petrol (2 mL) for 10 min. The petrol was
removed via cannula, then THF (2 mL) was added and the resul-
tant suspension was cooled to 0 °C. A solution of 45 (200 mg,
1.23 mmol, >99 : 1 dr) in THF (2 mL) was added via cannula
and the resultant mixture was allowed to warm to rt then stirred
for 45 min. BnBr (0.15 mL, 1.23 mmol) was then added and the
resultant solution was stirred at rt for 16 h. The reaction mixture
was diluted with Et2O (5 mL) and washed with satd aq NaHCO3
(2 × 10 mL). The aqueous layer was separated and extracted
with Et2O (2 × 6 mL) and the combined organic extracts were
dried and concentrated in vacuo. Purification via flash column
chromatography (eluent 30–40 °C petrol–Et2O, 7 : 3) gave 50 as
a pale yellow oil (170 mg, 55%, >99 : 1 dr);47 [α]D24 + 8.6 (c 1.0
in CHCl3); {lit.48 [α]D20 + 8.7 (c 1.2 in CHCl3)}; δH (400 MHz,
CDCl3) 1.47 (3H, s, MeCMe), 1.48 (3H, s, MeCMe), 2.23 (1H,
app q, J 4.3, OH), 3.61 (1H, dd, J 9.9, 5.8, CHAHBOBn),
3.70–3.77 (2H, m, CHAHBOH, CHAHBOBn), 3.83 (1H, dt, J
11.6, 4.4, CHAHBOH), 3.97–4.02 (1H, m, C(4)H), 4.09–
4.14 (1H, m, C(5)H), 4.64 (2H, app s, CH2Ph), 7.33–7.44
(5H, m, Ph).
tert-Butyl (S,S,E)-4,5-O-isopropylidene-4,5-dihydroxyhex-2-
enoate 32
10% Pd/C (60% w/w of substrate, 4.20 g) was added to a sol-
ution of 54 (7.00 g, 29.6 mmol) in EtOAc–AcOH (v/v 8 : 1,
144 mL) at rt. The resultant mixture was degassed and saturated
with H2, then left to stir under an atmosphere of H2 (1 atm) for
24 h. The reaction mixture was then filtered through Celite®
(eluent EtOAc) and the filtrate was concentrated in vacuo.
The residue was dissolved in CHCl3 (100 mL) and washed
with H2O (100 mL). The aqueous layer was extracted with
CHCl3 (3 × 50 mL) and the combined organic extracts were
dried and concentrated in vacuo to give 47 as a brown oil
(4.33 g); δH (400 MHz, CDCl3) 1.29 (3H, d, J 5.8, C(5)Me),
1.40 (3H, s, MeCMe), 1.43 (3H, s, MeCMe), 3.58–3.67 (2H, m,
CH2OH), 3.78–3.83 (1H, m, C(4)H), 3.98–4.05 (1H, m, C(5)H).
DMSO (2.74 mL, 38.5 mmol) was added dropwise to a solution
of (COCl)2 (3.00 mL, 35.5 mmol) in CH2Cl2 (260 mL) at
−78 °C and the resultant mixture was left to stir for 10 min.
A solution of 47 (4.33 g) in CH2Cl2 (20 mL) was added drop-
wise and the resultant mixture was left to stir at −78 °C for 1 h.
Et3N (11.5 mL, 59.2 mmol) was added and the reaction mixture
(4R,5S)-2,2-Dimethyl-4-iodomethyl-5-benzyloxymethyl-1,3-
dioxolane 52
PPh3 (1.00 g, 3.81 mmol) and imidazole (324 mg, 4.76 mmol)
were added to a solution of 50 (800 mg, 3.17 mmol, >99 : 1 dr)
in PhMe–MeCN (v/v 5 : 1, 10 mL). I2 (966 mg, 3.81 mmol) was
then added and the resultant solution was heated at 60 °C for
1 h. Et2O (6 mL) was added and the resultant mixture was
washed sequentially with satd aq Na2S2O3 (10 mL), H2O
(10 mL) and brine (10 mL), then dried and concentrated
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6186–6200 | 6197

View Article Online
was allowed to warm to rt over 2 h. tert-Butyl (triphenylphos-
phoranylidene)acetate (11.1 g, 29.6 mmol) was added and the
resultant mixture was left to stir at rt for 16 h. Satd aq Na2CO3
(150 mL) was then added and the aqueous layer was extracted
with CH2Cl2 (3 × 60 mL). The combined organic extracts were
dried and concentrated in vacuo to give a 68 : 32 [(E) : (Z)]
mixture of diastereoisomers. Purification via flash column
chromatography (eluent 30–40 °C petrol–Et2O, 20 : 1) gave 32
as pale yellow oil (2.69 g, 38% from 54, >99 : 1 dr); [α]2D4 + 11.1
(c 1.0 in CHCl3); νmax (ATR) 2982, 2934, 2875 (C–H), 1715
(CvO), 1661 (CvC); δH (400 MHz, CDCl3) 1.31 (3H, d, J 6.1,
C(6)H3), 1.42 (3H, s, MeCMe), 1.44 (3H, s, MeCMe), 1.49 (9H,
s, CMe3), 3.84 (1H, dq, J 8.5, 6.1, C(5)H), 4.05 (1H, ddd, J 8.5,
6.1, 1.3, C(4)H), 6.04 (1H, dd, J, 15.7, 1.3, C(2)H), 6.75 (1H,
dd, J 15.7, 6.1, C(3)H); δC (100 MHz, CDCl3) 16.6 (C(6)), 26.7,
27.3 (CMe2), 28.0 (CMe3), 76.4 (C(5)), 80.7 (CMe3), 81.7
(C(4)), 109.1 (CMe2), 124.8 (C(2)), 142.1 (C(3)), 165.2 (C(1));
126.5 (p-Ph), 128.1, 128.6 (o,m-Ph), 141.3 (i-Ph), 172.1 (C(1));
+
m/z (ESI+) 392 ([M + H]+, 100%); HRMS (ESI+) C23H38NO4
([M + H]+) requires 392.2795; found 392.2805. Further elution
gave 65 as a colourless oil (440 mg, 55%, >99 : 1 dr);
[α]2D4 − 24.2 (c 1.0 in CHCl3); νmax (ATR) 2977, 2933, 2875,
2718 (C–H), 1727 (CvO); δH (400 MHz, CDCl3) 1.05 (3H, d,
J 6.7, MeCHMe), 1.08 (3H, d, J 6.7, MeCHMe), 1.31 (3H, s,
MeCMe), 1.32 (3H, d, J 5.6, C(6)H3), 1.38 (3H, s, MeCMe),
1.49 (9H, s, CMe3), 2.43 (1H, dd, J 15.5, 5.7, C(2)HA), 2.61
(1H, dd, J 15.5, 6.6, C(2)HB), 3.01 (1H, septet, J 6.7, CHMe2),
3.41 (1H, app dt, J 6.1, 4.1, C(3)H), 3.69 (1H, d, J 14.7,
NCHAHBPh), 3.78 (1H, d, J 14.7, NCHAHBPh), 3.68–3.77 (2H,
m, C(4)H, C(5)H), 7.19–7.37 (5H, m, Ph); δC (100 MHz,
CDCl3) 18.7 (C(6)), 19.7, 20.5 (CHMe2), 26.9, 27.4 (CMe2),
28.1 (CMe3), 35.4 (C(2)), 48.4 (CHMe2), 50.1 (CH2Ph), 54.4
(C(3)), 75.7, 83.9 (C(4), C(5)), 80.0 (CMe3), 108.0 (CMe2),
126.6 (p-Ph), 128.0, 128.6 (o,m-Ph), 141.0 (i-Ph), 172.3 (C(1));
+
m/z (ESI+) 265 ([M
+
Na]+, 100%); HRMS (ESI+)
m/z (ESI+) 392 ([M + H]+, 100%); HRMS (ESI+) C23H38NO4
C13H22NaO4+ ([M + Na]+) requires 265.1410; found 265.1419.
([M + H]+) requires 392.2795; found 392.2790.
Representative procedure for hydrogenolytic chemical
correlation: tert-butyl (S,S,S)-3-N-isopropylamino-4,5-O-
isopropylidene-4,5-dihydroxyhexanoate 67
Representative procedure for lithium amide conjugate addition:
tert-butyl (S,S,S)-3-(N-benzyl-N-isopropylamino)-4,5-O-
isopropylidene-4,5-dihydroxyhexanoate 65 and tert-butyl
(3R,4S,5S)-3-(N-benzyl-N-isopropylamino)-4,5-O-isopropylidene-
4,5-dihydroxyhexanoate 66
Pd(OH)2/C (50% w/w of substrate, 125 mg) was added to a solu-
tion of 65 (250 mg, 0.64 mmol, >99 : 1 dr) in MeOH (12.9 mL)
at rt. The resultant mixture was degassed and saturated with H2,
then left to stir under an atmosphere of H2 (1 atm) for 16 h. The
reaction mixture was then filtered through Celite® (eluent
MeOH) and the filtrate was concentrated in vacuo to give 67 as a
pale yellow solid (176 mg, 92%, >99 : 1 dr);32 mp 38–42 °C;
[α]2D4 + 2.7 (c 1.0 in CHCl3); νmax (ATR) 3345, 3325 (N–H),
2975, 2932, 2873 (C–H), 1707 (CvO); δH (400 MHz, CDCl3)
1.01 (3H, d, J 6.1, MeCHMe), 1.03 (3H, d, J 6.1, MeCHMe),
1.34 (3H, d, J 5.9, C(6)H3), 1.37 (3H, s, MeCMe), 1.39 (3H, s,
MeCMe), 1.46 (9H, s, CMe3), 2.37 (1H, dd, J 15.3, 6.1, C(2)
HA), 2.50 (1H, dd, J 15.3, 5.1, C(2)HB), 2.91 (1H, septet, J 6.1,
CHMe2), 3.06 (1H, app q, J 5.6, C(3)H), 3.57 (1H, dd, J 8.1,
5.6, C(4)H), 3.89 (1H, dq, J 8.1, 5.9, C(5)H); δC (100 MHz,
CDCl3) 19.2 (C(6)), 22.9, 23.8 (CHMe2), 27.0, 27.3 (CMe2),
28.1 (CMe3), 37.2 (C(2)), 45.5 (CHMe2), 53.4 (C(3)), 75.2
(C(5)), 80.5 (CMe3), 84.1 (C(4)), 107.9 (CMe2), 171.8 (C(1));
BuLi (2.5 M in hexanes, 1.60 mL, 4.00 mmol) was added drop-
wise to
a stirred solution of N-benzyl-N-isopropylamine
(0.68 mL, 4.13 mmol) in THF (15 mL) at −78 °C and stirring
was continued for 30 min. A solution of 32 (500 mg,
2.06 mmol, >99 : 1 dr) in THF (25 mL) was then added via
cannula and the reaction mixture was stirred for 2 h. Satd aq
NH4Cl (2 mL) was then added and the reaction mixture was par-
titioned between Et2O (75 mL) and H2O (75 mL). The aqueous
layer was extracted with Et2O (3 × 50 mL) and the combined
organic extracts were washed sequentially with 10% aq citric
acid (200 mL), satd aq NaHCO3 (200 mL) and brine (200 mL),
then dried and concentrated in vacuo to give an 82 : 18 mixture
of 65 and 66. Purification via flash column chromatography
(eluent 30–40 °C petrol–Et2O, 20 : 1) gave 66 as a yellow oil
(114 mg, 14%, >99 : 1 dr); [α]2D4 − 15.5 (c 1.0 in CHCl3); νmax
(ATR) 2977, 2935, 2872, 2720 (C–H), 1725 (CvO); δH
(400 MHz, CDCl3) 0.98 (3H, d, J 6.4, MeCHMe), 1.01 (3H, d,
J 6.4, MeCHMe), 1.09 (3H, d, J 6.5, C(6)H3), 1.30 (3H, s,
MeCMe), 1.32 (3H, s, MeCMe), 1.48 (9H, s, CMe3), 2.67 (1H,
dd, J 14.8, 8.0, C(2)HA), 2.70 (1H, dd, J 14.8, 5.9, C(2)HB),
3.18 (1H, septet, J 6.4, CHMe2), 3.18–3.23 (1H, m, C(3)H),
3.48 (1H, dd, J 8.3, 2.8, C(4)H), 3.54 (1H, d, J 14.0,
NCHAHBPh), 4.06 (1H, d, J 14.0, NCHAHBPh), 4.22 (1H, dq,
J 8.3, 6.5, C(5)H), 7.17–7.38 (5H, m, Ph); δC (100 MHz,
CDCl3) 17.1, 17.3 (CHMe2), 22.7 (C(6)), 26.9, 27.2 (CMe2),
28.1 (CMe3), 35.8 (C(2)), 48.7 (CHMe2), 51.1 (C(3)), 51.3
(CH2Ph), 73.1 (C(5)), 80.5 (CMe3), 85.3 (C(4)), 107.3 (CMe2),
m/z (ESI+) 302 ([M + H]+, 100%); HRMS (ESI+) C16H32NO4
+
([M + H]+) requires 302.2326; found 302.2331.
X-ray crystal structure determination for 59, 63, 64 and 67‡
Data were collected using a Nonius κ-CCD diffractometer with
graphite monochromated Mo-Kα radiation using standard pro-
cedures at 150 K. The structures were solved by direct methods
(SIR92); all non-hydrogen atoms were refined with anisotropic
thermal parameters. Hydrogen atoms were added at idealised
positions. The structure was refined using CRYSTALS.51
6198 | Org. Biomol. Chem., 2012, 10, 6186–6200
This journal is © The Royal Society of Chemistry 2012

View Article Online
X-ray crystal structure data for 59 [C16H31NO4]:32 M =
301.43, monoclinic, space group P21, a = 5.7565(2) Å, b =
19.4300(8) Å, c = 8.1386(3) Å, β = 96.168(2)°, V = 905.02(6)
Å3, Z = 2, μ = 0.078 mm−1, colourless prism, crystal dimensions
= 0.07 × 0.08 × 0.26 mm. A total of 2122 unique reflections
were measured for 5 < θ < 27 and 1907 reflections were used
in the refinement. The final parameters were wR2 = 0.094 and
R1 = 0.043 [I > −3.0σ(I)].
9 A. Defoin, J. Pires, I. Tissot, T. Tschamber, D. Bur, M. Zehnder and
J. Streith, Tetrahedron: Asymmetry, 1991, 2, 1209.
10 S. G. Davies, A. C. Garner, R. L. Nicholson, J. Osborne, P. M. Roberts,
E. D. Savory, A. D. Smith and J. E. Thomson, Org. Biomol. Chem.,
2009, 7, 2604.
11 (a) F. Cardona, S. Valenza, A. Goti and A. Brandi, Eur. J. Org. Chem.,
1999, 1319; (b) T. Shibue, T. Hirai, I. Okamoto, N. Morita, H. Masu,
I. Azumaya and O. Tamura, Chem.–Eur. J., 2010, 16, 11678.
12 J. F. Costello, S. G. Davies and O. Ichihara, Tetrahedron: Asymmetry,
1994, 5, 1999.
13 For a review of the use of enantiopure lithium amides in synthesis, see: S.
G. Davies, P. D. Price and A. D. Smith, Tetrahedron: Asymmetry, 2005,
16, 2833.
14 S. G. Davies, M. J. Durbin, E. C. Goddard, P. M. Kelly, W. Kurosawa,
J. A. Lee, R. L. Nicholson, P. D. Price, P. M. Roberts, A. J. Russell,
P. M. Scott and A. D. Smith, Org. Biomol. Chem., 2009, 7, 761.
15 When THF was used as the solvent, the corresponding β,γ-unsaturated
ester was also isolated upon conjugate addition of lithium amides (R)-1
and (S)-1 to cis-dioxolane containing α,β-unsaturated esters 3 and 4 at
−78 °C. However, this deleterious side reaction was completely sup-
pressed when Et2O was used as the solvent and the conjugate addition
reactions were carried out at −20 °C.
16 The conjugate additions to α,β-unsaturated ester 3 were conducted in the
enantiomeric series but are represented here as the antipodes for ease of
comparison.
17 S. G. Davies and E. M. Foster, unpublished results; these data will be
reported in full in a forthcoming publication from this laboratory.
18 The conjugate additions to α,β-unsaturated ester 17 were conducted in the
enantiomeric series but are represented here as the antipodes for ease of
comparison.
19 L. A. Paquette and S. Bailey, J. Org. Chem., 1995, 60, 7849.
20 R. Appel, Angew. Chem., Int. Ed. Engl., 1975, 14, 801.
21 (a) K. Mori, Tetrahedron, 2009, 65, 3900; (b) U. Staempfli and
M. Neuenschwander, Helv. Chim. Acta, 1988, 71, 2022; (c) S. A. Ali and
M. I. M. Wazeer, J. Chem. Soc., Perkin Trans. 1, 1988, 3, 597.
22 (a) S. Thomas, T. Huynh, V. Enriquez-Rios and B. Singaram, Org. Lett.,
2001, 3, 3918; (b) G. B. Fisher, J. Harrison, J. C. Fuller, C. T. Goralski
and B. Singaram, Tetrahedron Lett., 1992, 33, 4533; (c) M. Harmata and
P. Zheng, Heterocycles, 2009, 77, 279.
23 (a) E. Nakamura, D. Machii and T. Inubushi, J. Am. Chem. Soc., 1989,
111, 6849; (b) W. F. Bailey and M. W. Carson, Tetrahedron Lett., 1999,
40, 5433; (c) K. I. Min, T. H. Lee, C. P. Park, Z. Y. Wu, H. H. Girault,
I. Ryu, T. Fukuyama, Y. Mukai and D. P. Kim, Angew. Chem., Int. Ed.,
2010, 49, 7063.
X-ray crystal structure data for 63 [C28H39NO4]:32 M =
453.62, monoclinic, space group P21, a = 7.9717(2) Å, b =
12.5518(3) Å, c = 13.0678(3) Å, β = 95.5854(9)°, V = 1301.35
(5) Å3, Z = 2, μ = 0.076 mm−1, colourless block, crystal dimen-
sions = 0.18 × 0.23 × 0.27 mm. A total of 3107 unique reflec-
tions were measured for 5 < θ < 27 and 2549 reflections were
used in the refinement. The final parameters were wR2 = 0.117
and R1 = 0.046 [I > −3.0σ(I)].
X-ray crystal structure data for 64 [C28H39NO4]:32 M =
453.62, orthorhombic, space group P212121, a = 9.3260(2) Å,
b = 13.7383(3) Å, c = 20.3169(6) Å, V = 2603.07(11) Å3, Z = 4,
μ = 0.076 mm−1, colourless block, crystal dimensions = 0.12 ×
0.15 × 0.16 mm. A total of 3316 unique reflections were
measured for 5 < θ < 27 and 2639 reflections were used in the
refinement. The final parameters were wR2 = 0.098 and R1 =
0.045 [I > −3.0(I)].
X-ray crystal structure data for 67 [C16H31NO4]:32 M =
301.43, monoclinic, space group C 2, a = 22.1832(6) Å, b =
5.8375(2) Å, c = 14.5119(4) Å, β = 105.8953(12)°, V = 1807.36
(9) Å3, Z = 4, μ = 0.078 mm−1, colourless prism, crystal dimen-
sions = 0.14 × 0.17 × 0.90 mm. A total of 2239 unique reflec-
tions were measured for 5 < θ < 27 and 2239 reflections were
used in the refinement. The final parameters were wR2 = 0.121
and R1 = 0.049 [I > −3.0(I)].
24 D. H. R. Barton and S. W. McCombie, J. Chem. Soc., Perkin Trans. 1,
1975, 1574.
Notes and references
25 W. F. Bailey and E. R. Punzalan, J. Org. Chem., 1990, 55, 5404.
26 N. Faucher, Y. Ambroise, J. C. Cintrat, E. Doris, F. Pillon and
B. Rousseau, J. Org. Chem., 2002, 67, 932.
1 S. Masamune, W. Choy, J. S. Petersen and L. R. Sita, Angew. Chem., Int.
Ed. Engl., 1985, 24, 1.
2 O. I. Kolodiazhnyi, Tetrahedron, 2003, 59, 5953.
27 It was necessary to add Et3N to the reaction mixture, as otherwise the
liberation of HI catalysed in situ deprotection of the acetonide protecting
group.
28 In an effort to improve the diastereoselectivity of the olefination step an
alternative procedure was also investigated whereby Swern oxidation of
47 was followed by a MeMgBr mediated Wadsworth–Emmons olefina-
tion (see ref. 29) of intermediate aldehyde 48. This gave (E)-32 as the
sole diastereoisomer (>99 : 1 dr), which was isolated in 7% overall yield
from diol 45. The poorer yield in this case is presumably due to the need
to isolate the volatile aldehyde 48.
3 (a) S. G. Davies, G. J. Hermann, M. J. Sweet and A. D. Smith, Chem.
Commun., 2004, 1128; (b) S. G. Davies, A. M. Fletcher, G. J. Hermann,
G. Poce, P. M. Roberts, A. D. Smith, M. J. Sweet and J. E. Thomson,
Tetrahedron: Asymmetry, 2010, 21, 1635; (c) S. G. Davies, J. A. Lee,
P. M. Roberts, J. E. Thomson and J. Yin, Tetrahedron, 2011, 67, 6382.
4 K. B. Sharpless, W. Amberg, Y. L. Bennani, G. A. Crispino, J. Hartung,
K. S. Jeong, H. L. Kwong, K. Morikawa, Z. M. Wang, D. Xu and
X. L. Zhang, J. Org. Chem., 1992, 57, 2768.
5 (a) S. Y. Ko, A. W. M. Lee, S. Masamune, L. A. Reed, K. B. Sharpless
and F. J. Walker, Tetrahedron, 1990, 46, 245; (b) P. C. Ray and
S. M. Roberts, Tetrahedron Lett., 1999, 40, 1779.
6 D. Meyer, J. C. Poulin, H. B. Kagan, H. L. Pinto, J. L. Morgat and
P. Fromageot, J. Org. Chem., 1980, 45, 4680.
29 T. D. W. Claridge, S. G. Davies, J. A. Lee, R. L. Nicholson, P.
M. Roberts, A.J. Russell, A. D. Smith and S. M. Toms, Org. Lett., 2008,
10, 5437.
30 In both cases the corresponding (Z)-configured diastereoisomer [(S,S,Z)-
4,5-O-isopropylidene-4,5-dihydroxyhex-2-enoate] was also isolated in
>99 : 1 dr (in 16% overall yield from 53, and in 17% overall yield from
54).
31 N. Sewald, K. D. Hiller, M. Körner and M. Findeisen, J. Org. Chem.,
1998, 63, 7263.
32 Crystallographic data (excluding structure factors) for compounds 59, 63,
64 and 67 have been deposited with the Cambridge Crystallographic Data
Centre as supplementary publication numbers CCDC 846621–846624,
respectively‡.
33 The absolute (3R,4S,5S,αS)-configuration within 3,4-syn-72 was also
independently verified by chemical correlation: an authentic sample of
(3S,4R,5R,αR)-72 was prepared via acetonide protection of tert-butyl
7 (a) M. P. Bonner and E. R. Thornton, J. Am. Chem. Soc., 1991, 113,
1299; (b) F. Calderón, E. G. Doyagüez and A. F. Mayoralas, J. Org.
Chem., 2006, 71, 6258; (c) F. Calderón, E. G. Doyagüez, P. H. Cheong,
A. Fernández-Mayoralas and K. N. Houk, J. Org. Chem., 2008, 73,
7916; (d) D. A. Evans, J. Bartroli and T. L. Shih, J. Am. Chem. Soc.,
1981, 103, 2127; (e) L. M. Geary and P. G. Hultin, Tetrahedron: Asymme-
try, 2009, 20, 131.
8 (a) S. G. Davies, C. J. R. Hedgecock and J. M. McKenna, Tetrahedron:
Asymmetry, 1995, 6, 827; (b) T. Ishikawa, K. Nagai, T. Kndoh and
S. Saito, Synlett, 1995, 1171; (c) T. Cailleau, J. W. B. Cooke,
S. G. Davies, K. B. Ling, A. Naylor, R. L. Nicolson, P. D. Price,
P. M. Roberts, A. J. Russell, A. D. Smith and J. E. Thomson, Org.
Biomol. Chem., 2007, 5, 3922.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6186–6200 | 6199

View Article Online
(3S,4R,5R,αR)-3-[N-benzyl-N-(α-methylbenzyl)amino]-4,5-dihydroxy-
hexanoate; this sample was found to have identical 1H and 13C NMR
spectroscopic data (and an equal and opposite specific rotation) to those
of (3R,4S,5S,αS)-72. For the preparation of tert-butyl (3S,4R,5R,αR)-3-
39 A. B. Pangborn, M. A. Giardello, R. H. Grubbs, R. K. Rosen and
F. J. Timmers, Organometallics, 1996, 15, 1518.
40 R. J. Anderson, R. M. Dixon and B. T. Golding, J. Org. Chem., 1992,
437, 227.
[N-benzyl-N-(α-methylbenzyl)amino]-4,5-dihydroxyhexanoate,
see:
41 N. A. Ivanova, Z. R. Valiullina, O. V. Shitikova and M. S. Muftakhov,
Russ. J. Org. Chem., 2007, 43, 742.
42 J. S. Yadav, B. V. S. Reddy and K. Srinivasa Reddy, Tetrahedron, 2003,
59, 5333.
43 B. Haefele, D. Schrocter and V. Jaeger, Angew. Chem., 1986, 98, 89.
44 B. Moon Kim, S. J. Bae, S. M. So, H. T. Yoo, S. K. Chang, J. H. Lee and
J. S. Kang, Org. Lett., 2001, 3, 2349.
45 B. Li, Y. Xuemer, K. Yang and E. Fu, Synth. Commun., 2005, 35, 2603.
46 W. Flezmann, D. Castagnolo, D. Rosenbeiger and J. Mulzer, J. Org.
Chem., 2007, 72, 2182.
K. Csatayová, S. G. Davies, J. A. Lee, P. M. Roberts, A. J. Russell,
J. E. Thomson and D. L. Wilson, Org. Lett., 2011, 13, 2606.
34 J. F. Costello, S. G. Davies and O. Ichihara, Tetrahedron: Asymmetry,
1994, 10, 1999.
35 (a) A. S. Galiano-Roth and D. B. Collum, J. Am. Chem. Soc., 1989, 111,
6772; (b) J. F. Remenar, B. L. Lucht and D. B. Collum, J. Am. Chem.
Soc., 1997, 119, 5567; (c) G. Hilmersson and O. Davidsson, J. Org.
Chem., 1995, 60, 7660; (d) R. Sott, J. Granander and G. Hilmersson,
Chem.–Eur. J., 2002, 8, 2081.
36 S. G. Davies, E. M. Foster, C. R. McIntosh, P. M. Roberts, T. E. Rosser, A.
D. Smith and J. E. Thomson, Tetrahedron: Asymmetry, 2011, 22, 1035.
37 (a) S. G. Davies and O. Ichihara, Chem. Commun., 1990, 1554;
(b) W. Bauer, T. Laube and D. Seebach, Chem. Ber., 1985, 118, 764.
38 Unfortunately, no firm conclusions could be drawn from 1H NMR NOE
analyses of both 31 and 32 (in CDCl3). A strong reciprocal enhancement
between the C(3)H and C(6)H3 protons within 31 was observed, although
this effect was far less pronounced for 32.
47 D. T. Fox and C. D. Poulter, J. Org. Chem., 2005, 70, 1978.
48 H. Kakinuma, Y. Tsuchiya, M. Tanaka, S. Horito and H. Hashimoto,
Carbohydr. Res., 1994, 264, 237.
49 K. Mori and T. Takeuchi, Tetrahedron, 1988, 44, 333.
50 Y. Kita, F. Itoh, O. Tamura, Y. Y. Ke, T. Miki and Y. Tamura, Chem.
Pharm. Bull., 1989, 37, 1446.
51 P. W. Betteridge, J. R. Carruthers, R. I. Cooper, C. K. Prout and
D. J. Watkin, J. Appl. Crystallogr., 2003, 36, 1487.
6200 | Org. Biomol. Chem., 2012, 10, 6186–6200
This journal is © The Royal Society of Chemistry 2012