Enantioselective homoallyl-cyclopropanation of dibenzylideneacetone by modified allylindium halide reagents-rapid access to enantioenriched 1-styryl-norcarene

Article abstract of DOI:10.1016/j.tet.2006.05.003

Dibenzylideneacetone (8) reacts with in situ-generated allylindium halide reagents to yield the product of a homoallyl-cyclopropanation reaction: 2-(3″-butenyl)-1,1-bis[(E)-2′-phenylethenyl]cyclopropane (9), which proceeds via step-wise cleavage of the C{double bond, long}O bond and delivery of two allyl fragments from the reagent. A range of enantiomerically enriched ligands have been tested as stoichiometric asymmetric modifiers for this process. Enantiopure compounds such as cinchona alkaloids, ephedra, aminoalcohols and tartaric acid derivatives, which have proven of utility as asymmetric modifiers for the indium-mediated allylation of aldehydes and ketones, were very inefficient in the process 8→9. However, mandelic acid derivatives, in particular mandelates, were found to be of significant potential. The absolute stereochemistry of the cyclopropane 9 has been determined by degradation to 1,1-dicarboxymethyl-2-butylcyclopropane, converging with an independent enantioselective synthesis starting from hexene. Under optimised conditions, viz. using allylindium iodide reagents and working-up with aqueous Na₂SO₃ to avoid iodine-mediated polymerisation, (S)-9 can be generated in 86% yield and with (S)-methyl mandelate as modifier useful enantiopurity (94/6 er) was observed. The cyclopropane product ((S)-9) undergoes RCM using standard conditions to afford a norcarene unit ((1S,6S)-1-(E)-2′-(phenylethenyl)-bicyclo[4.1.0]hept-2-ene) without loss of enantiopurity.

Full text of DOI:10.1016/j.tet.2006.05.003

Tetrahedron 62 (2006) 11402–11412
Enantioselective homoallyl-cyclopropanation of
dibenzylideneacetone by modified allylindium halide
reagents—rapid access to enantioenriched 1-styryl-norcarene
a
a
Guy C. Lloyd-Jones,a, Philip D. Wall, Jennifer L. Slaughter,
*
Alexandra J. Parkerb and David P. Laffanc
aThe Bristol Centre for Organometallic Catalysis, School of Chemistry, University of Bristol, Cantock’s Close, Bristol BS8 1TS, UK
bAstraZeneca, Avlon Works, Severn Road, Hallen, Bristol BS10 7ZE, UK
cAstraZeneca, Macclesfield Works, Hurdsfield Industrial Estate, Macclesfield, Cheshire SK10 2NA, UK
Received 3 March 2006; revised 29 April 2006; accepted 4 May 2006
Available online 2 June 2006
Abstract—Dibenzylideneacetone (8) reacts with in situ-generated allylindium halide reagents to yield the product of a homoallyl-cyclopro-
panation reaction: 2-(300-butenyl)-1,1-bis[(E)-20-phenylethenyl]cyclopropane (9), which proceeds via step-wise cleavage of the C]O bond
and delivery of two allyl fragments from the reagent. A range of enantiomerically enriched ligands have been tested as stoichiometric asym-
metric modifiers for this process. Enantiopure compounds such as cinchona alkaloids, ephedra, aminoalcohols and tartaric acid derivatives,
which have proven of utility as asymmetric modifiers for the indium-mediated allylation of aldehydes and ketones, were very inefficient in
the process 8/9. However, mandelic acid derivatives, in particular mandelates, were found to be of significant potential. The absolute stereo-
chemistry of the cyclopropane 9 has been determined by degradation to 1,1-dicarboxymethyl-2-butylcyclopropane, converging with an inde-
pendent enantioselective synthesis starting from hexene. Under optimised conditions, viz. using allylindium iodide reagents and working-up
with aqueous Na2SO3 to avoid iodine-mediated polymerisation, (S)-9 can be generated in 86% yield and with (S)-methyl mandelate as modifier
useful enantiopurity (94/6 er) was observed. The cyclopropane product ((S)-9) undergoes RCM using standard conditions to afford a norcarene
unit ((1S,6S)-1-(E)-20-(phenylethenyl)-bicyclo[4.1.0]hept-2-ene) without loss of enantiopurity.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
metal or indium(I) iodide, in combination with an allyl ha-
lide,8 to effect allylation,9 most often in water (Barbier type
conditions) or in an aprotic organic solvent such as THF or
DMF. The mildness of the conditions lend themselves to
the synthesis of complex natural products, particularly carbo-
hydrates, without the need for extensive use of protecting
groups.10 Unsurprisingly, the growing range of substrates,
reagents and conditions that have been developed to harness
the power of indium in organic synthesis has been the subject
of numerous reviews.11 Although the major focus in terms
of substrate has been carbonyl compounds,11 other electro-
philes including imines,12 alkenes and alkynes,13 acetals,14
epoxides,15 allylic sulfonium salts16 and pyridinium salts17
have proven of utility. There has also been progress towards
the development of more ‘green’ conditions for organo-
indium-mediated processes, which include solvent free reac-
tions,18 use of sc-CO219 and ionic liquids as solvent,20 and the
development of catalytic indium methodology.21
The use of an organoindium compound1 as a reagent to facil-
itate C–C bond formation was first described by Gilman and
Jones in 1940.2 Ph3In, generated by oxidative transfer of
phenyl from Ph2Hg to indium metal, was reacted with a range
of carbonyl compounds, including benzylideneacetone with
which it underwent clean 1,4-addition. The direct generation
of organoindium halide reagents from simple organohalides
and indium metal, thereby avoiding the use of R2Hg,3,4 is ex-
tremely slow. The procedure is generally only practical with
MeI and EtI,5 which function as both reactant and solvent. An
exception to this is the highly reactive ‘Riecke’ form of in-
dium, which facilitates the preparation of indium-based Re-
formatsky reagents.6 However, it was the pioneering work of
Araki et al.7 on the facile reaction of allylic bromides and
iodides with commercially available indium powder that
marked the onset of a period of extremely rapid development
in the field. The manifold applications that have emerged
since then have predominantly involved the use of indium
The indium-mediated allylation of aldehydes and asymmet-
ric ketones (1), leads, via alkoxide 2, to homoallylic alcohols
(3) in which a new stereogenic centre is generated at the hy-
droxyl-bearing carbon. The stereoselectivity of this reaction,
* Corresponding author. Fax: +44 117 929 8611; e-mail: guy.lloyd-jones@
bris.ac.uk
0040–4020/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.05.003

G. C. Lloyd-Jones et al. / Tetrahedron 62 (2006) 11402–11412
11403
Scheme 1, has been of much interest, with many investiga-
tions into the use of proximal functionality to ‘direct’ the
stereochemical course of reaction and thus engender dia-
stereoselectivity.11 The diastereoselective indium-mediated
allylation of other functionality, such as imines and aldi-
mines,22 hydrazones,23 azetidinones24 and cyclopropenes25
has also been achieved.
engender asymmetric allyl transfer (allyl iodide, indium,
THF, 4 A MS) to aryl hydrazones (up to 96/4 er),31 as well
˚
as the use of indium catalysts for asymmetric allyl transfer
from Sn,32 bode well for the developments in this area.33
In 1998 we reported that a,b-unsaturated ketones and
aldehydes (5) react with in situ-generated allylindium
halide reagents34 in THF to yield, on acidic aqueous
work-up of the indate (6+LiBr), homoallyl-vinylcyclopro-
panes 7, Scheme 3.35
Scheme 1. A generic scheme for indium-mediated allylation of an aldehyde
(R0¼H) or asymmetric ketone (RsR0) leading to a new stereogenic centre.
Scheme 3. The homoallyl-cyclopropanation of a,b-unsaturated ketones
(R¼alkyl, aryl) and aldehydes (R¼H) mediated by allylindium halide re-
agents in THF at ambient temperature. When R¼CH]CHR0, C(1) in 7 is
nonstereogenic and the configuration of C(2) is determined in step II.
The asymmetric allylation of ketones and aldehydes by way
of stoichiometrically-modified allyl–metal reagents (e.g.,
based on Zn, Sn, B) is a reaction of considerable current
interest, particularly in relation to the enantioselective
construction of quaternary carbon centres.26 It is perhaps
surprising that despite the extensive investigations into dia-
stereoselective indium-mediated allylation, the correspond-
ing developments in enantioselective allylation have been
much less marked. The first examples were from Araki
et al., who reported on the asymmetric modification of
indium-mediated Reformatsky reactions with cinchona
alkaloids.27 The same modifiers were subsequently used by
Loh et al. for the indium-mediated enantioselective allylation
(and propargylation) of aldehydes.28 For example, allylation
of PhCHO by allyl bromide/indium in THF/hexane in the
presence of (+)-cinchonine (2 equiv) proceeds in up to 86/
14 er at ꢀ78 ꢁC. The use of cinchonine as asymmetric
modifier in the indium-mediated allylation of imines has
also been reported (up to 72/28 er).29 More recently, Sin-
garam et al. reported on the application of simple aminoalco-
hols, e.g., 4, as modifiers, for aldehyde allylation Scheme 2.30
Under optimised conditions, the best modifier (4, 2.0 equiv)
facilitates selective generation of one enantiomer (up to 96.5/
3.5 er) in the allylation of a range of aldehydes at ꢀ78 ꢁC. At
higher temperatures, the selectivity drops markedly.
The addition of allylindium halide reagents to a,b-unsatu-
rated ketones and aldehydes (5) followed by acidic hydroly-
sis gives homoallyl alcohol-type products arising from
[1,2]-addition to the carbonyl (Step I, Scheme 3) rather
than [1,4]-addition to the enone.36 We thus proposed that
the generation of 7 involves an acid-catalysed C–O bond
cleavage37 of an indium alkoxide (6) concomitant with a sec-
ond allyl unit being delivered to the homoallylic terminus
from indium (Step II, Scheme 3).38 In the case of the reac-
tion of a symmetrically substituted bis-a,b-unsaturated
ketone, the intermediate 6, in which R¼CH]CHR0, is for-
mally achiral (ignoring issues of indium-based stereo-
centres) and the homoallylic alkene terminus is pro-chiral.
Consequently, the cyclopropane stereogenic centre39 (C(2)
in 7, Scheme 3) is generated not on the initial allylation
(step I), but on the C–O cleavage/ring-closure/allylation
(step II). In parallel with the efforts towards the discovery
and development of enantioselective indium-mediated allyl-
ations, vide supra, we have been testing whether the C–C
bond-forming event in the homoallyl-cyclopropanation
reaction can be rendered asymmetric by virtue of a simple
stoichiometric chiral modifier. The homoallyl-cyclopropa-
nation process is mechanistically distinct38 from the conven-
tional asymmetric indium-mediated allylation of carbonyls,
Scheme 2, and it is thus also of interest to determine whether
the same types of asymmetric modifier are effective. Herein
we report our preliminary studies in this area using the bis-
a,b-unsaturated ketone, dibenzylideneacetone (‘dba’, 8), as
a prototype substrate for the asymmetric homoallyl-cyclo-
propanation to generate 9 (Scheme 4; cf. 5 and 7, in
Scheme 2. Asymmetric indium-mediated allylation of benzaldehyde em-
ploying 1,2-diphenyl-2-aminoethanol (1S,2R-4) as super-stoichiometric
enantiopure modifier (indium/4/allyl bromide/pyr/PhCHO¼2/2/2/2/1).
A key issue of such reactions is the requirement of stoichio-
metric or super-stoichiometric quantities of enantiomerically
enriched, ideally enantiomerically pure, chiral modifier. In
practice, this limits the choice of modifiers that will be gen-
uinely applicable in synthesis to those derived from the chiral
pool or products of commercial resolution. Recent reports
from Cook et al., on the use of catalytic (10 mol %) quantities
of 2,20-dihydroxy-1,10-binaphthalene, and derivatives, to
Scheme 4. Reaction of dibenzylideneacetone (8) used to test potential
asymmetric modifiers (L*) in the homoallyl-cyclopropanation reaction
(/9); see also Table 1.

11404
G. C. Lloyd-Jones et al. / Tetrahedron 62 (2006) 11402–11412
Scheme 3, where R¼PhCH]CH, R0¼Ph). We present a
low-cost, commercially available and enantiomerically pure
modifier, which yields 9 under technically simple and scale-
able conditions, in up to 94/6 er40 and in good yield (86%).
that valinol and phenylglycinol gave no selectivity, we con-
cluded that the hydroxyl-bearing, rather than amine-bearing,
stereocentre was crucial. To further explore this issue, (S)-
mandelic acid ((S)-11), was converted to (S)-mandelamine
((S)-13), via (S)-mandelamide ((S)-12). However, like vali-
nol and phenylglycinol, using the aminoalcohol (S)-13 as
modifier gave racemic 9 (Table 1, entry 2). Somewhat unex-
pectedly, (S)-mandelamide ((S)-12) (1.0 equiv) gave the
most encouraging combination of selectivity and yield: 25/
75 er (S)-(+)-9, 27% yield, (Table 1, entry 3). However, al-
though this modifier was also efficient in a semi-catalytic
stoichiometry (Table 1, entry 4), use of super-stoichiometric
quantities to increase the enantioselectivity led to plummet-
ing yields (entries 5 and 6).
2. Results and discussion
Using the conditions described in our original report for the
homoallyl-cyclopropanation of a,b-unsaturated ketones
and aldehydes35 (see Scheme 3), we began by testing a range
of commercially available enantiomerically pure aminoalco-
hols and diols as modifiers for the reaction (8/9). An exten-
sive screening of chiral HPLC columns and conditions
facilitated the near base-line resolution of the chiral racemic
hydrocarbon (ꢂ)-9by use of a rigorously alcohol-free hexane
eluent in combination with an OD-H column, thus allowing
rapid assay of the enantioselectivity. Based on the work of
Araki et al.,27 and Loh et al.,28 an obvious starting point for
modifier screening was the cinchona alkaloids. Whilst there
was some selectivity in the reaction (8/9) it was low (max-
imum 44/56 er) and the yields were moderate (42–51%).
In order to address the issue of low yields in the asymmetric
modification of the homoallyl-cyclopropanation reaction
(8/9), we employed factorial based experimental design,41
to analyse the reaction both in the presence and absence of
1.0 equiv (S)-mandelamide modifier ((S)-12). Whilst some
minor interactions were discovered, a key factor that
emerged was a strong dependence of the yield, and an in-
verse dependence of the enantioselectivity, on the allyl-
indium halide reagent34 concentration at the point when
the ketone 8 is added: high concentrations gave higher ulti-
mate yields of 9 and lower concentrations gave higher enan-
tioselectivity. However, the efficient generation of the
allylindium halide reagent requires high concentration of
allyl halide. We thus compromised by generating the reagent
at 2.0 M (in indium) and then diluting to 0.11 M before add-
ing the ketone 8, resulting in an improved yield of 9 (73%)
A key point that emerged was that racemic cyclopropane
(ꢂ)-9 was obtained if the alkaloid was added after the ketone
8. Whether the modifier was added before or with the ketone
made no difference to the selectivity. Testing a range of
ephedra, tartaric acid derivatives and aminoalcohols led to
improved selectivities but with lower yields. The most prom-
ising selectivity (79/21 er; 21% yield) was obtained with
threonine methyl ester (10, Table 1, entry 1), and on the basis
Table 1. The effect of a range of enantiomerically pure alcohols, especially a-hydroxy carbonyl derivatives, as modifiers (L* in Scheme 4) for the asymmetric
homoallyl-cyclopropanation of dibenzylideneacetone (8, 0.11 M in THF, rt) to give cyclopropane 9
Entry
Modifier
Stoichiometry L*/8
Yield 9/%a
er 9 (R/S)b
1
2
3
4
5
6
7
8
(2S,3R)-10
(S)-13
(S)-12
(S)-12
(S)-12
(S)-12
(S)-12
(R)-14
(S)-15
(S)-16
(S)-17
(S)-11
(S)-18
(S)-18
(S)-19
(S)-20
(S)-21
(S)-22
(S)-23
1.0
1.0
1.0
0.5
2.0
4.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
2.0
1.0
1.0
1.0
1.0
1.0
21c
68c
27c
47c
8c
79/21c
50/50c
25/75c
32/68c
15/85c
c
0c,d
73
22
0
—
34/66
73/27
—
9
10
11
12
13
14
15
16
17
18
19
70
70
67
71
39
38
38
70
66
70
50/50
50/50
42/58
20/80
11/89
34/66
50/50
24/76
25/75
48/52
a
Isolated yield after silica-gel chromatography.
b
c
d
Determined by chiral HPLC (OD-H, see Section 4 for details).
Reaction conducted at 1.0 M 8.
1E-1-Phenyl-3-E-styryl-hexa-1,3,5-triene was isolated in 49% yield.

G. C. Lloyd-Jones et al. / Tetrahedron 62 (2006) 11402–11412
11405
with moderate selectivity (34/66 er) Table 1, entry 7. With
sufficient material in hand, we then established the absolute
configuration of (S)-(+)-9 (34/66 er) by degradation (9/
24/25) to converge with an asymmetric synthesis of (S)-
25 starting from hexene (/26/27/25), a key feature
being the double SN2-inversion (C(1) then C(2)) in 27 with
dimethyl malonate anion, Scheme 5.
stereogenic centre ((S)-22) was of similar efficacy (entry
18, compare entry 17). As with mandelamide, the hydroxy
group in the mandelate is crucial: the O-methylated methyl
mandelate((S)-23)gaveessentiallyracemicproduct, entry19.
Choosing then to explore a range of methyl mandelate ana-
logues in which the aryl ring is varied, we sought a high
yielding synthesis for such modifiers in high enantiomeric
purity. The vanadium-catalysed asymmetric addition of tri-
methylsilylcyanide (TMSCN) to aldehydes (28/29) devel-
oped by Belokon’ and North et al., Scheme 6,43 proved
efficient.
Scheme 6. The one-pot procedure employed to make methyl mandelate
derivatives (89/11/99/1 er) for use as modifiers, L* in asymmetric
homoallyl-cyclopropanation, Scheme 4. Conditions: (i) TMSCN, vanadyl
(1R,2R)-N,N0-bis(3,5-di-tert-butylsalicylidenato)-1,2-cyclohexane diamine
(0.6 mol %), CH2Cl2; (ii) Et2O, MeOH, HCl; (iii) H2O, NaHCO3.
Scheme 5. The converging degradation of 9 (34/66 er) and asymmetric
synthesis of (S)-25 employed to deduce the absolute configuration of (S)-
(+)-9. Conditions: (i) Cp2Zr(Cl)H, CH2Cl2, then H2O; (ii) cat. RuCl3,
NaIO4, MeCN, CCl4, H2O; (iii) TMSCHN2, MeOH, toluene; (iv) cat.
[(DHQD)2PYR], cat. KOsO4$H2O, K3Fe(CN)6, tert-BuOH, H2O; (v)
SOCl2, CCl4; (vi) cat. RuCl3$3H2O, NaIO4, H2O; (vii) CH2(CO2CH3)2,
DME, NaH (2.5 equiv).
Testing the methodology with benzaldehyde (28a), we found
that we could smoothly convert the resulting silylated
mandelonitrile ((S)-29a) in situ (MeOH/HCl; then H2O) to
(S)-methyl mandelate ((S)-18, 91/9 er) via the imidate
((S)-30a). To address the issue of screening nonenantiopure
mandelate modifiers ((S)-31) arising from the imperfect
enantioselectivity in the initial TMS-cyanohydrin genera-
tion (28/29), we tested the relationship between the enan-
tiopurity of the simple mandelate modifier (50, 60, 75, 90
and 100% (S)-18) and that of the product (S)-9 from the
asymmetric homoallyl-cyclopropanation reaction, which
was found to be linear, Figure 1.
Chiral GC analysis of the two samples of 25 (one from each
route), demonstrated conclusively that (S)-(+)-25 (34/66 er)
was derived from (+)-9, which establishes the S configura-
tion for the latter. Returning back to the asymmetric homo-
allyl-cyclopropanation, we screened a range of a-hydroxy
carbonyl derivatives and analogues of mandelamide (S)-
12, for their potential as modifiers, Table 1, entries 8–19.
Using N-alkylated analogues of the mandelamide (N-Bu
and N-Bn derivatives (R)-14 and (S)-15, respectively) re-
sulted in low or no yield, respectively (entries 8 and 9).
The phenyl and hydroxyl groups in mandelamide (S)-12
proved crucial for selectivity ((S)-lactic acid amide ((S)-
16) and (S)-O-acetylmandelamide ((S)-17) both giving race-
mic 9 in 70% yield, entries 10 and 11), however, the amide
group was not: the parent mandelic acid ((S)-11) gave 9 in
67% yield with some, albeit lower, selectivity, entry 12.
Somewhat surprisingly, the methyl ester ((S)-methyl mande-
late, (S)-18), a relatively inexpensive item of commerce,42
proved far superior, giving (S)-9 in good yield and high se-
lectivity, entry 13. Moreover, doubling the stoichiometry
of (S)-18, led to an improved selectivity without a dramatic
reduction in yield as was observed with the mandelamide
((S)-12, entry 14, compared to entry 5). The hydrogen at
the stereogenic centre in (S)-18 appears to be important:
methyl atrolactate ((S)-19) gave reduced selectivity (entry
15) whilst the homologated ester (b-hydroxy carbonyl, (S)-
20) was completely ineffective, entry 16. Ester variants of
mandelate (S)-18 were also screened, without improvement,
for example, the benzyl ester (S)-21 gave slightly lower se-
lectivity, entry 17. Interestingly, the homologue of this ester,
maintaining the a-hydroxy carbonyl relationship, but plac-
ing a methylene spacer between the aromatic ring and the
We were also concerned about the effect of any vanadium or
salen ligand contaminants in the potential modifiers, Scheme
6, but we found that the crude sample of (S)-18 (91/9 er) as
obtained after imidate hydrolysis gave (S)-9 in 75/25 er, fully
consistent with that predicted by the linear correlation,
Figure 1. On this basis, we then screened a range of aryl-
variants of 18 as modifiers, Table 2.
Despite some small improvements in selectivity, e.g., the
4-methoxy- and 1-naphthyl-derivatives (S)-31d and (S)-
31e, Table 2, entries 4 and 5, the major effect of aryl substi-
tution was a decrease in the yield of 9, and in balance the
simple methyl mandelate modifier ((S)-18) appeared best.
Since both enantiomers of mandelic acid and methyl mande-
late are commercially available,42 we focussed on optimising
the conditions to use this modifier. In our earlier work on the
homoallyl-cyclopropanation process, we reported that yields
were made more reproducible and substantially improved by
addition of LiBr to the reaction mixture after the initial reac-
tion of the ketone with the allylindium reagent and before the
aqueous acidic work-up.35 We proposed that the formation
of an indate type intermediate could stabilise allylindium
moieties towards acidic hydrolysis. Given the pronounced

11406
G. C. Lloyd-Jones et al. / Tetrahedron 62 (2006) 11402–11412
addition of LiBr, or LiI, resulted in the isolation of a dark, in-
tractable, tar-like product, Table 2, entry 7. The H NMR
1
spectra of these products suggested them to be polymeric,
however, TLC analysis prior to the aqueous acidic work-up
revealed the presence of some cyclopropane product 9.44
Given the ready reaction of cyclopropanes with electro-
philes,45 such as halonium ions or heavy metal halides, we
suspected that polymerisation, perhaps initiated by I2 or Iꢀ3 ,
was causing destruction of 9—certainly the reaction mixture
became yellow in colour on addition of the aqueous acid.
Consistent with this interpretation, termination of the work-
upbywashingwithsaturated aqueousNa2SO3, gaveacolour-
less organic phase from which the cyclopropane 9 was iso-
lated in good yield (81%) and also in higher selectivity (11/
89 er, Table 2, entry 8) than when bromide-based reagent
was employed. Moreover, when 2 equiv of modifier (S)-18
is employed, the yield and selectivity increased, giving 9 in
86% yield and 6/94 er, Table 2, entry 9.38c The most selective
modifier from the indium allyl bromide-based conditions
(the 1-naphthyl derivative, (S)-31e, entry 5) gave lower selec-
tivity using iodide (entry 10).
Figure 1. The linear relationship observed between the enantiopurity of
(S)-methyl mandelate ((S)-18, x-axis) used as modifier (1.0 equiv) in the
reaction of homoallyl-cyclopropanation reaction of dibenzylideneacetone
(8) by an allylindium bromide reagent system generated in situ, and the
enantiopurity of product ((S)-9, y-axis). Correlation by linear regression:
((S)-9 (%)¼0.58(ꢂ0.018), (S)-18 (%)¼+21.2(ꢂ1.3)).
3. Conclusion
In summary, we have developed a simple asymmetric modi-
fication of the indium-mediated homoallyl-cyclopropana-
tion of dibenzylideneacetone 8 by using enantiomerically
pure, commercially available and relatively inexpensive,
methyl mandelate ((S)-18). The asymmetric modifiers that
have been successfully employed in the mechanistically dis-
tinct allylation of aldehydes, ketones and imines (cinchona
alkaloids, aminoalcohols, binols)27–29 are rather poor modi-
fiers for the homoallyl-cyclopropanation process. The key to
effect that added halide has on the reaction it was interesting
to determine the effect of switching from a bromide-based re-
agent system to the one based on iodide. However, reaction of
ketone 8 with the analogous allylindium iodide reagent, in
the presence or absence of modifier (S)-18, followed by
Table 2. The effect of a range of enantiomerically enriched (S)-methyl mandelates (1.0 equiv) as modifiers for the asymmetric homoallyl-cyclopropanation of
dibenzylideneacetone (8, 0.11 M in THF) by allylindium reagents derived from indium metal and allyl bromide or allyl iodide at rt
Entry
Modifier (Ar¼)
Allyl reagent (halide¼)
Yield 9/%a
er 9 (R/S)b
1
2
3
4
5
6
7
8
9
10
(S)-18
Br
Br
Br
Br
Br
Br
I
I
I
I
71
62
68
41
20/80c
68/32
70/30
17/83
14/86
26/74
—
(R)-31bd
(R)-31ce
(S)-31df
(S)-31eg
(S)-31fh
(S)-18
50
53
0i
(S)-18
81j
86j,k
45j,k
11/89
6/94
11/89
(S)-18k
(S)-31el
a
Isolated yield after purification by silica-gel chromatography.
Determined by chiral HPLC (OD-H, see Section 4 for details), then extrapolated to er expected for 100% enantiopurity in the modifier.
Data from Table 1, entry 13.
>99/1 er.
>99/1 er.
89/11 er.
92/8 er.
90/10 er.
Product mixture consisted of an intractable dark brown tar. Analogous results obtained in absence of modifier or on replacement of LiI with LiBr.
Ether phase washed with saturated aqueous Na2SO3 after 1 M HCl.
Modifier (2.0 equiv).
98/2 er.
b
c
d
e
f
g
h
i
j
k
l

G. C. Lloyd-Jones et al. / Tetrahedron 62 (2006) 11402–11412
11407
attaining good yield in combination with high selectivity is
using the indium/allyl iodide reagent combination, with
aqueous Na2SO3 included in the work-up procedure to avoid
indium salt/iodine-mediated polymerisation of the cyclopro-
pane product 9.
with an FID detector and a Supelco g-TA column
0.25 mmꢃ30 m; carrier gas He; flow rate 1.0 mL minꢀ1
;
oven temperature 120 ꢁC. NMR spectra were acquired on
JEOL instruments (operating at 270, 300 and 400 MHz (1H
frequency)) and recorded in CDCl3. 1H and 13C{1H} NMR
spectra were referenced internally to CHCl3 (1H, 7.27 ppm)
and CDCl3 (13C, 77.0 ppm) or to TMS (0.00 ppm). Assign-
ments are based on HH COSY, HMQC (1JC,H) and DEPT.
Mass spectra were obtained using the EI source on a Fisons
Micromass Autospec mass spectrometer. Flash column chro-
matography: Merck silica gel 60. TLC: 0.25 mm, Merck sil-
ica gel 60 F254 visualising at 254 nm or with acidic (H2SO4)
aqueous KMnO4 solution (ca. 2%).
We have previously reported that homoallyl-vinylcyclo-
propanes undergo Ru-catalysed ring closing metathesis
(RCM) to generate norcarenes.46 The potential utility of
the enantioselective homoallyl-cyclopropanation reaction
is outlined in Scheme 7: using standard conditions (S)-9
(94/6 er) undergoes RCM to yield (1S,6S)-32 in 52% isolated
yield, chiral HPLC analysis confirmed that there was no
racemisation.
4.1.1. (S)-2-(300-Butenyl)-1,1-bis[(E)-20-phenylethenyl]-
cyclopropane35 (S)-(9). Underanitrogenatmosphere, indium
metal power (100 mesh, Mg-free) (459 mg, 4.0 mmol) was
added to 2.0 mL of dry THF with stirring, to give a grey
suspension. Addition of allyl iodide (0.55 mL, 6.0 mmol)
resulted in an immediate exothermic reaction affording a clear
solution, which was stirred for a further 10 min before adding
(S)-methyl mandelate ((S)-18, 332 mg, 2.0 mmol). After stir-
ring for a further 10 min, dibenzylideneacetone (8) (235 mg,
1.0 mmol) was added and the reaction mixture was stirred at
room temperature (approximately 40–60 min). LiI (133 mg,
1.0 mmol) was then added and the reaction mixture was
heated to 60 ꢁC for 1 h before cooling to ambient temperature.
Afteradditionof15 mLofEt2O/15 mLof1 MHCl(aqueous),
the aqueous layer was separated and extracted with Et2O
(3ꢃ5 mL). The combined organic extracts were washed
with saturated Na2SO3 solution, saturated brine, dried
(MgSO4), filtered and then concentrated invacuo. Purification
by silica gel column chromatography (hexane/ethyl acetate
50/1, 20ꢃ5 cm) collecting fractions with Rf value 0.67, gave
(S)-(9) as yellow oil, 256.5 mg (86%). Eluting with 9/1 hex-
ane/EtOAc gave (S)-18, 131.4 mg (39%). Chiral HPLC
method A: tR (R)-(9) 91 min (6%), (S)-(9) 96 min (94%);
[a]D23 +68(ꢂ4) (c 2.48 in CHCl3); The 1H and 13C NMR spec-
tra were identical to those previously reported,35b however, it
should be noted that the subsequent preparation of 13C-la-
belled samples of 9 (unpublished) has revealed (HMQC/
Scheme 7. Synthesis of enantioenriched 1-styryl-norcarene (S,S)-32 in three
steps from acetone/benzaldehyde. Conditions: (i) PhCHO (2.0 equiv),
NaOH, EtOH, H2O; (ii) In, allyl iodide, THF, then (S)-18 (2.0 equiv),
then Et2O, 1 M HCl, then Na2SO3; (iii) 5 mol % RuCl2(PCy3)2]CHPh,
CH2Cl2.
The route thus provides rapid access to an enantiomerically
enriched norcarene (bicyclo[4.1.0]hept-2-ene) building
block of potential synthetic utility and in only three steps
from very simple reactants: acetone, benzaldehyde and allyl
iodide.
We are currently investigating the mechanism of the asym-
metric homoallyl-cyclopropanation process36–38 as well as
exploring the range of ketones, aldehydes and other modi-
fiers that can be employed. These studies will be reported
in full in due course.
1
NOE data) a pair of cross-assignments in the H NMR and
13C NMR. The previously reported 1H NMR data
3
(400 MHz, CDCl3) is 6.18 (1H, d, J¼15.62 Hz, 20-Hanti),
6.35 (1H, d, 3J¼16.11 Hz, 10-Hanti), 6.39 (1H, d,
3
4. Experimental
4.1. General
3J¼16.11 Hz, 10-Hsyn), 6.47 (1H, d, J¼15.62 Hz, 20-Hsyn).
3
New assignment: 6.18 (1H, d, J¼15.62 Hz, 10-Hanti), 6.35
3
3
(1H, d, J¼16.11 Hz, 20-Hanti), 6.39 (1H, d, J¼16.11 Hz,
3
10-Hsyn), 6.47 (1H, d, J¼15.62 Hz, 20-Hsyn). Previously re-
13
Reactions were conducted under an inert atmosphere of dry
nitrogen using standard Schlenk/vacuum line techniques.
Solvents were HPLC quality and dried by passage through
an Anhydrous Technologies drying train (activated alumina)
followed by freeze-thaw degas cycles under vacuum/nitro-
gen as appropriate. Chiral HPLC Analysis was performed
on Agilent 1100 or 1050 series instruments under two sets
of conditions. Conditions A: Chiralcel OD-H; 4.6ꢃ250 mm;
0.5 mL minꢀ1 hexane; detecting at 254 nm; ambient temper-
ature; analyte loaded as solution in eluent; Conditions B:
Chiralcel OJ-H; 4.6ꢃ250 mm; 1.0 mL minꢀ1 hexane/iso-
propyl alcohol (9:1); detecting at 210 nm; ambient tempera-
ture; analyte loaded as solution in eluent. Chiral GC Analysis
was performed on a Shimadzu GC17A instrument equipped
ported C NMR data (68 MHz, CDCl3) is 127.4 (10-Csyn),
130.0 (10-Canti), 131.7 (20-Csyn), 136.2 (20-Canti). New assign-
ment: 127.4 (20-Canti), 130.0 (10-Csyn), 131.7 (20-Csyn), 136.2
(10-Canti).
4.1.2. (1S,6S)-1-(E)-20-(Phenylethenyl)-bicyclo[4.1.0]-
hept-2-ene46 (1S,6S)-(32). Following the published
procedure for preparation of (ꢂ)-3246 the homoallyl-cyclo-
propanation product (S)-(9) (6/94 er), (330 mg, 1.1 mmol)
underwent RCM catalysed by a solution of RuCl2(PCy3)2]
CHPh (27 mg, 3 mol %) in 20 mL of CH2Cl2 to afford, after
work-up and chromatography, a clear, colourless oil,
111.5 mg (52%). NMR data (1H, 13C) were identical to
that reported.46 Analysis by chiral HPLC method B: tR

11408
G. C. Lloyd-Jones et al. / Tetrahedron 62 (2006) 11402–11412
(1R,6R)-(32) 39 min (6%), (1S,6S)-(32) 47 min (94%);
[a]2D3 ꢀ64(ꢂ5) (c 2.58 in CHCl3).
was prepared as follows: under nitrogen 0.21 mL dimethyl
malonate (1.83 mmol) was added to a stirred suspension of
60% NaH in oil (183.5 mg, 4.59 mmol) and 5 mL DME.
After 30 min, (R)-(+)-4-butyl-[1,3,2]-dioxathiolane-2,2-
dioxide (R)-(27) (346.4 mg, 1.92 mmol) was dissolved in
4 mL DME and added via cannula. After 4 h, 2 mL aqueous
NH4Cl solution (2.8 M) was added and the aqueous phase
was extracted with 3ꢃ10 mL ethyl acetate, dried (MgSO4)
and filtered. Purification by silica gel column chromato-
graphy (ethyl acetate/hexane 1/25) and Kugelrohr distillation
(oven temperature 170 ꢁC, 22 Torr) gave a colourless oil,
198.4 mg (51%). [a]2D6 +44.5 (c 2.49 in CHCl3); Anal. Calcd
for C11H18O4 (214.26): C 61.7, H 8.5; found: C 61.8, H 8.5;
nmax (film)/cmꢀ1: 2955 m, 2930 m, 2960 m, 1725 s, 1440 s,
1330 s, 1280 s, 1250 m, 1210 s, 1130 s, 1080 m, 880 m,
4.1.3. (S)-2-Butyl-1,1-bis[(E)-20-phenylethenyl]cyclopro-
pane (S)-(24). Under nitrogen, at 0 ꢁC, cyclopropane (S)-
(9) (34/66 er, 862.9 mg, 2.87 mmol) was dissolved in
10 mL dry CH2Cl2 and added via cannula to Cp2Zr(H)Cl47
(1.04 g, 4.03 mmol) with stirring. After 16 h, 10 mL distilled
water was added and the aqueous phase was extracted with
Et2O (3ꢃ15 mL). The combined organics were dried
(MgSO4), filtered through a plug of silica gel and concen-
trated in vacuo. Purification by silica gel column chromato-
graphy (ethyl acetate/hexane 1/50) gave a colourless oil,
670.5 mg (77%). [a]2D6 +38 (c 1.07 in CHCl3); Anal. Calcd
for C23H26 (302.45): C 91.3, H 8.7; found: C 90.6, H 8.7;
nmax (film)/cmꢀ1: 3025, 2960, 2925, 1595, 1495, 1450,
1075, 1030, 960, 790, 745, 690; dH (400 MHz, CDCl3):
1.06–1.14 (4H, m), 1.29–1.43 (2H, m), 1.44–1.66 (6H, m),
6.36 (1H, d, 3J¼15.9 Hz, 10-Hanti), 6.54 (1H, d,
3
705 m; dH (300 MHz, CDCl3): 0.89 (3H, t, J¼7.1 Hz, 400-
H3), 1.13–1.50 (8H, m, 300-H2, 200-H2, 100-H2, 3-H2), 1.90
3
(1H, dddd, J¼8.3, 8.3, 8.3, 6.0 Hz, 2-H), 3.72 (3H, s, 20-
H3), 3.76 (3H, s, 20-H3); dC (75 MHz, CDCl3): 13.9 (400-C),
21.3 (CH2), 22.2 (CH2), 28.3 (CH2), 28.7 (3-C), 30.9
(CH2), 33.8 (1-C), 52.4, 52.5; Chiral GC analysis: tR (R)-
(25) 10 min (7%), (S)-(25) 11 min (93%).
3
3J¼15.9 Hz, 20-Hanti), 6.58 (1H, d, J¼15.9 Hz, 10-Hsyn),
6.65 (1H, d, 3J¼15.9 Hz, 20-Hsyn), 7.30–7.59 (10H, m,
10ꢃCHarom); dC (75 MHz, CDCl3): 14.2 (400-C), 20.0
(CH2), 22.5 (CH2), 29.0 (CH2), 29.3 (2-C), 29.8 (1-C), 32.0
(CH2), 125.8 (o-Carom), 126.0 (o-Carom), 126.7 (p-Carom),
127.0 (p-Carom), 127.2 (20-Canti), 128.5 and 128.6 (2ꢃ
m-Carom), 130.2 (10-Csyn), 131.5 (20-Csyn), 136.3 (10-Canti),
137.6 and 137.7 (2ꢃi-Carom); HRMS (EI) obsd: 302.2039
[M+]; calcd: 302.2035; MS (EI, %): 302 (75), 259 (6), 245
(90), 231 (30), 215 (53), 202 (35), 191 (12), 178 (13), 167
(57), 153 (56), 141 (56), 141 (65), 128 (59), 115 (56), 103
(25), 91 (100), 86 (14), 77 (23), 65 (14), 55 (10).
4.1.5. (R)-1,2-Hexandiol (R)-(26).50 Hydroquinidine
2,5-diphenylpyrimidine-4,6-diyldiether
[(DHQD)2PYR]
(250.0 mg, 0.280 mmol), potassium carbonate (12.8 g,
92.7 mmol), potassium ferricyanide (29.5 g, 92.9 mmol)
and potassium osmate dihydrate (89.5 mg, 0.244 mmol)
were stirred in 310 mL of 1/1 tert-BuOH/distilled water, at
room temperature. The mixture was cooled to 0 ꢁC and
then 3.9 mL 1-hexene (31.3 mmol) was immediately added.
The resulting slurry was stirred for 2 days and then stored in
a freezer for 2 weeks. Sodium sulfite (45.0 g, 479 mmol) was
added and the slurry was warmed to room temperature with
stirring. After 3 h, the slurry was extracted with 3ꢃ150 mL
ethyl acetate. The combined organic extracts were washed
with 300 mL distilled water, 200 mL saturated aqueous
brine, dried (MgSO4), filtered and evaporated. The residue
was purified by flash chromatography by eluting with Et2O
and then distilled (160 ꢁC, 22 Torr) to yield a colourless liq-
uid, 2.23 g (60%). [a]D23 +3.54 (neat), (lit.51 for (S)-(26) [a]2D8
ꢀ4.0 (neat)); nmax (film)/cmꢀ1: 3340 br, 2960 s, 2930 s, 2860
s, 1470 m, 1340 w, 1030 s, 1060 s; dH (400 MHz, CDCl3):
4.1.4. (S)-1,1-Dicarboxymethyl-2-butylcyclopropane (S)-
(25). Ruthenium trichloride trihydrate (5.06 mg, 19.4 mmol)
was added to a stirred, biphasic mixture of (S)-(24)
(317.4 mg, 1.05 mmol) and NaIO4 (1.85 mg, 8.65 mmol) in
4 mL CCl4, 4 mL MeCN and 6 mL distilled water.48 The
brown mixture was stirred for 24 h and then 20 mL CH2Cl2
and 20 mL distilled water were added. The aqueous phase
was extracted with 3ꢃ20 mL CH2Cl2. The combined organic
extracts were dried (MgSO4), concentrated in vacuo and
redissolved in 20 mL Et2O, filtered through Celite and con-
centrated in vacuo to yield off-white crystals (304.4 mg)
comprising a 2/1 mixture of benzoic acid and (S)-2-butyl-
cyclopropane-1,1-dicarboxylic acid, which was used without
further purification; selected NMR data for (S)-2-butylcyclo-
propane-1,1-dicarboxylic acid as confirmed by hydrolysis
(KOH/aqueous MeOH, reflux, 4 h) of a sample of pure
(ꢂ)-25: dH (400 MHz, CDCl3): 0.90 (3H, t, J¼7.1 Hz, 400-
H3), 1.25–1.49 (4H, m, 300-H2), 1.62–1.79 (2H, m, 200-H2),
3
0.92 (3H, t, J¼6.4 Hz, 6-H3), 1.29–1.49 (6H, m, 5-H2,
4-H2, 3-H2), 2.10 (2H, br s, 2ꢃOH), 3.44 (1H, dd,
3J¼7.8 Hz, 2J¼10.7 Hz, 1-H), 3.66 (1H, dd, 3J¼2.9 Hz,
2J¼10.7 Hz, 1-H), 3.71 (1H, m, 2-H); dC (75 MHz,
CDCl3): 14.0 (6-C), 22.7 (CH2), 27.7 (CH2), 32.9 (CH2),
66.8 (1-C), 72.3 (2-C).
3
2
1.89 (1H, dd, J¼8.8 Hz, J¼4.1 Hz, 3-H), 2.06 (1H, dd,
4.1.6. (R)-4-Butyl-[1,3,2]-dioxathiolane-2,2-dioxide (R)-
(27).52 Thionyl chloride (0.18 mL, 2.47 mmol) was added
to a stirred solution of (R)-(26) (448 mg, 4.08 mmol) in
4 mL CCl4. The mixture was heated to reflux for 2 h, cooled
to 0 ꢁC and 4 mL MeCN, RuCl3$3H2O (4.30 mg,
0.0164 mmol), NaIO4 (1.78 g, 8.32 mmol) and 6 mL dis-
tilled water were added.53 After 2 h, the reaction mixture
was diluted with 30 mL Et2O and the organic phase was
washed with 1 mL distilled water, 3ꢃ2 mL saturated aque-
ous NaHCO3, 1 mL saturated aqueous brine, dried
(MgSO4) and filtered through silica. After concentration
in vacuo, the residue was distilled (Kugelrohr, oven temp
200 ꢁC, 22 Torr) to yield a colourless liquid, 692.0 mg
2
3
3J¼9.3 Hz, J¼4.1 Hz, 3-H), 2.23 (1H, dddd, J¼7.3, 7.3,
8.8, 9.3 Hz, 2-H), 9.4–10.6 (2H, br m, 2ꢃ10-H); dC
(75 MHz, CDCl3): 13.9 (400-C), 22.2 (CH2), 26.0 (3-C),
26.9 (100-C), 30.5 (1-C), 31.1 (CH2), 39.5 (2-C), 173.9 (10-
C), 176.7 (10-C). The crude product was dissolved in a mix-
ture of 2.0 mL dry MeOH/7 mL toluene and then a hexane
49
solution of TMSCHN2 (2 M, 3.84 mmol) was added to
the stirred solution. After 16 h, 0.5 mL AcOH (8 mmol)
was added, and then the volatiles were removed in vacuo to
yield 341.27 mg of a mixture comprising 2/1 methylbenzoate
and (S)-(25). Chiral GC analysis: tR (R)-(25) 10 min (34%),
(S)-(25) 11 min (66%). In a separate procedure, (S)-(25)

G. C. Lloyd-Jones et al. / Tetrahedron 62 (2006) 11402–11412
11409
(94%). [a]2D6 +9.4 (c 2.59 in CHCl3), (lit.52 [a]2D2 +6.0 (c 5.11
in CHCl3)); nmax (film)/cmꢀ1: 2960 m, 2875 m, 2930 m, 1470
m, 1380 s, 1025 s, 970 s, 825 s; dH (300 MHz, CDCl3): 0.94
HPLC conditions B: tR (R)-(31e) 72 min (4%), (S)-(31e)
75 min (96%); nmax (film)/cmꢀ1: 3446 br, 3050 w, 2955 w,
1950 w, 1730 s, 1600 w, 1510 m, 1435 m, 1395 s, 1355 w,
1215 s, 1165 s, 1095 s, 1065 s, 1010 m, 970 m, 920 w, 900
w, 885 w, 865 w, 800 s, 785 s, 775 s, 655 m, 730 m; dH
(400 MHz, CDCl3): 3.71 (3H, s, 30-H3), 5.81 (1H, s, 20-H),
7.40–7.58 (4H, m, 6-H, 7-H, 8-H, 9-H), 7.84–7.91 (2H, m,
3
(3H, t, J¼7.1 Hz, 6-H3), 1.35–1.49 (4H, m, 5-H2, 4-H2),
1.77–1.82 (1H, m, 1-H), 4.74 (1H, dd, 3J¼5.9 Hz,
3
2J¼8.8 Hz, 1-H), 5.00 (1H, dddd, J¼8.1, 8.1, 5.9, 4.9 Hz,
2-H); dC (75 MHz, CDCl3, ppm): 13.7 (6-C), 22.2 (CH2),
26.6 (CH2), 31.9 (3-C), 73.2 (1-C), 83.5 (2-C).
3
2-H, 3-H), 8.16 (1H, d, J¼8.6 Hz, 4-H); dC (100 MHz,
CDCl3): 53.0 (30-C), 71.4 (10-C), 123.6 (CHarom), 125.2
(CHarom), 125.9 (CHarom), 126.6 (CHarom), 128.8 (CHarom),
129.5 (CHarom), 133.0 (Carom), 134.0 (Carom), 174.6 (20-C);
MS (EI, %) m/z 216 [M+] (45), 157 (100), 139 (7), 129
(90), 102 (8), 84 (12), 75 (8), 69 (14), 63 (10).
4.1.7. Vanadyl (1R,2R)-N,N0-bis(3,5-di-tert-butylsalicyli-
denato)-1,2-cyclohexane diamine.43 A solution of (1R,2R)-
N,N0-bis(3,5-di-tert-butylsalicylidene)-1,2-cyclohexane di-
amine (1.0 g, 18 mmol) in 20 mL THF and vanadyl sulfate
hydrate (0.6 g, 2.2 mmol) in 32 mL hot ethanol, were mixed
and the resulting solution was refluxed for 3 h. The solvent
was then removed in vacuo and the residue was extracted
with CH2Cl2, filtered, concentrated in vacuo and then ab-
sorbed onto silica. Elution with CH2Cl2 followed by ethyl
acetate/methanol (2/1) gave, after concentration, a green
crystalline solid, 0.48 g (44%); mp>320 ꢁC; [a]D23 ꢀ1200
(c 0.01 in CHCl3), (lit.43 [a]D23 ꢀ950 (c 0.01 in CHCl3));
nmax (film)/cmꢀ1: 2955 m, 1615 m, 1560 m, 1435 m, 1395
m, 1365 m, 1270 s, 1250 s, 1180 s, 1065 m, 1020 s, 955 s,
925 s, 870 s, 845 s, 815 m, 760 s, 660 m; dH (400 MHz,
CDCl3): 1.36 (9H, s, C(CH3)3), 1.38 (9H, s, C(CH3)3),
1.53 (9H, s, C(CH3)3), 1.54 (9H, s, C(CH3)3), 1.98 (4H, m,
2ꢃCH2), 1.92 (2H, m, CH2), 2.50–2.79 (2H, m, CH2),
3.80 (1H, m, HCN), 4.25 (1H, br m, HCN), 7.51 (1H, d, 4J¼
2.5 Hz, ArCH), 7.56 (1H, d, 4J¼2.5 Hz, ArCH), 7.71 (1H, d,
4.1.9. (S)-Methyl 2-naphthylglycolate (S)-(31f).55 Com-
pound (S)-(31f) was prepared by an identical procedure to
(S)-(31e), but starting from 2-napthylaldehyde (408.5 mg,
2.6 mmol) to yield a colourless, viscous liquid, 172.1 mg
(31%). [a]2D4 +149 (c 1.14 CHCl3), (lit.55 for (R)-(31f)
[a]2D3 ꢀ164.0 (c 1.00 in CHCl3)); chiral HPLC conditions
B: tR (R)-(31f) 40 min (10%), (S)-(31f) 41 min (90%);
nmax (film)/cmꢀ1: 3470 m, 3055 w, 2965 w, 1725 s, 1510
w, 1440 w, 1390 w, 1365 w, 1305 m, 1270 w, 1255 w,
1220 s, 1170 w, 1145 w, 1085 s, 985 m, 945 w, 925 w, 905
w, 870 wm, 860 w, 830 m, 775 w, 750 s, 735 m, 665 w;
dH (400 MHz, CDCl3) 3.67 (1H, br, OH), 3.76 (3H, s, 30-
3
H3), 5.36 (1H, d, J¼5.4 Hz, 20-H), 7.46–7.55 (3H, m,
CHarom), 7.81–7.88 (3H, m, CHarom), 7.91 (1H, br, CHarom);
dC (100 MHz, CDCl3) 53.0 (30-C), 73.0 (10-C), 124.1
(CHarom), 125.9 (CHarom), 126.3 (2ꢃCHarom), 127.7
(CHarom), 128.5 (CHarom), 133.1 (Carom), 133.3 (Carom),
135.5 (Carom), 174.1 (20-C); MS (EI, %) m/z 216 [M+]
(46), 157 (100), 139 (7), 29 (90), 102 (8), 83 (11), 75 (8),
69 (13), 63 (10).
4
4J¼2.5 Hz, ArCH), 7.76 (1H, d, J¼2.5 Hz, ArCH), 8.52
(1H, br s, HC]N), 8.76 (1H, br s, HC]N); dC (75 MHz,
CDCl3): 24.12 (CH2), 24.52 (CH2), 29.18 (C(CH3)3),
29.95 (C(CH3)3), 30.79 (CH2), 31.40 (C(CH3)3), 31.42
(C(CH3)3), 34.48 (C(CH3)3), 34.52 (C(CH3)3), 35.57
(C(CH3)3), 35.74 (C(CH3)3), 69.88 (C(H)N), 70.69
(C(H)N), 120.92 (ArC), 121.88 (ArC), 128.38 (ArCH),
129.15 (ArCH), 131.64 (ArCH), 131.91 (ArCH), 135.00
(ArC), 135.95 (ArC), 143.94 (ArC), 160.97 (ArCO),
161.14 (ArCO), 161.15 (HC]N), 164.91 (HC]N).
4.1.10. (S)-Methyl r-methoxymandelate (S)-(31d).56
Compound (S)-(31d) was prepared by an identical procedure
to (S)-(31f), but starting from p-anisaldehyde (446.5 mg,
3.28 mmol) to yield a crystalline solid, 221.1 mg (34%);
mp 51–52 ꢁC (lit.56 63–64 ꢁC); [a]2D3 +93 (c 1.08 in acetone),
(lit.56 [a]2D0 +119.7 (c 3.3 in acetone)); chiral HPLC condi-
tions B: tR (R)-(31d) 73 min (11%), (S)-(31d) 76 min
(89%); nmax (film)/cmꢀ1: 3435 br, 3010 w, 2970 w, 2940 w,
2840 w, 2050 w, 1900 w, 1725 s, 1610 m, 1585 w, 1510 m,
1440 m, 1450 m, 1385 m, 1330 m, 1300 m, 1250 s, 1215 s,
1185 s, 1170 s, 1115 m, 1075 s, 1025 s, 975 m, 940 w, 910
m, 865 w, 835 s, 820 m, 795 s, 750 s, 710 m; dH (400 MHz,
4.1.8. (S)-Methyl 1-naphthylglycolate (S)-(31e).54 The
following procedure43 is typical for the range of methyl
arylglycolates prepared by V-catalysed asymmetric addi-
tion of TMSCN. To a solution of vanadyl (1R,2R)-N,N0-
bis(3,5-di-tert-butylsalicylidenato)-1,2-cyclohexane diamine43
(36.2 mg, 0.06 mmol, 0.6 mol %) in CH2Cl2 (12 mL) was
added 1-naphthaldehyde (1.56 g, 10.0 mmol) followed by
trimethylsilylcyanide (1.7 g, 17.25 mmol) and the solution
was stirred for 24 h at room temperature to afford 1-napthyl-
trimethylsilyloxyacetonitrile. After filtration through silica
gel (ethyl acetate/hexane 1/5) and concentration in vacuo,
the crude silyloxyacetonitrile was dissolved in Et2O
(120 mL) and the solution was cooled to 0 ꢁC. Then HCl-
saturated MeOH (30 mL) was added dropwise and the solu-
tion was left to stand at <5 ꢁC for 24 h. The precipitated solid
was separated by decantation, washed with cold Et2O
(3ꢃ50 mL), dissolved in distilled water (100 mL) and ex-
tracted with Et2O (3ꢃ50 mL). After washing with saturated
aqueous NaHCO3 (100 mL), distilled water (2ꢃ50 mL)
and saturated brine (100 mL), the combined extracts were
dried (MgSO4), filtered and concentrated in vacuo to give
a viscous, colourless oil, 824 mg (38%). [a]2D3 +130 (c 1.2
acetone), (lit.54 [a]D23 +106.2 (c 1.0 in acetone)); chiral
3
CDCl3): 3.44 (1H, d, J¼5.6 Hz, OH), 3.74 (3H, s, 30-H3),
3
3.80 (3H, s, 100-H3), 5.12 (1H, d, J¼5.6 Hz, 10-H), 6.89
(2H, d, 3J¼9.0 Hz, o-CHarom), 7.32 (2H, d, 3J¼9.0 Hz,
m-CHarom); dC (100 MHz, CDCl3): 52.9 (30-C), 55.2 (100-
C), 72.4 (10-C), 114.0 (m-CHarom), 127.9 (o-CHarom), 130.4
(1-CHarom), 174.3 (20-C); MS (EI, %) m/z 196 [M+] (16),
137 (100), 131 (8), 119 (7), 109 (35), 94 (28), 83 (21), 77
(31), 69 (28).
4.1.11. (R)-Methyl-2-chloromandelate57 (R)-(31c). The
following procedure is typical for the range of methyl man-
delates prepared by esterification of the parent acid. (R)-2-
Chloromandelic acid (1.07 g, 5.72 mmol) was dissolved in
methanol (35 mL). After addition of concentrated sulfuric
acid (0.05 mL) the solution was refluxed until all the (R)-
2-chloromandelic acid had been consumed (monitored by

11410
G. C. Lloyd-Jones et al. / Tetrahedron 62 (2006) 11402–11412
TLC). After concentrating in vacuo, the residue was dis-
solved in tert-BuOMe (75 mL) and washed with saturated
aqueous Na2CO3 (2ꢃ35 mL) and saturated brine (2ꢃ
35 mL). The organic phase was dried (MgSO4), filtered
and concentrated in vacuo to yield a colourless liquid,
859.9 mg (75%; ca. 96% pure according to 1H NMR analysis
with benzylbenzoate internal standard) (R)-(31c); [a]D24
ꢀ179 (c 2.17 in CHCl3); nmax (film)/cmꢀ1: 3500 br, 2990
w, 2955 w, 1725 s, 1600 w, 1495 m, 1445 m, 1375 m,
1250 s, 1210 m, 1140 s, 1095 s, 1070 s, 1030 m, 975 m,
940 w, 915 w, 870 w, 785 m, 730 m, 695 s, 655 s; dH
(270 MHz, CDCl3, ppm): 3.77 (3H, s, 30-H3), 5.58 (1H, s,
10-H), 7.24–7.32 (2H, m, CHarom), 7.35–7.44 (2H,
m, CHarom); dC (68 MHz, CDCl3, ppm): 53.2 (30-C), 70.5
(10-C), 127.2 (CHarom), 128.9 (CHarom), 129.8 (CHarom),
130.0 (CHarom), 133.6 (Carom), 136.1 (Carom), 173.7 (20-C).
4.1.15. (S)-Mandelamide (S)-(12).61 (S)-Mandelic acid
(8.8 g, 57.8 mmol) was dissolved in 230 mL MeOH and
cooled to 0 ꢁC. Acetyl chloride (4.2 mL, 59.1 mmol) was
added and the solution was allowed to warm to room temper-
ature and stirred for 24 h. Concentration in vacuo gave a
colourless liquid, which was dissolved in 35 mL MeOH.
Aqueous 105 mL NH3 (35% w/v) was added and the solu-
tion was stored at <5 ꢁC for a further 24 h. Concentration
in vacuo gave a white solid, which was recrystallised from
hot EtOH to yield white plates, 3.88 g (45%); mp 112–
114 ꢁC, (lit.62 118–120 ꢁC); [a]2D5 +57 (c 1.05 in EtOH);
nmax (film)/cmꢀ1: 3345 br, 3180 br, 2930 w, 1955 w, 1680
m, 1655 m, 1610 m, 1495 m, 1450 m, 1420 m, 1340 m,
1290 m, 1250 m, 1190 m, 1100 m, 1080 m, 1055 s, 1025
m, 1000 m, 930 m, 895 m, 860 m, 760 m, 705 s, 690 s; dH
(300 MHz, CD3OD): 4.98 (1H, s, 10-H), 7.24–7.37 (3H, m,
CHarom), 7.46 (2H, ddd, 3J¼8.1 Hz, 4J¼1.7 Hz, 5J¼
0.5 Hz, o-CHarom); dC (75 MHz, CD3OD): 75.4 (10-C),
127.9 (o-CHarom), 129.1 (p-CHarom), 129.3 (m-CHarom),
141.7 (i-Carom), 178.6 (20-C).
4.1.12. (R)-Methyl-3-chloromandelate58 (R)-(31b). Com-
pound (R)-(31b) was prepared by an identical procedure to
(R)-(31c), but starting from (R)-(ꢀ)-3-chloromandelic acid
(502 mg, 2.69 mmol) to yield a colourless liquid, 242.5 mg
1
(45%; ca. 90% pure according to H NMR analysis with
4.1.16. (S)-Acetoxy mandelamide (S)-(17).63 Under a nitro-
gen atmosphere, (S)-(12) (606.4 mg, 4.01 mmol) was dis-
solved in 24 mL pyridine. Acetic anhydride (1.2 mL,
12.7 mmol) was added and the solution was stirred for
18 h. Concentration in vacuo gave an off-white solid, which
was recrystallised from hot EtOH to yield white crystals,
397.5 mg (51%); mp 133–134 ꢁC, (lit. 112–113 ꢁC, for non-
enantiomerically pure (R)-(17); [a]2D5 +139 (c 1.01 in EtOH),
(lit.63 for nonenantiomerically pure (R)-(17) [a]2D5 ꢀ145 (c
1.0 in CHCl3)); nmax (film)/cmꢀ1: 3380 m, 3160 m, 2160
w, 1980 w, 1740 s, 1665 s, 1630 m, 1590 w, 1495 w, 1455
w, 1420 m, 1380 m, 1320 w, 1295 w, 1265 m, 1220 s,
1190 s, 1105 m, 1080 m, 1030 s, 1005 m, 990 m, 950 s,
910 m, 855 w, 800 m, 755 s, 715 s, 695 s; dH (270 MHz,
CD3OD): 2.20 (3H, s, Ac-H3), 5.95 (1H, s, 10-H), 7.30–
7.41 (3H, m, CHarom), 7.42–7.55 (2H, m, o-CHarom); dC
(100 MHz, CD3OD): 20.7 (Ac-CH3), 76.8 (10-C), 128.6
(CHarom), 129.6 (CHarom), 130.0 (CHarom), 137.0 (i-Carom),
171.6 (Ac-C]O), 174.1 (20-C).
benzylbenzoate internal standard); [a]2D4 ꢀ121 (c 4.19 in
CHCl3); nmax (film)/cmꢀ1: 3500 br, 2955 w, 1740 s, 1600
m, 1580 m, 1475 m, 1435 s, 1360 m, 1215 s, 1190 s, 1105
s, 1075 s, 1000 m, 980 m, 885 m, 770 s, 705 s, 685 s; dH
(270 MHz, CDCl3): 3.77 (3H, s, 30-H3), 5.15 (1H, s, 10-H),
7.25–7.56 (4H, m, CHarom); dC (68 MHz, CDCl3): 54.4
(30-C), 73.4 (10-C), 125.9 (CHarom), 126.1 (CHarom), 127.9
(CHarom), 129.8 (CHarom), 131.0 (Carom), 141.3 (Carom),
174.7 (20-C).
4.1.13. (S)-Methyl atrolactate (S)-(19).59 Compound (S)-
(19) was prepared by an identical procedure to (R)-(31c),
but starting from (S)-atrolactic acid (268 mg, 1.61 mmol)
to yield a colourless oil, 156.7 mg (54%; ca. 88% pure ac-
1
cording to H NMR analysis with benzylbenzoate internal
standard); [a]D24 +55 (c 0.51 in CHCl3), (lit.59 [a]D26
+54.50 (c 2.84 in CHCl3)); nmax (film)/cmꢀ1: 3450 br,
2955 w, 1730 s, 1595 w, 1575 w, 1475 m, 1440 s, 1220 s,
1190 s, 1130 m, 1085 s, 1050 s, 1035 s, 980 m, 950 m,
910 w, 865 w, 785 m, 755 s, 705 s; dH (270 MHz, CDCl3):
1.79 (3H, s, 10-CH3), 3.75 (3H, s, 30-H3), 7.23–7.40 (3H,
m, CHarom), 7.50–7.58 (2H, m, CHarom); dC (68 MHz,
CDCl3): 26.8 (10-CH3), 53.2 (30-C), 76.0 (10-C), 125.2
(CHarom), 127.8 (CHarom), 128.4 (CHarom), 142.8 (Carom),
176.1 (20-C).
4.1.17. (R)-N-Butyl mandelamide (R)-(14).64 Following
a similar procedure to (S)-(12), but using butylamine, (R)-
mandelic acid (1.042 g, 6.86 mmol) gave, after column
chromatography, (MeCN), a clear colourless oil, 925 mg
(65%); [a]2D5 +23 (c 1.2 in EtOH); dH (270 MHz, CDCl3):
3
0.87 (1H, t, J¼7.15 Hz, 40-H3), 1.24–1.26 (2H, m, 30-H2),
1.40–1.43 (2H, m, 20-H2), 3.18–3.20 (1H, m, 10-H2), 4.13–
4.16 (1H, m, 2-OH), 4.90–4.93 (0.5H, m, 2-H rot. A),
4.95–4.98 (0.5H, m, 2-H rot. B), 6.33–6.45 (1H, br s, NH),
7.26–7.40 (5H, m, CHarom); dC (75 MHz, CDCl3): 13.6
(40-C), 19.7 (30-C), 31.4 (20-C), 39.1 (10-C), 73.95 (2-C rot.
A/B), 74.01 (2-C rot. A/B), 126.7, 128.4, 128.7 (CHarom),
139.6 (Carom), 172.2 (C]O); MS (EI, %) m/z 207 (M+, 8)
191 (3), 179 (33), 136 (25), 107 (100), 84 (40), 79 (77), 73
(25), 57 (56).
4.1.14. (S)-Methyl-a,a-methoxyphenylacetate (S)-(18).60
Compound (S)-(18) was prepared by an identical procedure
to (R)-(31c), but starting from O-methyl (S)-mandelic acid
(516 mg, 3.11 mmol) to yield a colourless liquid, 345.3 mg
1
(62%; ca. 84% pure according to H NMR analysis with
benzylbenzoate internal standard); [a]2D4 +107 (c 1.01 in
CHCl3) (lit.60 [a]D23 +125 (c 0.97 in CHCl3)); nmax (film)/
cmꢀ1: 2955 w, 2830 w, 2960 m, 1750 s, 1495 w, 1455 m,
1435 m, 1350 w, 1260 m, 1195 s, 1175 s, 1105 s, 1075 m,
1010 s, 930 w, 905 w, 850 w, 785 w, 730 s, 695 s; dH
(270 MHz, CDCl3): 3.39 (3H, s, 100-H3), 3.70 (3H, s, 30-
H3), 4.78 (1H, s, 10-H), 7.25–7.49 (5H, m, CHarom); dC
(68 MHz, CDCl3): 53.4 (100-C), 58.4 (30-C), 83.7 (10-C),
128.4 (CHarom), 129.8 (CHarom), 129.9 (CHarom), 137.6
(CHarom), 172.3 (20-C).
4.1.18. (S)-N-Benzyl mandelamide (S)-(15).65 Following
a similar procedure to (S)-(12), but using benzylamine,
(S)-mandelic acid (1.25 g, 7.89 mmol) gave a yellow oil,
which was crystallised from boiling MeOH to yield white
needles, 1.00 g (51%); mp 128–130 ꢁC (from MeOH)
(lit.65 136 ꢁC); [a]D24 +98 (c 1.06 in CHCl3) (lit.65 [a]2D4

G. C. Lloyd-Jones et al. / Tetrahedron 62 (2006) 11402–11412
11411
+82.2 (c 1.09 in CHCl3)); nmax(film)/cmꢀ1: 3405 m, 3185 br,
2940 w, 2895 w, 1900 w, 1645 s, 1600 w, 1585 w, 1535 s,
1495 m, 1450 m, 1440 m, 1340 w, 1310 w, 1290 m, 1260
w, 1240 m, 1210 w, 1195 w, 1180 w, 1095 m, 1080 m,
1065 s, 1030 m, 925 m, 850 w, 815 w, 780 w, 755 s, 735 s,
705 s; dH (400 MHz, CD3OD): 4.41 (2H, s, CH2Ph), 5.06
(1H, s, CHOH), 7.19–7.36 (8H, m, CHarom), 7.46 (2H, dd,
11. (a) Araki, S.; Butsugan, Y. Trends in Organometallic
Chemistry; Pandalai, S. C., Ed.; Research Trends:
Trivandrum, 1994; pp 1–8; (b) Cintas, P. Synlett 1995, 1087;
(c) Li, C. J. Tetrahedron 1996, 52, 5643–5668; Marshall,
J. A. Chem. Rev. 1996, 96, 31–47; (d) Marshall, J. A.
Chemtracts-Org. Chem. 1997, 10, 481–496; Coldham, I. J.
Chem. Soc., Perkin Trans. 1 1998, 1343; (e) Chan, T. H.; Li,
C. J. Tetrahedron 1999, 55, 11149–11176; (f) Chauhan,
K. K.; Frost, C. G. J. Chem. Soc., Perkin Trans. 1 2000,
3015–3019; (g) Ranu, B. C. Eur. J. Org. Chem. 2004, 55,
2347; (h) Babu, G.; Perumai, P. T. Aldrichimica Acta 2000,
33, 16–32; (i) Marshall, J. A. Chem. Rev. 2000, 100, 3163–
3186; (j) Podlech, J.; Maier, T. C. Synthesis 2003, 633–655;
(k) Nair, V; Ros, S.; Jayan, C. L.; Pillai, B. S; Tetrahedron
60, 1959–1982; (l) Kumar, S.; Kaur, P.; Kumar, V. Curr. Org.
Chem. 2005, 1205–1235.
12. For a recent example and leading references see: Jang, T.-S.;
Ku, I. W.; Jang, M. S.; Keum, G.; Kang, S. B.; Chung, B. Y.;
Kim, Y. Org. Lett. 2006, 8, 195–198.
13. (a) Araki, S.; Kenji, O.; Shiraki, F.; Hirashita, T. Tetrahedron
Lett. 2002, 43, 8033–8035; (b) Araki, S.; Imari, A.; Shimizu,
K.; Yamada, M.; Mori, A.; Butsugan, Y. J. Org. Chem. 1995,
60, 1841–1847 and references therein.
14. See for example: (a) Loh, T. P.; Li, X. R. Tetrahedron 1999, 55,
5611–5622; (b) Kwon, J. S.; Pae, A. N.; Choi, K. I.; Koh, H. Y.;
Kim, Y.; Cho, Y. S. Tetrahedron Lett. 2001, 42, 1957–1959.
15. (a) Araki, S.; Kameda, K.; Tanaka, J.; Hirashita, T.; Yamamura,
H.; Kawai, M. J. Org. Chem. 2001, 66, 7919–7921; (b) Yadav,
J. S.; Anjaneyulu, S.; Ahmed, M. M.; Reddy, B. V. S.
Tetrahedron Lett. 2001, 42, 2557–2559; (c) Araki, S.;
Kambe, S.; Kameda, S.; Hirashita, T. Synthesis 2003, 751–
754 and references therein.
16. Lee, K.; Kim, H.; Miura, T.; Kiyota, K.; Kusama, H.; Kim, S.;
Iwasawa, N.; Lee, P. H. J. Am. Chem. Soc. 2003, 125, 9682–
9688.
17. Loh, T. P.; Lye, P.-L.; Wang, R.-B.; Sim, K.-Y. Tetrahedron
Lett. 2000, 41, 7779–7783.
4
J¼8.1 Hz, JHH¼1.7 Hz, o-CHaromCHOH); dC (100 MHz,
CD3OD): 43.6 (CH2Ph), 75.4 (CHOH), 127.9, 128.2,
128.4, 129.1, 129.4, 129.5 (CHarom), 139.9 (i-Carom), 141.8
(i-Carom), 175.5 (CONHBn).
Acknowledgements
R. A. Stentiford (University of Bristol) contributed to the
early stages of this work and her input is gratefully acknowl-
edged. G.C.L.-J. warmly thanks AstraZeneca and the Royal
Society of Chemistry for generous and unrestricted research
support, in addition to an RSC Sir William Briggs Scholar-
ship (P.D.W.) and EPSRC Industrial CASE awards (P.D.W.
and J.L.S.). Dr. Bob Osborne and Dr. Ian Derrick (AstraZe-
neca, Avlon Works, UK) provided extensive advice and as-
sistance in factorial based experimental design, automation
and interpretation.
References and notes
1. Schumb, W. C.; Crane, H. I. J. Am. Chem. Soc. 1938, 60,
306–308 and references therein.
2. Gilman, H.; Jones, R. G. J. Am. Chem. Soc. 1940, 62, 2353–
2357.
3. (a) Dennis, L. M.; Work, R. W.; Rochow, E. G.; Chamot, E. M.
J. Am. Chem. Soc. 1934, 56, 1047–1049; (b) Visser, H. D.;
Oliver, J. P. J. Organomet. Chem. 1972, 40, 7–14.
4. Other organometallics were also employed, see for example:
(a) Clark, H. C.; Pickard, A. L. J. Organomet. Chem. 1967, 8,
427–434; (b) Maede, T.; Okawara, R.; Tada, H.; Yasuda, K.;
Okawara, R. J. Organomet. Chem. 1971, 27, 13–18.
5. (a) Deacon, G. B.; Parrott, J. C. Aust. J. Chem. 1971, 24, 1771–
1779; (b) Gynane, M. J. S.; Waterworth, L. G.; Worrall, I. J.
J. Organomet. Chem. 1972, 40, C9–C10.
6. Chao, L. C.; Rieke, R. D. J. Org. Chem. 1975, 40, 2253–2255.
7. (a) Araki, S.; Ito, H.; Butsugan, Y. J. Org. Chem. 1988, 53,
1831–1833; (b) Araki, S.; Ito, H.; Katsumura, N.; Butsugan,
Y. J. Organomet. Chem. 1989, 369, 291–296; (c) Araki, S.;
Shimizu, T.; Johar, P. S.; Jin, S. J.; Butsugan, Y. J. Org.
Chem. 1991, 56, 2538–2542.
18. Yi, X. H.; Li, C. J.; Haberman, J. X. Synth. Commun. 1998, 28,
2999–3009; (b) Andrews, P. C.; Peatt, A. C.; Raston, C. L.
Green Chem. 2001, 3, 313–315.
19. Haberman, J. X.; Irvin, G. C.; John, V. T.; Li, C. J. Green Chem.
1999, 1, 265–267.
20. Law, M. C.; Wong, K. Y.; Chan, T. H. Green Chem. 2002, 4,
161–164.
´
21. For leading references see: (a) Auge, J.; Lubin-Germain, N.;
Marque, S.; Seghrouchni, L. J. Organomet. Chem. 2003, 679,
79–83; (b) Araki, S.; Jin, S. J.; Idou, Y.; Butsugan, Y. Bull.
Chem. Soc. Jpn. 2002, 65, 1736–1738; (c) Takai, K.; Ikawa,
Y. Org. Lett. 2002, 4, 1727–1729.
8. For leading references to other routes to allylic indium reagents
and intermediates see: (a) Hinkle, K. W.; Marshall, J. A. J. Org.
Chem. 1996, 61, 105–108; (b) Araki, S.; Hirahsita, T.; Kamei,
T.; Yamamura, H.; Kawai, M. Org. Lett. 2000, 2, 847–849; (c)
Anwar, U.; Grigg, R.; Rasparini, M.; Savic, V.; Sridharan, V.
Chem. Commun. 2000, 645–646; (d) Hirashita, T.; Kambe,
S.; Tsuji, H.; Omori, H.; Araki, S. J. Org. Chem. 2004, 69,
5054–5059; (e) Norsikian, S. P.; Lubineau, A. Org. Biomol.
Chem. 2005, 3, 4089–4094.
9. Propargylation/allenylation has also been extensively explored.
For leading references see: Marshall, J. A.; Mulhearn, J. J. Org.
Lett. 2005, 7, 1593–1596.
10. For a recent example and leading references see: Palmelund,
A.; Masden, R. J. Org. Chem. 2005, 70, 8248–8251.
22. (a) See Vilaivan, T.; Winotapan, C.; Banphavichit, V.; Shinada,
T.; Ohfune, Y. J. Org. Chem. 2005, 70, 3464–3471 and refer-
ences therein; (b) see also: Ding, H.; Friestad, G. K.
Synthesis 2005, 2815–2829.
23. See: Cook, G. R.; Maity, B. C.; Kargbo, R. Org. Lett. 2004, 6,
1741–1743 and references therein.
24. Alcaide, B.; Almendros, P.; Aragoncillo, C.; Rodriguez-
Acebes, R. Synthesis 2003, 1163–1170 and references therein.
25. Araki, S.; Shiraki, F.; Tanaka, T.; Nakano, H.; Subburaj, K.;
Hirashita, T.; Yamamura, H.; Kawai, M. Chem.—Eur. J.
2001, 7, 2784–2790.
26. Reviews: (a) Martin, S. F. Tetrahedron 1980, 36, 419–460; (b)
Fuji, K. Chem. Rev. 1993, 93, 2037–2066; (c) Almstead,
N. G.; Denmark, S. E. Modern Carbonyl Chemistry; Otera,

11412
G. C. Lloyd-Jones et al. / Tetrahedron 62 (2006) 11402–11412
J., Ed.; Wiley-VCH: Weinheim, 2000; Chapter 10; (d)
Hartley, R. C.; Caldwell, S. T. J. Chem. Soc., Perkin Trans. 1
2000, 477–501; (c) Lebel, H.; Marcoux, J. F.; Molinaro, C.;
Charette, A. B. Chem. Rev. 2003, 103, 977–1050; (d) Doyle,
M. P.; Forbes, D. C. Chem. Rev. 1998, 98, 911–935;
(e) Lautens, M.; Klute, W.; Tam, W. Chem. Rev. 1996, 96,
Chemler, S. R.; Roush, W. R. Modern Carbonyl Chemistry;
Otera, J., Ed.; Wiley-VCH: Weinheim, 2000; Chapter 11;
(e) Christoffers, J.; Mann, A. Angew. Chem., Int. Ed. 2001,
40, 4591–4597; (f) Christoffers, J.; Baro, A. Angew. Chem.,
Int. Ed. 2003, 42, 1688–1690; (g) Denissova, I.; Barriault,
L. Tetrahedron 2003, 59, 10105–10146; (h) see also:
Ribeiro, C. M. R.; de Farias, F. M. C. Mini-Rev. Org.
Chem. 2006, 3, 1–10.
49–92; (f) Salaun, J. Chem. Rev. 1996, 96, 1247–1270.
¨
40. Gawley, R. E. J. Org. Chem. 2006, 71, 2411–2416.
41. The Computer Programme MODDE 7.0 (Umetrics UK Ltd,
Windsor, Berks, SL4 2DX, UK) was employed to design
a set of 19 experiments to identify inter-dependencies in the re-
action variables based on stoichiometry (In, allyl bromide, 8),
times between addition of components, THF/Et2O volumes,
and temperature.
42. Both enantiomers of methyl mandelate are available from
Aldrich, UK price £48.70/5g. Mandelic acid is substantially
cheaper ((R)- £30.40/25g; (S)- £124.40/100 g) and is conve-
niently converted to the methyl ester (MeOH, H2SO4) without
racemisation, see Section 4 for details.
43. Belokon’, Y. N.; Green, B.; Ikonnikov, N. S.; North, M.;
Parsons, T.; Tararov, V. I. Tetrahedron 2001, 57, 771–779.
44. It should be noted that in the homoallyl-cyclopropanation
reaction the acidic surface of the silica gel plate can induce
cyclopropane formation and that TLC analysis must thus be
interpreted with caution.
45. DuPuy, C. H. Top. Curr. Chem. 1973, 40, 73–118.
46. Lloyd-Jones, G. C.; Murray, M.; Stentiford, R. A.;
Worthington, P. A. Eur. J. Org. Chem. 2000, 975–985.
47. Labinger, J. A.; Schwartz, J. Angew. Chem. 1976, 88, 402–409.
48. Carlsen, P. H.; Katsuki, T.; Martin, V. S.; Sharpless, K. B.
J. Org. Chem. 1981, 46, 3936–3938.
27. Johar, P. S.; Araki, S.; Butsugan, Y. J. Chem. Soc., Perkin Trans.
1 1992, 711–713.
28. Allylation: (a) Loh, T.-P.; Zhou, J.-R.; Yin, Z. Org. Lett. 1999,
1, 1855–1857; (b) Propargylation: Loh, T.-P.; Lin, M.-J.;
Tan, K.-L. Tetrahedron Lett. 2003, 44, 507–509.
29. Han, R.; Choi, S.; Son, K.; Jun, Y. M.; Lee, B.; Kim, B. Synth.
Commun. 2005, 35, 1725–1733.
30. Hirayama, L. C.; Gamsey, S.; Knueppel, D.; Steiner, D.;
DeLaTorre, K.; Singaram, B. Tetrahedron Lett. 2005, 46,
2315–2318.
31. Cook, G. R.; Kargbo, R.; Maity, B. C. Org. Lett. 2005, 7, 2767–
2770.
32. (a) Teo, Y.-C.; Tan, K.-T.; Loh, T.-P. Chem. Commun. 2005,
1318–1320; (b) Teo, Y.-C.; Goh, E.-L.; Loh, T.-P.
Tetrahedron Lett. 2005, 46, 6209–6211; (c) Teo, Y.-C.; Goh,
J.-D.; Loh, T.-P. Org. Lett. 2005, 7, 2743–2745 and references
therein.
33. Reviews: (a) Corey, E. J.; Guzman-Perez, A. Angew. Chem.,
Int. Ed. 1998, 37, 388–401; (b) Yanagisawa, A.
Comprehensive Asymmetric Catalysis; Jacobsen, E. N.,
Pfaltz, A., Yamamoto, H., Eds.; Springer: Heidelberg, 1999;
Vol. II, Chapter 27; (c) Denmark, S. E.; Fu, J. Chem. Rev.
2003, 103, 2763–2793.
49. Aoyama, T.; Shiori, T. Chem. Pharm. Bull. 1981, 29, 3249–
3255.
34. There is some debate about the exact nature of the allylating
agent(s), see: Chan, T. H.; Yang, Y. J. Am. Chem. Soc. 1999,
121, 3228–3229.
35. (a) Hoppe, H. A. F.; Lloyd-Jones, G. C.; Murray, M.; Peakman,
¨
T. P.; Walsh, K. E. Angew. Chem., Int. Ed. 1998, 37, 1545–
50. Nelson, A.; Warren, S. J. Chem. Soc., Perkin Trans. 1 1999,
1963–1982.
51. Koine, N.; Okita, K.; Sakota, N.; Tanaka, S. Nippon Kagaku
Zasshi 1970, 91, 265–268.
52. Hayes, P.; Kitching, W.; Konig, W. A.; Paddon-Jones, G. C.;
McErlean, C. S. P. J. Org. Chem. 2001, 66, 7487–7495.
53. Gao, Y.; Sharpless, K. B. J. Am. Chem. Soc. 1988, 110, 7538–
7539.
54. Takahashi, I.; Koga, K.; Odashima, K. Chem. Pharm. Bull.
1985, 33, 3571–3573.
1547; (b) Capps, S. M.; Clarke, T. P.; Charmant, J. P. H.;
Hoppe, H. A. F.; Lloyd-Jones, G. C.; Murray, M.; Peakman,
¨
T. P.; Stentiford, R. A.; Walsh, K. E.; Worthington, P. A. Eur.
J. Org. Chem. 2000, 963–974.
36. Allyl indium halide reagents (generated from allyl iodide or
bromide, using In metal or In(I)I) are reported to add exclu-
sively [1,2]- to enones and enals, see Ref. 7a,b.
55. Kimura, M.; Kuboki, A.; Sugai, T. Tetrahedron: Asymmetry
2002, 13, 1059–1068.
37. For a mechanistically related process involving indoles, see: (a)
Kumar, S.; Kumar, V.; Chimni, S. S. Tetrahedron Lett. 2002,
43, 8029–8031; (b) Kumar, S.; Kumar, V.; Chimni, S. S.
Tetrahedron Lett. 2003, 44, 2101–2104.
56. McKenzie, A.; Pirie, D. J. C. Chem. Ber. 1936, 69, 861–863.
57. Neilson, D. G.; Zakir, U. J. Chem. Soc. C 1971, 898–903.
58. Mori, N.; Tanaka, Y.; Tsuzuki, Y. Bull. Chem. Soc. Jpn. 1966,
39, 1490–1493.
38. (a) Capps, S. M.; Lloyd-Jones, G. C.; Murray, M.; Peakman,
T. P.; Walsh, K. E. Tetrahedron Lett. 1998, 39, 2853–2856;
(b) Ongoing mechanistic studies suggest that the second allyl
group is delivered intermolecularly from indium: Wall, P. W.;
Slaughter, J. L.; Parker, A. J.; Laffan, D.; Lloyd-Jones, G. C.
Unpublished. (c) The beneficial effect of excess modifier under
certain conditions, e.g., Table 2, entry 9, does not necessarily
equate to the In/L* ratio in the modified reagent—as indeed
is suggested (but not ruled out) by the linear nature of the cor-
relation in Figure 1. The identity of the modified reagent(s) is
the subject of ongoing study.
59. Mash, E. A.; Fryling, J. A. J. Org. Chem. 1991, 56, 1094–
1098.
60. von Matt, P.; Lloyd-Jones, G. C.; Minidis, A. B. E.; Pfaltz, A.;
Macko, M.; Neuburger, M.; Zehnder, M.; Ruegger, H.;
Pregosin, P. S. Helv. Chim. Acta 1995, 78, 265–284.
61. Meyers, A. I.; Slade. J. Org. Chem. 1980, 45, 2785–2791.
62. Ebbers, E. J.; Ariaans, G. J. A.; Bruggink, A.; Zwanenburg, B.
Tetrahedron: Asymmetry 1999, 10, 3701–3718.
63. North, M.; Parkins, A. W.; Shariff, A. N. Tetrahedron Lett.
2004, 45, 7625–7627.
64. Aldea, R.; Alper, H. J. Org. Chem. 1998, 63, 9425–9426.
65. Carpentier, J. F.; Mortreaux, A. Tetrahedron: Asymmetry 1997,
8, 1083–1099.
39. For reviews discussing the synthesis of cyclopropanes see: (a)
Donaldson, W. A. Tetrahedron 2001, 57, 8589–8627; (b)