2-Halo-diketopiperazines as chiral glycine cation equivalents

Article abstract of DOI:10.1016/j.tetasy.2004.11.022

A range of (2S,5S)-5-isopropyl-2-halo-N,N′-bis-(p-methoxybenzyl)- piperazine-3,6-diones 8 (Cl), 11, 12 (F) and 13 (Br) have been prepared, either via electrophilic halogenation of the corresponding lithiated diketopiperazine, or via transhalogenation from

Full text of DOI:10.1016/j.tetasy.2004.11.022

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 15 (2004) 3989–4001
2-Halo-diketopiperazines as chiral glycine cation equivalents
Steven D. Bull, Stephen G. Davies,* A. Christopher Garner, Edward D. Savory,
Emma J. Snow and Andrew D. Smith
Department of Organic Chemistry, University of Oxford, Chemistry Research Laboratory, Mansfield Road, Oxford OX1 3TA, UK
Received 3 November 2004; accepted 11 November 2004
Abstract—A range of (2S,5S)-5-isopropyl-2-halo-N,N0-bis-(p-methoxybenzyl)-piperazine-3,6-diones 8 (Cl), 11, 12 (F) and 13 (Br)
have been prepared, either via electrophilic halogenation of the corresponding lithiated diketopiperazine, or via transhalogenation
from fluoro-11 and 12. The product distribution and stereoselectivity of additions of allyltrimethylsilane, sodium thiophenolate and
a range of organomagnesium reagents to chloro 8 are reported. In the reactions with Grignard reagents the observed stereo- and
regioselectivities are dependent on the reagent employed, with C-3 carbonyl addition products predominating upon addition of allyl
or methylmagnesium chloride and stereodivergent formal C-2 addition predominating with benzyl or isopropylmagnesium chloride.
A model to account for the different reactivity and stereoselectivity in these reactions is proposed.
ꢀ 2004 Published by Elsevier Ltd.
OMe
1. Introduction
The addition of Grignard reagents to imines is well doc-
umented in the literature,1 with stereoselective versions
of this protocol having found wide application in orga-
nic synthesis, including the asymmetric synthesis of
a-amino acids.2 While this methodology has been the
subject of many investigations, the reaction of Grignard
reagents with the corresponding cationic N-acyliminium
or iminium species has received much less attention.3
N
N
O
CO2H
R
NH2
O
(R)-α-amino acids
1
MeO
OMe
We have recently reported two new diketopiperazine de-
rived chiral auxiliaries 1 and 2 for the asymmetric syn-
thesis of homochiral a-amino acids based upon the
alkylation of chiral glycine anion equivalent 1,4 and
conjugate addition followed by protonation of chiral
dehydroalanine equivalent 2 (Fig. 1).5
N
N
O
CO2H
R
O
NH2
(S)-α-amino acids
2
MeO
As an extension of this methodology, we wished to
investigate the utility of N,N0-bis-(p-methoxybenzyl)-
protected diketopiperazines as asymmetric glycine cat-
ion equivalents. Although a number of cyclic auxiliary
based asymmetric syntheses of a-amino acids using this
strategy have been reported,6 our initial investigations in
this area showed that the addition of allyltrimethylsilane
to N-acyliminium ion 5 (derived from either of the dia-
Figure 1.
stereoisomeric acetates 3 or 4) gave an 80:20 mixture of
allylation products 6 and 7 (Scheme 1).7
While this approach demonstrated the viability of the
diketopiperazine template to act as a glycine cation
equivalent, it was envisaged that the addition of organo-
magnesium reagents to the chiral glycine cation 5 would
offer greater scope for the asymmetric synthesis of
a-amino acids, potentially allowing the stereoselective
*
Corresponding author. Tel.: +44 01865 275680; fax: +44 01865
275674; e-mail: steve.davies@chem.ox.ac.uk
0957-4166/$ - see front matter ꢀ 2004 Published by Elsevier Ltd.
doi:10.1016/j.tetasy.2004.11.022

3990
S. D. Bull et al. / Tetrahedron: Asymmetry 15 (2004) 3989–4001
OMe
OMe
(i)
OMe
(i)
OMe
N
N
O
N
N
O
N
N
O
N
N
O
H
5
R1
2
O
O
O
+
O
Cl
R2
5
1
MeO
MeO
MeO
8
MeO
3 R1 = OAc, R2 = H
4 R1 = H, R2 = OAc
aqueous
work-up
or moisture
OMe
OMe
OMe
N
N
O
H
N
N
O
N
N
O
+
O
H
O
R1
R2
O
MeO
MeO
9 R1 = OH, R2 = H
10 R1 = H, R2 = OH
:
20
80
MeO
6
7
Scheme 2. Reagents and conditions: (i) LHMDS, THF, À78ꢁC, C2Cl6.
Scheme 1. Reagents and conditions: (i) allyltrimethylsilane, BF3ÆOEt2,
CH2Cl2, À78ꢁC.
OMe
(i)
OMe
introduction of a wide variety of substituents. We report
herein our studies directed towards the asymmetric syn-
thesis of a-amino acids based upon addition of Grignard
reagents to 2-halo-diketopiperazines as glycine cation
equivalents.8
N
N
O
N
N
OLi
O
O
2. Results and discussion
1
MeO
MeO
2.1. Synthesis of 2-halo-diketopiperazines
95% combined yield
OMe
Initial investigations focused on the preparation of 2-
chloro-diketopiperazine 8 via an electrophilic chlorin-
ation protocol.6a Treatment of the lithium enolate of 1
with hexachloroethane at low temperature followed by
the removal of solvent revealed a crude reaction mixture
containing predominantly cis-(2S,5S)-chloro 8 (vide
infra) (Scheme 2). However, 2-chloro-diketopiperazine
8 was readily hydrolysed to the known hydroxy diketo-
piperazines trans-9 and cis-107 (in a 33:67 ratio, respec-
tively) upon exposure to moisture, and chloro 8
obtained without aqueous workup was necessarily con-
taminated with inorganic salts and excess reagents. With
the limitations of this protocol delineated, an alternative
approach to the synthesis of 8 was pursued.
OMe
N
N
O
H
N
N
O
F
5
5
+
2
2
F
H
O
O
MeO
MeO
:
33
67
11
12
Scheme 3. Reagents and conditions: (i) LHMDS, THF, À78ꢁC,
N-fluorobenzenesulfonimide.
The preparation of the potentially more stable fluoro
substituted diketopiperazine was next investigated.
Treatment of the lithium enolate of 1 with N-fluorodi-
benzenesulfonimide afforded a 67:33 mixture of epimeric
fluorides trans-(2R,5S)-11 and cis-(2S,5S)-12, respec-
tively, in good purity and high combined yield (95%)
(Scheme 3). This oily mixture was stable to aqueous
workup, but unstable to chromatographic purification,
with characterisation data and subsequent manipul-
ations being performed upon the mixture of 11 and 12.
The observed 67:33 mixture of trans-fluoro-11 and cis-
fluoro-12 is consistent with predominant trans selectivity
observed in alkylation of the enolate of 1,4 while the
poor diastereoselectivity of the reaction may reflect the
high reactivity of the fluorinating agent.
Chloro substituted diketopiperazine cis-8 was subse-
quently prepared from the 67:33 mixture of fluorides
11 and 12 by treatment with chlorotrimethylsilane
(TMSCl) in dichloromethane to afford, upon removal
of volatile residues, (2S,5S)-chloro 8 as a single diaste-
reoisomer (>95% de) and in excellent yield (95%), free

S. D. Bull et al. / Tetrahedron: Asymmetry 15 (2004) 3989–4001
3991
OMe
of inorganic salts and excess reagents. Similar treatment
2.4 %
4.0 %
of fluoride mixture 11 and 12 with bromotrimethylsilane
(TMSBr) afforded (2S,5S)-bromo 13 as a single diaste-
reoisomer (>95% de) in quantitative yield. As noted
above cis-chloro 8, and cis-bromo 13, obtained from this
route were moisture sensitive and unstable to chroma-
tography, and so were characterised without further
purification (Scheme 4).
Me
O
2.0 %
H
H
N
5
1.0 %
O
H
Me
2
N
Cl
MeO
8
OMe
OMe
Figure 2. Selected NOE difference enhancement for 8.
N
N
O
H
N
N
O
F
by the isopropyl group. However for N,N-dialkyl substi-
tuted diketopiperazines cis isomers have been shown to
be thermodynamically more stable, presumably due to
the introduction of 1,2 torsion strain caused by the inter-
action between the N-alkyl and ring substituents in the
corresponding trans isomers.8e,11
+
F
H
O
O
MeO
MeO
:
33
67
11
12
Chloro 8 and bromo 13 are obtained as single, cis diaste-
reoisomers and most probably derive from equilibration
of the kinetic addition products under the reaction con-
ditions. While initial delivery of halide onto the Re-face
of N-acyliminium ion 5, anti to the isopropyl group will
preferentially afford the (2R)-halo compounds 14 or 15,
epimerisation of these products is expected to provide
the thermodynamically more stable cis-(2S)-halo com-
pounds 8 or 13. This isomerisation may potentially
occur via either a N-acyliminium intermediate or an
enolisation pathway, although in support of a N-acyl-
iminium mediated epimerisation, the treatment of
chloro 8 with TMSBr gave bromo 13, suggesting that
the abstraction of chloride from 8 by TMSBr is a viable
process (Scheme 5).
(i) or (ii)
OMe
N
O
H
X
5
2
O
N
(i) 8 X = Cl, >95%, >95%d.e.
(ii) 13 X = Br, quant, >95%d.e.
MeO
Scheme 4. Reagents and conditions: (i) TMSCl, CH2Cl2, room
temperature; (ii) TMSBr, CH2Cl2, room temperature.
The relative configuration within (2S,5S)-chloro 8 was
assigned from H NMR spectroscopic data. The diag-
1
11 + 12
nostic difference in chemical shift between the diastereo-
topic isopropyl methyl groups (DdMe 0.09ppm)
indicated a cis relative configuration of the C-2 chloro
(i) or (ii)
OMe
1
OMe
and C-5 isopropyl groups. Similar H NMR spectro-
scopic data for (2S,5S)-bromo 13 (DdMe 0.09ppm) also
indicated a cis relative configuration for this com-
pound.9 Furthermore, 1H NMR NOE. difference exper-
iments for chloro 8 also revealed a small (1%) signal
enhancement of 2-H upon irradiation of 5-H, suggesting
that these hydrogen atoms lie on the same face of the
diketopiperazine ring, consistent with the assigned cis
configuration. The absolute (2S,5S)-configuration with-
in chloro 8 and bromo 13 follows from the known C-5
stereogenic centre derived from (S)-valine (Fig. 2).
N
N
O
X
H
N
N
O
O
X
O
F
Si
14 X = Cl
15 X = Br
MeO
MeO
5
OMe
OMe
The novel trimethylsilylhalide mediated transformation
of the mixture of epimeric fluorides 11 and 12 provides
cis-chloro 8 or cis-bromo 13 in high de (>95%). The
reaction presumably proceeds reversibly via N-acylimin-
ium intermediate 5, formed under Lewis acid catalysis
by the trimethylsilylhalide.10 Previous investigations
concerning the alkylation of diketopiperazine template
1 and addition of allyltrimethylsilane to N-acyliminium
species 5 under kinetic conditions have shown that the
corresponding trans-substituted diketopiperazines are
formed predominantly with diastereocontrol directed
N
N
O
N
N
O
X
H
X
O
O
+
X
Si
MeO
MeO
8 X = Cl
5
(ii)
13 X = Br
Scheme 5. Reagents and conditions: (i) Me3SiCl, CH2Cl2, room
temperature; (ii) Me3SiBr, CH2Cl2, room temperature.

3992
S. D. Bull et al. / Tetrahedron: Asymmetry 15 (2004) 3989–4001
OMe
In the reaction of the lithium enolate of 1 with hexachlo-
roethane the anomalous diastereoselectivity observed
may also result from initial chlorination of the enolate
on the Re-face of the auxiliary, to afford trans-
(2R,5S)-chloro 14, followed by equilibration to provide
cis-(2S,5S)-epimer 8.
N
N
O
H
5
2
Cl
O
8
2.2. Substitution reactions
(i) or (ii)
MeO
With the 2-fluoro-, 2-chloro- and 2-bromo-diketopiper-
azines in hand, their utility in synthetic transformations
was assessed. Preliminary investigations into the use of
2-halo-diketopiperazines 8, 11, 12 and 13 for synthesis
revealed that chloro 8 was the simplest, in practical
terms, for evaluation as a glycine cation equivalent.
The reaction of 8 with allyltrimethylsilane was initially
examined and, while no reaction was observed in the ab-
sence of halophilic reagents, treatment of 8 with SbCl5
and allyltrimethylsilane provided an 80:20 mixture of
trans-allyl 6 and cis-allyl 7 (Scheme 6).12 The observed
80:20 ratio of trans-allyl 6 and cis-allyl 7 is in accord
with the 80:20 ratio of 6 and 7 observed in the related
reaction of acetates 3 or 4 with allyltrimethylsilane in
the presence of BF3ÆOEt2, consistent with both reactions
proceeding via the same N-acyliminium intermediate 5.
OMe
OMe
N
O
N
O
5
5
SPh
H
H
2
2
+
O
N
O
N
SPh
MeO
(ii)
MeO
:
:
(i) 77
(ii)
7
23
93
17
16
16
Scheme 7. Reagents and conditions: (i) 0.9equiv NaSPh, THF;
(ii) 2equiv NaSPh, THF.
cis-(2S,5S)-sulfide 17 was isolated in 83% yield after
chromatography.
OMe
OMe
Attempts to prepare an authentic sample of trans-sulfide
16 via treatment of the lithium enolate of 1 with diph-
enyldisulfide, a protocol which has been shown to pre-
dominantly afford trans addition products in similar
alkylation reactions,4 afforded a 7:93 mixture of trans-
(2R,5S)-sulfide 16 and cis-(2S,5S)-sulfide 17 (17, 86%
de). Furthermore the reaction of the lithium enolate of
1 with S-phenyl benzenethiosulfonate gave a 20:80 mix-
ture of trans-16 and cis-17 (17, 60% de) (Scheme 8).
N
N
O
N
N
O
H
(i)
5
2
O
Cl
O
8
5
MeO
MeO
OMe
OMe
OMe
OMe
OMe
N
N
O
H
N
O
H
N
N
O
N
N
O
N
N
O
+
(i)
SPh
H
H
O
O
N
+
(ii) or (iii)
O
O
SPh
O
MeO
MeO
:
MeO
MeO
80
20
7
MeO
(ii)
:
:
93
80
17
1
7
6
(iii) 20
16
Scheme 6. Reagents and conditions: (i) CH2@CH2CH2SiMe3, SbCl5,
CH2Cl2, room temperature.
Scheme 8. Reagents and conditions: (i) LHMDS, THF, À78ꢁC; (ii)
PhS–SPh; (iii) PhS–SO2Ph.
The addition of sodium thiophenolate to chloro 8 was
next examined. Treatment of 8 with 0.9equiv of sodium
thiophenolate provided a 77:23 mixture of trans-
(2R,5S)-sulfide 16 and cis-(2S,5S)-sulfide 17 from which
trans-16 was isolated by chromatography (48% yield),
while treatment of chloro 8 with 2equiv of sodium thio-
phenolate gave a 7:93 mixture of trans-16 and cis-17,
respectively (17, 86% de) (Scheme 7). Furthermore,
treatment of trans-(2R,5S)-sulfide 16 (>98% de) with
2equiv of sodium thiophenolate also afforded a 7:93
mixture of trans-16 and cis-17, respectively, from which
The relative configurations of (2R,5S)-16 and (2S,5S)-17
were readily discerned from the H NMR spectroscopic
data via analysis of the characteristic isopropylmethyl
1
chemical
shift
differences
{trans-(2R,5S)-16,
DdMe = 0.31ppm; cis-(2S,5S)-17, DdMe = 0.07ppm}.7
The relative configuration within cis-(2S,5S)-sulfide 17
was unambiguously confirmed by single crystal X-ray
diffraction with the absolute configuration following
from the known configuration of the (S)-valine derived
stereogenic centre (Fig. 3).

S. D. Bull et al. / Tetrahedron: Asymmetry 15 (2004) 3989–4001
3993
OMe
OMe
Pmb
N
O
N
N
N
N
O
H
5
6
(i)
H
O
N
SPh
O
OR
O
Cl
Pmb
18 R = H
20 R = TMS
(ii), 77%
8
MeO
MeO
OMe
OMe
N
N
O
N
N
O
H
3
3
6
6
+
+
H
O
O
MeO
6
MeO
7
18 : 7 : 6 = 76: 12 : 12
Scheme 9. Reagents and conditions: (i) CH2@CH2CH2MgCl, THF,
À78ꢁC; (ii) trimethylsilylimidazole, DCM.
Figure 3. Chem 3Dꢂ representation of X-ray crystal structure of
(2S,5S)-sulfide 17 (some H omitted for clarity).
reaction with allyltrimethylsilane. In contrast 2-chloro-
diketopiperazine 8 should have the potential to react
with Grignard reagents, and as such the addition of a
range of organomagnesium species to chloro 8 was next
explored. The reaction of chloro 8 with allylmagnesium
chloride gave a 76:12:12 mixture of novel diallyl addi-
tion product (3S,6S)-18 and C-2 addition products
(3S,6R)-6 and (3S,6S)-7, respectively, from which the
major reaction product 18 was isolated in 60% yield
after chromatography (Scheme 9). The connectivity of
the carbon framework within diallyl 18 was determined
In the reaction of sodium thiophenolate with cis-chloro
8 the mixtures of epimeric sulfides 16 and 17 may plau-
sibly be formed via several different pathways. SN1 reac-
tion of chloro 8 can afford a mixture of trans- and cis-
sulfides 16 and 17, in which the trans-sulfide 16 may
be expected to predominate due to preferential attack
upon the Re-face of the N-acyliminium ion intermediate.
Alternatively, trans-sulfide 16 could arise from direct
SN2 displacement of chloride, with the cis-sulfide 17
deriving from C-2 epimerisation of the initially formed
trans-16. Thus, in the reaction of chloro 8 with 0.9equiv
of sodium thiolate the product ratio (77:23, 16:17) is
suggestive of a SN1 process via N-acyliminium ion 5, gi-
ven that the 77:23 ratio of diastereoisomers 16:17 is sim-
ilar to the 80:20 ratio of diastereoisomers 6:7 derived
from reaction of chloro 8 with allyltrimethylsilane.
However these results are also consistent with, and
indistinguishable from, an alternative mechanistic path-
way involving direct SN2 displacement of chloride to
afford trans-16, accompanied by epimerisation to afford
cis-17. While epimerisation in this system may occur
via N-acyliminium intermediates, the reaction of 8 with
excess sodium thiophenolate and the reaction of the lith-
ium enolate of 1 with diphenyl disulfide, predominantly
afford cis-sulfide 17 (7:93 ratio of 16:17), suggesting a
thiophenolate catalysed equilibration via an enolate to
give the thermodynamically more stable cis isomer.8e
The 20:80 ratio of 16 to 17 observed in the addition of
enolate of 1 to S-phenyl benzenethiosulfonate may rep-
resent a partially equilibrated mixture due to the lower
basicity of the lithium phenylsulfinate by-product.13
from H and H–13C correlation NMR data. HMBC
correlations indicated that both allyl groups were
attached to the C-5 quaternary carbon centre, while
1H–1H coupling between H-6 and the adjacent hydroxyl
proton established the presence of tertiary C-6, dis-
counting plausible alternative structure 19. The relative
configuration of the isopropyl and C-6 hydroxyl groups
within 18 was established from NOE data for trimethyl-
silyl ether derivative 20, prepared from 18 by treatment
with trimethylsilylimidazole. These studies showed a
NOESY cross peak from the trimethylsilyl methyl
groups to one isopropyl methyl signal, indicating a cis
arrangement of the C-3 isopropyl substituent and the
C-6 oxygen atom, with the absolute (3S,6S)-configura-
tion following from the known (S)-valine derived stereo-
genic centre (Fig. 4).
1
1
Treatment of chloro 8 with methylmagnesium chloride
afforded a 56:17:22:5 mixture of dimethyl (3S,6S)-21 in
>95% de, reduction product 1 (presumably arising from
transmetallation) and the known C-2 addition products
cis-(3S,6S)-22 and trans-(3S,6R)-23 (22, 63% de)4 from
which 21, 1 and 22 were isolated in 44%, 7% and 14%
yield, respectively after chromatography. The configura-
tion of 21 was assigned by analogy to (3S,6S)-18, while
the configuration of 22 has previously been established4
(Scheme 10).
2.3. Reactions with Grignard reagents
Previous investigations using the acetate substituted chi-
ral glycine cation equivalents 3 and 4 indicated that their
synthetic utility was restricted to Lewis acid catalysed

3994
S. D. Bull et al. / Tetrahedron: Asymmetry 15 (2004) 3989–4001
OMe
OMe
OMe
OMe
Me
O
N
N
O
N
N
O
H
N
N
N
N
(i)
Me
3
5
6
OH
H
6
Ph
45%
O
O
Cl
O
O
H
SiMe3
MeO
OMe
MeO
MeO
8
24 >95% d.e.
MeO
20
19
OMe
Figure 4. Selected NOESY correlations for 20.
N
N
O
N
N
O
H
3
(ii)
6
50%
OMe
O
O
Cl
MeO
MeO
N
N
O
8
25 >90% d.e.
O
Cl
Scheme 11. Reagents and conditions: (i) BnMgCl, THF, À78ꢁC;
(ii) (CH3)2CHMgCl, THF, À78ꢁC.
8
MeO
sium chloride gave cis configured (3S,6S)-diisopropyl 25
as the major product comprising 60% of the crude mix-
ture, in >90% de, and afforded 25 in 50% yield and
(i)
1
OMe
OMe
>98% de after chromatography (Scheme 11). H N MR
spectroscopic data for 25 were not consistent with the
known achiral trans-diisopropyl isomer derived from
the enolate alkylation protocol, and the H NMR spec-
Me
Me
H
1
N
N
N
O
3
23
+
+
troscopic data and specific rotation {½aŠ ¼ À253:6 (c
6
D
O
N
O
2.06, CHCl3)} indicated a chiral cis configured diisopro-
pyl substituted diketopiperazine. The relative configura-
tion of (3S,6S)-diisopropyl 25 was unequivocally
confirmed by X-ray crystallographic analysis, with the
absolute configuration following from the known
(S)-valine derived stereogenic centre (Fig. 5). This
OH
MeO
21
MeO
OMe
1
OMe
N
O
OMe
N
O
3
3
6
H
+
6
Me
O
N
Me
O
N
H
N
N
O
O
MeO
22
23
MeO
21 :1 : 22 : 23 = 56 : 17 : 22 : 5
MeO
Scheme 10. Reagents and conditions: (i) MeMgCl, THF, À78ꢁC.
Treatment of chloro 8 with benzylmagnesium chloride
provided C-2 addition product trans-(3S,6R)-benzyl 24
in >95% de as the major product, comprising 70% of
the crude mixture, along with a number of unidentified
minor products, from which 24 was isolated in 45% yield
and >98% de after chromatography (Scheme 11). The
relative configuration within (3S,6R)-benzyl 24 was
established by comparison with an authentic sample, ob-
tained from the alkylation of the lithium enolate of 1,
the configuration of which has been unequivocally
established previously by X-ray crystal structure deter-
mination and deprotection and hydrolysis to the constit-
uent a-amino acids, (S)-valine and (R)-phenylalanine.14
In contrast, treatment of chloro 8 with isopropylmagne-
Figure 5. Chem3DTM representation of X-ray crystal structure of
(3S,6S)-diisopropyl 25 (some H omitted for clarity).

S. D. Bull et al. / Tetrahedron: Asymmetry 15 (2004) 3989–4001
3995
MgCl
assignment was also confirmed by the preparation of an
authentic sample of 25 from (S,S)-cyclo-(val-val)
+
8
23
D
{½aŠ ¼ À257:4 (c 4.05, CHCl3)}. In this case the syn-
OMe
OMe
thetic protocol represents a formal self-reproduction of
chirality in which a second valine unit is generated under
the control of the starting (S)-valine stereogenic centre.
OMe
N
N
O
2.4. Model for the regio- and stereoselectivity of Grignard
additions to 8
O
N
O
N
Mg
O
Cl
Cl
26
The reaction of 8 with Grignard reagents affords prod-
ucts formally arising from addition of the alkyl magne-
sium halide to either the C-3 carbonyl group or C-2 of
chloro 8. The differences in the regioselectivity and stere-
oselectivity of Grignard additions must reflect reactivity
and structural differences between the organomagne-
sium species in these reactions. In the addition of Grig-
nard reagents to carbonyl compounds via a polar
mechanism, the relative reactivity order is determined
by a combination of steric bulk and carbanion stability
and has been established as allyl > benzyl >
methyl > isopropyl.15 The different modes of reaction
of these reagents with 8 may then be expected to reflect
this reactivity order. For all Grignard reagents em-
ployed in reaction with 8, the reaction reasonably pro-
ceeds by initial co-ordination of magnesium to the C-3
carbonyl oxygen followed by either: attack at the C-3
carbonyl and subsequent reaction (Fig. 6, path A), or
formation of an N-acyliminium species (by Lewis acid
catalysed extraction of chloride) and attack on this func-
tionality (Fig. 6, path B).
MeO
OMe
OMe
N
N
N
O
H
O
O
O
N
H
28
27
MeO
OMe
MeO
N
H
O
N
OH
MeO
18
Figure 7. Possible mechanism for allylmagnesium chloride addition
and hydroxyl migration to afford 18.
In the reaction with allylmagnesium chloride, fast attack
on the C-3 carbonyl group of 8 (Fig. 6, path A), anti to
the C-5 isopropyl group is promoted by a kinetically
favourable six membered cyclic transition state. Follow-
ing this addition loss of chloride affords N-acyliminium
intermediate 26, which undergoes intramolecular cycli-
sation to afford epoxy species 27. Subsequent reaction
with a further equivalent of Grignard reagent, via imin-
ium ion 28, gives diallyl 18 (Fig. 7). The diastereoiso-
meric minor addition products 6 and 7 presumably
arise from competitive direct addition to N-acyliminium
ion intermediate 5 (Fig. 6, path B).
The addition of methylmagnesium chloride to 8 affords
a mixture of 21, 22 and 23. This mixture must arise from
a competition between addition to the C-3 carbonyl
group, followed by hydroxyl migration and further
addition to afford 21, and direct addition to the N-acyl-
iminium carbon centre of 5 to give 22 and 23. The obser-
vation of products from both these reaction modes
reflects the lower reactivity of this reagent in carbonyl
additions allowing competitive formation and reaction
of the N-acyliminium intermediate 5.
OMe
OMe
X
Mg
X
N
N
O
N
N
O Mg
3
2
B
R
R
O
Cl
-Cl
O
8
5
MeO
MeO
A
OMe
OMe
Given the high reactivity of benzylmagnesium chloride
toward carbonyl addition, the reaction of benzylmagne-
sium chloride with 8 may reasonably be expected to
proceed via addition to the C-3 carbonyl group. Fur-
thermore the reaction of benzylic Grignard reagents
with a number of simple aldehydes and ketones has been
shown to afford predominantly products arising from
ortho addition.16 The observed C-2 addition product is
consistent with stereoselective ortho addition at C-3, anti
to the C-5 isopropyl group, followed by N-acyliminium
ion formation, to give unstable intermediate 29, which
R
N
N
R
N
O
O
O
Cl
-Cl
O
N
MeO
MeO
Figure 6. Potential reaction pathways in the reaction of 8 with
Grignard reagents.

3996
S. D. Bull et al. / Tetrahedron: Asymmetry 15 (2004) 3989–4001
MeO
ever, the observed stereoselectivity of the isopropylmag-
OMe
nesium chloride addition may be rationalised by a
mechanism involving initial co-ordination of the Grig-
nard reagent to the C-3 carbonyl group followed by for-
mation of the N-acyliminium species and attack by a
second equivalent of isopropyl magnesium chloride. In
this addition, the trajectory of approach to the N-acyl-
iminium carbon centre will be hindered by the proximal
isopropyl magnesium chloride co-ordinated on the
Re-face of the auxiliary, thus directing attack onto the
Si-face, affording cis-substituted-25 (Fig. 9). Presumably
a similar mode of reaction and stereocontrol operates to
afford the cis-methyl addition product 22.
OMe
R
Mg
N
N
O
Cl
O
N
O
Cl
N
Mg
O
Cl
Cl
8
R = Bn
MeO
MeO
OMe
N
N
N
O
O
Ph
3. Conclusion
O
N
O
+
In conclusion, reactive 2-chloro and 2-bromo diketopip-
erazine derivatives 8 and 13 have prepared from fluoro-
diketopiperazines 11 and 12 via a novel silylhalide
mediated transhalogenation reaction. The reactions of
2-chloro 8 with allyltrimethylsilane, sodium thiopheno-
late and Grignard reagents have been assessed. The
regio- and stereoselectivity observed in the addition of
Grignard reagents was found to be dependent on the
nature of the organomagnesium reagent and affords, in
high de, either rearranged C-3-carbonyl addition
products (18, 21 and 24) upon the addition of allyl-,
methyl- or benzylmagnesium chloride, or cis-substituted
diketopiperazine 25, from direct addition to the N-acyl-
iminium carbon centre, upon addition of isopropylmag-
nesium chloride.
trans-24
29
MeO
MeO
Figure 8. Possible mechanism of benzylmagnesium chloride addition
to 8.
subsequently undergoes preferential alkyl group migra-
tion to give trans-substituted diketopiperazine 24
(Fig. 8).
In the case of isopropyl magnesium chloride addition,
the lower reactivity and steric bulk of the reagent is ex-
pected to retard the rate of attack onto the C-3 carbonyl
group, resulting in the reaction of the N-acyliminium ion
predominating (Fig. 6, path B). The observed cis stereo-
selectivity of this addition contrasts with the predomi-
nant trans addition of allytrimethylsilane to the similar
N-acyliminium ion derived from acetates 3 or 4.7 How-
4. Experimental
4.1. General experimental
All reactions involving organometallic or moisture sen-
sitive reagents were performed under an atmosphere of
nitrogen via standard vacuum line techniques. All glass-
ware was flame dried and allowed to cool under vac-
uum. In all cases, the reaction diastereoselectivity was
MeO
OMe
R
N
N
O
1
Mg
N
N
O
assessed by peak integration in the H NMR spectrum
Cl
of the crude reaction mixture. Tetrahydrofuran was dis-
tilled under an atmosphere of dry nitrogen from sodium
benzophenone ketyl. All other solvents were used as
supplied (Analytical or HPLC grade), without prior
purification. Thin layer chromatography was performed
on aluminium sheets coated with 60 F254 silica. Sheets
were visualised using 10% phosphomolybdic acid in eth-
anol. Chromatography was performed on Kieselgel 60
silica. Nuclear magnetic resonance (NMR) spectra were
recorded on a Bruker AC 200 (1H: 200MHz and 13C:
50.3MHz) or Bruker DPX 400 (1H: 400MHz and 13C:
100.6MHz) spectrometer in the deuterated solvent
stated. All chemical shifts (d) are quoted in ppm and
coupling constants (J) inHz. Coupling constants are
quoted twice, each being recorded as observed in the
spectrum without averaging. Residual signals from the
solvents were used as an internal reference. 13C multi-
plicities were assigned using a DEPT sequence. Infrared
spectra were recorded on a Perkin–Elmer 1750 IR
O
O
Cl
8
R = iPr
cis-25
MeO
MeO
MeO
OMe
OMe
R
Mg
N
O
X
R
O
O
N
+
Mg
N
N
ClMg
R
+
O
Cl
Si-face
Cl
5
Mg
R
MeO
Figure 9. Possible mechanism of isopropylmagnesium chloride addi-
tion to 8.

S. D. Bull et al. / Tetrahedron: Asymmetry 15 (2004) 3989–4001
3997
Fourier Transform spectrophotometer using either thin
films on NaCl plates (film) or KBr discs (KBr) as stated.
Only the characteristic peaks are quoted in cmÀ1. Opti-
cal rotations were recorded on a Perkin–Elmer 241
polarimeter with a path length of 1dm. Concentrations
are quoted in g/100mL. Low resolution mass spectra
were obtained upon a VG micromass ZAB IF, a VG
MassLab 20-250, a VG Bio Q or an APCI Platform
spectrometer with only molecular ions, fragments from
molecular ions and major peaks being reported. High
resolution mass spectroscopic data was obtained upon
Micromass AutoSpec or Micromass ToFSpec spectro-
meter. Elemental analyses were performed by the
microanalysis service of the Inorganic Chemistry
Laboratory, University of Oxford.
1.24 (d, 1H, 6.90, CH3CHCH3), 2.58 (m, 1H,
CH3CHCH3), 3.75 (d, 1H, J 7.0, 5-H), 3.85 (s, 3H,
OMe), 3.85 (s, 3H, OMe), 3.98 (d, 1H, J 14.6, ArCH2),
4.01 (d, 1H, J 14.6, ArCH2), 5.32 (d, 1H, J 14.6,
ArCH2), 5.40 (d, 1H, J 14.6, ArCH2), 5.70 (s, CHCl),
6.88–6.95 (m, 4H, aromatic H), 7.12–7.15 (m, 2H, aro-
matic H), 7.20–7.31 (m, 2H, aromatic H); irradiation
at d 3.75 gave NOE enhancements at d 1.15 (2%), 1.24
(2.4%), 2.58 (4%), 3.98 (2%), 5.40 (1.2%), 5.70 (1%)
and 7.12–7.15 (2.4%); dC (125MHz, CDCl3): 19.3,
20.6, 31.9, 46.3, 49.0, 55.2 · 2, 64.9, 68.2, 114.3, 114.4,
126.0, 126.8, 129.4, 130.2, 159.4, 159.7, 161.1, 166.8;
m/z (CI) 448 (M+NH4+, 14%), 395 (76), 190 (100).
Found: M+ÀCl, 395.1962. C23H27N2O4 requires
+
395.1971.
4.2. (2R,5S)- and (2S,5S)-N,N0-Bis-(p-methoxybenzyl)-2-
fluoro-5-isopropyl-piperazine-3,6-dione 11 and 12
4.4. (2S,5S)-N,N0-Bis-(p-methoxybenzyl)-2-bromo-5-iso-
propyl-piperazine-3,6-dione 13
To a solution of 14b (2.0g, 5.05mmol) in dry THF
(30mL) was added lithium hexamethyldisilazide
(5.6mL, 1M THF solution, 5.6mmol) at À78ꢁC after
stirring for 1h the mixture was treated with N-fluoro-
benzenesulfonimide (1.75g, 5.6mmol). The mixture stir-
red for 30min then warmed to À30ꢁC and aqueous
saturated NH4Cl (10mL) was added. Water (100mL)
was then added and the mixture was extracted with
ether, the organic layer was dried (MgSO4) and the sol-
vent removed to afford a 67:33 mixture of trans and cis-
fluorides 11 and 12 as an oil (2.05g, 98%).
Fluorides 11 and 12 (2:1 mixture, 400mg, 0.97mmol)
and bromotrimethylsilane (1.0mL) were stirred in dry
dichloromethane (5mL) for 2h at room temperature.
The solvent and excess bromotrimethylsilane were then
removed in vacuo to afford bromo 13 as a viscous oil
1
(460mg, 100%), which was >95% pure as assessed H
23
NMR. ½aŠ ¼ À92:0 (c 2.10, CHCl3); tmax (film)/cmÀ1
D
2964, 2837, 1651, 1514; dH (400MHz, CDCl3): 1.13 (d,
1H,
J 6.8, CH3CHCH3), 1.21 (d, 1H, J 6.9,
CH3CHCH3), 2.71 (m, 1H, CH3CHCH3), 3.71 (d, 1H,
J 7.5, 3-H), 3.81 (s, 3H, OMe), 3.82 (s, 3H, OMe),
3.87 (d, 1H, J 14.5, ArCH2), 3.93 (d, 1H, J 14.9,
ArCH2), 5.29 (d, 1H, J 14.5, ArCH2), 5.37 (d, 1H, J
14.9, ArCH2), 5.88 (s, CHBr), 6.81–6.90 (m, 4H, aro-
matic CH), 7.06–7.21 (m, 4H, aromatic CH); dC
(100MHz, CDCl3): 19.6, 20.8, 31.2, 46.7, 49.4, 55.26,
55.30, 59.6, 65.1, 114.4, 114.5, 125.7, 126.9, 129.4,
130.2, 159.5, 159.8, 161.4, 167.1.
Data for mixture of 11 and 12*. tmax (film)/cmÀ1 2963,
2986, 2837, 1682, 1514; dH (500MHz, CDCl3) 0.70
(3H, d, J 6.9, CH3*CHCH3), 0.98 (3H, d, J 6.9,
CH3CHCH3), 1.07 (3H, d, J 7.0, CH3CHCH3*), 1.14
(3H, d, J 7.0, CH3CHCH3), 2.28 (2H, m, CH3CHCH3),
3.72–3.80 (2H, m, 2 · 3-H), 3.80 (12H, s, 4 · OMe), 3.87
(1H, d, J 14.7, ArCH2), 3.90 (1H, d, J 14.5, ArCH2*),
4.02 (1H, d, J 14.1, ArCH2*), 4.26 (1H, dd, J 14.6, 1.4
(4JHF), ArCH2N-1), 4.96 (1H, d, J 14.8, ArCH2), 5.27–
5.41 (3H, m, ArCH2*), 5.52 (1H, d, 58.7 (2JHF), 2-
H*), 5.60 (1H, d, 57.5 (2JHF), 2-H), 6.80–6.91 (8H, m,
aromatic CH), 7.09–7.29 (8H, m, aromatic CH); dC
(125MHz, CDCl3): 16.0*, 18.0, 19.3*, 20.1, 31.2*,
32.2, 45.2*, 46.4, 47.5 48.1, 55.2, 62.8, 64.6, 88.8 (d,
4.5. (3S,6R)- and (3S,6S)-N,N0-Bis-(p-methoxybenzyl)-6-
allyl-3-isopropyl-piperazine-2,5-dione 6 and 7
To chloride 8 (50mg, 0.12mmol) in dichloromethane
(5mL) was added antimony pentachloride (0.13mL,
1M in dichloromethane, 0.13mmol) then allyltrimeth-
ylsilane (50lL, 0.44mmol) and the mixture stirred 12h
at room temperature. The mixture was partitioned be-
tween water and dichloromethane, the organic phase
dried (MgSO4) and the solvent removed in vacuo to
afford a crude gum. Examination of the 1H N MR
spectrum of the reaction mixture indicated an 80:20
mixture of 6 and 7.
1
1JCF 197, CF), 91.9 (d, JCF 207, CF*), 113.6, 113.7,
114.2, 114.4, 114.9, 126.8, 129.6, 129.7, 130.0, 130.9,
159.5, 159.6, 160.0, 160.2, 164.4, 166.9.
4.3. (2S,5S)-N,N0-Bis-(p-methoxybenzyl)-2-chloro-5-iso-
propyl-piperazine-3,6-dione 8
Fluorides 11 and 12 (2:1 mixture, 1.0g, 2.41mmol) and
chlorotrimethylsilane (2mL) were stirred in dry dichlo-
romethane (20mL) for 2h at room temperature. The sol-
vent and excess chlorotrimethylsilane were then
removed in vacuo to afford a crude foam of chloride 8
(1.02g, 98%) which was >95% pure as assessed 1H
NMR. This moisture sensitive material was stable under
4.6. (2R,5S)-N,N0-Bis-(p-methoxybenzyl)-5-isopropyl-2-
thiophenyl-piperazine-3,6-dione 16
Chloride 8 (200mg, 0.46mmol) in THF (2mL) was
added to sodium phenylthiolate {prepared from
thiophenol (44mg, 0.40mmol) and sodium hydride
(16mg, 60% dispersion in oil, 0.4mmol)} in THF
(5mL). This mixture was stirred (12h, room tempera-
ture) then partitioned between ether and saturated aque-
ous copper sulfate solution, the organic phase dried
nitrogen at À20ꢁC for a several weeks and was generally
23
used immediately. ½aŠ ¼ À119:1 (c 1.11, CHCl3); tmax
D
(film)/cmÀ1 2963, 1681, 1612, 1513, 1249; dH
(500MHz, CDCl3): 1.15 (d, 1H, J 6.8, CH3CHCH3),

3998
S. D. Bull et al. / Tetrahedron: Asymmetry 15 (2004) 3989–4001
3
˚
(MgSO4) and the solvent removed in vacuo. Chroma-
tography (silica, 1:1 ether/hexane) gave 16 as a colour-
U = 2686A , Z = 4, l = 0.157, crystal dimensions
0.4 · 0.6 · 0.6mm, A total of 4315 unique reflections
were measured for 1 < h < 27 and 3816 reflections were
used in the refinement. The final parameters were
wR2 = 0.041 and R1 = 0.032 [I > 3r(I)].
23
less wax (112mg, 48%). ½aŠ ¼ À106:8 (c 1.10, CHCl3);
D
tmax (KBr disc) cmÀ1 2961, 1650, 1613, 1513, 1248; dH
(400MHz, CDCl3): 0.65 (3H, d, J 6.9, CH3CHCH3),
0.96 (3H, d, J 7.0, CH3CHCH3), 2.14 (1H, m,
CH3CHCH3), 3.17 (1H, d, J 2.7, 5-H), 3.79 (3H, s,
OMe), 3.81 (3H, s, OMe), 3.95 (1H, d, J 14.9, ArCH2),
4.39 (1H, d, J 14.1, ArCH2), 4.82 (1H, d, J 14.9,
ArCH2), 4.97 (1H, s, 2-H), 5.55 (1H, d, J 14.1, ArCH2),
6.75 (4H, m, aromatic CH), 6.86 (2H, m, aromatic CH),
7.24–7.40 (4H, m, aromatic CH), 7.45 (3H, m, aromatic
CH); dC (100MHz, CDCl3): 15.4, 19.6, 30.6, 45.7, 47.3,
55.26, 55.29, 62.6, 65.4, 114.0, 114.3, 126.6, 129.2, 129.3,
129.7, 129.9, 130.9, 136.1, 159.3, 159.5, 163.3, 164.6. m/z
(APCI+) 505 (MH+, 4%), 395 (MH+ÀSPh, 32), 121
(MeOC6H4CH2+, 100). [HRMS (TOF FI) Found: M+,
504.2091. C29H32N2O4S requires 504.2083].
Crystallographic data (excluding structure factors) has
been deposited with the Cambridge Crystallographic
Data Centre as supplementary publication number
CCDC 187478. Copies of the data can be obtained, free
of charge, on application to CCDC, 12 Union Road,
Cambridge CB2 1EZ, UK [fax: +44 01223 336033 or
deposit@ccdc.cam.ac.uk].
4.7.2. Alkylation of 1 with diphenyldisulfide. Lithium
hexamethyldisilazide (1.10mmol, 1.10mL, 1M in
THF) was added to 14b (400mg, 1.01mmol) in THF
(20mL) at À78ꢁC, under a nitrogen atmosphere. After
stirring for 1h at À78ꢁC, diphenyl disulfide (240mg,
1.1mmol) in THF (2mL) was added. The reaction mix-
ture was stirred at À78ꢁC for 30min and left to warm to
room temperature overnight before addition of excess
saturated aqueous ammonium chloride. The mixture
was partitioned between saturated copper sulfate solu-
tion and ethyl acetate and the aqueous phase extracted
with ethyl acetate. The combined organic layers were
dried (MgSO4), filtered and concentrated in vacuo to
yield the crude product. Chromatography (silica 1:3
ether/hexane) yielded 17 as a colourless crystalline solid
(246mg, 49%) with spectroscopic properties identical to
those described above.
4.7. (2S,5S)-N,N0-Bis-(p-methoxybenzyl)-5-isopropyl-2-
thiophenyl-piperazine-3,6-dione 17
trans-Sulfide 16 (35mg, 0.07mmol) in THF (1mL) was
added to sodium phenylthiolate {prepared from
thiophenol (15.2mg, 0.14mmol) and sodium hydride
(5.5mg, 60% dispersion in oil, 0.14mmol)} in THF
(3mL). This mixture was stirred (12h, room tempera-
ture) then partitioned between ether and saturated aque-
ous copper sulfate solution, the organic phase dried
(MgSO4) and the solvent removed in vacuo. Chroma-
tography (silica, 1:1 ether/hexane) afforded cis-sulfide
17 as a colourless solid (29mg, 83%). Mp 101ꢁC; (found:
C, 69.2; H, 6.4; N, 5.5. C29H32N2O4S requires C, 69.0;
4.7.3. Alkylation of 1 with S-phenyl benzenethiosulfo-
nate. Lithium
23
H, 6.4; N, 5.6); ½aŠ ¼ À255:2 (c 1.03, CH2Cl2); mmax
D
hexamethyldisilazide
(0.60mmol,
(KBr disc)/cmÀ1 1668 (NC@O), 1612, 1514, 1245; dH
(400MHz; CDCl3) 1.14 (3H, d, J 6.8, CH3CHCH3),
1.21 (3H, d, J 6.9, CH3CHCH3), 2.32 (1H, m,
CH3CHCH3), 3.70 (1H, d, J 7.2, 5-H), 3.74 (3H, s,
OMe), 3.79 (3H, s, OMe), 3.88 (1H, d, J 14.5, ArCH2),
3.91 (1H, d, J 14.8, ArCH2), 4.94 (1H, s, 2-H), 5.34 (1H,
d, J 14.5, ArCH2), 5.38 (1H, d, J 14.8, ArCH2), 6.67
(2H, m, aromatic CH), 6.74 (2H, m, aromatic CH),
6.84 (2H, m, aromatic CH), 7.01 (2H, m, aromatic
CH), 7.37 (3H, m, aromatic CH), 7.70 (2H, m, aromatic
CH); dC (100MHz, CDCl3): 19.3, 20.5, 33.1, 45.6, 49.3,
55.2, 55.3, 65.3, 67.1, 113.8, 114.0, 114.1, 114.3, 126.7,
127.5, 128.4, 128.5, 129.2, 129.4, 129.9, 132.8, 134.8,
159.3, 164.7, 165.6. m/z (APCI+) 505 (MH+, 4%), 121
(100%). Found 505.2161; C29H33N2O4S+ requires
505.2161.
1.10mL, 1M in THF) was added to 14b (200mg,
0.50mmol) in THF (10mL, degassed) at À78ꢁC, under
a nitrogen atmosphere. After stirring for 1h at À78ꢁC,
S-phenyl benzenethiosulfonate (153mg, 0.55mmol)
was added and the reaction mixture was stirred at
À78ꢁC for 4h then left to warm to room temperature
overnight before addition of excess saturated ammo-
nium chloride solution. The mixture was partitioned
between saturated copper sulfate solution and ethyl ace-
tate and the aqueous phase extracted with ethyl acetate.
The combined organic layers were dried (MgSO4),
filtered and concentrated in vacuo to yield a crude
1
gum (238mg). Analysis of the H NMR spectrum of
the crude reaction mixture showed a 80:20 mixture of
cis-17 and trans-16, respectively.
4.7.1. X-ray crystal structure data for 17. Data were
collected using a Nonius Kappa CCD diffractometer
with graphite monochromated Mo-Ka radiation using
standard procedures at 190K. The structure was solved
by direct methods (Sir92). All nonhydrogen atoms were
refined with anisotropic thermal parameters. Hydrogen
atoms were added at idealised positions. The structure
was refined using CRYSTALS.17
4.8. General procedure for the treatment of chloride 8 with
Grignard reagents
Alkylmagnesium chloride was added dropwise to freshly
prepared 8 in anhydrous THF (10mL) at À78ꢁC. This
mixture was stirred (4h, À78ꢁC), warmed to room tem-
perature over 12h then saturated NH4Cl (2mL) was
added. The mixture was then partitioned between water
and ether, the aqueous phase extracted with ether, the
organic layer dried (MgSO4) and the solvent removed
in vacuo to afford the crude product. Products were
isolated by chromatography.
Crystal data for 17 [C29H32N2O4S], colourless plate,
M = 504.64, orthorhombic, space group P212121,
˚
a = 9.3120(2)A, b = 16.9551(3)A, c = 17.0130(3)A,
˚
˚

S. D. Bull et al. / Tetrahedron: Asymmetry 15 (2004) 3989–4001
3999
4.9. (3S,6S)-N,N0-Bis-(p-methoxybenzyl)-5,5-diallyl-6-
hydroxy-3-isopropyl-piperazine-2-one 18
4.11. (3S,6S)-N,N0-Bis-(p-methoxybenzyl)-5,5-dimethyl-
6-hydroxy-3-isopropyl-piperazine-2-one 21, (3S,6S)-
N,N0-bis-(p-methoxybenzyl)-3-isopropyl-6-methyl-pipera-
zine-2,5-dione 22 and (3S)-N,N0-bis-(p-methoxybenzyl)-3-
isopropyl-piperazine-2,5-dione 1
Treatment of 8 (1.00g, 2.32mmol) with allylmagnesium
chloride (2.50mL, 2M solution in THF, 5.0mmol)
according to the General procedure gave after chroma-
tography (silica, 1:1 ether/hexane) 18 as a colourless
Treatment of 8 (430mg, 1.00mmol) with methylmagne-
sium chloride (0.50mL, 3M in ether, 1.5mmol) accord-
ing to the General procedure gave oily crude mixture
oil as the first eluting compound (672mg, 60%).
23
D
½aŠ ¼ À10:0 (c 1.05, CHCl3); tmax (film)/cmÀ1 3424,
2957, 2873, 1652 (NC@O), 1612, 1512, 1247; dH
(500MHz, CDCl3): 0.95 (d, 1H, J 6.8, CH3CHCH3),
1.26 (d, 1H, J 7.1, CH3CHCH3), 1.82 (dd, 1H, J 14.6,
7.4, CH2CH@CH2), 2.01 (m, 1H, (CH3CHCH3), 2.20
(dd, 1H, J 7.2 and 14.4, CH2CH@CH2), 2.43 (m, 2H,
CH2CH@CH2), 3.39 (d, 1H, J 17.4, ArCH2), 3.48 (d,
1H, J 1.7, 3-H), 3.55 (d, 1H, J 11.6, OH), 3.85 (s, 3H,
OMe), 3.85 (s, 3H, OMe), 4.17 (d, 1H, J 14.3, ArCH2),
4.30 (d, 1H, J 11.6, CHOH), 4.40 (d, 1H, J 17.4,
ArCH2), 5.03 (d, 1H, J 14.3, ArCH2), 5.08-5.12 (m,
4H, CH2CH@CH2), 5.68–5.74 (m, 1H, CH2CH@CH2),
5.74–5.87 (m, 1H, CH2CH@CH2), 6.87–6.93 (m, 4H,
aromatic H), 7.23–7.32 (m, 2H, aromatic H), 7.36–7.39
(m, 2H, aromatic H); dC (125MHz, CDCl3): 16.7,
21.6, 30.9, 34.5, 38.7, 47.8, 53.6, 55.3 · 2, 62.8, 71.3,
81.3, 113.8, 113.9, 119.2, 119.8, 127.0, 129.1, 130.5,
132.3, 133.5, 134.5, 158.3, 159.0, 169.2; m/z (APCI+)
479 (MH+, 18%), 357(4), 341(5), 121(100); [HRMS
(TOF, FI) found: M+, 478.2826. C29H38N2O4 requires
478.2832].
(400mg). Chromatography (silica, 1:1 ether/hexane)
23
afforded 21 as a clear oil (187mg, 44%). ½aŠ ¼ À13:1
D
(c 1.05, CHCl3); tmax (film)/cmÀ1 3441(OH), 2959,
2934, 2872, 2834, 1651, 1511, 1245; dH (500MHz,
CDCl3): 0.81 and 0.93 (s, 3H, Me2C), 1.01 (d, 1H, J
6.9, CH3CHCH3), 1.24 (d, 1H, 7.1, CH3CHCH3), 2.02
(m, 1H, CH3CHCH3), 3.20 (d, 1H, J 1.7, 3-H), 3.40
(d, 1H, J 16.9, ArCH2), 3.44 (d, 1H, J 11.9, CHOH),
3.79 (s, 3H, OMe), 3.80 (s, 3H, OMe), 3.93 (d, 1H, J
16.2, ArCH2), 3.96 (d, 1H, J 14.2, ArCH2), 4.05, (d,
1H, J 11.9, CHOH), 5.21 (d, J 14.3, ArCH2), 6.82–
6.91 (m, 4H, aromatic H), 7.20–7.23 (m, 2H, aromatic
H), 7.28–7.31 (m, 2H, ArH); dC (125MHz, CDCl3):
16.1, 17.4, 21.5, 26.1, 31.5, 47.0, 53.0, 55.1, 55.2, 58.3,
70.5, 84.0, 113.8, 113.81, 127.5, 129.1, 130.4, 134.1,
158.3, 159.0, 169.1; m/z (APCI) 427 (MH+, 48%), 121
(100); [HRMS (CI) found: MH+, 427.2597.
C25H35N2O4 requires 427.2597].
Further elution gave cis-methyl 22 as a colourless oil
23
D
(61mg, 14%). ½aŠ ¼ À194 (c 1.00, CHCl3) [lit.4b À202
(c 0.89, CHCl3)]; dH (400MHz, CDCl3): 1.04 (d, 1H, J
6.9, (CH3CHCH3), 1.17 (d, 1H, J 7.00, CH3CHCH3),
1.55 (d, J 7.1, NCHCH3), 2.20 (m, 1H, CH3CHCH3),
3.74 (d, 1H, J 5.6, 3-H), 3.81 (s, 3H, OMe), 3.81 (s,
3H, OMe), 3.87 (d, 1H, J 14.8, ArCH2), 3.99 (q, J 7.5,
6-H), 4.00 (d, 1H, J 14.9, ArCH2), 5.11 (d, 1H, J 14.7,
ArCH2), 5.35 (d, 1H, J 14.8, ArCH2), 6.83–6.86 (m,
4H, aromatic H), 7.08–7.16 (m, 4H, aromatic H); m/z
(APCI) 411 (MH+, 90%), 303 (10), 121 (100). Spectro-
scopic data was identical to authentic material.
4.10. (3S,6S)-N,N0-4-Bis-(p-methoxybenzyl)-6-trimethyl-
siloxy-5,5-diallyl-3-isopropyl-piperazine-2-one 20
Compound 18 (26mg, 0.048mmol) and trimethylsilylim-
idazole (0.5mL) in dichloromethane (0.5mL) were stir-
red for 12h at room temperature then partitioned
between water and ether, the organic layer dried
(MgSO4) and solvent removed in vacuo. Chromatogra-
phy (silica, 1:9 ether–hexane) gave 20 as a colourless
solid (23mg, 77%). Mp 136ꢁC (ether/hexane). Found:
C, 69.6; H, 8.2; N, 5.1. C32H46N2O4Si requires C, 69.8;
Further elution gave 1 as a colourless solid (28mg, 7%).
dH (200MHz, CDCl3): 0.92 (3H, d, J 7.0, CH3CHCH3),
23
H, 8.4; N, 5.1; ½aŠ ¼ À36:4 (c 1.03, CHCl3); tmax
D
(KBr disc) cmÀ1 2946, 1653 (NC@O), 1611, 1513,
1240; dH (400MHz, CDCl3): 0.31 (s, 9H, Si(CH3)3),
1.00 (d, 3H, J 6.7, CH3CHCH3), 1.14 (d, 3H, J 7.7,
CH3CHCH3), 1.52 (dd, 1H, J 14.5, 7.0, CH2CH@CH2),
1.81 (m, 1H, CH3CHCH3), 2.15 (dd 1H, J 14.5, 8.0
CH2CH@CH2), 2.32 (d, 2H, J 7.1, CH2CH@CH2),
3.31 (d, 1H, J 18.2, ArCH2), 3.45 (d, 1H, J 0.6Hz,
3-H), 3.61 (d, 1H, J 14.3, ArCH2), 3.80 (s, 3H, OMe),
3.81 (s, 3H, OMe), 4.45 (d, 1H, J 18.2, ArCH2),
4.45 (s, 1H, 6-H), 4.92 (m, 4H, CH2CH@CH2),
1.10 (3H, d, J 7.0, CH3CHCH3), 2.22 (1H, m,
CH3CHCH3), 3.75 (1H, d, J 5.0, 3-H), 3.80 (1H, d, J
17.0, 6-H), 3.81 (3H, s, OMe), 3.81 (3H, s, OMe), 3.85
(1H, d, J 15.0, ArCH2), 3.94 (1H, d, J 17.5, 6-H), 4.21
(1H, d, J 14.0, ArCH2), 4.82 (1H, d, J 14.0, ArCH2),
5.33 (1H, d, J 15.0, ArCH2), 6.85–7.19 (8H, m, aromatic
H). Spectroscopic data was identical to authentic
material.
4.12. (3S,6R)-N,N0-Bis-(p-methoxybenzyl)-3-isopropyl-6-
benzyl-piperazine-2,5-dione 24
5.43 (d, 1H,
J
14.3, ArCH2), 5.53–5.75 (m,
2H, CH2CH@CH2), 6.82–6.88 (m, 4H, aromatic
CH), 7.34 (m, 2H, aromatic CH), 7.43 (m, 2H, aro-
matic H). NOESY cross peak observed between
dH = 0.31 and dH = 1.00; dC (100MHz, CDCl3): 0.8,
16.8, 21.3, 31.5, 35.1, 37.4, 47.4, 53.1, 55.2 (· 2), 61.9,
71.2, 82.6, 113.2, 113.8, 118.7, 119.0, 127.1, 129.2,
Treatment of 8 (430mg, 1.0mmol) with benzylmagne-
sium chloride (1.20mmol, 0.60mL, 2M solution in
THF) according to the General procedure gave an oily
crude mixture. Chromatography (silica, 1:1 ether/hex-
ane) gave 24 as the first eluting compound (218mg,
23
D
130.4, 132.8, 134.0, 136.1, 157.9, 159.1, 170.6
;
45%). Mp 168ꢁC; ½aŠ ¼ þ54:6 (c 1.09, CHCl3) [lit.4b
23
½aŠ ¼ þ58:6 (c 0.99, CHCl3)]; dH (400MHz, CDCl3)
m/z (TOF FI) 551 (M+, 97%) 509 (M+ÀCH2@CHCH2,
D
0.74 (3H, d, J 7.0, CH3CHCH3), 0.99 (3H, d, J 7.0,
100).

4000
S. D. Bull et al. / Tetrahedron: Asymmetry 15 (2004) 3989–4001
CH3CHCH3), 2.16 (1H, m, CH3CHCH3), 3.31 (1H, d, J
3.0, 3-H), 3.36 (1H, dd, J 14.5 and 4.0, PhCH2CH), 3.40
(1H, dd, J 14.5 and 4.0, PhCH2CH), 3.79 (1H, d, J 15.0,
MeOC6H4CH2), 3.79 (3H, s, OMe), 3.83 (3H, s, OMe),
3.98 (1H, d, J 15.0, MeOC6H4CH2), 4.26 (1H, t, J 4.0, 6-
H), 5.08 (1H, d, J 15.0, MeOC6H4CH2), 5.66 (1H, d, J
15.0, MeOC6H4CH2), 6.58–7.49 (13H, m, aromatic H);
m/z (APCI+) 487 (MH+, 100%), 379 (MH+ÀMeOC6H4,
8), 121(32). Spectroscopic data was identical to authen-
tic material.
4.13.2. Preparation of 25from ( S,S)-cyclo (val-
val). (S,S)-cyclo (Val-val)18 (390mg, 1.97mmol) was
added portion wise to NaH (156mg, 60% suspension
in oil, 3.96mmol) in DMF (50mL) at 0ꢁC followed by
dropwise addition of p-methoxybenzyl chloride
(0.53mL, 3.96mmol) and the mixture stirred for 2h at
0ꢁC then warmed to room temperature and stirred for
a further 12h. Water was added and the mixture parti-
tioned between ethyl acetate and water, aqueous layers
were extracted with ethyl acetate and organic fractions
dried (MgSO4) and the solvent removed in vacuo. Chro-
matography (1:1 ether–pentane) then crystallisation
4.13. (3S,6S)-N,N0-Bis-(p-methoxybenzyl)-3,6-diiso-
propyl-piperazine-2,5-dione 25
from dichloromethane/hexane afforded 25 as colourless
23
D
Spectroscopic data identical to that reported above.
plates (398mg, 46%). ½aŠ ¼ À257:4 (c 4.05, CHCl3).
Treatment of 8 (430mg, 1.00mmol) with isopropylmag-
nesium chloride (1.0mL, 2.0M solution in THF,
2.0mmol) according to the General procedure gave a
mixture containing 25 (60%) as assessed by analysis of
Acknowledgements
1
The authors thank New College, Oxford for a Junior
Research Fellowship (A.D.S.).
the H NMR spectrum of the crude reaction mixture.
Chromatography (silica, 1:1 ether/hexane) gave 25 as a
colourless solid (210mg, 51%). Mp 110ꢁC (ether).
Found: C, 71.0; H, 7.7; N, 6.3. C26H34N2O4 requires
References
23
C, 71.2; H, 7.8; N, 6.4; ½aŠ ¼ À253:6 (c 2.06, CHCl3);
D
tmax (KBr disc)/cmÀ1 2971, 2874, 1661 (C@O), 1613,
1513, 1245; dH (500MHz, CDCl3) 1.16 (d, 6H, J 6.6,
CH3CHCH3), 1.17 (d, 6H, J 6.8, CH3CHCH3), 2.15
(m, 1H, CH3CHCH3), 3.55 (d, 2H, J 9.5, 3-H, 6-H),
3.69 (d, 2H, J 14.8, ArCH2), 3.80 (s, 6H, 2 · OMe),
5.44 (d, 2H, J 14.8, ArCH2), 6.79–6.82 (m, 4H, aromatic
H), 6.99–7.02 (m, 4H, aromatic H); dC (125MHz,
CDCl3): 20.7, 21.0, 34.1, 49.8, 55.2, 66.1, 114.2, 128.0,
129.2, 159.2, 167.4; m/z (APCI) 439 (MH+, 100%), 121
(10). Ee was assessed as >98% by examination of the
1H NMR spectrum of 25 with the chiral solvating
reagent (R)-1,1,1-trifluoro-2-(9-anthryl)-ethanol and
comparison to racemic sample.
1. For a recent review see: Bloch, R. Chem. Rev. 1998, 98,
1407.
2. For a review, see: (a) Williams, R. M. In Synthesis of
Optically Active a-Amino Acids. Organic Chemistry Series;
Baldwin, J. E., Magnus, P. D., Eds.; Pergamon, 1989;
Vol. 7; Also see: (b) Harwood, L. M.; Tyler, S. N. G.;
Anslow, A. S.; MacGilp, I. D.; Drew, M. G. B. Tetra-
hedron: Asymmetry 1997, 8, 4007; (c) Williams, R. M.;
Hendrix, J. A. J. Org. Chem. 1990, 55, 3723.
3. For a review of N-acyliminium chemistry in synthesis see:
Hiemstra, H.; Speckamp, W. N. In Comprehensive Organic
Synthesis; Trost, B. M., Ed.; Pergamon: Oxford, 1991;
Vol. 2, pp 1047–1082; Speckamp, W. N.; Moolenaar, M. J.
Tetrahedron 2000, 56, 3817; For an example of Lewis acid
catalysed addition of Grignard reagents to N-acyliminium
precursors see: Harrison, J. R.; OꢀBrien, P. Tetrahedron
Lett. 2000, 41, 6167; For N-acylpyridinium salts see:
Comins, Joseph, S. P.; Goehring, R. R. J. Am. Chem. Soc.
1994, 116, 4719; For examples of Grignard additions to
iminium ions see: Gommermann, N.; Koradin, C.; Kno-
chel, P. Synthesis 2002, 2143; Bortolussi, M.; Cinquin, C.;
Bloch, R. Tetrahedron Lett. 1996, 8729; Bishop, M. J.;
McNutt, R. W. Bioorg. Med. Chem. Lett. 1995, 5, 1311.
4. (a) Bull, S. D.; Davies, S. G.; Epstein, S. W.; Ouzman, J.
V. A. Chem. Commun. 1998, 659; (b) Bull, S. D.; Davies, S.
G.; Epstein, S. W.; Leech, M. A.; Ouzman, J. V. A. J.
Chem. Soc., Perkin Trans. 1 1998, 2321; (c) Bull, S. D.;
Davies, S. G.; Garner, A. C.; Mujtaba, N. Synlett 2001,
781; (d) Bull, S. D.; Davies, S. G.; Fox, D. J.; Garner, A.
C.; Sellers, T. G. R. Pure Appl. Chem. 1998, 70, 1501.
5. (a) Bull, S. D.; Davies, S. G.; OꢀShea, M. D. J. Chem. Soc.,
Perkin Trans. 1 1998, 3657; (b) Bull, S. D.; Davies, S. G.;
Garner, A. C.; OꢀShea, M. D. J. Chem. Soc., Perkin Trans.
1 2001, 3281.
4.13.1. X-ray crystal structure data for 25. Data were
collected using a Nonius Kappa CCD diffractometer
with graphite monochromated Mo-Ka radiation using
standard procedures at 190K. The structure was solved
by direct methods (Sir92). All nonhydrogen atoms were
refined with anisotropic thermal parameters. Hydrogen
atoms were added at idealised positions. The structure
was refined using CRYSTALS.17
Crystal data for 19 [C14H19NO2], colourless plate,
M = 233.31, orthorhombic, space group Pcan,
˚
˚
˚
a = 12.0566(2)A, b = 14.1730(3)A, c = 14.7730(3)A, U =
3
˚
2524.4A , Z = 8, l = 0.082, crystal dimensions
0.4 · 0.4 · 0.8mm, a total of 2872 unique reflections
were measured for 4.36 < h < 27.48 and 2367 reflections
were used in the refinement. The final parameters were
wR2 = 0.0187 and R1 = 0.0373 [I > 3r(I)].
6. For other examples of glycine cation equivalents employed
in the asymmetric synthesis of a-amino acids see: (a)
Scho¨llkopf, U.; Neubauer, H.-J.; Hauptrief, M. Angew.
Chem., Int. Ed. Engl. 1985, 24, 1066; Scho¨llkopf, U.;
Crystallographic data (excluding structure factors) has
been deposited with the Cambridge Crystallographic
Data Centre as supplementary publication number
CCDC 238325. Copies of the data can be obtained, free
of charge, on application to CCDC, 12 Union Road,
Cambridge CB2 1EZ, UK [fax: +44 01223 336033 or
e-mail: deposit@ccdc.cam.ac.uk].
Gruttner, S.; Anderskewitz, R.; Egert, E.; Dyrbusch, M.
¨
Angew. Chem., Int. Ed. Engl. 1985, 26, 683; (b) Schickli, C.
P.; Seebach, D. Liebigs Ann. Chem. 1991, 655; (c)
Williams, R. M.; Hendrix, J. A. J. Org. Chem. 1990, 55,
3723; (d) Zhai, D.; Zhai, W.; Williams, R. M. J. Am.

S. D. Bull et al. / Tetrahedron: Asymmetry 15 (2004) 3989–4001
4001
Chem. Soc. 1988, 110, 2501; (e) Sinclair, P. J.; Zhai, D.;
Reibenspies, J.; Williams, R. M. J. Am. Chem. Soc. 1986,
108, 1103; (f) Belokon, Y. N.; Popkov, A. N.; Cherno-
glazova, N. I.; Bakhmutov, V. I.; Saporovskaya, M. B.;
Belikov, V. M. Izv. Akad. Nauk SSSR, Ser. Khim. 1989,
1899 (in Russian CA 112:198996); (g) Agami, C.; Couty,
F.; Daran, J. F.; Princ, B.; Puchot, C. Tetrahedron Lett.
1990, 31, 2889; (h) Kardassis, G.; Brungs, P.; Nothfelder,
C.; Steckhan, E. Tetrahedron 1998, 54, 3479, and refer-
ences cited therein.
11. (a) Eguchi, C.; Kakuta, A. J. Am. Chem. Soc. 1974, 96,
3985–3989; (b) Benedetti, E.; Marsh, R. E.; Goodman, M.
J. Am. Chem. Soc. 1976, 98, 6676; (c) Radding, W.;
Donzel, B.; Ueyama, N.; Goodman, M. J. Am. Chem. Soc.
1980, 102, 5999.
12. For an example of chloride abstraction with antimony
pentachloride, see: (a) Auricchio, S.; Magnani, C.; Trus-
cello, A. M. Eur. J. Org. Chem. 2002, 14, 2411; (b) Reetz,
M. T.; Maier, W. F.; Heimbach, H.; Giannis, A.;
Anastassious, G. Chem. Ber. 1980, 113, 3734.
7. Bull, S. D.; Davies, S. G.; Garner, A. C.; OꢀShea, M. D.;
Savoury, E. D.; Snow, E. J. J. Chem. Soc., Perkin Trans. 1
2002, 2442–2448.
13. pKa PhSH: 10.3 in DMSO; pKa PhS(O)OH 7.1 in DMSO,
Bordwell, F. G.; Hughes, D. L. J. Org. Chem. 1982, 47,
3224.
8. For examples of halogen substituted chiral glycine equiv-
alents see: (a) Heredia, M. L.; Fernandez, M.; de la
Cuesta, E.; Avendano, C. Tetrahedron: Asymmetry 2001,
12, 411; (b) Williams, R. M.; Sinclair, P. J.; Zhai, W. J.
Am. Chem. Soc. 1988, 110, 482; (c) Williams, R. M.;
Sinclair, P. J.; Zhai, D.; Chen, D. J. Am. Chem. Soc. 1988,
110, 1547; (d) Williams, R. M.; Kwast, A. J. Org. Chem.
1988, 53, 5785; (e) Williams, R. M.; Armstrong, R. W.;
Maruyama, L. K.; Dung, J.-S.; Anderson, O. P. J. Am.
Chem. Soc. 1985, 107, 3246; (f) Williams, R. M.;
Armstrong, R. W.; Dung, J.-S. J. Am. Chem. Soc. 1984,
106, 5748; (g) Badran, T. W.; Chai, C. L. L.; Easton, C. J.;
Harper, J. B.; Page, D. M. Aust. J. Chem. 1995, 48, 1379;
(h) Chai, C. L. L.; Hockles, D. C. R.; King, A. R. Aust. J.
Chem. 1996, 49, 1229; (i) Chai, C. L. L.; Page, D. M.
Tetrahedron Lett. 1993, 34, 4373; (j) Badran, T. W.;
Easton, C. J. Aust. J. Chem. 1990, 43, 1455; (k) Badran, T.
W.; Easton, C. J.; Horn, E.; Kociuba, K.; May, B. L.;
Schliebs, D. M.; Tiekink, R. T. Tetrahedron: Asymmetry
1993, 4, 197; (l) Scho¨llkopf, U.; Neubauer, H. J.; Hauptr-
eif, M. Angew. Chem., Int. Ed. Engl. 1985, 24, 1066; (m)
Scho¨llkopf, U.; Gruttner, S.; Anderskewitz, R.; Egert, E.;
Dyrbusch, M. Angew. Chem., Int. Ed. Engl. 1987, 26, 683;
(n) Whitlock, C. R.; Cava, M. P. Tetrahedron Lett. 1994,
35, 371; (o) Yoshimura, J.; Nakamura, H.; Matsunari, K.
Bull. Chem. Soc. Jpn. 1975, 48, 605; (p) Yoshimura, J.;
Sugiyama, Y.; Nakamura, H. Bull. Chem. Soc. Jpn. 1973,
46, 2850.
14. Davies, S. G.; Thomson, J. E. unpublished data.
15. Polar additions of Grignard reagents are observed for
addition to carbonyl groups with high reduction potentials
and the reactivity order for shown is for reaction with
acetone, see: (a) Holm, T.; Crossland, I. In Grignard
Reagents New Developments; Richey, H. G., Ed.; John
Wiley and Sons: Chichester, 2000; pp 1–26; (b) Compre-
hensive Organometallic Chemistry; Wilkinson, G., Stone,
F. G. A., Abel, E. W., Eds.; Pergamon, 1982; Vol. 1,
Chapter 4, pp 155–252; (c) Comprehensive Organometallic
Chemistry II; Abel, E. W., Stone, F. G. A., Wilkinson, G.,
Eds.; Pergamon, 1995; Vol. 1, Chapter 3, pp 57–128; (d)
Ashby, E. C. Pure Appl. Chem. 1980, 52, 545; (e) Ashby,
E. C.; Laemmle, J.; Neumann, H. M. Acc. Chem. Res.
1974, 7, 272; (f) Holm, T. Acta Chem. Scand. 1991, 45,
276; (g) Holm, T. Acta Chem. Scand. 1969, 23, 579;
For additions following a SET mechanism a different
reactivity order is observed (h) Holm, T. J. Org. Chem.
2000, 65, 1188; For the discussion of the structures of
methyl- and benzylmagnesium chloride in solution, see: (i)
Sakamoto, S.; Imamoto, T.; Yamaguchi, K. Org. Lett.
2001, 3, 1793, and references contained therein; (j) Ashby,
E. C.; Becker, W. E. J. Am. Chem. Soc. 1963, 85, 118; (k)
Ashby, E. C.; Smith, M. B. J. Am. Chem. Soc. 1964, 86,
4363.
16. (a) Benkeser, R. A.; Snyder, D. C. J. Org. Chem. 1982, 47,
1243; (b) Benkeser, R. A.; DeTalvo, W.; Darling, D. J.
Org. Chem. 1979, 44, 225; (c) Benkeser, R. A.; Siklosi, M.
P.; Mozdzen, E. C. J. Am. Chem. Soc. 1978, 100, 2134; (d)
Bernardon, C.; Deberly, A. J. Org. Chem. 1982, 47, 463;
(e) Bernardon, C. J. Organomet. Chem. 1989, 367, 11; (f)
Bernardon, C.; Deberly, A. J. Chem. Soc., Perkin Trans. 1
1980, 2631.
17. Watkin, D. J.; Prout, C. K.; Carruthers, J. R.; Betteridge,
P. W.; Cooper, R. I. CRYSTALS, Issue 11; Chemical
Crystallography Laboratory: Oxford, UK, 2001.
18. Prepared by the method of Nitecki, D. E.; Halpern, B.;
Westley, J. W. J. Org. Chem. 1968, 33, 864.
9. trans-Alkyl and hydroxy substituted N,N0-bis-p-meth-
oxybenzyl-diketopiperazines consistently display a larger
chemical shift differences between the diastereotopic iso-
propyl methyl groups (trans-DdMe = 0.20–0.31ppm) than
the corresponding cis configured diastereoisomers (cis-
DdMe = 0.10–0.13ppm). See Ref. 7.
10. For an example of tetrafluorosilane catalysed oxonium ion
formation see: Hashimoto, S.; Hayashi, M.; Noyori, R.
Tetrahedron Lett. 1984, 1379; Noyori, R.; Hayashi, M.
Chem. Lett. 1987, 57.