Epoxidation of trans-4-aminocyclohex-2-en-1-ol derivatives: Competition of hydroxy-directed and ammonium-directed pathways

Article abstract of DOI:10.1071/CH14531

N-Substituted trans-4-aminocyclohex-2-en-1-ols undergo epoxidation upon treatment with Cl3CCO2H followed by meta-chloroperbenzoic acid (m-CPBA) via competitive pathways resulting from hydrogen-bonding delivery by both the hydroxy group and the in situ formed ammonium ion. The absence of epoxide ring-opening in these reactions renders these substrates a unique platform for analysing the effect of the two competing directing groups on the rate of the epoxidation reaction: the diastereoisomeric ratio of the epoxide products is also the ratio of the rate constants describing the competing epoxidation processes (the group with the higher directing ability dominating the stereochemical course of the reaction). Analysis of the diastereoisomeric epoxide mixtures obtained from these reactions allowed the following order of directing group ability to be defined: NHBn >> NMeBn>OH>NBn2. The large difference in rate between the secondary and tertiary amino groups is consistent with superior directing ability of the former due to the presence of two hydrogen-bond donor sites on the secondary ammonium ion and/or an increased conformational flexibility to adopt an optimum geometry. The rate of an ammonium-directed epoxidation proceeds ～10 times slower in the presence of an allylic hydroxy group than in its absence, consistent with the presence of the additional, inductively electron-withdrawing heteroatom abating the nucleophilicity of the olefin. The relative rate of the hydroxy-directed epoxidation process in the presence of a more sterically demanding ammonium substituent is greater than that in the presence of a less sterically demanding one: this effect is attributed to an increased bias for the half-chair conformer in which the bulky ammonium substituent and, hence, the hydroxy group occupy pseudo-equatorial sites, thus allowing the latter to direct the reaction more efficiently. Journal compilation

Full text of DOI:10.1071/CH14531

RESEARCH FRONT
CSIRO PUBLISHING
Aust. J. Chem. 2015, 68, 610–621
Full Paper
http://dx.doi.org/10.1071/CH14531
Epoxidation of trans-4-Aminocyclohex-2-en-1-ol
Derivatives: Competition of Hydroxy-Directed
and Ammonium-Directed Pathways*
A
A
,
B
A
M e´ abh B. Brennan, Stephen G. Davies,
Ai M. Fletcher,
A
A
A
James A. Lee, Paul M. Roberts, Angela J. Russell,
and James E. Thomson
A
A
Department of Chemistry, Chemistry Research Laboratory, University of Oxford,
Mansfield Road, Oxford OX1 3TA, UK.
B
Corresponding author. Email: steve.davies@chem.ox.ac.uk
N-Substituted trans-4-aminocyclohex-2-en-1-ols undergo epoxidation upon treatment with Cl CCO H followed by
3
2
meta-chloroperbenzoic acid (m-CPBA) via competitive pathways resulting from hydrogen-bonding delivery by both the
hydroxy group and the in situ formed ammonium ion. The absence of epoxide ring-opening in these reactions renders these
substrates a unique platform for analysing the effect of the two competing directing groups on the rate of the epoxidation
reaction: the diastereoisomeric ratio of the epoxide products is also the ratio of the rate constants describing the competing
epoxidation processes (the group with the higher directing ability dominating the stereochemical course of the reaction).
Analysis of the diastereoisomeric epoxide mixtures obtained from these reactions allowed the following order of directing
group ability to be defined: NHBn .. NMeBn . OH . NBn . The large difference in rate between the secondary and
2
tertiary amino groups is consistent with superior directing ability of the former due to the presence of two hydrogen-bond
donor sites on the secondary ammonium ion and/or an increased conformational flexibility to adopt an optimum geometry.
The rate of an ammonium-directed epoxidation proceeds ,10 times slower in the presence of an allylic hydroxy group
than in its absence, consistent with the presence of the additional, inductively electron-withdrawing heteroatom abating the
nucleophilicity of the olefin. The relative rate of the hydroxy-directed epoxidation process in the presence of a more
sterically demanding ammonium substituent is greater than that in the presence of a less sterically demanding one: this
effect is attributed to an increased bias for the half-chair conformer in which the bulky ammonium substituent and, hence,
the hydroxy group occupy pseudo-equatorial sites, thus allowing the latter to direct the reaction more efficiently.
Manuscript received: 30 August 2014.
Manuscript accepted: 22 October 2014.
Published online: 25 February 2015.
Introduction
ꢁ4
3
ꢁ1 ꢁ1
s ; this observed rate constant was
k ¼ 33.8 ꢀ 10 dm mol
divided into the dissected rate constants k ¼ 30.8 ꢀ 10
ꢁ4
First reported in 1909, the epoxidation of an olefin using a
1]
syn
ꢁ1 ꢁ1
to define
[
3
dm mol
ꢁ1 ꢁ1
ꢁ4
3
peracid (Prileschajew reaction) is now ubiquitous throughout
organic chemistry. Independent studies by Henbest and Wilson
s
and kanti ¼ 3.0 ꢀ 10 dm mol
s
the rate of epoxidation on the face of the olefin syn to the
hydroxy group and the face of the olefin anti to the hydroxy
group, respectively (Fig. 2). The following important trends
were apparent from these data: (i) incorporation of an allylic
oxy-substituent (OH, OMe or OAc within 2–6) retards the rate
of epoxidation compared with that of cyclohexene (1); (ii) the
relative rate of epoxidation of 2 on the face syn to the hydroxy
group is anomalously fast compared with the analogous process
within methyl ether 3 (i.e. epoxidation on the face syn to the
OMe group) and acetate ester 4 (i.e. epoxidation on the face syn
to the OAc group); (iii) an equatorial hydroxy group (i.e. within
cis-5) results in superior diastereoselectivity and faster reaction
compared with an axial counterpart (i.e. within trans-6).
Henbest and Wilson attributed the reduced reactivity to the
[
1957) and Whitham et al. (1970) explored both the dia-
2]
[3]
(
stereoselectivity and relative rate of epoxidation (by peroxy-
benzoic acid in benzene at 58C) of several cyclohexene
derivatives bearing a substituent in the allylic position. The
combined list of substrates investigated included cyclohexene
(1) itself, cyclohex-2-enol (2), the corresponding methyl ether 3
and acetate ester 4, and both diastereoisomers of 5-tert-butyl-
[
2,3]
cyclohex-2-en-1-ol cis-5 and trans-6 (Fig. 1).
From the
epoxidation diastereoselectivities, Whitham et al. dissected the
observed (second-order) rate constants to describe the rates
of reaction on each of the individual faces of the olefins. For
example, epoxidation of 2 gave a 91 : 9 mixture of syn- and
anti-epoxides, respectively, with an observed rate constant of
*
Dedicated to the memory of Professor Sir John Cornforth.
Journal compilation Ó CSIRO 2015
www.publish.csiro.au/journals/ajc

Hydroxy- and Ammonium-Directed Epoxdiation
611
X
X
X
H
ꢁ
H
ꢁ
1
NRBn
1
NRBn
NRBn
PhCO H
3
2
3
ꢁ
(i)
(ii)
3
2
C H , 5ꢂC
O
O
2
6
6
O
R
R
R
3
1
–6
syn-epoxide
anti-epoxide
1
7
, R ꢃ H
, R ꢃ Me
, R ꢃ Bn
10, R ꢃ H
11, R ꢃ Me
12, R ꢃ Bn
13, R ꢃ H, 95:5 dr
14, R ꢃ Me, ꢄ99:1 dr
15, R ꢃ Bn, 95:5 dr
Allylic
Diastereoselectivity
Rate constant Relative
[10 4 dm mol sꢀ1
] rate
ꢀ
3
ꢀ1
8
Substrate substituent (X)
(syn:anti)
9
1
H
OH
OMe
OAc
OH (eq)
OH (ax)
n.a.
63.3
33.8
4.3
2.9
47.8
6.7
100
53
7
5
76
11
2
3
91:9
(iii)
A
n.r.
43:57B
96:4
4
NRBn
NRBn
OH
cis-5
trans-6
3
84:16
2
OH
OH
(iv)
OH
OMe
OAc
OH
OH
1
OCOCCl3
1
0
9
, R ꢃ H, 95 %, 95:5 dr
, R ꢃ Me, 67 %, ꢄ99:1 dr 17, R ꢃ Me, ꢄ99:1 dr
16, R ꢃ H, 95:5 dr
2
tBu
tBu
21, R ꢃ Bn, quant, 95:5 dr
18, R ꢃ Bn, 95:5 dr
1
2
3
4
cis-5
trans-6
Scheme 1. Reagents and conditions: (i) Cl
3
CCO H, CH Cl
2
2
2
, room
temperature, 5 min; (ii) meta-chloroperbenzoic acid (m-CPBA), 21 h;
Fig. 1. Observed rate constants for epoxidation of substrates 1–6
A
[8]
(
iii) NaHCO
3
(0.1 M aq.); (iv) K
2
CO , MeOH, room temperature, 16 h.
3
(
n.a. ¼ not applicable; n.r. ¼ not reported). McKittrick and Ganem found
that epoxidation of 3 with meta-chloroperbenzoic acid (m-CPBA) in CH Cl
at 08C gave a 30 : 70 mixture of the corresponding syn- and anti-epoxide
ꢁ
For brevity, for ammonium ions 10–15, the Cl
shown (dr ¼ diastereomeric ratio).
3
CCO
2
counter-ions are not
2
2
B
[3]
diastereoisomers, respectively. Result of Whitham et al.; Henbest and
[2]
Wilson record an anti- to syn-diastereoselectivity of ‘approximately 4 : 1’
Although somewhat less well studied and utilised, the N–H
[
9]
[10]
[10–12]
for this transformation.
group within ureas,
amides,
sulfonamides,
[17–19]
and ammonium ions
carbamates,
[
10,11,13–16]
has proven able
k ꢃ 63.3
ksyn ꢃ 30.8
to direct this epoxidation reaction. We are currently engaged in a
research program aimed at the development and deployment
of ammonium ions (derived from the in situ treatment of the
corresponding allylic and homoallylic amines with a strong
3
1
OH
2
[
20–24]
H
Brønsted acid) as directing groups for this reaction;
this
kanti ꢃ 3.0
methodology has been used as one of the key steps in the
synthesis of a range of biologically significant, natural, and
1
, kobs ꢃ 63.3
ksyn ꢃ 45.9
2
, kobs ꢃ 33.8
[25,26]
[27,28]
non-natural amino
and imino
As part of our investigations to further our
sugars and related
[29]
compounds.
mechanistic understanding of this process, we determined that
kanti ꢃ 1.1
3
1
3
1
treatment of N-substituted cyclohex-2-enamines 7 (R ¼ H), 8
OH
H
(
R ¼ Me) and 9 (R ¼ Bn) with Cl CCO H followed by meta-
3
2
tBu
2
tBu
2
chloroperbenzoic acid (m-CPBA) gave the corresponding 1,2-
anti-2,3-syn-aminodiol trichloroacetate esters 16–18 in $95 : 5
diastereomeric ratio (dr) in all cases, with transesterification
upon treatment with K CO in MeOH giving the 1,2-anti-2,3-
H
OH
kanti ꢃ 1.9
ksyn ꢃ 5.6
cis-5
, kobs ꢃ 47.8
trans-6, kobs ꢃ 6.7
2
3
[
20,23]
syn-aminodiols 19–21.
This is consistent with a mecha-
Fig. 2. Rate constants and dissected rate constants for epoxidation of
ꢁ1 ꢁ1
nism involving directed epoxidation to the proximal face of the
olefin by hydrogen-bonding between the in situ formed ammo-
nium ion 10–12 and the oxidant (analogous to the proposal of
ꢁ
4
3
substrates 1, 2, cis-5, and trans-6 (in units of 10 dm mol s ).
[2]
inductively withdrawing oxy-substituent attenuating the
nucleophilicity of the olefin, and were first to advance that the
high syn-diastereoselectivity of epoxidation of cyclohex-2-enol
Henbest and Wilson),
opening of the intermediate epoxides 13–15 via an S
followed by regioselective ring-
2-type
N
process upon attack of trichloroacetate anion at C(1), distal to
the electron-withdrawing ammonium moiety where its destabi-
lising, inductively electron-withdrawing influence on the tran-
sition state is less pronounced. The observed rate constants for
(
and reagent in the transition state for the reaction (Bartlett’s
2) is due to the formation of a hydrogen bond between substrate
[
4]
butterfly’ model), with the allylic hydroxy group being the
‘
donor and the peroxy-oxygen of the peracid the acceptor,
which now constitutes perhaps the most well known substrate
[2]
ꢁ4
the epoxidation step were kobs (7) ¼ 360 ꢀ 10 , kobs (8) ¼ 37 ꢀ
ꢁ
4
ꢁ4
3
ꢁ1 ꢁ1
10 , and kobs (9) ¼ 7.3 ꢀ 10 dm mol
s (i.e. relative rates
[
5]
directed reaction reported to date. Whitham et al. suggested
that the superior diastereoselectivity (and reactivity) of cis-5
over trans-6 (and 2) was due to the hydroxy group within the
former being held in a favourable orientation to promote the
reaction (rather than being held in an unfavourable orientation
of ,49 : 5 : 1 for 7, 8, and 9, respectively). These data show that
the N-protecting groups have an impact on the ability of the
corresponding ammonium ions to promote the epoxidation
reaction and that the secondary ammonium moiety within 10
(derived from 7) is far superior over its tertiary counterparts
within 11 and 12 (derived from 8 and 9, respectively) in its
[3]
within trans-6 or conformationally more labile within 2),
although the precise transition state geometry (and, indeed, the
identity of the oxygen atom in the peracid which acts as the
[
23]
ability to promote the epoxidation reaction
(Scheme 1).
We have not thus far directly compared the ability of an
ammonium group versus a hydroxy group to direct the
[
2,6,7]
hydrogen-bond acceptor) is an ongoing subject of debate.

6
12
M. B. Brennan et al.
Ammonium-directed
epoxidation
Cl
NRBn
(
rate constant kN)
(i)
(ii)
ꢁ
H
NRBn
OH
NRBn
H ꢁ
H
Cl CCO H
3
2
NRBn
≡
OAc
26, ꢄ95:5 dr
OAc
O
cyclohexa-1,3-diene
27, R ꢃ H
H
H
28, R ꢃ Me
29, R ꢃ Bn
OH
23
Hydroxy-directed
epoxidation
2
2
R ꢃ H, Me, Bn
R ꢃ H, Me, Bn (rate constant k
O
)
(
iv)
m-CPBA
NRBn
kobs (ꢃ kN ꢁ kO)
NRBn
NRBn
NRBn
NRBn
(
iv)
(iii)
ꢁ
O
O
OH
, R ꢃ H, 35 %, 97:3 dr
OBz
OH
3
6
33, R ꢃ H
30, R ꢃ H, 43 %, 95:5 dr
OH
OH
37
, R ꢃ Me, 64 %, ꢄ99:1 dr 34, R ꢃ Me 31, R ꢃ Me, 32 %, 93:7 drA
A
2
4
25
38, R ꢃ Bn, 75 %, ꢄ99:1 dr 35, R ꢃ Bn 32, R ꢃ Bn, 43 %, 99:1 dr
Ratio 24:25 ꢃ k :kO
N
Scheme 2. Reagents and conditions: (i) Pd(OAc)
LiOAcꢂ2H O, AcOH, room temperature, 5 h; (ii) Pd(OAc)
Et N, THF, room temperature, 24 h; (iii) PPh , PhCO H, diethyl azodicar-
2
, p-benzoquinone, LiCl,
, RNHBn, PPh ,
3
Fig. 3. Competitive ammonium-directed and hydroxy-directed epoxida-
tion of N-substituted trans-4-aminocyclohex-2-en-1-ols 22.
2
2
3
3
2
boxylate (DEAD), PhMe, 08C, 1 h, then room temperature, 16 h; (iv) K
A
2
CO
3
,
MeOH, room temperature, 4 h. ,95 % purity.
stereochemical course of this reaction under analogous experi-
mental conditions (i.e. in the presence of a strong Brønsted acid).
A pilot experiment revealed that addition of Cl CCO H to a
3
2
in order to access the requisite substrates 36–38 directly, but was
unsuccessful. However, chloride displacement from 26 under
solution of cyclohex-2-enol (2) in CD Cl produced a very
2
2
1
broad and somewhat complex H NMR spectrum, suggesting
that 2 is unstable in the presence of a strong acid, perhaps due to
ionisation to the corresponding allylic cation being promoted,
and hence precluding analysis of the epoxidation of this sub-
strate under these conditions. However, it was envisaged that
treatment of a range of N-substituted 4-aminocyclohex-2-en-1-
ols 22 with Cl CCO H would result in N-protonation to give the
[32,33]
Tsuji–Trost conditions
methylamine, or N,N-dibenzylamine gave the corresponding
with benzylamine, N-benzyl-N-
[34]
cis-amino acetates 27–29
subsequent transesterification using MeOH and K CO fur-
in $93 : 7 dr in all cases, with
2
3
[35]
nishing the corresponding cis-amino alcohols 30–32,
which
were isolated in 32–43 % overall yield from cyclohexa-1,3-
diene, and in $93 : 7 dr in all cases (Scheme 2). The relative cis-
configuration within 32 was established unambiguously by
3
2
corresponding ammonium species 23 (analogous in structure
to 10–12) which were expected to be stable with respect to
decomposition, and thus provide a unique platform from which
to quantify the relative abilities of the secondary and tertiary
ammonium moieties versus a hydroxy group to promote epoxi-
dation. The substrates 22, and the derived ammonium ions 23,
were expected to favour a half-chair conformation in solution
with both substituents occupying pseudo-equatorial positions
[36]
single crystal X-ray diffraction analysis (Fig. 4), confirming
that the substitution reaction had proceeded with retention of
configuration (i.e. a stereoselective S 1-type process), as
N
expected. Chemoselective N-benzylation of 30 upon treatment
i
with benzylbromide (BnBr), Pr NEt, and 4-dimethylamino-
2
pyridine (DMAP) gave a sample of 32 in 86 % yield, thus
unambiguously establishing the relative cis-configuration
within 30. The relative cis-configuration within 31 was confi-
(i.e. equivalent environments, which are also directly analogous
to those within the expected favoured, half-chair conformations
of the monosubstituted systems 2 and 7–9, and the derived
ammonium ions 10–12). Assuming that the rate of background
(i.e. non-directed) epoxidation is negligible, then formation of
epoxide 24 is the result of hydrogen-bond-directed delivery of
[37–39]
dently assigned by analogy. Finally, a Mitsunobu reaction
of 30–32 using triphenylphosphine, diethyl azodicarboxylate
(DEAD), and benzoic acid gave access to the corresponding
trans-amino alcohols 36–38 in 35–75 % yield and $97 : 3 dr
[35]
in all cases (Scheme 2).
Following our previously reported procedure,
the oxidant by the (in situ formed) ammonium ion (defined by
the rate constant k ), while formation of epoxide 25 is the result
[20,23]
the for-
mation of ammonium species 41 upon addition of Cl CCO H to
N
3
2
of hydrogen-bond-directed delivery by the hydroxy group
(defined by the rate constant k ). The ratio of epoxides 24 to
1
a 0.36 M solution of 38 in CD Cl was monitored by H NMR
2
2
O
spectroscopy: the change in the chemical shift of the vinylic
C(2)H and C(3)H protons with increasing concentration of
Cl CCO H (versus the free amine) was monitored, and reached
2
absence of ring-opened products.
5 therefore corresponds directly to the ratio of k to k in the
N O
[
30]
We delineate herein the
3
2
results of our kinetic studies within this area (Fig. 3).
a plateau around 10 equiv, consistent with 41 being predominant
under these conditions; it was thus concluded that 10 equiv. of
Cl CCO H would be required to efficiently protect the nitrogen
Results and Discussion
3
2
[40]
A common and readily scalable route to a range of N-substituted
trans-4-aminocyclohex-2-en-1-ols was first developed. Palla-
dium catalysed 1,4-acetoxychlorination of cyclohexa-1,3-diene
atom from oxidation within this system.
The stability of
ammonium species 41 (which also contains an allylic hydroxy
functionality) under these strongly acidic conditions is in
contrast to the behaviour of cyclohex-2-enol (2) under identical
conditions but is consistent with initial and rapid N-protonation
of 38 to give 41 disfavouring further ionisation to the
[31]
gave cis-chloro acetate 26 in .95 : 5 dr.
The S 2-type dis-
N
placement of chloride from 26 with benzylamine, N-benzyl-N-
methylamine, or N,N-dibenzylamine was initially investigated

Hydroxy- and Ammonium-Directed Epoxdiation
613
H
ꢁ
NRBn
NRBn
4
3
2
(i)
1
OH
OH
3
6
, R ꢃ H, 97:3 dr
, R ꢃ Me, ꢄ99:1 dr 40, R ꢃ Me
8, R ꢃ Bn, ꢄ99:1 dr 41, R ꢃ Bn
39, R ꢃ H
3
7
3
ꢁ
H
ꢁ
H
NRBn
NRBn
O ꢁ
O
OH
OH
4
2
, R ꢃ H
45, R ꢃ H
46, R ꢃ Me
47, R ꢃ Bn
Fig. 4. X-Ray crystal structure of 32 (selected H atoms are omitted for
43, R ꢃ Me
44, R ꢃ Bn
clarity).
NRBn
NRBn
O
corresponding allylic cation (in fact a dicationic species in this
case). Subsequent reaction of ammonium species 41 under the
conditions used for the epoxidation of ammonium species 10–12
O ꢁ
(i.e. treatment with 1.6 equiv. of m-CPBA in CH Cl at room
temperature for 21 h)
2 2
[
20,23]
OH
OH
followed by basic aqueous work-up
4
8, R ꢃ H
51, R ꢃ H
52, R ꢃ Me
53, R ꢃ Bn
resulted in a mixture comprising ,5 % unreacted starting
material 41, 45 % of epoxides 50 and 53 (,25 : 75 ratio), and
4
5
9
0
, R ꢃ Me
, R ꢃ Bn
5
regio- and stereochemistry) derived from attack of Cl CCO H
0 % of species assigned as ring-opened products (of unknown
Time Conv. Epoxide
min
Yield
% (dr)
Yield
% (dr)
–
3
2
A
R
H
%
ratio
on the epoxide ammonium intermediates 44 and 47 (character-
1
istic signals were observed at ,5 ppm in the H NMR spectrum
B
C
36
30
96 ꢄ95:5 50, 64 (ꢄ99:1)
C
3
7
Me 180
38 Bn 180
91
88
82:18 49, 50 (ꢄ99:1) 52
,
6 (ꢄ99:1)
of the crude reaction mixture, due to the CH protons adjacent
to the trichloroacetate ester functionality). This result can be
compared with the behaviour of the parent system 12 (derived
from 9) under the same conditions, where complete consump-
tion of starting material occurs and epoxide ammonium inter-
mediate 15 does not accumulate to a significant amount at any
25:75 50, 19 (95:5)
53, 44 (ꢄ99:1)
Scheme 3. Reagents and conditions: (i) Cl
3
CCO
2
H (10 equiv.), meta-
2
, room temperature.
chloroperbenzoic acid (m-CPBA) (5 equiv.), CH
A
2
Cl
B
Ratio of 48 and 51, 49 and 52, or 50 and 53, respectively. Epoxide 48
was isolated as the corresponding N,N-dibenzyl substituted epoxide 50 after
i
treatment of the crude reaction mixture with BnBr/ Pr NEt. A 47 : 53
2
C
[20,23]
point during the reaction due to rapid in situ ring-opening.
mixture of 49 and 52 was also isolated in 7 % combined yield. For brevity,
ꢁ
The differing behaviour suggests both a slower rate of epoxida-
tion and a slower rate of ring-opening of the intermediate
epoxide(s) in the case of 41 as compared with 12. Both of these
observations may be attributable to the presence of a second,
electron-withdrawing allylic substituent in 41: this would have
the effects of first abating the nucleophilicity of the olefin (hence
diminishing the rate of epoxidation) and second retarding the
rate of epoxide ring-opening due to its destabilising (inductively
for ammonium ions 39–47, the Cl CCO
3
2
counter-ions are not shown.
substrate 38 and the diastereoisomeric epoxide 53. The epoxi-
dation of secondary N-benzyl substituted amine 36 under the
same conditions was essentially complete (96 % conversion)
after only 30 min, giving a single epoxide 48 (.95 : 5 dr).
Alternatively, epoxidation of 36 using 10 equiv. of Cl CCO H
and 1.6 equiv. of m-CPBA to give 48 (.95 : 5 dr) was complete
within 3.5 h. To facilitate purification, treatment of the crude
i
reaction mixture with BnBr and Pr NEt gave epoxide 50 in
.95 : 5 dr, which was isolated in 64 % yield (from 36) and
.99 : 1 dr, thus also unambiguously establishing the relative
configuration within 48. Epoxidation of tertiary N-benzyl-N-
methyl substituted amine 37 gave 91 % conversion after 3 h into
an 82 : 18 mixture of epoxides 49 and 52, from which the major
diastereoisomer 49 was isolated in 50 % yield and .99 : 1 dr and
the minor diastereoisomer 52 in 6 % yield and .99 : 1 dr,
3
2
[41,42]
electron-withdrawing) influence on the transition state.
Electrostatic repulsions between the incoming nucleophile and
the proximal allylic heteroatom may also be manifest in a
reduction in the rate of epoxide ring-opening. An optimal
procedure for the isolation of epoxides 50 and 53 was developed:
treatment of 38 with 10 equiv. of Cl CCO H followed by
2
3
2
5
2
equiv. of m-CPBA for 3 h gave 88 % conversion into a
5 : 75 mixture of epoxides 50 and 53, with no trace of ring-
opened products being observed. Purification gave 50 in 19 %
yield and 95 : 5 dr and 53 in 44 % yield and .99 : 1 dr
(
tion, was unambiguously established by single crystal X-ray
Scheme 3). The identity of 53, including its relative configura-
alongside a 47 : 53 mixture of 49 and 52 in 7 % combined yield.
i
Treatment of a crude sample of 48 with MeI and Pr NEt gave an
authentic sample of 49 in 55 % isolated yield (from 36), thus
establishing unambiguously the relative configuration within 49
and hence also allowing assignment of the relative configuration
within the diastereoisomer 52 (Scheme 3).
2
[
36]
diffraction analysis
(Fig. 5). The gross structure of 50 was
established by NMR spectroscopic analyses (including COSY
and HMBC), and its relative configuration was thus confidently
assigned from the known relative configurations within both the

6
14
M. B. Brennan et al.
products in the initial period of this reaction means that the
initially observed epoxidation diastereoselectivity directly cor-
responds to the ratio of the rate constants describing the rate of
reaction on each face of the olefin (i.e. the rate of ammonium-
directed epoxidation versus the rate of hydroxy-directed epoxi-
dation). Thus, the 25 : 75 ratio of epoxides 50 and 53 gives the
ꢁ
4
3
ꢁ1 ꢁ1
3
dissected rate constants k ¼ 0.7 ꢀ 10 dm mol
s
for the
ꢁ1 ꢁ1
N
ꢁ
4
ammonium-directed processand k ¼ 2.0 ꢀ 10 dm mol
s
O
for the hydroxy-directed process (Fig. 6).
In a directly analogous manner, addition of 10 equiv. of
Cl CCO H to a 0.36 M solution of 36 in CD Cl at 298 K gave
2
3
2
2
a solution of ammonium ion 39, with characteristic olefinic
signals at 5.97–6.07 and 6.22–6.32 ppm. As before, upon addi-
tion of 1.6 equiv. of m-CPBA to this sample, loss of these signals
(as well as those associated with m-CPBA at 7.88–7.91 and
Fig. 5. X-Ray crystal structure of 53 (selected H atoms are omitted for
clarity).
7
.95–7.97 ppm) was accompanied by the appearance of new
signals arising at 3.60–3.68 ppm, corresponding to the epoxide
ammonium salt 42 as a single diastereoisomer, the identity of
which was confirmed upon independent addition of 10 equiv. of
Cl CCO H to a 0.36 M solution of epoxide 48 in CD Cl . After
It is apparent from the diastereoselectivities elicited in the
epoxidations of 36 (.95 : 5 dr), 37 (82 : 18 dr), and 38 (25 : 75
dr) that the relative directing ability of the various allylic
substituents falls in the order: NHBn .. NMeBn . OH .
3
2
2
2
,1320 s (22 min), the intensity of these signals reached a
plateau and began to decrease, with concomitant appearance
and increase in intensity of signals at 5.46–5.52 ppm; these
signals were assigned to ring-opened products 55 of unknown
regio- and stereochemistry. A mass deficit was noted in the data
after ,7500 s (as with the epoxidation of 38 after ,10000 s).
Application of the integrated form of the second order rate law to
NBn . In an effort to quantify the relative abilities of these
2
groups to promote this epoxidation reaction, the rate constants
for the epoxidations of these substrates were determined, fol-
lowing our previously reported procedure by which we have
monitored analogous epoxidation/in situ ring-opening reactions
[
20,23]
(e.g. those of 7–9).
to a 0.36 M solution of 38 in CD Cl at 298 K gave a solution
Addition of 10 equiv. of Cl CCO H
3 2
1
the data generated from these H NMR spectroscopic studies
(decay of the starting material 36 and appearance of total prod-
2
2
of ammonium ion 41, with characteristic olefinic signals at
.01–6.03 and 6.35–6.37 ppm. Upon addition of 1.6 equiv. of
4
3
ꢁ1 ꢁ1
6
uct 42 and 55) gave kobs ¼ (37.5 ꢃ 1.0) ꢀ 10ꢁ dm mol
s .
Given that formation of only a single epoxide 42 (.95 : 5 dr) is
observed in this reaction (see above), it follows that k . 95k
m-CPBA to this sample, loss of these signals (as well as those
associated with m-CPBA at 7.88–7.91 and 7.95–7.97 ppm) was
accompanied by the appearance of new signals rising at 3.52–
N
O
ꢁ1 ꢁ1
s
ꢁ4
ꢁ4
3
and, hence, k . 36 ꢀ 10 and k , 1.5 ꢀ 10 dm mol
N
O
3
tively). Independent addition of 10 equiv. of Cl CCO H to
.59 and 3.59–3.64 ppm (in an approximate 75 : 25 ratio, respec-
(Fig. 7).
In the case of 37 (0.36 M solution in CD
Cl
at 298 K) the
3
2
2
2
0
5
.36 M solutions of authentic samples of the isolated epoxides
0 and 53 in CD Cl confirmed that these new signals were due
situation was somewhat more complex as addition of 10 equiv.
of Cl CCO H produced ammonium 40 as an approximate
50 : 50 mixture of epimers at the nitrogen atom, as observed
2
2
3
2
to the formation of the corresponding epoxide ammonium
intermediates 44 and 47 (in a 25 : 75 ratio, respectively). The
sum of both diastereoisomers was integrated within the range
1
by H NMR spectroscopic analysis. Upon addition of m-CPBA
to this sample, the consumption of 40 was monitored by the
reduction in intensity of the characteristic signals at 6.30–
6.44 ppm, which incorporated both of the epimeric ammonium
ions. Independent addition of 10 equiv. of Cl CCO H to 0.36 M
3
.52–3.64 ppm. After ,12000 s (3 h 20 min), the intensity of
these signals reached a plateau (and later began to decrease), but
with the concomitant appearance and subsequent increase in
intensity of signals at 4.96–5.13 and 5.40–5.47 ppm, consistent
with the formation of epoxide ring-opened products 54 (of
unknown regio- and stereochemistry). The decay of the starting
material 38 and the appearance of total product (the sum of 44,
3
2
solutions of authentic samples of epoxides 49 and 52 in CD Cl
2 2
gave an authentic sample of the corresponding epoxide ammo-
nium intermediates 43 and 46, respectively, as approximate
50 : 50 mixtures of epimers at the nitrogen atom in both cases.
The formation of these species was observed in the epoxidation
reaction of 37 although due to the simultaneous presence of
signals associated with the N-epimers of the starting material
ammonium 40 and formation of ring-opened products 56 (regio-
and stereochemistry unknown) indicated by the appearance of
characteristic signals at ,5 ppm, it proved impossible to accu-
rately quantify the amounts of the epoxide ammonium inter-
mediates 43 and 46 and ring-opened products 56 due to
significant signal overlap. Nonetheless, application of the inte-
grated form of the second order rate law to the data generated
4
7, and 54) can be analysed separately, to obtain two indepen-
dent values for the rate constant, kobs. In the latter case, the
accuracy of this analysis is crucially dependent upon there being
negligible mass deficit in the data: in the closed system, at all
stages of the reaction, the total concentration of all species
present must equal the starting concentration of the substrate 38
(i.e. 0.36 M in this case). The total concentration of all species
remains ,0.36 M before ,10000 s; after this point the devia-
tion (mass deficit) is attributed to the competitive formation
of the corresponding diols (which was not monitored), resulting
from either ring-opening of 44 or 47 by adventitious water, or
in situ hydrolysis of the trichloroacetate esters 54. Analysis
of these data by application of the integrated form of the
second order rate law gave two independent values for the rate
ꢁ
4
from monitoring the loss of 40 gave kobs ¼ (3.1 ꢃ 0.2) ꢀ 10
.
From the 82 : 18 mixture of diastereoisomeric epoxide products
49 and 52 formed in this reaction (see above), it follows that
4
ꢁ4
k
¼ 2.5 ꢀ 10ꢁ and k
¼ 0.6 ꢀ 10 (Fig. 8).
N
O
constant, with very good agreement: thus, k ¼ (2.7 ꢃ 0.2) ꢀ
obs
Comparison of the rate constants determined from these
kinetic studies on 36–38 with those for 7–9 determined under
ꢁ
4
3
dm mol s . The absence of formation of ring-opened
ꢁ1 ꢁ1
1
0

Hydroxy- and Ammonium-Directed Epoxdiation
615
ꢁ
ꢁ
ꢁ
ꢁ
H
H
H
H
NBn2
NBn2
NBn2
NBn2
NBn2
Cl CCO H
Cl CCO H
OR
OR
3
2
m-CPBA
3
2
ꢁ
O
O
kobs
OH
OH
OH
OH
OR
5
4
, R ꢃ H or COCCl3
3
8
41
44
47
4
4:47 ꢃ 25:75
0
0
0
0
0
0
0
0
.40
.35
.30
.25
.20
.15
.10
.05
0
0
5000
10000
15000
20000
25000
Time [s]
1
Fig. 6. Real-time ( H NMR data points) concentration profiles for Cl
promoted epoxidation of 38 at 298 K in CD Cl
3
CCO H (10 equiv.) and meta-chloroperbenzoic acid (1.6 equiv.)
2
2
2
. [41], blue; [44þ47], orange; [54], green; total [41þ44þ47þ54], black (the total
concentration of all species in the closed system must remain constant throughout, indicated by the hollow points). For brevity, for
ꢁ
ammonium ions 41, 44, 47, and 54, the Cl
3
CCO
2
counter-ions are not shown.
identical experimental conditions (Table 1) reveals, in all cases
examined, that the absolute value of the rate constant is lower
upon incorporation of a hydroxy group into the substrate (for
conformation-controlling ability of a tert-butyl group and so
enforce a conformation in which the hydroxy group is efficiently
locked in a pseudo-equatorial position within the favoured half-
chair conformation of 38, which is thus amenable to more
efficient hydrogen-bond direction from the latter.
example, compare k ¼ 7.3 ꢀ 10ꢁ for 9 versus kobs ¼ 2.5 ꢀ
4
obs
ꢁ
4
1
of the olefin and in accordance with the observations of Henbest
0
for 38), consistent with a decrease in the nucleophilicity
[
and Wilson, and Whitham et al. The variation in the
2]
[3]
Conclusion
dissected rate constants (k ) for the ammonium-directed process
N
occurring within 36–38 (Table 1) reveals that the inclusion of
the hydroxy substituent on the anti face retards the rate of the
ammonium-directed epoxidation by approximately the same
degree in each case: the relative rate order for the epoxidations
of 7, 8, and 9 is 49 : 5 : 1 while the relative rate order for the
epoxidations of 36, 37, and 38 is ,51 : 4 : 1 but the absolute
values of the latter are an approximate order of magnitude
less than those of the former (i.e. the epoxidation proceeds
The epoxidations of a range of N-substituted trans-4-amino-
cyclohex-2-en-1-ols upon treatment with Cl CCO H and m-
3
2
CPBA have been investigated. These substrates represent a
unique platform for probing the effect of the two competing
directing groups on both the rate of the epoxidation reaction and
the reaction diastereoselectivity. Competing direction by
hydrogen-bonding from both the hydroxy group and the in situ
formed ammonium ion was analysed from the diastereoisomeric
epoxide mixtures obtained from these reactions, allowing
the following order of directing group ability to be defined:
,10 times slower in the latter cases). Interestingly, the variance
in the dissected rate constants (k ) for hydroxy-directed epoxi-
O
dation occurring within 36–38 (Table 1) shows that the nature
of the ammonium-substituent and not merely its presence
affects the ability of the hydroxy group to act as a hydrogen-
bond donor: the rate of the hydroxy-directed process is greater
in the presence of the more sterically demanding N,N-
dibenzylammonium substituent than the less sterically demand-
ing N-benzyl-N-methylammonium and N-benzylammonium
substituents. In the former case, the N,N-dibenzyl-
ammonium substituent may be able to effectively mimic the
NHBn .. NMeBn . OH . NBn . The presence of two elec-
2
tron-withdrawing heteroatoms in the allylic positions results in a
relatively slow rate of in situ epoxide ring-opening (compared
with the rate of the epoxidation reaction), which not only
allowed for isolation of the intermediate epoxides but also
means that the observed epoxidation diastereoselectivity
directly corresponds to the ratio of the rate constants describing
the two competing hydrogen-bond directed processes. The
second-order rate constants for these reactions were thus

6
16
M. B. Brennan et al.
ꢁ
ꢁ
ꢁ
H
H
H
NHBn
OH
NHBn
NHBn
NHBn
Cl CCO H
Cl CCO H
3
2
m-CPBA
3
2
OR
OR
kobs
O
OH
OH
OH
3
6
39
42
55, R ꢃ H or OCOCCl3
0
0
0
0
0
0
0
0
.40
.35
.30
.25
.20
.15
.10
.05
0
0
5000
10000
15000
20000
25000
Time [s]
1
Fig. 7. Real-time ( H NMR data points) concentration profiles for Cl
1.6 equiv.) promoted epoxidation of 36 at 298 K in CD Cl
3 2
CCO H (10 equiv.) and meta-chloroperbenzoic acid
(
2
2
. [39], blue; [42], orange; [55], green; total [39þ42þ55], black (the total
concentration of all species in the closed system must remain constant throughout, indicated by the hollow points). For brevity, for
ꢁ
ammonium ions 39, 42 and 55, the Cl
3
CCO
2
counter-ions are not shown).
ꢁ
ꢁ
ꢁ
ꢁ
H
H
H
H
NMeBn
NMeBn
NMeBn
NMeBn
NMeBn
Cl CCO H
m-CPBA
Cl
3
CCO
2
H
OR
3
2
ꢁ
O
O
kobs
OR
OH
OH
OH
OH
OH
3
7
40
43
46
56, R ꢃ H or OCOCCl3
0
.40
.35
.30
.25
.20
.15
.10
.05
0
0
0
0
0
0
0
0
0
5000
10000
15000
20000
25000
Time [s]
1
Fig. 8. Real-time ( H NMR data points) concentration profiles for Cl CCO
3
2
H (10 equiv.) and meta-chloroperbenzoic acid (1.6 equiv.)
ꢁ
promoted epoxidation of 37 at 298 K in CD
Cl
2 2
. [40], blue (for brevity, for ammonium ions 40, 43, 46 and 56, the Cl
3
CCO
2
counter-ions
are not shown).

Hydroxy- and Ammonium-Directed Epoxdiation
617
Table 1. Rates of epoxidation of allylic cyclohexenylamines 7–9 and
(450 mL) at room temperature over 3 h. The resultant mixture
was then stirred at room temperature for a further 2 h. Saturated
aqueous NaCl (200 mL) was then added and the resultant mix-
ture was stirred at room temperature for 30 min, and then
3
6–38 (Cl
3
CCO
2
H, meta-chloroperbenzoic acid, CH
2
Cl
and k
O
2
, 298 K)
k
obs ¼ observed rate constant, dr ¼ diastereomeric ratio, k
N
are the
rates of hydrogen-bond-directed delivery of the oxidant by the (in situ
formed) ammonium ion and the hydroxy group, respectively
extracted with 30–408C petrol/Et O (v/v, 9 : 1, 3 ꢀ 400 mL). The
2
combined organic extracts were washed sequentially with H O
2
NRBn
NRBn
(
2 ꢀ 500 mL), saturated aqueous Na CO (500 mL), 2 M aque-
2
3
ous NaOH (2 ꢀ 500 mL), and saturated aqueous NaCl (500 mL),
and then dried and concentrated under vacuum (at room tem-
[
31]
perature) to give 26 as a yellow oil (18.7 g, .95 : 5 dr);
(400 MHz, CDCl ) 1.95–2.00 (2H, m, C(5)H , C(6)H ), 2.08
dH
OH
3
A
A
7
, R ꢃ H
, R ꢃ Me
36, R ꢃ H
37, R ꢃ Me
38, R ꢃ Bn
(3H, s, COMe), 2.09–2.15 (2H, m, C(5)H , C(6)H ), 4.53–4.58
B B
(1H, m, C(4)H), 5.26–5.31 (1H, m, C(1)H), 5.79–5.80 (1H, m,
C(2)H), 5.96–5.99 (1H, m, C(3)H).
8
9, R ꢃ Bn
R
Compd
k
obs
dr
Compd
k
obs
dr
k
N
k
O
(
RS,SR)-4-(N-Benzylamino)cyclohex-2-en-1-ol (30)
H
7
8
9
360
37
.95 : 5
.95 : 5
36
37
38
37.5 .95 : 5 .36
,1.5
0.6
Pd(OAc) (321 mg, 1.43 mmol) was added to a solution of PPh3
2
Me
Bn
3.1
2.5
82 : 18
25 : 75
2.5
0.7
(563 mg, 2.18 mmol), benzylamine (1.88 mL, 17.2 mmol), and
NEt (6.0 mL, 43.0 mmol) in THF (60 mL) and the resultant
7.3 .95 : 5
1.8
3
mixture was stirred at room temperature for 30 min. A solution
of 26 (2.5 g, 14.3 mmol, .95 : 5 dr) in THF (20 mL) was then
added, and the reaction mixture was stirred at room temperature
measured and dissected into individual rate constants to describe
the rates of both the ammonium-directed and the hydroxy-
directed processes. From these data, it is apparent that the rate of
an ammonium-directed epoxidation is retarded by the incorpo-
ration of an allylic hydroxy group (on the anti face) into the
substrate by an approximate order of magnitude. The hydroxy-
directed reaction proceeds more rapidly in the presence of a
more sterically demanding ammonium moiety than a less ste-
rically demanding one, which is attributed to increased con-
formational bias for the half-chair conformer in which both
substituents occupy pseudo-equatorial positions, allowing the
hydroxy group to act as a more efficient hydrogen-bond donor.
for 24 h. H O (100 mL) was then added and the resultant mixture
2
was extracted with Et O (3 ꢀ 150 mL). The combined organic
2
extracts were washed with saturated aqueous NaCl (300 mL),
dried, and then concentrated under vacuum. The residue was
dissolved in MeOH (30 mL), K CO (4.95 g, 35.8 mmol) was
2
3
added, and the resultant suspension was stirred at room tem-
perature for 4 h, filtered, and then concentrated under vacuum.
The residue was dissolved in CHCl (100 mL), washed with
3
H O (3 ꢀ 100 mL), dried, and then concentrated under vacuum.
2
Purification via flash column chromatography (eluent 30–408C
petrol/EtOAc, 1 : 1, increased to EtOAc) gave 30 as a yellow
[35]
oil (1.25 g, 43 % from 1,3-cyclohexadiene, 95 : 5 dr). dH
400 MHz, CDCl ) 1.62–1.86 (6H, m, C(5)H , C(6)H , OH,
(
NH), 3.14–3.17 (1H, m, C(4)H), 3.82–3.90 (2H, m, NCH Ph),
3 2 2
Experimental
2
General Experimental Details
4.12–4.15 (1H, m, C(1)H), 5.83–5.90 (2H, m, C(2)H, C(3)H),
7
.24–7.34 (5H, m, Ph).
m-CPBA was supplied as a 70–77 % slurry in water and titrated
[
43]
according to the procedure of Swern immediately before use.
Water was purified by an Elix UV-10 system. Organic solvents
were used as supplied (analytical or HPLC grade) without prior
purification. Organic layers were dried over Na SO . Thin-layer
(RS,SR)-4-(N-Benzyl-N-methylamino)cyclohex-
2-en-1-ol (31)
2
4
Pd(OAc) (321 mg, 1.43 mmol) was added to a solution of
2
chromatography was performed on aluminium plates coated
with 60 F254 silica. Plates were visualised using UV light
PPh (563 mg, 2.18 mmol), N-benzyl-N-methylamine (2.22 mL,
3
1
7.2 mmol), and NEt (6.0 mL, 43.0 mmol) in THF (60 mL) and
3
(254 nm), iodine, 1 % aqueous KMnO , or 10 % ethanolic
4
the resultant mixture was stirred at room temperature for 30 min.
A solution of 26 (2.5 g, 14.3 mmol, .95 : 5 dr) in THF (20 mL)
was then added, and the reaction mixture was stirred at room
phosphomolybdic acid. Flash column chromatography was
performed either on Kieselgel 60 silica on a glass column, or on
an automated flash column chromatography platform.
temperature for 24 h. H O (100 mL) was then added and the
2
Melting points are uncorrected. IR spectra were recorded
as a thin film on NaCl plates. Selected characteristic peaks are
resultant mixture was extracted with Et O (3 ꢀ 150 mL). The
2
combined organic extracts were washed with saturated aqueous
NaCl (300 mL), dried, and then concentrated under vacuum.
The residue was dissolved in MeOH (30 mL), K CO (4.95 g,
ꢁ1
reported in cm . NMR spectra were recorded in the deuterated
solvent stated. The field was locked by external referencing to
2
3
1
1
1
13
the relevant deuteron resonance. H– H COSY and H– C
HMQC analyses were used to establish atom connectivity.
Accurate mass measurements were run on a MicroTOF instru-
ment internally calibrated with polyalanine.
3
at room temperature for 4 h, filtered, and then concentrated
under vacuum. The residue was dissolved in CHCl (100 mL)
3
5.8 mmol) was added, and the resultant suspension was stirred
and the resultant solution was washed with H O (3 ꢀ 100 mL),
2
dried, and then concentrated under vacuum. Purification via
flash column chromatography (eluent 30–408C petrol/Et O,
2 : 8) gave 31 as a pale yellow oil (1.01 g, 32 % from 1,3-
(
RS,SR)-1-Acetoxy-4-chlorocyclohex-2-ene (26)
,3-Cyclohexadiene (11.9 mL, 125 mmol) was added via
syringe pump to a stirred solution of Pd(OAc)2 (1.40 g,
2
1
ꢁ1
cyclohexadiene, 93 : 7 dr, ,95 % purity). nmax/cm
3356,
3026, 2939. d (400 MHz, CDCl ) 1.66–1.93 (4H, m, C(5)H ,
6
2
.40 mmol), LiCl (10.5 g, 244 mmol), LiOAcꢂ2H O (25.5 g,
2
H
3
2
44 mmol), and p-benzoquinone (28.4 g, 260 mmol) in AcOH
C(6)H ), 2.25 (3H, s, NMe), 3.24–3.28 (1H, m, C(4)H), 3.53
2

6
18
M. B. Brennan et al.
3
˚
˚
(
4
7
(
1
1
1H, d, J 13.1, NCH H Ph), 3.68 (1H, d, J 13.1, NCH H Ph),
A
0.05 ꢀ 0.10 ꢀ 0.20 mm , a 6.1763(2) A, b 17.5995(6) A, c
B
A B
˚
.14–4.15 (1H, m, C(1)H), 5.94–5.95 (2H, m, C(2)H, C(3)H),
.23–7.36 (5H, m, Ph). d (100 MHz, CDCl ) 17.1 (C(5)), 30.1
17.9947(6) A, a 106.7590(14)8, b 93.9129(15)8, g 94.0362(14)8,
3
ꢁ1
˚
V 1860.25(11) A , Z 4, m 0.138 mm . A total of 14065 reflec-
tions, and 8243 unique reflections, were measured for 5 , y ,27
and 8243 reflections were used in the refinement. The final
parameters were GooF 0.9427, wR2 0.225 and R1 0.111
[I . ꢁ3.0s(I)].
C
3
C(6)), 37.9 (NMe), 57.6 (NCH Ph), 59.3 (C(4)), 63.7 (C(1)),
2
26.8 (p-Ph), 128.2, 128.7 (o,m-Ph), 130.7, 134.9 (C(2), C(3)),
þ
þ
39.8 (i-Ph). m/z (ESI ) 218 ([M þ H] , 100 %). m/z (HRMS
þ
þ
þ
ESI ) 218.1546; C H NO ([M þ H] ) requires 218.1539.
1
4 20
(
RS,SR)-4-(N,N-Dibenzylamino)cyclohex-2-en-1-ol (32)
(
RS,RS)-4-(N-Benzylamino)cyclohex-2-en-l-ol (36)
Method A
DEAD (40 wt-% solution in PhMe, 1.94 mL, 4.25 mmol) was
added dropwise via syringe pump over 1 h to a stirred solution of
30 (508 mg, 2.50 mmol, 95 : 5 dr), PPh (1.31 g, 5.0 mmol), and
Pd(OAc) (321 mg, 1.43 mmol) was added to a solution of
2
PPh3 (563 mg, 2.18 mmol), dibenzylamine (3.30 mL,
7.2 mmol), and NEt (6.0 mL, 43.0 mmol) in THF (60 mL)
3
1
PhCO H (452 mg, 3.75 mmol) in PhMe (10 mL) at 08C. The
2
3
and the resultant mixture was stirred at room temperature for
0 min. A solution of 26 (2.5 g, 14.3 mmol, .95 : 5 dr) in THF
20 mL) was then added, and the reaction mixture was stirred at
resultant mixture was allowed to warm to room temperature
over 16 h and then concentrated under vacuum. The residue was
dissolved in Et O (25 mL) and the resultant solution was washed
3
(
2
room temperature for 24 h. H O (100 mL) was then added and
2
sequentially with saturated aqueous Na CO (3 ꢀ 25 mL) and
2
3
the resultant mixture was extracted with Et O (3 ꢀ 150 mL). The
saturated aqueous NaCl (25 mL), dried, and then concentrated
under vacuum. The residue was dissolved in MeOH (10 mL) and
K CO (864 g, 12.5 mmol) was added. The resultant suspension
2
combined organic extracts were washed with saturated aqueous
NaCl (300 mL), dried, and then concentrated under vacuum.
The residue was dissolved in MeOH (30 mL), K CO (4.95 g,
2
3
was stirred at room temperature for 4 h, filtered, and then con-
centrated under vacuum. The residue was dissolved in Et O
(25 mL) and the resultant solution was washed sequentially with
2
3
35.8 mmol) was added, and the resultant suspension was stirred
at room temperature for 4 h, filtered, and then concentrated
2
under vacuum. The residue was dissolved in CHCl (100 mL)
3
H O (3 ꢀ 25 mL) and saturated aqueous NaCl (25 mL), dried,
2
and the resultant solution was washed with H O (3 ꢀ 100 mL),
and then concentrated under vacuum. Purification via flash
column chromatography (eluent 30–408C petrol/EtOAc, 1 : 1,
increased to EtOAc) gave 36 as an orange oil (178 mg, 35 %,
2
dried, and then concentrated under vacuum. Purification via
flash column chromatography (eluent 30–408C petrol/Et O,
2
[
35]
1
: 1) gave 32 as a white solid (1.82 g, 43 % from 1,3-cyclo-
97 : 3 dr). d (400 MHz, CDCl ) 1.38–1.52 (2H, m, C(6)H ),
H 3 2
ꢁ1
hexadiene, 99 : 1 dr); mp 49–508C. nmax/cm 3385, 2936. dH
1.68–1.97 (2H, m, OH, NH), 2.09–2.16 (2H, m, C(5)H ), 3.26–
2
(
400 MHz, CDCl ) 1.43–1.90 (5H, m, C(5)H , C(6)H , OH),
3.29 (1H, m, C(4)H), 3.81–3.88 (2H, m, NCH Ph), 4.25–4.27
2
3
2
2
3
3
5
.23–3.28 (1H, m, C(4)H), 3.58 (2H, d, J 14.0, N(CH H Ph) ),
A
(1H, m, C(1)H), 5.76–5.87 (2H, m, C(2)H, C(3)H), 7.23–7.34
(5H, m, Ph).
B
2
.76 (2H, d, J 14.0, N(CH H Ph) ), 4.12 (1H, br s, C(1)H),
A
B
2
.91–6.00 (2H, m, C(2)H, C(3)H), 7.21–7.43 (10H, m, Ph). dC
(100 MHz, CDCl ) 17.3 (C(5)), 30.1 (C(6)), 54.0 (N(CH Ph) ),
3 2 2
(RS,RS)-4-(N-Benzyl-N-methylamino)cyclohex-
5
1
4.8 (C(4)), 63.6 (C(1)), 126.8 (p-Ph), 128.2, 128.5 (o,m-Ph),
þ
2
-en-1-ol (37)
30.7 (C(2)), 135.7 (C(3)), 140.3 (i-Ph). m/z (ESI ) 294 ([M þ
DEAD (40 wt-% solution in PhMe, 1.94 mL, 4.25 mmol) was
added dropwise via syringe pump over 1 h to a stirred solution of
31 (543 mg, 2.50 mmol, 93 : 7 dr), PPh (1.31 g, 5.0 mmol), and
3
PhCO H (452 mg, 3.75 mmol) in PhMe (10 mL) at 08C. The
resultant mixture was allowed to warm to room temperature
over 16 h and then concentrated under vacuum. The residue was
dissolved in Et O (25 mL) and the resultant solution was washed
þ
þ
þ
þ
H] , 100 %). m/z (HRMS ESI ) 294.2854; C H NO ([M þ
2
0 24
H] ) requires 294.1852.
2
Method B
i
BnBr (88 mL, 0.74 mmol), Pr NEt (0.13 mL, 0.74 mmol),
2
and DMAP (3.0 mg, 0.025 mmol) were added sequentially to a
stirred solution of 30 (100 mg, 0.49 mmol, 95 : 5 dr) in CH Cl
2
2
2
sequentially with saturated aqueous Na CO (3 ꢀ 25 mL) and
2
3
(1.4 mL) at room temperature and the resultant mixture was
saturated aqueous NaCl (25 mL), dried, and then concentrated
under vacuum. The residue was dissolved in MeOH (10 mL) and
K CO (864 g, 12.5 mmol) was added. The resultant suspension
2
stirred at room temperature for 24 h, diluted with CH Cl2
(
5 mL), and washed with H O (2 ꢀ 5 mL). The combined
2
2
3
aqueous washings were extracted with CH Cl (10 mL) and
2 2
was stirred at room temperature for 4 h, filtered, and then con-
centrated under vacuum. The residue was dissolved in Et O
the combined organic extracts were washed with saturated
aqueous NaCl (20 mL), dried, and then concentrated under
vacuum. Purification via flash column chromatography (eluent
2
(25 mL) and the resultant solution was washed sequentially with
H O (3 ꢀ 25 mL) and saturated aqueous NaCl (25 mL), dried,
2
3
8
0–408C petrol/Et O, 1 : 1) gave 32 as a colourless oil (124 mg,
2
and then concentrated under vacuum. Purification via flash
column chromatography (eluent 30–408C petrol/Et O, 1 : 4)
6 %, .99 : 1 dr).
2
gave 37 as a pale yellow oil (348 mg, 64 %, .99 : 1 dr, ,95 %
[
36]
ꢁ1
X-Ray Crystal Structure Determination for 32
purity). nmax/cm 3346, 3026, 2862. dH (400 MHz, CDCl )
3
Data were collected using a Nonius k-CCD diffractometer with
graphite monochromated MoKa radiation using standard proce-
dures at 150 K. The structure was solved by direct methods
1.39–1.48 (1H, m, C(6)H
1.89–1.95 (1H, m, C(5)H
lapping 2.21 (3H, s, NMe), 3.39–3.43 (1H, m, C(4)H), 3.48 (1H,
d, J 13.3, NCH Ph), 3.62 (1H, d, J 13.3, NCH Ph), 4.28
(1H, ddd, J 9.4, 5.6, 3.4, C(1)H), 5.83 (2H, m, C(2)H, C(3)H),
7.23–7.33 (5H, m, Ph). d (100 MHz, CDCl ) 21.4 (C(5)), 32.3
(C(6)), 37.6 (NMe), 57.3 (NCH Ph), 59.3 (C(4)), 67.5 (C(1)),
126.8 (p-Ph), 128.2, 128.7 (o,m-Ph), 132.3, 133.7 (C(2), C(3)),
A
), 1.56–1.66 (2H, m, C(5)H
), 2.16–2.23 (1H, m, C(6)H
A
, OH),
) over-
B
B
(SIR92); all non-hydrogen atoms were refined with anisotropic
thermal parameters. Hydrogen atoms were added at idealised
A
H
B
H
A B
[44]
positions. The structure was refined using CRYSTALS. X-Ray
crystal structure data for 32 [C20.25H23.5Cl10.5NO]: M 629.63,
triclinic, space group P–1, colourless plate, crystal dimensions
C
3
2

Hydroxy- and Ammonium-Directed Epoxdiation
619
þ
þ
þ
1
39.8 (i-Ph). m/z (ESI ) 218 ([M þ H] , 100 %). m/z (HRMS
(1RS,2SR,3RS,4RS)- and (1RS,2RS,3SR,4RS)-2,3-Epoxy-4-
(N-benzyl-N-methylamino)cyclohexan-1-ol (49 and 52)
þ
þ
ESI ) 218.1535; C H NO ([M þ H] ) requires 218.1539.
1
4 20
Method A: Epoxidation of 36
(
RS,RS)-4-(N,N-Dibenzylamino)cyclohex-2-en-1-ol (38)
Step 1. Cl CCO H (804 mg, 4.92 mmol) was added to a
3
2
DEAD (40 wt-% solution in PhMe, 1.94 mL, 4.25 mmol) was
added dropwise via syringe pump over 1 h to a stirred solution of
stirred solution of 36 (100 mg, 0.49 mmol, 97 : 3 dr) in CH Cl
2 2
(1.4 mL, 0.36 M with respect to 36) at room temperature and
the resultant mixture was stirred at room temperature for 5 min.
m-CPBA (60 % by wt, 707 mg, 2.46 mmol) was subsequently
added and the resultant mixture was stirred at room temperature
3
PhCO H (452 mg, 3.75 mmol) in PhMe (10 mL) at 08C. The
2 (734 mg, 2.50 mmol, 99 : 1 dr), PPh (1.31 g, 5.0 mmol), and
3
2
resultant mixture was allowed to warm to room temperature
over 16 h and then concentrated under vacuum. The residue was
dissolved in Et O (25 mL) and the resultant solution was washed
for 30 min. Solid Na
and the resultant suspension was stirred until it solidified
(,5 min). The reaction mixture was then diluted with CH Cl
(5 mL) and washed with 10 % aqueous NaOH (3 ꢀ 5 mL). The
combined aqueous washings were extracted with CH Cl
SO (310 mg, 2.46 mmol) was then added
2
3
2
sequentially with saturated aqueous Na CO (3 ꢀ 25 mL) and
2
2
2
3
saturated aqueous NaCl (25 mL), dried, and then concentrated
under vacuum. The residue was dissolved in MeOH (10 mL) and
K CO (864 g, 12.5 mmol) was added. The resultant suspension
2
2
(2 ꢀ 5 mL). The combined organic extracts were washed with
saturated aqueous NaCl (10 mL), dried, and then concentrated
under vacuum to give 96 % conversion into a .95 : 5 mixture of
2
3
was stirred at room temperature for 4 h, filtered, and then con-
centrated under vacuum. The residue was dissolved in Et O
2
(
25 mL) and the resultant solution was washed sequentially with
48 and 51. Data for 48: d
C(5)H , C(6)H ), 1.70–1.76 (1H, m, C(5)H
m, C(6)H ), 2.12 (1H, br s, OH), 3.04–3.08 (1H, m, C(4)H), 3.22
(1H, d, J 3.8, C(2)H), 3.39 (1H, m, C(3)H), 3.96 (2H, app s,
NCH Ph), 3.96–4.01 (1H, m, C(1)H), 7.25–7.39 (5H, m, Ph).
Step 2. MeI (46 mL, 0.74 mmol), Pr NEt (0.13 mL,
2
H
(400 MHz, CDCl
3
) 1.17–1.23 (2H, m,
H O (3 ꢀ 25 mL) and saturated aqueous NaCl (25 mL), dried,
A
A
), 1.93–1.99 (1H,
B
2
and then concentrated under vacuum. Purification via flash
column chromatography (eluent 30–408C petrol/Et O, 1 : 1)
gave 38 as a colourless viscous oil (550 mg, 75 %, .99 : 1 dr).
B
2
2
ꢁ1
i
nmax/cm 3347, 3061, 2936, 1603, 1494, 1453. d (400 MHz,
CDCl ) 1.28–1.39 (1H, m, C(6)H ), 1.56–1.66 (2H, m, C(5)H ,
OH), 1.95–1.99 (1H, m, C(6)H ), 2.14–2.19 (1H, m, C(5)H ),
H
0.74 mmol), and DMAP (3.0 mg, 0.025 mmol) were added
sequentially to a stirred solution of the residue of 48 from the
3
A
A
B
B
3
3
5
.37–3.40 (1H, m, C(4)H), 3.53 (2H, d, J 14.0, N(CH H Ph) ),
A
previous step in CH
resultant mixture was stirred at room temperature for 24 h,
diluted with CH Cl (5 mL), and then washed with H
(2 ꢀ 5 mL). The combined aqueous washings were extracted
with CH Cl (10 mL). The combined organic extracts were
2 2
Cl (1.4 mL) at room temperature and the
B
2
.70 (2H, m, J 14.0, N(CH H Ph) ), 4.25–4.31 (1H, m, C(1)H),
2
A
B
.80–5.87 (2H, m, C(2)H, C(3)H), 7.21–7.40 (10H, m, Ph). dC
2
2
O
2
(100 MHz, CDCl ) 22.0 (C(5)), 32.4 (C(6)), 53.8 (N(CH Ph) ),
3 2 2
5
1
4.6 (C(4)), 67.6 (C(1)), 126.8 (p-Ph), 128.2, 128.4 (o,m-Ph),
þ
2
2
32.8, 133.9 (C(2), C(3)), 140.3 (i-Ph). m/z (ESI ) 294
þ
washed with saturated aqueous NaCl (20 mL), dried, and then
concentrated under vacuum. Purification via flash column
þ
þ
þ
([M þ H] , 100 %). m/z (HRMS ESI ) 294.1852; C H NO
2
0 24
([M þ H] ) requires 294.1852.
chromatography (eluent Et O) gave 49 as a colourless oil
2
ꢁ
1
(
63 mg, 55 % from 36, .99 : 1 dr). nmax/cm 3385, 2945,
2
1
2
864, 2796, 1603, 1495, 1454. dH (400 MHz, CDCl ) 1.16–
.27 (1H, m, C(5)H ), 1.47–1.71 (3H, m, C(5)H , C(6)H , OH),
.01–2.14 (1H, m, C(6)H ), 2.38 (3H, s, NMe), 3.07–3.16 (2H,
B
(
RS,RS)-4-(N-Benzylammonio)cyclohex-2-en-l-ol
Trichloroacetate (39)
Cl CCO H (588 mg, 3.6 mmol) was added to a solution of 36
73.2 mg, 0.36 mmol) in CD Cl (1.0 mL) to give 39. dH
2 2
500 MHz, CD Cl ) 1.75–1.81 (1H, m), 1.96–2.02 (1H, m),
2 2
.39–2.47 (2H, m), 4.17–4.19 (1H, m), 4.37–4.42 (2H, m), 4.64–
.65 (1H, m), 6.02 (1H, app d, J 10.3), 6.27 (1H, app d, J 10.3),
.48–7.49 (5H, m), 7.83 (1H, br s), 8.03 (1H, br s).
3
A
B
A
3
2
m, C(2)H, C(4)H), 3.43 (1H, app br s, C(3)H), 3.67–3.71 (1H, m,
NCH H Ph), 3.79–3.83 (1H, m, NCH H Ph), 3.99 (1H, app
dd, J 10.2, 6.5, C(1)H), 7.26–7.29 (1H, m, Ph), 7.33–7.39 (4H,
m, Ph). dC (100 MHz, CDCl ) 15.2 (C(5)), 31.3 (C(6)), 38.6
(
(
2
4
7
A
B
A B
3
(NMe), 54.7 (C(3)), 56.3 (C(4)), 58.5 (NCH Ph), 60.0 (C(2)),
2
6
5.9 (C(1)), 127.0 (p-Ph), 128.3, 128.8 (o,m-Ph), 135.0 (i-Ph).
þ
þ
þ
m/z (ESI ) 234 ([M þ H] , 100 %) m/z (HRMS ESI ) 256.1308;
(
RS,RS)-4-(N-benzyl-N-methylammonio)cyclohex-2-en-1-
þ
2
þ
C H NNaO ([M þ Na] ) requires 256.1308.
1
4
19
ol Trichloroacetate (40)
Cl CCO H (588 mg, 3.6 mmol) was added to a solution of 37
3
2
Method B: Epoxidation of 37
(78.2 mg, 0.36 mmol) in CD Cl (1.0 mL) to give 40. dH
2 2
Cl CCO H (606 mg, 3.70 mmol) was added to a stirred
3
2
(500 MHz, CD Cl ) 1.73–1.81 (1H, m), 1.90–2.00 (1H, m),
2 2
solution of 37 (80 mg, 0.37 mmol, .99 : 1 dr) in CH Cl2
2
2
4
.38–2.49 (2H, m), 2.82–2.88 (3H, m), 4.16–4.27 (1H, m), 4.36–
.43 (1H, m), 4.49–4.55 (1H, m), 4.66–4.69 (1H, m), 5.92–6.03
(
1.0 mL, 0.36 M with respect to 37) at room temperature and
the reaction mixture was stirred for 5 min. m-CPBA (65 % by wt,
89 mg, 1.84 mmol) was subsequently added at room tempera-
(
(
1H, m), 6.31–6.41 (1H, m), 7.49–7.54 (5H, m), 8.29–8.42
1H, m).
4
ture and the reaction mixture was then stirred for 3 h. Solid
Na SO (232 mg, 1.84 mmol) was then added and the resultant
solution was stirred until the reaction mixture solidified
2
3
(
RS,RS)-4-(N,N-dibenzylammonio)cyclohex-2-en-1-ol
Trichloroacetate (41)
(
,5 min). The reaction mixture was then diluted with CH Cl2
(5 mL) and washed with 10 % aqueous NaOH (3 ꢀ 5 mL). The
combined aqueous washings were extracted with CH Cl
2
2
Cl CCO H (588 mg, 3.6 mmol) was added to a solution of 38
3
2
(
106 mg, 0.36 mmol) in CD Cl (1.0 mL) to give 41. dH
2
2
2
(
2
4
500 MHz, CD Cl ) 1.71–1.77 (1H, m), 2.02–2.09 (1H, m),
2
.45–2.47 (2H, m), 4.26–4.30 (1H, m), 4.41–4.45 (3H, m), 4.50–
.54 (1H, m), 4.68–4.69 (1H, m), 6.01–6.03 (1H, m), 6.35–6.37
(2 ꢀ 5 mL). The combined organic extracts were washed with
saturated aqueous NaCl (10 mL), dried, and then concentrated
under vacuum to give 91 % conversion into an 82 : 18 mixture of
49 and 52. Purification via flash column chromatography (eluent
2
(
1H, m), 7.42–7.53 (10H, m), 8.06 (1H, br s).

6
20
M. B. Brennan et al.
Et O) gave 49 as a colourless oil (43 mg, 50 %, .99 : 1 dr).
2
m-CPBA (60 % by wt, 515 mg, 1.79 mmol) was subsequently
added and the resultant mixture was stirred at room temperature
for 3 h. Solid Na SO (226 mg, 1.79 mmol) was then added and
Further elution gave a 47 : 53 mixture of 49 and 52 as a colour-
less oil (6 mg, 7 %). Further elution gave 52 as a colourless oil
2
3
ꢁ
1
(
5mg, 6 %, .99 : 1 dr). nmax/cm 3385, 2943, 2865, 2793, 1603,
the resultant suspension was stirred until it solidified (,5 min).
1495, 1453. d (400 MHz, CDCl ) 1.26–1.43 (2H, m, C(5)H ,
C(6)H ), 1.71–1.82 (2H, m, C(5)H , C(6)H ), 1.92 (1H, br s,
The reaction mixture was then diluted with CH Cl (5 mL) and
washed with 10 % aqueous NaOH (3 ꢀ 5 mL). The combined
aqueous washings were extracted with CH Cl (2 ꢀ 5 mL). The
H
3
A
2
2
A
B
B
OH), 2.28 (3H, s, NMe), 2.93–2.97 (1H, m, C(4)H), 3.34–3.35
1H, m, C(2)H), 3.39 (1H, dd, J 3.9, 1.5, C(3)H), 3.64 (2H, ABq,
NCH Ph), 3.99 (1H, ddd, J 10.2, 5.4, 1.5, C(1)H), 7.24–7.36
2
2
(
combined organic extracts were washed with saturated aqueous
NaCl (10 mL), dried, and then concentrated under vacuum to
give 88 % conversion into a 25 : 75 mixture of 50 and 53.
Purification via flash column chromatography (eluent 30–
2
(
(
5H, m, Ph). d (100 MHz, CDCl ) 21.4 (C(5)), 25.6 (C(6)), 38.3
C 3
NMe), 56.5 (C(2)), 57.2 (C(4)), 58.5 (NCH Ph), 58.6 (C(3)),
2
6
9.0 (C(1)), 127.1 (p-Ph), 128.4, 128.6 (o,m-Ph), 139.1 (i-Ph).
þ
408C petroleum ether/Et O, 1 : 1) gave 50 as a colourless oil
2
þ
þ
m/z (ESI ) 234 ([M þ H] , 100 %). m/z (HRMS ESI )
(19 mg, 19 %, 95 : 5 dr). Further elution gave 53 as a white solid
(44 mg, 44 %, .99 : 1 dr); Anal. Calc. for C H NO : C 77.6,
H 7.5, N 4.5. Found C 77.7, H 7.6, N 4.4 %. Mp 93–968C.
þ
þ
2
Returned starting material 37 was also isolated in 5 % yield.
56.1312; C H NNaO ([M þ Na] ) requires 256.1308.
1
4
19
2
20 23
2
ꢁ1
nmax/cm 3417, 2926, 2853, 1643, 1494, 1454. d (400 MHz,
H
(
(
1RS,2SR,3RS,4RS)- and (1RS,2RS,3SR,4RS)-2,3-Epoxy-4-
N,N-dibenzylamino)cyclohexan-1-ol (50 and 53)
CDCl ) 1.20–1.28 (1H, m, C(6)H ), 1.37–1.43 (1H, m, C(5)H ),
1.72–1.80 (3H, m, C(5)H , C(6)H , OH), 3.01 (1H, app dd,
3
A
A
B
B
J 11.1, 6.0, C(4)H), 3.31–3-32 (1H, m, C(2)H), 3.40–3.42 (1H,
m, C(3)H), 3.70 (4H, ABq, N(CH Ph) ), 3.98 (1H, app dd, J 9.9,
Method A: Epoxidation of 36
2
2
Step 1. Cl CCO H (804 mg, 4.92 mmol) was added to a
3
2
4
.8, C(1)H), 7.23–7.41 (10H, m, Ph). d (100 MHz, CDCl ) 21.6
C 3
stirred solution of 36 (100 mg, 0.49 mmol, 97 : 3 dr) in CH Cl
2
2
(C(5)), 25.5 (C(6)), 52.4 (C(4)), 54.8 (N(CH Ph) ), 56.5 (C(2)),
2 2
(
1.4 mL, 0.36 M with respect to 36) at room temperature and the
resultant mixture was stirred for 5 min. m-CPBA (60 % by wt,
07 mg, 2.46 mmol) was subsequently added and the resultant
5
1
8.7 (C(3)), 69.1 (C(1)), 127.1 (p-Ph), 128.3, 128.4 (o,m-Ph),
þ
þ
39.5 (i-Ph). m/z (ESI ) 310 ([M þ H] , 100 %). m/z (HRMS
7
þ
þ
þ
ESI ) 332.1617; C H NNaO
2
([M þ Na] ) requires
0
23
2
mixture was stirred for 30 min at room temperature. Solid
Na SO (310 mg, 2.46 mmol) was then added and the resultant
3
1
32.1621. Returned starting material 38 was also isolated in
0 % yield.
2
3
suspension was stirred until it solidified (,5 min). The reaction
mixture was then diluted with CH Cl (5 mL) and washed with
1
2
2
[
36]
X-Ray Crystal Structure Determination for 53
0 % aqueous NaOH (3 ꢀ 5 mL). The combined aqueous wash-
ings were extracted with CH Cl (2 ꢀ 5 mL). The combined
Data were collected using a Nonius k-CCD diffractometer with
graphite monochromated MoKa radiation using standard pro-
cedures at 150 K. The structure was solved by direct methods
(SIR92); all non-hydrogen atoms were refined with anisotropic
thermal parameters. Hydrogen atoms were added at idealised
positions. The structure was refined using CRYSTALS.
X-Ray crystal structure data for 53 [C H NO ]: M 309.41,
tetragonal, space group I 4 /a, colourless plate, crystal
2
2
organic extracts were washed with saturated aqueous NaCl
10 mL), dried, and then concentrated under vacuum to give
(
9
6 % conversion into a .95 : 5 mixture of 48 and 51.
i
Step 2. BnBr (88 mL, 0.74 mmol), Pr NEt (0.13 mL,
2
[
44]
0.74 mmol), and DMAP (3.0 mg, 0.025 mmol) were added
sequentially to a stirred solution of the residue of 48 from the
2
0
23
2
previous step in CH Cl (1.4 mL) at room temperature and the
2
2
1
3
˚
resultant mixture was stirred at room temperature for 24 h,
diluted with CH Cl (5 mL), and then washed with H O
dimensions 0.13 ꢀ 0.17 ꢀ 0.29 mm , a ¼ b 33.1283(4) A,
3
˚
˚
c 6.34210(10) A, a ¼ b ¼ g 908, V 6960.35(16) A , Z 16,
2
2
2
ꢁ1
(
2 ꢀ 5 mL). The combined aqueous washings were extracted
m 0.076 mm . A total of 15034 reflections, and 3980 unique
reflections, were measured for 5 , y ,27 and 3980 reflections
were used in the refinement. The final parameters were GooF
1.0917, wR 0.091 and R 0.164 [I . ꢁ3.0s(I)].
with CH Cl (10 mL). The combined organic extracts were
2
2
washed with saturated aqueous NaCl (20 mL), dried, and then
concentrated under vacuum. Purification via flash column
chromatography (eluent 30–408C petrol/Et O, 1 : 1) gave 50 as
2
1
2
ꢁ
1
a colourless oil (76 mg, 50 %, .99 : 1 dr). nmax/cm 3407,
027, 2961, 1602, 1494, 1453. d (400 MHz, CDCl ) 1.05–1.15
Supplementary Material
3
H
3
1
13
Copies of H and C NMR spectra and thermal ellipsoid plots
are available on the Journal’s website.
(1H, m, C(6)H ), 1.49–1.59 (2H, m, C(5)H ), 1.88 (1H, br s,
OH), 2.03–2.10 (1H, m, C(6)H ), 3.04–3.09 (2H, m, C(2)H,
A 2
B
C(4)H), 3.40 (1H, app d, J 3.4, C(3)H), 3.68 (2H, d, J 14.2,
N(CH H Ph) ), 3.86 (2H, d, J 14.2, N(CH H Ph) ), 3.95 (1H,
app dd, J 10.2, 6.5, C(1)H), 7.21–7.41 (10H, m, Ph). dC
Acknowledgement
A
B
2
A
B
2
The authors would like to thank the National University of Ireland for a
Travelling Studentship (M.B.B.).
(100 MHz, CDCl ) 15.6 (C(6)), 31.5 (C(5)), 54.6 (N(CH Ph) ),
3 2 2
5
1
5.3 (C(2), C(3)), 56.3 (C(4)), 65.9 (C(1)), 126.8 (p-Ph), 128.2,
þ
þ
28.5 (o,m-Ph), 140.3 (i-Ph). m/z (ESI ) 310 ([M þ H] ,
References
þ
þ
þ
1
requires 310.1802.
00 %). m/z (HRMS ESI ) 310.1802; C H NO ([M þ H] )
2
0
24
2
[1] N. Prileschajew, Ber. Dtsch. Chem. Ges. 1909, 42, 4811. doi:10.1002/
CBER.190904204100
[
2] H. B. Henbest, R. A. L. Wilson, J. Chem. Soc. 1957, 1958. doi:10.1039/
JR9570001958
Method B: Epoxidation of 38
[
3] P. Chamberlain, M. L. Roberts, G. H. Whitham, J. Chem. Soc. B 1970,
1374. doi:10.1039/J29700001374
Cl CCO H (585 mg, 3.58 mmol) was added to a stirred
3
2
solution of 38 (105 mg, 0.36 mmol, .99 : 1 dr) in CH Cl2
2
(1.0 mL, 0.36 M with respect to 38) at room temperature and
the resultant mixture was stirred at room temperature for 5 min.
[4] P. D. Bartlett, Rec. Chem. Prog. 1950, 11, 47.
[5] A. H. Hoveyda, D. A. Evans, G. C. Fu, Chem. Rev. 1993, 93, 1307.
doi:10.1021/CR00020A002

Hydroxy- and Ammonium-Directed Epoxdiation
621
[
[
6] K. B. Sharpless, T. R. Verhoeven, Aldrichim Acta 1979, 12, 63.
7] For discussion and molecular modelling calculations, see:
[25] S. K. Bagal, S. G. Davies, A. M. Fletcher, J. A. Lee, P. M. Roberts,
P. M. Scott, J. E. Thomson, Tetrahedron Lett. 2011, 52, 2216.
doi:10.1016/J.TETLET.2010.12.035
[26] K. Csatayov a´ , S. G. Davies, J. G. Ford, J. A. Lee, P. M. Roberts, J. E.
Thomson, J. Org. Chem. 2013, 78, 12397. doi:10.1021/JO4020563
[27] S. K. Bagal, S. G. Davies, J. A. Lee, P. M. Roberts, A. J. Russell,
P. M. Scott, J. E. Thomson, Org. Lett. 2010, 12, 136. doi:10.1021/
OL902533B
a
M. Freccero, R. Gandolfi, M. Sarzi-Amad e` , A. Rastelli, J. Org.
Chem. 2000, 65, 8948. doi:10.1021/JO000900Y
8] B. A. McKittrick, B. Ganem, Tetrahedron Lett. 1985, 26, 4895.
doi:10.1016/S0040-4039(00)94979-7
9] D. K. Fukushima, M. Smulowitz, J. S. Liang, G. J. Lukacs, J. Org.
Chem. 1969, 34, 2702. doi:10.1021/JO01261A047
[
[
[
10] P. O’Brien, A. C. Childs, G. J. Ensor, C. L. Hill, J. P. Kirby, M. J.
Dearden, S. J. Oxenford, C. M. Rosser, Org. Lett. 2003, 5, 4955.
doi:10.1021/OL035873N
[28] S. K. Bagal, S. G. Davies, J. A. Lee, P. M. Roberts, P. M. Scott, J. E.
Thomson, J. Org. Chem. 2010, 75, 8133. doi:10.1021/JO101756G
[29] A. J. Cresswell, S. G. Davies, J. A. Lee, M. J. Morris, P. M. Roberts,
J. E. Thomson, J. Org. Chem. 2012, 77, 7262. doi:10.1021/JO301056R
[30] See pp. 191–193 in M. J. Pilling, P. W. Seakins, Reaction Kinetics
1996 (Oxford University Press: New York, NY).
[
11] P. Kocovsky, I. Stary, J. Org. Chem. 1990, 55, 3236. doi:10.1021/
JO00297A047
[
12] W. J. Brouillette, A. Saeed, A. Abuelyaman, T. L. Hutchison,
P. E. Wolkowicz, J. B. McMillin, J. Org. Chem. 1994, 59, 4297.
doi:10.1021/JO00094A049
[31] J.-E. B a¨ ckvall, J.-E. Nystr o¨ m, R. E. Nordberg, J. Am. Chem. Soc. 1985,
107, 3676. doi:10.1021/JA00298A043
[
[
[
[
13] L. Goodman, S. Winstein, R. Boschan, J. Am. Chem. Soc. 1958, 80,
[32] J. Tsuji, H. Takahashi, M. Morikawa, Tetrahedron Lett. 1965, 6, 4387.
doi:10.1016/S0040-4039(00)71674-1
4312. doi:10.1021/JA01549A049
14] G. Lukacs, D. K. Fukushima, J. Org. Chem. 1969, 34, 2707.
doi:10.1021/JO01261A048
15] R. Vince, S. Daluge, J. Med. Chem. 1974, 17, 578. doi:10.1021/
JM00252A003
16] J. E. Baldwin, R. M. Adlington, J. Chondrogianni, M. S. Edenborough,
[33] B. M. Trost, D. L. Van Vranken, Chem. Rev. 1996, 96, 395.
doi:10.1021/CR9409804
[34] The corresponding cis-amino alcohols 30–32 were also observed in the
1
H NMR spectra of the crude reaction mixtures.
[35] The preparation of both cis-30 and trans-36 has been previously
reported; see: R. G. P. Gatti, A. L. E. Larsson, J.-E. B a¨ ckvall,
J. Chem. Soc., Perkin Trans. 1 1997, 577. doi:10.1039/A600779A
[36] Full crystallographic data have been deposited with the Cambridge
Crystallographic Data Centre as supplementary publication numbers
CCDC 1021308 (32) and 1021309 (53). Copies of these data can be
obtained free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
J. W. Keeping, C. B. Ziegler, J. Chem. Soc., Chem. Commun. 1985,
816. doi:10.1039/C39850000816
[
17] G. Asensio, R. Mello, C. Boix-Bernardini, M. E. Gonz a` lez-N u´ n˜ ez,
G. Castellano, J. Org. Chem. 1995, 60, 3692. doi:10.1021/
JO00117A020
[
[
[
18] A. S. Edwards, R. A. J. Wybrow, C. Johstone, H. Adams, J. P. A.
Harrity, Chem. Commun. 2002, 1542. doi:10.1039/B203851J
19] V. K. Aggarwal, G. Y. Fang, Chem. Commun. 2005, 3448.
doi:10.1039/B503516C
20] C. Aciro, T. D. W. Claridge, S. G. Davies, P. M. Roberts, A. J. Russell,
J. E. Thomson, Org. Biomol. Chem. 2008, 6, 3751. doi:10.1039/
B808811J
21] C. Aciro, S. G. Davies, P. M. Roberts, A. J. Russell, A. D. Smith, J. E.
Thomson, Org. Biomol. Chem. 2008, 6, 3762. doi:10.1039/B808812H
22] C. W. Bond, A. J. Cresswell, S. G. Davies, W. Kurosawa, J. A. Lee, A.
M. Fletcher, P. M. Roberts, A. J. Russell, A. D. Smith, J. E. Thomson,
J. Org. Chem. 2009, 74, 6735. doi:10.1021/JO9012783
[37] O. Mitsunobu, M. Yamada, T. Mukaiyama, Bull. Chem. Soc. Jpn.
1967, 40, 935. doi:10.1246/BCSJ.40.935
[38] O. Mitsunobu, Y. Yamada, Bull. Chem. Soc. Jpn. 1967, 40, 2380.
doi:10.1246/BCSJ.40.2380
[39] K. C. Kumara Swamy, N. N. Bhuvan Kumar, E. Balaraman, K. V. P.
Pavan Kumar, Chem. Rev. 2009, 109, 2551. doi:10.1021/CR800278Z
[
3 2
[40] This method to determine the amount of Cl CCO H (or any Brønsted
[
acid) required to protect the nitrogen atom against oxidation has been
shown to give excellent agreement with results derived from experi-
ment (see Ref. [20]).
[
[
23] M. B. Brennan, T. D. W. Claridge, R. G. Compton, S. G. Davies, A. M.
Fletcher, M. C. Henstridge, D. S. Hewings, W. Kurosawa, J. A. Lee,
P. M. Roberts, A. K. Schoonen, J. E. Thomson, J. Org. Chem. 2012, 77,
[41] R. E. Parker, N. S. Isaacs, Chem. Rev. 1959, 59, 737. doi:10.1021/
CR50028A006
[42] J. K. Addy, R. E. Parker, J. Chem. Soc. 1963, 915. doi:10.1039/
JR9630000915
7241. doi:10.1021/JO3010556
24] (a) For reviews, see: W. Kurosawa, P. M. Roberts, S. G. Davies, Yuki
Gosei Kagaku Kyokaishi (J. Synth. Org. Chem. Jpn.) 2010, 68, 1295.
doi:10.5059/YUKIGOSEIKYOKAISHI.68.1295
[43] D. Swern, Org. React. 1953, VII, 392.
[44] P. W. Betteridge, J. R. Carruthers, R. I. Cooper, C. K. Prout,
D. J. Watkin, J. Appl. Cryst. 2003, 36, 1487. doi:10.1107/
S0021889803021800
(b) S. G. Davies, A. M. Fletcher, J. E. Thomson, Org. Biomol. Chem.
2014, 12, 4544. doi:10.1039/C4OB00616J