The reaction of 8-amino-p-menthene derivatives with electrophiles

Article abstract of DOI:10.1071/CH03158

Attempts to ring-close the nitrogen atom of 8-amino-p-menth-l-ene and of N-substituted 8-amino-p-menth-l-enes onto the C1-C2 double-bond carbons has led to a range of bicyclo[2.2.2] and bicyclo[3.2.1] products, together with the novel bicyclo[4.3.1]-1,3-o

Full text of DOI:10.1071/CH03158

Full Paper
CSIRO PUBLISHING
www.publish.csiro.au/journals/ajc
Aust. J. Chem. 2004, 57, 583–591
The Reaction of 8-Amino-p-menthene Derivatives with Electrophiles
Paul V. Bernhardt,A Raymond M. Carman,A,B and Roger P. C. DerbyshireA
A Department of Chemistry, University of Queensland, Brisbane QLD 4072, Australia.
B Author to whom correspondence should be addressed (e-mail: carman@chemistry.uq.edu.au).
Attempts to ring-close the nitrogen atom of 8-amino-p-menth-1-ene and of N-substituted 8-amino-p-menth-1-enes
onto the C1–C2 double-bond carbons has led to a range of bicyclo[2.2.2] and bicyclo[3.2.1] products, together with
the novel bicyclo[4.3.1]-1,3-oxazepine 9.
Manuscript received: 26 June 2003.
Final version: 3 November 2003.
Introduction
Unchanged starting material was the only isolatable com-
pound from the attempted reaction of N-acetyl derivative 5a
with one equivalent of NBS in acetone. Despite the use of
prolonged reaction times and refluxing solvent, no new prod-
ucts were observed by TLC, GC/MS, or NMR spectroscopy.
The acetamide nitrogen atom was evidently not nucleophilic
enough to attack any bromonium ion produced during the
reaction.
The N-BOC derivative 5b was reacted with NBS at room
temperature for 19 h, when no further product formation was
observed. A significant amount of starting material remained
(determined by TLC), but at least one new compound was
present. Flash chromatography resulted in significant degra-
dation of this new product but a pure sample was collected in
14% isolated yield. The high-resolution mass spectra, NMR
spectra, and microanalysis results of this new compound were
consistent with structure 7. Evidently, the double bond of
compound 5b reacted with NBS to give a bromonium ion,
which was not attacked by any nucleophile until bromine
atoms in the reaction mixture were converted into bromide
We recently discussed[1] the synthesis of ‘azacineole’ 4
from 8-amino-p-menth-1-ene 1 through the steps listed in
Scheme 1. The intermediate 2-bromo bicyclic compound 2
wastoounstabletoisolate, andthenucleophilicnitrogenatom
rapidly displaced the bromine to form aziridine 3.While com-
pound 3 could be readily converted into the desired bicycle
4, we were also attracted to the possibility that the employ-
ment of less nucleophilic nitrogen substituents might enable
the isolation of more stable 2-substituted azacineoles.
This approach, together with attempts to cyclize com-
pound 1 by using mercury electrophiles, is now discussed.
Results and Discussion
There are numerous examples in the literature of the
intramolecular ring closure of N-acyl-protected nitrogen
atoms onto activated C C double bonds.[2] In the present
example we anticipated that derivatives 5a–5c might, fol-
lowing electrophilic activation of the C C double bond, be
cyclized to structures 6 (Scheme 2). Products 6a–6c were
expected to be isolable since the non-nucleophilic amide
would be unlikely to displace the E group. The E group and
amide groups could then be modified or removed at will.
The N-acetyl 5a, N-tert-butyloxycarbonyl (BOC) 5b,
and N-benzoyl 5c derivatives of amine 1 were synthe-
sized. Each compound was then reacted separately with
N-bromosuccinimide (NBS), with greatly disparate results.
R
Eϩ/ϪHϩ
N
H
NH
R
E
5a R ϭ –CO–CH3
b R ϭ –CO–OC(CH3)3 (BOC)
c R ϭ –CO–Ph
6a Rϭ –CO–CH3
b Rϭ –CO–OC(CH3)3
c Rϭ –CO–Ph
d R ϭ –CH2–Ph
d Rϭ –CH2–Ph
Br
H
Br
Me
Me
H
H
8
Br
Br
N
1
4
1
N
N
2
H2/Pd
NBS
NH2
1
2
H
H
Br
NH BOC
8
Br
Br
1
2
3
4
7
8
Scheme 1.
Scheme 2.
10.1071/CH03158
© CSIRO 2004
0004-9425/04/060583

584
P. V. Bernhardt, R. M. Carman and R. P. C. Derbyshire
ions or molecular bromine. The trans diaxial stereochemistry
for the two bromine atoms was obtained by comparison with
the NMR spectra of limonene tetrabromides 8.[3–5] In both
compounds 7 and 8 (two diastereomers) the 1H NMR signal
for H2 exists as a doublet of doublets at δ 4.7 (J ≈ 3.5 and
3.0 Hz).
The maximum yield possible for the formation of
compound 7 is 50% when equimolar amounts of starting
5b and NBS are employed. A compound with an identical
mass spectrum to compound 7 was the major product from
the reaction of a sample of compound 5b with bromine in
dichloromethane.
the structure and absolute configuration were confirmed
by X-ray crystallography (Fig. 1). These data may sug-
gest some delocalization about the O1 C11 N1 region
of the molecule, with bond lengths of 1.25 and 1.37 Å,
respectively, for the C11 N1 and C11 O1 bonds, while the
dihedral angle (N1 C11 C12 C17 −9.6(7)◦) indicates that
the OCN group is almost coplanar with the phenyl ring. The
only analogous oxazepine for which an X-ray structure is
available[6] is a less sterically constrained compound carry-
ing a stabilizing cyano group rather than the phenyl group on
the equivalent C11 atom, and shows a shorter 1.33 Å C O
bond.
It is interesting to note that the double bond in the N-BOC
case, 5b, was brominated while under identical conditions
the double bond of the N-acetyl derivative 5a was completely
inert. In the former case, breakdown of the BOC group under
the reaction conditions must presumably allow production of
bromide ions.
In the N-benzoyl case, 5c, the course of the NBS reaction
was again entirely different. Reaction with one equivalent
of NBS at room temperature led to complete loss of start-
ing material after only four hours. There was no evidence
for dibromination of the C1–C2 double bond of the start-
ing material. The only isolable product after chromatography
was the bicyclic 4,5,6,7-tetrahydro-1,3-oxazepine deriva-
tive 9 in 32% yield. The compound gave crystals from
methanol and was somewhat unstable, as befits a tertiary
bromide. The compound showed NMR spectra consistent
with a carbon carrying a tertiary bromine, and a proton
geminal to the oxygen attachment. This appears to be the
first reported compound to contain a 1,3-oxazepine moi-
ety as part of a bicyclo[4.3.1]decane ring system. Both
The crystalline material was optically active and had [α]D
−54◦ at 25◦C. The starting limonene used in the synthesis of
the compound was >98% the R-(+) enantiomer, and hence
the product is expected to be enantiomerically pure.
1,3-Oxazepines with a C–N double bond identically
placed to that of compound 9 are known to be stabilized
if a phenyl or cyano group is attached to the carbon atom
between the oxygen and the nitrogen,[7] allowing delocaliza-
tion of electrons to occur. Examples do not exist where this
attached group is a methyl or a t-butyl ether, and presum-
ably such products are unstable, as is suggested by Avenosa
et al.[8] in work on simpler 1,3-oxazines.
A mechanism leading to compound 9 involves nucleo-
philic attack by the oxygen atom of the carbonyl group to the
backoftheC2 Brbondofthebromoniumion10(Scheme3).
Of the possible ring closures available to this ion 10 there are
two products, structures 11 and 12 (Scheme 4), that could
be formed by nitrogen attack, and two products that could
be formed by oxygen attack, structures 9 and the less-likely
structure 13 with an eight-membered ring. To study the ener-
getics of these pathways, the heats of formation for the six
possible transition states (there are two possibilities for each
of the nitrogen ring closures depending on whether or not
the carbonyl C O bond points in the same general direction
as the C Br bond) were calculated for the gas phase using
each of the three MOPAC (a semi-empirical molecular orbital
program, version 6.00)[9,10] force fields; MNDO, AM1, and
PM3. However calculated energy differences between the six
possible reaction pathways were small, and we are forced
to conclude that solvation effects must be playing a major
part in favouring the route leading to the observed product 9.
(Calculations with energies and input and output structures
are available from the authors.)
C9
C10
C8
C5
C16
C17
N1
C4
C3
C11
C12
C13
C6
C15
C1
Br1
O1
C2
C14
C7
For structure 11, superficially the anticipated product
from the cyclization of starting material 5c, Dreiding models
indicate an extremely hindered situation where the volume
Fig. 1. The X-ray structure of the bicyclic 1,3-oxazepine derivative 9.
Brϩ
NH
Cϩ
N
4
Brϩ
3
C
Ph
Ph
5c
O
O
2
1
H
NHCOPh
Br
Br
H
10
9
Scheme 3.

Reaction of 8-Amino-p-menthene Derivatives with Electrophiles
585
occupied by the C7, C9, and C10 methyl groups forces the
N CO bond to rotate so that the protons on the phenyl ring
are no longer in the same space as the methyl groups. This
rotation results in the loss of delocalization across the amide
group. The MOPAC results supported this view, with the
minimal-energy conformer for structure 11 having the planes
of the C8 N C1 atoms and the carbonyl–phenyl system at
roughly 90◦ to each other.
This compound is also a new product, and its structure was
verified by X-ray crystallography (Fig. 2).This determination
revealed the presence of one molecule of strongly hydrogen-
bonded water per product molecule in the unit cell, where it is
of interest that the lone pair of electrons on the nitrogen atom
are not directly involved in the hydrogen bonding (Fig. 3).
Inafurtherexploration, compound1wastreatedwithNBS
in the presence of perchloric acid, following the rationale
of Yardley and Rees[11] who argued that the bromonium
ion formed under aqueous acidic conditions would not be
attacked by the protonated amine, allowing, instead, the for-
mation of a bromohydrin. Subsequent basification might then
allow displacement of the bromine atom by the amino group,
leading to structures such as 14 (Scheme 4). In the event,
the only isolated product was epoxide 15, formed by alka-
line treatment of a corresponding bromohydrin. Compound
15, although a simple amino epoxy-p-menthane, appears to
be a new compound. Its stereochemistry was obtained by
comparison of NMR spectra with those of other 1,2-epoxy-
p-menthanes.[12–14] Compound 15 displays H2 as a doublet
(J 5.3 Hz) at δ 2.97, comparable with the literature and dif-
ferent from the alternate α-epoxide stereochemistry where
obvious doublets of doublets (J ≈ 2, ≈2 Hz) are observed.
The stereochemistry in compound 15 does not allow sub-
sequent SN2 displacement of the epoxide group by the amino
nitrogen.
C9
C2
C3
C4
C1
N1
O2
O1
C8
C7
C5
C6
C10
Fig. 2. X-Ray structure of compound 17.
b
In an attempt to gain further insight into the ring closure
of the various N-protected starting materials, amine 1 and its
derivatives 5 were then examined with mercuric acetate. The
direct reaction of mercuric acetate with unprotected amine 1
was very slow, with considerable reversion back to starting
materials. Demercuration afforded recovered starting amine
1 (35%), aziridine 3 (30%), and unstable acetate 16[1] (26%).
The reaction of compound 1 with mercuric trifluoro-
acetate, followed by hydride reduction under alkaline condi-
tions, proved cleaner and faster. Only 4% of starting material
1 was recovered, with aziridine 3 (12%) and a new crystalline
alcohol 17 (81%) predominating. Compound 17, a racemate,
is assumed to form by aTreibs-type mechanism[15–19] involv-
ing reversible allylic rearrangement about carbons 2, 1, and 6.
a
c
d
Fig. 3. The unit cell of allylic alcohol 17, viewed along the a axis,
showing hydrogen-bonding interactions.
10
9
N
C
O
H
PhCO
H
8
COPh
N
N
N
O
N
Ph
1
2
OH
H
NH2
7
OCOCH3
Br
Br
Br
11
12
13
14
15
16
Bn
Bn
H
Bn
H
N
N
H
N
O
1
2
6
HO
CH3COO
3
3
1
1
2
2
H
6
H
H
H
NHCOOC(CH3)3
NH2
H
OH
OH
OCOCH3
OCOCH3
17
18
19
20
21
22
Scheme 4.

586
P. V. Bernhardt, R. M. Carman and R. P. C. Derbyshire
The lack of formation of a trifluoroacetate corresponding to
acetate 16 in this reaction is presumably a reflection of the
poor nucleophilic properties of the trifluoroacetate anion.
The mercuration–demercuration of N-acetyl derivative
5a with mercuric acetate led to a complete recovery of
unchanged starting material.
The N-BOC compound 5b did react with mercuric acetate,
although not to completion and not to give a ring-closed
product. The isolated crystalline product was acetate 18
(Scheme 4). The racemic nature of this compound, which
is presumably also formed by a Triebs-type pathway, was
confirmed by X-ray crystallography (Fig. 4).
Mercuration–demercuration of N-benzoyl derivative 5c
with mercuric acetate under identical conditions to those
applied to the N-BOC derivative led only to complete
recovery of starting material. No Treibs-type reaction was
observed.
is probably derived from an HgH substituent.[23] It is then of
interest to compare the attack of acetate anion upon the two
aziridine rings present in compounds 3 and 23, where the first
compound 3 leads to the isolated [2.2.2]bicyclo product 16,[1]
while the benzyl derivative 23 provides the alternate [3.2.1]
derivative 19. Attack of a nucleophile might be expected to
be faster at the less-hindered C2 position of the aziridine 23
leading to the kinetic [2.2.2]bicyclo product 24 (Scheme 6).
However this un-isolated material is labile, reverting back
to structure 23. Slower attack at the more-hindered C1 end
of aziridine 23 leads to the thermodynamic [3.2.1] product
19. Protonation is not required in this sequence, which pro-
ceeds rapidly under alkaline conditions. In contrast, aziridine
3 requires protonation to become activated as salt 25 before
nucleophilic ring opening occurs (Scheme 7). Kinetic attack
at C2 affords salt 26, in which the protonated nitrogen is not
nucleophilic enough to displace the acetate group, so under
these conditions the product 26 is not in equilibrium with
The N-benzyl compound 5d reacted with mercuric acetate
in dichloromethane to provide, in 88% isolated yield after
reduction of the intermediate organomercurial, a ring-closed
product characterized as N-benzyl-2,8-azacineole-1-acetate
19. Compound 19 was distinguished from the other poten-
tial product, isomeric structure 20 (Scheme 4), on the basis
Bn
N
1. Hg(OAc)2
5d
2. HϪ
1
of the H2 coupling pattern in the H NMR spectrum. The
L
signal for H2 of compound 19 shows a doublet at δ 3.57,
with J 6.8 Hz to one H3, and with a coupling of J ≈ 0 Hz
to the other adjacent C3 proton with which it makes a tor-
sion angle of approximately 90◦. This pattern is similar to
that shown by pinol 21[20] where H2 provides a doublet at δ
3.82, with J 5 and 0 Hz to the two adjacent C3 protons. The
alternate bicyclo[2.2.2]structure 20 is expected, from many
examples,[1,21,22] to provide H2 as a doublet of doublets of
doublets, withobservablecouplingstotwoH3andlong-range
W to H6.
6d
Bn
Bn
Nϩ
N
CH3COOϪ
2
1
H
OCOCH3
23
19
Scheme 5.
The tertiary nature of the acetate group in structure 19
was chemically confirmed by reduction of the compound
with lithium aluminium hydride in ether to the alcohol 22
in 96% yield. This tertiary alcohol 22 was resistant to normal
oxidation.
Bn
Bn
Nϩ
:
N
C2 attack faster
2
1
ϪOAc
OAc
A mechanism for the formation of compound 19 is pro-
vided in Scheme 5, where the leaving group L is unclear but
23
24
Kinetic product
C1 attack slower
C11
Bn
N
:
N
Bn
C12
C3
Me
Thermodynamic
product
C10
H
H
O3
OAc
OAc
N1
C2
19
19a
C16
C4
O4
C13
C1
Scheme 6.
C14
C5
C7
C17
C15
H
H
O1
C8
H
C6
Nϩ
Nϩ
:
N
Hϩ
Ϫ
OAc
Kinetic
O2
OAc
C9
3
25
26
Scheme 7.
Fig. 4. X-Ray structure of compound 18.

Reaction of 8-Amino-p-menthene Derivatives with Electrophiles
587
starting material 3. In summary, structures 23 and 24 can
equilibrate under alkaline conditions, whereas under these
conditions structure 3 does not convert into acetate 26.
MOPAC calculations (gas phase) did not provide clear
evidenceforthelowerheatofformationofthe[3.2.1]bicycles
(e.g., structure 19) over the [2.2.2] system (e.g., 24), but they
do allow the prediction of conformer 19a to be the most stable
arrangement available to compound 19.
The above results did not lead to a useful synthesis of 1,8-
azacineole 4, partly because of participation at the C2 posi-
tion. A modified route starting from 8-amino-p-menth-1(7)-
ene 27 might lead to improvements since reaction of amine
27 with an electrophile (Scheme 8) would give a bicyclic
intermediate 28 uncomplicated by the possibility of further
displacement to give an aziridine skeleton, and removal of
the E group in a subsequent step should then be facile.
Amine 27 is an unknown compound. However, the
acetamide derivative 29, with the required exocyclic double
bond and a suitably placed nitrogen atom, is available[24] from
β-pinene 30 in a Ritter-type reaction by using NBS and trap-
ping the resultant carbocation 31 with acetonitrile to afford
cation 32. Subsequent aqueous workup to give amide 33,
followed by removal of the bromine from this surprisingly
stable allylic bromide with zinc powder in acetic acid affords
acetamide 29 (Scheme 9). We have now employed numerous
reagents in an attempt to hydrolyze the acetamide 29, but the
compound proved to be stable under non-acidic conditions
and amine 27 could not be successfully produced. Hydrolysis
of compound 29 did occur in 10% aqueous sulfuric acid, but
these conditions were then too harsh to allow the survival of
the resulting amine 27, containing a protonated amino group
attached to a tertiary carbon atom, and ammonia was lost
from the product to leave a mixture of isomeric monoterpene
dienes.Thus we have not obtained compound 27 by this route.
We also attempted to trap cation 32 with hydride, with
interesting results. Two new compounds, ethylamine 34 and
the [3.2.1] bicycle 35 were produced (Scheme 10). In pathway
A, two equivalents of hydride add to imine 36, providing
structure 34. The alternate pathway B features the addition
of one equivalent of hydride to imine 36 with a concomitant
cyclization (either concerted or in discrete steps) to generate
bicyclic amine 35. Pathways A and B compete with each
other. When excess hydride (3.0 equivalents of NaBH4) was
used, the product ratio of 34 to 35 was 85 : 15. When less
hydride was added (0.5 equivalents of NaBH4), the product
ratio of 34 to 35 was 10 : 90.
H
H
Eϩ
HϪ
N
N
ϪHϩ
All these attempts failed to provide the desired
bicyclo[2.2.2]-1,8-azacineole skeleton 4, and studies to
explore the ring closures of other N-substituted derivatives
have been abandoned at this stage.
NH2
CH2E
27
28
4
Scheme 8.
Experimental
Br
Br
1H and 13C NMR spectra in CDCl3 solution were recorded using either
a Bruker AMX400 or AV400 spectrometer. Multiplicities for some
obscured and overlapped 1H peaks were obtained from 1H–1H cor-
relation spectroscopy (COSY); where the magnitude of the coupling
constants could not be measured, there are indicated in the spec-
tral data as ‘?’. 13C multiplicities were assigned by the distortionless
enhancement by polarization transfer (DEPT) pulse sequence. Mass
spectra were recorded upon a Hewlett Packard MSD5970 spectrometer
using a GC inlet and BP5 column. High-resolution mass spectra were
NHCOCH3
Cϩ
NHCOCH3
29
30
31
Scheme 9.
33
HϪ
Br
Br
A
CH3
NϭCϩϪCH3
Nϭ
C
NHCH2CH3
H
HϪ
HϪ
32
36
34
Br
CH2CH3
N
B
CH3
H
Nϭ
C
HϪ
36
35
Scheme 10.

588
P. V. Bernhardt, R. M. Carman and R. P. C. Derbyshire
recorded on both Finnigan 2001 FT-MS and Kratos MS25RFA instru-
ments. Infrared spectra were measured using a Perkin–Elmer 1600 series
FT-IR spectrometer with NaCl disks. GC analyses were performed upon
a BP5 capillary column with flame ionization detection in aVarian 3300
instrument. Flash column chromatography was performed using Merck
silica, grade 60, and distilled solvents. New compounds were refined to
>99% (by NMR and GC) purity and analyzed by microanalysis and/or
high-resolution mass spectrometry.
Benzamide 5c
Amine 6 (766 mg, 5 mmol), in dichloromethane containing pyridine,
was treated with benzoyl chloride to provide a colourless powder (92%)
giving long fine needles of (R)-N-[1-methyl-1-(4-methylcyclohex-3-en-
1-yl)ethyl]benzamide [(R)-N-(p-menth-1-en-8-yl)benzamide] 5c, mp
134–135◦C (from benzene) (lit.[27] crystals from ethanol, mp 193◦C)
(Found: C 79.8, H 9.2, N 5.3. Calc. for C17H23NO: C 79.3, H 9.0,
N 5.4%). NMR spectra were consistent with the literature[27] and pro-
vided extra information. δH 7.60–7.56 (2H, m, meta H), 7.37–7.32 (1H,
m, para H), 7.31–7.26 (2H, m, ortho H), 5.80 (br s, NH), 5.27 (m, Wh/2
10, H2), 2.15 (dddd, J 12.2, 11.8, 5.0, 2.3, H4), 2.01–1.90 (2H, m,
1 × H6 and 1 × H3), 1.85 (dm, 1 × H3), 1.80–1.69 (2H, m, H5eq and
1 × H6), 1.53 (3H, br s, H7), 1.33 and 1.30 (2 × 3H, s, H9 and H10),
1.20 (dddd, J 12, 12, 12, 5.6, H5ax). δC see Table 1. vmax (Nujol)/cm−1
3300, 1630, 1599, 1576, 1539, 1316, 1183, 1163, 1076, 1027, 929, 874,
803, 774, 720, 693, 668. m/z (GCMS) 258 (M + 1, 1%), 257 (M, 4),
163 (15), 162 (26), 136 (24), 122 (23), 121 (26), 106 (8), 105 (100), 93
(17), 79 (11), 77 (52).
Acetamide 5a
Amine 1,[1] in ether containing triethylamine, was acetylated with
acetic anhydride to give (R)-N-{1-methyl-1-[4-methylcyclohex-3-en-
1-yl]ethyl}acetamide [(R)-8-acetamido-p-menth-1-ene] 5a (95%) as
colourless prisms, mp 113–114◦C (lit.[25] 111◦C) (Found: C 73.8, H
11.2, N 7.3. Calc. for C12H21NO: C 73.8, H 10.8, N 7.2%). NMR spectra
were consistent with the literature[25,26] and provided extra information.
δH 5.32 (m, Wh/2 12, H2), 5.28 (br s, Wh/2 10, NH), 2.08 (dddd, J 12.6,
12, 5, 2.6, H4), 2.02–1.89 (3H, m, 1 × H6 and 2 × H3), 1.88 (3H, s,
H12), 1.78–1.66 (2H, m, H5eq and 1 × H6), 1.59 (3H, br s, H7), 1.26
and 1.23 (2 × 3H, s, H9 and H10), 1.19 (dddd, J 12, 12, 12, 5.3, H5ax).
δC see Table 1. vmax (Nujol)/cm−1 3268, 3087, 2723, 1641, 1565, 1314,
1296, 1194, 1158, 1032, 965, 916, 800, 721. m/z (GCMS) 195 (M, 9%),
180 (M − 15, 1), 138 (7), 137 (11), 136 (95), 122 (12), 121 (100), 107
(16), 101 (19), 100 (55), 94 (13), 93 (81), 92 (17), 91 (24), 86 (15), 79
(32), 77 (25).
Benzamine 5d
In a variation of the literature synthesis describing the hydrochloride
salt of amine 5d,[28] anhydrous sodium sulfate (1.00 g) was added to
a solution of benzaldehyde (350 mg) in dry benzene (3 mL). Amine 1
(460 mg) in dry benzene (3 mL) was added. After 3 days the mixture
was decanted from the sodium sulfate and filtered through a cotton
wool plug. The residue was washed with further dry benzene. The com-
bined filtrates, after evaporation of solvent, provided an oily residue that
was dissolved in dry methanol (10 mL) and cooled (0◦C) while sodium
borohydride (0.52 g) was added. The reaction mixture was allowed to
warm to room temperature overnight.The methanol was removed, aque-
ous ammonia (25%, 40 mL) was added, and the solution was extracted
into diethyl ether (5 × 30 mL). These extracts were washed with brine,
dried over sodium sulfate, and the solvent was removed to give
the title compound, (R)-N-benzyl-1-methyl-1-(4-methylcyclohex-3-en-
1-yl)ethylamine [(R)-N-benzyl-8-amino-p-menth-1-ene] 5d, (0.71 g,
97%) as a viscous oil which was pure by GC analysis. δH 7.36–7.26
(5H, m, ortho- and meta-H and NH), 7.20 (1H, m, para-H), 5.40 (m,
Wh/2 11, H2), 3.68 (2H, AB system, JAB 14, ꢀAB 12.7, benzylic H),
2.08–1.91 (3H, m, 1 × H6 and 2 × H3), 1.86 (m, 1 × H6), 1.80 (m,
H5eq), 1.64 (3H, br s, H7), 1.64 (obscured dddd, J ≈ 12, ≈12, 4.9, 2.5,
H4), 1.25 (dddd, J 12.5, 12.2, 11.4, 5.9, H5ax), 1.08 and 1.07 (2 × 3H,
s, H9 and H10). δC see Table 1. vmax (neat)/cm−1 3354, 3061, 3026,
Carbamate 5b
Amine 1, in dichloromethane containing pyridine, was treated with
di-tert-butyl dicarbonate to afford, after flash chromatography over
silica (3% diethyl ether, 97% hexane), a viscous oil. Prolonged
exposure to high vacuum gave analytically pure (R)-tert-butyl 1-methyl-
1-[4-methylcyclohex-3-en-1-yl]ethylcarbamate [(R)-N-BOC-8-amino-
p-menth-1-ene] 5b (73%) (Found: C 71.0, H 11.2, N 5.3. C15H27NO2
requires C 71.1, H 10.7, N 5.5%). δH 5.34 (m, Wh/2 11, H2), 4.40 (br s,
Wh/2 13, NH), 2.04–1.88 (4H, m, 2 × H3, H4 and 1 × H6), 1.79–1.72
(2H, m, 1 × H5 and 1 × H6), 1.60 (3H, br s, H7), 1.39 (9H, s, t-butyl),
1.25–1.16 (m, one H5), 1.22 and 1.18 (2 × 3H s, H9 and H10). δC see
Table 1. vmax (neat)/cm−1 3452, 3362, 2972, 2925, 2835, 1722, 1559,
1497, 1455, 1388, 1364, 1308, 1252, 1224, 1173, 1118, 1070, 1034,
1023, 993, 938, 915, 884, 844, 799, 781, 759, 745. m/z (GCMS) 197
(M − 56, 3%), 182 (1), 180 (1), 158 (11), 137 (12), 136 (39), 121 (25), 95
(12), 93 (26), 92 (11), 91 (11), 81 (21), 79 (21), 77 (15), 58 (85), 57 (100).
Table 1. 13C chemical shifts for the p-menthanes (CDCl3 solution)
C2 C3 C4 C5 C6 C7 C8 C9, C10
133.9 120.4 31.0 40.9 24.0 26.5 23.2 56.2 23.8, 24.1
Compound
C1
Other chemical shifts
5a
5b
169.3 (C11), 24.5 (C12)
154.5 (C11), 78.5 (C12),
28.4 (C13, C14, C15)
133.9 120.6 31.2 41.5 24.15 26.6
23.3 54.9 24.13, 24.3
5c
5d
7
134.0 120.4 31.0 41.3 24.18 26.6
23.2 56.7 24.13, 24.18 166.8 (C11), 136.1 (C12),
131.0 (C15), 128.5 (C13,
C17), 126.6 (C14, C16)
134.0 121.1 31.4 41.5 24.1
26.7
36.5
23.3 54.6 24.3, 24.4
35.2 54.5 24.5, 24.9
141.7 (C12), 128.2/128.3
(C13, C14, C16, C17),
126.7 (C15), 46.2, (C11)
154.2 (C11), 78.8 (C12),
28.4 (C13, C14, C15)
71.0
57.4
61.1 33.0 38.1 23.6
58.8 25.5 42.8 19.9
15
17
18
30.6
68.7
70.9
22.7 53.9 25.0, 25.8
20.8 50.8 27.7, 28.6
20.5 54.4 24.2, 24.7
134.4 125.4 27.1 39.0 32.8
131.1 127.8 27.0 35.7 30.2
171.0 (C11), 154.4 (C13),
78.7 (C14), 28.5 (C15, C16,
C17), 21.2 (C12)
29
149.4
134.6 127.5 27.3 40.1 23.7
133.5 121.9 31.8 42.3 24.31 27.1
34.7 28.7 43.7 28.7
34.7 106.7 56.6 24.3, 24.3
169.5 (C11), 24.4 (C12)
169.4 (C11), 24.5 (C12)
35.9 (C11), 16.6 (C12)
33
26.8
38.8 56.1 24.0, 24.2
23.6 54.0 24.30, 24.5
34A
A In [D6]benzene.

Reaction of 8-Amino-p-menthene Derivatives with Electrophiles
589
2962, 2924, 1644, 1581, 1538, 1493, 1451, 1381, 1308, 1160, 1070,
1026, 914, 799, 752, 696. m/z (GCMS) 243 (M, 1%), 228 (2), 160 (2),
149 (12), 148 (100), 146 (6), 136 (2), 121 (3), 106 (2), 93 (3), 92 (8),
91 (86).
δC see Table 1. vmax (neat)/cm−1 3354, 2959, 1654, 1596, 1458, 1431,
1381, 1312, 1257, 1211, 1176, 1096, 1023, 974, 915, 848, 826, 779,
747, 667. m/z (GCMS) 170 (M + 1, 1%), 169 (10), 155 (3), 154 (32),
136 (6), 126 (5), 110 (11), 97 (8), 96 (100), 95 (9), 94 (23), 84 (9), 82
(10), 71 (12).
Dibromide 7
Aminomercuration of Amine 1 with Mercuric Acetate
Compound 5b (770 mg) and N-bromosuccinimide (590 mg) were
stirred (19 h) in acetone (40 mL). Flash chromatography of the prod-
uct (6% diethyl ether, 94% hexane) gave starting material 5b (RF 0.16),
followedbytert-butyl1-[(1R,3S,4S)-3,4-dibromo-4-methylcyclohexyl]-
1-methylethylcarbamate [N-BOC-(1S,2S,4R)-8-amino-1,2-dibromo-
p-menthane] 7, a highly viscous oil (RF 0.10) (14% yield) (Found:
C 43.9, H 6.7, N 3.2. C15H27Br2NO2 requires C 43.6, H 6.6, N 3.4%;
found m/z 416.0440, 414.0459, 412.0473. C15H2881Br2N (M + 1)
requires 416.0440, C15H2879Br81BrN (M + 1) requires 414.0461,
C15H2879Br2N (M + 1) requires 412.0481). δH 4.71 (dd, J 4.4, 2.9, H2),
4.35 (br s, Wh/2 12, NH), 2.52 (dddd, J 12.3, 12.2, 4.7, 2.9, H4), 2.38
(ddd, J.13.5, 12.3, 2.9, H3ax), 2.02 (1H, m, 1 × H6), 2.00 (m, H3eq),
1.96 (3H, s, H7), 1.95 (1H, m, 1 × H6), 1.70–1.61 (2H, m, 2 × H5), 1.41
(9H, s, t-butyl group), 1.28 and 1.19 (2 × 3H, s, H9 and H10). δC see
Table 1. vmax (neat)/cm−1 3445, 3358, 3281, 2974, 1731, 1694, 1667,
1504; with many strong peaks in the fingerprint region. m/z 416 (M + 1,
2%), 414 (M + 1, 4), 412 (M + 1, 2), 359 (10), 358 (3), 357 (20), 355
(9), 278 (5), 276 (4), 196 (9), 159 (3), 158 (38), 135 (7), 102 (36), 58
(100), 57 (21). The compound decomposed under GC conditions.
Amine 1 (500 mg, 3.26 mmol) in dichloromethane (20 mL) was stirred
(8 days) with mercuric acetate (1.04 g, 3.26 mmol). The mixture was
then added slowly to a stirred solution of sodium borohydride (250 mg,
6.52 mmol) in 5 m aqueous sodium hydroxide (2 mL) in an ice-salt
bath. After 1 h, saturated sodium chloride solution (20 mL) and diethyl
ether (50 mL) were added, and the mixture was transferred to a sepa-
ratory funnel. The contents were shaken and the organic (upper) layer
collected. The aqueous layer was further extracted with diethyl ether
when the organic layers were combined and dried (sodium carbonate).
GCMS analysis showed aziridine 3 (39%), starting amine 1 (35%),
and acetate 16 (26%). The mixture was chromatographed on silica
(dichloromethane/methanol, 95 : 4; containing 1% aqueous ammonia).
The products in order of increasing polarity were compounds 3 (30%),
16 (26%), and 1 (35%), although compounds 3 and 16 co-eluted in many
fractions. These three products were all identified by comparison (GC,
NMR) with authentic[1] compounds.
Alcohol 17
Amine 1 (380 mg, 2.48 mmol) in dry dichloromethane (30 mL), was
dried over molecular sieves (3 and 5 Å, approximately 0.5 g). The
dried solution was decanted from the sieves into an oven-dried flask
and mercuric trifluoroacetate (1.05 g, 2.48 mmol) was added. The
flask was sealed and the contents were stirred at room tempera-
ture (6 days) until comparison (GCMS) of a demercurated aliquot
with earlier samples showed that the reaction had reached equili-
brium. Volatile compounds present in the organic extract were start-
ing amine 1, aziridine 3, and a new compound 17 (4 : 12 : 81 by
GCMS). The reaction mixture was cooled in an ice bath and a pre-
chilled solution of sodium borohydride (310 mg, 8.19 mmol) in 2 m
NaOH (20 mL) was added. When the resultant solution had set-
tled (1 h, 0◦C) it was twice filtered through filter paper, which was
washed with dichloromethane. The organic layers provided crude crys-
talline material (220 mg) which contained an oily impurity. Trituration
with small quantities of diethyl ether removed the oily impurity and
left crystals (80 mg, 19%) of (1RS,5SR)-5-(1-amino-1-methylethyl)-
2-methylcyclohex-2-en-1-ol [(4RS,6SR)-8-amino-6-hydroxy-p-menth-
1-ene] 17, as colourless prisms, mp 108–109◦C (from diethyl ether)
(Found: m/z 154.1229, 152.1195. C9H16N1O1 requires 154.1232
(M − CH3, molecular ion was not observed), C10H16O1 requires
152.1201 (M − NH3)). X-ray analysis showed the presence of one water
of crystallization per product molecule. δH 5.55 (m, Wh/2 9, H2), 4.00
(dd, J 3.5, 2.3, H6), 2.09 (dm, J 16.7, ?, H3eq), 1.95 (ddd, J 13.5, 4.1,
2.3, H5eq), 1.76 (3H, br s, H7), 1.71 (obscured dm, J 16.7, ?, H3ax),
1.57 (dddd, J 13.5, 11.2, 4.1, 2.3, H4), 1.49 (3H, br s, NH2 and OH),
1.35 (ddd, J 13.5, 13.5, 3.5, H5ax), 1.06 and 1.04 (2 × 3H, s, H9 and
H10). δC seeTable 1. vmax (Nujol)/cm−1 3447, 3341, 3321, 3269, 3185,
2720, 1613, 1170, 1052, 1033, 964, 806. m/z (GCMS) 154 (M − CH3,
6%), 153 (3), 152 (M − NH3, 11%), 151 (M − H2O, 5%), 138 (2), 137
(23), 136 (34), 134 (19), 119 (30), 110 (12), 109 (100), 107 (12), 96
(11), 95 (48), 94 (23), 93 (29), 92 (11), 91 (50), 84 (11), 83 (11), 82
(12), 79 (40), 77 (53), 70 (67).
Compound 9
Benzamide 5c (0.99 g, 3.85 mmol) and N-bromosuccinimide (0.71 g,
4.00 mmol) in acetone (50 mL) were stirred (4 h) until no starting amide
remained (by TLC). The solvent was removed to give an oily residue
that was triturated with hexane/diethyl ether (95 : 5; 5 × 10 mL). The
combined extract was filtered and the solvent removed to give a vis-
cous oil (0.6 g), which was subjected to flash chromatography on silica
(hexane/diethyl ether; 95 : 5). Fractions containing the solid 9 were
combinedtogive(1R,6R,9R)-9-bromo-5,5,9-trimethyl-3-phenyl-2-oxa-
4-azabicyclo[4.3.1]dec-3-ene 9 (0.42 g, 32%) as colourless prisms, mp
112◦C (from methanol), [α]D –54◦ (CHCl3) (Found: C 60.8, H 6.7,
N 4.0. C17H22BrNO requires C 60.7, H 6.6, N 4.2%). ε245 nm (L mol−1
)
10628. δH 7.97–7.91 (2H, m, ortho H), 7.26–7.17 (3H, m, meta- and
para-H), 4.62 (m, Wh/2 7, H1), 2.70 (ddd, J 15.2, 5.1, 3.2, H10α), 2.46
(ddd, J 15.2, 5.0, 2.2, H10β), 2.11 (dm, J 14.3, 5.9, 3.8, 2.5, H7β),
2.01 (obscured m, H7α), 2.00 (obscured m, H6), 1.89 (obscured ddd,
J 15.4, 12.2, ?, H8α), 1.86 (3H, s, H19), 1.82 (obscured dm, H8β), 1.45
and 1.27 (2 × 3H, s, H17 and H18). δC 147.1 (C3), 136.3 (C11), 129.7
(C14), 128.3 (C12, C16), 127.9 (C13, C15), 81.6 (C1), 70.6 (C9), 59.5
(C5), 39.6 (C6), 35.3 (C8), 32.6 (C19), 30.4 and 30.9 (C17 and C18),
27.6(C10), 23.8(C7). vmax (Nujol)/cm−1 1720, 1667, 1447, 1381, 1360,
1331, 1243, 1191, 1157, 1106, 1093, 1057, 1027, 987, 804, 766, 745,
711, 694. m/z (GCMS) 337 (3%), 336 (M − Me from C17H2281BrNO,
7), 335 (3), 334 (M − Me from C17H2279BrNO, 7), 281 (2), 279 (2), 257
(4), 256 (30), 163 (7), 162 (50), 106 (8), 105 (100), 95 (29), 79 (12),
77 (39).
Epoxide 15
Amine 1 (490 mg, 3.2 mmol) and N-bromosuccinimide (730 mg,
4.1 mmol) were stirred (0◦C) in tetrahydrofuran (40 mL) containing
aqueous HClO4 (0.5 m, 15 mL). After 1 h, sodium hydroxide solu-
tion (5 m, 10 mL) was added. After a further 1 h (0◦C), water (40 mL)
was added and the solution extracted into diethyl ether. The ether was
washed with brine, dried over sodium sulfate, and the solvent was
removed. The residue was chromatographed over silica (methanol/
dichloromethane; 20 : 80) to provided 1-methyl-1-[(1R,3R,6S)-6-
methyl-7-oxabicyclo[4.1.0]hept-3-yl]ethylamine[(1S,2R,4R)-8-amino-
p-menthane-1,2-epoxide] 15 as an oil (360 mg, 67%), >99% pure by
GLC and NMR inspection (Found: m/z 169.1465. C10H19NO requires
169.1467). δH 2.97 (d, J 5.3 Hz, H2), 2.05–1.98 (2H, m), 1.69–1.58
(2H, m), 1.50–1.44 (1H, dm), 1.40–1.30 (1H, m), 1.29 (3H, s, H7),
1.25–1.15 (1H, obscured m), 1.18 and 1.16 (2 × 3H, s, H9 and H10).
Acetate 18
Compound 5b (600 mg, 2.37 mmol) and mercuric acetate (760 mg,
2.38 mmol) were stirred in a solution of dichloromethane (40 mL).
Periodically an aliquot was worked up by addition to excess sodium
borohydride in aqueous sodium hydroxide solution at 0◦C, extraction
into diethyl ether, and analysis by GC. After 7 days no further reac-
tion was observed and the solution was cooled (ice-salt bath) and
added carefully to a stirred solution of sodium borohydride (190 mg,
5 mmol) in aqueous sodium hydroxide (1 M, 20 mL) held at −5◦C.

590
P. V. Bernhardt, R. M. Carman and R. P. C. Derbyshire
After 4 h at −5◦C, the mixture (excluding the precipitated metal-
lic mercury) was transferred to a separatory funnel and extracted
into diethyl ether. Analysis of the extract showed starting mate-
rial 5b and a new compound 18 (58 : 42 by GC). Chromatography
over silica gel gave first starting material (330 mg, 55%, eluted with
hexane), followed by (1RS,5SR)-5-{1-[(tert-butoxycarbonyl)amino]-1-
methylethyl}-2-methylcyclohex-2-en-1-yl acetate [(4RS,6SR)-N-BOC-
8-amino-p-menth-1(2)-en-6-yl acetate] 18 (270 mg, 37%, in hexane/
diethyl ether 70 : 30) as colourless prisms, mp 132◦C (from hexane/
diethyl ether) (Found: C 65.9, H 9.7, N 4.6. C17H29NO4 requires C 65.6,
H 9.4, N 4.5%). δH 5.67 (ddd, J 7.2, 3.6, 1.8, H2), 5.21 (m, Wh/2 8,
H6eq), 4.33 (br s, Wh/2 10, NH), 2.34 (dddd, J 13.7, 11.3, 4.1, 2.0,
H4), 2.08 (dm, J 17.4, ?, H3eq), 2.02 (3H, s, acetate Me), 1.93 (ddd,
J 14.0, 4.1, 2.1, H5eq), 1.75 (dm, J 17.6, ?, H3ax), 1.65 (3H, m, Wh/2 7,
H7), 1.40 (9H, s, t-butyl Me), 1.41–1.32 (obscured m, H5ax), 1.23 and
1.16 (2 × 3H, s, H9 and H10). δC see Table 1. vmax (Nujol)/cm−1 3380,
1720, 1716, 1511, 1365, 1289, 1251, 1178, 1159, 1067, 1047, 1029,
1009, 912. m/z (GCMS) 196 (M − 115, molecular ion not observed,
1%), 195 (2), 158 (22), 135 (10), 134 (8), 119 (15), 103 (6), 102 (100),
94 (10), 93 (16), 91 (11), 79 (12), 77 (8).
state for further reactions, as chromatography led to significant decom-
position. Chromatography (ether/hexane) provided a small amount of
pure sample with 1H NMR, IR, and mass spectra identical with the
literature.[24] δC see Table 1, where some assignments, confirmed by
DEPT studies, are different from the literature.[24]
Compound 29
Crude bromide 33 was reduced with zinc powder in glacial acetic
acid by following the literature method.[24] The product, a mixture of
8-acetamido-p-menth-1(7)-ene 29 and its endocyclic double-bond iso-
mer (approximately 50 : 50) were chromatographed to afford compound
29 as colourless crystals (∼16% yield from compound 30), mp 113–
114◦C (from hexane/ether; lit.[24] 123–125◦C). All spectra, except 13
C
NMR assignments, were identical to the literature.[24] δC: see Table 1.
Thereagentswhichfailedtoconvertamide 29intoamine 27included
sodium peroxide,[29] potassium hydroxide in refluxing dioxan/water,[30]
refluxing methanol containing sodium acetate, and methyl lithium in
refluxing ether/dioxan.[31]
Amine 34
Acetate 19
β-Pinene (1.40 g, 10.1 mmol) and NBS (2.10 g, 11.8 mmol) were vigo-
rously stirred (2 h) in acetonitrile (anhydrous, 40 mL). Sodium boro-
hydride (1.15 g, 30.3 mmol) was added and the reaction mixture was
stirred for an additional hour. Sodium hydroxide solution (2%, 30 mL)
was added and after 1 h the pH was adjusted to 2 with hydro-
chloric acid, and the solution was then extracted with diethyl ether to
remove neutrals. The aqueous layer was adjusted to pH 14 (2% NaOH
solution) and the product extracted into diethyl ether. The resultant
colourless oil (1.00 g) comprised compound 34 and the bicycle 35
(90 : 10 by GCMS). Trituration of the oil with small volumes of pentane
(with the aid of an ultrasonic cleaning bath) removed all traces of the
minor isomer and left N-ethyl-1-methyl-1-(4-methylcyclohex-3-en-1-
yl)ethylamine [N-ethyl-8-amino-p-menth-1-ene] 34 (46% yield), >99%
pure by GCMS and NMR. A distilled sample (bp 79◦C/0.8 mmHg)
appeared to have attracted carbon dioxide from the atmosphere and
microanalysis showed approximately 0.2 mol equivalents of the bicar-
bonate salt (Found: C 75.6, H 12.3, N 7.4. C12H23N requires C 79.5,
H 12.8, N 7.7%. C12H23N + 0.2(H2CO3) requires C 75.6, H 12.2,
N 7.2%) (Found: m/z 182.1904, 181.1818. C12H24N (M + 1) requires
182.1908, C12H23N requires 181.1830). δH ([D6]benzene) 5.45 (m,
Wh/2 11, H2), 2.44 (ddd, J 14.2, 7.1, 7.1, 1 × H11), 2.41 (ddd, J 14.2,
7.1, 7.1, 1 × H11), 1.81–2.00 (4H, m, 2 × H3 and 2 × H6), 1.78 (dm,
width 28 Hz, H5eq), 1.65 (3H, br s, H7), 1.46 (dddd, J 12.2, 11.1, 4.9,
2.5, H4), 1.16 (dddd, J 12.2, 12.2, 12.2, 5.5, H5ax), 0.99 (3H, t, J11,12
7.1, H12), 0.93 and 0.92 (2 × 3H, s, H9 and H10). δC: see Table 1. vmax
(Nujol)/cm−1 3315, 1249, 1221, 1183, 1164, 1107, 1021, 916. m/z
(GCMS) 182 (M + 1, 0.1%), 181 (M, 0.3), 166 (1), 136 (1), 98 (5), 87
(6), 86 (100).
Amine 5d (0.88 g, 3.62 mmol) and mercuric acetate (1.16 g, 3.64 mmol)
were stirred (24 h) in dichloromethane (20 mL). Workup by cooling
and addition to a chilled solution of sodium borohydride (0.57 g,
15 mmol) in aqueous sodium hydroxide (1 m, 100 mL), as described
above for compound 18, followed by flash chromatography
(hexane/diethyl ether, 95 : 5) provided (1R,4R,5R)-6-benzyl-4,7,7-
trimethyl-6-azabicyclo[3.2.1]oct-4-yl acetate [(1R,2R,4R)-N-benzyl-
2,8-azacineole-1-acetate] 19 as a colourless viscous oil (0.95 g, 87%)
(Found: C 75.3, H 9.2, N 4.1. C19H27NO2 requires C 75.7, H 9.0,
N 4.6%). δH (CD2Cl2) 7.37–7.18 (five aromatic H), 3.96 and 3.66 (ABq,
JAB −13.4, two benzylic H), 3.57 (H2), 1.99 (H3β), 1.92 (acetate Me),
1.87 (H6α) 1.76 (H4), 1.67 (H6β), 1.66 (H5β), 1.65 (H3α), 1.58 (H5α),
1.19 and 1.05 (2 × 3H, s, H9 and H10), 0.95 (3H, H7); with J2,3α ≈ 0,
J2,3β 6.8, J3α,3β −11.0, J3α,4 ≈ 0, J3β,4 4.7, J3β,5β 2.2, J4,5α 3.0, J4,5β
2.2, J5α,5β −13.6, J5α,6α 13.6, J5α,6β 6.2, J5β,6α 6.9, J5β,6β unavailable,
and J6α,6β −14.0. δC 170.6 (C11), 142.8 (C14), 129.9 (C15 and C19);
128.3 (C16 and C18), 126.9 (C17), 85.7 (C1), 67.9 (C2), 64.1 (C8), 55.3
(C13), 46.1 (C4), 32.9 (C6), 30.4 (C3), 25.7 and 24.2 (C9 and C10), 24.8
(C5), 22.5 (C12), 22.4 (C7). vmax (neat)/cm−1 3026, 2972, 2926, 1731,
1495, 1455, 1366, 1301, 1258, 1209, 1154, 1114, 1087, 1022, 932, 828,
734, 700, 610. m/z (GCMS) 302 (M + 1, 3%), 301 (M, 10), 286 (6),
258 (7), 242 (6), 241 (4), 227 (7), 226 (40), 187 (5), 186 (30), 184 (5),
92 (8), 91 (100).
Alcohol 22
Acetate 19 was reduced (24 h, room temp.) with lithium aluminium
hydride in diethyl ether. Normal workup gave crude (1R,4R,5R)-6-
benzyl-4,7,7-trimethyl-6-azabicyclo[3.2.1]octan-4-ol [(1R,2R,4R)-N-
benzyl-1-hydroxy-2,8-azacineole] 22 (96% yield) which was sublimed
under vacuum to give small colourless chunks, mp 90–91◦C (Found:
C 79.1, H 10.1, N 5.1. C17H25NO requires C 78.7, H 9.7, N 5.4%). δH
7.32–7.16 (five aromatic H), 3.94 and 3.62 (ABq, JAB −13.2, 2 × H11),
2.70 (H2), 1.96 (H3α), 1.93 (H3β), 1.90 (H6α), 1.73 (H4), 1.64 (H5β),
1.60 (H5α), 1.23 (H6β), 1.16 and 1.05 (2 × 3H, s, H9 and H10), 0.72
(3H, H7); with J2,3α ≈ 0, J2,3β 5.0, J3α,3β −11.5, J3α,4 ≈ 0, J3β,4 4.5,
J3β,5β unavailable, J4,5α 3.1, J4,5β 2.5, J5α,5β −13.8, J5α,6α 13.8, J5α,6β
5.7, J5β,6α 7.9, J5β,6β small, and J6α,6β −13.8. δC 142.3 (C12), 129.5
(C13 and C17), 127.9 (C14 and C16), 126.6 (C15), 73.9 (C1), 70.5 (C2),
63.4 (C8), 55.0 (C11), 45.9 (C4), 34.0 (C6), 29.9 (C3), 28.1 (C7), 25.6
and 24.7 (C9 and C10), 24.6 (C5). vmax (Nujol)/cm−1 3329, 1298, 1251,
1221, 1163, 1091, 1070, 1025, 973, 910, 750, 720, 697. m/z (GCMS)
260 (M + 1, 2%), 259 (M, 10), 245 (6), 244 (30), 226 (10), 187 (9), 186
(62), 184 (11), 110 (6), 92 (9), 91 (100), 83 (12).
Amine 35
β-Pinene 30 (1.40 g, 10.1 mmol) and NBS (2.10 g, 11.8 mmol) were
vigorously stirred (2 h) in acetonitrile (anhydrous, 40 mL) exactly as
described above in the synthesis of compound 34, except that now
the reaction was quenched with excess sodium borohydride (191 mg,
5.05 mmol). The product, a colourless oil (0.95 g), was a mixture
of compounds 32 and 31 (85 : 15 by GCMS). Careful trituration of
the oil with small volumes of pentane (with the aid of an ultra-
sonic cleaning bath) extracted compound 35 from compound 34.
6-Ethyl-7,7-dimethyl-4-methylene-6-azabicyclo[3.2.1]octane [N-ethyl-
1,7-dehydro-2,8-azacineole] 35, a colourless volatile oil (41% yield),
was distilled (bp 45◦C/0.1 mmHg) to provide a sample >98% pure (by
GC and NMR) (Found: m/z 164.1440. C11H18N1 (M − CH3) requires
164.1439). δH ([D6]benzene;p-menthaneskeletalnumbering)4.60(2H,
br s, H7), 3.52 (d, H2), 2.56 and 2.52 (2 × dq, 2 × H11), 2.39 (ddd, H6α),
2.18 (ddd, H3β), 2.08 (br dd, H6β), 1.68 (br dddd, H5β), 1.61 (dddd, H4),
1.36 (dddd, H5α), 1.27 (dd, H3α), 1.08 (3H, t, H12), 1.05 and 1.04
(2 × 3H, s, H9 and H10); with J2,3α ≈ 0, J2,3β 5.9, J3α,3β −11.1, J3α,4
0.7, J3β,4 4.6, J3β,5β 2.7, J4,5α 3.2, J4,5β 2.9, J5α,5β −13.4, J5α,6α 12.3,
Compound 33
β-Pinene 30 and NBS in anhydrous acetonitrile afforded crude allylic
amide 33 by the literature method.[24] This material was used in a crude

Reaction of 8-Amino-p-menthene Derivatives with Electrophiles
591
J5α,6β 7.4, J5β,6α 8.1, J5β,6β ≈ 0, J6α,6β −15.4, J2,11a < 1, J2,11b < 1,
J11a,11b −12.6, and J11,12 7.1. δC ([D6]benzene) 152.4 (C1), 104.6
(C7), 65.5 (C2), 61.1 (C8), 45.4 (C4), 39.5 (C11), 36.9 (C6), 30.9 and
20.9 (C9 and C10), 28.8 (C3), 28.3 (C5), 16.8 (C12). vmax (neat)/cm−1
3064, 2962, 2099, 1776, 1645, 1454, 1381, 1262, 1185, 1116, 884. m/z
(GCMS) 180 (M + 1, 1%), 179 (M, 7), 165 (13), 164 (100), 136 (18),
124 (6), 122 (7), 119 (8), 93 (8), 91 (11), 79 (10), 77 (10).
[5] P. V. Bernhardt, R. M. Carman, B. P. Ross, Aust. J. Chem. 2000,
53, 611. doi:10.1071/CH00022
[6] J.-P. Praly, C. D. Stafano, G. Descotes, R. Faure, L. Somsak,
I. Eperjesi, Tetrahedron Lett. 1995, 36, 3329. doi:10.1016/0040-
4039(95)00520-M
[7] G. Capozzi, Heterocycles 1987, 26, 39.
[8] A. Avenosa, C. Cativiela, J. M. Peregrina, Tetrahedron 1994,
50, 10 021. doi:10.1016/S0040-4020(01)89617-4
[9] J. J. P. Stewart, J. Comput. Chem. 1989, 10, 209.
[10] J. J. P. Stewart, J. Comput. Chem. 1989, 10, 221.
[11] J. P. Yardley, R. W. Rees, Can. J. Chem. 1985, 63, 1013.
[12] R. M. Carman, M. T. Fletcher, Aust. J. Chem. 1983, 36, 1483.
[13] R. M. Carman, M. T. Fletcher, Aust. J. Chem. 1984, 37, 1117.
[14] R. M. Carman, K. D. Klika, Aust. J. Chem. 1991, 44, 1803.
[15] K. B. Wiberg, S. D. Nielsen, J. Org. Chem. 1964, 29, 3353.
[16] Z. Rappoport, P. D. Sleezer, S. Winstein, W. G. Young,
Tetrahedron Lett. 1965, 6, 3719. doi:10.1016/S0040-4039(01)
99552-8
Crystallography
Compound 9 (Fig. 1): C17H22BrNO, M 336.27, orthorhombic, space
group P212121 (no. 19), a 10.106(2), b 10.228(2), c 15.384(3) Å,
V 1590.2(5) Å3, F(000) 696, Dc 1.405 g cm−3, µ 25.81 cm−1, 2136
unique data (2θmax 50◦), R 0.0380 (for 1488 reflections with I > 2σ
(I)), wR2 0.1154 (all data).
Compound 17 (Fig. 2): C17H29NO4, M 311.41, monoclinic,
space group P21 (no. 4), a 10.010(3), b 9.3244(6), c 10.395(4) Å,
β 110.40(2)◦, V 909.4(4) Å3, F(000) 340, Dc(Z = 2) 1.137 g cm−3
,
µ 0.80 cm−1, 1910 unique data (2θmax 50◦), R 0.0373 (for 1318
reflections with I > 2σ(I)), wR2 0.1089 (all data).
[17] Z. Rappoport, S. Winstein, W. G. Young, J. Am. Chem. Soc.
1972, 94, 2320.
[18] W. Kitching, T. Sakakiyama, Z. Rappoport, P. D. Sleezer,
S. Winstein, W. G. Young, J. Am. Chem. Soc. 1972, 94, 2329.
[19] P. L. Ruddock, P. B. Reese, J. Chem. Res., Synop. 1994, 442.
[20] J. Wolinsky, J. H. Thorstenson, M. K. Vogel, J. Org. Chem.
1978, 43, 740.
[21] R. M. Carman, A. C. Garner, K. D. Klika, Aust. J. Chem. 1994,
47, 1509.
[22] R. M. Carman, A. C. Rayner, Aust. J. Chem. 1994, 47, 2087.
[23] K. Kwetkat, W. Kitching, J. Chem. Soc., Chem. Commun.
1994, 345.
[24] D. J. Brecknell, R. M. Carman, R. A. Edwards, K. A. Hansford,
T. Karoli, W. T. Robinson, Aust. J. Chem. 1997, 50, 689. doi:
10.1071/CH96188
[25] A. Pancrazi, I. Kabore, Q. Khong-Huu, Bull. Soc. Chim. Fr.
1977, 162.
[26] S. S. Koval’skaya, N. G. Koslov, G. V. Kalechits, Zh. Org. Khim.
1991, 27, 757.
[27] L. A. Popova, N. G. Koslov, Khim. Prir. Soedin. (Engl. Transl.)
1987, 235.
Compound 18 (Fig. 4): C10H21NO2, M 187.28, monoclinic, space
groupP21/c (no. 14), a6.116(2), b20.585(4), c 8.796(4) Å, β 94.99(2)◦,
V 1103.2(7) Å3, F(000) 416, Dc(Z = 4) 1.128 g cm−3, µ 0.77 cm−1
,
1943 unique data (2θmax 50◦), R 0.0673 (for 1038 reflections with
I > 2σ(I)), wR2 0.1820 (all data).
Intensity data at 296 K were collected on an Enraf–Nonius CAD4
four-circle diffractometer using graphite monochromated MoKα radi-
ation (λ 0.71073 Å) in the ω–2θ scan mode. Lattice dimensions were
determined by a least-squares fit of the setting parameters of 25 indepen-
dent reflections. Data reduction and empirical absorption corrections
(ψ-scans) were performed with theWINGX package.[32] Structures were
solved by direct methods with SHELXS and refined by full-matrix least-
squares analysis with SHELXL97.[33] All non-hydrogen atoms were
refined with anisotropic thermal parameters, and hydrogen atoms were
constrained at estimated positions using a riding model. Water hydrogen
atoms were first located from difference maps then restrained in these
positions. The atomic nomenclature is defined in Figs 1, 2, and 4 drawn
with ORTEP3.[34] The unit cell diagram (Fig. 3) was produced with the
program PLUTON.[35] Crystallographic data in CIF format are available
from the Cambridge Crystallographic Data Base (depositions: 205668,
compound 9; 205669, compound 17; and 205670, compound 18).
[28] K. Shishido, K. Hiroya, K. Fukumoto, T. Kametani, J. Chem.
Res., Synop. 1989, 100.
[29] H. L. Vaughn, M. D. Robbins, J. Org. Chem. 1975, 40, 1189.
[30] R. E. Banks, N. J. Lawrence, M. K. Besheesh, A. L. Popplewell,
R. G. Pritchard, Chem. Commun. 1996, 1629.
[31] K. Soai, M. Ishizaki, J. Org. Chem. 1986, 51, 3290.
[32] L. J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837. doi:10.1107/
S0021889899006020
References
[1] R. M. Carman, R. P. C. Derbyshire, Aust. J. Chem. 2003, 56,
319. doi:10.1071/CH02189
[2] S. Robin, G. Rousseau, Tetrahedron 1998, 54, 13 681.
doi:10.1016/S0040-4020(98)00698-X
[33] G. M. Sheldrick, SHELX97 Release 97-2 1998 (Universität
Göttingen: Göttingen).
[3] R. M. Carman, B. N. Venzke, Aust. J. Chem. 1971, 24, 1727.
[4] R. M. Carman, C. H. L. Kennard, W. T. Robinson, G. Smith,
B. N. Venzke, Aust. J. Chem. 1986, 39, 2165.
[34] L. J. Farrugia, J. Appl. Crystallogr. 1997, 30, 565. doi:10.1107/
S0021889897003117
[35] A. L. Spek, Acta Crystallogr. 1990, A46, C34.