Synthesis of symmetric dinitro-functionalised troeger's base analogues

Article abstract of DOI:10.1002/ejoc.200801032

The synthesis of six new examples of 2,8-dinitro-substituted Troeger's base analogues are reported, together with the first examples of 1,7-, 3,9- and 4,10-dinitro Troeger's base analogues and the first example of a tetranitro Troeger's base compound. Several of these dinitro compounds lack substituents at the 2- and 8-positions and therefore provide further examples of Troeger's base analogues derived from anilines lacking a para substituent.

Full text of DOI:10.1002/ejoc.200801032

FULL PAPER
DOI: 10.1002/ejoc.200801032
Synthesis of Symmetric Dinitro-Functionalised Tröger’s Base Analogues
M. Delower H. Bhuiyan,[a] Andrew B. Mahon,[a] Paul Jensen,[b] Jack K. Clegg,[b] and
Andrew C. Try*[a]
Keywords: Tröger’s base / Chirality / Aromatic substitution / Nitroanilines
The synthesis of six new examples of 2,8-dinitro-substituted
Tröger’s base analogues are reported, together with the first
ents at the 2- and 8-positions and therefore provide further
examples of Tröger’s base analogues derived from anilines
examples of 1,7-, 3,9- and 4,10-dinitro Tröger’s base ana- lacking a para substituent.
logues and the first example of a tetranitro Tröger’s base
compound. Several of these dinitro compounds lack substitu-
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
group. Tröger’s base compounds bearing a nitro group on
one of the aromatic rings have been prepared, but their syn-
thesis avoided the difficult bond-forming process by having
it in place in one of the starting materials.[6,7] Only two dini-
tro-substituted Tröger’s base compounds have previously
been prepared. The first of these was 2,8-dinitro Tröger’s
base 2, that was synthesised using two different methodolo-
gies. The initial report described the synthesis of 2 in 56%
yield as an unexpected product of a reaction aimed at pro-
ducing a cyclic imide 4, that was not observed at the time.
The reaction utilised 4-nitroaniline, diglycolic acid and pol-
yphosphoric acid (PPA) as the reagents.[8] In the second re-
port, 2 was obtained in 23% yield from the reaction of 4-
nitroaniline, hydrogen chloride and DMSO.[9] The synthesis
of a second dinitro analogue, 4,10-dimethyl-2,8-dinitro
Tröger’s base 3, in 80% yield, was reported more recently
from a reaction of 2-methyl-4-nitroaniline and paraformal-
dehyde in trifluoroacetic acid (TFA).[10]
Introduction
Tröger’s base (1, Figure 1) is a chiral, concave-shaped
molecule that is prepared by the acid-catalysed condensa-
tion of p-toluidine and formaldehyde.[1] It was the first com-
pound whose chirality is due solely to asymmetric nitrogen
atoms to be resolved and additionally was probably the first
compound to be resolved with the aid of a chiral stationary
phase.[2]
Figure 1. Chemical structure of Tröger’s base 1 and the conven-
tional numbering system.
Numerous analogues of 1 have been prepared, however
until very recently the type of substitution available on the
benzene rings was limited by the belief that electron-with-
drawing groups on the starting aniline units were poorly
tolerated, if at all. The synthesis of dihalogenated Tröger’s
base compounds[3–5] provided inspiration that other elec-
tron-withdrawing substitutents may also be incorporated on
aniline precursors.
Results and Discussion
In light of the synthetic utility of the nitro group, we
chose to further investigate the synthesis of dinitro Tröger’s
The synthesis of Tröger’s base analogues bearing the
more strongly electron-withdrawing nitro group has re-
mained a challenge due to the fact that it is difficult to form
the necessary aryl-methylene bond on a ring bearing a nitro
[a] Department of Chemistry and Biomolecular Sciences, Macqua-
rie University,
North Ryde, NSW 2109, Australia
Fax: +61-98508313
E-mail: andrew.try@mq.edu.au
[b] Crystal Structure Analysis Facility, School of Chemistry, The
University of Sydney,
Sydney, NSW 2006, Australia
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Figure 2. The four possible symmetric dinitro Tröger’s base ana-
logues.
Eur. J. Org. Chem. 2009, 687–698
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
687

M. D. H. Bhuiyan, A. B. Mahon, P. Jensen, J. K. Clegg, A. C. Try
FULL PAPER
base analogues and 4 types of symmetrically substituted the cyclic imide 4 that co-eluted with the desired Tröger’s
isomers were targeted as depicted in Figure 2.
base 2, as confirmed by examination of 1H NMR spectrum
of the crude product and H NMR spectra obtained after
1
repeated attempts of separation. The ratio of 2,8-dinitro
Tröger’s base 2/cyclic imide 4 was found to be 86:14 from
integration of signals in a 1H NMR spectrum after
chromatography, which was essentially the same product
distribution observed when Method A was employed
(83:17). In the case of Method C, unreacted 4-nitroaniline
was also recovered in a 21% yield.
The fact that very similar product distributions were ob-
served with both diglycolic acid (Method A) and glycolic
acid anhydride (Method C) suggests that the reaction pro-
ceeds via the anhydride (that in the case of diglycolic acid
could be formed from a dehydration reaction with polyp-
hosphoric acid). Method C was not employed with any
other nitro anilines.
The use of paraformaldehyde and trifluoroacectic acid
(TFA) (Method D, a reagent mixture that proved to be suc-
cessful in the synthesis of 3[10]) lead to the recovery of some
unreacted 4-nitroaniline, together with significant amounts
of intractable material. A similar result was obtained when
PPA was used in place of TFA. There was no trace of 2
present in the 1H NMR spectrum of the crude material ob-
tained after work-up in either case.
Reagent Investigation
As a starting point, several reagent combinations with 4-
nitroaniline were examined in attempts to form 2. The first
of these was a reproduction of the published methodology,
in which a mixture of 4-nitroaniline, polyphosphoric acid
and diglycolic acid were mixed together and heated at 80 °C
for 12 h; this will be referred to as Method A (Scheme 1).[8]
However, a yield of greater than 22% was never achieved
and the cyclic imide, 4-(4Ј-nitrophenyl)morpholine-3,5-di-
one (4) was always present as a co-eluting and inseparable
product after all attempted chromatographic separations.
No mention of 4 was made in the literature, although it was
the anticipated product.
In light of these results in the synthesis of 2, and those
obtained by other researchers in the synthesis of 3, a study
on the generality of the reaction was conducted where two
types of reagents and reactions conditions were used; in the
Scheme 1. Synthesis of 2,8-dinitro Tröger’s base 2 and morpholine
dione 4.
A slight modification to the published procedure (poly- first, diglycolic acid was employed as the formaldehyde
phosphoric acid and diglycolic acid were stirred together at equivalent in PPA at 80 °C and in the second set of condi-
80 °C for 2 h prior to the addition of 4-nitroaniline, which tions, paraformaldehyde was used with TFA at room tem-
will be referred to as Method B) resulted in the formation perature.
of 2 in 28% yield. Whilst this yield was again substantially
lower than the reported yield of 56%, the crude sample pro-
duced after work-up was devoid of 4 when this methodol-
ogy was employed.
Type II Compounds (2,8-Dinitro)
The initial point of variation in the nitroanilines was
based on Type II compounds and involved the incorpora-
tion substituents at the 2-position. It was anticipated that
the substituents would increase the yield of Tröger’s base
products by either enhancing the electron-richness of the
aniline, or increasing the solubility of the products, thereby
facilating their isolation and purification (Scheme 2,
Table 1).
It was subsequently found that for a variety of nitroani-
lines examined (see below) essentially the same yields were
obtained using either Method A or B, however the crude
samples obtained after work-up were easier to purify, and
generally free of any cyclic imide products, if Method B was
employed. It was also found that with some of the nitroani-
lines investigated, cyclic imides were not produced regardless
of whether Method A or Method B was employed. When
the reaction was conducted at 50 °C over the same time
period only unreacted starting material was recovered. In
contrast, when the reaction was performed at 70 °C the de-
sired product was obtained, although in lower yields than
reactions performed at 80 °C, and unreacted starting mate-
rial was also recovered. The use of temperatures greater
than 80 °C resulted in the formation of polymeric material
which made the isolation of the desired products more diffi-
cult.
Scheme 2. Synthesis of Type II compounds with various substitu-
tion patterns.
In order to learn more about the mechanism of the reac-
As mentioned previously, 2 was obtained in a 28% yield
tion, glycolic anhydride was employed in place of diglycolic in our hands. The presence of a substituent ortho to the
acid. The use of 1.5 equiv. of diglycolic anhydride in poly- amino group afforded several new 2,8-dinitro Tröger’s base
phosphoric acid (Method C) resulted in the formation of compounds, however the substituent had a negligible effect
688
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 687–698

Symmetric Dinitro Tröger’s Base Analogues
Table 1. Yields of the 2,8-dinitro Tröger’s base compounds 2, 3, 5
and 7–11.
R1
R2
R3
Yield (%)[a]
Method
2
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
28
B
D
B
D
A
B
D
B
D
B
D
B
D
A
D
B
2
trace[b]
34
3
CH3
CH3
OCH3
OCH3
OCH3
Br
3
78
5
20[c]
27
5
5
68
7
28
[d]
7
H
Br
–
8
CH3
CH3
Br
Br
CH3
CH3
Br
Br
CH3
CH3
CH3
CH3
H
H
H
16
92
23
Figure 3. An ORTEP diagram of 5 with 50% probability ellipsoids.
8
9
9
71
13[e]
14[f]
14
As was observed in the case of 3, 5 and 7, the presence
of substituents on compounds 8–11 increased their solu-
bility in organic solvents in comparison with 2, without in-
creasing the yields of the desired products when Methods
A or B were used, whilst yield of 8 and 9 were dramatically
enhanced with Method D.
10
10
11
11
[d]
H
–
D
[a] Isolated yields after column chromatography. [b] Only a trace
of a Tröger’s base product was observed in the H NMR spectrum
of the crude material after work-up. [c] When Method A was em-
ployed, a 34% yield of the 2Ј-methoxy analogue of 4, compound
6, was also obtained. [d] No Tröger’s base product was detected in
1
Compounds 10 and 11 were the products of anilines that
could each theoretically afford three different Type II
the 1H NMR spectrum of the crude material after work-up. [e] Tröger’s base products as they have two inequivalent posi-
Small amounts of the other two isomers were evident in a 1H NMR
tions ortho to the amino group (due to the presence of a
spectrum of the crude material. [f] The combined yield of an in-
substituent at the R1 site in Scheme 2). Indeed, all three
separable mixture of all 3 possible isomers.
isomers were formed, however the use of Method A led to
the formation of 10 as the major product which facilitated
its purification. In contrast, Method D resulted in the for-
mation of all three isomers in approximately equal amounts
on yields when either Method A or B was used. It is note-
and the Tröger’s base products proved to be inseparable by
worthy that the solubility of 3, 5 and 7 was considerably
either chromatography or recrystallisation, although the lat-
improved over that of 2 and in this regard the incorporation
ter technique did lead to partial separation with consider-
of the substituents exhibited the desired effect. The use of
able loss of material. As was the case in the attempted syn-
method D results in a considerable improvement in the
thesis of 7 using Method D, no trace of 11 was observed
yields of 3 and 5, with methyl and methoxy groups, respec-
using these conditions (the aniline employed in these two
tively, ortho to the amine. However, the inclusion of a
reactions are isomers of one another).
bromo group at this position failed to afford any trace of
the desired product, Tröger’s base 7, when Method D was
employed.
Type I Compounds (1,7-Dinitro)
When Method A was employed in the synthesis of 5, 4-
The reaction of m-nitroanilines was then examined
(Scheme 3, Table 2). In order to investigate the effect of the
presence of other substituents on the aromatic ring, four
different types of substitution patterns were studied; those
with ortho and para substituents, those with an ortho sub-
stituent only, those with a para substituent and lacking an
ortho substituent, and finally 3-nitroaniline. No single
method stands out as being superior for this group of ani-
lines.
(2Ј-methoxy-4Ј-nitrophenyl)morpholine-3,5-dione
6
was
also isolated in a 34% yield,[11] however this product was
1
not observed in a H NMR spectrum of the crude material
after work-up when Method B was employed.
An X-ray crystal structure of 5 is shown in Figure 3. The
dihedral angle between the least-squares planes of the two
aryl rings in this molecule is 103.4°, at the upper end of the
range of 82[12] – 110°[13] that has been reported for a wide
variety of simple dibenzo Tröger’s base analogues.
Scheme 3. Synthesis of Type I compounds with various substitu-
tion patterns.
Eur. J. Org. Chem. 2009, 687–698
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
689

M. D. H. Bhuiyan, A. B. Mahon, P. Jensen, J. K. Clegg, A. C. Try
FULL PAPER
Table 2. Yields of the 1,7-dinitro Tröger’s base compounds 12–24.
R1
R2
R3
Yield (%)[a] Method
12
12
13
13
14
14
15
15
16
16
17
17
18
18
19
19
20
20
21
21
22
22
23
23
24
24
CH3
CH3
Br
Br
CH3
CH3
H
H
H
H
H
H
H
H
CH3
CH3
CH(CH3)2
CH(CH3)2
CH2(CH3)2CH3
CH2(CH3)2CH3
Br
Br
Br
Br
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
CH3
CH3
H
H
CH3
CH3
CH3
CH3
Br
Br
CH3
CH3
CH(CH3)2 23
CH(CH3)2 34
OCH3
OCH3
Br
Br
H
H
H
H
H
H
H
H
H
51
45
46
45
B
D
A
D
A
D
A
D
A
D
A
D
A
D
B
D
A
D
B
D
B
12
trace[b]
6
56
Figure 4. An ORTEP diagram of 15 with 50% probability ellip-
5
soids.
22
10
[c]
–
32[d]
Compounds 19–24 were the products of anilines that
could each theoretically afford three different Tröger’s base
products (Type I, Type III and a Type I/III hybrid) as they
have two inequivalent positions ortho to the amino group
(i.e., R3 = H in Scheme 3). The anticipated different reactiv-
ities of the two vacant ortho sites is expected to be greatest
in the case of the anilines used to produce compounds 19–
22 and 24, where one of the vacant ortho sites is more steri-
cally hindered than the other, as illustrated in Figure 5.
[c]
–
8
45
18
trace[b]
6
[c]
–
D
A
D
B
18
H
H
H
trace[b]
22[e]
[c]
H
–
D
[a] Isolated yields after column chromatography. [b] Only a trace
of the symmetric Type I Tröger’s base product was observed in the
1H NMR spectrum of the crude material after work-up. [c] No
Tröger’s base product was detected in the 1H NMR spectrum of
the crude material after work-up. [d] Hybrid 25 was also isolated
in 9% yield. [e] Hybrid 26 was also isolated in 5% yield.
Figure 5. The two possible sites for diazocine bridge formation.
Tröger’s base compounds 12 and 13 were formed in ac-
ceptable yields. They were the result of reactions of anilines
that bear a substituent at the para position and that could
only afford a single Tröger’s base isomer as they had only
one vacant site ortho to the amino group. The yield of 14
was quite low, however the reason for this is unknown.
Four reactions involving 2-substituted-5-nitroanilines af-
forded Tröger’s base compounds 15–18 in poor to moderate
yields. In addition to halogen compounds,[5] the molecules
provide further examples of the successful synthesis of
Tröger’s base analogues from anilines lacking a substituent
in the p-position. A factor that may contribute to the poor
yields in these cases is the possibility that the formaldehyde
equivalent may condense at the p-position of such anilines
and then undergo subsequent reactions with amino groups
to afford non-Tröger’s base products. Indeed it was this pos-
sibility that gave rise to the belief that a substituent at the
para-position was a necessity if Tröger’s base compounds
were to be obtained.
Anilines where a halogen atom is present in the place of
the nitro group and R1 = H or CH3 have been reported
to afford all three possible Tröger’s base isomers (with the
exception of when the halogen is a fluorine atom).[5] In an
earlier report, 3,4-dimethylaniline was observed to form a
Type III compound together with a hybrid, in a 72:28 ratio,
with no evidence of a Type I compound.[14]
In these reactions, no Type III products where observed,
despite the fact that they would be the product of a reaction
at the less hindered ortho position. Infact, Type I com-
pounds were observed as the exclusive Tröger’s base prod-
ucts, with the exception of two anilines; 4-methyl-3-nitro-
aniline and 3-nitroaniline, which also afforded hybrids 25
and 26 in 9% and 5% yields, respectively, when Method B
was employed. No Tröger’s base products were formed
when these two anilines where used in Method D.
It is apparent from inspection of Table 2 that the yields
of 15, and to a lesser extent 17, are dramatically enhanced
with the use of Method D, but this method is not consist-
ently superior to Method A, even within this group of nitro-
anilines bearing the same pattern of substitution.
The structure of 15 was confirmed by X-ray crystallogra-
The identification of 25 and 26 as hybrid systems was
phy (Figure 4). The dihedral angle between the least- possible based solely upon examination of 1H NMR spectra
squares planes of the two aryl rings in this compound is as the bridging region of both compounds lacked the char-
99.1°.
acteristic splitting pattern that reflects the C2-symmetry
690
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 687–698

Symmetric Dinitro Tröger’s Base Analogues
present in all symmetrically substituted Tröger’s base ana- 2-nitroaniline and 4,5-dimethyl-2-nitroaniline, that afforded
logues. The structure of 25 was also confirmed by X-ray 29 (Method B) and 30 (Method A) in 36% and 8% yields,
crystallography.[15]
respectively, whilst the use of Method D conditions was un-
Interestingly, the use of 4-bromo-5-methyl-3-nitroaniline, successful with both anilines. In contrast, the use of 2-nitro-
an aniline with two vacant ortho sites, and two inequiva- aniline failed to result in the formation of a Tröger’s base
lently substituted meta sites, afforded a single Tröger’s base product under any of the examined conditions.
1
product 23, as evident from examination of the H NMR
spectrum of the crude material after work-up when Method
A was employed. The structural assignment of this com-
pound was only possible with the aid of two-dimensional
NMR experiments, which revealed the presence of an nOe
between the methyl group and an adjacent aryl proton (Fig-
ure 6) (see Supporting Information), as splitting patterns in
the aromatic region could not be used to provide an unam-
biguous assignment.
A Tetranitro Compound (1,3,7,9-Tetranitro)
In light of the success achieved with various nitroanil-
ines, a reaction was carried out with a dinitroaniline, 4-
methoxy-2-methyl-3,5-dinitroaniline, to yield the tetranitro
Tröger’s base analogue 31 in 11% yield with the use of
Method B, whilst Method D afforded 31 in a 4% yield. The
structure of 31 was confirmed by X-ray crystallography.[16]
This is yet another example of a Tröger’s base compound
that is multiply-substituted with electron-withdrawing sub-
stituents, the others being analogues with substituent pat-
terns like 4,10-dichloro-1,7-dimethoxy-2,8-dimethyl ester,[17]
4,10-dibromo-2,8-diethyl ester,[18] tetrahalo,[9,13,19] and oc-
tafluoro[20]. In the present case, it is thought that the elec-
tron-withdrawing nature of the two nitro groups in the ani-
line should be offset somewhat by the methyl and methoxy
groups that are also expected to aid in the solubility of 31.
Figure 6. The observed nOe that was use to confirm the structure
of 23.
The observed nOe would not be possible if the diazocine
bridge were formed at the site ortho to the methyl group as
this would result in the aryl proton in a Tröger’s base prod-
uct that was positioned para to the methyl group.
Type III Compound (3,9-Dinitro)
A single example of a Type III compound was prepared
from 2-methyl-3-nitroaniline. Unlike many of the 3-nitro-
anilines used in the synthesis of Type II compounds, 2-
methyl-3-nitroaniline could only afford a Type III Tröger’s
base analogue as the position ortho to both the amino and
nitro groups was not available for substitution. Compound
27 was prepared in 33% yield, together with another exam-
ple of an N-phenylmorpholine dione, 28 (cf. 4 and 6). In
this instance the two compounds were readily separable by
column chromatography. The use of Method D resulted in
the formation of 27 in 43% yield.
Amino Tröger’s Base Analogues
Three amino-substituted dibenzo analogues of Tröger’s
base have recently been reported; two of these were ob-
tained from a two-step reaction involving catalytic amin-
ation of preformed halogenated Tröger’s base compounds
and subsequent hydrolysis of the resultant imine[21] and the
third diamine was obtained from reduction of dinitro
Tröger’s base 3 with iron powder in acetic acid and etha-
nol.[10] We sought access to diamino Tröger’s base com-
pounds via reduction of the nitro groups using two different
reagent combinations, in order to establish the feasibility of
this general approach to diamino Tröger’s base analogues.
Type IV Compounds (4,10-Dinitro)
The use of 2-nitroanilines in the Tröger’s base forming Toward this end, compounds 2 and 12 were subjected to
reaction should theoretically result in the formation of Type two different sets of reduction conditions (Schemes 4 and
IV compounds. This was found to be the case for 4-methyl- 5).
Eur. J. Org. Chem. 2009, 687–698
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
691

M. D. H. Bhuiyan, A. B. Mahon, P. Jensen, J. K. Clegg, A. C. Try
FULL PAPER
Solvents and reagents were purified using standard techniques (see
the Supporting Information for the preparation of several anilines
used as starting materials in the work). All commercial solvents
were routinely distilled prior to use. Hexane refers to the fraction
of b.p. 60–80 °C. Where solvent mixtures are used, the portions are
given by volume.
Scheme 4. Synthesis of a Type II diamine.
Column chromatography was routinely carried out using the grav-
ity feed column techniques on Merck silica gel type 9385 (230–
400 mesh) with the stated solvent systems. Analytical thin-layer
chromatography (TLC) analyses were performed on precoated
plates (Merck aluminium sheets, silica gel 60F254, 0.2 mm). Visual-
isation of compounds was achieved by illumination under ultravio-
let light (254 nm).
Scheme 5. Synthesis of a Type I diamine.
Continuous-flow hydrogenation was carried out on a H-Cube
(ThalesNano), using a 30 mm 10% Pd on C catalyst cartridge, at
the conditions specified in the relevant sections.
Compound 32 was produced in 92% yield from the hy-
drogenolysis reaction, however no Tröger’s base material
was recovered from the reaction involving tin(II) chloride.
We believe that the reduction did take place, however we
could not extract 32 from the aqueous environment during
the work-up procedure.
In contrast, hydrogenolysis of 12 failed, presumably as a
result of the increased steric hindrance around the nitro
groups. 33 was obtained in 55% yield from the tin(II) chlo-
ride reduction. In this case, the hydrophilicity of the di-
amine is reduced by the presence of the methyl groups.
Both 2 and 12 were also subjected to continuous-flow
hydrogenation, (see Exp. Sect. for complete details) to af-
ford 32 and 33 in 85% and 97% yield, respectively, in dras-
tically reduced reaction times.
General Procedures
Method A: The nitroaniline (14.5 mmol), diglycolic acid (3.06 g,
23.3 mmol) and polyphosphoric acid (86%, d = 1.90 g/cm3, 20.0 g)
were heated at 80 °C for 12 h under a drying tube. After cooling,
water (100 mL) was added and the reaction mixture was neutralised
with sodium hydroxide (3 ). The mixture was extracted with
dichloromethane (2ϫ150 mL) and the combined organic layers
were washed with brine (100 mL), dried with anhydrous sodium
sulfate, filtered and the solvents evaporated to dryness. The crude
material was then purified as detailed below.
Method B: A mixture of diglycolic acid (3.06 g, 23.3 mmol) and
polyphosphoric acid (86%, d = 1.90 g/cm3, 20.0 g) was heated at
80 °C for 2 h under a drying tube. The nitroaniline (14.5 mmol)
was added to this clear solution and the mixture was heated at
80 °C for another 12 h. After cooling, water (100 mL) was added
and the reaction mixture was neutralised with sodium hydroxide
(3 ). The mixture was extracted with dichloromethane
(3ϫ100 mL) and the combined organic layers were washed with
brine (100 mL), dried with anhydrous sodium sulfate, filtered and
the solvents evaporated to dryness. The crude material was then
purified as detailed below.
Conclusions
We have prepared examples of all four types of symmetri-
cally substituted dinitro Tröger’s base compounds. In ad-
dition, we have synthesised the first example of a tetranitro
Tröger’s base analogue. We expect that these molecules will
become useful building blocks, via conversion to the corre-
sponding diamines, in the synthesis of more highly func-
tionalised compounds such as bis-,[22–25] tris-[26–28] or
higher-order fused Tröger’s base analogues and in the syn-
thesis of new double-stranded helicates.[10,29]
Method D: The nitroaniline (6.57 mmol) and paraformaldehyde
(414 mg, 13.8 mmol, 2.1 equiv.) were dissolved in TFA (15 mL) and
stirred in the dark under argon for the time specified below. The
reaction mixture was then poured onto ice (150 g), basified by the
addition of sodium hydroxide (6 , 100 mL) and extracted with
dichloromethane (3ϫ50 mL). The organic layers were combined,
washed with brine (100 mL), dried with anhydrous sodium sulfate,
filtered and the solvents evaporated to dryness. The crude material
was then purified as detailed below.
Type II Compounds
Experimental Section
2,8-Dinitro-6H,12H-5,11-methanodibenzo[b,f][1,5]diazocine (2) and
4-(4Ј-Nitrophenyl)morpholine-3,5-dione (4)
General: Melting points were recorded with either a Reichert melt-
ing-point stage and are uncorrected or a TA Instruments DSC 2010
Differential Scanning Calorimeter. Microanalyses were performed
by the Microanalytical Unit, University of Otago, New Zealand.
High-resolution mass spectrometry (HRMS) was obtained either
at the School of Chemistry, University of New South Wales (FAB)
or The Research School of Chemistry, Australian National Univer-
Method A: Starting with 4-nitroaniline (2.00 g, 14.5 mmol), the
crude material was chromatographed (silica gel, dichloromethane)
to afford a mixture of (Ϯ)-2 and 4 (468 mg) as a yellow solid in
a ratio of 83:17, as determined by integration of the 1H NMR
spectrum.
1
sity (EI, Fissons VG-Autospec, or ESI; Bruker Apex 3). H NMR
Method B: Starting with 4-nitroaniline (2.00 g, 14.5 mmol), the
crude material was chromatographed (silica gel, dichloromethane)
to afford (Ϯ)-2 (585 mg, 28%) as a pale yellow solid; m.p. 258–
spectra were recorded with a Bruker WM AMX 400 spectrometer
(400 MHz) at 300 K unless otherwise stated. Signals were recorded
in terms of chemical shifts and are expressed in parts per million
(δ), multiplicity, coupling constants (in Hz, rounded to one decimal
place) and assignments in that order.
260 °C (ref.[8] 258–260 °C). H NMR (400 MHz, CDCl3, 25 °C): δ
1
= 4.32 (d, J = 17.4 Hz, 2 H, CH2), 4.42 (app. s, 2 H, CH2), 4.82
(d, J = 17.1 Hz, 2 H, CH2), 7.27 (d, J = 8.9 Hz, 2 H, ArH), 7.88
692
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 687–698

Symmetric Dinitro Tröger’s Base Analogues
(d, J = 2.6 Hz, 2 H, ArH), 8.23 (dd, J = 2.6, 8.9 Hz, 2 H, ArH)
ppm. 13C NMR (100 MHz, CDCl3, 25 °C): δ = 58.6, 66.4, 122.9,
123.2, 125.7, 128.1, 143.9, 153.9 ppm.
Method B: Starting with 2-bromo-4-nitroaniline (2.00 g,
9.21 mmol), the crude material was chromatographed (silica gel,
dichloromethane) to afford (Ϯ)-7 (600 mg, 28%) as a pale yellow
solid; m.p. Ͼ298 °C (dec.). 1H NMR (400 MHz, CDCl3/DMSO,
25 °C): δ = 4.41 (app. s, 2 H, CH2), 4.52 (d, J = 17.5 Hz, 2 H,
CH2), 4.77 (d, J = 17.5 Hz, 2 H, CH2), 8.10 (d, J = 2.5 Hz, 2 H,
ArH), 8.29 (d, J = 2.5 Hz, 2 H, ArH) ppm. 13C NMR (100 MHz,
CDCl3/DMSO, 25 °C): δ = 30.7, 54.8, 119.5, 122.3, 126.1, 131.8,
143.6, 150.6 ppm. C15H10Br2N4O4 (470.07): calcd. C 38.33, H 2.14,
N 11.92; found C 38.38, H 1.96, N 11.72.
Method D: Starting with 4-nitroaniline (500 mg, 3.62 mmol) and
stirring for 7 d, 1H NMR of the crude material obtained upon
work-up (530 mg) indicated that no Tröger’s base product was pres-
ent.
4,10-Dimethyl-2,8-dinitro-6H,12H-5,11-methanodibenzo[b,f][1,5]di-
azocine (3)
Method B: Starting with 2-methyl-4-nitroaniline (2.00 g,
13.15 mmol), the crude material was chromatographed (silica gel,
dichloromethane) to afford (Ϯ)-3 (750 mg, 34%) as a pale yellow
Method D: Starting with 2-bromo-4-nitroaniline (500 mg,
2.30 mmol) and stirring for 6 d, 1H NMR of the crude material
obtained upon work-up (480 mg) indicated that no Tröger’s base
product was present.
1
solid; m.p. Ͼ312 °C (dec.) (ref.[10] Ͼ300 °C). H NMR (400 MHz,
DMSO, 25 °C): δ = 2.47 (s, 6 H, 2ϫCH3), 4.30 (d, J = 17.2 Hz, 2
H, CH2), 4.35 (app. s, 2 H, CH2), 4.68 (d, J = 17.2 Hz, 2 H, CH2),
7.81 (d, J = 2.6 Hz, 2 H, ArH), 7.96 (d, J = 2.6 Hz, 2 H, ArH)
ppm. 13C NMR (100 MHz, DMSO, 25 °C): δ = 16.8, 54.0, 66.1,
120.2, 123.4, 129.3, 134.5, 142.7, 151.9 ppm. C17H16N4O4 (340.33):
calcd. C 59.99, H 4.74, N 16.46; found C 59.74, H 4.78, N 16.31.
1,4,7,10-Tetramethyl-2,8-dinitro-6H,12H-5,11-methanodibenzo[b,f]-
[1,5]diazocine (8)
Method B: Starting with 2,5-dimethyl-4-nitroaniline (800 mg,
4.81 mmol), the crude product was chromatographed (silica gel,
dichloromethane) to afford (Ϯ)-8 (140 mg, 16%) as a yellow solid;
m.p. Ͼ277 °C (dec.). 1H NMR (400 MHz, CDCl3, 25 °C): δ = 2.24
(s, 6 H, 2ϫ CH3), 2.46 (s, 6 H, 2ϫCH3), 4.01 (d, J = 17.0 Hz, 2
H, CH2), 4.21 (app. s, 2 H, CH2), 4.51 (d, J = 17.0 Hz, 2 H, CH2),
7.63 (s, 2 H, ArH) ppm. 13C NMR (100 MHz, CDCl3, 25 °C): δ =
13.6, 17.1, 53.9, 65.4, 125.0, 127.8, 128.3, 131.8, 146.1, 150.2 ppm.
C19H20N4O4 (368.39): calcd. C 61.95, H 5.47, N 15.21; found C
62.21, H 5.59, N 15.18.
Method D: Starting with 2-methyl-4-nitroaniline (6.00 g,
39.45 mmol) the isolation procedure described in the literature[10]
was followed to afford (Ϯ)-3 (5.25 g, 78%) as a pale yellow solid;
m.p. Ͼ312 °C (dec.). The spectroscopic data on this material were
identical to those obtained for (Ϯ)-3 isolated from Method B.
4,10-Dimethoxy-2,8-dinitro-6H,12H-5,11-methanodibenzo[b,f][1,5]-
diazocine (5) and 4-(2Ј-Methoxy-4Ј-nitrophenyl)morpholine-3,5-di-
one (6)
Method D: Starting with 2,5-dimethyl-4-nitroaniline (1.00 g,
6.21 mmol) and stirring for 11 d, (Ϯ)-8 (1.05 g, 92%) was obtained
as a yellow solid that required no further purification; m.p. (dec.)
Ͼ277 °C. The spectroscopic data on this material were identical to
those obtained for (Ϯ)-8 isolated from Method B.
Method A: Starting with 2-methoxy-4-nitroaniline (1.00 g,
5.95 mmol), the crude material was chromatographed (silica gel,
dichloromethane) to afford 6 (546 mg, 34%) as a white solid; m.p.
212–214 °C. 1H NMR (400 MHz, DMSO, 25 °C): δ = 3.89 (s, 3 H,
OCH3), 4.54 (d, J = 16.3 Hz, 2 H, CH2), 4.62 (d, J = 16.3 Hz, 2
H, CH2), 7.56 (d, J = 8.8 Hz, 1 H, ArH), 7.94–7.96 (m, 2 H, ArH)
ppm. 13C NMR (100 MHz, DMSO, 25 °C): δ = 57.6, 68.1, 108.1,
116.5, 128.7, 132.3, 149.6, 156.6, 169.9 ppm. C11H10N2O6 (266.21):
calcd. C 49.63, H 3.79, N 10.52; found C 49.67, H 3.85, N 10.46.
1,7-Dibromo-4,10-dimethyl-2,8-dinitro-6H,12H-5,11-methanodibenzo-
[b,f][1,5]diazocine (9)
Method B: Starting with 5-bromo-2-methyl-4-nitroaniline (560 mg,
2.42 mmol), the crude product was chromatographed (silica gel,
ethyl acetate/hexane, 1:4) to afford (Ϯ)-9 (139 mg, 23%) as a yellow
1
solid; m.p. Ͼ288 °C (dec.). H NMR (400 MHz, DMSO, 25 °C): δ
A second compound, (Ϯ)-5 (224 mg, 20 %), was subsequently
1
= 2.43 (s, 6 H, 2ϫCH3), 4.04 (d, J = 17.3 Hz, 2 H, CH2), 4.31
(app. s, 2 H, CH2), 4.56 (d, J = 17.3 Hz, 2 H, CH2), 7.82 (s, 2 H,
ArH) ppm. 13C NMR (100 MHz, CDCl3, 25 °C): δ = 16.6, 55.9,
64.5, 111.3, 125.2, 129.5, 134.2, 145.9, 150.4 ppm. HRMS (FAB+):
m/z calcd. for C17H14Br2N4O4 [M + Na]+, 518.927401; found
518.926126. C17H14Br2N4O4 (489.13): calcd. C 40.99, H 2.83, N
11.25; found C 40.90, H 2.89, N 10.97.
eluted from the column as a yellow solid; m.p. Ͼ290 °C (dec.). H
NMR (400 MHz, DMSO, 25 °C): δ = 3.95 (s, 6 H, OCH3), 4.28
(app. s, 2 H, CH2), 4.36 (d, J = 17.4 Hz, 2 H, CH2), 4.60 (d, J =
17.4 Hz, 2 H, CH2), 7.56 (d, J = 2.3 Hz, 2 H, ArH), 7.71 (d, J =
2.3 Hz, 2 H, ArH) ppm. 13C NMR (100 MHz, DMSO, 25 °C): δ
= 36.3, 53.7, 56.2, 104.0, 115.0, 129.9, 142.0, 143.3, 152.9 ppm.
C17H16N4O6 (372.33): calcd. C 54.84, H 4.33, N 15.05; found C
54.53, H 4.46, N 14.76. Crystals suitable for X-ray diffraction were
obtained by crystallisation from dichloromethane.
Method D: Starting with 5-bromo-2-methyl-4-nitroaniline (1.00 g,
4.33 mmol) and stirring for 9 d, the crude product was recrystal-
lised from dichloromethane to afford (Ϯ)-9 (765 mg, 71%) as a
yellow solid; m.p. Ͼ288 °C (dec.). The spectroscopic data on this
material were identical to those obtained for (Ϯ)-9 isolated from
Method B.
Method B: Starting with 2-methoxy-4-nitroaniline (2.00 g,
11.89 mmol), the crude material was chromatographed (silica gel,
ethyl acetate/dichloromethane, 1:3) to afford (Ϯ)-5 (605 mg, 27%)
as a yellow solid; m.p. Ͼ290 °C (dec.). The spectroscopic data on
this material were identical to those obtained for (Ϯ)-5 isolated
from Method A.
1,7-Dimethyl-2,8-dinitro-6H,12H-5,11-methanodibenzo[b,f][1,5]di-
azocine (10)
Method D: Starting with 2-methoxy-4-nitroaniline (1.00 g,
5.95 mmol) and stirring for 5 d, the crude material was chromato-
graphed (silica gel, ethyl acetate/dichloromethane, 1:3) to afford
(Ϯ)-5 (755 mg, 68%) as a yellow solid; m.p. Ͼ290 °C (dec.). The
spectroscopic data on this material were identical to those obtained
for (Ϯ)-5 isolated from Method A.
Method A: Starting with 3-methyl-4-nitroaniline (800 mg,
5.26 mmol), the crude material was chromatographed (silica gel,
dichloromethane) to afford (Ϯ)-10 (112 mg, 13%) as a yellow solid;
m.p. 269–271 °C. 1H NMR (400 MHz, DMSO, 25 °C): δ = 2.21 (s,
6 H, 2ϫCH3), 4.23 (app. s, 2 H, CH2), 4.36 (d, J = 17.0 Hz, 2 H,
CH2), 4.67 (d, J = 17.0 Hz, 2 H, CH2), 7.26 (d, J = 8.9 Hz, 2 H,
4,10-Dibromo-2,8-dinitro-6H,12H-5,11-methanodibenzo[b,f][1,5]di- ArH), 7.72 (d, J = 8.9 Hz, 2 H, ArH) ppm. 13C NMR (100 MHz,
azocine (7)
DMSO, 25 °C): δ = 13.7, 56.9, 64.2, 123.1, 123.4, 128.4, 130.9,
Eur. J. Org. Chem. 2009, 687–698
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
693

M. D. H. Bhuiyan, A. B. Mahon, P. Jensen, J. K. Clegg, A. C. Try
FULL PAPER
145.3, 152.1 ppm. HRMS (FAB+): m/z calcd. for C17H16N4O4 [M
+ H]+, 341.124431; found 341.124252. C17H16N4O4 (340.33): calcd.
C 59.99, H 4.74, N 16.46; found C 59.72, H 4.81, N 16.43.
lised from acetone to afford (Ϯ)-13 (515 mg, 45%) as a pale yellow
solid; m.p. Ͼ304 °C (dec.). The spectroscopic data on this material
were identical to those obtained for (Ϯ)-13 isolated from Method
A.
Method D: Starting with 3-methyl-4-nitroaniline (1.00 g,
6.58 mmol) and stirring for 9 d, the crude material was chromato-
4,10-Dibromo-2,8-dimethyl-1,7-dinitro-6H,12H-5,11-methanodibenzo-
graphed (silica gel, dichloromethane) to afford a mixture of [b,f][1,5]diazocine (14)
Tröger’s base isomers (155 mg, 14%) that were inseparable.
Method A: Starting with 2-bromo-4-methyl-5-nitroaniline (2.00 g,
8.70 mmol), the crude material was chromatographed (silica gel,
hexane/dichloromethane, 2:3) to afford (Ϯ)-14 (266 mg, 12%) as a
pale yellow solid; m.p. 190–191 °C. 1H NMR (400 MHz, CDCl3,
25 °C): δ = 2.25 (s, 6 H, CH3), 4.24 (d, J = 17.9 Hz, 2 H, CH2),
4.31 (app. s, 2 H, CH2), 4.62 (d, J = 17.9 Hz, 2 H, CH2), 7.48 (s, 2
H, ArH) ppm. 13C NMR (100 MHz, CDCl3, 25 °C): δ = 17.4, 52.3,
66.4, 122.7, 122.8, 128.2, 134.4, 143.2, 148.4 ppm. HRMS (FAB+):
m/z calcd. for C17H14Br2N4O4 [M + Na]+, 518.927401; found
518.927064. C17H14Br2N4O4 (498.13): calcd. C 40.99, H 2.83, N
11.25; found C 41.75, H 3.01, N 11.00.
1,7-Dibromo-2,8-dinitro-6H,12H-5,11-methanodibenzo[b,f][1,5]di-
azocine (11)
Method B: Starting with 3-bromo-4-nitroaniline (1.00 g,
4.60 mmol), the crude product was chromatographed (silica gel,
dichloromethane) to afford (Ϯ)-11 (150 mg, 14%) as a yellow solid;
m.p. Ͼ258 °C (dec.). 1H NMR (400 MHz, CDCl3, 25 °C): δ = 4.13
(app. s, 2 H, CH2), 4.43 (d, J = 17.1 Hz, 2 H, CH2), 4.57 (d, J =
17.1 Hz, 2 H, CH2) 7.21 (d, J = 8.8 Hz, 2 H, ArH), 7.79 (d, J =
8.8 Hz, 2 H, ArH) ppm. 13C NMR (100 MHz, CDCl3, 25 °C): δ
= 53.4, 64.2, 106.1, 120.9, 122.4, 128.8, 133.4, 147.6 ppm. HRMS
(FAB+): m/z calcd. for C15H10Br2N4O4 [M + Na]+, 490.896101;
found 490.894969. C15H10Br2N4O4 (470.07): calcd. C 38.33, H
2.14, N 11.92; found C 38.62, H 2.17, N 11.62.
Method D: Starting with 2-bromo-4-methyl-5-nitroaniline (1.80 g,
7.80 mmol) and stirring for 5 d, 1H NMR of the crude material
obtained upon work-up (1.78 g) indicated that only a trace of
Tröger’s base product was present.
Method D: Starting with 3-bromo-4-nitroaniline (500 mg,
2.30 mmol) and stirring for 7 d, 1H NMR of the crude material
obtained upon work-up (507 mg) indicated that no Tröger’s base
product was present.
4,10-Dimethyl-1,7-dinitro-6H,12H-5,11-methanodibenzo[b,f][1,5]di-
azocine (15)
Method A: Starting with 2-methyl-5-nitroaniline (2.00 g,
13.15 mmol), the crude material was chromatographed (silica gel,
ethyl acetate/hexane, 1:4) to afford (Ϯ)-15 (126 mg, 6%) as a yellow
solid; m.p. Ͼ274 °C (dec.). 1H NMR (400 MHz, CDCl3/DMSO,
25 °C): δ = 2.27 (s, 6 H, CH3), 4.06–4.14 (m, 4 H, 2ϫCH2), 4.67
(d, J = 18.7 Hz, 2 H, CH2), 7.01 (d, J = 8.4 Hz, 2 H, ArH), 7.53
(d, J = 8.4 Hz, 2 H, ArH) ppm. 13C NMR (100 MHz, CDCl3/
DMSO, 25 °C): δ = 17.5, 53.5, 64.6, 120.3, 123.8, 129.0, 140.4,
144.9, 146.8 ppm. HRMS (FAB+): m/z calcd. for C17H16N4O4 [M
+ Na]+, 363.106376; found 363.106387. C17H16N4O4 (340.33):
calcd. C 59.99, H 4.74, N 16.46; found C 59.98, H 4.79, N 16.50.
Crystals suitable for X-ray diffraction were obtained by crystallisa-
tion from ethyl acetate/dichloromethane.
Type I Compounds
2,4,8,10-Tetramethyl-1,7-dinitro-6H,12H-5,11-methanodibenzo[b,f]-
[1,5]diazocine (12)
Method B: Starting with 2,4-dimethyl-5-nitroaniline (2.44 g,
15.04 mmol), the crude product was chromatographed (silica gel,
hexane/ethyl acetate, 3:1) to afford (Ϯ)-12 (1.37 g, 51%) as a pale
yellow solid; m.p. 226–227 °C. 1H NMR (400 MHz, CDCl3, 25 °C):
δ = 2.24 (s, 6 H, 2ϫCH3), 2.37 (s, 6 H, 2ϫCH3), 3.86 (d, J =
17.4 Hz, 2 H, CH2), 4.26 (app. s, 2 H, CH2), 4.61 (d, J = 17.4 Hz,
2 H, CH2), 7.02 (s, 2 H, ArH) ppm. 13C NMR (100 MHz, CDCl3,
25 °C): δ = 17.3, 17.7, 51.9, 66.3, 120.3, 120.8, 126.2, 132.1, 136.8,
144.3 ppm. HRMS (FAB+): m/z calcd. for C19H20N4O4 [M +
Na]+, 391.137676; found 391.137560. C19H20N4O4 (368.39): calcd.
C 61.95, H 5.47, N 15.21; found C 62.22, H 5.60, N 15.39.
Method D: Starting with 2-methyl-5-nitroaniline (1.00 g,
6.57 mmol) and stirring for 5 d, the crude material was chromato-
graphed (silica gel, ethyl acetate/hexane, 1:4) to afford (Ϯ)-15
(653 mg, 56%) as a yellow solid; m.p. Ͼ274 °C (dec.). The spectro-
scopic data on this material were identical to those obtained for
(Ϯ)-15 isolated from Method A.
Method D: Starting with 2,4-dimethyl-5-nitroaniline (500 mg,
3.01 mmol) and stirring for 6 d, the crude product was chromato-
graphed (silica gel, hexane/ethyl acetate, 3:1) to afford (Ϯ)-12
(251 mg, 45%) as a pale yellow solid; m.p. 226–227 °C. The spec-
troscopic data on this material were identical to those obtained for
(Ϯ)-12 isolated from Method B.
4,10-Diisopropyl-1,7-dinitro-6H,12H-5,11-methanodibenzo[b,f][1,5]-
diazocine (16)
Method A: Starting with 2-isopropyl-5-nitroaniline (1.00 g,
5.55 mmol), the crude material was chromatographed (silica gel,
ethyl acetate/hexane, 1:3) to afford (Ϯ)-16 (251 mg, 23%) as a pale
yellow solid; m.p. 241–242 °C. 1H NMR (400 MHz, CDCl3, 25 °C):
δ = 1.23 (d, J = 6.8 Hz, 6 H, CH3), 1.34 (d, J = 7.0 Hz, 6 H, CH3),
3.73 (m, 2 H, CH), 4.32–4.41 (m, 4 H, 2 ϫ CH2), 5.01 (d, J =
18.0 Hz, 2 H, CH2), 7.33 (d, J = 8.6 Hz, 2 H, ArH), 7.88 (d, J =
8.6 Hz, 2 H, ArH) ppm. 13C NMR (100 MHz, CDCl3, 25 °C): δ =
22.6, 24.9, 27.1, 56.3, 65.1, 121.8, 124.5, 125.7, 145.3, 146.4, 151.7
2,8-Dibromo-4,10-dimethyl-1,7-dinitro-6H,12H-5,11-methanodibenzo-
[b,f][1,5]diazocine (13)
Method A: Starting with 4-bromo-2-methyl-5-nitroaniline (1.00 g,
4.32 mmol), the crude material was chromatographed (silica gel,
ethyl acetate/hexane, 1:3) to afford (Ϯ)-13 (495 mg, 46%) as a pale
yellow solid; m.p. Ͼ304 °C (dec.). 1H NMR (400 MHz, CDCl3,
25 °C): δ = 2.38 (s, 6 H, CH3), 3.85 (d, J = 17.4 Hz, 2 H, CH2),
4.24 (app. s, 2 H, CH2), 4.57 (d, J = 17.4 Hz, 2 H, CH2), 7.41 (s, 2
H, ArH) ppm. 13C NMR (100 MHz, CDCl3, 25 °C): δ = 23.3, 55.0, ppm. HRMS (FAB+): m/z calcd. for C21H24N4O4 [M + Na]+,
65.8, 110.2, 118.6, 120.9, 128.5, 139.9, 147.3 ppm. HRMS (FAB+): 419.168976; found 419.16.168593. C21H24N4O4 (396.44): calcd. C
m/z calcd. for C17H14Br2N4O4 [M + Na]+, 518.927401; found
518.927064. C17H14Br2N4O4 (498.13): calcd. C 40.99, H 2.83, N
11.25; found C 41.10, H 2.91, N 11.00.
63.62, H 6.10, N 14.13; found C 63.76, H 6.15, N 14.15.
Method D: Starting with 2-isopropyl-5-nitroaniline (1.90 g,
10.50 mmol) and stirring for 14 d, the crude material was recrystal-
lised from acetone to afford (Ϯ)-16 (708 mg, 34%) as a pale yellow
solid; m.p. 241–242 °C. The spectroscopic data on this material
Method D: Starting with 4-bromo-2-methyl-5-nitroaniline (1.06 g,
4.59 mmol) and stirring for 13 d, the crude material was recrystal-
694
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 687–698

Symmetric Dinitro Tröger’s Base Analogues
were identical to those obtained for (Ϯ)-16 isolated from Method
A.
ArH) ppm. 13C NMR (100 MHz, CDCl3, 25 °C): δ = 17.8, 20.1,
55.7, 58.5, 66.2, 120.2, 120.9, 121.5, 126.7, 127.5, 129.3, 130.7,
130.9, 133.3, 146.1 ppm. C17H16N4O4 (340.33): calcd. C 59.99, H
4.74, N 16.49; found C 60.28, H 4.95, N 16.64.
4,10-Dimethoxy-1,7-dinitro-6H,12H-5,11-methanodibenzo[b,f][1,5]-
diazocine (17)
Method D: Starting with 4-methyl-3-nitroaniline (1.00 g,
6.57 mmol) and stirring for 7 d, 1H NMR of the crude material
obtained upon work-up (1.10 g) indicated that no Tröger’s base
product was present.
Method A: Starting with 2-methoxy-5-nitroaniline (2.00 g,
11.90 mmol), the crude material was chromatographed (silica gel,
dichloromethane) to afford (Ϯ)-17 (103 mg, 5%) as a pale yellow
1
solid; m.p. Ͼ308 °C (dec.). H NMR (400 MHz, CDCl3, 25 °C): δ
= 4.04 (s, 6 H, OCH3), 4.34 (app. s, 2 H, CH2), 4.67 (d, J = 18.6 Hz,
2 H, CH2), 4.91 (d, J = 18.6 Hz, 2 H, CH2), 6.85 (d, J = 9.2 Hz, 2
H, ArH), 7.99 (d, J = 9.2 Hz, 2 H, ArH) ppm. 13C NMR
(100 MHz, CDCl3, 25 °C): δ = 53.8, 56.5, 66.0, 108.5, 123.1, 126.5,
136.3, 140.7, 157.5 ppm. HRMS (FAB+ ): m/z calcd. for
C17H16N4O6 [M + Na]+, 395.096205; found 395.096136.
C17H16N4O6 (372.33): calcd. C 54.84, H 4.33, N 15.05; found C
54.91, H 4.44, N 14.98.
2,8-Diisopropyl-1,7-dinitro-6H,12H-5,11-methanodibenzo[b,f][1,5]-
diazocine (20)
Method A: Starting with 4-isopropyl-3-nitroaniline (900 mg,
4.99 mmol), the crude brown product was chromatographed (silica
gel, hexane/ethyl acetate, 3:1) to afford (Ϯ)-20 (75 mg, 8%) as a
pale yellow solid; m.p. 302–303 °C. 1H NMR (400 MHz, CDCl3,
25 °C): δ = 1.19 (d, J = 6.8 Hz, 6 H, 2ϫCH3), 1.23 (d, J = 6.8 Hz,
6 H, 2ϫCH3), 2.80–2.97 (m, 2 H, CH), 4.00 (d, J = 17.1 Hz, 2 H,
CH2), 4.25 (s, 2 H, CH2), 4.63 (d, J = 17.1 Hz, 2 H, CH2), 7.21 (d,
J = 8.5 Hz, 2 H, ArH), 7.26 (d, J = 8.5 Hz, 2 H, ArH) ppm. 13C
NMR (100 MHz, CDCl3, 25 °C): δ = 23.4, 23.8, 28.6, 55.0, 65.8,
118.9, 126.3, 127.7, 136.2, 146.3 ppm. HRMS (FAB+): m/z calcd.
for C21H24N4O4 [M + Na]+, 419.168976; found 419.16.168415.
Method D: Starting with 2-methoxy-5-nitroaniline (1.00 g,
5.95 mmol) and stirring for 5 d, the crude material was chromato-
graphed (silica gel, dichloromethane) to afford (Ϯ)-17 (254 mg,
22%) as a pale yellow solid; m.p. Ͼ308 °C (dec.). The spectroscopic
data on this material were identical to those obtained for (Ϯ)-17
isolated from Method A.
Method D: Starting with 4-isopropyl-3-nitroaniline (1.00 g,
4.59 mmol) and stirring for 7 d, the crude material was chromato-
graphed (silica gel, hexane/ethyl acetate, 4:1) to afford (Ϯ)-20
(516 mg, 45%) as a pale yellow solid; m.p. 302–303 °C. The spec-
troscopic data on this material were identical to those obtained for
(Ϯ)-20 isolated from Method A.
4,10-Dibromo-1,7-dinitro-6H,12H-5,11-methanodibenzo[b,f][1,5]di-
azocine (18)
Method A: Starting with 2-bromo-5-nitroaniline (1.00 g,
4.61 mmol), the crude material was chromatographed (silica gel,
hexane/dichloromethane, 1:9) to afford (Ϯ)-18 (103 mg, 10%) as a
pale yellow solid; m.p. Ͼ280 °C (dec.). 1H NMR (400 MHz,
CDCl3, 25 °C): δ = 4.40 (app. s, 2 H, CH2), 4.76 (d, J = 18.7 Hz,
2 H, CH2), 4.99 (d, J = 18.7 Hz, 2 H, CH2), 7.68 (d, J = 8.8 Hz, 2
H, ArH), 7.78 (d, J = 8.8 Hz, 2 H, ArH) ppm. 13C NMR
(100 MHz, CDCl3, 25 °C): δ = 54.6, 65.6, 122.3, 126.9, 127.5, 132.4,
146.4, 146.7 ppm. HRMS (FAB+): m/z calcd. for C15H10Br2N4O4
[M + Na]+, 490.896101; found 490.896087. C15H10Br2N4O4
(470.07): calcd. C 38.33, H 2.14, N 11.92; found C 38.12, H 2.32,
N 11.66.
2,8-Dibutyl-1,7-dinitro-6H,12H-5,11-methanodibenzo[b,f][1,5]diaz-
ocine (21)
Method B: Starting with 4-butyl-3-nitroaniline (950 mg,
4.89 mmol), the crude product was chromatographed (silica gel,
hexane/ethyl acetate, 3:1) to afford (Ϯ)-21 (182 mg, 18%) as a pale
yellow solid; m.p. 166–168 °C. 1H NMR (400 MHz, CDCl3, 25 °C):
δ = 0.89 (t, J = 7.3 Hz, 6 H, 2ϫCH3), 1.38–1.24 (m, 4 H, 2ϫCH2),
1.44–1.63 (m, 4 H, 2ϫCH2), 2.40–2.62 (m, 4 H, 2ϫCH2), 4.02 (d,
J = 17.2 Hz, 2 H, CH2), 4.27 (app. s, 2 H, CH2), 4.67 (d, J =
17.2 Hz, 2 H, CH2), 7.15 (d, J = 8.4 Hz, 2 H, ArH), 7.18 (d, J =
8.4 Hz, 2 H, ArH) ppm. 13C NMR (100 MHz, CDCl3, 25 °C): δ =
14.2, 22.9, 31.5, 33.2, 55.9, 66.2, 120.1, 128.1, 130.4, 131.5,
146.8 ppm. HRMS (FAB+): m/z calcd. for C23H28N4O4 [M +
Na]+, 447.200276; found 447.200812. C23H28N4O4 (424.49): calcd.
C 65.08, H 6.65, N 13.20; found C 65.35, H 6.77, N 13.29.
Method D: Starting with 2-bromo-5-nitroaniline (500 mg,
2.30 mmol) and stirring for 7 d, 1H NMR of the crude material
obtained upon work-up (514 mg) indicated that no Tröger’s base
product was present.
2,8–Dimethyl-1,7-dinitro-6H,12H-5,11-methanodibenzo[b,f][1,5]di-
azocine (19) and 2,8-Dimethyl-1,9-dinitro-6H,12H-5,11-methanodi-
benzo[b,f][1,5]diazocine (25)
Method D: Starting with 4-butyl-3-nitroaniline (1.00 g, 5.15 mmol)
and stirring for 11 d, 1H NMR of the crude material obtained upon
work-up (1.20 g) indicated that only a trace of Tröger’s base prod-
uct was present.
Method B: Starting with 4-methyl-3-nitroaniline (1.00 g,
6.57 mmol), the crude product was chromatographed (silica gel,
dichloromethane) to afford (Ϯ)-19 (315 mg, 32%) as an orange so-
lid; m.p. 230–232 °C. 1H NMR (400 MHz, CDCl3, 25 °C): δ = 2.76
(s, 6 H, CH3), 4.05 (d, J = 17.3 Hz, 2 H, CH2), 4.29 (app. s, 2 H,
CH2), 4.71 (d, J = 17.3 Hz, 2 H, CH2), 7.14 (d, J = 8.4 Hz, 2 H,
ArH), 7.18 (d, J = 8.4 Hz, 2 H, ArH) ppm. 13C NMR (100 MHz,
CDCl3, 25 °C): δ = 17.7, 55.7, 65.8, 120.2, 126.7, 127.7, 130.8,
146.4 ppm. C17H16N4O4 (340.33): C 59.99, H 4.74, N 16.49; found
C 60.06, H 4.96, N 16.71%.
2,8-Dibromo-1,7-dinitro-6H,12H-5,11-methanodibenzo[b,f][1,5]di-
azocine (22)
Method B: Starting with 4-bromo-3-nitroaniline (1.00 g,
4.61 mmol), the crude product was chromatographed (silica gel,
dichloromethane) to afford (Ϯ)-22 (57 mg, 6%) as a brown solid;
m.p. Ͼ300 °C (dec.). 1H NMR (400 MHz, CDCl3, 25 °C): δ = 4.16
(d, J = 17.1 Hz, 2 H, CH2), 4.36 (app. s, 2 H, CH2), 4.69 (d, J =
Another isomer, (Ϯ)-25 (91 mg, 9%) was subsequently eluted from
the column as a yellow solid; m.p. 242–244 °C. 1H NMR 17.1 Hz, 2 H, CH2) 7.49 (d, J = 8.8 Hz, 2 H, ArH), 7.69 (d, J =
(400 MHz, CDCl3, 25 °C): δ = 2.26 (s, 3 H, CH3), 2.47 (s, 3 H,
CH3) 4.07 (d, J = 17.3 Hz, 1 H, CH2), 4.17 (d, J = 17.3 Hz, 1 H,
CH2), 4.28 (app. s, 2 H, CH2), 4.67 (d, J = 17.3 Hz, 1 H, CH2),
4.72 (d, J = 17.3 Hz, 1 H, CH2), 6.87 (s, 1 H, ArH), 7.14 (d, J =
8.3 Hz, 1 H, ArH), 7.20 (d, J = 8.3 Hz, 1 H, ArH), 7.75 (s, 1 H,
8.8 Hz, 2 H, ArH) ppm. 13C NMR (100 MHz, DMSO, 25 °C): δ
= 54.4, 64.3, 106.4, 120.5, 122.0, 129.3, 132.5, 148.1 ppm. HRMS
(FAB+): m/z calcd. for C15H10Br2N4O4 [M + Na]+, 490.896101;
found 490.895153. C15H10Br2N4O4 (470.07): calcd. C 38.33, H
2.14, N 11.92; found C 38.33, H 2.18, N 11.68.
Eur. J. Org. Chem. 2009, 687–698
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
695

M. D. H. Bhuiyan, A. B. Mahon, P. Jensen, J. K. Clegg, A. C. Try
FULL PAPER
Method D: Starting with 4-bromo-3-nitroaniline (500 mg,
2.30 mmol); the 1H NMR of the crude material obtained upon
work-up (520 mg) indicated that no Tröger’s base product was pres-
ent.
132.9, 147.3 ppm. HRMS (EI+): m/z calcd. for C17H16N4O4 [M]+,
340.1172; found 340.1172.
Another compound, 28 (350 mg, 11%), was subsequently eluted
from the column as a white solid; m.p. 236–237 °C. 1H NMR
(400 MHz, CDCl3, 25 °C): δ = 2.29 (s, 3 H, CH3), 4.52–4.61 (m, 4
H, 2ϫCH2) 7.35 (dd, J = 1.0, 8.1 Hz, 1 H, ArH), 7.31–7.51 (m, 1
H, ArH), 8.00 (dd, J = 1.0, 8.2 Hz, 1 H, ArH) ppm. 13C NMR
(100 MHz, CDCl3, 25 °C): δ = 14.5, 68.5, 126.1, 127.7, 132.3, 133.9,
168.9 ppm. C11H10N2O5 (250.21): calcd. C 52.80, H 4.03, N 11.20;
found C 52.62, H 4.11, N 11.03.
2,8-Dibromo-3,9-dimethyl-1,7-dinitro-6H,12H-5,11-methanodibenzo-
[b,f][1,5]diazocine (23)
Method A: Starting with 4-bromo-3-methyl-5-nitroaniline (650 mg,
2.81 mmol), the crude material was chromatographed (silica gel,
dichloromethane) to afford (Ϯ)-23 (127 mg, 18%) as a pale yellow
1
solid; m.p. Ͼ339 °C (dec.). H NMR (400 MHz, CDCl3, 25 °C): δ
= 2.40 (s, 6 H, CH3), 4.01 (d, J = 17.0 Hz, 2 H, CH2), 4.21 (app.
s, 2 H, CH2), 4.58 (d, J = 17.0 Hz, 2 H, CH2), 7.12 (s, 2 H, ArH)
ppm. 13C NMR (100 MHz, CDCl3, 25 °C): δ = 23.3, 55.0, 65.8,
110.2, 118.6, 120.9, 128.5, 139.9, 147.3 ppm. HRMS (FAB+): m/z
calcd. for C17H14Br2N4O4 [M + Na]+, 518.927401; found
518.927064. C17H14Br2N4O4 (498.13): calcd. C 40.99, H 2.83, N
11.25; found C 40.99, H 2.91, N 11.02.
Method D: Starting with 2-methyl-3-nitroaniline (1.00 g,
6.57 mmol), the crude material was recrystallised from dichloro-
methane to afford (Ϯ)-27 (475 mg, 43%) as a pale yellow solid;
m.p. 279–280 °C. The spectroscopic data on this material were
identical to those obtained for (Ϯ)-27 isolated from Method B.
Type IV Compounds
2,8-Dimethyl-4,10-dinitro-6H,12H-5,11-methanodibenzo[b,f][1,5]di-
azocine (29)
Method D: Starting with 4-bromo-3-methyl-5-nitroaniline (1.00 g,
1
3.71 mmol) and stirring for 11 d, H NMR of the crude material
Method B: Starting with 4-methyl-2-nitroaniline (1.00 g,
6.57 mmol), the crude material was chromatographed (silica gel,
dichloromethane) to afford (Ϯ)-29 (400 mg, 36%) as a yellow solid;
obtained upon work-up (953 mg) indicated that only a trace of
Tröger’s base product was present.
1,7-Dinitro-6H,12H-5,11-methanodibenzo[b,f][1,5]diazocine (24) and
1,9-Dinitro-6H,12H-5,11-methanodibenzo[b,f][1,5]diazocine (26)
1
m.p. 242–243 °C. H NMR (400 MHz, CDCl3, 25 °C): δ = 2.32 (s,
6 H, 2ϫCH3), 4.26 (app. s, 2 H, CH2), 4.48 (d, J = 17.8 Hz, 2 H,
CH2), 4.72 (d, J = 17.8 Hz, 2 H, CH2), 7.02–7.06 (m, 2 H, ArH),
7.57–7.61 (m, 2 H, ArH) ppm. 13C NMR (100 MHz, CDCl3,
25 °C): δ = 20.9, 57.1, 67.1, 124.5, 131.0, 132.6, 135.0, 139.5,
145.3 ppm. HRMS (FAB+): m/z calcd. for C17H16N4O4 [M +
Na]+, 363.106376; found 363.105808. C17H16N4O4 (340.33): calcd.
C 59.99, H 4.74, N 16.46; found C 60.23, H 4.96, N 16.49.
Method B: Starting with 3-nitroaniline (2.00 g, 14.49 mmol), the
crude material was chromatographed (silica gel, dichloromethane)
to afford (Ϯ)-24 (450 mg, 22%) as a pale yellow solid; m.p. 294–
296 °C. 1H NMR (400 MHz, DMSO, 25 °C): δ = 4.39 (app. s, 2 H,
CH2), 4.57 (d, J = 17.8 Hz, 2 H, CH2), 4.97 (d, J = 17.8 Hz, 2 H,
CH2), 7.41 (app. t, J = 8.1 Hz, 2 H, ArH), 7.63 (dd, J = 1.2, 8.1 Hz,
2 H, ArH), 7.76 (dd, J = 1.2, 8.1 Hz, 2 H, ArH) ppm. 13C NMR
(100 MHz, DMSO, 25 °C): δ = 57.1, 64.1, 120.3, 123.9, 127.8,
131.6, 147.4, 149.5 ppm. C15H12N4O4 (312.28): C 57.69, H 3.87, N
17.94; found C 57.76, H 3.79, N 17.88%.
Method D: Starting with 4-methyl-2-nitroaniline (1.00 g,
6.57 mmol) and stirring for 5 d, 1H NMR of the crude material
obtained upon work-up (1.09 g) indicated that only a trace of
Tröger’s base product was present.
Another isomer, (Ϯ)-26 (100 mg, 5%) was subsequently eluted
from the column as a pale yellow solid; m.p. 222–224 °C. 1H NMR
(400 MHz, CDCl3, 25 °C): δ = 4.27–4.31 (m, 3 H, CH2), 4.57 (d, J
= 17.8 Hz, 1 H, CH2), 4.78 (d, J = 17.8 Hz, 1 H, CH2), 5.07 (d, J
= 17.8 Hz, 1 H, CH2), 7.09 (d, J = 8.3 Hz, 1 H, ArH), 7.36 (app.
t, J = 8.1 Hz, 1 H, ArH), 7.47 (dd, J = 2.2, 8.3 Hz, 1 H, ArH),
1,2,7,8-Tetramethyl-4,10-dinitro-6H,12H-5,11-methanodibenzo[b,f]-
[1,5]diazocine (30)
Method A: Starting with 4,5-dimethyl-2-nitroaniline (530 mg,
3.19 mmol), the crude material was chromatographed (silica gel,
dichloromethane) to afford (Ϯ)-30 (46 mg, 8%) as a yellow solid;
1
7.82–7.85 (m, 2 H, ArH), 8.01 (d, J = 2.2 Hz, 1 H, ArH) ppm. 13
C
m.p. 274–276 °C. H NMR (400 MHz, CDCl3, 25 °C): δ = 2.09 (s,
6 H, 2ϫCH3), 2.27 (s, 6 H, 2ϫCH3), 4.21 (app. s, 2 H, CH2), 4.42
(d, J = 17.8 Hz, 2 H, CH2), 4.64 (d, J = 17.8 Hz, 2 H, CH2), 7.67
(s, 2 H, ArH) ppm. 13C NMR (100 MHz, CDCl3, 25 °C): δ = 14.6,
19.8, 56.5, 65.4, 124.8, 128.7, 133.3, 140.2, 140.8, 142.6 ppm.
HRMS (FAB+ ): m/z calcd. for C1 9 H2 0 N4 O4 [M + Na]+ ,
391.137676; found 391.137325. C19H20N4O4 (368.39): calcd. C
61.95, H 5.47, N 15.21; found C 62.10, H 5.94, N 15.36.
NMR (100 MHz, CDCl3, 25 °C): δ = 57.9, 59.1, 65.7, 118.9, 120.7,
121.4, 124.2, 127.8, 127.9, 131.1, 134.7, 148.4, 149.8 ppm.
C15H12N4O4 (312.28): calcd. C 57.69, H 3.87, N 17.94; found C
57.80, H 3.91, N 17.66.
Method D: Starting with 3-nitroaniline (1.00 g, 7.25 mmol) and
stirring for 5 d, 1H NMR of the crude material obtained upon
work-up (1.07 g) indicated that no Tröger’s base product was pres-
ent.
Method D: Starting with 4,5-dimethyl-2-nitroaniline (1.00 g,
6.02 mmol) and stirring for 5 d, 1H NMR of the crude material
obtained upon work-up (1.26 g) indicated that only a trace of
Tröger’s base product was present.
Type III Compound
4,10-Dimethyl-3,9-dinitro-6H,12H-5,11-methanodibenzo[b,f][1,5]di-
azocine (27) and 4-(2Ј-Methyl-3Ј-nitrophenyl)morpholine-3,5-dione
(28)
Tetranitro Compound
2,8-Dimethoxy-4,10-dimethyl-1,3,7,10-tetranitro-6H,12H-5,11-meth-
anodibenzo[b,f][1,5]diazocine (31)
Method B: Starting with 2-methyl-3-nitroaniline (2.00 g,
13.15 mmol), the crude material was chromatographed (silica gel,
dichloromethane) to afford (Ϯ)-27 (732 mg, 33%) as a pale yellow
solid; m.p. 279–280 °C. 1H NMR (400 MHz, CDCl3, 25 °C): δ =
2.59 (s, 6 H, 2ϫCH3), 4.00 (d, J = 17.5 Hz, 2 H, CH2), 4.31 (app.
Method B: Starting with 4-methoxy-2-methyl-3,5-dinitroaniline
(500 mg, 2.20 mmol), the crude product was chromatographed (sil-
ica gel, hexane/ethyl acetate, 3:1) to afford (Ϯ)-31 (60 mg, 11%) as
1
s, 2 H, CH2), 4.66 (d, J = 17.5 Hz, 2 H, CH2), 6.93 (d, J = 8.4 Hz, a yellow solid; m.p. Ͼ236 °C (dec.). H NMR (400 MHz, CDCl3,
2 H, ArH), 7.56 (d, J = 8.4 Hz, 2 H, ArH) ppm. 13C NMR 25 °C): δ = 2.36 (s, 6 H, 2ϫCH3), 3.83 (d, J = 17.9 Hz, 2 H, CH2),
(100 MHz, CDCl3, 25 °C): δ = 13.5, 55.4, 67.1, 119.8, 124.9, 128.8,
696
3.90 (s, 6 H, OCH3), 4.28 (app. s, 2 H, CH2), 4.66 (d, J = 17.9 Hz,
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 687–698

Symmetric Dinitro Tröger’s Base Analogues
2 H, CH2) ppm. 13C NMR (100 MHz, CDCl3, 25 °C): δ = 12.3, (400 MHz, CDCl3, 25 °C): δ = 2.06 (s, 6 H, 2ϫCH3), 2.33 (s, 6 H,
51.9, 64.8, 65.9, 123.1, 130.6, 140.9, 142.2 ppm. HRMS (FAB+): 2ϫCH3), 3.24 (br. s, 4 H, 2ϫNH2), 3.83 (d, J = 16.2 Hz, 2 H,
m/z calcd. for C19H18N6O10 [M + Na]+, 513.097662; found CH2), 4.25 (app. s, 2 H, CH2), 4.32 (d, J = 16.2 Hz, 2 H, CH2),
513.098404.
6.80 (s, 2 H, ArH) ppm. 13C NMR (100 MHz, CDCl3, 25 °C): δ =
16.4, 16.9, 51.9, 66.2, 113.1, 117.2, 120.8, 121.9, 130.5, 138.9 ppm.
Method D: Starting with 4-methoxy-3,5-dinitro-2-methylaniline
(1.00 g, 4.41 mmol) and stirring for 10 d, the crude product was
chromatographed (silica gel, hexane/ethyl acetate, 3:1) to afford
(Ϯ)-31 (45 mg, 4%) as a yellow solid; m.p. Ͼ236 °C (dec.). The
spectroscopic data on this material were identical to those obtained
for (Ϯ)-31 isolated from Method B.
+
HRMS (ESI): m/z calcd. for C19H24N4 [M H]+, 309.2074; found
309.2081.
Reduction of (؎)-12 with Continuous-Flow Apparatus: (Ϯ)-12
(100 mg, 0.27 mmol) was dissolved in ethyl acetate (10 mL) and
ethanol (10 mL) and the mixture passed through a H-Cube
equipped with a 30 mm 10% Pd/C cartridge (1 mL/min; 20 °C;
70 bar). Fractions were collected, which were monitored by TLC,
and evaporation of solvent under reduced pressure afforded (Ϯ)-
33 (81 mg, 97%) as an off-white solid that had identical spectral
properties to those listed above.
2,8-Diamino-6H,12H-5,11-methanodibenzo[b,f][1,5]diazocine (32):
Reduction of (Ϯ)-2 with Pd/C in ethanol: A mixture of (Ϯ)-2
(100 mg, 0.32 mmol), ethanol (20 mL), dichloromethane (5 mL)
and 10% palladium on carbon (50 mg) was stirred in the dark un-
der hydrogen for 3 d. The reaction was monitored by TLC and
when no evidence of the starting material was observed, the mix-
ture was filtered through Celite and the solvents evaporated to dry-
ness. The crude solid was chromatographed (silica gel, ethyl ace-
tate:methanol, 4:1) to afford (Ϯ)-32 (74 mg, 92%) as a white solid;
m.p. 303,–304 °C (ref.[21] 266 °C with decomposition). 1H NMR
(400 MHz, DMSO, 25 °C): δ = 3.76 (d, J = 16.3 Hz, 2 H, CH2),
4.04 (app. s, 2 H, CH2), 4.37 (d, J = 16.2 Hz, 2 H, CH2), 4.64 (br.
s, 4 H, 2ϫNH2), 6.06 (d, J = 2.2 Hz, 2 H, ArH), 6.35 (dd, J = 2.2,
8.4 Hz, 2 H, ArH), 6.72 (d, J = 8.4 Hz, 2 H, ArH) ppm. 13C NMR
(100 MHz, DMSO, 25 °C): δ = 58.3, 67.1, 110.8, 113.5, 124.9,
128.3, 137.5, 144.5 ppm. HRMS (EI+): m/z calcd. for C15H16N4
[M]+, 252.1375; found 252.1377. C15H16N4·0.5H2O (261.32): calcd.
C 68.94, H 6.56, N 21.44; found C 69.26, H 6.45, N 20.79.
Supporting Information (see also the footnote on the first page of
this article): Experimental details for the synthesis of all non-com-
mercially available anilines used in this work, together with a
NOESY spectrum of compound 23.
CCDC-645980 (for 5) and -645981 (for 15) contain the supplemen-
tary crystallographic data. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Acknowledgments
We thank the Australian Research Council for a Discovery Project
grant to A. C. T. (DP0345180) and Macquarie University for the
award of PhD scholarships to M. D. H. B. and A. B. M.
Reduction of (؎)-2 with Tin(II) Chloride: (Ϯ)-2 (300 mg,
0.96 mmol) and tin(II) chloride (2.5 g, 11.1 mmol) were suspended
in absolute ethanol (50 mL) and refluxed at 75 °C for 2 d. The reac-
tion mixture was cooled then poured onto ice and neutralised with
sodium hydrogen carbonate solution (10%). The reaction mixture
was extracted with dichloromethane, ethyl acetate and finally ether,
however very little organic material was present after evaporation
of the solvents and no Tröger’s base products were observed follow-
[1] J. Tröger, J. Prakt. Chem. 1887, 36, 225–245.
[2] V. Prelog, P. Wieland, Helv. Chim. Acta 1944, 27, 1127–1134.
[3] J. Jensen, K. Wärnmark, Synthesis 2001, 1873–1877.
[4] J. Jensen, M. Strozyk, K. Wärnmark, J. Heterocycl. Chem.
2003, 40, 373–375.
[5] A. P. Hansson, J. Jensen, O. F. Wendt, K. Wärnmark, Eur. J.
Org. Chem. 2003, 3179–3188.
[6] T. H. Webb, C. S. Wilcox, J. Org. Chem. 1990, 55, 363–365.
[7] C. Pardo, M. Ramos, A. Fruchier, J. Elguero, Magn. Reson.
Chem. 1996, 34, 708–710.
[8] W. W. K. R. Mederski, M. Baumgarth, M. Germann, D. Kux,
T. Weitzel, Tetrahedron Lett. 2003, 44, 2133–2136.
[9] Z. H. Li, X. Xu, Y. Peng, Z. Jiang, C. Ding, X. Qian, Synthesis
2005, 1228–1230.
1
ing H NMR analysis of the crude material.
Reduction of (؎)-2 with Continuous-Flow Apparatus: (Ϯ)-2 (100 mg,
0.32 mmol) was dissolved in warm ethanol (150 mL) and the warm
mixture was passed through a H-Cube equipped with a 30 mm 10%
Pd/C cartridge (1 mL/min; 35 °C; 70 bar). Fractions were collected,
which were monitored by TLC, and evaporation of solvent under
reduced pressure afforded (Ϯ)-32 (69 mg, 85%) as a white solid
that had identical spectral properties to those listed above.
[10] U. Kiehne, T. Weilandt, A. Lützen, Org. Lett. 2007, 9, 1283–
1286.
[11] M. D. H. Bhuiyan, P. Jensen, P. Turner, A. C. Try, Acta Crys-
tallogr., Sect. E 2007, 63, o1115-o1116.
[12] C. Solano, D. Svensson, Z. Olomi, J. Jensen, O. F. Wendt, K.
Wärnmark, Eur. J. Org. Chem. 2005, 3510–3517.
[13] K.-X. Zhu, D. C. Craig, A. C. Try, Acta Crystallogr., Sect. E
2008, 64, o1797.
[14] I. Sucholeiki, V. Lynch, L. Phan, C. S. Wilcox, J. Org. Chem.
1988, 53, 98–104.
[15] M. D. H. Bhuiyan, P. Jensen, A. C. Try, Acta Crystallogr., Sect.
E 2007, 63, o908-o909.
[16] M. D. H. Bhuiyan, P. Jensen, A. C. Try, Acta Crystallogr., Sect.
E 2007, 63, o4393.
[17] D. P. Becker, P. M. Finnegan, P. W. Collins, Tetrahedron Lett.
1993, 34, 1889–1892.
[18] M. D. H. Bhuiyan, A. C. Try, J. Klepetko, P. Turner, Acta Crys-
tallogr., Sect. E 2006, 62, o4887-o4888.
1,7-Diamino-2,4,8,10-tetramethyl-6H,12H-5,11-methanodibenzo-
[b,f][1,5]diazocine (33): Reduction of (Ϯ)-12 with Pd/C in ethanol:
A mixture of (Ϯ)-12 (1.0 g, 2.7 mmol), ethanol (75 mL) and 10%
palladium on carbon (50 mg) was stirred in the dark under hydro-
gen for 3 d. No evidence of a reaction was observed by TLC moni-
toring. The mixture was then filtered through Celite and unreacted
(Ϯ)-12 was recovered in quantitative yield.
Reduction of (؎)-12 with Tin(II) Chloride: (Ϯ)-12 (1.0 g, 2.7 mmol)
and tin(II) chloride (7.09 g, 31.4 mmol) were suspended in absolute
ethanol (75 mL) and refluxed at 75 °C for 2 d. The reaction mixture
was cooled to room temperature, then poured into the ice and neu-
tralised with sodium hydrogen carbonate. The reaction mixture was
extracted with dichloromethane (3ϫ100 mL). The combined or-
ganic layers were washed with brine, dried with anhydrous sodium
sulfate, filtered and the solvents evaporated to dryness. The crude
solid was chromatographed (silica gel, ethyl acetate) to afford (Ϯ)-
33 (456 mg, 55%) as an off-white solid; m.p. 102–104 °C. 1H NMR
[19] M. Faroughi, A. C. Try, P. Turner, Acta Crystallogr., Sect. E
2006, 62, o3893-o3894.
Eur. J. Org. Chem. 2009, 687–698
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
697

M. D. H. Bhuiyan, A. B. Mahon, P. Jensen, J. K. Clegg, A. C. Try
FULL PAPER
[20] C. M. L. Vande Velde, D. Dider, F. Blockhuys, S. Sergeyev,
Acta Crystallogr., Sect. E 2008, 64, o538.
[21] D. Didier, S. Sergeyev, Tetrahedron 2007, 63, 3864–3869.
[22] C. Pardo, E. Sesmilo, E. Gutiérrez-Puebla, A. Monge, J. El-
guero, A. Fruchier, J. Org. Chem. 2001, 66, 1607–1611.
[23] M. Valík, B. Dolenský, H. Pertrícková, V. Král, Collect. Czech.
Chem. Commun. 2002, 67, 609–621.
[24] T. Mas, C. Pardo, J. Elguero, ARKIVOC 2004, 4, 86–93.
[25] T. Mas, C. Pardo, F. Salort, J. Elguero, M. R. Torres, Eur. J.
Org. Chem. 2004, 1097–1104.
[26] T. Mas, C. Pardo, J. Elguero, Mendeleev Commun. 2004, 235–
237.
[27] T. Mas, C. Pardo, J. Elguero, Helv. Chim. Acta 2005, 88, 1199–
1207.
[28] B. Dolenský, M. Valík, D. Sýkora, V. Král, Org. Lett. 2005, 7,
67–70.
[29] U. Kiehne, A. Lützen, Eur. J. Org. Chem. 2007, 5703–5711.
Received: October 22, 2008
Published Online: December 19, 2008
698
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 687–698