Bromination and iodination of 6H,12H-5, 11-methanodibenzo[b,f][1,5] diazocine: A convenient entry to unsymmetrical analogs of Troeger's base

Article abstract of DOI:10.1002/ejoc.200700253

6H,12H-5,11-Methanodibenzo[b,f][1,5]diazocine can easily be prepared from aniline and paraformaldehyde. Its reactions with either N-bromosuccinimide (NBS) or with iodine monochloride (ICl) in the presence of a suitable activator smoothly afford 2-bromo- a

Full text of DOI:10.1002/ejoc.200700253

FULL PAPER
DOI: 10.1002/ejoc.200700253
Bromination and Iodination of 6H,12H-5,11-Methanodibenzo[b,f][1,5]diazocine:
A Convenient Entry to Unsymmetrical Analogs of Tröger’s Base
Delphine Didier[a] and Sergey Sergeyev*[a]
Keywords: Heterocycles / Electrophilic substitution / Tröger’s base / Bromination / Iodination
6H,12H-5,11-Methanodibenzo[b,f][1,5]diazocine can easily
be prepared from aniline and paraformaldehyde. Its reac-
tions with either N-bromosuccinimide (NBS) or with iodine
monochloride (ICl) in the presence of a suitable activator
smoothly afford 2-bromo- and 2-iodo- derivatives, respec-
tively. The combination of these halogenation methods pro-
vides access to the 8-bromo-2-iodo derivative, which is an
interesting and otherwise inaccessible intermediate for the
synthesis of unsymmetrical analogs of Tröger’s base. The re-
ported halogenations represent the first examples of electro-
philic substitution in 6H,12H-5,11-methanodibenzo[b,f][1,5]-
diazocine.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2007)
Generally, symmetrical analogs of Tröger’s base (Ϯ)-1
are synthesized by the condensation of anilines 2 with form-
aldehyde or its synthetic equivalent (Scheme 1).[3] It was ac-
cepted for a long time that anilines have to bear an electron-
donating substituent in the para position to the amino
group in order to give Tröger’s base analogs in this reaction.
However, the synthesis of symmetrical dihalogen analogs of
Tröger’s base (Ϯ)-1b,c was recently developed by
Wärnmark[4,5] and successfully optimized for multigram
preparations.[6,7] Dihalides (Ϯ)-1b,c are particularly valu-
able since they can be converted into a large variety of func-
tional derivatives by metal-catalyzed transformations.[8–10]
Introduction
Tröger’s base, (Ϯ)-2,8-dimethyl-6H,12H-5,11-methanodi-
benzo[b,f][1,5]diazocine [(Ϯ)-1a, Figure 1], was first synthe-
sized in 1887 from para-methylaniline and formaldehyde.
Tröger’s base is a chiral diamine with two stereogenic
bridge-head nitrogen atoms. A unique set of structural fea-
tures (C2 symmetry and a rigid V-shape geometry with the
two aromatic rings nearly perpendicular to each other) has
encouraged many elegant applications of Tröger’s base de-
rivatives in supramolecular chemistry and molecular re-
cognition.[1] Recently, increasingly sophisticated molecules
containing two or more fused methanodibenzodiazocine
units were also prepared.[2]
Figure 1. Tröger’s base: structural formula with numbering of
atoms according to IUPAC rules (left) and optimized geometry of
the (S,S) enantiomer (right).
Scheme 1. General synthesis for symmetrical (top) and unsymmet-
rical (bottom) analogs of Tröger’s base.
[a] Université Libre de Bruxelles (ULB), Laboratoire de Chimie
des Polymères,
In support of our studies of new building blocks for H-
bonded supramolecular assemblies, we are interested in an-
alogs of Tröger’s base bearing various substituents in one
aromatic ring with another aromatic ring being unsubsti-
C. P. 206/01, Boulevard du Triomphe, 1050 Bruxelles, Belgium
Fax: +32-2-6505410
E-mail: sserguee@ulb.ac.be
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Eur. J. Org. Chem. 2007, 3905–3910
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3905

D. Didier, S. Sergeyev
FULL PAPER
tuted. The corresponding unsymmetrical halogen deriva- However, in this crucial step, transformation of 7 into (Ϯ)-
tives of 6H,12H-5,11-methanodibenzo[b,f][1,5]diazocine ap- 4 occurred with a disappointingly poor yield (25%). In ad-
peared to be attractive intermediates for these molecules. dition, the yields of intermediates 6 and 7 were modest, and
However, known syntheses of such unsymmetrical halogen the separation of 7 from minor amounts of side products,
derivatives still leave room for improvement. Indeed, the ap- which resulted from the reductive debromination of 6 and
proach based on desymmetrization by metal-catalyzed the action of LiAlH4 upon 7, was somewhat tedious. This
cross-coupling reactions[11] is useful but limited in its scope route to (Ϯ)-4 thus appears to be impractical and we de-
as it introduces only particular functional groups in one of cided to develop an alternative, shorter and more practical
the aromatic rings. The desymmetrization of a 2,8-dibromo synthesis.
analog of Tröger’s base (Ϯ)-1b by selective monolithiation
To this end, 6H,12H-5,11-methanodibenzo[b,f][1,5]di-
with nBuLi followed by the quenching with a proton azocine (Ϯ)-8 appeared to be an attractive starting point.
source[12] is complicated by the formation of side products, Indeed, (Ϯ)-8 can be viewed as a structural analog of N,N-
requires very careful temperature control because of the in- dimethylaniline and should therefore be prone to aromatic
stability of the Li intermediate, and is therefore difficult to electrophilic substitutions. However, to the best of our
reproduce.[11] Therefore, we decided to develop a new, ex- knowledge, no example of electrophilic substitution on this
perimentally simple and potentially scalable method for the heterocyclic system has been published earlier. The most
synthesis of 2-halo-6H,12H-5,11-methanodibenzo[b,f][1,5]- straightforward access to (Ϯ)-8 could obviously be granted
diazocines.
by a direct condensation of aniline with formaldehyde or
its equivalent. However, such a condensation of anilines
lacking an electron-donating substituent in the para posi-
tion to the amino group was previously considered not feas-
ible.[3] At the same time, the above-discussed reaction of
anilines 2b,c bearing moderately electron-withdrawing hal-
ogens with paraformaldehyde in TFA gives fair-to-excellent
yields of the halogen analogs of Tröger’s base (Ϯ)-1b,c
(Scheme 1). As expected, the direct condensation of aniline
with paraformaldehyde in TFA gave 78% of (Ϯ)-8. Pleas-
antly, we found this reaction to be well reproducible up to
a 50-mmol scale (Scheme 3). With this practical synthesis
of (Ϯ)-8 in hand, we investigated a few halogenation meth-
ods to access 2-bromo- and 2-iodo-6H,12H-5,11-methano-
dibenzo[b,f][1,5]diazocine [(Ϯ)-4 and (Ϯ)-9, respectively].
Results and Discussion
The general method for the synthesis of unsymmetrical
Tröger’s base analogs bearing different substituents in the
2- and 8-positions relies on the sequential buildup of amine
derivatives such as 3a followed by the condensation with
formaldehyde to deliver the intact polycyclic system 1.[13]
All reported examples of this transformation deal with
amines
3
bearing electron-donating substituent
X
(Scheme 1). However, the above-mentioned synthesis of 2,8-
dihalogen analogs of Tröger’s base clearly demonstrates fea-
sibility of electrophilic substitution in the aromatic ring
bearing a halogen atom in the para position to the amino
group. Therefore, we investigated the possibility to use this
approach in the synthesis of bromo derivative (Ϯ)-4
(Scheme 2).
Scheme 2. Synthesis of monobromide (Ϯ)-4 from isatoic anhydride
and 4-bromoaniline; TFA = trifluoroacetic acid, CF3COOH.
Scheme 3. Synthesis of unsymmetrical dihalogen derivatives of
6H,12H-5,11-methanodibenzo[b,f][1,5]diazocine; TFA = trifluoro-
acetic acid, DMF = N,N-dimethylformamide, NBS = N-bromosuc-
cinimide, Tf = trifluoromethylsulfonyl, CF3SO2.
The reaction of isatoic anhydride (1H-benzo[d][1,3]oxaz-
ine-2,4-dione, 5) with 4-bromoaniline (2b) afforded amide 6
(45%), which was then reduced to diamine 7 with LiAlH4
in THF (56%). The subsequent condensation of diamine
7 with paraformaldehyde in trifluoroacetic acid (TFA) was
expected to deliver bromide (Ϯ)-4 without any problem
Among the plethora of reagents used for the bromination
since the bromo substituted aromatic ring in 7 should be of arenes, N-bromosuccinimide (NBS) seems to remain the
sufficiently activated towards electrophilic substitutions. most widely used one due to its versatility, low toxicity, and
3906
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 3905–3910

Bromination and Iodination of 6H,12H-5,11-Methanodibenzo[b,f][1,5]diazocine
FULL PAPER
ease of handling. Bromination of N,N-dialkylanilines with Under these conditions, only 28% of iodide (Ϯ)-9 was iso-
NBS to give the corresponding 4-bromo derivatives pro- lated together with 53% of the starting material (Table 1,
ceeds at room temp. in almost quantitative yield.[14]
Entry 1). The prolongation of the reaction time to 120 h
When (Ϯ)-8 was treated with 2 equiv. of NBS in DMF, (Table 1, Entry 2) had nearly no effect on the yield. In an
30% of monobromide (Ϯ)-4 together with 30% of dibro- attempt to overcome the low reactivity of (Ϯ)-8 we studied
mide (Ϯ)-1b were isolated after 18 h at room temp. These the influence of the amount of catalyst on the yield of io-
results indicated much lower reactivity of (Ϯ)-8 relative to dination products (Table 1, Entries 3–5). The best results in
that of N,N-dialkylanilines. This is undoubtedly due to poor terms of conversion of the starting material were obtained
conjugation between the aromatic ring and the lone pair of in the presence of 1–2 equiv. of Hg(OTf)2. However, notable
electrons of the bridgehead nitrogen atom in (Ϯ)-8. How- amounts of diiodide (Ϯ)-1c resulting from the iodination
ever, this conjugation is not negligible since substitution oc- of both aromatic rings in (Ϯ)-8 were isolated. A possible
curs exclusively at the 2- and 8-positions of 6H,12H-5,11- explanation to these observations can be that complexation
methanodibenzo[b,f][1,5]diazocine and no isomeric prod- between (Ϯ)-8 and a Hg2+ cation may inhibit the catalytic
ucts were isolated. When the amount of NBS was decreased activity of the latter and at least 1 equiv. of Hg(OTf)2 is
to 1.2 equiv., only traces of dibromide (Ϯ)-1b were detected, necessary. This is supported by the fact that iodination of
the yield of bromide (Ϯ)-4 was 37%, and 40% of the start- acetanilide required at least 0.5 equiv. of an activating
ing material were recovered. Finally, prolonged reaction Lewis acid.[18] A further increase in the amount of Hg-
times (90 h) resulted in an increase in the yield of (Ϯ)-4 (OTf)2 was not necessary and resulted in a lower yield of
to 57%. In all reactions, (Ϯ)-4 was easily separated from (Ϯ)-9.
dibromide (Ϯ)-1b and from starting material (Ϯ)-8 by col-
When the iodination of (Ϯ)-8 was conducted with
umn chromatography on SiO2.
3 equiv. of ICl in the presence of 2 equiv. of Hg(OTf)2, the
Iodide (Ϯ)-9 appears to be an even more appealing syn- complete conversion to diiodide (Ϯ)-1c (65%) was achieved
thetic intermediate than bromide (Ϯ)-4 as iodides generally (Table 1, Entry 6). This transformation is of relatively little
demonstrate superior reactivity in many synthetically useful preparative value since diiodide (Ϯ)-1c can be prepared in
transformations. At the same time, the direct iodination of comparable yield (56–62% on 20 to 100-mmol scale)[7] di-
aromatic substrates is known to be more challenging due to rectly from commercially available 4-iodoaniline (2c). How-
the poorer electrophilicity of molecular iodine relative to ever, it is methodologically interesting as a first example of
chlorine and bromine. To generate more reactive iodine spe- the twofold electrophilic substitution of 6H,12H-5,11-meth-
cies, many iodination procedures use harsh conditions and anodibenzo[b,f][1,5]diazocine. On the contrary, when the
involve strong oxidative agents[15] that are likely incompat- amount of ICl was decreased to 1.1 equiv. while keeping the
ible with 6H,12H-5,11-methanodibenzo[b,f][1,5]diazocine. amount of Hg(OTf)2 unchanged, iodide (Ϯ)-9 was isolated
In contrast to aromatic bromination and chlorination, ca- in an appreciable 55% yield (Table 1, Entry 7). Under these
talysis with Lewis acids is less efficient in iodination with optimized conditions, nearly complete conversion of the
molecular iodine. Nevertheless, some softer Lewis acids starting material was observed and only 7% of twofold io-
such as mercury[16] and silver[17] salts have been used in the dination product (Ϯ)-1c was isolated.
iodinations of activated substrates with I2. Recently, inter-
Next to Hg(OTf)2, less toxic In(OTf)3 was reported to
halogen compounds such as ICl and IBr in combination be a highly efficient activator in the iodination of various
with various Lewis acids were suggested as a mild, yet electron-rich aromatic substrates upon the action of ICl.
highly active, iodination system. In a model study, Hg- However, with the use of the optimized procedure
(OTf)2 was shown to be the most potent activator in the (1.1 equiv. of ICl, 2 equiv. of Lewis acid) and In(OTf)3 in
iodination of acetanilide with ICl.[18]
exchange for Hg(OTf)2, only 32% of iodide (Ϯ)-9 was iso-
Following the published results with acetanilide, we lated together with 51% of the unreacted starting material.
started experiments on the iodination of (Ϯ)-8 with On the other hand, a larger excess of the iodinating reagent
1.5 equiv. of ICl in the presence of 0.5 equiv. of Hg(OTf)2. resulted in an increase in the yield of (Ϯ)-9 to 51% without
Table 1. Yields of products for the reaction of (Ϯ)-8 (1.0 mmol) with ICl in MeCN (5 mL) at room temp.[a]
Entry
ICl [mmol]
Time [h]
Additive [mmol]
(Ϯ)-8 [%][b]
(Ϯ)-9 [%][b]
(Ϯ)-1c [%][b]
1
2
3
4
5
6
7
8
9
1.5
1.5
1.5
1.5
1.5
3
1.1
1.1
2.5
24
120
72
72
72
72
72
72
72
Hg(OTf)2 (0.5)
Hg(OTf)2 (0.5)
Hg(OTf)2 (1.0)
Hg(OTf)2 (2.0)
Hg(OTf)2 (5.0)
Hg(OTf)2 (2.0)
Hg(OTf)2 (2.0)
In(OTf)3 (2.0)
In(OTf)3 (1.0)
53
51
8
10
28
25
48
47
traces[c]
traces[c]
24
27
26
6
34
traces[c]
traces[c]
43
traces[c]
55
65
7
32
51
traces[c]
traces[c]
31
1
[a] No other products were detected by TLC or H NMR spectroscopy. [b] Isolated yield of analytically pure substances. [c] Less than
5%.
Eur. J. Org. Chem. 2007, 3905–3910
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3907

D. Didier, S. Sergeyev
FULL PAPER
detectable formation of diiodide (Ϯ)-1c (Table 1, Entries 8, benzo[b,f][1,5]diazocine. Although our study revealed rela-
9). Thus, In(OTf)3 appears to be a less potent activator in tively low reactivity of this heterocyclic system compared to
the iodination of (Ϯ)-8 relative to Hg(OTf)2. Nevertheless, N,N-dialkylanilines, it clearly demonstrated the feasibility
optimized isolated yields of (Ϯ)-9 obtained with both acti- of the perfectly regioselective substitution in the para posi-
vators were comparable (cf. Table 1, Entries 7 and 9).
tion to the nitrogen atom. Moreover, we showed that such
In order to eliminate the necessity for toxic metal salts, a substitution can be performed in either one or two aro-
we decided to test other mild procedures for the iodination matic rings of the heterocyclic system. Therefore, methano-
of (Ϯ)-8. For example, nearly quantitative iodination of dibenzodiazocine (Ϯ)-8 probably has an even greater syn-
N,N-dimethylaniline upon the action of I2 and pyridine/di- thetic potential as a substrate for other electrophilic substi-
oxane was reported.[19] Unfortunately, (Ϯ)-8 was unreactive tutions. The preparative utility of (Ϯ)-8 can certainly be
under these conditions: the starting material was nearly limited by its diminished reactivity caused by the poor con-
quantitatively recovered after 24 h at room temp. and no jugation between the aromatic ring and the nitrogen atom.
iodination products were detected by TLC or 1H NMR Nevertheless, we are currently examining other electrophilic
spectroscopy of the reaction mixture. This results confirmed substitutions of (Ϯ)-8 that may result in interesting building
the much lower reactivity of (Ϯ)-8 in electrophilic halogena- blocks and intermediates.
tion reactions relative to N,N-dialkylanilines.
Finally, interesting derivative (Ϯ)-10 bearing bromine
and iodine atoms in two different aromatic rings of
Experimental Section
6H,12H-5,11-methanodibenzo[b,f][1,5]diazocine was pre-
General: All chemicals were purchased from Aldrich or Acros and
pared in 68% yield by bromination of iodide (Ϯ)-9 with
used without further purification unless stated otherwise. Technical
NBS in DMF; this compound is not accessible by other
grade solvents were distilled before use. THF was heated at reflux
synthetic routes. Both bromine and iodine can serve as leav-
over sodium and benzophenone until a blue–violet color persisted
ing groups in many cross-coupling reactions. At the same
and then distilled directly into the reaction flask. Iodine mono-
chloride (ICl) was purchased from Aldrich as a 1.0  solution in
dichloromethane. Column chromatography was performed with
SiO2 Kieselgel 60 (Macherey–Nagel, particle size 0.04–0.063 mm).
time, the differences in their reactivity allow selective substi-
tution of iodine and leave bromine intact for subsequent
transformations (representative examples: see ref.[20] and
references cited therein). Therefore, (Ϯ)-10 can undoubtedly
TLC was performed with precoated SiO2 plates Kieselgel 60F254
be considered as a very promising intermediate in the syn- (Merck). 1H NMR (300 MHz) and 13C NMR (75 MHz) spectra
were recorded with a Bruker Avance 300 spectrometer; chemical
shifts (δ) are given in ppm relative to Me4Si (internal standard);
coupling constants (J) are given in Hz. Electron impact mass spec-
tra (EIMS) were recorded with a Waters AutoSpec 6F instrument;
m/z with the lowest isotopic mass are reported unless stated other-
wise. Melting points were determined with a Nikon Eclipse 80i
microscope equipped with a Mettler hot stage. Elemental analyses
were carried out at the University of Mainz, Germany.
thesis of complex unsymmetrical derivatives of Tröger’s
base.
Conclusions
We developed convenient methods for the synthesis of
6H,12H-5,11-methanodibenzo[b,f][1,5]diazocine
[(Ϯ)-8]
2-Amino-N-(4-bromophenyl)benzamide (6): 4-Bromoaniline (2b;
3.44 g, 20 mmol) was added in portions to the suspension of isatoic
anhydride (5; 13.05 g, 80 mmol) in dry EtOH (160 mL). The mix-
ture was heated to reflux for 24 h, then allowed to reach room
temp., filtered, and poured into H2O (1 L). The precipitated solid
was collected by filtration and dissolved in CH2Cl2 (100 mL). The
resulting solution was filtered, dried with MgSO4, and concen-
and its transformation into unsymmetrical bromo- and
iodo derivatives (Ϯ)-4 and (Ϯ)-9. In our opinion, this is a
useful and experimentally simple approach to these valuable
synthetic intermediates for many valuable transformations
that were earlier reported for symmetrical dihalo analogs of
Tröger’s base.[8,9,11] Bromide (Ϯ)-4 was prepared in 44% to-
tal yield starting from aniline. This can be viewed as a com- trated in vacuo. Purification of the residue by column chromatog-
raphy (CH2Cl2) afforded 6 (2.61 g, 45%) as a colorless solid. Rf =
0.31 (CH2Cl2). H NMR (300 MHz, CDCl3, 25 °C): δ = 5.48 (br.
petitive alternative to the known synthesis of (Ϯ)-4 by de-
symmetrization of dibromide (Ϯ)-1b (52% starting from 4-
bromoaniline).[12] As to iodide (Ϯ)-9 (43% starting from
aniline), its synthesis was not described until now, but diffi-
culties in the attempted desymmetrization of diiodide (Ϯ)-
1c were reported.[11] Unsymmetrical derivative (Ϯ)-10 is an
even more interesting synthetic intermediate since it allows
for the exploration of the different reactivities of bromine
and iodine in carbon–carbon or carbon–heteroatom bond
formation. In addition, electrophilic halogenation followed
by Pd-catalyzed amination[8] can be a very useful iterative
strategy in the recently described rational synthesis of
“fused tris-Tröger’s bases”.[21]
1
s, 2 H, NH2), 6.68–6.73 (m, 2 H), 7.23–7.28 (m, 1 H), 7.42–7.46
(m, 5 H), 7.75 (s, 1 H, NHCO) ppm. 13C NMR (75 MHz, CDCl3,
25 °C): δ = 115.7, 116.8, 117.0, 117.6, 121.9 (2 C), 127.1, 131.9 (2
C), 132.9, 136.9, 148.9, 167.4 ppm. HREIMS: calcd. for
C13H1181BrN2O [M]+ 292.0034; found 292.0011.
2-Amino-N-(4-bromophenyl)benzylamine (7): A solution of amide 6
(291 mg, 1 mmol) in THF (5 mL) was added to the suspension of
LiAlH4 (76 mg, 2 mmol) in THF (5 mL). The mixture was heated
to reflux for 5 h, allowed to reach room temp., and quenched by
the dropwise addition of H2O (0.2 mL). The resulting mixture was
filtered, and the solid on the filter was washed twice with CH2Cl2.
Combined filtrates were concentrated in vacuo, and the residue was
The reported halogenation represents the first example
of an electrophilic substitution of 6H,12H-5,11-methanodi- 56%) as a yellow solid. Rf = 0.31 (CH2Cl2). H NMR (300 MHz,
purified by column chromatography (CH2Cl2) to give 7 (151 mg,
1
3908
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 3905–3910

Bromination and Iodination of 6H,12H-5,11-Methanodibenzo[b,f][1,5]diazocine
FULL PAPER
CDCl3, 25 °C): δ = 3.91 (br. s, 3 H, NH, NH2), 4.14 (s, 2 H, CH2),
6.56 and 7.27 (AAЈBBЈ system, 4 H), 6.68–6.76 (m, 2 H), 7.12–7.16
a pale yellow solid. Rf = 0.36 (CH2Cl2/AcOEt, 80:20); m.p. 121–
1
2
123 °C. H NMR (300 MHz, CDCl3, 25 °C): δ = 4.12 (d, JH,H
=
=
=
=
2
2
(m, 2 H) ppm. 13C NMR (75 MHz, CDCl3, 25 °C): δ = 46.4, 109.6, 16.8 Hz, 1 H), 4.14 (d, JH,H = 16.8 Hz, 1 H), 4.25 (d, JH,H
2
2
114.6 (2 C), 115.6, 118.1, 128.6 (2 C), 129.6, 131.6 (2 C), 145.1,
146.8 ppm. HREIMS: calcd. for C13H1381BrN2 [M]+ 278.0242;
found 278.0229.
12.7 Hz, 1 H), 4.32 (d, JH,H = 12.7 Hz, 1 H), 4.64 (d, JH,H
2
3
16.6 Hz, 1 H), 4.69 (d, JH,H = 16.6 Hz, 1 H), 6.88 (d, JH,H
3
4
8.5 Hz, 1 H), 6.90 (dd, JH,H = 7.6 Hz, JH,H = 0.8 Hz, 1 H), 6.99
(td, 3JH,H = 7.6 Hz, JH,H = 1.5 Hz, 1 H), 7.12 (td, 3JH,H = 8.0 Hz,
4
(؎)-6H,12H-5,11-Methanodibenzo[b,f][1,5]diazocine [(؎)-8]: Ani-
line (4.68 g, 50.2 mmol) and then paraformaldehyde (3.06 g,
102 mmol) were added in portions with vigorous stirring to TFA
(100 mL) at –15 °C. The resulting mixture was stirred for 20 min
at –15 °C. The mixture was allowed to reach room temp., stirred
for 48 h, and then slowly added to a stirred mixture of ice and an
excess of NH3 (25% solution in water) at 0 °C. The resulting mix-
ture was extracted with CH2Cl2 (3ϫ200 mL), dried with MgSO4,
and concentrated in vacuo. Purification of the residue by column
chromatography (CH2Cl2/AcOEt, 95:5) afforded (Ϯ)-8 (4.34 g,
78%). Rf = 0.24 (CH2Cl2/AcOEt, 95:5); m.p. 130–131 °C (ref.[22]
m.p. 127–128 °C). Spectral data of (Ϯ)-8 were identical to those
published earlier.[12]
4JH,H = 1.5 Hz, 1 H), 7.18 (dd, JH,H = 8.0 Hz, JH,H = 1.5 Hz, 1
3
4
4
3
4
H), 7.24 (d, JH,H = 2.1 Hz, 1 H), 7.44 (dd, JH,H = 8.4 Hz, JH,H
= 2.1 Hz, 1 H) ppm. 13C NMR (75 MHz, CDCl3, 25 °C): δ = 58.1,
58.7, 66.7, 87.4, 124.2, 125.0, 127.0, 127.1, 127.53, 127.55, 130.5,
135.7, 136.2, 147.7, 148.0 ppm. HREIMS: calcd. for C15H13IN2
[M]+ 348.0123; found 348.0122. C15H13IN2 (348.0): calcd. C 51.74,
H 3.76, N 8.05; found C 51.27, H 3.81, N 7.90.
(؎)-2,8-Diiodo-6H,12H-5,11-methanodibenzo[b,f][1,5]diazocine
[(؎)-1c]: A solution of ICl in CH2Cl2 (3.0 mL, 3.0 mmol) was
added under an atmosphere of argon to a stirred solution of (Ϯ)-
8 (223 mg, 1.0 mmol) and Hg(OTf)2 (997 mg, 2.0 mmol) in dry
CH3CN (5 mL). The resulting mixture was stirred for 72 h at room
temp., and then triethylamine (3.0 mL) was added followed by the
addition of a saturated aqueous solution of Na2S2O3 until disap-
pearance of the brown color. The resulting mixture was filtered
trough Celite, which was then additionally washed with CH2Cl2
(50 mL). The organic layer was separated and washed with water
(3ϫ50 mL), dried with MgSO4, and concentrated in vacuo. Purifi-
cation of the residue by column chromatography (gradient from
CH2Cl2 to CH2Cl2/AcOEt, 95:5) afforded (Ϯ)-1c (310 mg, 65%) as
a pale yellow solid. Rf = 0.54 (CH2Cl2/AcOEt, 80:20); m.p. 178.5–
180 °C (ref.[4] m.p. 180.5–181.5 °C). Other analytical data were
identical with those published earlier.[4]
(؎)-2-Bromo-6H,12H-5,11-methanodibenzo[b,f][1,5]diazocine [(؎)-
4]: Method A: Amine 7 (138 mg, 0.5 mmol) and then paraformalde-
hyde (30 mg, 1 mmol) were added in portions to TFA (2 mL) at
–15 °C. The mixture was allowed to reach room temp., stirred for
72 h, and then slowly added to a stirred mixture of ice and an
excess of NH3 (25% solution in water) at 0 °C. The resulting mix-
ture was extracted with CH2Cl2 (3ϫ50 mL), dried with MgSO4,
and concentrated in vacuo. Column chromatography (SiO2;
CH2Cl2/AcOEt, 5:1) afforded (Ϯ)-4 (37 mg, 25%). Method B: A
solution of NBS (220 mg, 1.24 mmol) in DMF (2.0 mL) was added
at 0 °C to a stirred solution of (Ϯ)-8 (227 mg, 1.02 mmol) in DMF
(2.0 mL). The resulting mixture was stirred for 15 min at 0 °C, then
allowed to reach room temp., and stirred for another 92 h. The
reaction mixture was then diluted with CH2Cl2 (50 mL), the or-
ganic layer was washed with water (3ϫ80 mL), dried with MgSO4,
and concentrated in vacuo. Purification of the residue by column
chromatography (CH2Cl2/AcOEt, 95:5) afforded (Ϯ)-4 (175 mg,
57%). Pale yellow solid; Rf = 0.36 (CH2Cl2/AcOEt, 5:1); m.p. 119–
121 °C (ref.[12] m.p. 123.5–125 °C). 1H NMR (300 MHz, CDCl3,
25 °C): δ = 4.12 (d, 2JH,H = 16.7 Hz, 1 H), 4.15 (d, 2JH,H = 16.8 Hz,
(؎)-8-Bromo-2-iodo-6H,12H-5,11-methanodibenzo[b,f][1,5]diaz-
ocine [(؎)-10]: A solution of iodide (Ϯ)-9 (162 mg, 0.465 mmol)
and NBS (91.0 mg, 0.511 mmol) in DMF (2.5 mL) was stirred for
72 h at room temp. The resulting mixture was diluted with CH2Cl2
(10 mL). The organic layer was separated, washed with water
(3ϫ10 mL), dried with MgSO4, and concentrated in vacuo. Purifi-
cation of the residue by column chromatography (gradient from
CH2Cl2 to CH2Cl2/AcOEt, 95:5) afforded (Ϯ)-10 (134 mg, 68%) as
a pale yellow solid. Rf = 0.50 (CH2Cl2/AcOEt, 5:1); m.p. 134–
2
2
1 H), 4.25 (d, JH,H = 12.9 Hz, 1 H), 4.32 (d, JH,H = 12.9 Hz, 1
1
2
136 °C. H NMR (300 MHz, CDCl3, 25 °C): δ = 4.08 (d, JH,H
=
=
=
=
2
2
H), 4.65 (d, JH,H = 16.8 Hz, 1 H), 4.69 (d, JH,H = 16.7 Hz, 1 H),
6.90 (dd, JH,H = 6.9 Hz, JH,H = 0.9 Hz, 1 H), 6.98 (td, JH,H
7.5 Hz, JH,H = 1.6 Hz, 1 H), 7.01 (d, JH,H = 8.6 Hz, 1 H), 7.04
(d, JH,H = 2.2 Hz, 1 H), 7.12 (td, JH,H = 7.5 Hz, JH,H = 1.5 Hz,
2
4
16.8 Hz, 1 H), 4.09 (d, JH,H = 16.8 Hz, 1 H), 4.25 (t, JH,H
1.2 Hz, 2 H), 4.62 (d, JH,H = 16.8 Hz, 1 H), 4.63 (d, JH,H
3
4
3
=
2
2
4
3
3
3
16.8 Hz, 1 H), 6.87 (d, JH,H = 8.4 Hz, 1 H), 6.99 (d, JH,H
4
3
4
8.7 Hz, 1 H), 7.05 (d, 4JH,H = 2.1 Hz, 1 H), 7.24 (d, 4JH,H = 2.0 Hz,
3
4
1 H), 7.19 (dd, JH,H = 8.0 Hz, JH,H = 1.5 Hz, 1 H), 7.26 (dd,
3
4
1 H), 7.27 (dd, JH,H = 8.6 Hz, JH,H = 2.2 Hz, 1 H), 7.46 (dd,
3JH,H = 8.6 Hz, JH,H = 2.2 Hz, 1 H) ppm. 13C NMR (75 MHz,
4
3JH,H = 8.4 Hz, JH,H = 2.1 Hz, 1 H) ppm. 13C NMR (75 MHz,
4
CDCl3, 25 °C): δ = 58.4, 58.7, 66.7, 116.5, 124.2, 125.0, 126.8,
127.0, 127.5 (2 C), 129.7, 130.0, 130.4, 147.2, 147.7 ppm. HREIMS:
calcd. for C15H1381BrN2 [M]+ 302.0242; found 302.0247.
CDCl3, 25 °C): δ = 58.1, 58.3, 66.6, 87.6, 116.8, 126.7, 127.0,
129.67, 129.70, 130.1, 130.6, 135.7, 136.4, 146.8, 147.6 ppm. HRE-
IMS: calcd. for C15H1281BrIN2 [M]+ 427.9208; found 427.9196.
C15H12BrIN2 calcd. C 42.18, H 2.83, N 6.56; found C 41.96, H
2.92, N 6.48.
(؎)-2-Iodo-6H,12H-5,11-methanodibenzo[b,f][1,5]diazocine [(؎)-9]:
A solution of ICl in CH2Cl2 (1.1 mL, 1.1 mmol) was added under
an atmosphere of argon to a stirred solution of (Ϯ)-8 (222 mg,
1.0 mmol) and Hg(OTf)2 (997 mg, 2.0 mmol) in dry CH3CN
(5 mL). The resulting reaction mixture was stirred at room temp.
for 72 h, and then triethylamine (1.5 mL) was added followed by
the addition of a saturated aqueous solution of Na2S2O3 until dis-
appearance of the brown color. The resulting mixture was filtered
through Celite, which was then additionally washed with CH2Cl2
(50 mL). The organic layer was separated and washed with water
(3ϫ50 mL), dried with MgSO4, and concentrated in vacuo. Purifi-
cation of the residue by column chromatography (gradient from
Supporting Information (see footnote on the first page of this arti-
1
cle): H and 13C NMR spectra of (Ϯ)-4, (Ϯ)-9, and (Ϯ)-10.
Acknowledgments
S. S. is indebted to Prof. Y. Geerts (Université Libre de Bruxelles)
for the opportunity to conduct an independent research program
in his laboratory and for generous financial support. We thank A.
CH2Cl2 to CH2Cl2/AcOEt, 95:5) afforded (Ϯ)-9 (192 mg, 55%) as Mujawase and J.-F. Picron for experimental assistance.
Eur. J. Org. Chem. 2007, 3905–3910
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3909

D. Didier, S. Sergeyev
FULL PAPER
[11] F. Hof, M. Schär, D. M. Scofield, F. Fischer, F. Diederich, S.
Sergeyev, Helv. Chim. Acta 2005, 88, 2333–2344.
[12] J. Jensen, J. Tejler, K. Wärnmark, J. Org. Chem. 2002, 67, 6008–
6014.
[13] T. H. Webb, C. S. Wilcox, J. Org. Chem. 1990, 55, 363–365.
[14] K. Staub, G. A. Levina, S. Barlow, T. C. Kowalczyk, H. S.
Lackritz, M. Barzoukas, A. Fort, S. R. Marder, J. Mater.
Chem. 2003, 13, 825–833.
[15] a) E. B. Merkushev, Synthesis 1988, 923–937; b) J. R. Hanson,
J. Chem. Res. 2006, 277–280.
[16] a) K. Orito, T. Hatakeyama, M. Takeo, H. Suginome, Synthesis
1995, 1273–1277; b) A. Bachki, F. Foubelo, M. Yus, Tetrahe-
dron 1994, 50, 5139–5146.
[1] a) M. Valík, R. M. Strongin, V. Král, Supramol. Chem. 2005,
17, 347–367; b) B. Dolensky, J. Elguero, V. Král, C. Pardo, M.
Valík, Adv. Heterocycl. Chem. 2007, 93, 1–56; c) J. C. Adrian Jr,
C. S. Wilcox, J. Am. Chem. Soc. 1992, 114, 1398–1403; d) S.
Paliwal, S. Geib, C. S. Wilcox, J. Am. Chem. Soc. 1994, 116,
4497–4498; e) F. Hof, D. M. Scofield, W. B. Schweizer, F. Died-
erich, Angew. Chem. Int. Ed. 2004, 43, 5056–5059; f) S.
Sergeyev, F. Diederich, Angew. Chem. Int. Ed. 2004, 43, 1738–
1740.
[2] a) J. Artacho, P. Nilsson, K.-E. Bergquist, O. F. Wendt, K.
Wärnmark, Chem. Eur. J. 2006, 12, 2692–2701; b) M. Havlik,
V. Kral, B. Dolensky, Org. Lett. 2006, 8, 4867–4870; c) T. Mas,
C. Pardo, J. Elguero, Helv. Chim. Acta 2005, 88, 1199–1207.
[3] I. Sucholeiki, V. Lynch, L. Phan, C. S. Wilcox, J. Org. Chem.
1988, 53, 98–104.
[4] J. Jensen, K. Wärnmark, Synthesis 2001, 1873–1877.
[5] A. Hansson, J. Jensen, O. F. Wendt, K. Wärnmark, Eur. J. Org.
Chem. 2003, 3179–3188.
[6] J. Jensen, M. Strozyk, K. Wärnmark, J. Heterocycl. Chem.
2003, 40, 373–375.
[7] S. Sergeyev, M. Schär, P. Seiler, O. Lukoyanova, L. Echegoyen,
F. Diederich, Chem. Eur. J. 2005, 11, 2284–2294.
[8] D. Didier, S. Sergeyev, Tetrahedron 2007, 63, 3864–3869.
[9] a) J. Jensen, M. Strozyk, K. Wärnmark, Synthesis 2002, 2761–
2765; b) U. Kiehne, A. Lützen, Synthesis 2004, 1687–1695.
[10] S. P. Bew, L. Legentil, V. Scholier, S. V. Sharma, Chem. Com-
mun. 2007, 389–391.
[17] W. W. Sy, Tetrahedron Lett. 1993, 34, 6223–6224.
[18] R. Johnsson, A. Meijer, U. Ellervik, Tetrahedron 2005, 61,
11657–11663.
[19] C. Monnereau, E. Blart, F. Odobel, Tetrahedron Lett. 2005, 46,
5421–5423.
[20] a) D. J. Sinclair, M. S. Sherburn, J. Org. Chem. 2005, 70, 3730–
3733; b) A. Blaszczyk, M. Chadim, C. von Haenisch, M.
Mayor, Eur. J. Org. Chem. 2006, 3809–3825.
[21] A. Hansson, T. Wixe, K.-E. Bergquist, K. Wärnmark, Org.
Lett. 2005, 7, 2019–2022.
[22] F. C. Cooper, M. W. Partridge, J. Chem. Soc. 1955, 991–994.
Received: March 21, 2007
Published Online: June 18, 2007
3910
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 3905–3910