Rearrangement of fused tetracyclic heterocycles induced by alkyl halides and formation of a new type of 'proton sponge'

Article abstract of DOI:10.1016/j.tet.2008.11.057

Intramolecular criss-cross cycloaddition of 3-substituted symmetrical homoallenyl azines 2 by heating in xylene lead to interesting fused heterocyclic systems consisting of four five-membered rings with two nitrogen atoms in the skeleton 3. These compound

Full text of DOI:10.1016/j.tet.2008.11.057

Tetrahedron 65 (2009) 792–797
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Rearrangement of fused tetracyclic heterocycles induced by alkyl halides
and formation of a new type of ‘proton sponge’
*
´
´ ˇ
Hana Zachova, Stanislav Man, Jan Taraba, Milan Potacek
ˇ
´
´
Department of Chemistry, Faculty of Science, Masaryk University of Brno, Kotlarska 2, CZ-611 37 Brno, Czech Republic
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 29 July 2008
Received in revised form 5 November 2008
Accepted 20 November 2008
Available online 25 November 2008
Intramolecular criss-cross cycloaddition of 3-substituted symmetrical homoallenyl azines 2 by heating in
xylene lead to interesting fused heterocyclic systems consisting of four five-membered rings with two
nitrogen atoms in the skeleton 3. These compounds 3 were found to be sensitive to attack by alkyl
halides. Their presence, depending on the reaction conditions, resulted in a new type of rearrangement
leading to compounds 4 and 5, respectively. With an excess of alkyl halide and in the presence of
NaBH3CN a new structure 6 with signs of the molecule corresponding a ‘proton sponge’ moiety was
created. The scope of the rearrangement and reaction products structure was investigated.
Ó 2008 Elsevier Ltd. All rights reserved.
n
1. Introduction
n
n
N
N
lateral
N
N
N
N
central
In the past we have reported some of our achievements in the
applications of allene derivatives, especially those starting with
homoallenyl aldehyde 1 and its derivatives. Thus, we have recently
published a paper on an easy and interesting transformation of
substituted homoallenyl oximes to stable nitrones,1 and their ap-
plication either in further transformation with exploitation of the
nitrone moiety, or other transformations on the cyclic skeleton.2
A rather rich chapter deals with application of symmetrical
homoallenyl azines 2 in criss-cross cycloaddition reaction.
Criss-cross cycloadditions represent a way to the formation of
fused heterocyclic compounds in a ‘one pot’ arrangement, but most
of them described in the literature are intermolecular processes
leading to two fused five-membered rings.3,4 Only few are intra-
molecular reactions. Thus, Suschitzky and Marthur5 performed
intramolecular reactions on benzalazines and naphthalene azines,
respectively, having in the ortho position to the azine, allyloxy and
propargyloxy groups, respectively. In the case where the other
ortho position to the alkenyloxy group was blocked in order to
prevent Claisen rearrangement, the reaction led to formation of
cycloadducts with a lateral connection of rings. Another type of
intramolecular reaction we found on homoallenyl azines.6 During
this reaction a completely new type of compound was formed,
having centrally fused four five-membered heterocyclic rings. Ap-
parently, the distance between the azine group and the multiple
bonds determines whether a ‘lateral’ or ‘central’ type cyclization is
preferred (Scheme 1).
n
n = 2
n = 1
n
n
Scheme 1.
One of our papers deals with the central type of intramolecular
criss-cross cycloadditions and the influence of substitution upon
the reaction.7 The prepared symmetrical azines 2 with a sub-
stitution in position 3 are stable compounds and were able to
undergo intramolecular criss-cross cycloaddition reactions. Intra-
molecular criss-cross cycloadditions consist of two successive
1,3-dipolar cycloadditions in the same molecule.8 By this reaction
two new stereogenic centres are formed. These compounds exist as
racemic mixtures of enantiomers with configuration (R,R) and (S,S).
Generally, we can observe that an electron donating substituent in
the 3-position increases the reactivity towards criss-cross cyclo-
addition. As soon as a bulky substitution appears, steric hindrance
leads to decreasing reactivity in the shown direction. The thermally
initiated criss-cross reaction is a source of interesting fused
heterocyclic compounds consisting of four fused five-membered
heterocycles 3 (Scheme 2).7
R
R
R
2
xylene
N2H4.H2O
CH2Cl2
CH2
C
N
N
C
N
O
C
N
reflux, 2h
R
CH2
CH2
R
1
2 60 - 97%
3 44 - 77%
* Corresponding author. Tel.: þ420 549496615; fax: þ420 549492688.
Scheme 2.
´ ˇ
E-mail address: potacek@chemi.muni.cz (M. Potacek).
0040-4020/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2008.11.057

´
H. Zachova et al. / Tetrahedron 65 (2009) 792–797
793
Table 1
When investigating their chemical properties, we found very
interesting behaviour of these structures when treated with acid in
the presence of reducing agent NaBH3CN (Scheme 3).
Reaction times and yields of compounds 5 found during the rearrangement of 3
(Scheme 4)
Educt
R1
Reagent
Product
Yield [%]
Reaction
time [h]
3a
3a
3a
3a
3b
3b
3b
3b
3b
–Ph
–Ph
–Ph
–Ph
–SiMe3
–SiMe3
–SiMe3
–SiMe3
–SiMe3
Methyl iodide
Benzyl iodide
Chlorallyl iodide
Allyl bromide
Methyl iodide
Benzyl iodide
Benzyl chloride
Chlorallyl iodide
Allyl bromide
5a
5b
5c
5d
5e
5f
5g
5h
5i
82
88
86
97
81
70
90
90
97
2
2
4
24
2
2
48
4
24
R
R
H+
N
R
NaBH3CN
R
N
MeOH
NH
NH
3
4
R
phenyl
a
b
c
86%
82%
77%
morpholinomethyl
benzyl
Scheme 3. Rearrangement of cycloadduct 3 in acidic medium and in the presence of
NaBH3CN.
also other electrophiles may act as catalysts in a similar reaction.
Therefore several alkyl halides were tested (Scheme 4).
They underwent a completely new reaction, breaking the N–N
bond in the molecule and rearranging to a caged compound 4,
having two five-membered and two six-membered rings.9
In this paper, we describe our attempts to extend our knowledge
of the behaviour of these ‘basket like’ cyclic structures 3. Some of
the molecules that were formed during the rearrangement repre-
sent a new structure with signs of the molecule corresponding to
a ‘proton sponge’.
The reaction was carried out in dry chloroform. This reaction
proceeds very well when methyl iodide or benzyl iodide is used as
reagent because of their rather high reactivity with nucleophilic
nitrogen atom. Reaction was monitored by thin layer chromatog-
raphy. After the reaction, we found rearranged products 5 as a sta-
ble quaternary iminium salts (Scheme 4).
For full identification and description of the new structure, 1D
and 2D NMR spectroscopy was used. Although NMR identification
supported our expectations concerning the structure, the elemen-
tal analyses did not fit. Problems arose with identification of the
contra anions, especially when the starting compounds were io-
dides. Finally, after X-ray analysis it was found that instead of pure
iodides a mixture of iodides and triiodides is formed. But we have
not succeeded with their separation. We have not faced such
problems in the case of chlorides and bromides. On the other hand,
the reaction times of reactions with alkyl chlorides and bromides
had to be prolonged due to their lower reactivity.
In Table 1, the tested alkyl halides are presented with the re-
action times and yields of products 5. The influence of substitution
R upon the reactivity was more pronounced in case of the reaction
with less reactive benzyl chloride. Probably conjugation of phenyl
group (R¼Ph) with the nitrogen atom in compound 3a diminish
nucleophilicity of the nitrogen atom entering as nucleophile into
reaction with benzyl chloride and causes the fact that benzyl
chloride does not react with compound 3a (R¼Ph) at all. But when
R is trimethylsilyl, compound 3b afforded with the same reagent
a high yield of the product 5b, however, within longer reaction
time.
There is rather rich list of papers related to such structures.10
Anyway, as mostly cited structure is 1,8-bis(dimethylamino)naph-
thalene and its derivatives. This compound was firstly reported by
Alder and co-workers11 with all its manifestations. Since then,
many other families of proton sponges have been created and new
are appearing.12 We believe that also our structure fit among them.
All are characterized with a high basicity and this property is
according up to date knowledge produced by a strong crowding of
unshared electron pairs on nitrogen atoms kept in close proximity
by the molecular structure. Formation of intramolecular hydrogen
bond between the nitrogen atoms –N/H/N– in protonated form
brings relief of the steric strain upon protonation. Although most of
the described structures contain an aromatic ring, our structure has
got a polycyclic aliphatic skeleton where the whole structure is
rigid. A close structure of a proton sponge is described and studied
13
´
by Estrada and Simon-Manso and another was described and
studied by Alder.14
2. Results and discussion
Our experiments were started by preparation of homoallenyl
aldehyde 1. Compounds 1 and 2 for the proposed research were
prepared accordance with the procedure in our previous paper.7
Heating of azines 2 in xylene under inert atmosphere led to criss-
cross cycloaddition and formation of compounds 3 (Scheme 2). It is
generally accepted that the reaction proceeds via 1,3-dipole formed
by the attack of one of the nitrogen atoms to Csp atom of the allenyl
skeleton present. A successive attack of the dipole on the second
allenyl group then leads to formation of criss-cross adduct
consisting of four five-membered rings. During the reaction two
stereogenic centres are formed. A proton catalyzed rearrangement
had already been observed. Now we tried to investigate whether
Mechanism of the reaction can be explained by the following
Scheme 5. Nucleophilic nitrogen atom attacks the carbon atom next
to halogen in alkyl halide.
Formation of ammonium nitrogen leads to polarization of the
bond N–N in the molecule 3, which leads to its fission under
R1
N
R1
N
7
1
N
N
R
R1- X
N
R
R
R
R
R
N
2
6
3
4
5
X
X
3
R
R
6
2
1
7
R
6
R1X
4
R
R
R
R
2
5
7
1
3
N
N
4
2
3
CHCl3
X
9
8
7
5
10
11
R
N
N
12
R1
10
N
9
8
X
NR1
11
R
N
5
X
3
N
R1
phenyl
trimethylsilyl
a
70 - 97%
12
b
5
Scheme 5. Proposed mechanism of the rearrangement of compound 3 to compound 5.
Scheme 4. Rearrangement of cycloadducts 3 to products 5 induced by alkyl halides.

´
H. Zachova et al. / Tetrahedron 65 (2009) 792–797
794
formation of enamine. After a flip of the molecule, the enamine
then reacts in -position as a nucleophile with the formed posi-
enables preparation of proton sponges with different substitution
on both nitrogen atoms.
b
tively charged carbon atom in the second part of the molecule,
which gets to a close distance and a new bond C–C is created. This
way a new molecule 5 is formed.
3. Conclusions
When the reaction was carried out with compound 3a with an
excess of benzyl iodide and with reducing agent NaBH3CN (again in
chloroform), compound 6a was identified as a product. In this case,
probably the primary formed iminium salt is reduced, then alkyl-
ated and product as ammonium salt isolated. In case of compound
6b (R1¼allyl), the procedure of its preparation had to be carried out
stepwise. The monoalkylated intermediate was immediately re-
duced under acid catalysis with NaBH3CN and after treatment of
NaOH and isolation the crude product was alkylated in the excess of
allyl bromide and afforded compound 6b (Scheme 6).
In continuation of chemical properties’ investigation of the
criss-cross cycloaddition products 3 we found that products con-
sisting of four five-membered rings undergo rearrangement not
only in the presence of proton and NaBH3CN to cyclic secondary
amines 4 with skeleton formed by two six and two five-membered
rings. But similar transformation of compounds 3 is initiated by
various reactive alkyl halides. The reaction proceeds smoothly at
room temperature, leads to a new structure 5 and proceeds in high
yield and purity. The products are N-substituted iminium salts 5. In
an excess of alkyl halide and a presence of reducing agent
NaBH3CN, new type of tertiary amines 6 with a rigid skeleton,
which is supposed to work as proton sponge, were created.
4. Experimental section
4.1. General remarks
1. R1X / CHCl3
N
N
2. HCl, NaBH3CN
X
N
N
R1
H
R1 - X
R1
Melting points were measured on a Boetius Rapido PHMK 73/
2106 (Wa¨getechnik) instrument with TM-1300K thermometer. TLC
was carried out on Silica gel 60 F254 (Merck), detection was made by
Fluotest Universal (Quazlampen, Hanau) or in I2 vapors. NMR
spectra were recorded on Bruker Avance DRX 300 in chloroform.
CHCl3 was used as an internal standard for 1H NMR spectra
(7.27 ppm) and CDCl3 was used as an internal standard for 13C NMR
spectra (triplet 77.23 ppm). The measured 13C and 1H NMR spectra
were correlated with those obtained by simulation (Advanced
Chemistry Development, Inc., Toronto, Canada). FTIR spectra were
measured at GENESIS ATI (Unicam) in KBr pallets (wave numbers in
cmꢀ1), mass spectra (EI, 30 eV) were determined on FISONS TRIO
1000. Elemental analyses were performed with a FlashEA 1112, MS
at Thermo Finnigan and HRMS at Waters-Micromas, Q-Tof Micro
(ESSI positive). The yields of the products are calculated to the
described compounds, but they can be only informative in respect
to the changeable composition of the anion in some cases. For
basicity of compound 6a estimation, commercially available
(Aldrich) ‘proton sponge’ was used.
benzyl iodide 6a 67%
allyl bromide 6b 65%
3a
6
Scheme 6. Product of compound 3a rearrangement in the presence of benzyl iodide
excess and NaBH3CN.
This way we have succeeded to prepare compounds with
a structure characteristic for proton sponges. The first indications of
such behaviour we have already observed at compounds 4 with
similar skeleton after rearrangement by acid. When we tried to
transfer the prepared ammonium salts to a free base we were faced
problems with not easy procedure. We succeeded only with con-
centrated NaOH (c¼1 mol/L) under heating. The structure of 6 is
rigid and contains tertiary amines in a suitable neighbourhood
(distance N–N over space according with X-ray analysis is 2.607 Å)
and the bound proton is situated between both nucleophilic
nitrogen atoms. We tried to estimate the basicity of compound 6 by
NMR experiment. In the case of D2O addition, full exchange of
the ammonium proton in compound 6a was observed and after
addition of a strong base tetramethylquanidine (pKBHþ¼23.3) the
protonation of our compound 6a completely disappeared. The
transprotonation experiment on 6a and the known proton
sponge 1,8-bis(dimethylamino)naphthalene (pKBHþ¼18.2–18.7 in
CH3CN)15,16 was carried out by 1H NMR spectroscopy. The NMR
experiments using different ratio mixtures of 6a and proton sponge
showed that our compound 6a pKBHþ ranges between 20 and 21.
The described method in this paper, when compared with
rearrangement of compound 3 carried out under acid catalysis,
4.2. X-ray crystallographic study
A perspective view of the molecular structures of 5b and 6a is
shown in Figures 1 and 2, respectively. Organic cations and triio-
dide anions in both structures are situated on coplanar planes.
The asymmetric unit of 5b consists of one organic cation (con-
taining two six-membered rings and two five-membered rings in
the central molecular skeleton and four methyl, two phenyl and
one benzyl exoskeleton groups) and one (triiodide) anion each with
Figure 2. Molecular structure of 6a. The thermal ellipsoids were drawn at the 50%
Figure 1. Molecular structure of 5b. The thermal ellipsoids were drawn at the 50%
probability level and hydrogen atoms and phenyl groups were omitted for clarity.
probability level and hydrogen atoms and phenyl groups were omitted for clarity.

´
H. Zachova et al. / Tetrahedron 65 (2009) 792–797
795
Table 2
Table 3
Selected bond lengths and angles of 5b and 6a
Crystal data and structure refinement
Bond length [Å]
N(1)–N(1)#1
5b
6a
Ref. 1
Structure
5b
6a
Empirical formula
Formula weight
Temperature (K)
Wavelength (Å)
Crystal system
Space group
C33H35I3N2
840.33
120
0.71069
Triclinic
P-1
C40H44I3N2
933.47
120
0.71069
Monoclinic
C2/c
2.702
1.287
1.475
2.885
1.515
1.515
2.607
1.508
1.518
2.919
1.557
1.520
1.497
1.430
1.502
d
N(1)–C(2)
N(1)–C(3)
I(1)–I(2)
*
*
,#2
*
C(1)–C(2)
C(3)–C(4)
*
*
1.329
1.563
Unit cell
dimensions
a¼10.792(2) Å
b¼10.848(2) Å
c¼13.548(3) Å
a¼16.915(3) Å
b¼18.231(4) Å
c¼11.471(2) Å
Angle [ꢁ]
5b
6a
Ref. 1
C(2)–N(1)–C(3)
C(1)–C(2)–N(1)
C(4)–C(3)–N(1)
*
*
*
106.79
115.41
103.89
98.04
99.45
175.67
106.44
107.19
104.33
97.87
100.52
178.17
103.76
113.71
102.83
108.53
100.59
d
a
b
¼83.28(3)ꢁ
a
b
¼90ꢁ
¼82.34(3)ꢁ
¼96.03(3)ꢁ
g
¼74.60(3)ꢁ
g
¼90ꢁ
C(2)–C(1)–C(4)
C(3)–C(4)–C(1)
*
Volume (Å3)
Z
1509.9(5)
2
3517.7(12)
*
I(2)–I(1)–I(2)#2
4
Calculated
density
1.848
1.763
*Average symmetry: #1 ꢀxþ1, y, ꢀzþ1/2; #2ꢀx, y, ꢀzþ1/2.
(Mg mꢀ3
)
Absorption
3.130
2.697
coefficient
(mmꢀ1
F(000)
Crystal size
(mm)
)
a point symmetry Cs. The point-group symmetry of molecule 6a
consists of one organic cation (containing two six-membered rings
and two five-membered rings in the central molecular skeleton and
four methyl, two phenyl and two benzyl exoskeleton groups) and
one (triiodide) anion is C2.
Related structures for bond lengths and angle data compari-
son are presented in papers.6,7 Tables 2 and 3 then show their
comparison with 5b and 6a. As concerns of the five-membered
rings’ geometry in 5b and 6a they have shown only small vari-
ation in the bond lengths and angles, which can be easily
explained by changes in exoskeleton groups. However, the
comparison of 5b and 6a with the two referred structures can be
approximate only, because the bonding system in the central
cage is a quite different.
The highest difference can be found in the N–N distance, which
is in our present structures 2.607 Å and 2.702 Å (non-bonding
distance), but in referred structures 1.477 Å and 1.497 Å as a typical
single bond. This difference is slightly shown also around nitrogen
atoms (N–C bonds and all angles around). The rest of the central
molecular cage is quite similar and small difference can be
explained by measurement errors. The geometry of triiodide anion
is in good accordance with related structures in papers17,18 and
ranges in length near 2.911 Å and in angle near 177.7ꢁ. The crystal
structures of 5b and 6a have not shown any H-bridges or other
shorter interactions.
812
1828
0.10ꢂ0.07ꢂ0.05
0.30ꢂ0.07
ꢂ0.05
q
range for data
collection
3.03ꢁ–28.71ꢁ
3.30ꢁ–28.74ꢁ
Limiting indices
ꢀ14ꢃhꢃ14, ꢀ14ꢃ
kꢃ14, ꢀ14ꢃlꢃ18
ꢀ22ꢃhꢃ22,
ꢀ18ꢃk
ꢃ24, ꢀ15ꢃlꢃ13
16,429/4201
[Rint¼0.0807]
99.8
Reflections
collected/unique
Completeness
18,117/7022
[Rint¼0.1453]
99.6
to 2
q
¼25.00 (%)
Absorption correction
Max. and min.
Psi-scan
1.000 and 0.8669
Psi-scan
1.000 and
0.5680
Full-matrix
least-squares
on F2
transmission
Refinement method
Full-matrix
least-squares
on F2
Data/restrains/
parameters
7022/0/332
4201/0/206
Goodness-of-fit on F2
0.807
0.867
Final R indices [I>2
s(I)]
R1¼0.0607, wR2¼0.1120
R1¼0.0394,
wR2¼0.0687
R1¼0.1135,
wR2¼0.0854
1.559 and
ꢀ0.713
R indices (all data)
R1¼0.1725, wR2¼0.1560
2.603 and ꢀ2.572
Largest diff.
peak and hole
(e Åꢀ3
)
Diffraction data were collected on a KUMA KM-4 four-circle
4.3. General procedure for the rearrangement of criss-cross
cycloadducts 3
single crystal diffractometer (l¼0.71069 Å) equipped with CCD
camera using -scan mode and corrected for Lorentz and polari-
u
zation effects. The temperature during data collection was 120(2) K.
The intensity data were corrected for Lorentz and polarization ef-
fects. All structures were solved by direct methods and refined by
full-matrix least-squares methods using anisotropic thermal pa-
rameters for the non-hydrogen atoms. The hydrogen atoms were
inserted from geometry and refined as riding. The software pack-
ages used were Xcalibur CCD system for the data collection/re-
duction,19 and Shelxtl for the structure solution, refinement and
drawing preparation.20 In figures the thermal ellipsoids were
drawn at the 50% probability level and hydrogen atoms were
omitted for clarity. Selected bond lengths and angles are listed in
Table 2. Details of the data collection and structure refinement are
listed in Table 3.
CCDC 671034 and 671035 contain the supplementary crystal-
lographic data for this paper.
These data can be obtained free of charge at www.ccdc.cam.a-
c.uk/conts/retrieving.html [or from the Cambridge Crystallographic
Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: þ44 1223
336 033; e-mail: deposit@ccdc.cam.ac.uk].
Cycloadduct 3 was dissolved in dry chloroform under argon
atmosphere and then an alkyl halide was added. The reaction
mixture was stirred at ambient temperature till the reaction com-
pleted. Then, solvent was removed under vacuum and the solid
residue was washed with diethyl ether. Compounds were purified
by column chromatography (diethylether, methanol).
4.3.1. 11-N-Methyl-1,6-diphenyl-2,2,7,7-tetramethyl-11,12-
diazatetracyclo[4.4.0.13,10.15,8]dodeca-5(12),10-diene-11-ium (5a)y
Reaction mixture: cycloadduct 3a (100 mg, 0.27 mmol) in dry
chloroform (5 mL), methyl iodide (38.5 mg, 0.27 mmol). Yield
170 mg (82%). Yellow solid, mp 182–183 ꢁC. 1H NMR (CDCl3):
d
¼0.22 (s, 3H, CH3), 0.55 (s, 3H, CH3), 1.06 (s, 3H, CH3), 1.13 (s, 3H,
CH3), 3.61 (dd, J¼12.9, 4.5 Hz, 1H, CH2), 3.64 (d, J¼12.9 Hz, 1H, CH2),
3.88 (d, J¼13.2 Hz, 1H, CH2), 4.08 (dd, J¼13.2, 5.2 Hz, 1H, CH2), 4.17
y
The procedure indicates that the described ammonium cation may be accom-
panied by a mixture of anions in dependence on the used halogenides.

´
H. Zachova et al. / Tetrahedron 65 (2009) 792–797
796
(s, 3H, CH3), 4.47 (d, J¼5.2 Hz, 1H, CH), 4.91 (d, J¼4.5 Hz, 1H, CH),
7.4–7.8 (m, 10H, Har.) ppm. 13C NMR (CDCl3):
27.2, 36.4, 38.6, 42.4, 46.7, 49.1, 83.7, 83.8, 128.9, 129.2, 129.9, 135.6,
137.9, 179.7, 199.4 ppm. IR (KBr): nmax 704, 748, 1182, 1396, 1468,
1576, 1624, 2937, 2972, 3498 cmꢀ1. For C27H31N2þ: calcd Mþ
383.2487. HRMS: found Mþ 383.2488.
141.0, 149.9, 151.4 ppm. IR (KBr): nmax 763, 849, 1026, 1095, 1252,
1444, 1639, 2960, 3415, 3491 cmꢀ1. MS (EI, 30 eV): m/z (%) 360 (30),
345 (10),180 (50),142 (22),108 (8), 73 (100), 59 (5). For C21H39N2Siþ2 :
calcd Mþ 375.2652. HRMS: found Mþ 375.2653.
d
¼18.9, 19.5, 25.9,
4.3.6. 11-N-Benzyl-1,6-di(trimethylsilyl)-2,2,7,7-tetramethyl-11,12-
diazatetracyclo[4.4.0.13,10.15,8]dodeca-5(12),10-diene-11-ium (5f)y
Reaction mixture: cycloadduct 3b (100 mg, 0.28 mmol) in dry
chloroform (5 mL), benzyl iodide (60 mg, 0.28 mmol). Yield
160 mg (70%). Yellow solid, mp 169–171 ꢁC. 1H NMR (CDCl3):
4.3.2. 11-N-Benzyl-1,6-diphenyl-2,2,7,7-tetramethyl-11,12-
diazatetracyclo[4.4.0.13,10.15,8]dodeca-5(12),10-diene-11-ium (5b)y
Reaction mixture: cycloadduct 3a (100 mg, 0.27 mmol) in dry
chloroform (5 mL), benzyl iodide (59.2 mg, 0.27 mmol). Yield
200 mg (88%). Yellow solid, mp 112–118 ꢁC. 1H NMR (CDCl3):
d
¼0.16 (s, 9H, CH3), 0.29 (s, 9H, CH3), 1.21 (s, 3H, CH3), 1.22 (s, 3H,
CH3),1.29 (s, 3H, CH3),1.32 (s, 3H, CH3), 2.61 (dd, J¼15.5,1.6 Hz,1H,
CH2), 2.69 (dd, J¼16.8, 2.8 Hz, 1H, CH2), 2.94 (dd, J¼15.5, 10.6 Hz,
1H, CH2), 3.20 (dd, J¼16.8, 10.2 Hz, 1H, CH2), 4.22 (dd, J¼10.2,
2.8 Hz,1H, CH), 4.86 (dd, J¼10.6,1.6 Hz,1H, CH), 5.00 (d, J¼13.5 Hz,
1H, CH2), 5.59 (d, J¼13.5 Hz,1H, CH2), 7.4–7.5 (m, 3H, Har.), 7.6–7.7
d
¼0.27 (s, 3H, CH3), 0.40 (s, 3H, CH3), 1.01 (s, 3H, CH3), 1.12 (s, 3H,
CH3), 3.51 (d, J¼2.6 Hz, 2H, CH2), 3.87 (d, J¼2.6 Hz, 1H, CH), 4.10 (d,
J¼13.2 Hz, 1H, CH2), 4.42 (dd, J¼13.2, 5.1 Hz, 1H, CH2), 4.55 (d,
J¼5.1 Hz, 1H, CH), 4.90 (d, J¼15.2 Hz, 1H, CH2), 6.74 (d, J¼15.2 Hz,
1H, CH2), 7.4–7.7 (m, 15H, Har.) ppm. 13C NMR (CDCl3):
d
¼18.6, 19.3,
(m, 2H, Har.) ppm. 13C NMR (CDCl3):
d
¼0.3, 0.8, 25.5, 25.7, 26.2,
25.7, 26.7, 35.9, 39.8, 45.9, 49.0, 56.9, 79.2, 83.9, 128.8, 129.0, 129.7,
129.8, 130.1, 130.1, 130.9, 135.1, 137.8, 179.8, 200.1 ppm. IR (KBr):
nmax 705, 752, 1182, 1454, 1600, 2925, 3426 cmꢀ1. For C33H35N2þ:
calcd Mþ 459.2800. HRMS: found Mþ 459.2798.
27.3, 29.2, 30.5, 56.1, 56.2, 66.0, 76.6, 81.7,126.3,127.6,129.5,131.3,
132.5, 144.0, 148.8, 150.5 ppm. IR (KBr): nmax 760, 843, 1034, 1249,
1367, 1456, 1641, 2897, 2967, 3433 cmꢀ1. MS (EI, 30 eV): m/z (%)
360 (33), 345 (14), 180 (58), 166 (15), 108 (13), 91 (100), 73 (95),
65 (18). For C27H43N2Si2þ: calcd Mþ 451.2965. HRMS found Mþ
451.2965.
4.3.3. 11-N-(3-Chloroallyl)-1,6-diphenyl-2,2,7,7-tetramethyl-11,12-
diazatetracyclo[4.4.0.13,10.15,8]dodeca-5(12),10-diene-11-ium (5c)y
Reaction mixture: cycloadduct 3a (100 mg, 0.27 mmol) in dry
chloroform (5 mL), 3-chloroallyl iodide (54.9 mg, 0.27 mmol). Yield
190 mg (86%). Light yellow solid, mp 158–162 ꢁC. 1H NMR (CDCl3):
4.3.7. 11-N-Benzyl-1,6-di(trimethylsilyl)-2,2,7,7-tetramethyl-11,12-
diazatetracyclo[4.4.0.13,10.15,8]dodeca-5(12),10-diene-11-ium
chloride (5g)
d
¼0.23 (s, 3H, CH3), 0.49 (s, 3H, CH3), 1.06 (s, 3H, CH3), 1.12 (s, 3H,
Reaction mixture: cycloadduct 3b (100 mg, 0.28 mmol) in dry
chloroform (5 mL), benzyl chloride (35 mg, 0.28 mmol). Yield
CH3), 3.62 (d, J¼12.9 Hz, H, CH2), 3.70 (dd, J¼12.9, 4.6 Hz, 1H, CH2),
3.96 (d, J¼13.2 Hz, 1H, CH2), 4.27 (dd, J¼13.2, 5.3 Hz, 1H, CH2), 4.52
(d, J¼5.0 Hz, 1H, CH), 4.6–4.7 (m, 2H, CH2), 5.91 (dd, J¼14.9, 6.9 Hz,
1H, CH), 6.3–6.4 (m, 1H, CH), 7.18 (d, J¼13.2 Hz, 1H, CH), 7.3–7.7 (m,
128 mg (90%). White solid, mp 113–115 ꢁC. 1H NMR (CDCl3):
d
¼0.11
(s, 9H, CH3), 0.23 (s, 9H, CH3), 1.18 (s, 6H, CH3), 1.22 (s, 3H, CH3), 1.23
(s, 3H, CH3), 2.55 (dd, J¼15.5, 1.6 Hz, 1H, CH2), 2.62 (dd, J¼16.5,
2.6 Hz, 1H, CH2), 2.88 (dd, J¼15.5, 10.6 Hz, 1H, CH2), 3.15 (dd, J¼16.5,
10.2 Hz, 1H, CH2), 4.15 (dd, J¼10.2, 2.6 Hz, 1H, CH), 4.81 (dd, J¼10.6,
1.6 Hz, 1H, CH), 4.92 (d, J¼13.5 Hz, 1H, CH2), 5.64 (d, J¼13.5 Hz, 1H,
CH2), 7.3–7.4 (m, 3H, Har.), 7.6–7.7 (m, 2H, Har.) ppm. 13C NMR
10H, Har.) ppm. 13C NMR (CDCl3):
d
¼18.7, 19.3, 25.7, 26.8, 36.3, 39.0,
46.3, 48.9, 52.8, 79.7, 83.0, 122.2, 128.8, 129.1, 130.0, 135.1, 137.6,
178.8 ppm. IR (KBr): nmax 707, 750, 1182, 1450, 1490, 1606, 2919,
2970, 3027, 3420 cmꢀ1. For C29H32N2Clþ: calcd Mþ 443.2254.
HRMS: found Mþ 443.2255.
(CDCl3):
d
¼0.4, 0.9, 25.6, 25.8, 26.6, 27.5, 29.0, 30.5, 55.9, 56.0, 66.3,
76.7, 81.8, 125.1, 128.9, 129.0, 130.7, 132.9, 143.1, 149.5, 151.4 ppm. IR
(KBr): nmax 702, 756, 840, 1033, 1251, 1367, 1456, 1639, 2898, 2962,
3423 cmꢀ1. MS (EI, 30 eV): m/z (%) 451 (Mþ, 1), 360 (15), 180 (45),
166 (7), 126 (42), 108 (5), 91 (100), 83 (26), 73 (24), 65 (16). For
C27H43ClN2Si2 (487.27): calcd C 66.55, H 8.89, N 5.75, Found C 66.45,
H 8.37, N 5.50.
4.3.4. 11-N-Allyl-1,6-diphenyl-2,2,7,7-tetramethyl-11,12-
diazatetracyclo[4.4.0.13,10.15,8]dodeca-5(12),10-diene-11-ium
bromide (5d)
Reaction mixture: cycloadduct 3a (50 mg, 0.14 mmol) in dry
chloroform (3 mL), allyl bromide (16.4 mg, 0.14 mmol). Yield 64 mg
(97%). Light yellow solid, mp 210–215 ꢁC.1H NMR (CDCl3):
d¼0.23 (s,
3H, CH3), 0.51 (s, 3H, CH3), 1.07 (s, 3H, CH3), 1.12 (s, 3H, CH3), 3.59 (d,
J¼3.8 Hz, 2H, CH2), 3.97 (d, J¼13.3 Hz, 1H, CH2), 4.28 (dd, J¼13.3,
5.3 Hz, 1H, CH2), 4.49 (d, J¼5.3 Hz, 1H, CH), 4.55 (dd, J¼14.8, 9.0 Hz,
1H, CH2), 4.59 (d, J¼4.3 Hz,1H, CH), 4.60(d, J¼9.9 Hz,1H, CH), 5.78 (d,
J¼17.0 Hz,1H, CH), 5.90 (dd, J¼14.8, 5.3 Hz,1H, CH2), 7.2–7.4 (m,10H,
4.3.8. 11-N-(3-Chloroallyl)-1,6-di(trimethylsilyl)-2,2,7,7-
tetramethyl-11,12-diazatetracyclo[4.4.0.13,10.15,8]dodeca-5(12),
10-diene-11-ium (5h)y
Reaction mixture: cycloadduct 3b (100 mg, 0.28 mmol) in dry
chloroform (5 mL), 3-chloroallyl iodide (56.1 mg, 0.28 mmol). Yield
200 mg (90%). Light yellow solid, mp 172–174 ꢁC. 1H NMR (CDCl3):
Har.) ppm. 13C NMR (CDCl3):
d¼18.7, 19.3, 25.8, 26.9, 36.0, 38.7, 46.1,
48.9, 55.9, 79.7, 83.1, 125.5, 127.8, 128.8, 129.7, 135.4, 137.8,
179.2 ppm. IR (KBr): nmax 705, 750,1180,1450,1496,1604,1639, 2917,
2969, 3421 cmꢀ1. MS (EI, 30 eV): m/z (%) 409 (Mþ, 8), 226 (77), 225
(100), 210 (74), 184 (93), 170 (25), 128 (28), 83 (38). C29H33BrN2
(484.49): calcd C 71.16, H 6.80, N 5.72. Found: C 71.23, H 6.76, N 5.70.
d
¼0.17 (s, 9H, CH3), 0.28 (s, 9H, CH3), 1.24 (s, 3H, CH3), 1.27 (s, 3H,
CH3), 1.29 (s, 3H, CH3), 1.52 (s, 3H, CH3), 2.6–2.7 (m, 2H, CH2), 3.04
(dd, J¼15.7, 10.7 Hz, 1H, CH2), 3.46 (dd, J¼16.8, 10.2 Hz, 1H, CH2),
4.34 (dd, J¼10.1, 2.5 Hz, 1H, CH), 4.41 (dd, J¼14.0, 8.1 Hz, 1H, CH),
4.81 (dd, J¼8.9, 2.0 Hz, 1H, CH), 5.29 (dd, J¼13.9, 7.6 Hz, 1H, CH),
6.2–6.3 (m, 1H, CH), 7.14 (d, J¼13.2 Hz, 1H, CH) ppm. 13C NMR
4.3.5. 11-N-Methyl-1,6-di(trimethylsilyl)-2,2,7,7-tetramethyl-11,12-
diazatetracyclo[4.4.0.13,10.15,8]dodeca-5(12),10-diene-11-ium (5e)y
Reaction mixture: cycloadduct 3b (100 mg, 0.28 mmol) in dry
chloroform (5 mL), methyl iodide (39.4 mg, 0.28 mmol). Yield
(CDCl3):
d¼0.2, 1.0, 25.8, 26.0, 26.3, 27.8, 30.0, 30.6, 56.2, 56.3, 61.4,
76.4, 81.7, 114.6, 116.5, 121.3, 131.8, 149.3, 151.0 ppm. IR (KBr): nmax
761, 841, 1253, 1640, 2159, 2952, 3461 cmꢀ1. For C23H40N2Si2Clþ
calcd Mþ 435.2419. HRMS found Mþ 435.2420.
170 mg (81%). Yellow solid, mp 198–200 ꢁC.1H NMR (CDCl3):
d
¼0.04
(s, 9H, CH3), 0.15 (s, 9H, CH3),1.12 (s, 3H, CH3),1.15 (s, 3H, CH3),1.18 (s,
3H, CH3),1.43 (s, 3H, CH3), 2.5–2.6 (m, 2H, CH2), 3.1–3.2 (m, 2H, CH2),
3.71 (s, 3H, CH3), 4.21 (dd, J¼10.4, 2.6 Hz, 1H, CH), 5.12 (dd, J¼10.4,
4.3.9. 11-N-Allyl-1,6-di(trimethylsilyl)-2,2,7,7-tetramethyl-11,12-
diazatetracyclo[4.4.0.13,10.15,8]dodeca-5(12),10-diene-11-ium
bromide (5i)
Reaction mixture: cycloadduct 2b (100 mg, 0.28 mmol) in dry
chloroform (5 mL), allyl bromide (33.5 mg, 0.28 mmol). Yield
2.0 Hz, 1H, CH) ppm (signals overlapped). 13C NMR (CDCl3):
d¼0.1,
0.5, 24.9, 25.2, 25.6, 27.2, 29.6, 29.9, 51.0, 55.5, 55.8, 75.7, 84.6,124.9,

´
H. Zachova et al. / Tetrahedron 65 (2009) 792–797
797
170 mg (97%). Light yellow solid, mp 201–202 ꢁC. 1H NMR (CDCl3):
vacuum removed and the solid residue dissolved in diethyl ether
and washed with 1 M NaOH. Organic layer was concentrated in
vacuo and solid phase was dissolved in chloroform and stirred with
another 2 equiv of allyl bromide. After overnight stirring, the sol-
vent was evaporated and product was purified by column chro-
matography (CH2Cl2 with 1% of MeOH).
d
¼0.01 (s, 9H, CH3), 0.02 (s, 9H, CH3), 0.99 (s, 3H, CH3), 1.02 (s, 3H,
CH3), 1.05 (s, 3H, CH3), 1.27 (s, 3H, CH3), 3.59 (d, J¼14.2 Hz, 2H, CH2),
2.75 (dd, J¼15.8, 10.9 Hz, 1H, CH2), 3.29 (dd, J¼16.8, 10.2 Hz, 1H,
CH2), 3.90 (dd, J¼13.7, 8.1 Hz, 1H, CH), 4.09 (dd, J¼10.2, 2.6 Hz, 1H,
CH), 4.66 (d, J¼10.2 Hz, 1H, CH2), 4.91 (dd, J¼13.7, 6.1 Hz, 1H, CH),
5.43 (d, J¼10.2 Hz, 1H, CH2), 5.55 (d, J¼16.8 Hz, 1H, CH) ppm. 13C
Yield 188 mg (65%). Light yellow solid, mp 254–256 ꢁC. 1H NMR
NMR (CDCl3):
d¼0.1, 0.3, 25.1, 25.3, 25.4, 27.0, 29.2, 29.8, 55.4, 55.5,
(CDCl3):
d
¼0.51 (s, 6H, CH3), 1.09 (s, 6H, CH3), 2.58 (dd, J¼15.5,
63.6, 75.8, 80.7, 124.9, 125.4, 127.3, 141.4, 149.5, 150.9 ppm. IR (KBr):
nmax 759, 841, 1009, 1252, 1639, 2951, 3382, 3463 cmꢀ1. MS (EI,
30 eV): m/z (%) 401 (Mþ, 1), 360 (32), 345 (12), 180 (50), 166 (10),
108 (12), 73 (100), 41 (28). For C23H41BrN2Si2 (481.66): calcd C
57.35, H 8.58, N 5.82. Found C 57.30, H 8.56, N 5.63.
3.6 Hz, 2H, CH2), 2.79 (dd, J¼15.5, 2.6 Hz, 2H, CH2), 3.45 (br s, 2H,
CH), 3.84 (dd, J¼13.0, 7.6 Hz, 2H, CH2), 4.11 (dd, J¼13.0, 6.1 Hz, 2H,
CH2), 4.63 (br s, 2H, CH), 5.3–5.4 (m, 4H, CH2), 6.1–6.2 (m, 2H, CH),
6.8–7.4 (m, 10H, Har.), 13.26 (br s, 1H, NH) ppm. 13C NMR (CDCl3):
d
¼22.9, 27.6, 29.0, 49.8, 58.6, 58.7, 67.7, 69.7, 122.2, 126.0, 127.7,
128.4, 129.1, 132.4, 139.2 ppm. IR (KBr): nmax 714, 750, 1018, 1087,
1444, 1639, 2877, 2926, 2979, 3440 cmꢀ1. For C32H41N2þ: calcd Mþ
453.3270. HRMSþ found Mþ 453.3271.
4.3.10. 11,12-N,N0-Dibenzyl-1,6-diphenyl-2,2,7,7-tetramethyl-11,12-
diazatetracyclo [4.4.0.13,10.15,8] dodecane-11-ium (6a)y
Compound 3a (0.15 g, 0.41 mmol) was dissolved in dry chloro-
form (10 mL) under argon and then benzyl iodide (0.18 g,
0.81 mmol) was added. The reaction mixture was stirred at ambi-
ent temperature till the reaction completed. Then, to the reaction
mixture NaBH3CN (0.10 g, 1.63 mmol) and finally few drops of
hydrochloric acid were added. After 2 h of stirring, the solvent was
removed under vacuum and the solid residue was washed with
water and extracted with dichloromethane. The solvent was
evaporated and product was washed with diethyl ether.
Acknowledgements
The authors would like to acknowledge Professor Milan Kra-
´
tochvıl for fruitful discussions and bright ideas.
References and notes
Yield 253 mg (67%). White solid, mp 278–281 ꢁC. 1H NMR
ˇ
1. Man, S.; Buchlovic, M.; Potacek, M. Tetrahedron Lett. 2006, 47, 6961–6963.
´ ˇ
(CDCl3):
d
¼0.60 (s, 6H, CH3), 1.08 (s, 6H, CH3), 2.29 (dd, J¼15.7,
ˇ
´ ˇ
2. Buchlovic, M.; Man, S.; Potacek, M. Tetrahedron 2008, 64, 9953–9961.
3. Grashey, R. Azomethine Imines. In 1,3-Dipolar Cycloaddition Chemistry; Padwa,
A., Ed.; Willey: New York, NY, 1984; pp 757–763.
3.5 Hz, 2H, CH2), 2.46 (dd, J¼15.7, 3.8 Hz, 2H, CH2), 3.47 (d, J¼3.6 Hz,
2H, CH2), 4.37 (dd, J¼12.7, 3.1 Hz, 2H, CH), 4.50 (dd, J¼12.7, 1.8 Hz,
2H, CH), 4.72 (br s, 2H, CH2), 6.87 (d, J¼7.8 Hz, 2H, Har), 7.13 (t,
J¼7.8 Hz, 2H, Har.), 7.3–7.5 (m, 16H, Har.), 12.53 (br s, 1H, NH) ppm.
´
4. Radl, S. Aldrichimica Acta 1997, 30, 97–100.
5. Marthur, S.; Suschitzky, H. J. Chem. Soc., Perkin Trans. 1 1975, 2479–2483.
ˇ
´ ˇ
´
ˇ
6. Potacek, M.; Marek, R.; Zak, Z.; Trottier, J.; Janousek, Z.; Viehe, H. G. Tetrahedron
13C NMR (CDCl3):
d¼22.9, 28.0, 29.5, 50.0, 58.6, 58.2, 60.4, 70.0,
Lett. 1993, 34, 8341–8344.
7. Zachova´, H.; Man, S.; Necas, M.; Potacek, M. Eur. J. Org. Chem. 2005, 2548–2557.
8. Huisgen, R. Angew. Chem., Int. Ed. Engl. 1963, 2, 365–598.
9. Zachova´, H.; Marek, R.; Man, S.; Taraba, J.; Pota´cek, M. Tetrahedron Lett. 2006, 47,
ˇ
´ ˇ
126.1, 127.9, 128.5, 129.1, 129.3, 129.6, 130.1, 132.4, 134.1, 139.0 ppm.
IR (KBr): nmax 700, 754, 1026, 1250, 1466, 1670, 2860, 2933, 2966,
3059, 3421 cmꢀ1. For C40H45N2þ: calcd Mþ 553.3580. HRMSþ found
Mþ 553.3580.
ˇ
8157–8159.
10. Staab, H. A.; Saupe, T. Angew. Chem. 1988, 100, 895–909.
11. Alder, R. W.; Bowman, P. S.; Steele, W. R.; Winterman, D. R. J. Chem. Soc., Chem.
Commun. 1968, 723–724.
4.3.11. 11,12-N,N0-Diallyl-1,6-diphenyl-2,2,7,7-tetramethyl-11,12-
diazatetracyclo [4.4.0.13,10.15,8] dodecane-11-ium bromide (6b)
Compound 3a (0.2 g, 0.54 mmol) was dissolved in dry chloro-
form (10 mL) and under argon allyl bromide (0.13 g,1.09 mmol) was
added. The reaction mixture was stirred at ambient temperature till
the starting compound disappeared. Then to the reaction mixture
NaBH3CN (0.14 g, 2.17 mmol) and finally few drops of hydrochloric
acid were added. After stirring overnight, the solvent was under
12. Alder, R. W. Chem. Rev. 1989, 89, 1215–1223.
´
13. Estrada, E.; Simon-Manso, Y. Angew. Chem., Int. Ed. 2006, 45, 1719–1721.
14. Alder, R. W. J. Am. Chem. Soc. 2005, 127, 7924–7931.
15. Kaljurand, I.; Ku¨tt, A.; Soova¨li, L.; Rodima, T.; Ma¨emets, V.; Leito, I.; Koppel, I. A.
J. Org. Chem. 2005, 70, 1019–1028.
16. Kanbara, T.; Suzuki, Y.; Yamamoto, T. Eur. J. Org. Chem. 2006, 3314–3316.
17. Tebbe, K. F.; Dombrowski, I. Z. Anorg. Allg. Chem. 1999, 625, 167.
ˇ
18. Taraba, J.; Za´k, Z. Inorg. Chem. 2003, 42, 3591–3594.
19. Xcalibur CCD System, CrysAlis Software System, Version 1.170. Oxford, England.
20. Shelxtl, Version 5.10; Bruker AXS: Madison, WI, USA, 1997.