Synthesis of unsymmetrically substituted 4,6,10-trihydroxy-1,4,6,10-tetra- azaadamantanes via intramolecular cyclization of tris(-oximinoalkyl)amines

Article abstract of DOI:10.1055/s-0031-1289735

The cyclotrimerization of oximino groups in unsymmetrically substituted tris(-oximinoalkyl)amines was studied. A general approach to the synthesis of 4,6,10-trihydroxy-1,4,6,10-tetraazaadamantanes containing different substituents at bridgehead carbon ato

Full text of DOI:10.1055/s-0031-1289735

PAPER
1095
Synthesis of Unsymmetrically Substituted 4,6,10-Trihydroxy-1,4,6,10-tetra-
azaadamantanes via Intramolecular Cyclization of Tris(b-oximinoalkyl)amines
U
nsymmetricallyS
r
ubstitut
t
ed1,4,6
e
,10-Tetraa
m
z
aadamantane
s
N. Semakin, Alexey Yu. Sukhorukov,* Yulia V. Nelyubina, Sema L. Ioffe, Vladimir A. Tartakovsky
N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, Leninsky prosp. 47, 119991 Moscow, Russian Federation
Fax +7(499)1355328; E-mail: sukhorukov@server.ioc.ac.ru
Received 30 December 2011; revised 8 February 2012
access to a large library of 1,4,6,10-tetraazaadamantanes
Abstract: The cyclotrimerization of oximino groups in unsymmet-
so that their properties and, in particular, their biological
rically substituted tris(b-oximinoalkyl)amines was studied. A gen-
activity, can be explored.3 Recently, unsymmetrically
eral approach to the synthesis of 4,6,10-trihydroxy-1,4,6,10-
substituted tris-oximes 1 became available using a strate-
tetraazaadamantanes containing different substituents at bridgehead
carbon atoms was developed from available aliphatic nitro com- gy that involved the silylation of aliphatic nitro com-
pounds (Scheme 2).4 According to this approach, tris-
pounds.
oximes 1 containing two or three different oximinoalkyl
groups are assembled in several steps from two or three
different N,N-bis(siloxy)enamines 3. The latter enamines
are obtained by double silylation of the corresponding ni-
tro compounds.5
Key words: cyclization, heterocycles, amines, oximes, adaman-
tanes
1,4,6,10-Tetraazaadamantane is a structural isomer of the
well-known 1,3,5,7-tetraazaadamantane (urotropin),
which is widely applied in various fields of science and in-
dustry.1 Although 1,4,6,10-tetraazaadamantane itself has
not yet been obtained, the synthesis of its derivatives 2 by
an unusual intramolecular cyclization of tris(b-oximino-
alkyl)amines 1 was recently reported (Scheme 1).2
In the present work, the cyclotrimerization of oximino
groups in unsymmetrically substituted tris-oximes 1 to
give 1,4,6,10-tetraazaadamantanes 2 was studied.
R1
R1
Me
NO2
2 Me3SiX, Et3N
N(OSiMe3)2
3
N
OH
N
AcOH
R2
R
R
MeOH–H2O
R2
Me
R
N
2 Me3SiX, Et3N
R
OH
HO
N
N
N(OSiMe3)2
N
N
HO
OH
NO2
R
R
N
3
R4
1
2
OH
R3
R3
R = H, Me, Et, Bn, CH2CH2CO2Me
R = Ph, CO2Et - no cyclization
R4
2 Me3SiX, Et3N
X = Cl, Br, OTf
N(OSiMe3)2
NO2
Scheme 1 Cyclization of symmetrical tris-oximes 1
3
3–4 steps
ref. 4
R4
However, according to this procedure, only symmetrical-
ly substituted 1,4,6,10-tetraazaadamantanes 2 possessing
all equal substituents R in bridgehead positions could be
obtained, because only symmetrically substituted tris-
oximes 1 were available until recently. Furthermore, the
reversible cyclization of 1 into 2 is sensitive to the nature
of the substituent R. Thus, whereas the cyclization of tris-
oximes 1 with R = H or alkyl to 1,4,6,10-tetraazaadaman-
tanes 2 proceeds smoothly, formation of the correspond-
ing adamantanes from tris-oximes 1 with R = Ph or CO2Et
was not observed.2 Clearly, for a more detailed study of
the effects of the substituent on the equilibrium between 1
and 2 it would be beneficial to explore the cyclization of
‘mixed’ tris-oximes 1 containing substituents that favor or
disfavor the transformation into tetraazaadamantanes 2.
Furthermore, the successful realization of the cyclization
of unsymmetrically substituted tris-oximes 1 will provide
R1
R3
N
N
N
R2
N
OH
HO
OH
1
Scheme 2 Synthesis of unsymmetrically substituted tris-oximes 1
A series of ‘unsymmetrical’ tris-oximes 1a–l obtained by
a known method4 were chosen as model objects to study
the cyclotrimerization reaction (Scheme 3 and Table 1).
Tris-oximes 1a and 1b partially underwent cyclization to
the corresponding tetraazaadamantanes 2a and 2b at room
temperature.4 Other tris-oximes 1c–l were stable and did
not cyclize to tetraazaadamantanes 2 spontaneously.
However, in the presence of acetic acid (see Scheme 3,
equation 1 and Table 1, procedure 1) most of the tris-
oximes 1 produced the corresponding tetraazaadaman-
tanes 2, which were isolated in good yields in some cases
(products 2a–e). For tris-oxime 1f, the formation of tet-
raazaadamantane 2f was detected in trace amounts by 1H
SYNTHESIS 2012, 44, 1095–1101
x
x.
x
x.
2
0
1
2
Advanced online publication: 05.03.2012
DOI: 10.1055/s-0031-1289735; Art ID: T119911SS
© Georg Thieme Verlag Stuttgart · New York

1096
A. N. Semakin et al.
PAPER
NMR analysis (Table 1, entry 6). No cyclization of tris- alents of benzyl bromide, which reacts with tetraazaada-
oximes 1h–j was observed under these conditions mantanes 2 to give the corresponding quaternary salts 4
(Table 1, entries 8–10).
(see Scheme 3, equation 2 and Table 1, procedure 2). The
majority of tris-oximes 1 afforded the corresponding qua-
ternary salts of tetraazaadamantanes 4 in high yields ac-
cording to this protocol. However, under these conditions,
tris-oximes 1h–j produced mixtures of unidentified prod-
ucts in which quaternary salts of 1,4,6,10-tetraazaada-
mantanes 4 were not detected.
R4
R4
R1
R3
N
OH
N
N
i
R1
HO
R3
OH
(1)
N
R2
N
HO
OH
N
N
N
R2
OH
1
2
Thus, the transformation of tris-oximes 1 into quaternary
salts 4 proceeds more efficiently and with higher yields
than the cyclization to free tetraazaadamantane bases 2.
Interestingly, tetraazaadamantanes 2 decompose into the
corresponding tris-oximes 1 quite slowly at room temper-
ature, demonstrating their kinetic stability. Based on this
observation, we developed a two-step procedure for the
preparation of tetraazaadamantanes 2 from tris-oximes 1
that consists of the initial preparation of quaternary salts 4
by using procedure 2, followed by catalytic debenzylation
with hydrogen over Pd/C6 (procedure 3, Scheme 4). Ac-
cording to this route, product 2f, which is practically un-
available by using procedure 1, could be obtained in good
yield (cf. Scheme 4 and Table 1, entry 6).
R4
CH2Ph
R4
R3
R1
R3
N
OH
Br
R1
N
ii
(2)
N
N
R2
N
OH
HO
HO
N
N
OH
N
R2
OH
1
4
Scheme 3 Reagents and conditions: (i) AcOH (3 equiv), MeOH, r.t.
(procedure 1); (ii) AcOH (3 equiv), BnBr (1.5 equiv), MeOH, r.t.
(procedure 2).
Previously, we demonstrated that quaternization of the
bridgehead nitrogen atom in tetraazaadamantanes 2 re-
sults in a shift in the equilibrium between 1 and 2 towards
the adamantane structure.2 Therefore, the cyclization of
tris-oximes 1a–l was studied in the presence of 1.5 equiv-
As seen from the data presented in Table 1, the cyclization
of tris-oximes 1 to adamantanes 2 is sensitive to steric ef-
fects of substituents R1–R3 in the oximinoalkyl fragments.
Table 1 Scope of the Cyclization
Entry
1
R1
R2
R3
R4
Procedure 1
Procedure 2
Time (%)a Yield 2 (%)b Time (h) Yield 4 (%)b
1
2
1a
1b
1c
1d
1e
1f
Me
Me
H
H
Me
H
H
H
H
H
H
H
H
H
H
10
12
89
3
4
75
92
86
87
90
83
88
Me
Me
H
85
3
Me
Me
(CH2)2CO2Me
24
78
24
20
25
30
48
96
96
96
72
48
4
Me
Me
Bn
48
68c
60d
<5e
31f
n.r.
n.r.
n.r.
24h
32i
5
Me
Me
CO2Et
Ph
72
6
Me
Me
240
96
7
1g
1h
1i
Bn
Bn
Me
Me
Me
Bn
g
8
CO2Et
CO2Et
240
240
240
96
–
g
9
Ph
Ph
–
g
10
11
12
1j
Ph
Ph
–
1k
(CH2)2CO2Me
Me
(CH2)2CO2Me
(CH2)2CO2Me
Bn
75
72
1l
Bn
96
a The time corresponds to either the point of full consumption of tris-oxime 1 or the time after which its residual concentration remained constant
(TLC analysis).
b Isolated yield.
c Conversion of 1d: 78%.
d Conversion of 1e: 75%, see ref. 2.
e Determined by 1H NMR analysis.
f Conversion of 1g: 55%.
g A complex mixture of unidentified products.
h Conversion of 1k: 32%.
i Conversion of 1l: 40%.
Synthesis 2012, 44, 1095–1101
© Thieme Stuttgart · New York

PAPER
Unsymmetrically Substituted 1,4,6,10-Tetraazaadamantanes
1097
Bn
N
OH
basicity of the nitrogen atom in the oxime group. Further-
more, carboxyethyl groups destabilize cations B and C
(shifting the equilibrium to the starting oxime), as well as
complicate the cyclization reactions of intermediates A
and B by reducing the nucleophilicity of nitrogen atom of
oxime group (Scheme 5).
N
OH
Br
R1
i
R1
HO
R3
OH
R3
OH
N
N
HO
N
N
N
N
R2
R2
4d
4f
4l
2d, 1 h, 80%
2f, 45 min, 78%
2l, 50 min, 82%
N
OH
N
R
R
Scheme 4 Reagents and conditions: (i) H2 (1 atm), Pd/C, K2CO3,
H
N
R
MeOH, r.t. (procedure 3).
R
N
N
HO
HO
OH
N
N
OH
R
N
R
A
H
1
The smaller these substituents are, the more efficiently the
intramolecular cyclotrimerization proceeds. Thus, tris-
oximes 1a and 1b, which possess one sterically unhin-
dered aldoxime group, enter the cyclization reaction most
easily. Tris-oximes 1c–e, which contain two methyl
groups and a more bulky substituent R3 (CH2CH2CO2Me,
Bn, or CO2Et), produced the corresponding tetraazaada-
mantanes 2c–e in high yields in the presence of acetic
acid. In the series of tris-oximes 1a–f, containing different
substituents R3 (H, CH2CH2CO2Me, Bn, CO2Et, or Ph)
and equal substituents R1 and R2 (methyl group), substrate
1f with the bulkiest R3 group (phenyl) produced the tet-
raazaadamantane in the lowest yield (cf. Table 1, entries
1–6).7 The presence of two bulky substituents R1 and R2
(for example, benzyl groups) in tris-oximes 1 led to a sig-
nificant decrease in the yield of tetraazaadamantanes 2 un-
der the conditions of procedure 1 (cf. Table 1, entries 4
and 7). Tris-oximes 1i and 1j, bearing two phenyl groups,
did not cyclize to the tetraazaadamantanes at all (Table 1,
entries 9 and 10). We suppose that such an effect of sub-
stituent size in the oximino group on the cyclotrimeriza-
tion reaction is associated with the thermodynamic
stability of tetraazaadamantanes 2. The more sterically
hindered tetraazaadamantane 2 is, the less thermodynam-
ically preferable it is compared to the corresponding tris-
oxime 1 (see also Semakin et al.2).
OH
N
OH
N
N
OH
N
R
R
R
R
HO
HO
N
N
N
N
OH
OH
R
R
C
H
H
B
N
OH
N
OH
R
R
N
R
N
R
HO
HO
N
N
N
N
OH
– H
OH
R
R
H
2
Scheme 5 Proposed mechanism of cyclization of tris-oximes 1
The structures of previously unknown tetraazaadaman-
13
tanes 2 and 4 were confirmed by 1H and C NMR spec-
troscopy, high-resolution mass spectrometry and
elemental analysis for quaternary salts 4. In the 1H and 13C
NMR spectra of tetraazaadamantanes 2 and 4, a broaden-
ing of signals, which is characteristic for this class of com-
pounds, was observed.2,4 The structure of salt 4e was
established by X-ray diffraction analysis of its hydrate ob-
tained by crystallization of 4e from a solution in aqueous
methanol (Figure 1).9
In the presence of benzyl bromide (procedure 2), the na-
ture of substituents mainly influences the time required
for full conversion of tris-oximes 1. However, when tris-
oxime 1 contained at least two phenyl substituents R [see
Table 1, entry 10 and Scheme 1 (R = Ph)], the corre-
sponding tetraazaadamantane salts 4 were not formed
even under these conditions.
Tris-oximes 1, possessing two (see Table 1, entry 8) or
three (see Scheme 1 and Semakin et al.2) CO2Et groups
did not cyclize to tetraazaadamantanes 2 using either pro-
cedure 1 or 2. This fact may be explained in terms of
Brønsted acid catalysis of the oxime cyclotrimerization
reaction between 1 and 2 (Scheme 5). Apparently, the ma-
jor role of acetic acid is the protonation of the nitrogen
atom in the oximino group, with reversible generation of
cation A.8 The latter reacts with a nucleophilic nitrogen
atom of a neighboring oximino group (Scheme 5). Two
subsequent cyclizations of intermediate cations B and C
furnish the tetraazaadamantane cage. The carboxyethyl
group, due to its electron-withdrawing character, does not
favor the generation of cation A because it decreases the
Figure 1 X-ray diffraction analysis of product 2(4e)·7H2O9
Thus, tris-oximes 1, containing two or three different ox-
iminoalkyl fragments, were involved in the cyclotrimer-
ization reaction leading to 1,4,6,10-tetraazaadamantanes
2. Only tris-oximes 1 containing two (three) ester groups
© Thieme Stuttgart · New York
Synthesis 2012, 44, 1095–1101

1098
A. N. Semakin et al.
PAPER
3,5-Dimethyl-1,4,6,10-tetraazatricyclo[3.3.1.13,7]decane-4,6,10-
triol (2a)4
or two (three) phenyl substituents connected to oximino
groups (substituents R1–R3) do not undergo cyclization to
tetraazaadamantanes 2. Using this methodology, a wide
spectra of 1,4,6,10-tetraazaadamantanes 2 and 4 contain-
ing different combinations of substituents at bridgehead
positions can be obtained starting from a small set of
available nitro compounds.
Yield: 192 mg (89%; procedure 1). All NMR data are in accordance
with literature data for 2a.4
2,5,7-Trimethyl-1,4,6,10-tetraazatricyclo[3.3.1.13,7]decane-
4,6,10-triol (2b)4
Yield: 196 mg (85%; procedure 1). All NMR data are in accordance
with literature data for 2b.4
Analytical TLC was performed on Silufol silica gel plates; visual-
ization was accomplished by using a solution of ninhydrin in EtOH.
Flash chromatography was performed on a thin pad (20–30 mm) of
silica gel 40–60 mm (Acros). MeOH was distilled over Mg. Hexane
and EtOAc were distilled without drying agents. Acetic acid was
twice recrystallized. The commercial reagents BnBr (Sigma–
Aldrich) and palladium on activated charcoal (10% Pd/C; Sigma–
Aldrich) were used as received. Tris-oximes 1a–l were prepared ac-
cording to literature procedures.4 Catalytic hydrogenation reactions
were carried out in a round-bottom flask equipped with a balloon
and a magnetic stirrer.
Methyl 3-{4,6,10-Trihydroxy-5,7-dimethyl-1,4,6,10-tetraazatri-
cyclo[3.3.1.13,7]decan-3-yl}propanoate (2c)
Yield: 236 mg (78%; procedure 1); colorless solid; mp 75–86 °C;
Rf = 0.35 (EtOAc–MeOH, 3:1).
1H NMR: d = 1.16 [s, 6 H, (7)], 1.93 [br s, 2 H, (4)], 2.38 [s, 2 H,
(3)], 2.5–3.2 [br s, 6 H, (6 and 9)], 3.57 [s, 3 H, (1)], 7.6–8.2 [2 × br
s, 3 H, (10 and 11)].
13C NMR: d = 21.1 (7), 27.6 and 28.4 (3 and 4), 51.5 (1), 49–63 (br,
6 and 9), 74.7 (8), 75.5 (5), 174.4 (2).
HRMS (+ve): m/z [M + Na]+ calcd for C12H22N4O5: 325.1482;
1H (300.13 MHz) and 13C NMR (75.47 MHz; with complete proton
decoupling) spectra were recorded at r.t. with a Bruker AM 300
NMR spectrometer; samples were dissolved in DMSO-d6. Chemi-
cal shifts are reported in ppm relative to the residual solvent peak as
internal standard.11 Elemental analyses were performed by the
Analytical Laboratory of the N.D. Zelinsky Institute of Organic
Chemistry. High-resolution mass spectra were recorded with a
MicroTOFF spectrometer. Melting points (uncorrected) were deter-
mined with a Kofler apparatus. Atom numberings in products 2 and
4 are shown, respectively, in Figure 2 and Figure 3. The numbers in
parentheses in the characterization data refer to these atom num-
bers.
found: 325.1482.
3-Benzyl-5,7-dimethyl-1,4,6,10-tetraazatricyclo[3.3.1.13,7]dec-
ane-4,6,10-triol (2d)
Adamantane 2d was obtained by procedure 1 from tris-oxime 1d
(306 mg, 1.0 mmol) and by procedure 3 from benzyl salt 4d (119
mg, 0.25 mmol).
Yield: 68% (208 mg; procedure 1); 80% (61 mg; procedure 3); col-
orless oil; Rf = 0.12 (EtOAc).
1H NMR: d = 1.19 [s, 6 H, (8)], 2.3–3.2 [br, 6 H, (7 and 10)], 3.13
[s, 2 H, (5)], 7.1–7.3 [m, 5 H, (1, 2 and 3)], 7.8–8.2 [2 × br s, 3 H,
(11 and 12)].
13C NMR: d = 21.3 (8), 38.5 (5), 51–60 (br, 7 and 10), 75.1 (9), 76.7
Adamantanes 2a–e, 2g, 2k, and 2l; General Procedure 1
AcOH (0.172 mL, 3.00 mmol) was added to a stirred solution of the
corresponding tris-oxime 1 (1.00 mmol) in MeOH (5.0 mL). The re-
action mixture was kept with periodic stirring at r.t. until either full
conversion or constant residual concentration of tris-oxime 1 was
observed (TLC analysis, time indicated in Table 1). The mixture
was evaporated in vacuo at 5–20 °C. The residue was preadsorbed
on silica gel and purified by flash chromatography (silica gel;
EtOAc–hexane, 1:1 then EtOAc–MeOH, 3:1). Yields of target ada-
mantanes 2 and conversions of tris-oximes 1 are presented in
Table 1.
(6), 126.3, 128.0 and 131.2 (1, 2 and 3), 137.2 (4).
HRMS (+ve): m/z [M + H]+ calcd for C15H23N4O3: 307.1765; found:
307.1769.
Ethyl 5,7-Dimethyl-1,4,6,10-tetraazatricyclo[3.3.1.13,7]decane-
4,6,10-triol-3-carboxylate (2e)2
Adamantane 2e was obtained by procedure 1 from tris-oxime 1e (51
mg, 0.177 mmol).2
Yield: 60% (31 mg); white foam. All NMR data are in accordance
with literature data for 2e.2
1
N
N
N
6
5
6
5
7
6
O
10
9
3
3
OH
9
2
OH
2
OH
4
Me
Me
HO
5
Me
HO
2
4
3
N
N
N
N
4
1
O
10HO
1
Me
N
N
N
N
N
OH 12
8
9
8
10
11
OH 11
OH
7Me
Me
8 Me
11
7
2f
2c
2d
1
11Me
O
N
N
MeO
O
1
N
3
2
7
16
15
12
13
13
14
19
10
2
7
10
OH
OH
2
OH
6
4
1
Bn
11HO
6
3
4
O
Me
OH 12
N
N
3
5
N
N
HO
5
N
N
N
N
17 HO
9
12
14
OH
15
N
9
OH 18
8
11
O
Me
10
8
7
8
9
6
Me
O
4
5
2g
2k
2l
Figure 2 Atom numbering in products 2
Synthesis 2012, 44, 1095–1101
© Thieme Stuttgart · New York

PAPER
Unsymmetrically Substituted 1,4,6,10-Tetraazaadamantanes
1099
9
10
13
8
10
11
11
12
14
15
12
3
9
12
5
13
6
16
9
Me 3
Br
Br
6
N
N
Br
6'
N
N
2
2
O
OH
N
4'
OH
OH
1
5'
5
1
Me
HO
Me
HO
H
Me
HO
N
2
H
N
N
Me 1
4
O
N
OH
N
N
4
N
6
10
OH 8
8
7
5
7
OH
Me
Me
Me
11
7
3
4
4b
4c
4a
13
12
14
12
14
11
13
14
13
15
16
15
16
9
15
8
17
Br
Br
N
Br
7
1
N
N
6
5
3
OH
2
10
OH
2
5
OH
O
4
4
Me
6
Me
HO
5
N
Me
HO
N
N
4
3
N
N
HO
1
3 O
N
N
8
10
N
6
2
9
7
N
OH
OH
11
9
OH
10
Me
Me 1
7
11
12
Me
8 Me
4f
4e
4d
14
13
17
15
16
21
18
19
16
20
22
23
17
20
24
Br
N
11
Br
1
2
Br
3
2
1
MeO
O
N
MeO
O
N
10
16
OH
7
12
13
14
7
13
19
10
Me 1
12
OH
2
Bn
HO
OH
6
N
15
6
4
3
4
N
N
3
N
N
OH
N
N
HO
5
11
9
5
HO
N
14
12
N
8
OH
9
8
OH
17
15
11
O
Me
18
7
10
6
9
8
4
Me
5
O
4g
4l
4k
Figure 3 Atom numbering in products 4
3,5-Dimethyl-7-phenyl-1,4,6,10-tetraazatricyclo[3.3.1.13,7]dec-
ane-4,6,10-triol (2f)
Yield: 57 mg (78%; procedure 3); white solid; mp 102–115 °C;
Rf = 0.68 (EtOAc–MeOH, 3:1).
Methyl 3-{5-Benzyl-4,6,10-trihydroxy-7-(3-methoxy-3-oxopro-
pyl)-1,4,6,10-tetraazatricyclo[3.3.1.13,7]decan-3-yl}propanoate
(2k)
Adamantane 2k was obtained by procedure 1 from tris-oxime 1k
(290 mg, 0.64 mmol).
1H NMR: d = 1.28 [s, 6 H, (7)], 2.69 and 2.83 [2 × br s, 6 H, (6 and
9)], 7.1–7.4 [m, 5 H, (1, 2 and 3)], 7.98 and 8.04 [2 × s, 3 H, (10 and
11)].
13C NMR: d = 21.1 (7), 54–56 (br, 6 and 9), 75.0 and 79.0 (5 and 8),
125.7, 126.5 and 127.5 (1, 2 and 3), 140.5 (4).
Yield: 24% (70 mg); white foam; Rf = 0.35 (EtOAc–MeOH, 3:1).
1H NMR: d = 1.7–2.1 [br s, 4 H, (11)], 2.2–2.6 [br s, 4 H, (10)], 2.9–
3.8 [br m, 8 H, (5, 7 and 13)], 3.57 [s, 6 H, (8)], 7.1–7.4 [m, 5 H, (1,
2 and 3)], 7.9–8.2 [br s, 3 H, (14 and 15)].
HRMS (+ve): m/z [M + H]+ calcd for C14H21N4O3: 293.1608; found:
293.1601.
13C NMR: d = 27.5 and 28.8 (10 and 11), 38.3 (5), 49–55 (br, 7 and
13), 51.6 (8), 74.0 (12), 76.7 (6), 126.4, 127.9 and 131.2 (1, 2 and
3), 137.0 (4), 174.4 (9).
HRMS (+ve): m/z [M + H]+ calcd for C21H31N4O7: 451.2187; found:
451.2182.
3,5-Dibenzyl-7-methyl-1,4,6,10-tetraazatricyclo[3.3.1.13,7]dec-
ane-4,6,10-triol (2g)
Yield: 118 mg (31%; procedure 1); colorless oil; Rf = 0.40 (EtOAc).
1H NMR: d = 1.23 [s, 3 H, (1)], 2.5–3.9 [br m, 10 H, (3, 8 and 10)],
7.0–7.4 [m, 10 H, 4, 5 and 6], 7.8–8.0 [br s, 1 H, (11)], 8.1–8.4 [br
s, 2 H, (12)].
13C NMR: d = 21.6 (1), 38.3 (8), 50–58 (br, 3 and 10), 75.7 (2), 76.9
(9), 126.3, 127.8 and 131.2 (4, 5 and 6), 137.0 (7).
Methyl 3-{5-Benzyl-4,6,10-trihydroxy-7-methyl-1,4,6,10-tetra-
azatricyclo[3.3.1.13,7]decan-3-yl}propanoate (2l)
Adamantane 2l was obtained by procedure 1 from tris-oxime 1l
(150 mg, 0.39 mmol) and by procedure 3 from benzyl salt 4l (38 mg,
0.069 mmol).
Yield: 32% (47 mg; Procedure 1); 82% (22 mg, Procedure 3); white
amorphous solid; Rf = 0.32 (EtOAc).
HRMS (+ve): m/z [M + H]+ calcd for C21H27N4O3: 383.2078; found:
383.2075.
© Thieme Stuttgart · New York
Synthesis 2012, 44, 1095–1101

1100
A. N. Semakin et al.
PAPER
1H NMR: d = 1.19 [s, 3 H, (8)], 2.2–2.6 [br s, 4 H, (13 and 14)], 2.7–
3.6 [3 × br, 8 H, (5, 7, 10 and 16)], 3.58 [s, 3 H, (11)], 7.0–7.4 [m, 5
H, (1, 2 and 3)], 7.7–8.2 [2 × br, 3 H, (17, 18 and 19)].
13C NMR: d = 20.9 (8), 27.3 and 28.5 (13 and 14), 37.9 (5), 51.2
(11), 50–60 (br, 7, 10 and 16), 74.8, 75.5 and 76.2 (6, 9 and 15),
125.9, 127.5 and 130.8 (1, 2 and 3), 136.7 (4), 174.0 (12).
Anal. Calcd for C19H29BrN4O5: C, 48.21; H, 6.18; N, 11.84. Found:
C, 47.80; H, 6.29; N, 11.67.
1,3-Dibenzyl-4,6,10-trihydroxy-5,7-dimethyl- 1,4,6,10-tetraaza-
tricyclo[3.3.1.13,7]decan-1-ium Bromide (4d)
Yield: 415 mg (87%); white solid; mp 218–225 °C (dec.).
1H NMR: d = 1.26 [s, 6 H, (8)], 3.15, 3.29 and 3.38 [3 × s br, 8 H,
(5, 7 and 10)], 4.58 [s, 2 H, (17)], 7.15–7.35 and 7.40–7.55 [2 × m,
10 H, (1, 2, 3, 13, 14, 15)], 8.5–9.2 [br s, 3 H, (11 and 12)].
HRMS (+ve): m/z [M + H]+ calcd for C18H27N4O5: 379.1976; found:
379.1974.
Quaternary Salts of 1,4,6,10-Tetraazaadamantanes 4a–g, 4k,
4l; General Procedure 2
13C NMR: d = 21.3 (8), 38.5 (5), 50-57 (br, 7 and 10), 67.9 (17), 76.5
(9), 77.7 (6), 126.3 (16), 127.1, 128.4, 129.3, 130.9, 131.3 and 133.5
(1, 2, 3, 13, 14 and 15), 135.1 (4).
AcOH (0.172 mL, 3.00 mmol) and benzyl bromide (257 mg, 1.50
mmol) were consecutively added to a stirred solution of the corre-
sponding tris-oxime 1 (1.00 mmol) in MeOH (5.0 mL). The reaction
mixture was kept at r.t. with periodic stirring until full consumption
of starting material was observed (TLC analysis, see Table 1). The
reaction mixture was evaporated and the residual solid was washed
with EtOAc and dried in vacuo to give pure quaternary salts 4 (for
analytical purposes, samples were recrystallized from Et2O–MeOH
or Et2O–acetone). Yields of products 4 are presented in Table 1.
+
HRMS (+ve): m/z calcd for C22H29N4O3 : 397.2234; found:
397.2235.
Anal. Calcd for C22H29BrN4O3: C, 55.35; H, 6.12; N, 11.74. Found:
C, 55.14; H, 6.30; N, 11.64.
1-Benzyl-3-(ethoxycarbonyl)-4,6,10-trihydroxy-5,7-dimethyl-
1,4,6,10-tetraazatricyclo[3.3.1.13,7]decan-1-ium Bromide (4e)
Yield: 413 mg (90%); white solid; mp 195–207 °C (dec.).
1H NMR: d = 1.23 [br s, 9 H, (1 and 6)], 3.3–3.4 [br m, 2 H, (5)],
3.5–3.7 [br m, 4 H, (8)], 4.15 [q, J = 7.0 Hz, 2 H, (2)], 4.76 [s, 2 H,
(15)] 7.5–7.7 [m, 5 H, (11, 12 and 13)], 8.0–8.5 [br s, 1 H, (9)], 9.1–
9.3 [br s, 2 H, (10)].
13C NMR: d = 14.3 (1), 20.2 (6), 45.9 (br, 5), 55.5 (br, 8), 59.8 (2),
68.1 (15), 76.0 (7), 83.9 (4), 126.3 (14), 129.5, 131.0 and 136.6 (11,
12 and 13), 166.8 (3).
1-Benzyl-4,6,10-trihydroxy-3,5-dimethyl-4,6,10-triaza-1-azo-
niatricyclo[3.3.1.13,7]decane bromide (4a)
Yield: 290 mg (75%); white solid; mp 236–238 °C.
1H NMR: d = 1.23 [s, 6 H, (3)], 3.43 [br s, 4 H, (5)], 3.56 [br s, 2 H,
(2)], 4.28 [s, 1 H, (1)], 4.67 [s, 2 H, (12)], 7.5–7.7 [m, 5 H, (8, 9 and
10)], 8.4–9.2 [br s, 3 H, (6 and 7)].
13C NMR: d = 20.1 (3), 49–58 (2 × br, 2 and 5), 67.3 (12), 74.4 and
74.9 (1 and 4), 126.0 (11), 129.0, 130.5 and 133.1 (8, 9 and 10).
+
HRMS (+ve): m/z calcd for C18H27N4O5 : 379.1976; found:
+
379.1981.
HRMS (+ve): m/z calcd for C15H23N4O3 : 307.1765; found:
307.1761.
Salt 4e was recrystallized from MeOH–H2O to give crystalline
2(4e)·7H2O that was suitable for X-ray diffraction analysis
(Figure 1).9
Anal. Calcd for C15H23N4O3Br: C, 46.52; H, 5.99; N, 14.47. Found:
C, 46.01; H, 6.49; N, 13.97.
1-Benzyl-4,6,10-trihydroxy-5,7-dimethyl-3-phenyl-1,4,6,10-tet-
raazatricyclo[3.3.1.13,7]decan-1-ium Bromide (4f)
Yield: 384 mg (83%); white solid; mp 234–239 °C (dec.).
1-Benzyl-4,6,10-trihydroxy-2,5,7-trimethyl-4,6,10-triaza-1-azo-
niatricyclo[3.3.1.13,7]decane bromide (4b)
Yield: 369 mg (92%); white solid; mp 167–173 °C (dec.).
1H NMR: d = 1.28 [s, 6 H, (7)], 3.62 [br s, 4 H, (9)], 3.93 [br s, 2 H,
(6)], 4.79 [s, 2 H, (16)], 7.2–7.8 [m, 10 H, (1, 2, 3, 12, 13 and 14)],
8.2–8.7 [br s, 3 H, (10 and 11)].
13C NMR: d = 21.1 (7), 47–50 (br, 6), 54–57 (br, 9), 68.5 (16), 76.2
(8), 79.6 (5), 126.5 (15), 128.0, 128.3, 129.5, 131.0 and 133.7 (1, 2,
3, 4, 12, 13 and 14).
1H NMR (COSY, HSQC): d = 1.14 and 1.23 [2 × s, 6 H, (4 and 4¢)],
1.64 [d, J = 6.4 Hz, 3 H, (3)], 3.03 and 3.52 [2 × d, J = 12.0 Hz, 2
H, (6)], 3.25 and 3.79 [2 × d, J = 12.6 Hz, 2 H, (6¢)], 4.21 [br s, 2 H,
(1 and 2)], 4.52 and 4.60 [2 × d, J = 12.6 Hz, 2 H, (13)], 7.4–7.7 [m,
5 H, (9, 10 and 11)], 8.47, 8.64 and 8.99 [3 × s, 3 H, (7 and 8)].
13C NMR (HSQC): d = 12.2 (3), 20.5 and 20.7 (4 and 4¢), 50-55 (br,
6 and 6¢), 58-61 (br, 2), 64.3 (13), 74.9 and 76.9 (5 and 5¢), 79.2 (1),
126.7 (12), 129.4, 130.9 and 133.6 (9, 10 and 11).
+
HRMS (+ve): m/z calcd for C21H27N4O3 : 383.2078; found:
383.2076.
+
HRMS (+ve): m/z calcd for C16H25N4O3 : 321.1921; found:
Anal. Calcd for C21H27BrN4O3: C, 54.43; H, 5.87; N, 12.09. Found:
C, 54.92; H, 6.03; N, 11.84.
321.1916.
Anal. Calcd for C16H25BrN4O3: C, 47.89; H, 6.28; N, 13.96. Found:
C, 47.52; H, 6.80; N, 13.55.
1,3,5-Tribenzyl-4,6,10-trihydroxy-7-methyl-1,4,6,10-tetraaza-
tricyclo[3.3.1.13,7]decan-1-ium Bromide (4g)
Yield: 487 mg (88%); white solid; mp 233–241 °C (dec.).
1-Benzyl-4,6,10-trihydroxy-3-(3-methoxy-3-oxopropyl)-5,7-
dimethyl-1,4,6,10-tetraazatricyclo[3.3.1.13,7]decan-1-ium Bro-
mide (4c)
1H NMR: d = 1.32 [s, 3 H, (1)], 2.5–4.0 [3 × br s, 10 H, (3, 8 and
10)], 4.53 [s, 2 H, (17)], 7.1–7.6 [m, 15 H, (4, 5, 6, 13, 14 and 15)],
7.9–8.6 [br s, 1 H, (11)], 8.8–9.2 [br s, 2 H, (12)].
Yield: 407 mg (86%); white solid; mp 201–206 °C (dec.).
1H NMR: d = 1.22 [s, 6 H, (7)], 1.86 [br s, 2 H, (4)], 2.53 [br s, 2 H,
(3)], 3.38 [br s, 6 H, (6 and 9)], 3.58 [s, 3 H, (1)], 4.65 [s, 2 H, (16)],
7.4–7.7 [m, 5 H, (12, 13 and 14)], 8.4–9.0 [br s, 3 H, (10 and 11)].
13C NMR: d = 21.5 (1), 38.4 (8), 50–57 (br, 3 and 10), 67.5 (17), 76–
78 [br, (2 and 9)], 126.2 (16), 127.1, 128.3, 129.2, 130.8, 131.2 and
133.3 (4, 5, 6, 13, 14 and 15), 134.8 (7).
13C NMR: d = 21.1 (7), 26.4 and 27.6 (2 × br, 3 and 4), 51.7 (1), 53–
57 (br, 6 and 9), 67.9 (16), 75.7 (br, 8), 77.3 (br, 5), 126.4 (15),
129.4, 130.9 and 133.6 (12, 13 and 14), 174.0 (2).
+
HRMS (+ve): m/z calcd for C28H33N4O3 : 473.2547; found:
473.2531.
Anal. Calcd for C28H33BrN4O3: C, 60.76; H, 6.01; N, 10.12. Found:
C, 60.83; H, 6.04; N, 9.84.
+
HRMS (+ve): m/z calcd for C19H29N4O5 : 393.2132; found:
393.2128.
Synthesis 2012, 44, 1095–1101
© Thieme Stuttgart · New York

PAPER
Unsymmetrically Substituted 1,4,6,10-Tetraazaadamantanes
1101
1,3-Dibenzyl-4,6,10-trihydroxy-5,7-bis(3-methoxy-3-oxopro-
References
pyl)-1,4,6,10-tetraazatricyclo[3.3.1.13,7]decan-1-ium Bromide
(1) For application of urotropin, see, for example: (a) Seneca,
H.; Peer, P. Med. Times 1969, 97, 243. (b) Salamone, J. S.
Polymeric Material Encyclopedia; CRC Press: New York,
1996, 5036.
(2) Semakin, A. N.; Sukhorukov, A. Yu.; Lesiv, A. V.; Ioffe, S.
L.; Lyssenko, K. A.; Nelyubina, Yu. V.; Tartakovsky, V. A.
Org. Lett. 2009, 11, 4072.
(3) For application of 1,4,6,10-tetraazaadamantanes in the
design of water-soluble bioactive molecules, see: Tolbin, A.
Yu.; Sukhorukov, A. Yu.; Ioffe, S. L.; Lobach, O. A.; Nosik,
D. N.; Tomilova, L. G. Mendeleev Commun. 2010, 1, 25.
(4) Semakin, A. N.; Sukhorukov, A. Yu.; Ioffe, S. L.;
Tartakovsky, V. A. Synthesis 2011, 1403.
(5) (a) Ioffe, S. L. In Nitrile Oxides, Nitrones and Nitronates in
Organic Synthesis; Feuer, H., Ed.; Wiley Interscience:
Hoboken, 2008, 604. (b) Dilman, A. D.; Tishkov, A. A.;
Lyapkalo, I. M.; Ioffe, S. L.; Strelenko, Yu. A.; Tartakovsky,
V. A. Synthesis 1998, 181. (c) Dilman, A. D.; Tishkov, A.
A.; Lyapkalo, I. M.; Ioffe, S. L.; Kachala, V. V.; Strelenko,
Yu. A.; Tartakovsky, V. A. J. Chem. Soc., Perkin Trans. 1
2000, 2926.
(4k)
Benzyl salt 4k was obtained from tris-oxime 1k (290 mg, 0.64
mmol).
Yield: 75% (298 mg); white solid; mp 115–123 °C.
1H NMR: d = 1.88 [br s, 4 H, (11)], 2.57 [br s, 4 H, (10)], 2.9–3.6 [3
× br s, 8 H, (5, 7 and 13)], 3.6 [s, 6 H, (8)], 4.56 [s, 2 H, (20)], 7.18–
7.36 and 7.40–7.55 [2 × m, 10 H, (1, 2, 3, 16, 17 and 18)], 8.7–9.0
[br s, 3 H, (14 and 15)].
13C NMR: d = 25.9 and 27.4 (2 × br, 10 and 11), 37.7 (5), 50–57 (br,
7 and 13), 51.3 (8), 67.8 (20), 76.4 (br, 6), 77.2 (br, 12), 126.7 (19),
125.8, 128.0, 128.9, 130.5, 130.75 and 133.1 (1, 2, 3, 16, 17 and 18),
134.5 (4), 173.5 (9).
+
HRMS (+ve): m/z calcd for C28H37N4O7 : 541.2657; found:
541.2655.
1,3-Dibenzyl-4,6,10-trihydroxy-5-(3-methoxy-3-oxopropyl)-7-
methyl-1,4,6,10- tetraazatricyclo[3.3.1.13,7]decan-1-ium Bro-
mide (4l)
Benzyl salt 4l was obtained from tris-oxime 1l (110 mg, 0.29
mmol).
(6) Bernier, D.; Blake, A.; Woodward, S. J. Org. Chem. 2008,
73, 4229.
Yield: 72% (115 mg); white solid; mp 207–210 °C (dec.).
(7) For the relative steric size of different substituents, see:
Meyer, A. Y. J. Chem. Soc., Perkin Trans. 2 1986, 1567.
(8) An additional role of acetic acid may involve quaternization
of the bridgehead nitrogen atom in tetraazaadamantanes 2,
which shifts the equilibrium between 1 and 2 towards 2.2
(9) X-ray data for 2(4e)·7H2O: Empirical formula:
C36H68Br2N8O17; M = 1044.80; triclinic; space group P1;
T = 103 K; a = 8.4154(6), b = 10.0049(7), c = 14.4240(11)
Å, a = 87.414(2), b = 86.468(2), g = 83.935(3)°;
V = 1204.43(15) Å3; Z = 1; dcalc = 1.440 gcm–3;
1H NMR: d = 1.23 [s, 3 H, (8)], 1.89 [br s, 2 H, (14)], 2.59 [br s, 2
H, (13)], 2.8–3.5 [4 × br s, 8 H, (5, 7, 10 and 16)], 3.60 [s, 3 H, (11)],
4.54 [s, 2 H, (24)], 7.18–7.35 and 7.40–7.55 [2 × m, 10 H, (1, 2, 3,
20, 21 and 22)], 7.9–9.2 [br s, 3 H, (17, 18 and 19)].
13C NMR: d = 21.2 (8), 26.4 and 27.6 (13 and 14), 38.3 (5), 51.7
(11), 50–57 and 58–62 (2 × br; 7, 10 and 16), 68.1 (24), 76-79 (br;
6, 9 and 15), 126.2 (23), 127.1, 128.4, 128.5, 129.3, 130.9 and 133.5
(1, 2, 3, 20, 21 and 22), 134.2 (4), 174.0 (12).
+
HRMS (+ve): m/z calcd for C25H33N4O5 : 469.2445; found:
m(MoKa) = 17.57 cm–1; F(000) = 546. Intensities of 13428
reflections were measured with a Bruker SMART APEX2
CCD diffractometer [l(MoKa) = 0.71072Å, w-scans,
2q<58°], and 6386 independent reflections [Rint = 0.0342]
were used in further refinement. The structure was solved by
direct methods and refined by the full-matrix least-squares
technique against F2 in the anisotropic-isotropic
469.2440.
Anal. Calcd for C25H33BrN4O5: C, 54.65; H, 6.05; N, 10.20. Found:
C, 54.36; H, 5.74; N, 10.00.
Debenzylation of Quaternary Tetraazaadamantane Salts 4d, 4f
and 4l; General Procedure 3
Palladium on activated charcoal (10%, 100 mg) and anhydrous
K2CO3 (34 mg, 0.25 mmol) were added to a stirred solution of N-
benzyl adamantane 4d, 4f or 4l (0.25 mmol) in MeOH (6 mL) under
argon. The reaction mixture was hydrogenated with H2 (1 atm) at
r.t. for 1 h (4d), 45 min (4f), or 50 min (4l). The resulting solution
was filtered through Celite and the filtrate was evaporated at 35 °C
in vacuo. The residue was preadsorbed on silica gel and purified by
flash chromatography (silica gel; EtOAc–hexane, 3:1→EtOAc) to
give pure adamantanes 2d, 2f or 2l.
approximation. The hydrogen atoms of OH groups and those
of water molecules were found in difference Fourier
synthesis. The H(C) atom positions were calculated. All
hydrogen atoms were refined in the isotropic approximation
within the riding model. For 2(4e)·7H2O, the refinement
converged to wR2 = 0.0840 and GOF = 1.002 for all the
independent reflections (R1 = 0.0494 was calculated against
F for 3820 observed reflections with I>2s(I)). All
calculations were performed using SHELXTL PLUS 5.0.10
CCDC-843973 contains the supplementary crystallographic
data for 2(4e)·7 H2O. These data can be obtained free of
charge via http://www.ccdc.cam.ac.uk/conts/retrieving.html
(or from the CCDC, 12 Union Road, Cambridge, CB21EZ,
UK; or deposit@ccdc.cam.ac.uk).
Primary Data for this article are available online at
http://www.thieme-connect.com/ejournals/toc/synthesis and can be
cited using the following DOI: 10.4125/pd0026th.
(10) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found.
Crystallogr. 2008, 64, 112.
(11) Gotlieb, H. E.; Korlyar, V.; Nudelman, A. J. Org. Chem.
1997, 62, 7512.
Acknowledgment
We are grateful for financial support from RFBR (grant 09-03-
00676a), Council for Grants of the President of Russian Federation
(grant MK-1361.2011.3) and RAS Presidium program 7P.
© Thieme Stuttgart · New York
Synthesis 2012, 44, 1095–1101