CsOH-promoted chemoselective mono-N-alkylation of diamines and polyamines

Article abstract of DOI:10.1016/S0040-4039(00)01747-0

Selective N-alkylation of diamines and polyamines was carried out using cesium hydroxide, 4 ? molecular sieves, and DMF. This protocol was highly chemoselective, favoring mono-N-alkylation over overalkylations. (C) 2000 Published by Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)01747-0

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 9705–9708
CsOH-promoted chemoselective mono-N-alkylation of
diamines and polyamines
Ralph N. Salvatore,a Shaun E. Schmidt,a Seung Il Shin,a Advait S. Nagle,a
Jay H. Worrella and Kyung Woon Junga,b,
*
aDepartment of Chemistry, University of South Florida, 4202 E. Fowler Avenue, Tampa, FL 33620-5250, USA
bDrug Discovery Program, H. Lee Moffitt Cancer Center & Research Institute, Tampa, FL 33612-9497, USA
Received 15 September 2000; accepted 5 October 2000
Abstract
,
Selective N-alkylation of diamines and polyamines was carried out using cesium hydroxide, 4 A
molecular sieves, and DMF. This protocol was highly chemoselective, favoring mono-N-alkylation over
overalkylations. © 2000 Published by Elsevier Science Ltd.
Naturally occurring diamines and polyamines have attracted considerable attention in syn-
thetic organic chemistry due to their interesting biological activities,1 potential pharmacological
applications,2 and asymmetric catalysis.3 However, their synthetic use is quite limited mainly due
to lack of efficient methods for monofunctionalization, although this problem has been averted
by employing amines in large excess, which can be rather expensive. Recently, we reported a
cesium hydroxide-promoted chemoselective protocol for the mono-N-alkylation of a primary
amine,4 and discussed herein is the application of this methodology to diamines and polyamines,
resulting in high chemoselectivity.
In our laboratories, we set out to synthesize 1,8-diamino-3-thia-6-azaoctane (NNSN) 3, in
efforts to study metal complexation effects with cobalt(III)5 (Scheme 1). Although extensive
studies for the synthesis of this compound have been reported, they lack in efficiency since the
procedures are toxic6 and low yielding.7 In our initial synthesis, 2-thioethylamine·HCl 1 and
N-(2-bromoethyl)phthalimide 2 were reacted using potassium carbonate in refluxing acetonitrile8
to afford the coupled product in low yield. Upon deprotection of the phthalimide moiety,
NNSN was produced in extremely low yield (9% yield over two steps), and other side products
were formed concomitantly. Much to our surprise, the N-alkylation in the presence of cesium
carbonate proved more problematic, and the desired product was not produced. As demon-
strated using a model substrate such as phenethylamine 5, 2-oxazoline 6 was isolated along with
cyclic amide 7, stemming from aminolysis followed by intramolecular alkylations.
* Corresponding author.
0040-4039/00/$ - see front matter © 2000 Published by Elsevier Science Ltd.
PII: S0040-4039(00)01747-0

9706
O
1+
H2N
Co
Br
O2N
N
S
H
N
(2)
NH2
O
O
NH2·HCl
H2N
O
S
O2N
HS
K2CO3, CH3CN, reflux,
then H2NNH2, EtOH, reflux
9% in two steps
NH
3
1
H2N
H
N
4
N
2, Cs2CO3
NH2
H
N
+
Ph
DMF, 23 °C, 43 h
N
Ph
5
Ph
O
O
6 (67%)
7 (12%)
Scheme 1.
In order to avoid aminolysis, unmasked electrophiles including 2-bromoethylamine·HBr 8
were screened using a variety of bases, which met with failure. However, when CsOH was
employed (Scheme 2),4 the S-alkylated product 9 and the desired product 3 were acquired in 44
and 24% yields, respectively. In addition, NSN 9 was recycled for direct N-alkylation to deliver
NNSN 3 predominantly in 73% isolated yield (56% combined yield), which is far superior to the
existing methods.9 It is noteworthy that the improved yield originated from the high chemoselec-
tivity in the monoalkylation of diamine 9, prompting us to further develop this technique for
general applications.
CsOH·H2O
NH2·HCl
H2N
NH2
+
3 (24%)
+
1
Br
S
DMF, 4 Å MS, 23 °C, 44 h
9 (44%)
8
CsOH·H2O, 8
H2N
NH2
3 (73%)
S
DMF, 4 Å MS, 23 °C, 12 h
9
Scheme 2.
As representatively depicted in Table 1, various electrophiles as well as numerous diamines
and polyamines were subjected to the N-alkylation conditions to evaluate the efficiency and
chemoselectivity. In the presence of a catalytic amount of CsOH (0.5 equiv.), ethylene diamine
was found to react with TsCl to produce the N-sulfonamide in high yield. Similarly, unreactive
bromide 13 reacted with symmetrical diamines including 12 and 14 (entries 2 and 3), offering
similar yields. Polyamines encompassing diethylenetriamine 15 still proved pragmatic, reacting
at the primary amine and leaving the secondary amine untouched (entry 4). Using a reactive
halide such as benzyl chloride, bis-tetramine 16 (entry 5) and putrescine 18 (entry 6) were still
facile under the developed conditions, delivering the mono-N-alkylated polyamine and diamine,
respectively. Within our detection limits, dialkylation on the same nitrogen was not detected,
and bisalkylation on the different nitrogens was minimized (<10%) by performing the reaction
at 0°C. Diamines containing elements of stereochemistry were also found to be compatible.
(1R,2R)-(+)-1,2-Diphenyl-1,2-ethanediamine 19 underwent N-benzylation, breaking the symmet-
rical element within the molecule (entry 7) and no overalkylation was seen. Although the
starting materials were not consumed, they were usually eliminated during aqueous work-up,
offering the desired mono-N-alkylation product predominantly in the crude product.

9707
Table 1
yielda
entry
diamine or polyamine
time
2 h
halide
NH2
H2N
75%b
65%
(10)
(12)
(14)
(15)
TsCl (11)
1
2
(13)
Br
24 h
H2N
H2N
NH2
NH2
Ph
13
13
3
4
14 h
24 h
61%
58%
8
H
N
H2N
NH2
H
N
NH2
52%c
65%
(16)
(18)
5
6
BnCl (17)
12 h
12 h
H2N
N
H
NH2
17
H2N
Ph
Ph
(19)
7
17
12 h
79%
H2N
NH2
a Isolated yields of mono-N-alkylated amines. The starting amines were not consumed completely,
accounting for the additional mass balance. b The reaction was run using a catalytic amount of
CsOH at 0 °C. c Overalkylation was minimized by performing the reaction at 0 °C.
Unsymmetrical diamines also reacted regioselectively under the developed conditions (Scheme
3). 1,2-Diaminopropane 20 was found to alkylate exclusively at the primary amine, leaving the
more sterically hindered secondary amine intact. Chiral diamine, (
L
)-lysine methyl ester 22, was
selectively N-benzylated at the primary center, and no racemization was detected.10
NH2
NH2
H
N
CsOH·H2O, 13
NH2
Ph
4 Å MS, DMF, 23 °C, 28 h
20
21 (77%)
NHBn
NH2·HCl
CsOH·H2O, 17
4 Å MS, DMF, 0 °C-rt, 12 h
H2N
CO2Me
23 (70%)
HCl·H2N
CO2Me
22
Scheme 3.
To further attest issues of chemoselectivity, N-alkylation of amino alcohols was performed for
the preparation of higher order peptidomimetics such as trimer 26 (Scheme 4). For example,
amino bromide 25 was coupled with unsymmetrical dimer 24, where the secondary bromide was
rearranged to the desired primary form via the corresponding aziridinium salt 27 during the
N-alkylation.11 The primary amine was found to react exclusively with the aziridinium salt,
leaving the secondary amine and the alcohol functionality intact. Besides high chemoselectivity,
racemizations were not detected during any alkylations of these chiral peptidomimetic
substrates.12

9708
via
H
N
NBn2
H
N
(25)
Br
Bn2N
Br
N
H
OH
H2N
OH
Bn2N
27
CsOH·H2O, 4 Å MS
24
26
DMF, 23 °C, 36 h, 46%
Scheme 4.
In conclusion, we have developed a convenient and efficient protocol for the chemoselective
mono-N-alkylation of diamines and polyamines using a wide variety of halides. Our protocols
offered moderate to high yields, and the use of protecting groups proved to be unnecessary.
Furthermore, our technology was applicable to the efficient synthesis of peptidomimetic
compounds.
Acknowledgements
We gratefully acknowledge financial supports from the H. Lee Moffitt Cancer Center &
Research Institute and the American Cancer Society (Institutional Research Grant c032).
References
1. For recent reviews on polyamines and their derivatives, see: (a) Advances in Polyamine Research 1 and 2;
Campbell, R. A.; Morris D. R.; Bartos, D.; Davies, G. D.; Bartos F., Eds.; Raven Press: New York, 1978. (b)
Ganem, B. Acc. Chem. Res. 1982, 15, 290 and references cited therein.
2. For a comprehensive review on the neuropharmacology of polyamines, see: The Neuropharmacology of
Polyamines; Carter, C., Ed.; Academic Press: San Diego, 1994.
3. Ojima, I. Catalytic Asymmetric Synthesis; VCH Verlag: Weinheim, 1993.
4. Salvatore, R. N.; Nagle, A. S.; Schmidt, S. E.; Jung, K. W. Org. Lett. 1999, 1, 1893.
5. McClure, M. R.; Jung, K. W.; Worrell, J. H. Coord. Chem. Rev. 1998, 174, 33.
6. Teumac, F. U. S. Patent 3,362,996, 1968; Chem. Abstr. 1968, 68, 77738.
7. Weiss, A. L.; Chen, S.-F.; Reddy, P. S.; Mittakanti, M.; Dexter, D. L.; Woynarowski, J. M. U. S. Patent
5,561,042, 1996.
8. Glennon, R. A.; Naiman, N. A.; Pierson, M. E.; Smith, J. D.; Ismaiel, A. M.; Titeler, M.; Lyon, R. A. J. Med.
Chem. 1989, 32, 1921.
,
9. Representative experimental procedure: To activated powdered dry 4 A molecular sieves (5.0 g) in anhydrous
N,N-dimethylformamide (45 mL), were added cesium hydroxide monohydrate (3.7 g, 22 mmol) and 1,5-diamino-
3-thiopentane 9 (1.1 g, 8.8 mmol). With vigorous stirring, 2-bromoethylamine hydrobromide 8 (2.7 g, 13 mmol)
was added and the mixture was stirred at room temperature for 12 h, at which time the reaction was quenched
with 100 mL of 1 N NaOH. Following filtration to remove solids, the solution was concentrated by blowing air.
The resulting solid was then taken up in a small amount of methanol, triturated with Et2O, subsequently filtered,
and reduced to dryness in vacuo. Trituration was repeated two more times to ensure the removal of the inorganic
salts. The resulting thick yellow oil was then subjected to silica gel chromatography using 5% NH4OH–MeOH,
followed by 15% NH4OH–MeOH, which gave 3 (1.04 g, 73% yield).
10. Product 23 was converted back to the starting diamine 22 by hydrogenolysis, followed by salt formation using
dry HCl gas. The optical rotation of the resultant salt was 16.5°, whereas the reported value is 15.6° (c=1.3;
H2O). See: J. Org. Chem. 1963, 2898.
11. Nagle, A. S.; Salvatore, R. N.; Chong, B. D.; Jung, K. W. Tetrahedron Lett. 2000, 41, 3011.
12. 1H, 13C and 2-D NMR indicated the product as a single diastereomer.