N-Cyanomethyl-β-chloro amines: chiral building blocks for the synthesis of azacrown ethers

Article abstract of DOI:10.1016/j.tetlet.2006.05.046

(1R,2S)-Ephedrine and norephedrine derived N-cyanomethyl-β-chloro amines react with tri-, tetra- and penta-ethylene glycol to stereoselectively give the corresponding amino ethers. Further transformation into chiral monoaza 12-, 15- and 18-crown-4, -5 and

Full text of DOI:10.1016/j.tetlet.2006.05.046

Tetrahedron Letters 47 (2006) 4817–4821
N-Cyanomethyl-b-chloro amines: chiral building blocks
for the synthesis of azacrown ethers
Franc¸ois Couty,* Gwilherm Evano, Laurence Menguy,
Vincent Steimetz and Mathieu Toumi
Institut LAVOISIER, UMR CNRS 8180, Universite´ de Versailles, 45, Avenue des Etats-Unis, 78035 Versailles Ce´dex, France
Received 22 March 2006; revised 2 May 2006; accepted 10 May 2006
Abstract—(1R,2S)-Ephedrine and norephedrine derived N-cyanomethyl-b-chloro amines react with tri-, tetra- and penta-ethylene
glycol to stereoselectively give the corresponding amino ethers. Further transformation into chiral monoaza 12-, 15- and 18-
crown-4, -5 and -6 ethers was realized in three steps and good overall yields by: (i) mesylation, (ii) deprotection of the N-cyanomethyl
group and (iii) intramolecular alkylation. Binding affinities of these azacrown ethers for alkali cations was studied by FAB-MS.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Chiral cryptands have recently found a growing number
of applications in many fields such as enantioselective
catalysis,1 selective transport of enantiomers2 and enan-
tiomeric differentiation.3 Enantiopure b-amino alcohols
being among the most widely used members of the chiral
pool,4 one would expect their incorporation in many
azamacrocycles. Surprisingly, a careful examination of
the literature shows that they have only been scarcely
used for the preparation of azacrownethers in which
they systematically act either as N,N-bis nucleophiles,
which lead to a chiral azamacrocycle bearing the chiral-
ity outside the ring or as N,O-bis nucleophile, that gives
access to macrocycles possessing the chirality within the
ring (Fig. 1).5
We recently reported that amino alcohols-derived
N-cyanomethyl-b-chloroamines act as efficient, easy to
use and readily available building blocks for the synthesis
of chiral amino ethers or diamines by their simple reac-
tion with alcohols and amines, a reaction that was shown
to proceed via an aziridinium intermediate.6 In this letter,
we report the use of these chiral synthons for the efficient
preparation of azacrown ethers. This strategy is com-
pletely different from those reported in the sense that
the amino alcohol now acts as both an electrophilic and
nucleophilic partner. This results in a highly efficient
formation of the macrocycle with no modification of
the absolute and relative configuration of the starting
amino alcohol. Importantly, this procedure allows for
1a
Incorporation of chirality ouside the macrocycle
O
K CO
OH
2
3
N
O
OH
(
)
n
+
I
I
NH
(
2
MeCN
(
n
3b
Incorporation of chirality within the macrocycle
O
O
1) Ethylene oxide
O
OH
O
NBn
(
)
2
2)
NHBn
TsO
OTs
O
Figure 1. Use of chiral amino alcohols as bis-nucleophiles in azacrown synthesis: the possibilities.
*
Corresponding author. Tel.: +33 139254455; fax: +33 139254452; e-mail: couty@chimie.uvsq.fr
0040-4039/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.05.046

4818
F. Couty et al. / Tetrahedron Letters 47 (2006) 4817–4821
)
O
OH
O
n
(
)
n
(
2 steps6
HO
OH
Ph
Ph
O
N
Cl
N
(1R,2S)-Ephedrine
neat, 80 ºC
CN
CN
Me
Me
Me
Me
1
2 n = 2 (82%)
3 n = 3 (84%)
4 n = 4 (60%)
(
)
O
OMs
n
(
1) MsCl, Et3N
2) AgNO3, THF/H2O
Ph
Ph
O
O
O
Et3N, CH3CN
(
n-1
70 ºC
NH
Me
Me
O
Me
NMe
8
9
n = 2 (60%)
n = 3 (79%)
5 n = 2
6 n = 3
7 n = 4
10 n = 4 (84%)
Scheme 1. Ephedrine-derived azacrown ethers.
the use of highly substituted aminoalcohols which would
typically act as poor nucleophiles. Using this protocol,
ephedrin-derived aminoalcohols, which are inexpensive
and available in both enantiomeric forms, can therefore
be incorporated in the macro- cyclic core of azacrown
ethers.
the first step (typically 20 mL/g) and the more expensive
penta-ethylene glycol was used only in twofold molar
excess (2 mL/1.5 g) without a severe drop of the yield;
if the cyclization step was quite sluggish (typically
5 days), it did not require high dilution to avoid side
reactions.
The first step of our synthesis consists in the neat reac-
tion of 1 with tri-, tetra- or penta-ethylene glycol at
80 ꢁC: in fair to good yields, the corresponding ethers
2–4 were obtained with retention of configuration at
the reacting centre.7 Activation of the primary alcohol
as a mesylate was then followed by deprotection of the
N-cyanomethyl protecting group using silver nitrate in
a H2O/THF solvent mixture. Intramolecular alkylation
was finally initiated by treating the crude amine with tri-
ethylamine in acetonitrile at 80 ꢁC: to our delight, the
desired azamacrocycles 8–10 were obtained in good
overall yields (Scheme 1).8
We next wanted to further test our reaction sequence
and to determine whether another functionalized/func-
tionalizable group could be introduced on the nitrogen
atom without affecting its overall efficiency or not.
Therefore, (1R,2S)-norephedrine 11 was selectively
monoalkylated with t-butylbromoacetate (BrCH2CO2-
t-Bu, DBU, toluene, rt), which was followed by intro-
duction of the N-cyanomethyl protecting group by
alkylation with bromoacetonitrile. The resulting amino
alcohol 13 was chlorinated to give 14 with retention of
configuration as previously demonstrated for related
substrates,9 which was next reacted with tetra-ethylene
glycol. Using the conditions described above, ether 15
was isolated in only 33% yield, which was attributed
to a partial cleavage of the t-butyl ester by the hydro-
chloric acid released during the process. Addition of tri-
ethylamine thus allowed to limit this side reaction,
improving the yield to 56%. Mesylation, deprotection
and final ring-closure were next conducted using the
Key features of these syntheses are their straightfor-
wardness and their easy scale up. Interestingly, the yield
of the cyclization increases with the size of the ring
formed during the process. Moreover, the dilution did
not have a pronounced effect on this cyclization since
tri- and tetra-ethylene glycols were used as solvents in
Ph
Ph
OH
OH
N
Ph
OH
BrCH CN, K CO
3
BrCH CO tBu
2
2
2
2
NH
CN
Me
BuCO
Me
NH
2
MeCN, reflux
88%
Me
t
BuCO
t
2
2
11
12
13
54%
SOCl CH Cl
2,
2
2
reflux
(
quant.
CN
1) MsCl, Et N
)
3
O
OH
Tetraethylene
3
Ph
O
O
O
(
Ph
Ph
(
Cl
N
2) AgNO , THF/H O
O
N
glycol, Et N, 80 ºC
3
2
3
2
3) Et N, MeCN
3
reflux
Me
N
CN
Me
Me
t
BuCO
2
t
BuCO
2
t
BuCO
2
60% overall
56%
16
15
14
Scheme 2. Norephedrine-derived azacrown ethers: functionalized azacrown.

F. Couty et al. / Tetrahedron Letters 47 (2006) 4817–4821
4819
presence of the additional bulky t-butyl ester was found
to slow down the rate of the cyclization reaction
(10 days vs 5 days for the cyclization to 9) but a fair
60% yield of 16 was still obtained (Scheme 2).
relative
intensities
100
90
80
70
60
50
40
30
20
10
0
Binding abilities of macrocycles 8–10 and 16 for Li+,
Na+ and K+ cations were next determined using the
FAB-MS competition technique. Data obtained with
this technique are collected in Figure 2 where relative
peak intensities of [macrocycle + metal]+ ions reflect
the relative cation binding affinities.10
Li+ e1
Na+
K+
e3
8
9
10
16
Quite unexpectedly, increasing the ring size resulted in a
higher affinity for the lithium cation (compare 8 and 10).
However, modifying the N-substituent did not affect a
very good binding selectivity for the sodium cation
(compare 9 and 16).
Figure 2. Relative binding affinities of macrocycles 8–10 and 16 for
alkali cations. Conditions: macrocycle (10ꢀ5 M in acetonitrile/H2O: 9/
1) was complexed with equimolar amounts of LiCl, NaCl and KCl
(10ꢀ7, 2 · 10ꢀ7 and 10ꢀ6 M each in acetonitrile/H2O: 9/1). Values
reported in the graph are an average of the peak intensities for MS
spectra recorded for these three concentrations.
Finally, NMR data of 8 displayed an unexpected feature:
we found that the resolution of the 1H NMR spectrum of
8 in CDCl3 at rt was dependent on the concentration of
the sample: while sharp well-resolved signals were re-
corded at high concentration (0.3 mol Lꢀ1), the spectrum
conditions optimized for the ephedrine-derived sub-
strates and gave the functionalized macrocycle 16. The
Figure 3. 1H NMR of 8 in CDCl3 at rt. (A): C = 0.3 mol Lꢀ1. (B): C = 1.410ꢀ2 mol Lꢀ1
.

4820
F. Couty et al. / Tetrahedron Letters 47 (2006) 4817–4821
6. Couty, F.; Evano, G.; Prim, D. Tetrahedron Lett. 2005, 46,
2253.
showed broadened resonances at low concentration
(Fig. 3). This unusual observation suggests the occur-
rence of a slow conformational exchange of the macro-
cycle whose rate is dependent on the concentration.
The well-resolved spectrum at high concentration might
be due to a freezing of this conformational exchange that
could result from a supramolecular organization of the
macrocycles due to intermolecular p-stacking interac-
7. The observed retention result from the regioselective
opening of an intermediate aziridinium ion, as previously
demonstrated with simple alcohols, and azide anions
involved as nucleophiles, see: (a) Couty, F.; Evano, G.;
Prim, D. Tetrahedron Lett. 2005, 46, 2306; (b) Couty, F.;
Durrat, F.; Prim, D. Tetrahedron Lett. 2004, 45,
3725.
8. Representative experimental procedures: General procedure
for the reaction of N-cyanomethyl-b-chloroamines with
alcohols. A solution of the chloride (1 mmol) in the required
alcohol (10 mL) was heated for 1 h at 80 ꢁC. The reaction
mixture was next diluted with water (30 mL) and diethyl
ether (50 mL) and basified with a saturated aqueous
solution of NaHCO3. Aqueous layer was extracted with
ether, combined organic layers were washed with brine,
dried over MgSO4, filtered and concentrated under reduced
pressure. The residue was finally purified by flash column
chromatography (silica gel, elution: ether followed ether/
methanol: 5/95).
1
tions between phenyl substituents. Resolution of the H
NMR spectrum recorded at a low concentration was also
restored by addition of NaI or of 1 equiv of mandelic
acid.
In conclusion, we have described a new strategy for the
synthesis of monoazamacrocycles of various sizes from
b-amino alcohols. Since amines were also shown to react
efficiently with N-cyanomethyl-b-chloro amines, this
strategy may be extended to prepare other classes of
polyazamacrocycles. The evaluation of these com-
pounds for enantiomeric differentiation is currently
under study.
(1S,2R)-{[2-(2-{2-[2-(2-Hydroxyethoxy)-ethoxy]-ethoxy}-
ethoxy)-1-methyl-2-phenylethyl]-methylamino}-acetonitrile
20
3. Yield: 84%; ½aꢁD ꢀ35 (c 1.4, CHCl3); 1H NMR
(300 MHz, CDCl3): d 1.08 (d, J = 6.9 Hz, 3H), 2.54 (s,
3H), 2.82 (qd, J = 6.9 and 3.0 Hz, 1H), 2.98 (br s, 1H), 3.47–
3.82 (m, 18H), 4.59 (d, J = 3.0 Hz, 1H), 7.25–7.38 (m, 5H);
13C NMR (75 MHz, CDCl3): 8.9, 40.0, 41.8, 61.7, 64.0,
68.3, 70.4, 70.5, 70.6, 70.8, 72.6, 72.8, 83.4, 117.5, 126.8,
127.4, 128.3, 140.2; MS (CI, NH3): 381, 354, 97.
Acknowledgements
CNRS is acknowledged for generous support.
General procedure for the mesylation. To a solution of the
alcohol (1 mmol) and triethylamine (280 lL, 2 mmol) in
dichloromethane (10 mL) was added methanesulfonylchlo-
ride (175 lL, 1.5 mmol) dropwise at 0 ꢁC. The reaction
mixture was warmed to rt, stirred for 1 h, and quenched by
addition of water. Aqueous layer was extracted with
dichloromethane, combined organic layers were washed
with brine, dried over MgSO4, filtered and concentrated
under reduced pressure to give the crude mesylate deriva-
tives which were used in the next step without further
purification.
References and notes
1. For enantioselective catalysis involving monoazamacro-
cycles, see: (a) Juanes, O.; Rodriguez, J. C.; Brunet, E.;
Pennemann, H.; Kossenjans, M.; Martens, J. Eur. J. Org.
Chem. 1999, 3323; (b) Brunet, E.; Poveda, A. M.;
Rabasco, D.; Oreja, E.; Font, L. M.; Singh Batra, M.;
´
Rodrıguez-Ubis, J. C. Tetrahedron: Asymmetry 1994, 5,
´
´
´
935; (c) Bako, T.; Bako, P.; Keglevich, G.; Bathori, N.;
´
Czugler, M.; Tatai, J.; Novak, T.; Parlagh, G.; To¨ke, L.
Tetrahedron: Asymmetry 2003, 14, 1917.
2. Newcomb, M.; Toner, J. L.; Helgeson, R. C.; Cram, D. J.
J. Am. Chem. Soc. 1979, 101, 4941.
General procedure for the deprotection/cyclization step. To a
solution of mesylate (0.623 mmol) in THF (5 mL) was
added a solution of silver(I)nitrate (136 mg, 0.8 mmol) in
water (1 mL). Upon addition, a brownish precipitate was
immediately formed and the suspension was stirred in the
dark at rt for 5 h. Crude reaction mixture was then filtered
through a plug of Celiteꢂ, which was thoroughly washed
with DCM and then with methanol to avoid loss of product.
Filtrate was concentrated under reduced pressure to give
crude N-deprotected compound as a nitrate salt. This
residue was dissolved in acetonitrile (20 mL), and triethyl-
amine (175 lL, 1.25 mmol) was added. The solution was
heated at 70 ꢁC for 5 days (or until consumption of starting
material), cooled to rt, filtered through a plug of Celiteꢂ,
which was again thoroughly washed with DCM and then
with methanol to avoid loss of product. Filtrate was
concentrated under reduced pressure and the residue was
treated with water and with 2 M aqueous NaOH solution.
Aqueous layer was extracted with ether (3 · 20 mL), com-
bined organic layers were dried over MgSO4, filtered and
concentrated under reduced pressure. The residue was finally
purified by flash column chromatography (silica gel, elution:
ether then ether/30% NH4OH: 98/2, then ether/30%
NH4OH/ethanol: 93/2/5) to give the desired macrocycle.
(11R,12S)-12,13-Dimethyl-11-phenyl-1,4,7,10-tetraoxa-13-
3. For recent examples, see: (a) Colera, M.; Costero, A. M.;
Gavina, P.; Gil, S. Tetrahedron: Asymmetry 2005, 16, 2673;
˜
(b) Togrul, M.; Askin, M.; Hosgoren, H. Tetra-
hedron: Asymmetry 2005, 16, 2771; (c) Ragusa, A.; Rossi,
S.; Hayes, J. M.; Stein, M.; Kilburn, J. D. Chem. Eur. J.
2005, 11, 5674; (d) Lee, C.-S.; Teng, P.-F.; Wong, W.-L.;
Kwong, H.-L.; Chan, A. S. C. Tetrahedron 2005, 61, 7924;
(e) Nakatsuji, Y.; Nakahara, Y.; Muramatsu, A.; Kida, T.;
Akashi, M. Tetrahedron Lett. 2005, 46, 4331; (f) Togrul,
M.; Turgut, Y.; Hosgo¨ren, H. Chirality 2004, 16, 351; (g)
Wong, W.-L.; Huang, K.-H.; Teng, P.-F.; Lee, C.-S.;
Kwong, H.-L. Chem. Commun. 2004, 384; (h) Demirel, N.;
Bulut, Y. Tetrahedron: Asymmetry 2003, 14, 2633.
4. (a) Cruz, A.; Juarez-Juarez, M. Curr. Org. Chem. 2004, 8,
671; (b) Rizzacasa, M.; Perkins, M. Stoichiometric Asym-
metric Synthesis. Postgraduate Chemistry Series; Sheffield
Academic Press: Sheffield, 2000.
5. For examples of syntheses of achiral monoazamacrocycles,
see: (a) Ahmed, S. A.; Tanaka, M.; Ando, H.; Iwamoto,
H.; Kimura, K. Eur. J. Org. Chem. 2003, 2437; (b)
Bratton, L. D.; Strzelbicka, B.; Bartsch, R. A. Arkivoc
2003, xiii, 80; (c) Elshani, S.; Kobzar, E.; Bartsch, R. A.
Tetrahedron 2000, 56, 3291; (d) Tsukube, H.; Inoue, T.;
Hori, K. J. Org. Chem. 1994, 59, 8047; (e) Krakowiac, K.
E.; Bradshaw, J. S. J. Org. Chem. 1991, 56, 3723.
20
aza-cyclopentadecane 9. Yield: 79%; ½aꢁD ꢀ525 (c 0.36,
CHCl3); 1H NMR (300 MHz, CDCl3):
d 0.93 (d,
J = 7.0 Hz, 3H), 2.40 (s, 3H), 2.63–2.76 (m, 2H), 3.21–

F. Couty et al. / Tetrahedron Letters 47 (2006) 4817–4821
4821
3.30 (m, 1H), 3.46–3.77 (m, 14H), 4.71 (d, J = 1.5 Hz, 1H),
7.25–7.42 (m, 5H); 13C NMR (75 MHz, CDCl3): 7.1, 42.3,
51.2, 65.6, 68.5, 70.45, 70.49, 70.52, 70.62, 70.9, 71.0, 84.1,
126.6, 126.9, 128.2, 141.6; MS (CI, NH3): 324.
9. Agami, C.; Couty, F.; Evano, G. Tetrahedron: Asymmetry
2002, 13, 297.
10. Tsukube, H.; Inoue, T.; Hori, K. J. Org. Chem. 1994, 59,
8047.