A convenient synthesis of chiral dioxocyclens and application as chiral solvating agents

Article abstract of DOI:10.1016/S0040-4039(02)00623-8

This paper reports a very short and efficient synthesis of chiral dioxocyclens starting from natural amino acids. NMR experiments were undertaken to assess the chiral recognition properties of these chiral macrocycles. The NMR spectra of mandelic acid or its derivatives in the presence and absence of the chiral dioxocyclens showed that these macrocycles have different enantiomeric discriminating ability. It was revealed that this type of dioxocyclen may be promising hosts for chiral discrimination.

Full text of DOI:10.1016/S0040-4039(02)00623-8

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 3935–3937
A convenient synthesis of chiral dioxocyclens and application as
chiral solvating agents
Quan Yuan,a Enqin Fu,a,* Xiaojun Wu,a Maohai Fang,a Peng Xue,a Chengtai Wua and
Jiahua Chenb
aDepartment of Chemistry, Wuhan University, Wuhan 430072, PR China
bCollege of Chemistry and Molecular Engineering, Beijing University, Beijing 100087, PR China
Received 19 December 2001; revised 22 March 2002; accepted 28 March 2002
Abstract—This paper reports a very short and efficient synthesis of chiral dioxocyclens starting from natural amino acids. NMR
experiments were undertaken to assess the chiral recognition properties of these chiral macrocycles. The NMR spectra of mandelic
acid or its derivatives in the presence and absence of the chiral dioxocyclens showed that these macrocycles have different
enantiomeric discriminating ability. It was revealed that this type of dioxocyclen may be promising hosts for chiral discrimination.
© 2002 Elsevier Science Ltd. All rights reserved.
Chiral macrocyclic compounds have been recognized as
successful and promising chiral selectors for molecular
recognition mainly because of their inherent reduced
flexibility and complexation ability.1 Macrocyclic amides
that bear the dual features of macrocyclic polyamines
and oligopeptides, can act as both hydrogen bond
acceptors and donors and can form complexes with
neutral molecules. So this kind of chiral dioxocyclen may
have potential applications for enantiomeric recognition,
especially with neutral molecules and anions, such as
drugs and biomolecules.2 However, the preparation of
these molecules is often fraught with difficult procedures,
low yields and laborious purifications. Only a few reports
about the synthesis and application of chiral peraza-
macrocycles have appeared.3 Thus, the development of
an efficient synthesis of these highly desirable chiral
compounds remains of key interest. In this paper we
report a very short and efficient synthesis of chiral
dioxocyclens (Scheme 1) and their application as chiral
solvating agents (CSAs) for enantiomeric discrimination.
Scheme 1. The synthesis of dioxocyclens starting from amino acids.
* Corresponding author. Tel.: +86-27-87684117; fax: +86-27-87647617; e-mail: fueq@chem.whu.edu.cn
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)00623-8

3936
Q. Yuan et al. / Tetrahedron Letters 43 (2002) 3935–3937
L
-Amino acids were converted into methyl esters, and
Many approaches are available for studying the struc-
ture and dynamics of multimolecular complexes. The
NMR method is perhaps most often used for the
examination of such complexes in solution because it
can provide direct structural and dynamic information.6
Studies on the molecular recognition were carried out
on a 300 MHz NMR spectrometer using the com-
pounds 1–4 as CSAs.
reacted with methyl chloroacetate to give the intermedi-
ates 6, which then reacted with diethylenetriamine to
afford the desired chiral products after recrystallization
twice from CH3CN. The total yields were about 10%.
The structures were confirmed by MS, H, 13C NMR
1
spectra and microanalysis.4
This synthetic approach has some advantages in that all
the products were synthesized without N-protection
and high dilution conditions are not required. The
introduction of a chiral unit into a macrocyclic ring is
expected to give steric rigidity at the chiral carbon that
may be helpful in chiral recognition, and different
substituents at the chiral center will provide informa-
tion on the mechanism of the chiral recognition. Fur-
thermore, the imido NH can be functionalized easily.5
The racemates of mandelic acid and its derivatives were
chosen as substrates. The signal of the methine hydro-
gen for these substrates is a sharp singlet and does not
overlap with the peaks of the other proton signals in
1
their H NMR spectra. Therefore, it is an ideal probe
for discrimination. Samples for analysis were prepared
by mixing equimolar amounts of the substrate and the
chiral dioxocyclen (the concentrations were normally 20
mM) in CDCl3.
The methine proton signals of all substrates were
shifted upfield about 0.1 ppm in the presence of the
chiral dioxocyclen. Some were split into two peaks due
to the different interactions of the two enantiomers of
the substrate with the CSA. This confirmed that chiral
recognition had occurred. Fig. 1 (a and b) shows the
methine signal in the 1H NMR spectra of mandelic acid
in the absence and presence of an equimolar amount of
the compound 2. The non-equivalence (DDl, Hz) is the
difference of the chemical shifts of corresponding pro-
tons of the two enantiomers in the presence of the CSA.
In Fig. 1b, the methine signal has clearly split into two
peaks (DDl 8.4 Hz). The methine signal of the (R)-
enantiomer is more intense than that of the (S)-enan-
tiomer because the (R)-mandelic acid was added in
slight excess to the solution. The fact that the (R)-enan-
tiomer has larger chemical shift changes than the (S)-
enantiomer reveals that the (R)-mandelic acid has a
stronger interaction with compound 2. Table 1 summa-
rizes the DDl values of the methine signals of the
enantiomers of the four substrates in the presence of
the dioxocyclens 1ꢀ4, respectively. It can be seen that
Figure 1. (a) The methine proton signal of mandelic acid; (b)
the methine proton signals of mandelic acid in the presence of
an equimolar amount of compound 2; (c) part of phenyl
proton signal of 4-methoxymandelic acid; (d) part of phenyl
proton signal of 4-methoxymandelic acid in the presence of
an equimolar amount of compound 2.
Table 1. Differences (DDl) between the chemical shifts of the methine signals of the enantiomers of the substrates in the
presence of the dioxocyclens 1–4 (host:guest=1:1)

Q. Yuan et al. / Tetrahedron Letters 43 (2002) 3935–3937
3937
dioxocyclens 1ꢀ4 exhibit different enantiomeric dis-
criminating ability. In particular, for recognition of
mandelic acid, less steric hindrance at the chiral carbon
of the dioxocyclen leads to weaker chiral recognition.
As for compound 1, although the proton signals of the
substrates shifted over 0.1 ppm, little chiral recognition
occurred.
Verleysen, K.; Vandijck, J.; Schelfaut, M.; Sandra, P. J.
High Resolut. Chromatogr. 1998, 21, 323–331.
2. Hu, S.; Fu, E.; Li, P. H. J. Chromatogr., A. 1999, 844,
439–446.
3. (a) Kim, B. M.; So, S. M. Tetrahedron Lett. 1999, 40,
7687–7690 and references cited therein; (b) Achmatowicz,
M.; Jurczak, J. Tetrahedron: Asymmetry 2001, 12, 111; (c)
Bhattacharyya, T.; Nilsson, N. J. Tetrahedron Lett. 2001,
42, 2873.
1
The chemical shifts of the amide NH signals in the H
NMR spectra of the dioxocyclens move downfield by
0.1–0.3 ppm. Meanwhile, both phenyl ring proton sig-
nals of the four substrates and compound 2 were
shifted. This implies that a p–p interaction might occur
between the phenyl rings of the substrates and that of
the host. As to 4-methoxymandelic acid, the peaks of
the phenyl ring proton signals which were not overlap-
ping with the host phenyl ring proton signals were
clearly split into quartets in the presence of compound
2 (Fig. 1 c and d), and the proton signals of the MeO
group also split into two peaks (DDl=3.6) because of
the change in electronic density on the phenyl ring.
4. Compound 1: [h]2D0=23.71 (c=0.8, MeOH), 1H NMR (300
MHz, CDCl3): l=7.55 (br, 1H, CONH), 7.38 (br, 1H,
3
CONH), 3.17–3.24 (q, J=6.9 Hz, 1H, NHCHCO), 3.38–
3.68 (m, 4H, 2CONHCHH, NCH2CO), 2.58–3.11 (m, 6H,
2CONHCHH·2NHCH2), 1.83 (br, 2H, 2NH), 1.34–1.36
3
(d, J=6.6 Hz, 3H, CH3), 13C NMR (75 MHz, CDCl3): l
174.97, 172.03, 62.25, 53.94, 45.34, 45.05, 38.31, 37.63,
20.59; MS (FAB): 215 [M+1]+, mp: 181–183°C. Anal.
calcd for C9H18N4O2, C 50.45; H, 8.47; N, 26.15. Found:
C, 50.6; H, 8.1; N, 25.4%. Compound 2: [h]2D0=40.34
(c=1.0, MeOH), 1H NMR (300 MHz, CDCl3): l=7.54
(br, 1H, CONH), 7.49 (br, 1H, CONH), 7.26–7.42 (m, 5H,
Ph), 3.27–3.64 (m, 5H, 2CONHCHH, NHCH2CO,
NHCHCO), 3.03–3.22 (m, 2H, 2CONHCHH), 2.64–3.01
(m, 6H, 2NHCH2, PhCH2), 1.94 (br, 2H, 2NH); 13C NMR
(75 MHz, CDCl3): l 173.69, 172.01, 129.32, 129.26,
127.48, 67.49, 57.31, 45.32, 45.04, 39.90, 38.36, 37.63: MS
(FAB): 291 [M+1]+, mp: 192–194°C. Anal. calcd for
C15H22N4O2: C, 62.05; H, 7.64; N, 19.30. Found: C, 61.6;
H, 7.3; N, 19.3%. Compound 3: [h]2D0=39.11 (c=1.0,
MeOH), 1H NMR (300 MHz, CDCl3): l=7.69 (br, 1H,
CONH), 7.30 (br, 1H, CONH), 3.04–3.77 (m, 6H,
NHCH2CO, 2CONHCH2), 2.67–2.99 (m, 5H, NHCHCO,
2NHCH2), 1.86 (br, 2H, 2NH), 1.61–2.12 (m, 1H,
CH3CHCH3), 0.97–1.05 (m, 6H, 2CH3), 13C NMR (75
MHz, CDCl3): l 173.89, 171.86, 72.53, 54.72, 45.69, 45.33,
38.12, 31.72, 19.70, 18.46, MS (FAB): 243 [M+1]+, mp:
201–203°C. Anal. calcd for C11H22N4O2: C, 54.52; H, 9.15;
N, 23.12. Found: C, 53.8; H, 9.2; N, 23.3%. Compound 4:
[h]2D0=22.32 (c=1.0, MeOH), 1H NMR (300 MHz,
CDCl3): l=7.63 (br, 1H, CONH), 7.26 (br, 1H, CONH),
2.98–3.71 (m, 6H, NCH2CO, 2CONHCH2), 2.60–2.97 (m,
5H, 2NHCH2, NHCHCO), 1.72–1.78 (m, 3H, 2NH,
CH2CHCH3) 1.08–1.60 (m, 2H, CHCH2CH3), 0.89–0.97
(m, 6H, 2CH3), 13C NMR (75 MHz, CDCl3): l 174.11,
172.14, 71.43, 54.72, 45.45, 45.07, 38.37, 37.91, 37.88,
25.28, 15.91, 11.56: MS (FAB): 257 [M+1]+, mp: 213–
215°C. Anal. calcd for C12H24N4O2: C, 56.22; H, 9.44; N,
21.86. Found: C, 56.0; H, 9.1; N, 21.4%.
In conclusion, the chiral dioxocyclen 2 is the best CSA
for the four substrates we chose. The phenyl group
plays an important role in chiral recognition. We have
used an efficient and simple synthetic methodology for
novel chiral dioxocyclen synthesis and we have intro-
duced this kind of chiral dioxocyclen into molecular
recognition research as CSAs. All chiral dioxocyclens
are amphiphiles, and they may be further functional-
ized, providing opportunities for the design and investi-
gation of more effective chiral macrocyclic hosts that
may be used for chiral recognition, enantiomeric sepa-
ration and enzyme models.
Acknowledgements
This work was supported by the Hubei Province Sci-
ence Foundation, and by a grant from The Education
Ministry of China for the visiting scholar to the Key
Laboratory of Bioorganic Chemistry and Molecular
Engineering.
References
5. Xue, P.; Fu, E.; Wang, G.; Wu, C. Synth. Commun. 1999,
29, 1585.
6. Pirkle, W. H.; Pochapsky, T. C. Chem. Rev. 1989, 89,
347–362.
1. (a) Zhang, X. X.; Bradshaw, J. S.; Izatt, R. M. Chem. Rev.
1997, 97, 3313–3361; (b) Webb, T. H.; Wicox, C. S. Chem.
Soc. Rev. 1993, 22, 383–395; (c) Naemura, K.; Tobe, Y.;
Kaneda, T. Coord. Chem. Rev. 1996, 148, 199–219; (d)