4
D. Pál et al. / Tetrahedron: Asymmetry xxx (2016) xxx–xxx
be observed, which was the highest among the set. These results
suggest that the enantiomeric differentiation is based on very sub-
tle effects.19 One of these effects is the type of the receptor units
(urea or thiourea) in the anion sensors, which significantly influ-
enced the recognition ability toward the enantiomers of Boc-Phg.
The structure of the guest molecules also affected the enantioselec-
tivity: the presence of a hydroxyl group (in Man) instead of the pro-
tected amino group (in Boc-Phg) and the presence of a benzyl
(aralkyl) group (in Boc-Phe) or a methyl (alkyl) group (in Boc-Ala)
instead of the phenyl group (in Boc-Phg) decreased the recognition
ability of receptor (S,S)-9. This preference of receptor (S,S)-9 is sim-
ilar to that of the reported bis(thiourea) derivative 5 containing glu-
copyranosyl groups, in which case the highest enantioselectivity
was also observed for (R)-Boc-Phg over its (S)-isomer
Avance spectrometer. 1H (300 MHz) and 13C (75.5 MHz) NMR spec-
tra were obtained on a Bruker 300 Avance spectrometer. The sig-
nals of NH protons in the 1H NMR spectra were helped to
identify by shaking the NMR samples with D2O. Mass spectra were
recorded on an Agilent-6120 Single Quadrupole LC/MS instrument
using ESI method. Elemental analyses were performed in the
Microanalytical Laboratory of the Department of Organic Chem-
istry, Institute for Chemistry, L. Eötvös University, Budapest,
Hungary.
UV–vis spectra were taken on a Unicam UV4-100 spectropho-
tometer. Quartz cuvettes with path length of 1 cm were used. Flu-
orescence spectra were recorded on a Perkin–Elmer LS 50B
luminescent spectrometer. Emission spectra were corrected by
the spectrometer software. Quartz cuvettes with path length of
1 cm were used. Fluorescence quantum yields were determined
relative to quinine sulfate (Uf = 0.53 in 0.1 M H2SO4).38 Stability
constants of the complexes were determined by global nonlinear
regression analysis using SPECFIT/32TM software.
The enantiomers of Man and Boc-protected amino acids were
purchased from Sigma–Aldrich Corporation. The tetrabutylammo-
nium salts of the anions were prepared by adding 1 equiv of car-
boxylic acid to 1 equiv of Bu4NOH dissolved in MeOH. After
evaporating MeOH, the salts were dried under reduced pressure
over P2O5. During the fluorescence titrations, the concentrations
of the solutions of receptors (S,S)-6–(S,S)-9 were 2, 3 and 5 lM,
and the concentrations of the titrant solutions of chiral carboxy-
lates were 0.2, 0.5, 1 and 10 mM.
(D
logK = 0.22) among the same chiral carboxylates.47 Another
trend can be observed: the stability constants of complexes with
the enantiomers of Man are lower than those with the enantiomers
of amino acid derivatives (Boc-Phg, Boc-Phe, Boc-Ala) in the cases of
5,5-dioxophenothiazine-based anion sensors 5, (S,S)-8 and (S,S)-9.
The complexation properties of receptors (S,S)-6 and (S,S)-7
toward the chiral carboxylates were also examined. Sensor mole-
cules (S,S)-6 and (S,S)-7 showed similar, but considerably smaller
absorption and fluorescence spectral changes compared to those
of receptors (S,S)-8 and (S,S)-9. These slight changes did not allow
the accurate determination of the stability constants of complexes,
therefore, the enantiomeric recognition abilities of receptors (S,S)-
6 and (S,S)-7 could not be evaluated.
4.2. General procedure for the synthesis of receptors (S,S)-6–
(S,S)-9
3. Conclusion
We synthesized four 5,5-dioxophenothiazine derivatives as
potential enantioselective anion sensors, and their absorption
and fluorescence behaviour in the presence of the enantiomers of
To a stirred solution of diamine 1045 (200 mg, 0.535 mmol) in
pyridine (3 mL) was added the solution of the appropriate (S)-1-
arylethyl isocyanate (1.12 mmol, R = phenyl: 165 mg, R = 1-naph-
thyl: 222 mg) or (S)-1-arylethyl isothiocyanate (1.61 mmol,
R = phenyl: 262 mg, R = 1-naphthyl: 342 mg) in pyridine (2 mL)
under Ar at rt. The mixture was stirred at rt for 1 h (X = O) or 2 days
(X = S) (see Scheme 1). After the reaction was completed, the reac-
tion mixture was poured into a water–ice mixture, and acidified to
pH 2 using concentrated hydrochloric acid. The precipitate was fil-
tered off, washed with water, and the crude product was purified
as described below for each compound.
tetrabutylammonium salts of a-hydroxy and N-protected a-amino
acids were studied. Receptors (S,S)-6 and (S,S)-7 containing phenyl
groups at their stereogenic centres gave considerably small absorp-
tion and fluorescence spectral changes upon addition of the car-
boxylates, which did not allow the accurate determination of the
stability constants of these complexes. However, derivatives (S,
S)-8 and (S,S)-9 containing 1-naphthyl groups at their stereogenic
centres showed larger spectral changes in the presence of the chi-
ral carboxylates, and the enantiomeric recognition abilities were
evaluated based on the fluorescence spectral changes. The highest
enantioselectivity, which is a moderate one, could be observed in
the case of receptor (S,S)-9 and the enantiomers of Boc-Phg.
4.2.1. 1,10-(3,7-Di-tert-butyl-5,5-dioxo-5,10-dihydro-5k6-
phenothiazine-1,9-diyl)bis{3-[(1S)-1-phenylethyl]urea} (S,S)-6
The crude product was triturated with ethanol to give receptor
(S,S)-6 (203 mg, 57%) as off-white crystals. Mp: 216–217 °C; Rf:
4. Experimental
4.1. General
27
0.46 (silica gel TLC, MeOH–CH2Cl2 1:20); [
a
]
27 = +65.2, [
a
]
=
D
578
27
27
+69.5, [
3372, 3085, 3063, 3030, 2965, 2907, 2871, 1665, 1609, 1547,
1495, 1453, 1365, 1240, 1137, 901, 875, 760, 701, 600, 547 cmꢀ1
a]546 = +81.3, [a]436 = +154 (c 1.00, DMF); IR (KBr) mmax
;
Starting materials were purchased from Sigma–Aldrich Corpo-
ration unless otherwise noted. (S)-1-(1-Naphthyl)ethyl isothio-
cyanate was obtained from Alfa Aesar. Silica gel 60 F254 (Merck)
plates were used for TLC. Silica gel 60 F254 (Merck) plates (1 mm)
were used for PLC. Silica gel 60 (70–230 mesh, Merck) was used
for column chromatography. Ratios of solvents for the eluents
are given in volumes (mL/mL). Solvents were dried and purified
according to well established methods.52 Evaporations were car-
ried out under reduced pressure.
1H NMR (500 MHz, DMSO-d6) d 1.30 (s, 18H), 1.39 (d, J = 6 Hz,
6H), 4.82–4.97 (m, 2H), 7.18–7.47 (m, 10H + 2H, NH), 7.51 (s,
2H), 7.98 (s, 2H), 9.08 (br s, 2H, NH), 9.20 (br s, 1H, NH); 13C
NMR (75.5 MHz, DMSO-d6) d 23.55, 30.86, 34.36, 49.08, 111.68,
121.34, 123.88, 125.73, 126.61, 127.64, 128.27, 128.40, 144.25,
145.09, 154.80; MS calcd for C38H45N5O4S: 667.3, found (M+H)+:
668.3; Anal. calcd for C38H45N5O4S: C 68.34, H 6.79, N 10.49, S
4.80, found: C 68.11, H 6.91, N 10.24, S 4.53.
Melting points were taken on a Boetius micro-melting point
apparatus and are uncorrected. Optical rotations were taken on a
Perkin–Elmer 241 polarimeter, which was calibrated by measuring
the optical rotations of both enantiomers of menthol. IR spectra
were recorded on
(500 MHz) NMR spectra were obtained on a Bruker DRX-500
4.2.2. 1,10-(3,7-Di-tert-butyl-5,5-dioxo-5,10-dihydro-5k6-
phenothiazine-1,9-diyl)bis{3-[(1S)-1-phenylethyl]thiourea}
(S,S)-7
The crude product was purified by preparative layer chro-
matography using 1:20 MeOH–CH2Cl2 as an eluent to give receptor
(S,S)-7 (220 mg, 59%) as yellow crystals. Mp: 155–156 °C; Rf: 0.85
a
Bruker Alpha-T FT-IR spectrometer. 1H