A new series of Cs+, K+ and Na+ chelators: Synthesis, kinetics, thermodynamics and modeling

Article abstract of DOI:10.1016/j.ica.2012.08.009

The synthesis of two molecules, B1 and B2, based on elements of norbadione A, the natural Cs⁺ chelator in mushrooms, associated, in the case of B2, with an 18-crown-6 ether is reported. Thermodynamic and kinetic analyses performed in water, ethanol and ethanol/water 9/1 v/v (M1) show in M1 and ethanol that B1 and B2 form stable complexes with Na⁺, K⁺ and Cs⁺. Affinity constants, measured spectrophotometrically in ethanol and M1, by the use of the Specfit program, are in the 10⁵ and 10⁶ range for B1 and B2, respectively. The second-order rate constants are in the 10⁶-10⁷ M^-1 s^-1 range and the first-order rate constants about unity. The ratios of the second-order/first-order rate constants confirm the thermodynamic results in EtOH. The kinetic processes become much too fast to allow runs in M1. Molecular simulations in EtOH imply the existence of two isomers for each of the Cs ⁺/B1 and Cs⁺/B2 complexes. With B1, the more stable one is that in which the two enolates are parallel and mimic the alkali-metal inclusion cavity already envisaged for norbadione A. With B2, two similar structures are extracted, in both of which Cs⁺ is included in the crown ether and capped by the enolate. The affinity of B1 for Cs⁺ is comparable to that of norbadione A, whereas that of B2 is higher. These results are encouraging as they introduce a new series of alkali chelators which can lead to molecules capable of complexing ¹³⁷Cs⁺ for radioactive decontamination.

Full text of DOI:10.1016/j.ica.2012.08.009

Inorganica Chimica Acta 394 (2013) 45–57
Contents lists available at SciVerse ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
A new series of Cs+, K+ and Na+ chelators: Synthesis, kinetics, thermodynamics
and modeling
Alexandre Korovitch a, Antoine Le Roux b, Florent Barbault a, Miryana Hémadi a,
Nguyêt-Thanh Ha-Duong a, Claude Lion a, Alain Wagner b, Jean-Michel El Hage Chahine a,
⇑
a Université Paris Diderot Sorbonne Paris Cité – CNRS, Interfaces, Traitements, Organisation et Dynamique des Systèmes, UMR 7086, Bâtiment Lavoisier, 15 Rue Jean-Antoine de Baïf,
75205 Paris Cedex 13, France
b Laboratoire des Systèmes Chimiques Fonctionnels, UMR 7199, Faculté de Pharmacie, 74 Route du Rhin, BP 24, 67401 Illkirch-Graffenstaden, France
a r t i c l e i n f o
a b s t r a c t
The synthesis of two molecules, B1 and B2, based on elements of norbadione A, the natural Cs+ chelator in
mushrooms, associated, in the case of B2, with an 18-crown-6 ether is reported. Thermodynamic and
kinetic analyses performed in water, ethanol and ethanol/water 9/1 v/v (M1) show in M1 and ethanol
that B1 and B2 form stable complexes with Na+, K+ and Cs+. Affinity constants, measured spectrophoto-
metrically in ethanol and M1, by the use of the SPECFIT program, are in the 105 and 106 range for B1 and B2,
respectively. The second-order rate constants are in the 106–107 Mꢀ1 sꢀ1 range and the first-order rate
constants about unity. The ratios of the second-order/first-order rate constants confirm the thermody-
namic results in EtOH. The kinetic processes become much too fast to allow runs in M1. Molecular sim-
ulations in EtOH imply the existence of two isomers for each of the Cs+/B1 and Cs+/B2 complexes. With
B1, the more stable one is that in which the two enolates are parallel and mimic the alkali-metal inclusion
cavity already envisaged for norbadione A. With B2, two similar structures are extracted, in both of which
Cs+ is included in the crown ether and capped by the enolate. The affinity of B1 for Cs+ is comparable to
that of norbadione A, whereas that of B2 is higher. These results are encouraging as they introduce a new
series of alkali chelators which can lead to molecules capable of complexing 137Cs+ for radioactive
decontamination.
Article history:
Received 19 June 2012
Accepted 5 August 2012
Available online 29 August 2012
Keywords:
Radioactive decontamination
Norbadione
Relaxation kinetics
Complex formation
Stopped-flow
Mechanism
Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction
related to a family of mushroom pigments, the pulvinic acids
(Fig. 1) [10,14].
Alkali metals, such as sodium and potassium, are absolutely
essential for life, whereas others, such as cesium or lithium, are
not required for biological processes. However, despite this fact,
up to 1.5 mg of cesium is present in the human body [1]. Its stable
isotope, 133Cs is considered innocuous or slightly toxic because of
it interference with sodium potassium pump [2,3]. Nonetheless,
38 other radioactive isotopes, such as 135Cs and 137Cs, are generated
during nuclear reactions. The most dangerous to health is 137Cs with
a half-life of 30 years. 137Cs is produced in nuclear power plants by
the chain decay of 235U to 137Te and then 137Cs [4]. During the Cher-
nobyl disaster in 1986 and very recently during that of Fukushima,
large amounts of 137Cs were released [5–8]. It should also be noted
that more than 20 million people live within a range of 30 km of a
nuclear power plant. This renders 137Cs accumulation a potential
major health problem [7–9]. In Europe, after Chernobyl, 137Cs was
partly accumulated in norbadione A (NbA), the essential pigment
of the bay boletus mushroom [10–13]. NbA is a naphtholactone
Complex formations of alkali metals are considered to be extre-
mely fast processes, which are practically diffusion-controlled
[15–17]. However, with some sterically hindered and capped
calix[4-8]arenes, or with NbA, they can be slow to very slow
[18–20]. Most of the other investigations concerning these com-
plexes were performed by 1H, 13C and/or 133Cs NMR. They always
indicated fast kinetic processes occurring in the sub-second time
range [21]. These are also assumed to be host–guest processes that
occur with crown ethers and with calixarenes [22–26]. On the
other hand, apart from NbA, alkali metal complexes with classical
chelators, that do not have an inclusion cavity, are quite rare
[10,11,16,18,27]. Furthermore, the elementary tetronic and pulvi-
nic acid building blocks that constitute NbA do not complex alkali
metals (Fig. 1) [18]. Complex formation occurs because of the par-
ticular structure of NbA [18,28]. The aim of this work is to mimic
NbA in complex formation with alkali metals by synthesizing a
structure based on tetronic acid, to associate a tetronic acid with
crown ether, and to analyze by means of chemical relaxation, fast
kinetics and molecular simulation the mechanisms involved in
complex formation with Na+, K+ and Cs+.
⇑
Corresponding author. Tel.: +33 157277238; fax: +33 157277263.
E-mail address: chahine@univ-paris-diderot.fr (J.-M. El Hage Chahine).
0020-1693/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ica.2012.08.009

46
A. Korovitch et al. / Inorganica Chimica Acta 394 (2013) 45–57
O
O
O
HO
HO
R1
OH
O
CO2H
O
O
O
O
O
O
HO
HO2C
HO
R2
HO2C
OH
Norbadione A
Pulvinic acid
Tetronic acid
OH
Fig. 1. Molecular structures of norbadione A and its components; pulvinic and tetronic acids.
2.1.1. 4-Hydroxy-3-(4-methoxyphenyl)-furan-2(5H)-one (1)
2. Experimental
Ethyl glycolate (2.9 g, 28 mmol, 1 eq) and potassium tert-butox-
ide (6.3 g, 56 mmol, 2 eq) were added to a solution of methyl 4-
methoxyphenylacetate (5 g, 28 mmol, 1 eq) in THF (180 mL) and
the mixture was refluxed overnight. After cooling to room temper-
ature, concentrated HCl was added until acidic and the mixture ex-
tracted with ethyl acetate (3 ꢁ 100 mL). The combined organic
phases were dried over magnesium sulfate and concentrated in va-
cuo. Crystals formed during the evaporation process were collected
and washed with diethyl ether. The filtrate was concentrated, and
the operation repeated until the end of precipitation. Yield: 78%;
1H NMR (300 MHz, DMSO-d6): d = 3.75 (s, 3H), 4.74 (s, 3H), 6.94
(d, 2H, J = 9.0 Hz), 7.84 (d, 2H, J = 9.0 Hz), 12.58 (s, 1H); 13C NMR
(DMSO-d6): d = 55.0, 66.0, 97.3, 113.5, 122.9, 127.6, 157.6, 173.1,
173.3; MS (ESI) m/z 207.1 ([M+H]+).
2.1. Synthesis
Commercial reagents were used without further purification.
Anhydrous tetrahydrofuran (THF) was obtained by distillation over
sodium and benzophenone. Analytical thin layer chromatography
(TLC) was performed using plates cut from glass sheets (silica gel
60F-254 from Merck). Detection was performed under a 254 or
365 nm UV light and by immersion in an ethanol solution of cer-
ium sulfate, followed by heating. Compounds were purified on a
Silica gel 60 chromatography column. IR spectra were recorded
in CH2Cl2 solutions or in the solid state on a diamond plate on a
Nicolet 380 FT-IR spectrometer from Thermo Electron Corporation.
1H and 13C NMR spectra were recorded at 23 °C on a Bruker
400 MHz, Bruker 300 MHz or Bruker 200 MHz spectrometers. Re-
corded shifts are reported in parts per million (d) and calibrated
with residual undeuterated solvent. Data were represented as fol-
lows: Chemical shift, mutiplicity (s = singlet, d = doublet, t = triplet,
m = multiplet), integration and coupling constant (J, Hz). High
resolution mass spectra (HRMS) were obtained using an Agilent
Q-TOF (time of flight) 6520 instrument and low resolution mass
spectra using an Agilent MSD 1200 SL. Electrospray ionization/
atmospheric pressure chemical ionization (ESI/APCI) was per-
formed on an Agilent HPLC1200 SL.
2.1.2. 4-Methoxy-3-(4-methoxyphenyl)-furan-2(5H)-one (2)
Potassium carbonate (2.8 g, 20 mmol, 1 eq) and dimethyl sulfate
(2.5 g, 20 mmol, 1 eq) were added to a suspension of 1 (4.1 g,
20 mmol, 1 eq) in acetone (85 mL), and the suspension refluxed
for 4 h. After cooling to room temperature, the solid was filtered
off over Celite. The filtrate was concentrated to give 2 as a white so-
lid and which was used for the next step without further purifica-
tion. Yield: 99%; 1H NMR (200 MHz, CDCl3): d = 3.76 (s, 3H), 3.86
(s, 3H), 4.72 (s, 2H), 6.86 (d, 2H, J = 8.8 Hz), 7.77 (d, 2H, J = 8.8 Hz);
O
O
O
O
O
O
O
O
O
O
HO
O
O
O
O
OH
O
O
O
O
1
2
3
O
O
O
O
O
O
O
O
H2N
NH2
NH
OH
N
H
OH
O
OH
O
O
O
O
O
HO
O
O
6
4
5
O
B1

A. Korovitch et al. / Inorganica Chimica Acta 394 (2013) 45–57
47
13C NMR (CDCl3): d = 55.3, 58.0, 64.6, 102.1, 113.7, 121.9, 128.9,
2.1.5. 3-(4-Methoxyphenyl)-6-methylfuro[3,2-b]furan-2,5-dione (5)
[29,30]
158.9, 172.3, 173.1; MS (ESI) m/z 221.1 ([M+H]+).
4 (1 g, 3.6 mmol) was suspended in acetic anhydride (17 mL),
heated to reflux for 1 h and then cooled to room temperature. Dur-
ing cooling, yellow crystals appeared which were collected by fil-
tration and washed with pentane to give 5. Yield: 94%; 1H NMR
(300 MHz, CDCl3): d = 2.06 (s, 3H); 3.86 (s, 3H); 6.98 (d, 2H
J = 9.1 Hz); 7.94 (d, 2H J = 9.1 Hz,); 2.06 (s, 3H); 3.86 (s, 3H); 6.98
(d, 2H, J = 9.1 Hz,); 7.94 (d, 2H, J = 9.1 Hz,); 13C NMR (CDCl3):
d = 7.64, 55.5, 98.8, 101.1, 114.8, 119.0, 129.9, 155.6, 159.3, 161.0,
166.1, 168.8; MS [ESI + APCI] m/z 258.0 (M+).
2.1.3. Methyl 2-(3-methoxy-4-(4-methoxyphenyl)-5-oxofuran-2(5H)-
ylidene)-propanoate (3)
A solution of diisopropylamine (4.2 mL, 30 mmol, 1.5 eq) in
anhydrous THF (125 mL) was cooled to -20 °C, and then nBuLi
(1.6 M solution, 18.7 mL, 30 mmol, 1.5 eq) was added dropwise.
The mixture was stirred for 30 min and afterwards cooled to -
78 °C. A solution of 2 (4.4 g, 20 mmol, 1 eq) in anhydrous THF
(65 mL) was added dropwise and the mixture was stirred for
30 min before dropwise addition of methyl pyruvate (6.1 mL,
60 mmol, 3 eq). The mixture was stirred for 30 min at ꢀ78 °C
and then warmed to room temperature. A saturated solution of
ammonium chloride was added, the aqueous layer extracted with
ethyl acetate (3 ꢁ 60 mL), and the combined organic phases dried
over magnesium sulfate and concentrated in vacuo. The resulting
oil was diluted in dichloromethane (DCM, 200 mL) and cooled to
0 °C. Triethylamine (16.7 mL, 120 mmol, 6 eq), 4-dimethylamino-
pyridine (240 mg, 2 mmol, 0.1 eq) and trifluoroacetic anhydride
(1.26 g, 60 mmol, 3 eq) were added dropwise, successively. The
mixture was stirred overnight at room temperature and then
acidified by 1 M HCl. The aqueous phase was extracted with
DCM, (3 ꢁ 50 mL), and the combined organic phases dried over
magnesium sulfate and concentrated in vacuo. The crude was then
purified by chromatography (cyclohexane/ethyl acetate 90/10 to
50/50) to give 3 as a mixture of E and Z (E/Z = 70/30). Yield:
88%; 1H NMR (300 MHz, CDCl3): isomer E: d = 2.15 (s, 3H), 3.72
(s, 3H), 3.82 (s, 3H), 3.83 (s, 3H), 6.92–6.95 (m, 2H), 7.41–7.43
(m, 2H); isomer Z: d = 2.27 (s, 3H), 3.80 (s, 3H), 3.83 (s, 3H),
3.86 (s, 3H), 6.93–6.96 (m, 2H), 7.40–7.43 (m, 2H); 13C NMR
(CDCl3): isomer E: d = 15.2, 52.8, 55.6, 61.1, 109.0, 113.4, 114.2,
120.8, 131.4, 143.3, 160.3, 161.3, 168.2, 168.7; MS (ESI) m/z
305.1 ([M+H]+).
2.1.6. (2E,20E)-N,N0-(Propane-1,3-diyl)bis(2-(3-hydroxy-4-(4-
methoxyphenyl)-5-oxofuran-2(5H)-ylidene) propanamide) (6) [31]
A suspension of 5 (200 mg, 0.77 mmol, 1 eq) in anhydrous THF
(15 mL) was cooled to ꢀ35 °C, before adding a solution of tetra-N-
butylammonium fluoride (TBAF) (1.0 M in THF, 1.55 mL, 1.55 mmol,
2 eq). The mixture was stirred for 15 min and then cooled to ꢀ78 °C.
A solution of 1,3-diaminopropane (33 lL, 0.39 mmol, 0.5 eq) in
5 mL THF was added dropwise and the mixture stirred for 15 min.
At room temperature, ethyl acetate was added and the organic
phase washed with 1 M HCl and brine, dried over MgSO4, filtered
and concentrated. The yellow residue was washed with diethyl-
ether and compound 6 collected as a yellow solid. Yield: 57%. 1H
NMR (pyridine-d5): d = 1.97 (t, 2H, J = 6.8 Hz), 2.20 (s, 6H), 3.56 (t,
4H, J = 6.8 Hz), 3.67 (s, 6H), 7.10 (d, 4H, J = 8.9 Hz), 8.57 (d, 4H,
J = 8.9 Hz); 13C NMR: d = 14.4, 29.5, 38.8, 55.6, 101.9, 113.6, 114.7,
124.6, 129.4, 153.1, 159.6, 163.6, 168.3, 170.2: HRMS (ESI): Anal.
Calc. for C31H31N2O10 ([M+H]+): 591.19732. Found: 591.1991%; IR:
(cmꢀ1): 464, 499, 515, 555, 582, 634, 655, 698, 743, 792, 811, 833,
912, 1027, 1088, 1117, 1151, 1177, 1247, 1295, 1417, 1455, 1485,
1507, 1557, 1602, 1619, 1660, 1739, 2836, 2932, 3340. 3,6-Bis(4-
methoxyphenyl)-furo[3,2-b]furan-2,5-dione (7) [32]. Sodium
(3.6 g, 156 mmol, 2 eq) was dissolved in cooled absolute ethanol
(50 mL). Diethyl oxalate (10.8 mL, 80 mmol, 1 eq) and 4-methoxy-
phenylacetonitrile (22.6 mL, 156 mmol, 2 eq) were added after
complete consumption of the sodium. The mixture was stirred at
70 °C for 90 min, and cooled to room temperature. Water (8 mL)
were added, followed by acidification with acetic acid down to pH
5. This led to the precipitation of 9.55 g of an orange product. This
was dissolved in 20 mL of acetic acid and 10 mL of water before add-
ing dropwise 8.5 mL of concentrated sulfuric acid. A red solid ap-
peared, and the mixture was heated at reflux for 30 min after
which it was cooled in an ice bath. The solid was collected by filtra-
tion, suspended in 15 mL of acetic anhydride and heated at reflux for
30 min. The bis-lactone 7 was collected at room temperature by fil-
tration and washed with heptane. Yield: 10%. 1H NMR (400 MHz,
CDCl3): d = 3.89 (s, 6H), 7.01 (d, 4H, J = 8.6 Hz), 8.01 (d, 2H,
J = 8.6 Hz); 13C NMR (CDCl3): d = 54.8, 96.4, 100.3, 115.3, 117.6,
130.0, 134.4, 159.5; IR (cmꢀ1): 595, 869, 1019, 1157, 1257, 1354,
1506, 1600, 1657, 1782, 1909.
2.1.4. 2-(3-Hydroxy-4-(4-methoxyphenyl)-5-oxofuran-2(5H)-
ylidene)-propanoic acid (4)
Magnesium bromide (184 mg, 1 mmol, 1 eq) was added to a
solution of 3 (304 mg, 1 mmol, 1 eq) in dimethylformamide (DMF)
(10 mL) and the mixture stirred at 120 °C until no more starting
materiel remained. The solvent was removed under reduced pres-
sure, and the crude diluted in water. The organic phase was washed
with dichloromethane (3 ꢁ 15 mL), and acidified with 1 M HCl. The
aqueous phase was extracted with ethylacetate (3 ꢁ 15 mL). The
combined organic phases were washed with brine, dried over so-
dium sulfate and concentrated to give 4. Yield: 78%; 1H NMR
(200 MHz, MeOD): d = 2.11 (s, 3H), 3.81 (s, 3H), 6.92 (d, 2H,
J = 9.0 Hz,), 8.02 (d, 2H, J = 9.0 Hz); 13C NMR: d = 14.5, 55.6, 103.9,
114.1, 114.6, 123.4, 129.9, 154.6, 160.6, 160.9, 168.3, 174.7; MS
(ESI) m/z 231.0 ([MꢀCO2]ꢀ).
O
O
O
O
O
O
O
H2N
O
O
O
O
O
O
O
CN
O
O
O
O
O
O
O
+
OH
O
O
7
N
H
O
O
O
O
O
8
B2

48
A. Korovitch et al. / Inorganica Chimica Acta 394 (2013) 45–57
2.1.7. (E)-N-(2-(1,4,7,10,13-Pentaoxa-16-azacyclooctadecan-16-
yl)ethyl)-2-(3-hydroxy-4-(4-methoxyphenyl)-5-oxofuran-2(5H)-
ylidene)-2-(4-methoxyphenyl)acetamide (8)
2.3. Spectrophotometric measurements
Absorption measurements were performed at 25 0.1 °C on a
Cary 4000 spectrophotometer equipped with Peltier thermostated
cell-carriers. Fluorimetric measurements were performed at
25 0.5 °C on an Amino-Bowman series 2 luminescence spectrom-
eter equipped with a thermostated cell carrier. The fluorescence
intensity was corrected for the inner filter effect of the absorption
of B1 and B2.
2-Aminomethyl-18-crown-6 (250 mg, 0.85 mmol, 1.1 eq) was
added to a suspension of compound 7 (270 mg, 0.77 mmol) in
15 mL of DCM. The mixture was stirred for 1 h at room tempera-
ture and concentrated. The crude was purified by chromatography
(DCM/MeOH 98/2) to give 8 as an orange solid. Yield: 93%. 1H NMR
(CDCl3): d = 3.48–3.72 (m, 25H), 3.82 (s, 3H), 3.83 (s, 3H), 6.76 (s,
1H), 6.94 (d, 2H, J = 9.2 Hz), 6.97 (d, 2H, J = 8.8 Hz), 7.22 (d, 2H,
J = 8.8 Hz), 8.12 (d, 2H, J = 8.8 Hz); 13C NMR: d = 42.4, 55.4, 55.4,
69.5, 70.7–71.0 (m), 72.2, 103.1, 113.9, 114.7, 117.1, 122.6, 124.1,
129.0, 131.6, 152.7, 159.1, 160.3, 160.9, 167.4, 168.9; HRMS
(ESI): Anal. Calc. for C33H42NO12 ([M+H]+): 644.27015. Found:
644.27263%; IR: (cmꢀ1): 463, 486, 522, 549, 582, 634, 650, 699,
752, 765, 828, 880, 912, 956, 1024, 1110, 1182, 1247, 1292,
1417, 1444, 1463, 1506, 1556, 1601, 1766, 1860, 2863, 2895,
2942, 3373, 3398.
2.4. pH Measurements
The pHs were measured at 25 0.5 °C with a Jenco pH-meter. In
M1 (ethanol/water 9/1) and EtOH, the pH values were corrected
according to published procedures [33].
2.5. Protodissociation and affinity constants measurements
Protodissociations and affinity constants were determined
spectrophotometrically by the use of the Global Analysis program
2.2. Stock solutions
SPECFIT32. SPECFIT32 is a multivariate data analysis program for data
The ethanol-to-water ratio is given in volume. The solubilities
of B1 and B2 are very poor in water. Therefore, both molecules
were first dissolved in pure ethanol (5 ꢁ 10ꢀ4 M) and then diluted
in the final media to [1–100] ꢁ 10ꢀ7 M and [5–100] ꢁ 10ꢀ7 M,
respectively. All other products were of the purest possible grade
(Sigma, Merck, Acros, or Aldrich). Ethanol was Merck spectroscopy
grade, and water was demineralized and doubly distilled.
sets that are obtained from multiwavelength spectrophotometric
measurements. The program utilizes a specially adapted version
of the Levenberg–Marquardt method. This procedure returns opti-
mized model parameters, their standard errors, and the predicted
spectra of the unknown colored species [34]. For protodissocia-
tions, the spectra were measured at different pH values in M1.
Neutralization and acidification were achieved with NaOH
Fig. 2. Molecular structures of B1 and B2. The numbers represent the atom ID used in Tables 3–6.

A. Korovitch et al. / Inorganica Chimica Acta 394 (2013) 45–57
49
(61 ꢁ 10ꢀ5 M) and HCl (ꢂ1 ꢁ 10ꢀ3 M). The affinity constants were
checked when possible by a variant of the Benesi and Hildebrand
method. Unless specified otherwise, they were measured in etha-
nol in the absence of alkali cations other than Cs+, Na+ or K+. In
M1, they were measured in media buffered with 5 ꢁ 10ꢀ4 M of bo-
ric acid at pH 9.3.
In all simulations, potential, kinetic and total energy were main-
tained constant. Furthermore, the restraint did not cause any steric
hindrance. Cation interaction with the ligands occurred in less than
10 ps and the complexes did not dissociate after release of the re-
straint. This confirms the recognition potential of B1 and B2 for the
three alkali cations.
2.6. T-jump kinetics
3. Results
Experiments were run on a modified Joule effect Messanlagen
und Studien absorption and fluorescence emission T-jump spectro-
photometer. The apparatus was equipped with a 200 W Xe/Hg
light source, a Jarrel Ash monochromator, and a thermostated
cell-holder maintained at 20 1 °C [18]. The temperature jump of
3.1. Synthesis
For the synthesis of B1 and B2, we expected to make use of the
particular reactivity of bis-lactone type compounds. These com-
pounds are obtained by dehydration of pulvinic acids (Scheme 1)
[32]. In the presence of a nucleophile, and especially amines, one
of the lactone functions will react by ring-opening, giving thus
the two amides without any control of the regioselectivity (Scheme
2). In the case of bis-lactone 1, we recently reported that the addi-
tion of 2 equivalents of TBAF allows controlling the regioselectivity,
giving compound 2 as the major product (Scheme 3) [31].
For compound B1, initially one equivalent of the bis-lactone 1
reacted with 0.5 equivalent of 1,3-diaminopropane in DCM. How-
ever, this led to the precipitation of intermediate 4, thus preventing
the second attack of the free amine of 4. The addition of DMSO to
solubilise the mixture allowed this second reaction, but led to the
formation of three isomers (Scheme 4), which were inseparable by
chromatography on silica gel. The reaction was then performed in
the presence of TBAF in THF, which prevented the formation of iso-
mers 5 and 6. B1 was isolated in 57% yield.
5–6 °C was achieved by discharging an 0.05 lF condenser charged
at 10 kV in solutions containing 0.1 M NH4Cl.
2.7. Stopped-flow kinetics
Experiments were realized on a Hi-Tech Scientific SF61DX2
stopped-flow spectrophotometer equipped with a Xe/Hg light
source and a thermostated bath at 25 1 °C. Solutions of B1 and
B2 in water, M1 and EtOH (1 ꢁ 10ꢀ7 M < c0 < 5 ꢁ 10ꢀ7 M) were
mixed with solutions of Cs+ (0 < c1 < 1 ꢁ 10ꢀ3 M). All signals were
accumulated at least 10 times. The excitation wavelength was set
to 365 nm, which is one of the emission peaks of the light source.
Detection was set to kem P 400 nm.
2.8. Data analysis
We then focused our attention on the synthesis of derivatives
bearing a crown ether, choosing the commercially available amine
7 for an initial attempt. The reaction with bis-lactone 1 in the pres-
ence of TBAF in THF was not regioselective, giving the two isomers,
which again were inseparable by chromatography on silica gel
(Scheme 5).
In order to avoid regioselectivity problems, we selected bis-lac-
tone 10, in which the two lactone moieties are identical. The reac-
tion of amine 7 on 10 in DCM led to compound B2 in 93% yield
(Scheme 6).
The kinetic data were analyzed by linear and nonlinear least-
squares regressions, and all uncertainties are twice the standard
deviations. Brouillard established the conditions in which chemical
relaxation approximations can apply with the use of high perturba-
tions, such as concentration-jumps [35]. Therefore, all the experi-
mental conditions were set so as to allow the use of the methods
and techniques of chemical relaxation. This led to the fact that
all the observed kinetic processes were pure exponentials and as-
sumed as relaxation modes [18,19,35–38].
2.9. Molecular analysis simulation
3.2. Thermodynamics
SYBYL software was used for the graphical construction of B1
and B2 (Fig. 2) [39]. All calculations were performed with the AM-
BER v11 software package [40]. B2 contains an asymmetric carbon
which led us to consider the two possible B2-R and B2-S isomers in
our computational approach. The GAFF (general AMBER force field)
[41] set of parameters was used with the atomic partial charges
produced by AM1–BCC computations [42,43]. B1 and B2 were first
optimized with 5000 steps of minimization in an ethanol implicit
solvent with a dielectric constant of 24.4. A single cation (Na+, K+
or Cs+) ion, provided by the AMBER software [44], was afterwards
randomly added. Molecular dynamics simulations at 300 K were
carried out for 10 ns in ethanol implicit solvent, with each of the
three cations for B1 and B2. Computational details are reported
elsewhere [45]. To initiate the systems and facilitate the interac-
tion between the cations and the ligands, weak restraints
(10 kcal/mol max) were imposed for 2 ns. These were defined as
a square-bottom potential well with parabolic sides up to an estab-
lished distance, and linear afterwards. The distances between each
ion and the oxygen atoms carrying the negative charges of B1 and
the enolate of B2 were chosen as indicators for complex formation.
This is based on the fact that electrostatic effects are long-range
interactions. After 2 ns, all restraints were removed. Molecular
imaging and structural analysis were then performed for the
remaining 8 ns with the VMD software [46].
The results reported here are obtained in three different media:
(1) water, (2) ethanol/water 9/1 in volume (M1) and (3) ethanol.
We began by analyzing the protodissociations of the acid–base
centers involved in B1 and B2, in the absence or presence of Cs+,
Na+ or K+, in water and M1. B1 and B2 have typical absorption
and fluorescence emission spectra in water, M1 and EtOH. There-
fore, spectrophotometric detection was used in most of our
experiments.
3.2.1. Protodissociation constants
In M1, the absorption spectra of both B1 and B2 vary with pH
(Fig. 3A and B). The analysis of the spectrophotometric data by
the SPECFIT32 program allowed us to determine 2 pKa values (Eqs.
(1) and (2)) for B1 and one pKa value (Eq. (3)) for B2, which, as
for NbA and its subunits, were ascribed to the enol functions (Table
1) [18]
B12ꢀ þ Hþ ꢀ B1Hꢀ
B1Hꢀ þ Hþ ꢀ B1H2
B2ꢀ þ Hþ ꢀ B2H
ð1Þ
ð2Þ
ð3Þ
where B12ꢀ is the fully deprotonated B1 species, B2ꢀ the fully
deprotonated B2 species and K1a = [H+][B12ꢀ]/[B1Hꢀ], K2a = [H+][-
B1Hꢀ]/[B1H2] and K3a = [H+][B2ꢀ]/[B2H].

50
A. Korovitch et al. / Inorganica Chimica Acta 394 (2013) 45–57
O
O
O
O
R2
R2
R1
R1
OH
OH
OH
OH
O
O
OH
or
OH
O
O
O
O
- H2O
+ H2O
- H2O
R2
and
R1
O
O
O
O
O
O
R1
R1
R2
R2
OH
OH
Scheme 1.
O
O
O
O
R3
N
R4
O
O
R2
R1
R2
H
R3
R3
R1
R2
R1
N
N
O
OH
OH
O
O
R4
R4
O
Scheme 2.
O
O
O
O
O
nBu2NH
O
O
N
O
OH
O
O
N
O
O
OH
O
1
2
3
without TBAF :
2 eq TBAF :
58 %
98 %
42 %
2 %
Scheme 3.
O
HO
O
O
O
O
O
O
O
O
HO
O
O
N
0.5 eq
H2
N
NH2
DMSO
O
H
NH2
N
H
O
O
O
O
OH
O
O
N
O
DCM, rt, 12h
O
N
H
inseparable mixture
OH
H
O
O
O
N
H
1
OH
O
O
4
5
B1
0.5 eq
2 eq TBAF
H2
N
NH2
O
HO
O
O
THF, -35°C to -78°C, 1h
O
O
O
57 %
N
H
N
H
OH
O
O
O
6
HO
O
O
O
O
O
N
H
N
H
OH
O
O
B1
Scheme 4.
The same measurements were performed in an excess of Cs+,
3.2.2. Complex formation
Na+ or K+ (not shown) at c1 = 1 ꢁ 10ꢀ2
M
with B1 and
In the 7–9.5 pH range and at fixed pH values, adding Cs+, Na+ or
K+ to a B1 or B2 solutions in an aqueous medium does not lead to
any change in the absorption or emission spectra. This does not im-
ply that the Cs+ complexes are not formed with B1 or B2 in water. It
means that if a complex is formed, it is not detected by the spectro-
photometric techniques.
c1 = 5 ꢁ 10ꢀ5 M with B2. In M1 and in the case of B1 and B2 in
the presence of Cs+, pK1a, pK2a and pK3a are decreased by 0.5, 2.6
and 0.15 (Table 1), respectively. They kept the same values with
Na+ and K+. This implies that complex formation occurs between
the alkali metal and the enolates [47].

A. Korovitch et al. / Inorganica Chimica Acta 394 (2013) 45–57
51
O
OH
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
N
H
2 eq TBAF
OH
H2N
O
O
O
O
O
O
THF, -35°C to -78°C, 1h
O
O
O
O
O
O
N
H
1
O
7
8
9
Scheme 5.
O
O
O
O
O
O
O
O
O
O
OH
H2N
O
O
DCM
O
O
O
O
O
N
O
O
H
O
O
10
7
O
B2
O
Scheme 6.
Table 1
0.4
0.2
0.0
A
Protodissociation constants of B1 and B2 in the absence or presence of Cs+
(c1 = 1 ꢁ 10ꢀ2 M with B1 and 5 ꢁ 10ꢀ5 M with B2) in M1 at 25.0 0.5 °C.
Without Cs+
With Cs+
B1
B2
pK1a
pK2a
pK3a
5.6 0.1
8.00 0.15
4.60 0.05
5.1 0.3
5.4 0.1
4.40 0.05
pH = 1.2
spectra of both ligands, with the metal-free ligands taken as refer-
ences (Fig. 4A and B for Cs+, not shown for K+ and Na+). If we as-
sume that these changes are the results of complex formation
between the alkali cations and B1 or B2, we can use the SPECFIT32
program to determine the stoechiometries and affinities involved
[18,19,34]. In each case, the analysis implies a single complex for-
mation (Eqs. (4) and (5)):
pH = 12.8
300
400
500
λ (nm)
B1 þ Akþ ꢀ B1Akþ
B2 þ Akþ ꢀ B2Akþ
ð4Þ
ð5Þ
0.20
0.15
0.10
0.05
0.00
B
pH = 2.5
with Ak+ = Cs+, Na+ or K+; K1B1 = [B1Ak+]/[B1][Ak+] and K1B2 = [-
B2Ak+]/[B2][Ak+]. In ethanol and in M1, K1B1EtOH, K1B2EtOH, K1B1M1
and K1B2M1 are reported in Table 2 for the three Ak+. These values
were confirmed by a Benesi and Hildebrand analysis [18,19,48].
The addition of variable concentrations (c1) of alkali cations to B1
increases the fluorescence intensity up to a plateau at about c1 ꢂ 1 ꢁ
10ꢀ4, 1 ꢁ 10ꢀ4 and 5 ꢁ 10ꢀ4 M, for Cs+, Na+ and K+, respectively. This
isaccompaniedbya 10 nmblueshift(Fig. 5AforCs+, notshownforK+
and Na+). In M1, the fluorescence intensity increases up to a plateau
for c1 ꢂ 1.5 ꢁ 10ꢀ4, 1 ꢁ 10ꢀ4 and 5 ꢁ 10ꢀ4 M, for Cs+, Na+ and K+,
respectively. This is accompanied by a 5 nm blue shift (Fig. 5B for
Cs+, not shown for K+ and Na+). The analysis of the emission spectra
by the SPECFIT32 program allowed us to determine and confirm the
pH = 8.5
300
400
500
λ (nm)
K1B1EtOH and K1B1M1 values (Table 2). The same experiments were re-
Fig. 3. Evolution of the absorption spectra with pH at 25.0 0.5 °C. (A) B1
(c0 = 5 ꢁ 10ꢀ6 M) in M1, (B) B2 (c0 = 5 ꢁ 10ꢀ6 M) in M1.
peated with B2. The addition of the alkali cations manifested by an
increase in the fluorescence intensities without, however, the blue
shifts observed for B1 (Fig. 6A and B for Cs+, not shown for K+ and
Na+). These results confirm those obtained by differential absorption
detection (Table 2).
In M1 and EtOH, adding Cs+, Na+ or K+ to solutions of B1 and B2
leads to changes in the emission and differential absorption

52
A. Korovitch et al. / Inorganica Chimica Acta 394 (2013) 45–57
in M1 and EtOH are too fast to allow kinetic detection by conven-
A
tional techniques. Indeed, complex formation with alkalis is
assumed to be an extremely fast process, which can be diffusion-
controlled [15,17,49]. Our kinetic investigations were, therefore,
first performed by T-Jump and then by the stopped-flow mixing
technique [18,19]. We did not succeed in obtaining exploitable
results by T-jump (low signal-to-noise ratio and very small signal
variations).
0.03
0.00
c1 = 0
In EtOH, when solutions of B1 or B2 are mixed with solutions
of Cs+, Na+ or K+, a single kinetic process occurs as an exponential
increase in the fluorescence intensity with time (excitation max-
imum, kex = 365 nm; emission kem > 400 nm) (Fig. 7A and B for
Cs+, not shown for K+ and Na+). The experimental reciprocal
relaxation times associated with these processes depend on
[Ak+].
-0.03
-0.06
c1 = 7 x 10-4 M
300
400
500
The reciprocal relaxation time equation associated with Eqs. (4),
(5) is expressed as [18,19,36,37]:
λ (nm)
sꢀ1 ¼ k1ð½Akþꢃ þ ½BꢃÞ þ k
ð6Þ
ꢀ1
with, B = B1 or B2.
-4
Since, under our experimental conditions, [Ak+] ꢄ [B], Eq. (6)
c
= 5 x 10
M
1
0.08
0.04
0.00
-0.04
B
simplifies to Eq. (7):
sꢀ1 ¼ k1c1 þ k
ð7Þ
ꢀ1
with c1, the analytical concentration of the alkali metal salt.
A very good linear least-squares regression of the data was ob-
tained for each of the two ligands with Cs+, Na+ or K+ in EtOH
(Fig. 8A and B for Cs+, not shown for K+ and Na+). From the slopes
and intercepts of the best lines, k1, k and K = k1/k values were
ꢀ1
ꢀ1
determined for B1 and B2 with each of the three alkali cations (Ta-
ble 2). The K values are, within the limits of uncertainty, identical
to those determined spectrophotometrically.
c1 = 0
In M1, mixing solutions of B1 or B2 with solutions of Cs+, Na+ or
K+ leads to very fast processes (<1 ms) resulting in an increase in
the fluorescence emission. Attempts to use the T-jump technique
to investigate these processes were fruitless, because the signal-
to-noise ratios were too low to allow any interpretation. To obtain
further information about the structures that may be involved in
the alkali complexes, modeling of the B1- and/or B2-Cs+ complexes
was performed by molecular dynamics simulations.
300
400
500
λ (nm)
Fig. 4. Differential absorption spectra. (A) B1 in ethanol at different Cs+ concen-
trations with B1 taken as reference, with c0 = 1 ꢁ 10ꢀ5 M and 0 < c1 < 7 ꢁ 10ꢀ4 M at
25.0 0.5 °C, (B) B2 in ethanol at different Cs+ concentrations with B2 taken as
reference, with c0 = 1 ꢁ 10ꢀ5 M and 0 < c1 < 5 ꢁ 10ꢀ4 M at 25.0 0.5 °C.
3.4. Modeling
3.3. Kinetics
Two main conformations of the B1–Ak+ complexes are extracted
from the molecular dynamics trajectory. They correspond to a
‘‘cisoid/transoid’’ of the aromatic moieties (Fig. 9). The ‘‘cisoid’’s
The spectrophotometric variations describing complex forma-
tion upon the addition of alkali cations to solutions of B1 and B2
Table 2
Kinetic and thermodynamic data related to complex formation between Cs+, Na+, K+ and B1 or B2 in EtOH and in M1 at pH 9.3.
a
b
a
Molecule
B1
Metal
Cs+
logKBEtOH
k1 (Mꢀ1 sꢀ1
)
k
(sꢀ1
)
logKBEtOH
4.7 0.3
logKBM1
ꢀ1
4.8 0.1c
5.0 0.3d
6.4 0.1c
6.3 0.1d
5.0 0.1c
5.1 0.2d
6.0 0.1c
5.8 0.2d
(1.2 0.1) ꢁ 105
(2.1 0.1) ꢁ 107
(9.3 0.3) ꢁ 105
(1.4 0.1) ꢁ 107
2
1
4.7 0.1c
4.9 0.2d
6.1 0.1c
6.0 0.2d
4.9 0.1c
5.0 0.2d
5.8 0.1c
5.7 0.2d
B2
B1
B2
11
17
18
4
2
4
6.3 0.2
4.7 0.2
5.9 0.2
Na+
B1
B2
K+
4.3 0.1c
4.3 0.1d
5.5 0.2c
5.6 0.3d
(4.7 0.4) ꢁ 105
(9.1 0.2) ꢁ 106
38
26
1
6
4.1 0.2
5.5 0.2
4.2 0.1c
4.1 0.2d
5.3 0.1c
5.1 0.2d
a
b
c
Is related to the spectrophotometric determination of the K values.
Is related to their kinetic determination.
Is related to determination of the K values using differential absorption.
Is related to their determination using fluorescence emission.
d

A. Korovitch et al. / Inorganica Chimica Acta 394 (2013) 45–57
53
c1 = 5 x 10-4 M
c1 = 1 x 10-4 M
75
A
A
12
60
45
30
15
c1 = 0
8
4
c1 = 0
500
600
700
500
600
700
λ (nm)
λ (nm)
c1 = 1 x 10-3 M
c1 = 1 x 10-4 M
B
B
60
16
40
20
c1 = 0
c1 = 0
8
500
700
600
700
500
600
λ (nm)
λ (nm)
Fig. 6. Emission spectra of solutions of B2 at different Cs+ concentrations for
kex = 391 nm, c0 = 1 ꢁ 10ꢀ6 M at 25 0.5 °C. (A) EtOH with 0 < c1 < 1 ꢁ 10ꢀ4 M, (B)
Fig. 5. Emission spectra of solutions of B1 at different Cs+ concentrations for
kex = 391 nm, c0 = 1 ꢁ 10ꢀ6 M at 25 0.5 °C. (A) EtOH with 0 < c1 < 5 ꢁ 10ꢀ4 M, (B)
M1 with 0 < c1 < 1 ꢁ 10ꢀ4
.
M1 with 0 < c1 < 1 ꢁ 10ꢀ3
.
between the enolate and the positive charge of the cation. Similar
3D structures for the B2R- and B2S-Ak+ are not surprising (Table 5).
In fact, the crown ether linked to the asymmetric carbon is very
flexible. It adopts, though variation of dihedral angle, two struc-
tures in which, the oxygen atoms are equally exposed to provide
the highest electrostatic stabilization. This is confirmed by the in-
ter-atomic distances (Table 6) and by the energies (Table 5). These
show that B2R is less stable by about 3 kcal/mol than B2S. This dif-
ference can be explained by the dihedral energy variation. It is,
moreover, weak which allows the interaction of both enantiomers
with the cations. Furthermore, B2 interacts strongly with the three
cations. The differences in energy follow the ionic radius. Although
the electrostatic effect is the major partner in these complexes, the
van der Waals interactions are responsible for these variations in
energy, as they slightly favor the Na+ complex at the expense of
those with K+ and Cs+.
conformers are present for about 80% of the complex lifetime,
whereas the ‘‘transoid’’s are present for the remaining 20%. The
‘‘transoid’’s are less stable than the ‘‘cisoid’’s by about 4 kcal/mol
(Table 3). These differences in energy are the consequence of van
der Waals and electrostatic effects, which stabilize the ‘‘cisoid’’s.
The values of average energies of the systems are practically identi-
cal for the three cations, which implies that B1 binds all three (Table
3). In all three cases, the electrostatic term remains by far the main
component of the potential energy, because of the two negative
charges of the ligand. Comparison of the energies of the three Ak+
complexes is slightly in favor of Na+ (about 2 kcal/mol, Table 4)).
This is probably due to the van der Waals interactions which favor
smaller ionic radii. Moreover, the average distances between the
Ak+/heteroatoms and the ligands (Table 4) are longer in the ‘‘tran-
soid’’ conformation (6.41 Å for Cs+) than in the ‘‘cisoid’’ (6.29 Å for
Cs+). This highlights a more favorable electrostatic effect in the ‘‘ci-
soid’’ conformer. Furthermore, the lifetime of the ‘‘transoid’’ is
slightly longer with Na+. This is probably due to its smaller ionic
radius, which leads to less steric hindrance than with K+ and Cs+.
The enantiomers B2R-Ak+ and B2S-Ak+ complexes each adopt a
conformation in which the cation is surrounded by the oxygens of
the crown ether and capped with the aromatic part of the molecule
(Fig. 10). This capping maximizes the electrostatic interaction
4. Discussion
4.1. Complex formation
In Table 2, we summarize the kinetic and thermodynamic data
determined here. The kinetic runs were performed by the

54
A. Korovitch et al. / Inorganica Chimica Acta 394 (2013) 45–57
75
A
A
17
16
15
14
50
25
0
0.0
2.0x10-4
4.0x10-4
6.0x10-4
0.0
0.1
0.2
c1 (M)
Time (s)
B
160
24
B
120
80
22
20
18
40
2.0x10-6
4.0x10-6
6.0x10-6
8.0x10-6
0.00
0.03
0.06
0.09
Time (s)
c1 (M)
Fig. 7. Variation of the fluorescence emission with time for k P 400 nm at
kex = 365 nm after stopped-flow mixing in ethanol at 25.0 0.5 °C of a solution of:
Fig. 8. Plot of sꢀ1 against c1 after stopped-flow mixing in ethanol at 25.0 0.5 °C of
a solution of: (A) B1 (c0 = 5 ꢁ 10ꢀ7 M) with solutions of Cs+. Intercept, 2 1 sꢀ1
;
(A) B1 (c0 = 5 ꢁ 10ꢀ7 M) with
a
solution of Cs+ (c1 = 4 ꢁ 10ꢀ4 M) and (B) B2
slope, (1.22 0.07) ꢁ 105 Mꢀ1 sꢀ1; r = 0.99529, (B) B2 (c0 = 5 ꢁ 10ꢀ7 M) with solu-
tions of Cs+. Intercept, 11 4 sꢀ1; slope, (2.1 0.1) ꢁ 107 Mꢀ1 sꢀ1; r = 0.99178.
(c0 = 5 ꢁ 10ꢀ7 M) with a solution of Cs+ (c1 = 4 ꢁ 10ꢀ6 M).
stopped-flow mixing technique, which allows to investigate kinetic
processes occurring in the 10ꢀ3 – 1 s range. Complex formation with
alkali metals usually occurs by inclusion of the cation in the cavity of
a crown ether or a calixarene [22,25,50,51]. Very few examples con-
cerning complex formation between more classical ligands not
bearing an inclusion cavity, such as NbA, are known [16,18,52].
The case of NbA is, therefore, rather unusual, as ligands such as enols
and carboxylates form stable complexes with Cs+ and less stable
ones with Na+ and K+ [18,52]. Furthermore, complex formation with
alkali is believed to occur in the nano- to microsecond range and
was studied mainly by the use of fast kinetic techniques, such as
pulse and resonation ultrasonic relaxation or T-jump [15–19,53].
However, in the case of some capped calixarenes and NbA, complex
formation with Cs+ is slowed down [18–20]. The rate constants in
Table 2 are comparable to some of those reported for complex for-
rigidities or steric hindrances as in NbA or capped calixarenes. Nev-
ertheless, although, highly improbable, if conformational changes
were to rate-limit complex formation, the experimental relaxation
times would be first order processes. In this case, they should be
independent of the concentrations of the species present in the
medium [18,19,35–38]. This would be in a total contradiction with
our data as it is clearly shown that the experimental reciprocal
relaxation times depend on alkali metal concentrations and are re-
lated to complex formation (Figs. 7 and 8; Eqs. (4), (5), and (7)). This
is confirmed by the fact that the thermodynamic constants deter-
mined kinetically are within the limits of uncertainty identical to
those reported by spectrophotometric titration (Table 2). In M1,
these complex formations become much too fast (probably diffu-
sion-controlled) to be measured.
The structure of B1 resembles that of NbA, with the two tetronic
moieties assembled through a propyldiamide but without the
naphthoquinone and the carboxylic ligand involved in the second
Cs+ uptake by NbA [18]. However, in B1 the tetronic acids have cer-
tain flexibility, which is not the case in the separate moieties. Na+,
K+ and Cs+ can be considered as rather hard metals, which implies
the important role played by the oxygen ligands in complex forma-
tion with NbA [19,54]. The flexibility of the two tetronic moieties
and the fact that the apparent pKas of the enols are affected by
complex formation with Cs+ (Table 1) lead to the assumption of a
mation between cesium and NbA as well as a series of
2
crown[6]calyx[4]arenes in water, water–ethanol and ethanol. These
rate constants increased from 105 Mꢀ1 sꢀ1 to 6 ꢁ 109 Mꢀ1 sꢀ1 with
the ethanol/water ratio up to M1. This was attributed to several fac-
tors, such as the rigidity of the calixarene or slow conformational
changes, which allow the pulvinic acids of NbA to mimic an inclu-
sion cavity [18,19]. This, however, is not the case here because in
B1 and B2, conformational changes, would be extremely fast pro-
cesses. Indeed, these two molecules do not present any structural

A. Korovitch et al. / Inorganica Chimica Acta 394 (2013) 45–57
55
Fig. 9. Structures of the ‘‘cisoid’’ and ‘‘transoid’’ conformations of the Cs+ complex of B1, with ball and stick (top) and VDW (bottom) representations.
Table 3
Molecular energies of B1-Ak+, in kcal/mol, for the ‘‘cisoid’’ and ‘‘transoid’’ conformations. The time percentage stands for the presence ratio during the molecular dynamics
trajectories.
Ion
Cisoid
Na+
Transoid
Na+
Average
Na+
K+
Cs+
85
K+
Cs+
15
K+
Cs+
% time
Epot
78
80
22
20
–
–
–
ꢀ280.14
ꢀ278.17
ꢀ276.81
ꢀ276.82
ꢀ274.47
ꢀ273.67
ꢀ279.41
ꢀ277.43
ꢀ276.34
Ebond
Eangle
Edihed
Evdw
Eelec
16.95
17.04
52.98
31.13
17.06
16.71
16.07
53.20
32.88
17.19
16.90
16.85
17.08
51.74
31.67
ꢀ3.35
ꢀ377.22
0.07
53.08
31.05
ꢀ0.82
ꢀ377.35
0.18
52.02
32.01
ꢀ2.53
ꢀ375.12
0.08
52.74
32.02
ꢀ0.06
ꢀ375.72
0.16
51.80
31.74
ꢀ3.17
ꢀ376.75
0.07
53.03
53.03
31.48
31.20
ꢀ2.11
ꢀ377.34
0.13
ꢀ1.11
ꢀ375.62
0.11
ꢀ1.91
ꢀ377.00
0.13
ꢀ0.71
ꢀ377.11
0.18
Erestr
4.2. Modeling and proposed structures
Table 4
Distances, in Å, between Ak+ and heteroatoms in the B1 complex. The inter atomic
distance ID is displayed in Fig. 1.
Although, molecular dynamics simulations provide a supple-
mentary temporal dimension to explore molecular flexibility, they
are always performed in the ns scale with initial (2 ns in our case)
weak restraints imposed on the system in explicit solvents. This is
meant to impose an initial approach between the metal and the li-
gand. When the restraints are removed, the metal is released and
becomes free to interact or not with the ligand for the remaining
time (8 ns in our case). This time is short when compared to that
of complex formation which, as already mentioned, occurs in the
ID/
Na+
K+
Cs+
conformation
Cisoid Transoid Cisoid Transoid Cisoid Transoid
1
2
3
4
5
6
7
8
9
10
11
12
9.14
7.43
6.02
4.01
5.52
4.41
5.22
6.71
8.53
7.64
4.03
6.69
6.28
8.41
7.19
6.09
3.79
6.61
5.85
5.34
7.19
8.27
7.41
4.1
9.13
7.47
6.01
3.99
5.59
4.39
5.19
6.81
8.51
7.67
4.11
6.64
6.29
8.42
7.18
5.98
3.91
6.49
5.74
5.39
7.21
8.31
7.49
4.01
6.68
6.40
9.16
7.41
6.05
4.04
5.51
4.41
5.19
6.74
8.54
7.62
4.06
6.71
6.29
8.44
7.2
6.02
3.85
6.53
5.7
5.41
7.22
8.28
7.39
4.06
6.79
6.41
ls to the ms range (Table 2, Figs. 7 and 8). However, the molecular
dynamics simulations available in the ms scales with no initial
restraints validated approaches, such as ours [55–57]. These
simulations were all performed with ANTON, a massively parallel
supercomputer designed for molecular dynamics simulations of
proteins and other biological macromolecules [58]. Therefore,
although our models do not pretend to propose structures for the
final complexes, they, nevertheless constitute a realistic approach
of these structures. In B1, the affinities are surprisingly high for
alkali metals complexes, which do not involve an inclusion process
6.81
6.42
Average
complex with the enols in the deprotonated states. Therefore, the
affinities for the three alkali cations were measured at pH 9.3,
where all the enols are deprotonated.

56
A. Korovitch et al. / Inorganica Chimica Acta 394 (2013) 45–57
Fig. 10. Structures of B2–Cs+ complexes.
acids are missing, which implies a monocomplex formation be-
tween the alkali metal and the enolates and leads to the cisoid
and transoid conformations, with a preference for the cisoid’s be-
cause it mimics an inclusion cavity (Fig. 9).
Table 5
Molecular energies of B1, in kcal/mol, for the ‘‘cisoid’’ and ‘‘transoid’’ conformations.
The time percentage stands for the presence ratio during the molecular dynamics
trajectories.
B2-R
Na+
B2-S
Na+
K+
Cs+
K+
Cs+
4.3. Role of the crown ether
10
ꢀ71.70
ꢀ68.66
ꢀ65.52
ꢀ75.65
ꢀ71.22
ꢀ68.15
Ebond
Eangle
Edihed
Evdw
Eelec
Erestr
18.86
52.13
19.19
53.61
18.72
51.86
18.69
51.64
19.17
53.45
19.10
52.69
B2 contains a tetronic acid moiety and an 18-crown-6 ether.
Crown ethers are good ligands for alkali cations, with which they
form stable inclusion complexes [15,16,49,53,60]. Our aim was to
synthesize a specific Cs+ ligand by associating host–guest inclusion
with a more classical coordination. The result was a ligand with
higher affinities for Cs+ than crown ethers, which can be compared
to that of more specific Cs+ ligands, such as calixarenes
[22,23,51,61–63]. B2 is, however, also a good ligand for K+ and
Na+ in M1 and a bit more selective for Cs+ in EtOH (about one order
of magnitude) (Table 2). Although molecular dynamics calculations
are based on electrostatic and van der Waals interactions, they
clearly imply that the B2–Cs+ complex is present in ethanol in
two conformations (Fig. 10, Table 6). The inter-atomic distances
(Table 6) between Cs+ and the oxygens of the crown ether are very
close to those reported in the literature for the crystal structures of
similar molecules [64]. The addition of the tetronic acid moiety sta-
bilizes the inclusion complex by forming in both conformers a
structure in which Cs+ is included in the crown ether and interacts
at the same time with amide 13 and oxygens 2–4. The interatomic
distances involved in the more stable conformer B2S–Ak+ (Fig. 10)
and its lifetime (80% of that of the complex) indicate that the enol
of the tetronic moiety acts as a supplementary arm which tends to
cap Cs+ and maintains it included in the crown ether. This also oc-
curs with B2R–Ak+ with, however, a lower lifetime. This explains
the higher affinity of B2 for Cs+ as compared to that of simple
18-crown-6 ethers [17].
38.63
38.35
39.65
35.61
36.03
36.57
ꢀ9.19
ꢀ172.22
0.08
ꢀ6.35
ꢀ173.67
0.21
ꢀ5.52
ꢀ170.78
0.55
ꢀ8.90
ꢀ172.80
0.11
ꢀ6.10
ꢀ173.82
0.05
ꢀ5.03
ꢀ171.73
0.25
Table 6
Distances, in Å, between Ak+ and the heteroatoms in the B2 complex. The ID is
displayed in Fig. 1.
ID/ion
B2-R
Na+
B2-S
Na+
K+
2.93
Cs+
4.03
K+
3.03
Cs+
3.57
1
2
3
4
5
6
7
8
3.04
2.83
2.86
2.91
2.41
2.59
10.58
6.28
7.17
3.86
8.09
6.10
5.63
2.86
3.02
3.06
3.12
2.74
2.56
10.11
6.10
7.31
3.87
8.23
5.83
5.34
3.11
3.12
3.09
2.97
3.06
11.45
7.16
7.75
3.78
7.68
6.27
6.81
3.61
3.68
3.68
3.73
4.09
11.70
7.61
8.27
4.08
7.23
6.76
6.27
3.14
2.96
3.32
3.28
3.22
10.24
7.12
7.79
3.80
7.94
6.01
5.54
3.96
3.80
4.11
4.09
4.08
10.48
7.32
8.31
3.86
7.62
6.16
5.58
9
10
11
12
13
Average
4.95
5.32
5.75
4.93
5.18
5.61
in a crown ether or a calixarene [16]. The models proposed in Fig. 9
show that in B1 (Fig. 1 and Table 4), the alkali metals are very close
to oxygens 4 and 11 (enolates), close to oxygen 5 (one of the two
carbonyls) and to nitrogens 6 and 7 (diamide). This probably im-
plies that a complex is formed between the alkali metals and the
two enolates, one of the carbonyls and the diamide, which can in
this case explain the values of the rather high affinity constants
(Table 2). In NbA, complex formation occurs with two Cs+, the first
with the enolates and the second with the carboxylates at higher
pH values [18]. Complex formation between Cs+ and NbA involves
the Z and E conformers with a broad diversity of binding modes
[59]. In the case of B1, the naphthoquinone and the carboxylic
4.4. Mimicking NbA?
NbA forms strong complexes with Cs+ in alcohol and water–
alcohol mixtures. These complexes involve the two enolates and
the two carboxylates. Complex formation occurs with one Cs+
when only the enols are in the ionized state and with two Cs+ when
both the enol and the carboxylic acid moieties are deprotonated.
This affinity of NbA for Cs+ is probably the result of the particular
structure in which the two pulvinic acid arms adopt a conforma-
tion that forms two complexation sites consisting of the two eno-
lates and/or the two carboxylates [18]. In B1, a flexibility of the
pulvinic ligand was sought to mimic an inclusion pseudocavity

A. Korovitch et al. / Inorganica Chimica Acta 394 (2013) 45–57
[16] M. Eigen, G. Maass, Z. Phys. Chem. 49 (1966) 163.
57
or to adopt the structure of a pair of jaws that would surround the
alkali metal. This is also implied by molecular dynamics, where the
B1-Cs+ complex in ethanol appears as two conformers, ‘‘transoid’’
and ‘‘cisoid’’ (Table 4 and Fig. 9). The latter is more stable than
the ‘‘transoid’’ (Table 3). The ‘‘cisoid’’ conformation mimics, as ex-
pected here and assumed for NbA, an inclusion cavity in which the
alkali cation is complexed by the enolate and carboxylate ligands.
However, this flexibility does not increase the affinity of B1 for Cs+
as compared to NbA [18]. With B1 a monocomplex is formed with
K+, Na+ and Cs+ and the two enols. Furthermore, the affinities in-
volved for the three alkali metals are within the same order of
magnitude, whereas NbA shows specificity for Cs+ (Table 2) [18].
The lack of the carboxylates does not explain the lack of selectivity
for Cs+. With NbA this selectivity is the result of the positioning of
the pulvinic acids imposed by the naphthol and the presence of the
two carboxylic acid moieties which enables the molecules to com-
plex two Cs+ (Table 2) [18].
[17] G.W. Liesegang, M.M. Farrow, N. Purdie, E.M. Eyring, J. Am. Chem. Soc. 98
(1976) 6905.
[18] A. Korovitch, J.B. Mulon, V. Souchon, I. Leray, B. Valeur, A. Mallinger, B. Nadal, T.
Le Gall, C. Lion, N.T. Ha-Duong, J.M. El Hage Chahine, J. Phys. Chem. B 114
(2010) 12655.
[19] A. Korovitch, J.B. Mulon, V. Souchon, C. Lion, B. Valeur, I. Leray, N.T. Ha-Duong,
J.M. El Hage Chahine, J. Phys. Chem. B 113 (2009) 14247.
[20] Y. Suzuki, H. Otsuka, A. Ikeda, S. Shinkai, Tetrahedron Lett. 38 (1997) 421.
[21] U.C. Meier, C. Detellier, J. Phys. Chem. A 103 (1999) 9204.
[22] Z. Asfari, C. Bressot, J. Vicens, C. Hill, J.F. Dozol, H. Rouquette, S. Eymard, V.
Lamare, B. Tournois, Anal. Chem. 67 (1995) 3133.
[23] Z. Asfari, C. Naumann, J. Vicens, M. Nierlich, P. Thuery, C. Bressot, V. Lamare, J.F.
Dozol, New J. Chem. 20 (1996) 1183.
[24] H.F. Ji, R. Dabestani, G.M. Brown, R.A. Sachleben, Chem. Commun. (2000) 833.
[25] V. Souchon, I. Leray, B. Valeur, Chem. Commun. (2006) 4224.
[26] H.M. Chawla, S.P. Singh, S. Upreti, Tetrahedron 62 (2006) 2901.
[27] S. Garaudée, M. Elhabiri, D. Kalny, C. Robiolle, J.M. Trendel, R. Hueber, A.V.
Dorselaer, P. Albrecht, A.M. Albrecht-Gary, Chem. Commun. 9 (2002) 944.
[28] P. Kuad, R. Schurhammer, C. Maechling, C. Antheaume, C. Mioskowski, G.
Wipff, B. Spiess, Phys. Chem. Chem. Phys. 11 (2009) 10299.
[29] A. Le Roux, S. Meunier, T. Le Gall, J.M. Denis, P. Bischoff, A. Wagner, Chem. Med.
Chem. 6 (2011) 561.
[30] Y. Bourdreux, E. Bodio, C. Willis, C. Billaud, T. Legall, C. Mioskowski,
Tetrahedron 64 (2008) 8930.
[31] D. Habrant, A. Le Roux, S. Poigny, S. Meunier, A. Wagner, C. Mioskowski, J. Org.
Chem. 73 (2008) 9490.
The results obtained with B2 are interesting, as the affinities for
Cs+ are higher than those with NbA and those with the crown ether
alone (K1crownEtOH ꢂ 1 ꢁ 105).
[32] J. Volhard, Justus Liebigs Ann. Chem. 282 (1894) 1.
[33] R.G. Bates, Determination of pH – theory and practice, Wiley-Interscience,
New York, 1973.
5. Conclusion
[34] R.A. Binstead, A.D. Zuberbühler, B. Jung, SPECFIT global analysis system version
3.04.34, 2003.
The decontamination of 137Cs+ following nuclear incidents, such
as Chernobyl and Fukushima, remains a major public health prob-
lem. To the best of our knowledge, the elimination of this radioiso-
tope is not yet possible. In nature, NbA is among the few molecules
which form stable complexes with Cs+. To mimic its structure in
order to achieve such a purpose is of interest. This is shown here
by the capabilities of B1 and B2 to form stable complexes with al-
kali metals. Even if with B2 a slight specificity for Cs+ is obtained, it
remains that both B1 and B2 are not as specific for Cs+ as envis-
aged. However, they constitute a framework which, with the
involvement of the naphthol present in NbA, can lead to more spe-
cific Cs+ chelators.
[35] R. Brouillard, J. Chem. Soc., Faraday Trans. 1 (76) (1980) 583.
[36] C.F. Bernasconi, Relaxation Kinetics, Academic Press, New York, 1976.
[37] M. Eigen, L. DeMaeyer, Techniques of Chemistry, Wiley, New York, 1973.
[38] J.M. El Hage Chahine, J.E. Dubois, J. Am. Chem. Soc. 105 (1983) 2335.
[39] SYBYL, in: Tripos Inc., 1699 South Hanley Rd., St. Louis, MO 63144, USA.
[40] D.A. Case, T.A. Darden, T.E. Cheatham, C.L. Simmerling, J. Wang, R.E. Duke, R. Luo,
R.C. Walker, W. Zhang, K.M. Merz, B. Roberts, B. Wang, S. Hayik, A. Roitberg, G.
Seabra, I. Kolossvary, K.F. Wong, F. Paesani, J. Vanicek, J. Liu, X. Wu, S. Brozell, T.
Steinbrecher, H. Gohlke, Q. Cai, X. Ye, J. Wang, M.-J. Hsieh, G. Cui, D.R. Roe, D.H.
Mathews, M.G. Seetin, C. Sagui, V. Babin, T. Luchko, S. Gusarov, A. Kovalenko,
P.A. Kollman, AMBER 11, University of California, San Francisco, 2010.
[41] J. Wang, R.M. Wolf, J.W. Caldwell, P.A. Kollman, D.A. Case, J. Comput. Chem. 25
(2004) 1157.
[42] A. Jakalian, B.L. Bush, D.B. Jack, C.I. Bayly, J. Comput. Chem. 21 (2000) 132.
[43] A. Jakalian, D.B. Jack, C.I. Bayly, J. Comput. Chem. 23 (2002) 1623.
[44] J. Aaqvist, J. Phys. Chem. 94 (1990) 8021.
[45] C. Teixeira, N. Serradji, F. Maurel, F. Barbault, Eur. J. Med. Chem. 44 (2009)
3524.
Acknowledgment
[46] W. Humphrey, A. Dalke, K. Schulten, J. Mol. Graph. 14 (1996) 33.
[47] J.M. El Hage Chahine, A.M. Bauer, K. Baraldo, C. Lion, F. Ramiandrasoa, G.
Kunesch, Eur. J. Inorg. Chem. (2001) 2287.
[48] H.A. Benesi, J.H. Hildebrand, J. Am. Chem. Soc. 71 (1949) 2703.
[49] P.B. Chock, Proc. Natl. Acad. Sci. USA 69 (1972) 1939.
[50] F. Arnaud-Neu, Z. Asfari, B. Souley, J. Vicens, New J. Chem. 20 (1996) 453.
[51] I. Leray, Z. Asfari, J. Vicens, B. Valeur, J. Chem. Soc., Perkin Trans. 2 (2002) 1429.
[52] P. Kuad, M. Borkovec, M. Desage-El Murr, T. Le Gall, C. Mioskowski, B. Spiess, J.
Am. Chem. Soc. 127 (2005) 1323.
[53] C. Chen, W. Wallace, E.M. Eyring, S. Petrucci, J. Chem. Phys. 88 (1984) 2541.
[54] R.G. Pearson, J. Am. Chem. Soc. 110 (1988) 7684.
[55] K. Lindorff-Larsen, P. Maragakis, S. Piana, M.P. Eastwood, R.O. Dror, D.E. Shaw,
PLoS One 7 (2012) e32131.
This work was supported by DGA (Délégation Générale pour
l’Armement) Grant PEA PROPERGAL N°. 06.70.110. The authors
are grateful to Dr. John S. LOMAS for helpful discussions.
References
[1] Toxicology profile for cesium in: Agency Toxic Substances and Disease Registry
(Ed.), US Department of Health and Human Services, 2004.
[2] P. Melnikov, L.Z. Zanoni, Biol. Trace Elem. Res. 135 (2010) 1.
[3] Y. Jiang, Y. Feng, Y. Wang, J. Lu, M. Hu, S. Li, Chem. Biol. Interact. 171 (2008)
325.
[4] F. Brown, G.R. Hall, A.J. Walter, J. Inorg. Nucl. Chem. 1 (1955) 241.
[5] Chernobyl’s Legacy in: Health, Environmental and Socio-Economic Impacts
and Recommendations to the Governments of Belarus, the Russian Federation
and Ukraine (Ed.), International Atomic Energy Agency D.K.I. Division of Public
Information Austria, 2006.
[6] A. Bolsunovsky, D. Dementyev, J. Environ. Radioact. 102 (2011) 1062.
[7] D. Butler, Nature 472 (2011) 13.
[8] D. Butler, Nature 471 (2011) 555.
[56] K. Lindorff-Larsen, S. Piana, R.O. Dror, D.E. Shaw, Science 334 (2011) 517.
[57] K. Lindorff-Larsen, N. Trbovic, P. Maragakis, S. Piana, D.E. Shaw, J. Am. Chem.
Soc. 134 (2012) 3787.
[58] D.E. Shaw, R.O. Dror, J.K. Salmon, J.P. Grossman, K.M. Mackenzie, J.A. Bank, C.
Young, M.M. Deneroff, B. Batso, K.J. Bowers, E. Chow, M.P. Eastwood, D.J. Ieradi,
J.L. Klepeis, J.S. Kuskin, R.H. Larson, K. Lindorff-Larsen, P. Maragakis, M.A.
Moraes, S. Piana, Y. Shan, B. Towles, Proceedings of the ACM/IEEE Conference
on Supercomputing (SC09) (2009) 1.
[9] D. Butler, Nature 472 (2011) 400.
[10] D.C. Auman, G. Clooth, B. Steffan, W. Steglich, Angew. Chem., Int. Ed. Engl. 28
(1989) 453.
[59] R. Schurhammer, R. Diss, B. Spiess, G. Wipff, Phys. Chem. Chem. Phys. 10
(2008) 495.
[60] H.K. Frensdorff, J. Am. Chem. Soc. 93 (1971) 600.
[11] B. Steffan, W. Steglich, Angew. Chem., Int. Ed. Engl. 23 (1984) 445.
[12] W. Steglich, W. Furtner, A. Prox, Z. Naturforsch. 23b (1968) 1044.
[13] W. Steglich, W. Furtner, A. Prox, Z. Naturforsch. 25b (1970) 557.
[14] M. Winner, A. Giménez, H. Schmidt, B. Sontag, B. Steffan, W. Steglich, Angew.
Chem., Int. Ed. Engl. 43 (2004) 1883.
[61] V. Arora, H.M. Chawla, T. Francis, M. Nanda, S.P. Singh, Ind. J. Chem. 42A (2003)
3041.
[62] V. Arora, H.M. Chawla, S.P. Singh, Arkivoc ii (2007) 172.
[63] I. Leray, Z. Asfari, J. Vicens, B. Valeur, J. Fluorescence 14 (2004) 451.
[64] T.C. Kim, S.S. Lee, J.S. Kim, J. Kim, Anal. Sci. 17 (2001) 573.
[15] P.B. Chock, F. Eggers, M. Eigen, R. Winkler, Biophys. Chem. 6 (1977) 239.