New water-soluble calix[4]arene-bis(benzocrown-6) for caesium-sodium separation by nanofiltration-complexation

Article abstract of DOI:10.1016/j.tet.2003.10.051

Calix-bis(benzocrown-6) 6 and 7 were converted into the water-soluble receptors 9, 10, 12 and 15 by introducing hydroxy, carboxy, sulfato or diethanolamino groups at the para position of the phenolic ring and/or on the benzo-ether moieties. The complexati

Full text of DOI:10.1016/j.tet.2003.10.051

Tetrahedron 59 (2003) 10313–10324
New water-soluble calix[4]arene-bis(benzocrown-6) for
caesium–sodium separation by nanofiltration–complexation
´ ´ ´
Stephane Pellet-Rostaing, Frederic Chitry, Jean-Alexis Spitz, Antoine Sorin,
*
´
Alain Favre-Reguillon and Marc Lemaire
` ´
Laboratoire de Catalyse et Synthese Organique, Universite Claude Bernard Lyon I, CPE Lyon, 43 bd du 11 Novembre 1918,
69100 Villeurbanne, France
Received 17 April 2003; revised 12 September 2003; accepted 1 October 2003
Abstract—Calix-bis(benzocrown-6) 6 and 7 were converted into the water-soluble receptors 9, 10, 12 and 15 by introducing hydroxy,
carboxy, sulfato or diethanolamino groups at the para position of the phenolic ring and/or on the benzo-ether moieties. The complexation
properties of these ionophores were studied for all alkali cations in methanolic and aqueous media. Stability constants were calculated by
UV–Vis spectroscopy. All ligands showed a more or less affinity for the larger cations, depending on the nature and the position of the
substituents grafted on the benzo-ether chain only or both on the calixarene ring and the benzo-ether loop. For selective Csþ/Naþ separation,
the efficiency of the ligands was evaluated by means of a nanofiltration system. In comparison with the known tetrahydroxylated bis-crown-6
calix[4]arene 1, compounds 9, 12 and 15 represent the most selective ligands for the Csþ cation in a moderate salted medium
([NaNO3]¼85 g/L).
q 2003 Elsevier Ltd. All rights reserved.
1. Introduction
Nanofiltration techniques showed great potential for the
separation of metal cations in aqueous phase. It allowed the
separation of different ion valences, but selectivity was not
generally ensured for the separation of ions with same
valences. This could be improved by associating the
nanofiltration step to a preliminary selective complexation
step with water-soluble ligands of molecular weight higher
than the ‘cut off’ of the membrane (300–1000 g mol). In
previous reports we described the synthesis and the
evaluation of several water-soluble calix[4]arene-bis-
(crown-6)33,34 incorporating neutral or proton-ionizable
substituents such as sulfonamide, sulfonate, carboxy and
hydroxy groups at the para position of the aromatic ring.
The selective complexation of caesium towards alkali ions
(Naþ, Kþ, Rbþ, Csþ) and with these ligands were studied in
basic aqueous media by UV–Vis analysis, NMR spec-
troscopy and nanofiltration–complexation tests.35 Some of
these ligands showed a more or less pronounced affinity for
the target ion. In fact, these studies allowed us to prove that
the efficiency of these macrocycles depended on the number
and the nature of the grafted hydrophilic functionalities.
These play a major role in the complexation and
stabilization of the caesium ion both in the cavity of the
hydrophobic aromatic ringligand and in the crown ether
chain as also described by Vicens et al.36 with para-
sulfonated 1,2;3,4-calix[4]arene-bis(crowns) in the 1,2-
alternate conformation or by Bartsch et al.37,38 in a
liquid–liquid extraction process with different calixcrowns
mono-substituted. With a Csþ/Naþ selectivity of 6570, the
Due to their interesting selective binding abilities towards
alkali cations, calixcrowns1 constitute an important class
of macrocyclic compounds combining calixarene2 and
crown ether3 frameworks. Particularly important are the
dialkoxycalix[4]arene(crown-6) derivatives4–14 and ana-
logous bis-(crown-6)15–19 in 1,3-alternate conformation
which showed in many cases a high selectivity towards
caesium ions. Thus numerous studies concerning solvent
extraction6,20 or liquid membrane transport,18 solution
thermodynamics,6,10 detection,21,22 X-ray crystal structure
and molecular modeling8,23–29 have been performed on
these receptors. This has opened up new possibilities for
radioactive 137Cs separation from nuclear fuel reproces-
sing solutions containing much larger amounts of
sodium30 (20£103 times as much than caesium) which
have become of crucial interest for the nuclear industry.
Since becoming involved in this project, we chose to use the
technique of nanofiltration–complexation (NF) as a mem-
brane separation process,31,32 especially dedicated to the
basic effluents reprocessing, resulting from the nuclear
facilities washing.
Keywords: calix[4]arene; (benzo)crown ether; water-soluble; caesium;
nanofiltration–complexation.
*
Corresponding author; e-mail: marc.lemaire@univ-lyon1.fr
0040–4020/$ - see front matter q 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2003.10.051

10314
S. Pellet-Rostaing et al. / Tetrahedron 59 (2003) 10313–10324
soluble benzo-crown-6 derivatives of calix[4]arene, a study
of their affinity towards Csþ, Rbþ, Kþ and Naþ cations in
methanol and in aqueous media, and finally the various
aspects of the use of these ligands in a nanofiltration–
complexation process. The water-soluble macrocycles were
obtained by the preliminary preparation of two new
lipophilic calixcrowns, the calix[4]arene-bis-[(4-methyl-
1,2-phenylene)crown-6] 6 and the bis-[(4-ethoxycarbonyl-
1,2-phenylene)crown-6] analogue 7. These macrobicycle
ligands were characterised by the presence of a calixarene
ring and two substituted catechol subunits which could be
modified separately or together. Thus, water-soluble
derivatives were obtained either by modifying the ester
function of 7 or by the incorporation of hydrophilic groups
on the methyl-phenylene moiety and the calixarene ring of 6
at once.
Figure 1. 5,11,17,23-Tetrahydroxycalix[4]arene-bis(crown-6).
5,11,17,23-tetrahydroxycalix[4]arene-bis(crown-6)
1
(Fig. 1) was the best receptor for the caesium.35 This
achieved the best compromise between preorganization,
steric hindrance and cooperative effect of the cation–arene
and cation–phenolate interactions. Nevertheless, we esti-
mated that the Csþ/Naþ selectivity could be enhanced using
water-soluble calix[4]arene-bis(benzo-crown-6) of which
lipophilic analogues exhibited better selectivities in organic
medium.18,39–43
2. Results and discussion
2.1. Synthesis and characterisation
Calix[4]arenes bis(crown-6) 6 and 7 were synthesized
according to the established method of Vicens et al.18,44
They were obtained as a mixture of two enantiomers
(Scheme 1). 4-Methyl-catechol and ethyl-3,4-dihydroxy-
benzoate were reacted with 2-(2-chloroethoxy)ethanol in
45,44
We report in this paper the synthesis of several new water-
the presence of K2CO3
to give the diols 2 and 3 in
Scheme 1.

S. Pellet-Rostaing et al. / Tetrahedron 59 (2003) 10313–10324
10315
Scheme 2.
acceptable yields. Ditosylates 4 and 5 were respectively
obtained in good yields by reaction of tosylchloride with
2 and NaOH in THF–H2O or 3 in CH2Cl2 and Et3N.
Finally, macrocyclisation with the calix[4]arene and
ditosylates 4 and 5 were performed in refluxing CH3CN in
the presence of K2CO3 to give calixcrowns 6 and 7 with
respectively 66 and 41% yields. Symmetrical 1,3-alternate
structures of these compounds were determinated by NMR
spectroscopy which showed very simple patterns. Thus,
derivatives 6 and 7 exhibited well defined 1H NMR spectra.
The 1,3-alternate conformations were deduced from the
presence of one singlet at 3.78 ppm for 6 and 3.76 ppm for 7
corresponding to the resonance signals of the ArCH2Ar
protons. 13C NMR spectroscopy confirmed these struc-
tures,46 since only one signal assigned to the bridging
methylene carbons was present at 37.92 ppm for 6 and
37.78 ppm for 7.
the usual conditions (NBS, benzoyl peroxide or AIBN in
CCl4) but these synthetic routes failed. Vicens et al.1
synthesised a dibromomethyl analogue by the reduction of
calix[4]arene1,3;2,4-bis-[(5-ethoxycarbonyl-1,3-phenyl-
ene)crown-6] with 40 equiv. of LiAlH4 in Et2O followed by
the conversion of the diol into the dibromomethyl derivative
with an excess of PBr3 in Et2O. By using the same
procedure, reduction of 7 gave dihydroxymethyl 13 in a low
yield of 10% even though the reaction time was extended to
5 days. On the other hand, when 1,2-dimethoxyethane was
used instead of Et2O, 13 was obtained in 80% yield.
Treatment of diol with PBr3 in THF afforded 14 in 90%
yield which was finally converted into the bis-diethanola-
mine 15 with HN(CH2CH2OH)2 in DMSO in 50% yield. In
the case of compound 12, 1H NMR (Fig. 2) showed that the
aromatic signals of the calixarene ring were slightly
different to those expected (spectrum A). Indeed, calixarene
protons were characterised by two overlapped triplets at
0
6.57 and 6.59 ppm (Ha, Ha ) and two overlapped doublets at
0
7.11 and 7.13 ppm (Hb, Hb ) demonstrating an unexpected
Water-soluble derivatives 9 and 10 were obtained in two
steps as depicted in Scheme 2. Subsequent bromination of 6
with Br2 in CH2Cl2 at 08C afforded the hexa-bromo
derivative 8 (75%). Treatment of compound 8 with an
excess of t-BuLi in dry THF at 2788C for 90 min followed
by quenching with B(OMe)3, oxidation with 3 M NaOH–
35% H2O2 or quenching with CO2 gave polyol 9 and
polyacid 10 in 62 and 76% yield, respectively.
dissymmetry of the structure of the receptor. This anomaly
disappeared when a drop of DCl (spectrum B) was added in
the NMR sample and phenolic protons of the calixarene ring
were characterised by a triplet at 6.53 ppm and a doublet at
7.09 ppm. This phenomenon could be explained by the
possible complexation of one of the potassium ion of
the sulfonate chain which could be trapped inside the
corresponding crown ether chain. Furthermore, the presence
of well defined signals indicated a rapid cation–crown ether
exchange between both sulfonates substituents.
Exhaustive substitution of 6 in the fixed 1,3-alternate cone
conformation simplified the characterisation of the reaction
products by NMR in CD3Cl. One singlet for the ArH meta of
the calixarene ring, two singlets for the residual ArH of the
benzo(crown) moiety and one singlet for the bridged
methylene allowed the characterisation of each macrocyclic
Due to their structural analogy, calixcrowns derivatives 11,
13–15 were characterised by similar 1H and 13C NMR
resonance signals in CD3Cl (14) or DMSO-d6 (11, 13, 15)
corresponding to the ArCH, ArC, ArCH2Ar and OCH2CH2O
protons and carbons.
1
frame by H NMR.
Modification of the ester groups of derivative 7 afforded
ligands 12 and 15 soluble in water at pH 7, in two and three
steps (Scheme 3). The synthesis of disulfato derivative 12
involved the hydrolysis of the ester group with KOH in
refluxing ethanol followed by the reaction of the carboxylic
acid intermediate 11 with 1,2,3-dioxathiolane-2,2-dioxyde
in the presence of K2CO3 in CH2Cl2 to give 12 in an overall
yield of 64%. Compound 15 was obtained via the synthesis
of dibromomethyl intermediate 14. In order to prepare the
latter, radical bromination of ligand 6 was attempted under
The removal of the triplet and the quadruplet of ethyl ester
1
protons of 7 in the H NMR spectra and the presence of
signals assigned to the grafted substituents confirmed the
modification of the benzo ether chains (COOH of 11
appeared at 12.71 ppm and benzylic CH2 of 13, 14 and 15 at
4.43, 4.52 and 3.64 ppm, respectively).
Mass measurements, performed by electrospray technique

10316
S. Pellet-Rostaing et al. / Tetrahedron 59 (2003) 10313–10324
Scheme 3.
1
Figure 2. H NMR of compound 18 (A: in DMSO-d6, B: in DMSO-d6þDCl).

S. Pellet-Rostaing et al. / Tetrahedron 59 (2003) 10313–10324
10317
Table 1. Determination of Cs or Na-ligands 9, 10, 11, 15 complexes
association constants by UV spectrometry in methanol
in some cases, the appearance of isobestic points. For every
ligand, a 1:1 Mþ–ligand ratio was deduced from the straight
line corresponding to the plot of Dabs/[MCl] versus Dabs
which allowed the calculation of the association constant b
from the slope. Association constants without units, b0¼
brMeOH are given in Table 1. As a general rule, the
introduction of benzo groups in the crown loop leads to a
decrease in the stability of the Naþ–ligand complexes while
no change was observed in the stability of Csþ–11 and
Csþ–15 complexes as already mentioned by Arnaud-Neu
et al.48 with the calix[4]arene-dibenzo(crown-6). On the
other hand, the stability of the Csþ–9 complex was
enhanced, certainly due to the good compromise between
the rigidification of the crown part and the cooperative effect
of the cation–arene and cation–OH interactions. The
stability of the Csþ–10 complex was lower due to the
likely steric hindrance generated by the presence of the four
carboxy groups grafted on the calixarene part. Finally, the
presence of ionizable substituents on the phenolic ring
Ligand
135
9
10
11
15
log bC0 s–L 5.79^0.01 5.89^0.01 5.44^0.01 5.81^0.01 5.80^0.01
log bN0 a–L 3.5^0.2
3.2^0.2 3.3^0.2 2.9^0.3 3.0^0.3
were in accordance with the hosts’ expected structures, par-
ticularly the mass spectrum of the dipotassium salt 12 which
showed two signals at m/z 1359.2 and 1375.2 corresponding to
the monocharged species ([12þNa]þ and [12þK]þ).
2.2. Complexation studies
2.2.1. UV–Vis studies. The molecular recognition proper-
ties of the hydrophilic calix[4]arene-bis(benzocrowns)
compounds 9–12 and 15 towards alkali ions were studied
by UV–Vis spectroscopy analysis and compared with the
complexation properties of the macrocycle 1, in methanol
and/or in basic aqueous medium depending on the solubility
of ligands. The association constants b of the metal–ligand
complexes were determined by absorption method, in which
b is a constant ratio ([MLþ]/[Mþ][L]) corresponding to the
equilibrium MþþL$ML (Mþ is Naþ, Kþ, Rbþ or Csþ and
L is the ligand). The absorption changes in the UV spectra of
the ligand were recorded as alkali ions were progressively
added to the medium containing a fixed amount of receptor.
These modifications allowed the calculation of bML stability
constants, according to the Foster–Hammick–Wardley
method.47
seemed to increase the stability of Naþ–ligand with bN0 a/1
,
bN0 a/9 and bN0 a/10 values higher than those obtained with
unsubstituted derivatives 11 and 15.
2.2.3. Aqueous medium. The complexing ability of water-
soluble ligands 9, 10, 12 and 15 toward Csþ, Rbþ and Kþ
ions was also studied in aqueous medium in the same
conditions. The association constants given in Table 2 are
apparent constants b p, because they depend on Naþ
amounts in the medium. Meanwhile the complexing ability
of calixarenes toward other alkali ions such as Csþ, Rbþ and
Kþ could be estimated. Carboxyl derivative 10 induced a
hypochromic effect of the absorption band at 246 nm owing
to the interaction of ligands with caesium and rubidium
cations. The presence of only one isobestic point at 278 nm
confirmed the 1:1 Csþ–10 ratio, as well as the straight line
corresponding to the plot of Dabs/[MCl] versus Dabs in
accordance with the Foster–Hammick–Wardley procedure
(Fig. 3).
2.2.2. Methanolic medium. Ligands 9, 10, 11 and 15 were
tested by UV–Vis analysis in methanol toward Csþ and
Naþ cations and recorded between 240 and 330 nm.
Addition of CsCl to the ligands solution induced batho-
chromic effects, whereas NaCl additions induced hypso-
chromic effects on the band of each ligand at 285^3 nm and
Table 2. Logarithms of the apparent association constants (log b p^sn21) of alkali–ligand complexes by UV spectrometry for ligands 9, 10, 12 and 15 in
aqueous medium at 208C (standard deviation s obtained from n experiments with 2,n,4)
Ligands
135
9
10
0.001
12
0.01
15
0.01
[NaOH] (mol/L)
lmax (nm)
0.01
304
5.63^0.01
5.27^0.02
4.60^0.06
0.01
305
246
284
286
Csþþ
5.75^0.01
5.34^0.02
4.50^0.06
5.38^0.01
4.78^0.04
N.d.
5.71^0.01
5.37^0.02
4.35^0.06
5.70^0.01
5.36^0.02
4.34^0.06
Rb
Kþ
Figure 3. Plot of Dabs/[CsCl]¼f(Dabs) with ligand 10 in NaOH 0.1 M at room temperature.

10318
S. Pellet-Rostaing et al. / Tetrahedron 59 (2003) 10313–10324
Figure 4. Changes in the absorption spectrum of a solution of ligand 9 with CsCl addition in water ([9]¼2£1024mol/L; pH¼12.5).
Interactions of CsCl with compound 9 resulted in significant
hyperchromic and bathochromic shifts. Variations of
absorption resulted in three isobestic points at 245, 275
and 302 nm (Fig. 4) and the straight line from the plot of
1H NMR spectroscopy in DMSO-d6. Addition of NaNO3 in
excess to solutions of ligands showed that no complexation
of Naþ occurred in view of the insignificant changes from
the free ligands spectra observed. On the other hand,
addition of CsNO3 in excess to solutions of ligands led to
appreciable changes due to the proximity of the Csþ cation
nearby the altered protons. Thus, due to the p–Csþ
interactions,48 downfield shifts of the aromatic proton’s
resonance of the phenolic rings were observed (Table 3) and
were characterised by relatively sharp signals. Similarly, the
changes in the level of the signals of the protons of the
glycolic chains were also indicative of the interaction of
Csþ with oxygen donor atoms. As an example, the low
downfield shift of the meta phenolic protons’ resonance of
10 (Dd¼0.03 ppm) and the higher one corresponding to the
multiplet of the OCH2CH2OPh (Dd¼0.23 ppm) could be
interpreted by weak p–aromatic interactions which let
suppose the presence of an exo complex with the cation
located within the polyether loop and stabilized by the
COOH groups. A second example showed that the addition
of CsNO3 in excess to a solution of 11 was accompanied by
downfield shifts of the triplet and doublet assigned to the
para and meta phenolic protons (Dd¼0.07 and 0.16 ppm,
respectively) suggesting the presence of an endo complex.
Similar effects were observed with ligands 15 and 12.
D
abs/[MCl] versus Dabs proved the formation of 1:1
complexes with Csþ. Finally, the spectra of solutions of
ligands 12 and 15 containing increasing amount of CsCl and
RbCl present the same pattern. The addition of metal ion
induced an increase of the intensity of the two main
absorption bands of each ligand (252 and 284 nm) with a
low bathochromic shift and the appearance of two
isobestic points at 248 and 265 nm for 12 and 245 and
260 nm for 15. With Kþ, 9, 12 and 15 showed a moderate
affinity (log b pK–L¼4.50, 4.35, 4.34, respectively). As
observed in methanol, the addition of 9 with four hydroxyl
side arms on the calixarene ring or 12 and 15 with no
substituents on it exhibited much higher complexation
efficiency towards metal cations (log b pCs–L¼5.75, 5.71,
5.70 and log b pRb–L¼5.34, 5.37, 5.36, respectively) than
10 with carboxy substituents (log b pCs – 10¼5.38, log
b pRb–10¼4.78). Finally, the complexation level of Csþ by
calixcrowns 1, 12, 15 was practically the same and lower
than the one obtained with 9.
Results confirmed that the efficiency of the ligand involved
a good synergy between the nature of the polyether chain
and the number and the nature of the substituents grafted on
the phenolic ring. In both solvents, the selectivity increase
according to the sequence 10,1,12¼15,9. On the other
hand, the presence of the aryl group in the polyether chain
and the OH function resulted in a polar and rigid
macrocyclic cavity enough to maximise the cation–ligand
interactions in basic aqueous media with a higher selectivity
for the polyhydroxy derivative 9.
Furthermore, to confirm the 1:1 stoichiometry of the Csþ
1
complexes, a H NMR study was also performed with the
1:1 complex of L–CsþNO3 at room temperature. In
comparison of the spectral modifications obtained upon
the addition of CsNO3 in excess, no significant changes
were observed with a ratio ligand/caesium of 1:1, assuming
the 1:1 stoichiometry for the cation complex in polar solvent
as predicted by Wipff.25 For all ligands, the presence of only
one singlet for 9 and 10 or one triplet and one doublet for 11
and 15 corresponding to the signals of the phenolic protons
seemed to indicate either a rapid cation–ligand exchange or
2.2.4. NMR study. The ability of the receptors 9–12 and 15
to complex with Naþ and Csþ was also studied by the use of
1
Table 3. Changes in the H NMR aromatic shifts of 9–12 and 15 upon addition of CsNO3 in DMSO-d6
Compound
9
9,Csþ
10
10,Csþ
11
11,Csþ
12
12,Csþ
15
15,Csþ
H in meta
H in para
6.40
–
6.64
–
7.76
–
7.79
–
7.03
6.58
7.19
6.64
7.11–7.13
6.57–6.59
7.19
6.67
6.99
6.58
7.10
6.67

S. Pellet-Rostaing et al. / Tetrahedron 59 (2003) 10313–10324
10319
Figure 6. Caesium retention¼f([ligand]/[caesium]) ([caesium¼15 mg/L,
[NaNO3]¼0.1 mol/L, T¼293 K, DP¼0.06 MPa).
of permeate and retentate were taken half an hour after each
ligand addition.
Nanofiltration–complexation experiments were also per-
formed in the conditions of the solubility of each ligand.
This solubility was determined for each receptor before
testing them in the nanofiltration–complexation process.
Thus, water-soluble ligands 9, 10, 12 and 15 were
preliminary dissolved in a 0.1 M sodium nitrate solution
(50 mL) in the conditions of pH and temperature ensuring a
total solubility of the quantity of the ligands, required to
reach a ratio [ligand]/[caesium] of 5 (Table 4).
Figure 5. Schematic flow diagram of the lab-scale membrane system, 1:
feed tank, 2: pH meter, 3: heat exchanger, 4: pump, 5: cell body, 6:
membrane, 7: feed pressure gauge, 8: feed flow control valve, 9: retentate
flux meter and 10: permeate flux meter.
a symmetrical position in the aromatic ring of the ligand, as
suggested by Arnaud-Neu et al.48 for the calix[4]arene-bis-
crown-6-CsþI2 complex studied in deuterated acetonitrile.
2.2.5. Caesium–sodium separation by nanofiltration–
complexation. The water-soluble calixcrown derivatives
were subsequently used as specific ligands for caesium in a
nanofiltration–complexation process. Nanofiltration–com-
plexation tests were carried out to measure the ligand
ability in order to help improve the separation of caesium
from sodium by the nanofiltration process. In the range of
300–1000 g/mol molecular weights, nanofiltration mem-
brane ‘cut offs’ are located between ultrafiltration and
reverse osmosis. Nanomax 50 membrane allowed only 10%
retention of Naþ and Csþ ions. Then selective complexation
of caesium with a ligand of larger size and mass than the
target should increase the retention of this latter preventing
this ion from passing through the membrane. A schematic
representation of the nanofiltration (NF) loop used in our
experiments is given in Figure 5. By totally recycling the
permeate and the retentate, the feed remains at constant
composition during the experiment.
In the absence of ligand, the performances of the nanofiltration
membrane used (Nanomax 50 membrane) were not really
affected by the experimental conditions (pH and temperature)
andwerenotsufficient (Fig. 6). Witharatio[ligand]/[caesium]
equal to 5, caesium retention was respectively of 91, 92 and
93% for hydrosolublederivatives 9, 12and 15which showed a
goodaffinitytowardcaesiumaswellasthecompound1. These
were much higher than with carboxylic analogue 10 of which
retention rates varied identically until 50% with a [ligand]/
[caesium] molar ratio of 5.
Since the purpose of this work was to improve the separation
of traces of caesium contained in high concentrated sodium
salt media, a second experiment was performed to evaluate
and to compare the Csþ/Naþ selectivity of the ligands by
addition of known amounts of sodium nitrate to a solution
First, the selective caesium affinity of ligands was studied in
an aqueous medium (400 mL) containing low levels of
sodium nitrate. Several known amounts of a selected ligand
were progressively added to a solution containing fixed
concentrations of caesium (15 mg/L) and sodium nitrates
(0.1 mol/L) which was filtered on a NF membrane. Samples
Table 4. Experimental conditions of the solubility (pH and the temperature)
of 9, 10, 11, 15 in a solution of NaNO3 (50 mL, 0.1 M)
Ligands
135
9
10
12
15
[L] (g/L)
pH
T (K)
4.03
12
298
4.73
12.5
298
5.50
11
298
6.00
7
303
5.23
7
303
Figure 7. Caesium retention¼f([NaNO3]) ([caesium¼15 mg/L,
[ligand]/[caesium]¼5, T¼293 K, DP¼0.06 MPa).

10320
S. Pellet-Rostaing et al. / Tetrahedron 59 (2003) 10313–10324
containing a fixed ratio [ligand]/[caesium] of 5, which was
then filtered on the NF membrane.
or AM300 (chemical shifts in ppm, J in Hz). Mass
spectrometry was performed by electrospray technique,
positive and negative mode. Elemental analysis was
performed at Service Central d’Analyse, CNRS, Vernaison
(France).
WhenNaNO3 wasadded,theCsþ/Naþ selectivityofligands9,
12and 15decreased (Fig. 7). Infact, ina 1 mol/LNaNO3 basic
aqueous medium with a ratio [ligand]/[caesium] of 5, the
caesium retention by the NF membrane was still close to 80%.
Surprisingly, when the sodium nitrate concentration varied
from 1 to 4 mol/L, the experimental Csþ/Naþ selectivity of
these ligands decreased abnormally. Although the tests were
performed in closed loop and the caesium measuring out
should remain constant in the retentate, an unusual decrease of
the caesium concentration in the retentate was noticed. Thus,
with absorption values theoretically constant in the retentate,
caesium retention should rise to 76, 65, 60 and 57% for 9 at
sodium nitrate concentrations, respectively, of 1, 2, 3 and
4 mol/L. With experimental values of 75, 61, 52 and 46%, the
gradual precipitation both of the free ligands and the caesium
complex at the surface of the membrane could be suggested.
At [NaNO3].1 M, ligands settle on the surface of the
membrane involving the decreasing of the permeate flux
(generally from 46 L/h bar m2 at 1 M to 1.5 L/h bar m2 at
2 M). In fact, particle deposition is one of the fouling causes
from these decreases in permeate flux.49,50 With the use of
ligands 12 and 15, this phenomenon was amplified. Indeed,
experimental values at 3 mol/L in sodium nitrate were of 39%
although the theoretical retention rate should be close to 60%
for both ligands.
4.2. Synthesis
Solvents were commercial reagents and used without
further purification. Dissolute compound 445 was prepared
according to reported procedures. Vicens procedure was
used to prepare 5, 6 and 7. The synthesis of compounds 1
was described in a previous paper of water-soluble
calixarenes-bis-crown-6.34
4.2.1. Ethyl 3,4-bis[2-(2-hydroxyethoxy)ethoxy]benzo-
ate, 3. Ethyl 3,4-dihydroxybenzoate (20 g, 109 mmol) and
potassium carbonate (63 g, 450 mmol) were stirred at room
temperature in acetonitrile (540 mL) for 2 h. The mixture
was then heated to reflux and 2-(2-chloroethoxy)ethanol
(31.64 g, 254 mmol) was added. After 3 days, more
potassium carbonate (24 g, 171 mmol) and 2-(2-chloro-
ethoxy)ethanol (16.5 g, 132 mmol) were added and the
mixture was kept to reflux 4 more days. The solvent was
then evaporated and the residue dissolved in dichloro-
methane, washed successively with 1N HCl, water, brine
and dried over magnesium sulfate. Chromatography on
silica column (60/40 CH2Cl2–acetone) afforded pure diol 3
as an oil (12 g, 30%). 1H NMR spectrum (CDCl3) d (ppm)
1.37 (t, J¼7.2 Hz, 3H, OCH2CH3); 3.55 (s, 2H, OH); 3.66–
3.69 (m, 4H, OCH2CH2O); 3.74–3.77 (m, 4H, OCH2CH2O);
3.91–3.94 (m, 4H, OCH2CH2O); 4.18–4.21 (m, 4H,
OCH2CH2O); 4.33 (q, J¼6.6 Hz, 2H, OCH2CH3); 6.87 (d,
J¼8.8 Hz, ArH on C-5); 7.57 (d, J¼2.2 Hz, ArH on C-2);
7.66 (dd, J¼8.1, 2.2 Hz, ArH on C-6). ES-MS (ESþ): 397.3
([MþK]þ); 381.3 ([MþNa]þ).
Finally, when the concentration of sodium nitrate varied
from 0.1 to 4 mol/L, no precipitation of 10 occurred,
however the Csþ retention decreased rapidly until 16%,
suggesting a weak Csþ/Naþ selectivity.
3. Conclusion:
4.2.2. Ethyl 3,4-bis[2-[2-[[(4-methylphenyl)sulfonyl]-
oxy]ethoxy]ethoxy]benzoate, 5. To a solution of diol 3
(10.5 g, 29.30 mmol) and tosylchloride (27 g, 142 mmol) in
dichloromethane (300 mL) at 08C, triethylamine (14 g,
142 mmol) in dichloromethane (100 mL) was added drop-
wise. After stirring at room temperature for 6 days, water
(200 mL) was added. The organic layer was dried over
magnesium sulfate, filtered, concentrated and purified by
silica gel column chromatography (95/5 CH2Cl2–acetone)
The new synthesised host molecules 9–12 and 15 were
obtained by different synthetic routes from both calix-
(benzo)crown 6 and 7. Their binding efficiencies for alkali-
metal cations were determined by UV spectrophotometry in
methanol and aqueous media. The results show that the
efficiency of these hydrosoluble calix[4]benzocrowns-6
depends on the nature and number of grafted hydrophilic
groups. Every ligand shows a more or less pronounced affinity
for the studied alkali-metal cation, decreasingin all cases from
the larger (Csþ) to the smaller (Kþ). Apart from the ligand 10,
the hydrosoluble receptors achieve a good compromise
between preorganisation, steric hindrance and the cooperative
effect of cation–arene and/or cation–phenolate interactions.
Results concerning Csþ/Naþ separation by nanofiltration
combined with these organic compounds fit well with UV–
Vis analysis. Compounds 9, 12 and 15 showed high Csþ/Naþ
selectivity and allowed an efficient separation of traces of
caesium separation by nanofiltration–complexation from a
moderate salted medium ([NaNO3]¼85 g/L).
1
to give pure 5 (27 g, 86%) as an oil. H NMR spectrum
(CDCl3) d (ppm) 1.34 (t, J¼7 Hz, 3H, OCH2CH3); 2.34 (s,
6H, CH3); 3.70–3.77 (m, 8H, OCH2CH2O); 4.02–4.08 (m,
4H, OCH2CH2O); 4.11–4.15 (m, 4H, OCH2CH2O); 4.31 (q,
J¼7 Hz, OCH2CH3); 6.82 (d, J¼8.8 Hz, ArHbenzo); 7.23 (d,
J¼8.1 Hz, 4H, ArHtosyl); 7.48 (d, J¼1.8 Hz, ArHbenzo); 7.62
(dd, J¼8.5, 1.8 Hz, ArHbenzo); 7.72 (d, J¼8.1 Hz, 4H,
ArHtosyl). 13C NMR (CDCl3) d (ppm) 14.43, 21.59 (CH3);
60.83 (OCH2CH3); 68.60, 68.87, 68.92, 68.97, 69.47, 69.59,
69.70, (OCH2CH2O), 112.7, 114.9 (ArCH), 123.0, 144.9,
148.2, 152.8 (ArC); 166.2 (CO). ES-MS (ESþ): 397.3
([MþK]þ); 381.3 ([MþNa]þ).
4. Experimental
4.2.3. Calix[4]arene-bis-[(4-methyl-1,2-phenylene)-
crown-6], 6.
8.24 mmol), potassium carbonate (11.4 g, 82.4 mmol) and
ditosylate 4 (5.14 g, 8.24 mmol) in dry acetonitrile
A
mixture of calix[4]arene (3.5 g,
4.1. Analytical instruments
1H and 13C NMR spectra were recorded on Brucker AC200

S. Pellet-Rostaing et al. / Tetrahedron 59 (2003) 10313–10324
10321
(300 mL) was refluxed under Argon for 7 days. The same
quantities of potassium carbonate and 4 in acetonitrile
(50 mL) were added and reflux was continued for additional
10 days. After cooling to room temperature, the solvent was
removed in vacuo. Dichloromethane (400 mL) was added
and the suspension was neutralised by 1N HCl. The organic
layer was washed with water, brine and dried over
magnesium sulfate. The filtered solution was then concen-
trated and precipitated with methanol–diethylether to yield
(CDCl3) d (ppm) 2.31 (s, 6H, CH3); 3.39 (br t, 8H,
OCH2CH2O); 3.51 (br t, 8H, OCH2CH2O); 3.74 (s, 8H,
ArCH2Ar); 3.81 (m, 8H, OCH2CH2O); 4.15 (m, 8H,
OCH2CH2O); 6.81 (s, 2H, ArHbenzo); 7.08 (s, 2H, ArHbenzo);
7.21 (s, 8H, ArHcalix). 13C NMR (CDCl3) d (ppm) 22.75
(CH3); 37.52 (ArCH2Ar); 70.08, 70.10, 70.23, 70.26, 70.59,
70.70, 71.04, 71.10 (OCH2CH2O); 118.6, 120.1, 132.5
(ArCH), 115.9, 131.3, 136.7, 148.7, 149.5, 151.3 (ArC). ES-
MS (ESþ): 1449.1 ([MþNa]þ). Anal. calcd for
8
1
pure 6 as a white solid (5.18 g, 66%). H NMR spectrum
(C58H58O12Br6): C, 48.85; H, 4.10; O, 13.46; Br, 33.62.
Found: C, 48.76; H, 4.14; O, 13.85; Br, 33.50.
(CDCl3) d (ppm) 2.32 (s, 6H, CH3); 3.47–3.72 (m, 24H,
OCH2CH2O); 3.78 (s, 8H, ArCH2Ar); 4.07–4.13 (m, 8H,
OCH2CH2O); 6.69 (t, J¼7.5 Hz, 4H, ArHcalix); 6.79 (d,
J¼7.6 Hz, 2H, ArHbenzo); 6.80 (s, 2H, ArHbenzo); 6.88 (d,
J¼7.6 Hz, 2H, ArHbenzo); 7.10 (d, J¼7.5 Hz, 8H, ArHcalix).
13C NMR spectrum (CDCl3) d (ppm) 21.03 (CH3); 37.92
(ArCH2Ar); 69.93, 70.11, 70.18, 70.31, 70.37 (OCH2CH2O);
116.1, 120.4, 122.6, 130.0 (ArCH); 131.9, 134.1, 146.9, 149.2,
156.5 (ArC). ES-MS (ESþ): 975.7 ([MþNa]þ). Anal. calcd
for 6·0.5H2O (C58H65O12.5): C, 72.41; H, 6.80; O, 20.78.
Found: C, 72.23; H, 6.81; O, 20.21.
4.2.6. 5,11,17,23-Tetrahydroxycalix[4]arene-bis-[(4-
hydroxy-5-methyl-1,2-phenylene)crown-6], 9. To
a
solution of 8 (0.5 g, 0.35 mmol) in dry tetrahydrofuran
(100 mL) cooled at 2788C under argon was added tert-
butyllithium (8.4 mL of a 1.7 M solution in pentane,
14.28 mmol). After 2 h, trimethyl borate (2.55 g,
24.47 mmol) was added and the solution was allowed to
warm to room temperature. After 4 h under stirring, the
reaction mixture was cooled again at 2788C and 3 M
sodium hydroxide–hydrogen peroxide 35% (25 mL) was
added dropwise. The mixture was again warmed to room
temperature and stirred overnight. Sodium thiosulfate was
carefully added to the solution and cooled to 08C until
saturation. After evaporation of tetrahydrofuran, water
(50 mL) was added and the mixture was acidified with 2N
HCl until 6,pH,7. The resulting precipitate was filtered
off, rinsed with water and dissolved in methanol. Insoluble
impurities were removed by filtration. Filtrate was evapor-
ated to dryness and the residue dissolved in tetrahydrofuran.
Precipitation with diethylether and filtration afforded pure 9
(0.228 g, 62%) as a white powder. 1H NMR spectrum
(DMSO-d6) d (ppm) 2.01 (s, CH3); 3.16–4.00 (m, 32H,
OCH2CH2O and 8H, ArCH2Ar); 6.41 (br s, 2H, ArHbenzo
and 8H, ArHcalix); 6.70 (s, 2H, ArHbenzo); 8.69 (br s, 6H,
OHbenzo and OHcalix). 13C NMR (DMSO-d6) d (ppm) 15.54
(CH3); 37.48 (ArCH2Ar); 68.92, 69.50, 69.86, 70.15, 70.48
(OCH2CH2O); 103.8, 115.4, 120.39 (ArCH), 116.0, 134.2,
141.5, 148.2, 149.1, 150.3, 151.8 (ArC). ES-MS (ESþ):
1071.6 ([MþNa]þ). Anal. calcd for 9·H2O (C58H66O19): C,
65.28; H, 6.23; O, 28.49. Found: C, 65.33; H, 6.15; O, 28.86.
4.2.4. Calix[4]arene-bis-[(4-ethoxycarbonyl-1,2-phenyl-
ene)crown-6], 7. A mixture of calix[4]arene (3.5 g,
8.24 mmol), potassium carbonate (11.4 g, 82.4 mmol) and
ditosylate 5 (5.64 g, 8.42 mmol) in dry acetonitrile
(300 mL) was refluxed under Argon for 7 days. The same
quantities of potassium carbonate and 5 in acetonitrile
(50 mL) were added and reflux was maintained for
additional 7 days. After cooling to room temperature, the
solvent was removed under reduce pressure. Dichloro-
methane (200 mL) was added and the suspension was
neutralised by 1N HCl. The organic layer was washed with
water, brine and dried over magnesium sulfate. Pure 7
(3.6 g, 41%) was obtained as a white foam by purification
on chromatography column (SiO2, CH2Cl2–acetone 90/10).
1H NMR spectrum (CDCl3) d (ppm) 1.40 (t, J¼7.3 Hz, 6H,
OCH2CH3); 3.57–3.62 (m, 16H, OCH2CH2O); 3.74–3.78
(m, 16H, ArCH2Ar and OCH2CH2O); 4.18–4.20 (m, 8H,
OCH2CH2O); 4.37 (q, J¼7.3 Hz, 4H, OCH2CH3); 6.67 (t,
J¼7.3 Hz, 4H, ArHcalix); 6.97 (d, J¼8.5 Hz, 2H, ArHbenzo);
7.07 (d, J¼7.3 Hz, 8H, ArHcalix); 7.66 (d, J¼2.2 Hz, 2H,
ArHbenzo); 7.74 (d, J¼8.5 Hz, 2H, ArHbenzo). 13C NMR
spectrum (CDCl3) d (ppm) 14.47 (CH3); 37.78 (ArCH2Ar);
60.90 (OCH2CH3); 69.96, 69.99, 70.01, 70.07, 70.26, 70.44
(OCH2CH2O); 113.0, 115.6, 122.6, 124.3, 130.3 (ArCH);
123.8, 134.1, 148.4, 153.0, 156.4 (ArC); 166.4 (CO). ES-
MS (ESþ): 1107.5 ([MþK]þ); 1091.4 ([MþNa]þ). Anal.
calcd for 7·0.25CH2Cl2 (C62H68O16, 0.25 CH2Cl2): C,
68.56; H, 6.33; O, 23.47. Found: C, 68.64; H, 6.29; O,
23.29.
4.2.7. 5,11,17,23-Tetracarboxycalix[4]arene-bis-[(4-car-
boxyl-5-methyl-1,2-phenylene)crown-6], 10. To a stirred
solution of 8 (0.6 g, 0.42 mmol) dissolved in dry tetra-
hydrofuran (100 mL) at 2788C under argon was added tert-
butyllithium in pentane (12.3 mL 1.7 M, 20.9 mmol). The
resulting red solution was stirred at 2788C for 30 min and
quenched with a large excess of CO2 (g). HCl (10 mL, 6 M)
was added to the reaction mixture and tetrahydrofuran was
evaporated. The resulting precipitate was filtered, washed
with water and dried under reduce pressure at 508C
overnight. Precipitation in acetone and filtration afforded
pure 10 as a white powder (0.23 g, 45%). 1H NMR spectrum
(DMSO-d6) d (ppm) 2.49 (s, 6H, CH3); 3.01 (br t, 8H,
OCH2CH2O); 3.49–3.57 (m, 16H, OCH2CH2O); 3.62–4.01
(m, 8H, OCH2CH2O and 8H, ArCH2Ar); 6.86 (s, 2H,
ArHbenzo); 7.40 (s, 2H, ArHbenzo); 7.76 (s, 8H, ArHcalix);
12.54 (s, 6H, COOH). 13C NMR (DMSO-d6) d (ppm) 21.18
(CH3); 36.78 (ArCH2Ar); 68.01, 68.83, 69.29, 69.90, 70.00
(OCH2CH2O); 117.6, 119.5, 130.61 (ArCH), 122.1, 124.8,
133.4, 135.1, 145.9, 152.5, 160.0 (ArC); 167.1, 167.9
4.2.5. 5,11,17,23-Tetrabromocalix[4]arene-bis-[(4-
bromo-5-methyl-1,2-phenylene)crown-6], 8. Compound
6 (2 g, 2.09 mmol) was stirred in dichloromethane (350 mL)
at 08C. Bromine (6.4 g, 40 mmol) in dichloromethane
(20 mL) was added. After 5 h at room temperature,
additional dichloromethane (200 mL) was added and the
organic layer was washed with a solution of sodium
metabisulfite 10%, water and brine, then dried over
magnesium sulfate. Evaporation of solvent and purification
by chromatography (SiO2, CH2Cl2–MeOH 98/2) afforded
1
pure 8 (2.18 g, 75%) as a white foam. H NMR spectrum

10322
S. Pellet-Rostaing et al. / Tetrahedron 59 (2003) 10313–10324
(COOH). ES-MS (ESþ): 1241.1 ([MþNa]þ). Anal. calcd
for 10·2H2O (C64H62O26): C, 61.64; H, 5.01; O, 33.35.
Found: C, 61.95; H, 4.83; O, 33.46.
NaOH 15% (0.5 mL) and an additional quantity of water
(1.5 mL). After 15 min under stirring, celite (5 g) was added
and the mixture was stirred for 30 min. After filtration, the
solvent was removed under reduced pressure and the residue
chromatographed on silica gel (CH2Cl2–MeOH, 90/10) to
give pure 13 (0.23 g, 80%) at a white foam. 1H NMR
spectrum (DMSO-d6) d (ppm) 3.28–3.35 (m, 8H, OCH2-
CH2O and 2H, OH); 3.46–3.49 (m, 8H, OCH2CH2O); 4.1–
4.2 (m, 8H, OCH2CH2O); 3.56–3.59 (m, 8H, OCH2CH2O);
3.73 (s, 8H, ArCH2Ar); 4.01–4.03 (m, 8H, OCH2CH2O);
4.43 (s, 4H, CH2OH); 6.58 (t, J¼7.7 Hz, 4H, ArHcalix); 6.87
(d, J¼8.2 Hz, 2H, ArHbenzo); 6.96 (d, J¼8.2 Hz, 2H,
ArHbenzo); 6.98 (s, 2H, ArHbenzo); 7.02 (d, J¼7.4 Hz, 8H,
ArHcalix). 13C NMR spectrum (DMSO-d6) d (ppm) 36.92
(ArCH2Ar); 62.66 (ArCH2OH); 69.21, 69.42 (OCH2CH2O);
113.8, 115.2, 119.5, 121.9, 129.5 (ArCH); 133.4, 136.1,
147.4, 148.6, 156.1 (ArC). ES-MS (ESþ): 1023.7
([MþK]þ); 1007.7 ([MþNa]þ). Anal. calcd for 13
(C58H64O14): C, 70.72; H, 6.55; O, 22.73. Found: C,
70.34; H, 6.48; O, 23.19.
4.2.8. Calix[4]arene-bis-[(4-carboxyl-1,2-phenylene)-
crown-6], 11. A mixture of 7 (0.4 g, 0.38 mmol) and
potassium hydroxide (0.084 g, 1.52 mmol) in ethanol
(15 mL) was stirred at reflux for 3 h. After cooling to
room temperature, water (15 mL) was added followed by
concentrated HCl (1.5 mL) and water (15 mL). The
resulting white precipitate was filtered off, rinsed with
water and dissolved in chloroform. The solution was dried
over magnesium sulfate and evaporated to dryness to give
pure 11 as a white powder (0.346 g, 90%). 1H NMR
spectrum (DMSO-d6) d (ppm) 3.32 (br s, 8H, OCH2CH2O);
3.43–3.48 (m, 8H, OCH2CH2O); 3.62–3.74 (m, 8H,
ArCH2Ar and 8H, OCH2CH2O); 4.1–4.2 (m, 8H, OCH2-
CH2O); 6.57 (t, J¼7.4 Hz, 4H, ArHcalix); 7.03 (d, J¼7.4 Hz,
8H, ArHcalix); 7.13 (d, J¼8.1 Hz, 2H, ArHbenzo); 7.52 (s, 2H,
ArHbenzo); 7.63 (d, J¼8.5 Hz, 2H, ArHbenzo); 12.71 (s, 2H,
COOH). 13C NMR spectrum (DMSO-d6) d (ppm) 36.81
(ArCH2Ar); 68.90, 69.12, 69.34, 69.52, 69.76 (OCH2CH2O);
113.2, 115.1, 121.9, 123.5, 129.7 (ArCH); 133.7, 147.7, 152.5,
156.0, 156.1 (ArC); 166.9(CO). ES-MS (ESþ): 1035.5
([MþNa]þ). Anal. calcd for 11 (C58H60O16): C, 68.83; H,
5.98; O, 25.29. Found: C, 68.34; H, 5.90; O, 24.98.
4.2.11. Calix[4]arene-bis-[(4-bromomethyl-1,2-phenyl-
ene)crown-6], 14. Phosphorus tribromide (150 mL,
1.62 mmol) was added to a stirred solution of 13 (0.53 g,
0.54 mmol) in dichloromethane (25 mL), cooled at 08C.
After 5 h of vigorous stirring, the reaction mixture was
quenched with 50 mL of ice water. The organic layer was
separated and the aqueous layer was further extracted with
dichloromethane (50 mL). The combined extracts were
washed with brine, dried over magnesium sulphate, and
evaporated under reduced pressure. The residue was
chromatographed on silica gel (CH2Cl2–MeOH, 95/5) to
4.2.9. Calix[4]arene-bis-[[(4-(2-sulfooxy)ethylester)-1,2-
phenylene]crown-6], dipotassium salt, 12. To a suspen-
sion of 11 (0.2 g, 0.197 mmol) and potassium carbonate
(0.11 g, 0.788 mmol) in dry dichloromethane (50 mL) under
argon was added 1,2,3-dioxathiolane-2,2-dioxide (0.1 g,
0.788 mmol). The mixture was heated at reflux for 48 h and
then cooled to room temperature. The resulting precipitate
was filtered off, washed with dichloromethane and triturated
in dimethylsulfoxide. After filtration of insoluble salts,
acetone was added to the filtrate and the resulting white
precipitate was filtered off, washed with acetone and dried
under vacuo to give pure 12 as a white powder (0.187 g,
1
give pure 14 (0.54 g, 90%) as a white powder. H NMR
spectrum (CDCl3) d (ppm) 3.44–3.61 (m, 16H, OCH2CH2O);
3.66–3.73 (m, 8H, OCH2CH2O); 3.79 (s, 8H, ArCH2Ar);
4.14–4.16 (m, 8H, OCH2CH2O); 4.52 (s, 4H, CH2Br); 6.68 (t,
J¼7.4 Hz, 4H, ArHcalix); 6.91–7.09 (m, 6H, ArHbenzo and 8H,
ArHcalix). 13C NMR spectrum (CDCl3) d (ppm) 34.12
(CH2Br); 37.86 (ArCH2Ar); 69.88, 69.97, 70.17, 70.25
(OCH2CH2O); 114.9, 116.3, 122.6, 122.7, 130.2 (ArCH);
131.4, 134.1, 149.1, 149.4, 156.4 (ArC). ES-MS (ESþ):
1050.3 ([MþK]þ); 1033.2 ([MþNa]þ). Anal. calcd for 14
(C58H62O12Br2): C, 68.91; H, 6.18; O, 18.99; Br, 15.81.
Found: C, 68.96; H, 6.16; O, 19.24; Br, 15.84.
1
71%). H NMR spectrum (DMSO-d6) d (ppm) 3.52 (br t,
8H, OCH2CH2O); 3.63 (br t, 8H, OCH2CH2O); 3.70–3.74
(m, 8H, OCH2CH2O); 3.81 (s, 8H, ArCH2Ar); 4.03 (t,
J¼5.2 Hz, 4H, OCH2CH2OSO3K); 4.13–4.20 (m, 16H,
OCH2CH2O); 4.39 (t, J¼5.2 Hz, 4H, OCH2CH2OSO3K);
6.56 (t, J¼7.7 Hz, 2H, ArHcalix); 6.59 (t, J¼7.7 Hz, 2H,
ArHcalix); 7.11 (br d, 4H, ArHcalix); 7.13 (br d, 4H, ArHcalix);
7.18 (d, J¼8.4 Hz, 2H, ArHbenzo); 7.65 (s, 2H, ArHbenzo);
7.70 (d, J¼8.4 Hz, 2H, ArHbenzo). 13C NMR spectrum
(DMSO-d6) d (ppm) 36.39 (ArCH2Ar); 63.8, 63.84, 68.53,
68.71, 69.34, 69.60, 70.71, 70.76 (OCH2CH2O); 112.7,
114.2, 122.1, 122.4, 123.9, 130.1 (ArCH); 134.1, 134.2,
147.6, 152.0, 156.1 (ArC). ES-MS (ESþ): 1375.2
([MþK]þ); 1359.2 ([MþNa]þ). Anal. calcd for 12·H2O
(C62H68O25S2K2): C, 54.95; H, 5.06; O, 29.52; S, 4.73.
Found: C, 54.59; H, 5.10; O, 29.35; S, 5.18.
4.2.12. Calix[4]arene-bis-[[4-((N,N-diethanol)amino-
methyl)-1,2-phenylene]crown-6], 15. Diethanolamine
(0.382 g, 0.36 mmol) in dry dimethylsulfoxide (5 mL) was
added to a stirred solution of 14 (0.2 g, 0.18 mmol) in
dimethylsulfoxide (20 mL). The mixture was heated at 408C
under argon for 48 h. The solvent was evaporated under
reduced pressure and the residue purified by chromato-
graphy on silica gel (CH2Cl2–MeOH–Et3N 90/5/5) to give
pure 15 (0.208 g, 50%). 1H NMR spectrum (CDCl3) d (ppm)
2.68 (t, J¼5.1 Hz, 8H, NCH2CH2OH); 3.43–3.45 (m, 8H,
OCH2CH2O); 3.49–3.51 (m, 8H, OCH2CH2O); 3.55–3.59
(m, 8H, OCH2CH2O and 8H, NCH2CH2OH); 3.64 (s, 4H,
ArCH2N); 3.69 (s, 8H, ArCH2Ar); 4.01–4.05 (m, 8H,
OCH2CH2O); 5.28 (bs, 4H, OH); 6.60 (t, J¼7.4 Hz, 4H,
ArHcalix); 6.84 (s, 2H, ArHbenzo); 6.96–7.00 (m, 4H,
ArHbenzo and 8H, ArHcalix). 13C NMR spectrum (CDCl3) d
(ppm) 37.82 (ArCH2Ar); 55.77, 58.96, 59.44 (NCH2CH2-
OH and ArCH2N); 69.94, 70.10, 70.22 (OCH2CH2O);
4.2.10. Calix[4]arene-bis-[(4-hydroxymethyl-1,2-phenyl-
ene)crown-6], 13. Under an atmosphere of argon, a solution
of 7 (0.31 g, 0.289 mmol) in dry dimethoxyethane (20 mL)
was added dropwise to a suspension of LiAlH4 (0.46 g,
12.1 mmol) in dimethoxyethane (30 mL). The mixture was
heated at reflux for 6 h. The excess of LiAlH4 was destroyed
by addition of water (0.5 mL), followed by addition of

S. Pellet-Rostaing et al. / Tetrahedron 59 (2003) 10313–10324
10323
¨
5. Yamamoto, H.; Shinkaı, S. Chem. Lett. 1994, 1115–1117.
6. Ghidini, E.; Ugozzoli, F.; Ungaro, R.; Harkema, S.; Abu
El-Fadl, A.; Reinhoudt, D. N. J. Am. Chem. Soc. 1990, 112,
6979–6985.
115.2, 116.3, 122.4, 122.5, 130.1 (ArCH); 132.2, 134.0,
148.3, 149.1, 156.4 (ArC). ES-MS (ESþ): 1198.6
([MþK]þ); 1181.1 ([MþNa]þ); 1159.8 ([MþH]þ). Anal.
calcd for 15·H2O (C66H84O17N2): C, 67.32; H, 7.19; N, 2.38.
Found: C, 67.88; H, 7.25; N, 2.52.
7. Brzozka, Z.; Lammerink, B.; Reinhoudt, D. N.; Ghidini, E.;
Ungaro, R. J. Chem. Soc., Perkin Trans. 2 1993, 1037–1041.
8. Ungaro, R.; Casnati, A.; Ugozzoli, F.; Pochini, A.; Dozol,
J.-F.; Hill, C.; Rouquette, H. Angew. Chem., Int. Ed. Engl.
1994, 33, 1506–1509.
4.3. Stability constant measurement
UV–Vis studies were carried out in methanol and an
aqueous medium containing 0.1 M NaOH at 258C. Spectral
changes of calixarenes solutions (2–3 mL) upon stepwise
addition of alkali-metal salt into the measurement cell were
recorded from 200 to 350 nm with a HP 8453 spectro-
photometer using 1 cm path length quartz vessel. Chloride
salts (KCl, RbCl, CsCl, Acros, p.a.) were preferred to nitrate
salts to prevent NO23 overlapping of the ligand spectrum by
the broad NO23 band.
9. Ungaro, R.; Casnati, A.; Ugozzoli, F.; Pochini, A.; Dozol,
J.-F.; Hill, C.; Rouquette, H. Angew. Chem., Int. Ed. Engl.
1994, 106, 1551–1553.
10. Casnati, A.; Pochini, A.; Ungaro, R.; Ugozzoli, F.; Arnaud, F.;
Fanni, S.; Schwing, M.-J.; Egberink, R. J. M.; De Jong, F.;
Reinhoudt, D. N. J. Am. Chem. Soc. 1995, 117, 2767–2777.
11. Kim, J. S.; Cho, M. H.; Yu, I. Y.; Pang, J. H.; Kim, E. T.; Suh,
I. H.; Oh, M. B.; Ra, D. Y.; Cho, N. S. Bull. Kor. Chem. Soc.
1997, 18, 677–680.
4.4. Nanofiltration tests
12. Sachleben, R. A.; Urvoas, A.; Bryan, J. C.; Haverlock, T. J.;
Hay, B. P.; Moyer, B. A. Chem. Commun. 1999, 17,
1751–1752.
Apparatus. A SEPA CF Membrane Cell (OSMONICS) was
used with Nanomax 50 plane membrane (Millipore) which
presents a surface area of 0.015 m2, designed for tangential
filtration. The retention (%) of a substance i was calculated
as Ri¼100(12Cip/Cir) where Cip is the concentration of i in
the permeate and Cir the concentration in the retentate. For
all the NF results, sodium retention remained under 10%
because the Nanomax membrane was chosen to be very
permeable to monovalent ions. Tests were carried out using
a transmembrane pressure of 0.06 MPa and a temperature of
293 or 333 K.
´
13. Guillon, J.; Leger, J.-M.; Sonnet, P.; Jarry, C.; Robba, M. J.
J. Org. Chem. 2000, 65, 8283–8289.
14. Casnati, A.; Sansone, F.; Dozol, J.-F.; Rouquette, H.; Arnaud-
Neu, F.; Byrne, D.; Fuangswasdi, S.; Schwing-Weill, M.-J.;
Ungaro, R. J. Incl. Phenom. 2001, 41, 193–200.
15. Asfari, Z.; Harrofield, J. M.; Sobolev, A. N.; Vicens, J. Aust.
J. Chem. 1994, 47, 757–762.
16. Saadioni, M.; Asfari, Z.; Vicens, J. Tetrahedron Lett. 1997, 38,
1187–1190.
´
17. Saadioni, M.; Asfari, Z.; Thuery, P.; Nierlich, M.; Vicens, J.
Tetrahedron Lett. 1997, 38, 5643–5646.
18. Asfari, Z.; Bressot, C.; Vicens, J.; Miel, C.; Dozol, J.-F.;
Rouquette, H.; Eymard, S.; Lamare, V.; Tournois, B. Anal.
Chem. 1995, 67, 3133–3139.
Chemicals and reagents. The metallic salts used for the NF
tests were nitrates NaNO3 (99%, Aldrich) and CsNO3
(99.99%, Aldrich). Ligands were synthesised as described
above.
19. Asfari, Z.; Lamare, V.; Dozol, J.-F.; Vicens, J. Tetrahedron
Lett. 1999, 40, 691–694.
Analytical. Caesium concentration was determined by
atomic absorption spectroscopy (AAS) in an air–acetylene
flame with a PU 9100X PHILIPS atomic absorption
spectrometer. Each result is assumed to have a 5% standard
deviation. Standards for AAS analysis were made of pure
Csþ (caesium standards for ICP-AES analysis) and
dissolved in the corresponding NaNO3 and NaOH aqueous
matrix.
20. Groenen, L. C.; Brunink, J. A. J.; Baker, W. I. I.; Harkema, S.;
Wijmenga, S. S.; Reinhoudt, D. N. J. Chem. Soc., Perkin
Trans. 2 1992, 1899–1906.
21. Ji, H.-F.; Finot, E.; Dabestani, R.; Thundat, T.; Brown, G. M.;
Britt, P. F. Chem. Commun. 2000, 457–458.
22. Ji, H.-F.; Dabestani, T.; Brown, G. M.; Sachleben, R. A. Chem.
Commun. 2000, 833–834.
23. Asfari, Z.; Naumann, C.; Vicens, J. New J. Chem. 1996, 20,
1183–1194.
24. Lamare, V.; Dozol, J.-F.; Ugozzoli, F.; Casnati, A.; Ungaro, R.
Eur. J. Org. Chem. 1998, 1559–1568.
References
25. Varnek, A.; Wipff, G. J. Mol. Struct., Theochem 1996, 363,
67–85.
1. Casnati, A.; Ungaro, R.; Asfari, Z.; Vicens, J. Crown Ether
Derived from Calix[4]Arenes. In Calixarenes; Asfari, Z.,
26. Lauterbach, M.; Wipff, G.; Mark, A.; Van Gunsteren, W. F.
Gazz. Chim. Ital. 1997, 127, 699–709.
¨
Bohmer, V., Harrowfield, J., Vicens, J., Eds.; Kluwer
Academic: Dordrecht, The Netherlands, 2001.
27. Lauterbach, M.; Engler, E.; Muzet, N.; Troxler, L.; Wipff, G.
J. Phys. Chem. B 1998, 102, 245–256.
28. Lamare, V.; Dozol, J.-F.; Fuangswasdi, S.; Arnaud-Neu, F.;
2. Gutsche, C. D. In Calixarenes, Monographs in Supra-
molecular Chemistry; Stoddart, J. F., Ed.; The Royal Society
of Chemistry: Cambridge, 1989.
´
Thuery, P.; Nierlich, M.; Asfari, Z.; Vicens, J. J. Chem. Soc.,
Perkin Trans. 2 1999, 271–278.
´
29. Tuery, P.; Nierlich, M.; Lamare, V.; Dozol, J.-F.; Asfari, Z.;
3. Gutsche, C. D. In Calixarenes Revisited, Monographs in
Supramolecular Chemistry; Stoddart, J. F., Ed.; The Royal
Society of Chemistry: London, UK, 1998.
Vicens, J. J. Incl. Phenom. 2000, 36, 375–408.
30. Radioactive Waste Management and Disposal; Cecille, L.,
Ed.; Elsevier: New York, 1991.
4. Gokel, G. Crown Ethers and Cryptands. In Monographs in
Supramolecular Chemistry No. 3; Stoddart, J. F., Ed.; The
Royal Society of Chemistry: Cambridge, 1991.
´
31. Gaubert, E.; Barnier, H.; Nicod, L.; Favre-Reguillon, A.; Foos,

10324
S. Pellet-Rostaing et al. / Tetrahedron 59 (2003) 10313–10324
J.; Guy, A.; Bardot, C.; Lemaire, M. Sep. Sci. Technol. 1997,
32(14), 2309–2320.
41. Asfari, Z.; Lamare, V.; Dozol, J.-F.; Vicens, J. Tetrahedron
Lett. 1999, 40, 691–694.
32. Gaubert, E.; Barnier, H.; Maurel, A.; Foos, J.; Guy, A.; Bardot,
C.; Lemaire, M. Sep. Sci. Technol. 1997, 32(1–4), 585–597.
33. Nicod, L.; Pellet-Rostaing, S.; Chitry, F.; Lemaire, M.;
Barnier, H.; Federici, V. Tetrahedron Lett. 1998, 39,
9443–9446.
42. Haverlock, T. J.; Bonnesen, P. V.; Sachleben, R. A.; Moyer,
B. A. J. Incl. Phenom. 2000, 36, 21–37.
43. Casnati, A.; Sansone, F.; Dozol, J.-F.; Rouquette, H.; Arnaud-
Neu, F.; Byrne, D.; Fuangswasdi, S.; Schwing-Weill, M.-J.;
Ungaro, R. J. Incl. Phenom. 2001, 41, 193–200.
34. Pellet-Rostaing, S.; Chitry, F.; Nicod, L.; Lemaire, M. J. Chem.
Soc., Perkin Trans. 2 2001, 1426–1432.
44. Abibi, R.; Asfari, Z.; Harrowfield, J. M.; Nauman, C.;
Sobolev, A. N.; Vicens, J. An. Quim. Int. Ed. 1996, 92, 51–56.
´
45. Draye, M.; Favre-Reguillon, A.; Chomel, R.; Faure, R.; Guy,
35. Chitry, F.; Pellet-Rostaing, S.; Nicod, L.; Gass, J.-L.; Foos, J.;
Guy, A.; Lemaire, M. J. Phys. Chem. A 2000, 104,
4121–4128.
A.; Foos, J.; Lemaire, M. Bull. Soc. Chim. Fr. 1996, 133,
183–197.
´
36. Mathieu, A.; Asfari, Z.; Tuery, P.; Nierlich, M.; Faure, S.;
46. Jaime, C.; Mendoza, J.; Prados, P.; Nieto, P. M.; Sanchez, C.
J. Org. Chem. 1991, 56, 3372–3376.
Vicens, J. J. Incl. Phenom. 2001, 40, 173–181.
37. Talanov, V. S.; Talanova, G. G.; Bartsch, R. A. Tetrahedron
Lett. 2000, 41, 8221–8224.
47. Foster, R.; Hammick, D. L.; Wardley, A. A. J. Chem. Soc.
1953, 3817–3820.
38. Talanov, V. S.; Talanova, G. G.; Gorbunova, M. G.; Bartsch,
R. A. J. Chem. Soc., Perkin Trans. 2 2002, 209–215.
39. Haverlock, T. J.; Bonnesen, P. V.; Sachleben, R. A.; Moyer,
B. A. Radiochim. Acta 1997, 76, 103–108.
48. Arnaud-Neu, F.; Asfari, Z.; Souley, B.; Vicens, J. New
J. Chem. 1996, 20, 453–463.
49. Seidel, A.; Elimelech, M. J. J. Membr. Sci. 2002, 23, 245–255.
50. Zhao, Y.; Zhong, J.; Li, H.; Xu, N.; Shi, J. J. Membr. Sci. 2002,
208, 331–341.
40. Kim, J. S.; Yu, I. Y.; Pang, J. H.; Kim, J. K.; Lee, Y. I.; Lee,
K. W.; Oh, W.-Z. Microchem. J. 1998, 58, 225–235.