Anion recognition by phosphonium calix[4]arenes: Synthesis and physico-chemical studies

Article abstract of DOI:10.1080/10610270903437051

p-tert-Butylcalix[4]arenes, in the cone conformation, di- and tetra-substituted at the narrow rim with charged phosphonium groups, have been synthesised and characterised. Their interactions with a wide range of anions have been investigated in chloroform and acetonitrile solutions by means of ¹H and ³¹P NMR and isothermal titration microcalorimetry. These compounds have also been incorporated as sensing material in poly(vinyl chloride) ion-selective electrodes. The results showed that they interact strongly with the more lipophilic anions ClO^-, SCN^- and I^-, in solution as in the electrode membranes. The origin of this selectivity is discussed and, in particular, the role of the salt counterion is examined.

Full text of DOI:10.1080/10610270903437051

This article was downloaded by: [University of Chicago Library]
On: 25 December 2014, At: 16:36
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK
Supramolecular Chemistry
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/gsch20
Anion recognition by phosphonium calix[4]arenes:
synthesis and physico-chemical studies
Radoslaw Pomecko a b , Zouhair Asfari a , Véronique Hubscher-Bruder a , Maria Bochenska b &
Françoise Arnaud-Neu a
a IPHC-DSA, ULP, CNRS, ECPM , 25 rue Becquerel, 67087, Strasbourg Cedex 2, France
b Department of Chemical Technology , Chemical Faculty, Gdansk University of Technology ,
ul. Narutowicza 11/12, 80-264, Gdansk, Poland
Published online: 18 Feb 2010.
To cite this article: Radoslaw Pomecko , Zouhair Asfari , Véronique Hubscher-Bruder , Maria Bochenska & Françoise Arnaud-
Neu (2010) Anion recognition by phosphonium calix[4]arenes: synthesis and physico-chemical studies, Supramolecular
Chemistry, 22:5, 275-288, DOI: 10.1080/10610270903437051
To link to this article: http://dx.doi.org/10.1080/10610270903437051
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

Supramolecular Chemistry
Vol. 22, No. 5, May 2010, 275–288
Anion recognition by phosphonium calix[4]arenes: synthesis and physico-chemical studies
ab
Radoslaw Pomecko , Zouhair Asfari , Veronique Hubscher-Bruder , Maria Bochenska * and Franc¸oise Arnaud-Neu *
a
a
b
a
´
b
aIPHC-DSA, ULP, CNRS, ECPM, 25 rue Becquerel, 67087 Strasbourg Cedex 2, France; Department of Chemical Technology,
Chemical Faculty, Gdansk University of Technology, ul. Narutowicza 11/12, 80-264 Gdansk, Poland
(Received 24 July 2009; final version received 21 October 2009)
p-tert-Butylcalix[4]arenes, in the cone conformation, di- and tetra-substituted at the narrow rim with charged phosphonium
groups, have been synthesised and characterised. Their interactions with a wide range of anions have been investigated in
chloroform and acetonitrile solutions by means of 1H and 31P NMR and isothermal titration microcalorimetry. These
compounds have also been incorporated as sensing material in poly(vinyl chloride) ion-selective electrodes. The results
showed that they interact strongly with the more lipophilic anions ClO24 , SCN2 and I2, in solution as in the electrode
membranes. The origin of this selectivity is discussed and, in particular, the role of the salt counterion is examined.
Keywords: phosphonium calix[4]arenes; anion binding properties; microcalorimetry; ion-selective electrodes
Introduction
grafted. Calixpyrroles are also known as efficient anion
receptors (20, 21).
Anions play an important role in many biological
processes, such as regulation of cell activity, synthesis of
proteins and transport of hormones (1–3). They are also
frequently used in many industrial technologies, which
very often generate an increase of the concentration of
anions in the environment or even introduce anionic
species which were unknown so far in ecosystems. The
presence of these anions is crucial in environmental and
medical concerns, as they are pollutants and may have
harmful effects on living organisms and human health
(4–6). Therefore, there is a need of fast and selective anion
detection methods allowing real-time monitoring of anion
concentration changes and of efficient clean up processes.
Design of anion receptors for such applications remains a
great challenge for chemists because they have to take into
consideration the specific anion properties, such as a large
range of shapes and geometries, small electric charges vs.
sizes, high free energies of solvation and, in some cases,
multiple oxidation states of the central atoms in oxoanions
or pH dependence. In many artificial anion hosts, non-
covalent interactions are responsible for host–guest
recognition. They include electrostatic interactions,
hydrogen bonding, hydrophobic effects and coordination
to a metal ion or combinations of these interactions. The
hosts can be neutral, containing urea (7–10), thiourea (11,
12) or amide functions (13). They can also be positively
charged, containing pyridinium (14), polyammonium (15)
or quaternary ammonium (16) binding sites. Calix[4]ar-
enes (17, 18) and porphyrins (19) are often used as
scaffolds into which these functional groups can be
Recently, we described the synthesis and characteris-
ation of a new calix[4]arene derivative (5) bearing four
positively charged triphenyl phosphonium groups (22).
The presence of these highly polarisable moieties, where
the charge is spread over the three aromatic rings, was
expected to favour the interaction with lipophilic anions.
Preliminary binding studies showed that this ligand
interacted selectively with some anions, namely ClO42
and SCN2. This compound was also incorporated, as
ionophore-sensing material, in ion-selective electrodes
(ISEs) which exhibited a selectivity order similar to the
Hofmeister series. The disubstituted derivative 1 was also
synthesised as the hexafluorophosphate (23).
In order to get more information on the mechanisms
involved in the recognition process and to optimise its
selectivity, we have now extended this study to new di-
and tetra-substituted phosphonium calix[4]arene deriva-
tives (compounds 2–4, 6 and 7) and re-examined the
properties of 1 and 5 (Figure 1). In some of these
compounds, one of the phenyl rings on the phosphonium
groups has been replaced by a methyl radical or a
hydrogen atom. The presence of such small substituents is
expected to increase the charge density on the phosphorus
atoms and their accessibility (24). Moreover, the presence
of hydrogen atoms may induce the formation of hydrogen
bonds with anions. The possibility of tuning the charge
density on the phosphorus atoms by changing the nature
of their substituents could allow the design of receptors
able to distinguish lipophilic anions such as ClO24 , SCN2,
I2 or NO23 from other anions and between them. The
*Corresponding authors. Email: farnaud@chimie.u-strasbg.fr; marboch@chem.pg.gda.pl
ISSN 1061-0278 print/ISSN 1029-0478 online
q 2010 Taylor & Francis
DOI: 10.1080/10610270903437051
http://www.informaworld.com

276
R. Pomecko et al.
O
O
O
OH
OH
O
O
O
O
R
O
O
O
O
R
R
P+
R
R
R
P+
R
P+
R
P+
R
P+
R
P+
R'
R
P+
R'
R
+
R
R
P
R
R
R
R
P+
R'
R'
R'
R'
R'
R
R'
2Br–
2Br–
Br–
4A
1: R, R’ = Ph
2: R = Ph, R’ = Me
3: R, R’ = Ph
4: R = Ph, R’ = Me
8: R = Ph
5: R, R’ = Ph, A = Br–
5a: R, R’ = Ph, A = ClO4
–
–
5b: R, R’ = Ph, A = PF6
6: R = Ph R’ = Me, A = Br–
6a: R = Ph R’ = Me, A = ClO4
–
–
6b: R = Ph R’ = Me, A = PF6
7: R = Ph R’ = H, A = Br–
Figure 1. Chemical structures of the ligands under study.
binding properties of these compounds towards a variety
of anions have been followed by 1H and 31P NMR and by
isothermal titration microcalorimetry. In particular, the
role of the salt counterion was examined using these
techniques.
The more lipophilic molecules 3 and 4, where the free
phenolic protons are substituted with propyl groups, were
prepared in order to increase their stability in the lipophilic
membrane of ISEs. The reaction of S3 with 10 equivalents
of 1,4-dibromobutyl, refluxed for 4 days in dimethylfor-
mamide in the presence of 7 equivalents of NaH, gave 1,3-
bis-(propoxy)-2,4-bis-(butoxy-4-bromide)-tetrakis-p-tert-
butylcalix[4]arene (S4) in 57% yield. Compounds 3 and 4
were obtained by the reaction of S4 with 10 equivalents of
triphenylphosphine and 10 equivalents of diphenylmethyl-
phosphine in chloroform in 76 and 64% yield, respectively
(Scheme 1).
Results and discussion
Synthesis
The ligands synthesised in this work (Figure 1) are based
on a tetrakis-p-tert-butylcalix[4]arene platform which
presents several advantages. It has a well-defined size and
is readily amenable to substitution at its lower rim where
ligating groups can be attached and in some extent pre-
organised. The amphiphilicity of such derivatives should
allow their introduction into the membranes of ISEs.
Diphosphonium ligands were synthesised in two or
three steps according to the known procedure (23). The
first step was the selective bromoalkylation of the tetrakis-
p-tert-butylcalix[4]arene (S1), leading to the intermediate
molecules S2 or S3 in 71 and 55% yield, respectively
(Scheme 1). 1,3-Bis-(4-triphenylphosphonium-butoxy)-p-
tert-butyl-calix[4]arene dibromide (1) was obtained by the
reaction of S2 with 10 equivalents of triphenylphosphine.
After 6 days under reflux in chloroform, the product was
precipitated from a dichloromethane–hexane mixture in
76% yield. 1,3-Bis-(4-(P,P-diphenyl-P-methylphospho-
nium)-butoxy)-p-tert-butylcalix[4]arene dibromide (2)
was obtained during the reaction of S2 with 10 equivalents
of diphenylmethylphosphine in the same conditions in
69% yield.
Tetraphosphonium ligands 6 and 7 were synthesised
according to the procedure already described for 5 (22)
from the intermediate molecule tetrakis-(butoxy-4-bro-
mide)-p-tert-butylcalix[4]arene (S5) by substitution of
the bromine atoms of the alkyl chains (Scheme 2).
The tetrakis-(4-(P,P-diphenyl-P-methylphosphonium)-
butoxy)-p-tert-butylcalix[4]arene tetrabromide (6) was
obtained from the reaction of S5 with 20 equivalents of
diphenylmethylphosphine in chloroform in 59% yield.
Ligand 7 was synthesised in three steps. The reaction
of S5 with 5 equivalents of KPPh2 gave the product S6 in
40% yield (25–27). S6 was then protonated with an excess
of HBr giving ligand 7 in 52% yield.
In order to study the influence of the ligand counterion
on the ligand–anion interactions, the tetra-substituted
phosphonium ligands were also synthesised as perchlor-
ates (5a and 6a) and hexafluorophosphates (5b and 6b).
These compounds were obtained by the reaction of ligands
5 and 6 with the appropriate silver salts.

Supramolecular Chemistry
277
Scheme 1. Synthesis of diphosphonium ligands.
For comparison purpose, the monomeric subunit of
ligand 5, triphenylphosphonium-butoxy-p-tert-butylphenol
bromide (8), was synthesised as previously described (22).
The cone conformation of the disubstituted calix[4]-
and 1.28 ppm for 1 and 0.96 and 1.29 ppm for 2) and an AB
system for the methylene protons (at 3.19 and 3.90 ppm for
1 and at 3.23 and 4.02 ppm for 2). The 1H NMR spectra of
ligands 5 and 6 are characteristic of tetra-substituted
derivatives of calix[4]arenes in the cone conformation. For
instance, in the case of 6, it is indicated by the presence of
1
arenes was indicated by their presence in the H NMR
spectra of two singlets for the tert-butyl protons (at 0.99
Scheme 2. Synthesis of tetraphosphonium ligands.

278
R. Pomecko et al.
the singlet corresponding to the protons of the tert-butyl
groups at 1.03 ppm and the AB system of the methylene
protons of the calixarene at 4.26 and 3.03 ppm. Two well-
defined multiplets for the protons of the aromatic
phosphonium groups can be observed as well as one
doublet for the methyl protons in the direct neighbourhood
of the phosphorus atoms.
Similar but smaller changes were induced in the
spectrum of ligand 2 by the presence of the same anions
(see Table S2, Supplementary Material). As with ligand 1,
no change is observed for the aromatic protons (c,d) and
for the protons (e,f) of the methylene bridge, indicating
that the conformation of the calixarene unit is not
disturbed. This may be explained by the high rigidity of
these disubstituted derivatives due to hydrogen bonds
involving free phenolic groups.
The spectrum of the protonated phosphonium ligand 7
presents several broad peaks corresponding to multiplets
in coalescence. Only two singlets could be clearly
observed, one for the tert-butyl groups and one for the
aromatic protons of the calix[4]arene, suggesting also the
cone conformation of this molecule.
In the 31P NMR spectra of the free ligands 1 and 2, and
of these ligands in the presence of sodium iodide,
thiocyanate and perchlorate, the phosphorus atoms appear
as a singlet indicating that the two phosphonium groups
are chemically equivalent and participate in the anion–
ligand interaction. The most important changes in
chemical shifts are observed for perchlorate: Dd ¼ 0.602
and 0.620 ppm with 1 and 2, respectively (Table 1).
With the tetra-substituted calix[4]arene 5 previously
studied (22) and ClO42, SCN2 and I2, the most important
changes in chemical shifts corresponded to the signals of
the methylene bridge protons and the aromatic protons of
the calixarene, indicating changes in the conformational
tensions of the calixarene scaffold. The signals correspond-
ing to protons of the butyl chains were also shifted as well
as the aromatic protons of the phosphonium. On the
contrary, no change was observed with monomer 8 (22).
The spectrum of ligand 6 was also modified in the
presence of these anions (see Table S3, Supplementary
Material). In particular, the signals corresponding to the
protons of the calix[4]arene scaffold are shifted in a
comparableway for the three anions. The changes observed
Binding studies
1H and 31P NMR studies in chloroform
The interactions between the phosphonium calixarenes
and the anions NO23 , ClO42, I2, SCN2, SO422, HCO23 and
1
Cr2O272 provided as sodium salts were studied by H and
31P NMR in CDCl3. Among these anions, only perchlorate,
1
thiocyanate and iodide salts induced changes in the H
NMR spectra of the ligands. The changes observed in the
case of the disubstituted ligand 1 upon addition of sodium
perchlorate are illustrated in Figure 2. The main signals
affected were those of the CH2 protons (i,j) directly bound
to the carbon atoms next to the phosphorus atoms and the
aromatic protons (k) of the phosphonium groups (see Table
S1, Supplementary Material). The shifts of these signals
due to changes in the charge density on the phosphorus
atoms reflect interactions with these anions.
a
c
b
d
e
f
2
O
OH
g
h
n
1+ClO4–
i
j
P+
2 Br–
k
l
m
1
1
Figure 2. H NMR spectra of ligand 1 alone and in the presence of ClO24 in CDCl3.

Supramolecular Chemistry
279
Table 1. Changes (Dd in ppm) in the 31P NMR spectra of
phosphonium ligands in the presence of sodium iodide,
thiocyanate and perchlorate in CDCl3.
suggest stronger interactions (Table 1). With all ligands, the
most important values were observed for perchlorate.
1
2
5
6
1H NMR studies in acetonitrile
Ligand (d)
25.797
0.215
0.470
0.602
25.962
0.203
0.530
0.620
25.756
0.210
0.767
0.992
25.982
0.175
0.965
1.170
1H NMR experiments were repeated with 5 and sodium
perchlorate in deuterated acetonitrile, a more dissociating
solvent, in which association phenomena are not as
important as in chloroform (22). The spectra of the ligand
are very similar in both the solvents. In acetonitrile, the
addition of NaClO4 induces shifts of the same signals as in
chloroform. Moreover, this study showed the influence of
the counterion of the salts on the shifts observed in the
spectra (see Table S4, Supplementary Material). The
signal of the methylene protons (g) adjacent to the
phenolic oxygen atoms was shifted only with NaClO4 and
LiClO4. With both salts, the shifts corresponding to the
signals of the protons (j) next to the phosphorus atoms
(which are supposed to interact with perchlorate) were
similar, whereas with Et4NClO4 and especially with
CsClO4, the values were very small. The changes observed
suggested that the ligand–anion interaction was connected
with the nature of the counterion and its affinity for the
ligand.
LigandþI2 (Dd)
LigandþSCN2 (Dd)
LigandþClO24 (Dd)
in the spectrum of this ligand in the presence of ClO24 are
illustrated in Figure 3.
With this anion, the multiplets corresponding to the
protons (g) and (j) from 3.87 to 3.70 ppm give two
multiplets from 4.36 to 4.17 ppm for (g) and from 3.28 to
3.11 ppm for (j). The multiplet from 8.17 to 8.02 ppm for
proton (k) is shifted upfield and gives one multiplet from
7.90 to 7.70 ppm, whereas the multiplet from 7.78 to
7.54 ppm of protons (l,m) does not move significantly. The
doublet of the CH3 protons (o) adjacent to the phosphorus
atoms at 2.91 ppm is strongly shifted to 2.50 ppm.
The addition of the anions studied to 7 does not induce
any change in its 1H NMR spectrum. Especially, the
signals of the protons of the phosphonium moieties,
expected to be involved in hydrogen bond formation, were
not shifted.
In order to observe the influence of the ligand
counterion, bromide anions were replaced by the more
lipophilic perchlorate (5a and 6a) or hexafluorophosphate
(5b and 6b) anions. The chemical shifts (d) of selected
protons, given in Table 2, show only slight differences for
protons (c,e,f) and (g) close to the calix[4]arene scaffold
(Dd in the range 0.02–0.07 ppm). In contrast, the signals
of the protons (j) next to the charged phosphorus atoms are
With ligand 6 as with 5, only one singlet for the
phosphorus atoms was detected in its 31P NMR spectrum,
indicatingall thefourphosphonium groupsbeingchemically
equivalent. In the presence of sodium perchlorate and
thiocyanate, the changes in chemical shifts are larger
than those observed with the disubstituted derivatives and
a
c
e
4
f
O
6+ClO4–
g
h
4 Br–
i
j
P+
CH3O
k
l
m
6
1
Figure 3. H NMR spectra of ligand 6 alone and in the presence of ClO24 in CDCl3.

280
R. Pomecko et al.
Table 2. Differences (Dd) in the proton chemical shifts (d) (ppm) in the spectra of ligands 5 and 6, 5a and 6a, 5b and 6b in CD3CN.
c
e
f
g
j
o
d (5)
d (5a)
Dd (5a–5)
6.87
6.83
20.04
2.93
2.91
20.02
4.15
4.11
20.04
3.74
3.76
0.02
3.81
3.17
20.64
–
–
–
d (5b)
Dd (5b–5)
6.80
20.07
2.87
20.06
4.09
20.06
3.77
0.03
3.06
20.75
–
–
d (6)
d (6a)
Dd (6a–6)
6.92
6.89
20.03
3.02
2.98
20.04
4.19
4.15
20.04
3.70
3.73
0.03
3.58
2.85
20.73
2.78
2.37
20.41
d (6b)
Dd (6b–6)
6.88
20.04
2.97
20.05
4.12
20.07
3.75
0.05
2.75
20.83
2.32
20.46
greatly shifted (Dd ¼ 20.64 ppm for 5a, 20.75 ppm for
5b and Dd ¼ 20.73 ppm for 6a, 20.83 ppm for 6b). With
ligands 6a and 6b, the signals of the protons (o) of the
methyl substituents are also displaced (Dd ¼ 20.41 ppm
for 6a and 20.46 ppm for 6b). These results show that the
chemical shifts of the protons next to the phosphonium
groups are influenced by the ligand counterion.
models. With sodium and both ligands, the best fit was
obtained by considering the presence of ML and ML2
complexes. The same species were found with LiClO4 and
5b, whereas only a 1:1 complex was formed with 5. The
formation of ML2 species could be explained by the
complexation of the ion pair NaþClO42 by two ligands.
The stability constants of these complexes are given in
Table S5 (see the Supplementary Material).
On the other hand, it was also shown that when the
lipophilic PF26 anions were replacing the Br2 counterions
of the ligand, there was still a significant shift of the
protons (j) close to the phosphorus atoms for ligand 5b in
the presence of ClO24 (Table S4) (22). This result
suggested complexation of perchlorate with this ligand,
where no exchange is normally possible.
An important heat effect was observed during the
titration of ligands 5 and 5b against LiBr, which should be
related directly to cation complexation as no anion
exchange is possible with these ligands. The data
interpretation led to species of the same stoichiometry as
with LiClO4 (Table S5). The values of the stability
constants of 1:1 species formed in the presence of LiClO4
and LiBr are comparable (with 5, log b ¼ 3.24 and 3.15,
respectively, and with 5b, log b ¼ 4.29 and 4.32,
respectively). They are lower than that of the complex
formed by the tetra-methylated p-tert-butylcalix[4]arene
with lithium (log b ¼ 5.10 in acetonitrile (28)).
Microcalorimetric studies in acetonitrile
In order to get more information on the influence of the
counterion of the salt, microcalorimetric titrations were
carried out with ligands 5 and 5b against NaClO4, LiClO4
and Et4NClO4 in acetonitrile. The thermograms recorded
during the titration of these ligands with NaClO4 and
LiClO4 showed significant exothermic heat effects, whereas
their titration with Et4NClO4 led to no thermal effect (see
Figure S1, Supplementary Material). For comparison
purpose, the titration of monomer 8 with NaClO4 was
also carried out showing no significant heat effect.
If only the ClO24 anion were involved in the
complexation, a similar heat effect should be observed in
all the titrations. The fact that an effect is only observed for
NaClO4 and LiClO4 suggests that it is not only related to
the anion interaction (complexation or anion exchange).
Assuming that the large tetraethylammonium cation
cannot be complexed with a calix[4]arene, the heat effects
observed during the titration with NaClO4 and LiClO4
would rather be due to the complexation of the cations
with the two ligands. This is supported by the fact that no
heat effect was detected with monomer 8, supposed to be
unable to complex these cations.
All that considered, the UV spectrophotometric
titrations of ligand 5 against ClO24 previously performed
may certainly be interpreted in terms of cation rather than
anion complexation (22). The values of the stability
constants of the 1:1 complexes with sodium perchlorate
(log b ¼ 3.81 ^ 0.02) and with lithium perchlorate
(log b ¼ 3.71 ^ 0.04) (22) are of the same order of
magnitude as those obtained from microcalorimetric
measurements.
Potentiometric studies
Only few ligands containing phosphorus atoms have been
studied so far as active material in ion-selective membrane
electrodes (29–31). They showed a selective response for
ClO24 , with, however, little discrimination with respect to
I2 and SCN2. The results suggested a particular affinity of
ligands containing phosphorus atoms for ClO24 , SCN2 and
I2 anions. By attaching phosphonium moieties to a
calix[4]arene scaffold and taking advantage of the
preorganisation of the ligand, it was expected to enhance
Calorimetric data obtained with NaClO4 and LiClO4
were interpreted assuming different cation complexation

Supramolecular Chemistry
281
the selectivity for tetrahedral or spherical anions. Such
selectivity (especially for ClO24 over I2) is hard to obtain
with other kinds of receptors.
more than 3 weeks. They showed close to Nernstian
response for ClO42, I2, NO23 and SCN2 and no significant
response for SO242, CO322, HPO422, PO432. A similar
behaviour was already observed for electrodes incorporat-
ing ligand 5 (Table S8). The highest selectivity was
obtained for ClO24 ions in buffered solution (pH 5.5) and in
water (pH 6.5). The over-Nernstian slope of the electrode
response for Cr2O272 could indicate a mechanism where
both processes, anion complexation and anion exchange,
play an important role. It can also be explained as the
presence of different forms of chromate in the sample.
While the addition of lipophilic anionic sites (KTClPB)
to the membranes in the case of ligand 5 did not change
much the properties of the electrodes, it affected the
properties of the electrodes containing 6 (Table 3). Without
the salt, the slope of the electrode is 254.2 mV, and slightly
decreases to 251.8 and 249.9 mV, respectively, in the
presence of the salt. According to the literature data (29,
31), such results suggest that none of the ligands works as a
neutral carrier because the addition of the lipophilic anion
to the membrane does not induce a cationic response of the
potentiometric cell. Ligand 6 seems to work in the
membrane as a typical anion exchanger, whereas ligand 5
could be considered as a charged ligand despite the small
influence of KTClPB.
The influence of the ligand counterions (Br2, ClO42,
PF26 ) on the properties of the membrane electrodes was
also studied. Table 4 compares the responses of
perchlorate of electrodes based on ligands 5a and 6a
(perchlorates) and on ligands 5b and 6b (hexafluoropho-
sphates) to the corresponding electrodes based on ligands
5 and 6 (bromides). The less good slope of the electrodes
containing 6a and 6b as compared to that of the electrodes
containing 6 suggests rather the anion exchange nature of
the latter ligand. In this case, the presence of more
lipophilic anions (perchlorate or hexafluorophosphate)
Disubstituted phosphonium ligands 1 and 2 were tested
as ionophores in the membrane electrodes. The electrodes
were sensitive to perchlorate, thiocyanate, iodide and
nitrate, showing fast, near-Nernstian responses (Figure 4
and Table S6 of the Supplementary Material). However,
their characteristics changed with time. Attempts to
optimise the composition of the conditioning solutions as
well as the conditioning time did not improve the situation,
which might be due to slow leakage of the ionophores
from the membrane. These ligands had also the tendency
to crystallise in the membrane phase. Crystallisation of our
ligands within the membranes depends strongly on the
kind of plasticiser used. The ligands in the membranes
based on bis-(2-ethylhexyl)sebacate (BEHS) crystallised
strongly, which decreased their stability. This phenomenon
originates from the higher lipophilicity of this plasticiser
(log P ¼ 10.1) as compared to 2-nitrophenyloctylether (o-
NPOE) (log P ¼ 5.9) (32), which is more suitable for
charged ligands. Electrodes with membranes based on o-
NPOE had the best lifetime and response characteristics
and were chosen for further studies.
The more lipophilic ligands 3 and 4, in which n-propyl
chains replace the two phenolic OH groups, were also
synthesised. The lifetime of electrodes incorporating these
ligands for perchlorate was increased to at least 3 weeks.
The repeatability of the measurements was also good, but
their detection limits increased (see Table S7, Supplemen-
tary Material).
The membranes of electrodes incorporating the tetra-
substituted ligand 6 showed rather quick (within 15–20 s),
stable and fully reversible responses (Figure 5 and Table
S8 of the Supplementary Material). The repeatability of
the measurements was also good and their lifetime was
150
100
50
ClO4–
SCN–
I–
NO3–
0
Br–
–50
–100
–150
HPO32–
Cr2O72–
–7
–6
–5
–4
–3
–2
–1
log [A]
Figure 4. Potentiometric anion responses of electrodes with the PVC/NPOE membrane containing ligand 1 in MES buffer at pH 5.5.

282
R. Pomecko et al.
150
100
50
ClO4–
SCN–
I–
NO3–
0
Br–
–50
–100
–150
HCO32–
Cr2O72–
–7
–6
–5
–4
–3
–2
–1
log [A]
Figure 5. Potentiometric anion responses of electrodes with the PVC/NPOE membrane containing ligand 6 in MES buffer at pH 5.5.
slows down the anion exchange process. In contrast, the
properties of electrodes with the bromide ligand 5 and with
the perchlorate ligand 5a are comparable. The presence of
highly lipophilic perchlorate anions does not disturb the
electrode response, while the presence of hexafluoropho-
sphate anions in the membrane phase (electrode contain-
ing ligand 5b) decreases the slope and the linearity range.
The poorer properties of the electrode containing 5b
and 6b could be explained by the higher lipophilicity of
hexafluorophosphate anions which hinders the process of
anion exchange. The complexation of ClO24 by ligand 5
could explain the good response of electrodes based on 5
and 5a to perchlorate. Such interpretation is consistent with
the results of previous experiments (Table 4) and indicates
that ligand 5 behaves more like a charged carrier for ClO24 ,
while ligand 6 behaves more as an anion exchanger.
The order of selectivity observed for all phosphonium
ligands 1–6 and 8 follows the Hofmeister series:
The electrodes based on tetraphosphonium derivatives
show higher selectivities for perchlorate over thiocyanate,
iodide and nitrate than those based on their diphosphonium
counterparts. These selectivities are also better than the
selectivity of electrodes based on monomer 8.
The replacement of one phenyl substituent on the
phosphorus atoms by one methyl group does not change
significantly the selectivity pattern and the values of the
selectivity coefficients of the electrodes containing either
di- or tetraphosphonium ligands. The only exception is the
selectivity of the electrodes based on compound 5 against
Cr2O272, which increases from 2.6 to 3.2 log units.
Electrodes based on alkylated compounds 3 and 4 are
also selective for perchlorate but the selectivity over iodide
and thiocyanate is decreased (Table 5).
Ligands 5 and 6 display better potentiometric proper-
ties than the protonated cyclam (33) or its copper complex
(34) than a phosphodithiamacrocycle (35). For instance,
the detection limit is 2.5 £ 1027 M with ligand 5 when
ClO42 . SCN2 . I2 . Cr2O722. NO32
compared to 4.2 £ 1026 M for the cyclam and 8 £ 1027
M
. Br2. HCO32. HPO422. SO422
:
for the phosphodithiamacrocycle. Ligand 6 as 5 presents
generally higher selectivities than TDMACl (36), the
The highest selectivity is observed for perchlorate (Table 5).
protonated cyclam and [Cu(cyclam)]2þ
.
Table 3. Characteristics of potentiometric responses for
perchlorate of PVC/NPOE electrodes containing ligands 5 and 6
and different amounts of KTClPB.
Table 4. Characteristics of potentiometric responses for
perchlorate of PVC/NPOE electrodes containing tetra-substituted
phosphonium ligands with different counterions.
Ligand
KTClPB (mol%)
S (mV/decade)
LR (log[A])
Ligand
Counterion
S (mV/decade)
LR (log[A])
5
5
5
6
6
6
0
40
120
0
40
120
255.6
254.6
255.9
254.2
251.8
249.9
26.0
26.0
25.7
26.0
26.0
25.7
5
Br2
256.5
256.1
238.3
254.4
240.4
236.2
26.0
26.0
25.5
25.7
26.0
26.0
5a
5b
6
6a
6b
ClO24
PF26
Br2
ClO24
PF26
Notes: Inner and conditioning electrolyte, MES/NaOH, pH ¼ 5.5/1022 M NaCl.
Notes: Inner and conditioning electrolyte, MES/NaOH, pH ¼ 5.5/1022 M NaCl.

Supramolecular Chemistry
283
pot
ClO24 ;X
Table 5. Selectivity coefficients as log K
monomer 8.
of the PVC/NPOE membrane electrodes based on phosphonium calixarenes 1–6 and
pot
ClO24 ;X
Log K
Anion X
1
2
3
4
5
6
8
ClO24
SCN2
I2
NO23
HCO23
0
0
0
0
0
0
0
21.2
21.7
22.6
24.6
21.4
21.7
22.6
24.5
21.1
21.3
22.4
24.6
20.6
20.9
21.9
24.4
21.3
22.0
22.9
24.6
21.4
22.0
23.0
24.4
21.1
21.6
22.5
ND
Cr2O272
22.2
22.4
22.6
21.7
22.6
23.2
21.9
HPO242
SO242
24.5
24.9
24.5
24.7
24.5
24.7
24.4
24.7
24.5
24.7
24.4
24.5
ND
ND
Note: ND, not determined.
Concluding remarks
complexed in the cavity of the calixarene. With perchlorate,
the best interaction takes place with Naþ and Liþ, whereas
little or no interaction occurs with the larger Et4Nþ and
Csþ. It can also be noted that the anion also plays a role in
the complexation of the cation, since the spectrum of the
calixarene part is not affected in the presence of NO23 ,
SO242, HCO23 and Cr2O722, i.e. the less lipophilic ones.
The different techniques used to assess the binding
properties of phosphonium derivatives showed strong
interactions with SCN2, I2 and especially with ClO24 , and
pointed out the important role played by the salt counterion
(Naþ or Liþ), which may be complexed by the
calix[4]arene. Incorporated in the PVC membrane electro-
des, these molecules are efficient sensing material for
anions with a selectivity order following the Hofmeister
series generally observed for ion exchangers. However, the
electrodes based on tetraphosphonium derivatives showed
better selectivities than those based on the diphosphonium
analogues or on the monomeric unit, indicating the
importance of the ligand preorganisation which cannot be
observed in the case of simple anion exchangers.
Experimental
FAB mass spectra were obtained on a VG analytical ZAB
HF instrument. All reagents and solvents were commercial
and used without further purification.
Chromatography columns were prepared from
Kieselgel Merck Si 60 (40–63 mm). TLC was performed
on 250 mm silica gel plates (Merck, Darmstadt, Germany)
containing a fluorescent indicator.
A question which must be addressed concerns the
nature of the interaction between the ligands and the
anions, e.g. ClO24 . Is it a simple ion exchange between the
bromides of the ligand and this more lipophilic anion, or is
it complexation within the charged phosphonium groups?
What is the exact role of the salt counterion?
Synthesis of intermediate compounds
1,3-Bis-(butoxy-4-bromide)-p-tert-butylcalix[4]arene (S2)
Into a 250 cm3 flask containing tetrakis-p-tert-butylca-
lix[4]arene S1 (3.244 g, 5.00 mmol) and acetone (50 cm3),
K2CO3 (1.383 g, 10.00 mmol) was added. The mixture was
stirred at room temperature for 2 h. Then, 1,4-butyl-
dibromide (3.236 g, 15.00 mmol) in acetone (50 cm3) was
added. The mixture was left for 4 days under reflux. After
4 days, methanol (5 cm3) was added. The solvents were
evaporated, and the reaction mixture was dissolved in
dichloromethane (100 cm3). After extraction with water
(150 cm3), the organic phase was dried with Na2SO4 and
evaporated. The crude product was purified by crystal-
lisation from a 1:10 dichloromethane–methanol mixture
giving compound S2 (3.252 g, 3.54 mmol) in 71% yield.
If NMR gives some indications on the changes in the
molecule, suggesting interactions, it does not tell if there is
complexation or anion exchange, since the nature of the
counterion of the ligand has been shown to influence the
chemical shifts of the protons near the charged atoms. In
favour of ion exchange is the fact that the most important
shifts are observed with the more lipophilic anions ClO24 ,
SCN2 and I2. The behaviour of selective electrodes is also
consistent with this assumption. However, the fact that no
change occurred in the spectrum of the monomer, where
only exchange is possible, is against this hypothesis. In
favour of complexation is the fact that, in the presence of
NaClO4, strong shifts are observed for the signals of the
protons next to the phosphonium groups in the spectrum of
the hexafluorophosphate ligand where no exchange is
possible.
1
Mp . 2808C. H NMR (300 MHz, CDCl3) d (ppm): 0.98
(s, 18H, CZ(CH3)3), 1.30 (s, 18H, CZ(CH3)3), 2.15 (qn,
4H, J ¼ 4.4 Hz, CH2ZCH2ZO), 2.32 (qn, 4H, J ¼ 4.2 Hz,
CH2ZCH2ZBr), 3.32 (d, 4H, J ¼ 13.0 Hz, Ar-CH2-Ar),
3.65 (t, 4H, J ¼ 6.6 Hz, CH2ZCH2ZBr), 4.01 (t, 4H,
1
On the other hand, H NMR and microcalorimetry
emphasised the importance of the cation, which can be

284
R. Pomecko et al.
J ¼ 5.8 Hz, CH2ZCH2ZO), 4.25 (d, 4H, J ¼ 13.0 Hz, Ar-
CH2-Ar), 6.78 (s, 4H, Ar-H), 7.08 (s, 4H, Ar-H), 7.40 (s,
2H, OH). Anal. Calcd for C52H70O4Br2: C, 67.97; H, 7.68.
Found: C, 68.21; H, 7.94.
Tetrakis-(butoxy-4-bromide)-tetrakis-p-tert-
butylcalix[4]arene (S5)
The suspension of p-tert-butylcalix[4]arene S1 (1.947 g,
3.00 mmol) and NaH in oil washed three times with
hexane (0.700 g, 29.17 mmol) was stirred at room
temperature in DMF (50 cm3) for 1 h. Then, 1,4-
dibromobutane (12.947 g, 59.06 mmol) was added and
the mixture was heated to 808C. After 4 days of heating,
the mixture was cooled and MeOH (20 cm3) was added.
After removal of the solvent, the residue was dissolved in
dichloromethane and water and acidified with 1 M HCl.
The organic layer was dried over Na2SO4, filtered and
evaporated. After precipitation from methanol, the pure
compound S5 (1.520 g, 1.28 mmol) was obtained in 43%
1,3-Bis-(propoxy)-p-tert-butylcalix[4]arene (S3)
Into a 250 cm3 flask containing tetrakis-p-tert-butylca-
lix[4]arene S1 (3.244 g, 5.00 mmol) and acetone (50 cm3),
K2CO3 (1.383 g, 10.00 mmol) was added. The mixture was
stirred at room temperature for 2 h. Then, bromopropane
(1.845 g, 15.00 mmol) in acetone (40 cm3) was added. The
mixture was left for 4 days under reflux. After 4 days,
methanol (5 cm3) was added. The solvents were evapor-
ated, and the reaction mixture was dissolved in
dichloromethane (100 cm3). After extraction with water
(150 cm3), the organic phase was dried with Na2SO4 and
evaporated. The crude product was purified by crystal-
lisation from a 1:9 acetone–methanol mixture giving the
pure compound S3 (2.016 g, 2.75 mmol) in 55% yield.
1
yield. Mp 1808C. H NMR (300 MHz, CDCl3) d (ppm):
1.09 (s, 36H, ZC(CH3)3), 1.98–2.09 (m, 8H, ZCH2Z),
2.16–2.23 (m, 8H, ZCH2Z), 3.15 (d, 4H, J ¼ 13.0 Hz,
Ar-CH2-Ar), 3.53 (t, 8H, J ¼ 6.9 Hz, ZCH2ZBr), 3.91 (t,
8H, J ¼ 6.9 Hz, ZCH2ZOZ), 4.36 (d, 4H, J ¼ 13.0 Hz,
Ar-CH2-Ar), 6.79 (s, 8H, Ar-H). Anal. Calcd for
C60H84O4Br4: C, 60.61; H, 7.12; Found: C, 60.87; H, 7.32.
1
Mp . 2808C. H NMR (200 MHz, CDCl3) d (ppm): 1.02
(s, 18H, CZ(CH3)3), 1.27 (t, 6H, J ¼ 5.0 Hz, CH3ZCH2-
Z), 1.28 (s, 18H, CZ(CH3)3), 2.06 (sx, 4H, J ¼ 4.9 Hz,
CH3ZCH2ZCH2ZO), 3.32 (d, 4H, J ¼ 12.8 Hz, Ar-CH2-
Ar), 3.96 (t, 4H, J ¼ 5.9 Hz, OZCH2ZCH2), 4.31 (d, 4H,
J ¼ 12.8 Hz, Ar-CH2-Ar), 6.86 (s, 4H, Ar-H), 7.05 (s, 4H,
Ar-H), 7.89 (s, 2H, OH). Anal. Calcd for C50H68O4: C,
81.92; H, 9.35. Found: C, 82.05; H, 9.40.
Tetrakis-(4-(diphenylphosphine)-butoxy)-p-tert-
butylcalix[4]arene (S6)
Into a 100 cm3 flask containing compound S5 (1.510 g,
1.27 mmol) of freshly distilled THF (10 cm3), KP(Ph)2
(1.282 g, 5.72 mmol) in THF (15 cm3) was added via a
syringe. The mixture was stirred for 2 h at room
temperature, during which the colour of the reaction
mixture changed from red to dark yellow. The mixture was
then evaporated and extracted twice with dichloromethane
(30 cm3). Purification of the crude product on silica
column with a 3:7 dichloromethane–hexane mixture as
the eluent gave compound S6 (0.409 g, 0.26 mmol) in 20%
1,3-Bis-(propoxy)-2,4-bis-(butoxy-4-bromide)-tetrakis-p-
tert-butylcalix[4]arene (S4)
Into a 250 cm3 flask containing S3 (2.016g, 2.75mmol) and
DMF (50 cm3), NaH (0.480g, 20.00 mmol) was added. NaH
was washed twice with hexane before addition. The mixture
was stirred at room temperature for 4 h. After this time, 1,4-
dibromobutyl (5.940g, 27.5mmol) in DMF (40cm3) was
added. The mixture was left for 2 days at 80–908C. After
2 days, methanol (30 cm3) was added. The solvents were
evaporated and the reaction mixture was dissolved in
dichloromethane (100cm3). After extraction with water
(250cm3), the organic phase was dried with Na2SO4 and
evaporated. The crude product was purified by crystal-
lisation from a 1:10 dichloromethane–methanol mixture
giving compound S4 (1.584g, 1.57mmol) in 57% yield.
1
yield. Mp 1108C. H NMR (300 MHz, CDCl3) d (ppm):
1.08 (s, 36H, ZC(CH3)3), 1.42–1.63 (m, 8H, ZCH2
ZCH2ZP), 2.00–2.18 (m, 8H, ZCH2ZCH2ZO), 2.00–
2.18 (m, 8H, ZCH2ZP), 3.05 (d, 4H, J ¼ 12.9 Hz, Ar-
CH2-Ar), 3.81 (t, 8H, J ¼ 5.4 Hz, ZCH2ZCH2ZO), 4.29
(d, 4H, J ¼ 12.9 Hz, Ar-CH2-Ar), 6.75 (s, 8H, Ar-H),
7.20–7.46 (m, 40H, P-Ar-H). 31P NMR (400 MHz,
CDCl3) d (ppm): 214.91 [P]. m/z (MALDI) 1610.88
(MþH)þ. Anal. Calcd for C108H124O4P4: C, 80.57; H,
7.76; Found: C, 80.73; H, 7.87.
1
Mp 1708C. H NMR (200MHz, CDCl3) d (ppm): 1.01 (t,
6H, J ¼ 5.2 Hz, CH3ZCH2Z), 1.05 (s, 18H, CZ(CH3)3),
1.12 (s, 18H, CZ(CH3)3), 1.99–2.05 (m, 4H, CH3ZCH2Z),
2.07–2.25 (m, 8H, ZCH2ZCH2ZCH2ZBr), 3.13
(d, 4H, J ¼ 13.0Hz, Ar-CH2-Ar), 3.51 (t, 4H, J ¼ 6.6 Hz,
ZCH2ZBr), 3.81 (t, 4H, J ¼ 5.8 Hz, CH3ZCH2ZCH2ZO),
3.91 (t, 4H, J ¼ 5.8 Hz, BrZCH2ZCH2ZCH2ZCH2ZO),
4.39 (d, 4H, J ¼ 13.0Hz, Ar-CH2-Ar), 6.74 (s, 4H, Ar-H),
6.83 (s, 4H, Ar-H). Anal. Calcd for C58H82O4Br2: C, 69.45;
H, 8.24. Found: C, 69.65; H, 8.30.
Synthesis of phosphonium ligands
1,3-Bis-(4-triphenylphosphonium-butoxy)-p-tert-
butylcalix[4]arene dibromide (1)
Into a 100 cm3 flask containing S2 (1.184 g, 2.00 mmol) in
chloroform (30 cm3), triphenylphosphine (5.248 g,
20.00 mmol) in chloroform (20 cm3) was added. After
6 days under reflux, the mixture was cooled and the solvent

Supramolecular Chemistry
285
was evaporated. The residue was dissolved in dichlor-
omethane. The excess of triphenylphosphine was pre-
cipitated from methanol and filtered out. The filtrate was
evaporated. The pure product 1 (2.179 g, 1.51 mmol) was
obtained by precipitation from a 1:9 dichloromethane–
hexane mixture, as a white–light green powder in 76%
yield: Mp . 2808C. 1H NMR (300 MHz, CDCl3) d (ppm):
0.99 (s, 18H, ZC(CH3)3), 1.28 (s, 18H, ZC(CH3)3), 2.05–
2.29 (m, 8H, CH2ZCH2ZCH2ZCH2ZO), 3.19 (d, 4H,
J ¼ 12.8 Hz, Ar-CH2-Ar), 3.80–4.00 (m, 4H, ZCH2ZO),
3.90 (d, 4H, J ¼ 12.8 Hz, Ar-CH2-Ar), 3.90–4.08 (m, 4H,
ZCH2ZP), 6.79 (s, 4H, Ar-H), 6.99 (s, 4H, Ar-H), 7.49 (s,
2H, OH), 7.54–7.64 (m, 12H, P-Ar-H meta), 7.65–7.74
(m, 6H, P-Ar-H para), 7.80–7.93 (m, 12H, P-Ar-H ortho).
31P NMR (400 MHz, CDCl3) d (ppm): 25.80. m/z (FABþ)
721.7 (Mþ2H)2þ; m/z (MALDI) 1361.7 (M2Br)þ. Anal.
Calcd for C88H100O4P2Br2: C, 73.22; H, 6.98. Found: C,
73.46; H, 7.20.
ZC(CH3)3), 1.65–1.87 (m, 4H, CH3ZCH2ZCH2ZO),
1.77–1.94 (m, 4H, ZCH2ZCH2ZP), 2.28–2.43 (m, 4H,
ZCH2ZCH2ZO), 2.95 (d, 4H, J ¼ 13.0 Hz, Ar-CH2-Ar),
3.63 (t, 4H, J ¼ 5.9 Hz, CH3ZCH2ZCH2ZO), 3.85–4.00
(m, 4H, ZCH2ZP), 3.85–4.00 (m, 4H, ZCH2ZO), 4.18
(d, 4H, J ¼ 13.0 Hz, Ar-CH2-Ar), 6.61 (s, 4H, Ar-H),
6.76 (s, 4H, Ar-H), 7.60–7.92 (m, 30H, P-Ar-H ortho,
meta, para). 31P NMR (400 MHz, CDCl3) d (ppm): 25.84.
m/z (MALDI) 1447.6 (M2Br)þ. Anal. Calcd for
C94H112O4P2Br2: C, 73.91; H, 7.39. Found: C, 73.67;
H, 7.66.
1,3-Bis-propoxy-2,4-bis-(4-(P,P-diphenyl-P-
methylphosphonium)-butoxy)-p-tert-butylcalix[4]arene
dibromide (4)
Compound 4 was obtained according to the same procedure
as for 3 with S4 (3.050 g, 3.04 mmol) and diphenylmethyl-
phosphine (6.006 g, 30.00 mmol) as a white powder in 64%
1
yield. Mp 1488C. H NMR (300 MHz, CDCl3) d (ppm):
1,3-Bis-(4-(P,P-diphenyl-P-methylphosphonium)-butoxy)-
p-tert-butylcalix[4]arene dibromide (2)
0.79 (s, 18H, ZC(CH3)3), 1.03 (t, 6H, J ¼ 6.9 Hz,
CH3ZCH2ZCH2ZO), 1.31 (s, 18H, ZC(CH3)3),
1.55–1.75 (m, 4H, ZCH2ZCH2ZP), 1.88–2.04
(m, 4H, CH3ZCH2ZCH2ZO), 2.40–2.58 (m, 4H, CH2Z
CH2ZCH2ZO), 3.05 (d, 4H, J ¼ 12.5 Hz, Ar-CH2-Ar),
3.23 (d, 6H, J ¼ 14.3 Hz, CH3ZP), 3.62–3.80 (m, 8H,
ZCH2ZP and ZCH2ZO), 3.88 (t, 4H, J ¼ 5.9 Hz,
ZCH2ZCH2ZCH2ZO), 4.32 (d, 4H, J ¼ 12.5 Hz, Ar-
CH2-Ar), 6.43 (s, 4H, Ar-H), 7.08 (s, 4H, Ar-H), 7.60–7.80
(m, 12H, P-Ar-H meta, para), 8.01–8.13 (m, 8H, P-Ar-H
ortho). 31P NMR (400 MHz, CDCl3) d (ppm): 25.98. m/z
(MALDI) 1323.7 (M2Br)þ. Anal. Calcd for C84H108
O4P2Br2: C, 71.88; H, 7.76. Found: C, 71.99; H, 7.85.
Compound 2 was prepared following the same procedure as
for compound 1 with S2 (2.753 g, 3.00 mmol) and
diphenylmethylphosphine (6.006g, 30.00 mmol) in 69%
yield. Mp . 2808C. 1H NMR (300MHz, CDCl3) d (ppm):
0.96 (s, 18H, ZC(CH3)3), 1.29 (s, 18H, ZC(CH3)3),
1.91–2.10 (m, 4H, ZCH2ZCH2ZP), 2.11–2.27 (m, 4H,
ZCH2ZCH2ZO), 2.99 (d, 6H, J ¼ 13.8Hz, CH3ZP),
3.23 (d, 4H, J ¼ 13.5Hz, Ar-CH2-Ar), 3.58–3.75 (m, 4H,
ZCH2ZP), 3.90 (t, 4H, J ¼ 5.3 Hz, ZCH2ZO), 4.02 (d, 4H,
J ¼ 13.5Hz, Ar-CH2-Ar), 6.76 (s, 4H, Ar-H), 7.01 (s,
4H, Ar-H), 7.31 (s, 2H, OH), 7.48–7.59 (m, 8 H, P-Ar-H
meta), 7.61–7.70 (m, 4H, P-Ar-H para), 7.90–8.06 (m, 8H,
P-Ar-H ortho). 31P NMR (400MHz, CDCl3) d (ppm):
25.96. m/z (MALDI) 1239.58 (M2Br)þ. Anal. Calcd for
C78H96O4P2Br2: C, 71.01; H, 7.33. Found: C, 71.21; H, 7.52.
Tetrakis-(4-triphenylphosphonium-butoxy)-tetrakis-p-
tert-butylcalix[4]arene tetrabromide (5)
Compound S5 (1.184 g, 1.00 mmol) was dissolved in
chloroform (30 cm3). After a few minutes of stirring,
triphenylphosphine (5.248 g, 20.00 mmol) and chloroform
(20 cm3) were added. After 6 days of refluxing, the mixture
was cooled and the solvent was evaporated. The residue
was dissolved in dichloromethane. The excess of
triphenylphosphine was precipitated from methanol and
filtered. The organic layer was evaporated. Chromatog-
raphy on a silica column with a 90:10 dichloromethane–
methanol mixture as the eluent gave compound 5 (0.67 g,
0.30 mmol) in 30% yield. Mp 1328C. 1H NMR (300 MHz,
CDCl3) d (ppm): 1.02 (s, 36H, ZC(CH3)3), 1.56–1.72 (m,
8H, ZCH2Z), 2.24–2.41 (m, 8H, ZCH2Z), 2.91 (d, 4H,
J ¼ 13.0 Hz, Ar-CH2-Ar), 3.78–4.01 (m, 16H, ZCH2ZP
and Ar-OZCH2), 4.23 (d, 4H, J ¼ 13.0 Hz, Ar-CH2-Ar),
6.63 (s, 8H, Ar-H), 7.59–7.71 (m, 36H, P-Ar-H, meta,
para), 7.76–7.88 (m, 24H, P-Ar-H, ortho). 31P NMR
1,3-Bis-(4-triphenylphosphonium-butoxy)-2,4-bis-
propoxy-p-tert-butyl-calix[4]arene dibromide (3)
Into a 100 cm3 flask containing S4 (2.012 g, 2.00 mmol) in
chloroform (30 cm3), triphenylphosphine (5.248 g,
20.00 mmol) in chloroform (20 cm3) was added and left
for 6 days under reflux. After that time, the mixture was
cooled and the solvent was evaporated. The residue was
dissolved in dichloromethane. The excess of triphenylpho-
sphine was precipitated from methanol and filtered off. The
filtrate was evaporated. The pure product 3 (2.179 g,
1.51 mmol) was obtained by precipitation from a 1:9
dichloromethane–hexane mixture, as a white–light green
1
powder in 76% yield. Mp 1208C. H NMR (300 MHz,
CDCl3) d (ppm): 0.86 (t, 6H, J ¼ 5.3 Hz, CH3ZCH2-
ZCH2ZO), 0.99 (s, 18H, ZC(CH3)3), 1.11 (s, 18H,

286
R. Pomecko et al.
(400 MHz, CDCl3) d (ppm): 25.76. m/z (FABþ) 479.5
(M)4þ; m/z (MALDI) 2157.7 (M2Br)þ. Anal. Calcd
for C132H144O4P4Br4: C, 70.84; H, 6.49. Found: C, 70.97;
H, 6.69.
mixture to give compound 6 (1.190 g, 0.59 mmol) in
59% yield. Mp 1608C. 1H NMR (300 MHz, CDCl3)
d (ppm): 1.03 (s, 36H, ZC(CH3)3), 1.55–1.75 (m, 8H,
ZCH2ZCH2ZP), 2.30–2.48 (m, 8H, ZCH2ZCH2ZO),
2.91 (d, 12H, J ¼ 13.5 Hz, CH3ZP), 3.03 (d, 4H,
J ¼ 12.8 Hz, Ar-CH2-Ar), 3.70–3.87 (m, 16H, ZCH2ZP
and CH2ZO), 4.26 (d, 4H, J ¼ 12.8 Hz, Ar-CH2-Ar), 6.71
(s, 8H, Ar-H), 7.54–7.78 (m, 24H, P-Ar-H meta, para),
8.02–8.17 (m, 16H, P-Ar-H ortho). 31P NMR (400 MHz,
CDCl3) d (ppm): 25.98. m/z (MALDI) 1909.68 (M2Br)þ.
Anal. Calcd for C112H136O4P4Br4: C, 67.61; H, 6.89.
Found: C, 68.59; H, 7.10.
Tetrakis-(4-triphenylphosphonium-butoxy)-p-tert-
butylcalix[4]arene tetraperchlorate (5a)
Into a 10 cm3 flask, 5 (0.100 g, 0.045 mmol) was
dissolved in acetonitrile (1 cm3). AgClO4 (0.050 g,
0.241 mmol) in acetonitrile (1 cm3) was added dropwise
to the ligand solution. After 24 h of stirring at room
temperature, the precipitate of AgBr was filtered off. The
filtrate was evaporated to give compound 5a (0.088 g,
0.038 mmol) in 84% yield. Mp . 1208C decomposition.
1H NMR (300 MHz, CDCl3) d (ppm): 1.07 (s, 36H,
ZC(CH3)3), 1.70–1.85 (m, 8H, ZCH2ZCH2ZP), 2.45–
2.53 (m, 8H, ZCH2ZCH2ZO), 2.95 (d, 4H, J ¼ 13.0 Hz,
Ar-CH2-Ar), 3.43–3.60 (m, 8H, ZCH2ZP), 4.20–4.30
(m, 8H, ZCH2ZO), 4.44 (d, 4H, J ¼ 13.0 Hz, Ar-CH2-
Ar), 6.97 (s, 8H, Ar-H), 7.56–7.90 (m, 60H, P-Ar-H,
ortho, meta, para). 31P NMR (400 MHz, CDCl3) d (ppm):
24.54. m/z (MALDI) 2214.2 (M2ClO4)þ. Anal. Calcd
for C132H144O4P4(ClO4)4: C, 68.45; H, 6.27. Found: C,
68.70; H, 6.45.
Tetrakis-(4-(P,P-diphenyl-P-methylphosphonium)-
butoxy)-p-tert-butylcalix[4]arene tetraperchlorate (6a)
Compound 6a was obtained according to the same
procedure as for 5a with 6 (0.100 g, 0.05 mmol) and
AgClO4 (0.050 g, 0.24 mmol) in 90% yield. Mp . 1208C
decomposition. 1H NMR (300 MHz, CDCl3) d (ppm): 1.09
(s, 36H, ZC(CH3)3), 1.65–1.80 (m, 8H, ZCH2ZCH2ZP),
2.35–2.50 (m, 8H, ZCH2ZCH2ZO), 2.50 (d, 12H,
J ¼ 12.0 Hz, CH3ZP), 3.18–3.28 (m, 8H, ZCH2ZP),
3.38 (d, 4H, J ¼ 12.8 Hz, Ar-CH2-Ar), 4.15–4.28 (m, 8H,
ZCH2ZO), 4.45 (d, 4H, J ¼ 12.8 Hz, Ar-CH2-Ar), 6.97 (s,
8H, Ar-H), 7.60–7.87 (m, 40H, P-Ar-H, ortho, meta,
para). 31P NMR (400 MHz, CDCl3) d (ppm): 24.77.
m/z (MALDI) 1966.24 (M2ClO4)þ. Anal. Calcd for
Tetrakis-(4-triphenylphosphonium-butoxy)-tetrakis-p-
tert-butylcalix[4]arene tetrahexafluorophosphate (5b)
C112H136O4P4(ClO4)4: C, 65.05; H, 6.63. Found: C, 65.13;
H, 6.72.
Compound 5 (0.100 g, 0.045 mmol) was dissolved in
acetonitrile (1 cm3). Then, AgPF6 (0.062 g, 0.245 mmol)
was dissolved in acetonitrile and added dropwise to the
ligand solution. After 24 h, the precipitate of NaBr was
removed and the solution was evaporated. Compound 5b
(0.090 g, 0.036 mmol) was obtained in 80% yield. Mp
1288C. 1H NMR (500 MHz, CDCl3) d (ppm): 1.12 (s, 36H,
ZC(CH3)3), 1.65–1.81 (m, 8H, ZCH2Z), 2.38–2.49 (m,
8H, ZCH2Z), 3.25–3.38 (m, 8H, ZCH2ZP), 3.47 (d, 4H,
J ¼ 13.0 Hz, Ar-CH2-Ar), 4.15–4.24 (m, 8H, Ar-OZCH2),
4.47 (d, 4H, J ¼ 13.0 Hz, Ar-CH2-Ar), 7.04 (s, 8H, Ar-H),
7.66–7.70 (m, 60H, P-Ar-H). 31P NMR (400 MHz, CDCl3)
d (ppm): 24.72 [Pþ], 2143.39 [PF6]. m/z (MALDI)
1447.60 (M2PF6)þ. Anal. Calcd for C132H144O4P4(PF6)4:
C, 63.46; H, 5.81. Found: C, 63.30; H, 5.78.
Tetrakis-(4-(P,P-diphenyl-P-methylphosphonium)-
butoxy)-p-tert-butylcalix[4]arene
tetrahexafluorophosphate (6b)
Compound 6b was obtained according to the same
procedure as for 5b with 6 (0.100g, 0.05mmol) and
1
AgPF6 (0.069g, 0.27mmol) in 80% yield. Mp 1558C. H
NMR (300MHz, CDCl3) d (ppm): 1.11 (s, 36H, ZC(CH3)3),
1.55–1.70 (m, 8H, ZCH2ZCH2ZP), 2.24–2.38 (m, 8H,
ZCH2ZCH2ZO), 2.42 (d, 12H, J ¼ 12.0Hz, CH3ZP),
2.95–3.10 (m, 8H, ZCH2ZP), 3.39 (d, 4H, J ¼ 12.8Hz, Ar-
CH2-Ar), 4.05–4.20 (m, 8H, ZCH2ZO), 4.42 (d, 4H,
J ¼ 12.8Hz, Ar-CH2-Ar), 7.00(s, 8H, Ar-H), 7.58–7.80(m,
40H, P-Ar-H, ortho, meta, para). 31P NMR (400MHz,
CDCl3) d (ppm): 24.50 [Pþ], 2143.07 [PF6], m/z (MALDI)
2104.80 (M2PF6)þ. Anal. Calcd for C112H136O4P4(PF6)4:
C, 59.79; H, 6.09. Found: C, 59.87; H, 6.03.
Tetrakis-(4-(P,P-diphenyl-P-methylphosphonium)-
butoxy)-p-tert-butylcalix[4]arene tetrabromide (6)
Into a 100 cm3 flask containing S5 (1.184 g, 1.00 mmol)
dissolved in chloroform (30 cm3), diphenylmethylpho-
sphine (4.004 g, 20.00 mmol) in chloroform (20 cm3) was
added. After 6 days under reflux, the mixture was cooled
and the solvent was evaporated. The product was purified
by precipitation from a 1:9 dichloromethane–hexane
Protonated tetrakis-(4-(P,P-diphenyl-phosphine)-butoxy)-
p-tert-butylcalix[4]arene tetrabromide (7)
Into a 50 cm3 flask containing S6 (1.510 g, 0.94 mmol) in
dichloromethane (10 cm3), HBr (5 cm3; solution of 33%

Supramolecular Chemistry
287
wt. in glacial acetic acid) in dichloromethane (15 cm3) was
added. The mixture was left stirred for 24 h at room
temperature. The mixture was then evaporated to give the
pure product 7 (0.954 g, 0.49 mmol) in 52% yield. Mp
1758C. 1H NMR (300 MHz, CDCl3) d (ppm): 1.22 (s, 36H,
ZC(CH3)3), 1.80–2.15 (m, 16H, OZCH2ZCH2ZCH2-
ZCH2ZP), 2.92–3.10 (m, 4H, Ar-CH2-Ar), 3.18–3.30
(m, 8H, CH2ZO), 3.40–3.55 (m, 8H, ZCH2ZP), 4.05–
4.27 (m, 4H, Ar-CH2-Ar), 7.05 (s, 8H, Ar-H), 7.60–8.05
(m, 40H, P-Ar-H, ortho, meta, para), 10.30–10.40 (m, 4H,
H-P). 31P NMR (400 MHz, CDCl3) d (ppm): 46.32 [Pþ].
m/z (MALDI) 1849.04 (M2Br)þ. Anal. Calcd for
complexation (DS) were then derived from the expressions
DG ¼ 2RT ln b and DG ¼ DH 2 TDS.
Ion-selective electrodes
THF was dried and freshly distilled before use for the
preparation of the ion-selective membranes. PVC, o-
NPOE, BEHS and (2-morpholino)ethanesulphonic acid
monohydrate (MES) were obtained from Fluka Selecto-
phore (Derbyshire, UK). The LiClO4, CsClO4 and sodium
salts, Cl2, Br2, I2, ClO24 , SCN2, NO23 , SO422, CO322
,
HPO422, PO432, Cr2O722, citrate, acetate, benzoate and
oxalate, were of p.a. grade. All aqueous salt solutions were
prepared with demineralised water (conductivity
,1.0 mS/cm).
C108H128O4P4Br4: C, 67.08; H, 6.67. Found: C, 66.84;
H, 6.92.
The membranes were composed of 4 mg of ionophores
1–6 and 8, 60 mg of PVC and 120 mg of plasticiser. All the
components were dissolved in 1.5 cm3 of dried, freshly
distilled THF and the solutions were poured into glass rings
of 24 mm in diameter. The solutions were left for 24 h for
slow solvent evaporation giving the mother membranes of a
thickness of about 0.1 mm. Several membranes of 7 mm in
diameter were cut from each mother membrane and were
incorporated into the Ag/AgCl electrode bodies of IS 561
type (Moeller SA, Zurich, Switzerland). The two
plasticisers, BEHS and o-NPOE, were used for the
preparation of the membranes. However, electrodes with
membranes based on NPOE had the best lifetime and
response characteristics. The EMF measurements were
carried out at zero current conditions using a Lawson Lab
16 EMF station (multichannel millivoltmeter) or a
Metrohm 654 millivoltmeter. A double-junction reference
Radelkis 0P0820P electrode with a 1 M CH3COOLi
solution in the bridge cell was used. The measurements
were carried out using cells of the type
NMR studies
Five milligrams of ligand were introduced with 6
equivalents of the solid alkali metal salts (Liþ, Naþ,
Csþ) or tetraethylammonium perchlorate in a glass vessel
and dissolved in a small volume of deuterated chloroform
or acetonitrile. After manual shaking for a few minutes,
the mixture was left in contact for 24 h before filtration of
the excess of salt, if necessary. The solution (or the filtrate)
was then sampled in a NMR tube and its spectrum
recorded on Bruker SY300 MHz or SY400 MHz spec-
trometers equipped for H or 31P resonances.
1
Microcalorimetric studies
Microcalorimetric titrations were performed using a 2277
Thermal Activity Monitor Microcalorimeter (Thermo-
metric). Titration was carried out at 258C on 2.7 cm3 of
1025 to 5 £ 1024 M solutions of the ligand in acetonitrile
using a glass cell of 4 cm3. The heat changes were
Ag=AgClj1 M KClj1 M CH3COOLijsamplej
jmembranej
j0:05 M MES=NaOH; 0:01 M NaCljAgCl=Ag:
measured after injection of 15 £ 15 ml of 1023 and 1022
M
At least three identical electrodes of the same membrane
composition and containing the same inner electrolyte
were prepared (39). The studies were repeated several
times over the period of 1 month.
LiClO4, LiBr, NaClO4, NaPF6 or Et4NClO4 solutions in
the same solvent. Chemical calibration was made by the
determination of the complexation enthalpy of Ba2þ with
18C6 in water or of Rbþ with 18C6 in methanol, as
recommended (37). Values of the stability constants (b)
and of the enthalpies of complexation (DH) were refined
simultaneously from these data using the ligand-binding
analysis program DIGITAM version 4.1 (38) and after
correction for the heat of dilution determined in separate
experiments by adding the salt solutions to 2.7 cm3 of the
pure solvent. The values of the corresponding entropies of
To reduce the pH changes during the titrations,
solutions were prepared with 0.05 M MES/NaOH buffer of
pH 5.5 (MES). All salt solutions contained 1022 M NaCl
as the supporting electrolyte (22).
pot
The selectivity coefficients K of the electrodes were
A;B
determined by the separate solution method and in some
cases by the fixed interference method (40–42). The
calibration curves were obtained by the addition of standard

288
R. Pomecko et al.
solutions of different anions to 50 cm3 of 0.01 M NaCl in
0.05 M MES/NaOH buffer solution of pH 5.5. The
concentration of the primary anion [A] was increased
from 1027 to 1022 M. They were also measured by
successive dilution of initial 5 £ 1022 M salt solutions until
further dilution resulted in no potential change.
(18) Miao, R.; Zheng, Q.Y.; Chen, C.F.; Huang, Z.T.
Tetrahedron Lett. 2004, 45, 4959–4962.
(19) Jagessar, R.C.; Shang, M.; Scheidt, W.R.; Burns, D.H.
J. Am. Chem. Soc. 1998, 120, 11684–11692.
(20) Gale, P.A.; Sessler, J.L.; Kral, V.; Lynch, V. J. Am. Chem.
Soc. 1996, 118, 5140–5141.
(21) Gale, P.A.; Sessler, J.L.; Allen, W.E.; Tvermoes, N.A.;
Lynch, V. Chem. Commun. 1997, 665–666.
(22) Pomecko, R.; Asfari, Z.; Hubscher-Bruder, V.; Bochenska,
M.; Arnaud-Neu, F. Supramol. Chem. 2007, 19, 459–466.
(23) Hamdi, A.; Nam, K.C.; Ryu, B.J.; Kim, J.S.; Vicens, J.
Tetrahedron Lett. 2004, 45, 4689–4692.
Acknowledgements
R.P. gratefully acknowledges the Gdansk University of
(24) Broder, C.K.; Davidson, M.G.; Forsyth, V.T.; Howard,
J.A.K.; Lamb, S.; Mason, S.A. Crys. Grow. Des. 2002, 2,
163–169.
`
Technology, Poland, and the Ministere de la Recherche et des
Nouvelles Technologies, France (Cotutelles de theses), for
financial support.
`
(25) Issleib, K.; Muller, D.-W. Chem. Berichte 1959, 92,
3175–3182.
(26) Clark, P.W. Org. Prep. Proc. Int. 1979, 11, 103–106.
(27) Mohr, W.; Horn, C.R.; Stahl, J.; Gladysz, J.A. Synthesis
2003, 8, 1279–1285.
(28) Baklouti, L.; Cherif, J.; Abidi, R.; Arnaud-Neu, F.;
Harrowfield, J.; Vicens, J. Org. Biomol. Chem. 2004, 2,
2786–2792.
(29) Savage, P.B.; Holmgren, S.K.; Gellman, S.H. J. Am. Chem.
Soc. 1994, 116, 4069–4070.
(30) Errachid, A.; Perez-Jimenez, C.; Casabo, J.; Escriche, L.;
Munoz, J.A.; Bratov, A.; Baussels, J. Sens. Actuators B
1997, 43, 206–210.
(31) Shamsipur, M.; Soleymanpour, A.; Akhond, M.; Sharghi,
H.; Hasaninejad, A.R. Sens. Actuators B 2003, 89, 9–14.
(32) Dinten, O.; Spichiger, U.E.; Chaniotakis, N.; Gehrig, P.;
Rusterholz, B.; Morf, W.E.; Simon, W. Anal. Chem. 1991,
63, 596–603.
References
(1) Bianchi, A., Bowman-James, K., Garcia-Espana, E., Eds.
Supramolecular Chemistry of Anions; John Wiley-VCH:
New York, 1997; Chapter 3, pp 63–78.
(2) Kanyo, Z.F.; Christianson, D.W. J. Biol. Chem. 1991, 266,
4264–4268.
(3) Pardee, A.P. J. Biol. Chem. 1966, 241, 5886–5896.
(4) Puljak, L.; Kilic, G. Biochim. Biophys. Acta 2006, 1762,
404–413.
(5) Weiner, M.L.; Salminen, W.F.; Larson, P.R.; Barter, R.A.;
Kranetz, J.L.; Simon, G.S. Food Chem. Toxicol. 2001, 39,
759–786.
(6) Motzer, W.E. Environ. Forensic 2001, 2, 301–311.
(7) Bondy, C.R.; Gale, P.A.; Loeb, S.J. J. Am. Chem. Soc. 2004,
126, 5030–5031.
´
(8) Amendola, V.; Boiocchi, M.; Esteban-Gomez, D.;
(33) Lizondo-Sabater, J.; Segui, M.J.; Loris, J.M.; Martinez-
Manez, R.; Pardo, T.; Sancenon, F.; Soto, J. Sens. Actuators
B 2004, 101, 20–27.
Fabbrizzi, L.; Monzani, E. Org. Biomol. Chem. 2005, 3,
2632–2639.
(9) Brooks, S.J.; Edwards, P.R.; Gale, P.A.; Light, M.E. New
J. Chem. 2006, 30, 65–70.
(34) Segui, M.J.; Lizondo-Sabater, J.; Martinez-Manez, R.;
Soto, J. Analyst 2002, 127, 387–390.
(10) Turner, D.R.; Paterson, M.J.; Steed, J.W. J. Org. Chem.
2006, 71, 1598–1608.
(35) Casabo, J.; Escriche, L.; Perez-Jimenez, C.; Munoz, J.A.;
Teixidor, F.; Baussells, J.; Errachid, A. Anal. Chim. Acta
1996, 320, 63–68.
(36) Schaller, U.; Bakker, E.; Spichiger, U.E.; Pretsch, E. Anal.
Chem. 1994, 66, 391–398.
(11) Boerrigter, H.; Grave, L.; Nissink, J.W.M.; Chrisstoffels,
L.A.J.; van der Maas, J.H.; Verboom, W.; de Jong, F.;
Reinhoudt, D.N. J. Org. Chem. 1998, 63, 4174–4180.
(12) Lee, K.H.; Hong, J.I. Tetrahedron Lett. 2000, 41,
6083–6087.
(37) Arnaud-Neu, F.; Delgado, R.; Chaves, S. Pure Appl. Chem.
2003, 75, 71–102.
(13) Chmielewski, M.J.; Jurczak, J. Tetrahedron Lett. 2004, 45,
6007–6010.
(14) Beer, P.D.; Drew, M.G.B.; Gradwell, K. J. Chem. Soc.
Perkin Trans. 2, 2000, 511–519.
(38) Hallen, D. Pure Appl. Chem. 1993, 65, 1527–1532.
(39) Bochenska, M.; Zielenska, A.; Pomecko, R.; Hubscher-
Bruder, V.; Arnaud-Neu, F. J. Incl. Phenom. 2005, 52,
129–134.
(15) Dietrich, B. Pure Appl. Chem. 1993, 65, 1457–1464.
(16) Egorov, V.V.; Rakhman’ko, E.M.; Okaev, E.B.; Pomele-
nok, E.V.; Nazarov, V.A. Talanta 2004, 63, 119–130.
(17) Matthews, S.E.; Beer, P.D. In Calixarenes 2001; Asfari, Z.,
(40) Buck, R.P.; Linder, E. Pure Appl. Chem. 1994, 66,
2527–2536.
(41) Umezawa, Y.; Buhlmann, P.; Umezawa, K.; Tohda, K.;
Amemiya, S. Pure Appl. Chem. 2000, 72, 1851–2082.
(42) Lindner, E.; Umezawa, Y. Pure Appl. Chem. 2008, 80,
85–104.
¨
Bohmer, V., Harrowfield, J., Vicens, J., Eds.; Kluwer
Academic: Dordrecht, 2001; Chapter 23, pp 421–439.