The first acid constant of calix[4]arenes

Article abstract of DOI:10.1039/a607259c

A series of calix[4]arenes containing a single p-nitrophenol unit (or two p-nitrophenol units) have been synthesized by fragment condensation. Their first acid constant (pK_a1) has been determined in 2-methoxyethanol-water (9:1) by optical titration. Relative to the corresponding linear trimers with a p-nitrophenol in the middle, a decrease of pK_a1 by 2.1 units or more is observed, which can be explained entirely by intramolecular hydrogen bonds stabilising the monoanion. In this way electron-withdrawing p-substituents in the opposite phenolic unit lead to a further decrease in pK_a1, while a distortion of the cone conformation by m-methyl groups causes a slight increase. The structure of one calix[4]arene was further confirmed by single crystal X-ray analysis showing the molecule in the usual cone conformation.

Full text of DOI:10.1039/a607259c

View Article Online / Journal Homepage / Table of Contents for this issue
The first acid constant of calix[4]arenes
Michaela Backes,a Volker Böhmer,*,a George Ferguson,b Cordula Grüttner,a
Christian Schmidt,a Walter Vogt a and Kadija Ziat a
a
Institut für Organische Chemie, Johannes-Gutenberg-Universität, J.-J.-Becher-Weg 34, SB1,
D-55099 Mainz, Germany
b Department of Chemistry and Biochemistry, University of Guelph, Guelph, Ontario, Canada
N1G 2W1
A series of calix[4]arenes containing a single p-nitrophenol unit (or two p-nitrophenol units) have been
synthesized by fragment condensation. Their first acid constant (pKa1) has been determined in 2-
methoxyethanol–water (9 :1) by optical titration. Relative to the corresponding linear trimers with a
p-nitrophenol in the middle, a decrease of pKa1 by 2.1 units or more is observed, which can be explained
entirely by intramolecular hydrogen bonds stabilising the monoanion. In this way electron-withdrawing
p-substituents in the opposite phenolic unit lead to a further decrease in pKa1, while a distortion of the
cone conformation by m-methyl groups causes a slight increase. The structure of one calix[4]arene was
further confirmed by single crystal X-ray analysis showing the molecule in the usual cone conformation.
and we present in this paper pKa1 values, which are consistent
within this series, showing steric and electronic effects exerted by
substituents in the other phenolic units.
The aim of our studies was the comparison of different
calix[4]arenes (CalH4) and their corresponding model com-
pounds with respect to their acidity [equilibrium (1)], or more
Introduction
Calixarenes are macrocyclic molecules in which phenolic units
are linked together via methylene bridges ortho to the phenolic
hydroxy groups.1 Calix[4]arenes,† the smallest members of the
family, normally assume the so-called cone conformation which
is stabilised by a cyclic array of intramolecular hydrogen bonds
between the endo hydroxy groups. This close vicinity of OH
groups must have a pronounced influence on their acidity due
to both intramolecular hydrogen bonding and electrostatic
interactions.
CalH4 ϩ B
CalH3Ϫ ϩ BHϩ
(1)
precisely their first acid constants [eqn. (2)]. In principle this
[CalH3Ϫ][BHϩ]
The determination of pKa values for calixarenes is compli-
cated by the fact that one is dealing with at least tetraprotonic
acids. Furthermore, most of the usual calixarenes are only spar-
ingly soluble, especially in water or water-containing solvents,
which seriously hampers the application of standard techniques
like potentiometric titration. Thus, the first studies of p-
nitrocalix[4]arene led to rather divergent values,2,3 especially for
pKa2, although they were in agreement with theoretical calcu-
lations4 that pKa1 is much lower than in comparable model
compounds. Meanwhile more reliable values were found for the
dissociation of the first two phenolic OH groups in calix[4]arene-
tetrasulfonate. For instance pKa values of 3.34 and 11.55 and
3.26 and 11.8,6 respectively, were reported by two groups.
Our own interest has been directed towards calix[4]arenes
containing just one p-nitrophenol unit, which represents the
most acidic phenolic unit, the dissociation of which can be eas-
ily monitored by the change in UV absorption. Thus, structural
influences on pKa1 caused by the other phenolic units can be
studied conveniently, following a strategy which has been suc-
cessfully applied to various linear oligomers.7 An early result
obtained in methanol–water (1:1) indicated the influence of the
p-substituent in the phenolic unit opposite to the p-nitrophenol
unit,8 which seemed to be even more pronounced for two series
of calix[4]arenes in methanol–water (9:1).9 Although a tenta-
tive explanation could be a conformational distortion of the
cone conformation, this explanation remained unsatisfactory
since such a distortion could not be found by other methods,
including single crystal X-ray analysis.10
(2)
Ka1
=
[CalH4]
equilibrium, where B represents the (basic) solvent (or even a
mixture of solvent bases) being present in excess and BHϩ the
protonated solvent (or the respective mixture of protonated
solvents), can be studied in any solvent (mixture). Relative
values for Ka1, which can be compared with a series of com-
pounds, are obtained as long as all members of this series are
studied under identical conditions.
The choice of an appropriate solvent was, nevertheless, a dif-
ficult problem. Most of the calixarenes are sparingly soluble,
especially in aqueous solvents, while many buffer systems are of
low solubility (at least in some pH regions) in more organic
mixtures. After numerous tests we have chosen for the present
study 2-methoxyethanol–water (9:1) [only some control
measurements were made for comparison in methanol–water
(9:1)], which enabled a convenient continuous variation in
[BHϩ] by the titration of a mixture of weak acids by a
strong base. In this solvent, a glass electrode assumes a stable
potential within reasonably short times. Thus, a potentiometric
characterisation of [BHϩ] was possible by an apparent pH
value, while the ratio [CalH3Ϫ]/[CalH4] of monoanion and cal-
ixarenes was determined by measurement of the UV–VIS
absorption.
Results and discussion
We therefore extended these studies to other calix[4]arenes
Syntheses
All compounds were prepared by the well known fragment
condensation 11–14 of linear trimers 1, containing the p-nitro-
phenol unit in the middle, with the bis(bromomethyl)ated [or
† IUPAC name: calix[4]arene = 12,32,52,72-tetrahydroxy-1,3,5,7(1,3)-
tetrabenzena-cyclooctaphane.
J. Chem. Soc., Perkin Trans. 2, 1997
1193

View Article Online
Table 1 Apparent pKa1 values of p-nitrophenol (pN), selected trimers
and calix[4]arenes in methanol–water 9:1
Compound
pN
pKa1
Compound
pKa1
8.90
3a
3b
3c
3d
4.99
4.77
4.97
4.82
1b
1c
6.97
7.03
Table 2 Apparent pKa1 values obtained in 2-methoxyethanol–water
9:1
Compound
pN
pKa1
Compound
pKa1
8.73
3f
3g
3h
3i
3.27
3.15
3.27
2.40
1b
1c
1d
1e
6.15
6.11
6.14
6.31
4
1.98
2a
2b
2c
2d
3.24
3.30
3.87
3.75
5a
5b
5c
4.01
3.30
3.58
6a
6b
6c
6d
6e
4.37
4.68
3.88
4.03
4.18
3a
3b
3c
3d
3e
4.01
3.95
4.01
4.00
3.76
the bis(bromomethyl)ated phenol and in the trimer (3% for 6b
and 6d) and drops to 0.6% for 6e. Compound 4 was obtained in
a similar way by 2 ϩ 2-fragment condensation 11 of the p-tert-
butylphenol dimer and the bis(bromomethyl)ated dimer of p-
nitrophenol. The acids 2b, 2d and 3h were obtained by alkaline
hydrolysis of the respective esters 2a, 2c and 3g.
The structure of all calix[4]arenes was unambiguously con-
firmed by 1H NMR and mass spectra and in the case of 3e, also
by a single crystal X-ray analysis.
pKa values
First determinations of pKa1 were made for compounds 3a–d in
methanol–water (9:1), the solvent mixture formerly used.9
These results are collected in Table 1. As expected, the trinuclear
compounds 1b,c show a decrease in pKa1 of ca. 1.9 units when
compared with p-nitrophenol, which is easily explained by a
stabilisation of the monoanion by intramolecular hydrogen
bonds. A further decrease of ca. 2.0–2.2 units (3.9–4.1 relative
to p-nitrophenol) is observed for all calix[4]arenes studied and
an alkyl substituent in the p-position of the opposite phenolic
unit [CH3C(CH3)3, C6H11, C12H25 in this series] obviously has
no remarkable influence.
To obtain a broader view, we switched to 2-methoxyethanol–
water (9:1) as the solvent system, and the pKa1 values obtained
here are collected in Table 2. For compounds 3a and 4, the UV
spectra are shown as a function of the apparent pH (optical
titration) in Fig. 1.The spectra of all the compounds containing
one p-nitrophenol unit show an isosbestic behaviour over the
whole pH range [Fig. 1(a)]. For the calix[4]arenes 3i and 4 con-
taining two p-nitrophenol units no deviations from a first set of
isosbestic points are found for pH < (pKa1 ϩ 3), while all
spectra pass a second set of isosbestic points above pH = 10 [cf.
Fig. 1(b)].
bis(hydroxymethyl)ated] phenol that has to form the opposite
phenolic unit in the calix[4]arene. Compounds 1 themselves
were obtained in yields of 65–75% by condensation of 2,6-
bis(bromomethyl)-4-nitrophenol with an excess of a 4-alkyl- or
3,4-dialkyl-phenol. The macrocyclisation reaction was carried
out in dioxane using TiCl4 as the Friedel–Crafts catalyst and
(probably) as template. The calix[4]arenes 2–6 are easily isolated
and purified by flash chromatography (in the case of 2a, 2c and
3g after complete reesterification of the crude product); the
yield of pure product, which is not optimised for the special
examples, varies between 8–27% for compounds 2 and 3, the
lowest yields (2a and 3g) being obtained with the bis(bromo-
methyl)ated ester of p-hydroxybenzoic acid. However, no
general conclusions on the influence of the p-substituent in the
bromomethylated phenol should be drawn from these results.
The yield is significantly lower if m-methyl groups are present in
The evaluation of the pKa1 values is illustrated by typical
plots in Fig. 2. From eqn. (2), eqn. (3) may be deduced, where
ε Ϫ ε(HA)
ε(AϪ) Ϫ ε
(3)
log
= pH Ϫ pKa
1194
J. Chem. Soc., Perkin Trans. 2, 1997

View Article Online
Fig. 2 Evaluation of the pKa1 values for the calix[4]arenes 2a, 3a, 3e
and 3i; typical plots using eqn. (2)
bond from the opposite hydroxy group seems less likely, since
the distance between distal phenolic oxygens in calix[4]arenes is
in the range of 3.7–3.8 Å; compare the X-ray structure below.)
The increased acidity of the opposite phenolic unit in 3e–3i
favours or strengthens the H-bond (β) and consequently the
H-bonds (α), which explains the observed decrease in pKa1 for
these compounds.
Stronger intramolecular hydrogen bonding in the monoan-
ions of the calix[4]arenes is also indicated by the wavelength of
the absorption maximum, which is found at lower values (396–
404 nm) than in the case of the linear trimers (409–411 nm). A
similar hypsochromic effect was observed for the linear oligo-
mers with the p-nitrophenol at the end for an increasing number
of phenolic units.7
Not unexpectedly, the pKa1 value of 4, with two adjacent p-
nitrophenol units, is further decreased by 0.42 in comparison
with the isomeric 3i, while the pKa2 value for both compounds is
13.2 (±0.2). This is in agreement with the most probable hydro-
gen bond pattern for the mono- and di-anion, as sketched in
Scheme 1.
Fig. 1 (a) UV spectra of compound 3a as a function of the apparent
pH (pH values of the single curves: 2.21, 2.71, 3.29, 3.57, 3.83, 4.19,
4.40, 4.66, 5.10 and 5.74); (b) UV spectra of compound 4 as function of
the apparent pH (pH values of the single curves: 1.29, 1.47, 1.55, 1.62,
1.68, 1.74, 1.81, 1.88, 1.94, 2.00, 2.08, 2.14, 2.23, 2.40, 2.81, 3.43, 3.99,
5.05, 10.12, 10.23, 11.58, 12.20, 12.40, 12.60, 12.79, 12.98 and 13.10)
Most probably the carboxylic groups in 2b and 3h are de-
protonated in a similar pH range as the p-nitrophenol unit.§
However, obviously this dissociation, if it occurs, does not
intervene with our determinations in a measurable way, since
neither the isosbestic behaviour of the UV spectra is disturbed,
nor the linearity or the slope of plots according to eqn. (3).
Perhaps the slightly higher pKa1 values for 2b and 3h in com-
parison with 2a and 3g are an indication for the dissociation of
the carboxylic groups. However, like the opposite difference in
2c/2d, this should not be overinterpreted. Comparison of 2a/2b
and 3g/3h shows that the alkyl substituent R1 (Me/But) has no
remarkable influence on pKa1, which is also the basis for the
following discussion.
Methyl groups in the meta position to the phenolic hydroxy
groups should not change the acidity of this phenolic unit but
should generally cause a distortion of the C4 symmetrical cone
conformation of calix[4]arenes. This has been demonstrated by
the X-ray structures for octamethyl calix[4]arenes derived from
3,4-dimethylphenol15 (four m-methyl groups) and 3,5-di-
methylphenol16 (eight m-methyl groups), and in the latter case
also by low temperature 1H NMR spectroscopy in solution. We
were interested if such a distortion, caused by the opposite
phenolic unit (cf. ref. 10), influences the dissociation of the p-
nitrophenol unit. Comparison of 3a with 5a and 3b with 5b
shows that a single m-methyl group has no detectable effect,
which is in agreement with the X-ray structure of a similar
calix[4]arene with a single m-methylphenol unit.17 Two m-
methyl groups at the opposite phenolic unit, e.g. as in 5b, cause
an increase in pKa1 of 0.3 in comparison with 3c.
ε is the molar absorption coefficient obtained at a given wave-
length at the apparent pH, and ε(HA) and ε(AϪ) are the molar
absorption coefficients in acidic and in alkaline solutions,
respectively (see Experimental section). Plots of eqn. (3) are
linear with a slope of 1, and the apparent pKa is obtained as
the intercept with the pH axis. Repeated determinations and
evaluations at different wavelengths usually led to deviations
<0.05 (for details see the Experimental section), but for the
following discussion only differences greater than 0.10 are
considered.
As in methanol–water (9:1), the pKa1 values of the trimers
1b–e are nearly identical (6.1–6.3), but are lower than p-nitro-
phenol (8.7) by ca. 2.5 units. Again a further decrease in pKa1
by more than 2 units is observed when this trimeric unit is
incorporated in a calix[4]arene. Comparison of the pKa1 values
of compounds 3a–d (3.9–4.0) shows again that different alkyl
groups in the para position of the opposite phenolic unit have
no influence on the acidity of the p-nitrophenol unit. A slight
decrease is observed for R2 = C6H5 (3e, 3.75) and for electron-
withdrawing substituents like R2 = Cl, COOR and NO2, the
pKa1 decreases by 0.75, 0.85 and 1.30.‡
All these results can be entirely understood by stabilisation
of the monoanion by two intramolecular hydrogen bonds (α),
as indicated in Scheme 1. Such hydrogen bonds are possible
also in the trimers 1, but their influence is less pronounced due
to the higher rotational freedom. Additional stabilisation stems
from the hydrogen bond (β) of the opposite phenolic unit, an
effect which was also observed in linear compounds.7 (The for-
mation of a ‘three-centred’ hydrogen bond or a direct hydrogen
The deformation may be stronger if two meta positions
adjacent to each other are substituted, which prompted us, to
‡ When comparing the calix[4]arene 3i containing two p-nitrophenol
units with the other compounds, it must be kept in mind that already for
statistical reasons the pKa1 of a diprotonic acid is lower by log 2 = 0.3
(while pKa1 is higher by log 2) in comparison with the corresponding
monoprotonic acid, even if there is no mutual influence of the two
acidic functions.
§ The pKa1 of p-hydroxybenzoic acid is lower by ca. 2.5 than the pKa of
p-nitrophenol (in aqueous solution), which is not too different from the
decrease in the pKa1 observed for the p-nitrophenol unit in a calix-
[4]arene in comparison with p-nitrophenol itself.
J. Chem. Soc., Perkin Trans. 2, 1997
1195

View Article Online
Table 3 Summary of crystal data, data collection, structure solution
and refinement details
3e
(a) Crystal data
Formula
Molar mass
Colour, habit
Crystal size/mm
Crystal system
a/Å
2(C42H43NO6), 0.57(CHCl3)
1383.7
Colourless, block
0.40 × 0.36 × 0.33
Triclinic
13.893(3)
b/Å
15.266(2)
c/Å
20.065(4)
α/Њ
75.14(2)
β/Њ
75.40(2)
γ/Њ
81.548(4)
3964.9(13)
¯
P1
V/Å3
Space group
Z
2
F(000)
1466
1.159
0.132
Dcalc/g cmϪ3
µ/mmϪ1
(b) Data acquisition a
Temp./K
294(1)
Unit-cell reflcns (θ rangeЊ)
Max. θ/Њ for reflcns
hkl Range of reflcns
Variation in 3 standard reflcns
Reflcns measured
25 (16.2 20.5)
26.9
Ϫ17 to 17; 0 to 19; Ϫ24 to 25
1.0%
17 226
17 226
6695
Scheme
1
Schematic representation of intramolecular hydrogen
bonds in (a) trimers 1, (b) calix[4]arenes 2, 3, (c) 4 and of (d) the
dianions of 3i (only one direction is shown) and (e) 4
Unique reflcns
Reflcns with I > 2σ(I)
synthesize calix[4]arenes 6. Comparison of 3a with 6a and
especially of 3f with 6c shows an increase in pKa1 by 0.36 and
0.6, respectively, which most probably is due to a similar steric
influence of the two m-methyl groups as discussed above, caus-
ing a distortion of the cone conformation. (A slight increase in
pKa1 is seen, however, also for the trimer 3e in comparison with
3b.) If now one or two methyl groups are ‘introduced’ in posi-
tions R1 and R3 of 6c, pKa1 increases by 0.15 (6d) and 0.3 (6e).
In conclusion, a deformation of the calix[4]arene skeleton by
m-methyl groups in the phenolic unit opposite to the dissoci-
ating p-nitrophenol unit weakens the intramolecular hydrogen
bonds and causes changes in pKa1 up to 0.9, the difference for 3f
and 6e. It is reasonable to assume that a similar deformation of
the calixarene skeleton by m-methyl groups ‘between’ the p-
nitrophenol and the adjacent phenol units leads to a larger
increase in pKa1.
(c) Structure solution and refinement b
Refinement on
F 2
Solution method
H-atom treatment
No. of variables in L.S.
Weights:
Direct methods
Riding
928
k in w = 1/[σ2(Fo2) ϩ k]
(0.1077P)2
2
[P = (Fo2 ϩ 2Fc )/3]
R(obs, F), RwF 2 (all), g.o.f.
Density range in final ∆-map/e ÅϪ3
Final shift/error ratio
0.079, 0.233, 1.01
Ϫ0.352, 0.433
Ϫ0.002
a Data collection on an Enraf-Nonius CAD4 diffractometer with graph-
ite monochromatised Mo-Kα radiation (λ 0.7107 Å). All calculations
were done on an IBM Aptiva 166 MHz system with the NRCVAX
system of programs (E. J. Gabe, Y. Le Page, J.-P. Charland, F. L. Lee
and P. S. White, J. Appl. Cryst., 1989, 22, 384) for refinement with
observed data on F, or with SHELXL-93 (G. M. Sheldrick, 1993) for
refinement with all data on F 2.
b
X-Ray structure of 3e
Single crystals of 3e suitable for X-ray diffraction studies were
obtained by allowing a chloroform solution to evaporate to dry-
ness. Details of the X-ray experimental conditions, cell data,
structure solution and refinement are shown in Table 3.¶
The asymmetric unit contains two crystallographically in-
dependent molecules of 3e which have similar open-cone con-
formations as a result of the formation of eight-membered
intramolecular hydrogen-bonded rings with O ؒ ؒ ؒ O 2.652(4)–
2.701(4) Å (Fig. 3). The calix[4]arene conformation is described
by the interplanar angles between the aromatic rings and the
plane of the CH2 groups which join them. For the plane C17,
C27, C37, C47, these values are 122.9(1)Њ [C11–C16], 126.8(1)Њ
[C21–C26], C124.6(1)Њ [C31–C36] and 124.5(1)Њ [C41–C46].
The corresponding values for the second molecule and the C57,
C67, C77, C87 plane are 122.9(1)Њ [C51–C56], 129.5(1)Њ [C61–
C66], C121.2(1)Њ [C71–C76] and 131.0(1)Њ [C81–C86]. The exo-
cyclic phenyl rings make angles of 38.1(1)Њ [C38–C313] and
30.4(2)Њ [C78–C713] with the calixarene aromatic ring to which
they are bonded. The nitro group in each molecule is rotated
away from coplanarity with its associated aromatic ring, but in
different directions [ϩ15.8(3)Њ (N18, O19, O110) and Ϫ8.7(3)Њ
(N58, O59, O510)].
The two calixarenes pack around independent inversion
centres such that a tert-butyl group of one molecule is inside the
cavity of the inversion related one and vice versa (Fig. 4). In one
such pair, the interplanar distance between ring C21–C26 and
that related by the inversion centre at (0, 0.5, 0.5) is 3.53 Å; the
corresponding distance for the second pair between ring
C81–C86 and the molecule related by the inversion centre
at (0.5, 0, 0) is much larger (4.95 Å). This difference found for
the two analogous pairs is clearly caused by packing effects. It
demonstrates how carefully such intermolecular distances of
host–guest interactions must be discussed. The shortest intra-
molecular contacts are between disordered methyl C atoms
(C211, C214) and atom C23 (at Ϫx, 1 Ϫ y, 1 Ϫ z) with dis-
tances 3.60 and 3.61 Å; in the other dimer the corresponding
separations (3.71 and 3.62 Å) are between (C811, C814) and
C81 and C86 (at 1 Ϫ x, Ϫy, Ϫz). Comparable packing of
dimers has been previously observed 10 in a calix[4]arene very
similar to 3e with the exocyclic phenyl ring replaced by a methyl
group.
¶ Atomic coordinates, displacement parameters and dimension tables
have been deposited at the Cambridge Crystallographic Data Centre.
See ‘Instructions for Authors’, J. Chem. Soc., Perkin Trans. 2, 1997,
Issue 1. Any request to the CCDC for the material should quote the full
literature citation and reference number 188/68.
1196
J. Chem. Soc., Perkin Trans. 2, 1997

View Article Online
glacial acetic acid was heated to 90 ЊC and 2,6-bis(bromo-
methyl)-4-nitrophenol (10 g, 31 mmol) was added. After 5 h the
solution was cooled to room temperature and 80 ml water was
added. The precipitate was filtered, washed with water and
purified as indicated for the individual compounds.
2,6-Bis(2-hydroxy-5-methylbenzyl)-4-nitrophenol (1a). Meth-
od A: recrystallisation from chloroform gave 8.4 g (71%) of
light-yellow crystals, mp 219 ЊC (lit.20: 51%; mp 219 ЊC) (Found:
C, 69.0; H, 5.65; N, 3.7. C22H21NO5 requires C, 69.65; H, 5.6; N,
3.7%).
2,6-Bis(2-hydroxy-5-tert-butylbenzyl)-4-nitrophenol
(1b).
Method A: recrystallisation from chloroform gave 9.7 g (68%)
of white crystals, mp 212 ЊC; δH (200 MHz; CDCl3) 8.11 (2 H, s,
ArH), 7.28 (2 H, d, J 2, ArH), 7.10 (2 H, dd, J 2, 9, ArH), 6.74
(2 H, d, J 9, ArH), 3.97 (4 H, s, ArCH2Ar) and 1.25 [18 H, s,
C(CH3)3].
2,6-Bis(2-hydroxy-5-cyclohexylbenzyl)-4-nitrophenol
(1c).
Method A: recrystallisation from chloroform gave 11.1 g (70%)
of light-yellow crystals, mp 216 ЊC; δH (200 MHz; CDCl3) 8.11
(2 H, s, ArH), 7.10 (2 H, s, ArH), 6.91 (2 H, d, J 8, ArH), 6.72 (2
H, d, J 8, ArH), 3.94 (4 H, s, ArCH2Ar), 2.40 (2 H, br, ArCH),
1.77 (10 H, br, CH2) and 1.31 (10 H, br s, CH2).
Fig. 3 View of one of the two independent calixarene molecules 3e
and our numbering scheme. Anisotropic displacement ellipsoids are
drawn at the 30% probability level.
2,6-Bis(2-hydroxy-4,5-dimethylbenzyl)-4-nitrophenol
(1d).
Method A: recrystallisation from chloroform gave 9.5 g (76%)
of white crystals, mp 224 ЊC (decomp.); δH (200 MHz; [2H6]-
DMSO) 10.03 (1 H, s, ArOH), 9.31 (2 H, s, ArOH), 7.62 (2 H, s,
ArH), 6.85 (2 H, s, ArH), 6.65 (2 H, s, ArH), 3.83 (4 H, s,
ArCH2Ar), 2.12 (6 H, s, CH3) and 2.07 (6 H, s, CH3).
2,6-Bis(2-hydroxy-4-methyl-5-isopropylbenzyl)-4-nitrophenol
(1e). Method B: the precipitate was dissolved in chloroform,
extracted three times with water and the solution dried over
sodium sulfate. The solvent was evaporated and the crude
product purified by flash chromatography with light petroleum
(bp 40–60 ЊC)–acetone (4:1) to give 9.2 g (65%) of pale yellow
crystals, mp 191 ЊC (Found: C, 72.2; H, 7.15; N, 3.0. C28H33NO5
requires C, 72.55; H, 7.2; N, 3.0%); δH (200 MHz; CDCl3) 8.09 (2
H, s, ArH), 7.11 (2 H, s, ArH), 6.62 (2 H, s, ArH), 3.93 (4 H, s,
ArCH2Ar), 3.01 [2 H, sept., J 6.8, CH(CH3)2], 2.21 (6 H, s, CH3)
and 1.20 [12 H, d, J 6.8, CH(CH3)2].
General procedure for the synthesis of calix[4]arenes
To a suspension of TiCl4 (3.3 ml, 30 mmol) in dry dioxan (250
ml) a solution of the appropriate 2,6-bis(bromomethyl)ated
phenol (5 mmol) [for the synthesis of 5b,c and 6a,c–e the 2,6-
bis(hydroxymethyl)ated phenol was used instead] and 1 (5
mmol) was slowly added (over 24 h) at 100 ЊC. The reaction
mixture was stirred at 110 ЊC for a further 12 h. The dark red
solution was evaporated in vacuum (eventually reesterified), the
residue was dissolved in 400 ml chloroform, silica gel (50 g) was
added and the resultant mixture was evaporated again. The
silica gel was extracted for 24 h with chloroform in a Soxhlet
apparatus. The crude product thus obtained after evaporation
was purified by flash chromatography. Further details are given
for the individual compounds.
17-Ethoxycarbonyl-11,23-dimethyl-5-nitro-25,26,27,28-tetra-
hydroxycalix[4]arene (2a). The crude product was reesterified
by refluxing with ethanol for 3 h prior to the Soxhlet extraction.
Purification by flash chromatography with chloroform–ethyl
acetate (10:1) and recrystallisation from chloroform–methanol
gave 235 mg (12%) of colourless crystals, mp 327 ЊC (decomp.)
(Found: C, 69.7; H, 5.45; N, 2.35. C33H31NO8 requires C, 69.6;
H, 5.5; N, 2.5%); δH (200 MHz; [2H6]DMSO) 8.08 (2 H, s, ArH),
7.72 (2 H, s, ArH), 6.86 (2 H, s, ArH), 6.81 (2 H, s, ArH), 4.23 (2
H, q, J 7, CH2CH3), 3.88 (8 H, br s, ArCH2Ar), 2.01 (6 H, s,
CH3) and 1.27 (3 H, t, J 7, CH2CH3).
Fig. 4 View of the dimeric arrangements shown for one of the two
independent calixarene molecules 3e
The disordered trichloromethane molecules lay in a volume
element centred on (0.5, 0.5, 0.5) in the unit cell effectively
between calixarene dimers centred at (0.5, 0, 0) and (0, 0.5, 0.5)
molecules. As a result of the dimer packing adopted by the
calixarenes effectively filling the calix[4]arene cones, it was not
possible for any solvent molecule to take part in C–H ؒ ؒ ؒ π
interactions inside the cones as we have reported previously.
Experimental
Syntheses
Hydroxymethylated and bromomethylated phenols,12,13 the
dimers of p-tert-butylphenol18 and p-nitrophenol,19 its bis-
(bromomethyl)ated derivative and the trimer 1a 20 were pepared
as described in the literature or in analogy to literature
prescriptions.
General procedure for the synthesis of linear trimers
Method A. 2,6-Bis(bromomethyl)-4-nitrophenol (10 g, 31
mmol) was added to an excess of molten p-alkylphenol (0.3
mol). After 5 h at 120 ЊC, the evolution of HBr finished, the
mixture was cooled to room temperature and the excess phenol
was removed by steam distillation. The residue was purified as
described for the individual compounds.
17-(1-Methoxycarbonylmethyl)-11,23-dimethyl-5-nitro-
25,26,27,28-tetrahydroxycalix[4]arene (2c). The crude product
was reesterified by refluxing with methanol for 3 h prior to the
Soxhlet extraction. Purification by flash chromatography with
chloroform–ethyl acetate (10:1) and recrystallisation from
Method B. A solution of p-alkylphenol (0.3 mol) in 80 ml
J. Chem. Soc., Perkin Trans. 2, 1997
1197

View Article Online
chloroform–methanol gave 210 mg (8%) of colourless crystals,
mp 332 ЊC (Found: C, 69.55; H, 5.6; N, 2.35. C33H31NO8
requires C, 69.6; H, 5.5; N, 2.5%); δH (200 MHz; [2H6]DMSO)
8.07 (2 H, s, ArH), 6.97 (2 H, s, ArH), 6.88 (2 H, s, ArH), 6.80 (2
H, s, ArH), 3.88 (4 H, br s, ArCH2Ar), 3.77 (4 H, br s,
ArCH2Ar), 3.56 (3 H, s, COOCH3), 3.41 (2 H, s, ArCH2) and
2.02 (6 H, s, CH3).
11,23-Di-tert-butyl-17-methyl-5-nitro-25,26,27,28-tetra-
hydroxycalix[4]arene (3a). Purification by flash chromato-
graphy with chloroform and recrystallisation from chloroform–
methanol gave 450 mg (15%) of colourless crystals, mp 352 ЊC
(Found: C, 74.5; H, 7.1; N, 2.15. C37H41NO6 requires C, 74.6; H,
6.95; N, 2.35%); δH (200 MHz; CDCl3) 10.20 (4 H, s, ArOH),
7.96 (2 H, s, ArH), 7.10 (4 H, s, ArH), 6.81 (2 H, s, ArH), 4.21 (4
H, br m, ArCH2Ar), 3.54 (4 H, br m, ArCH2Ar), 2.09 (3 H, s,
CH3) and 1.24 [18 H, s, C(CH3)3].
11,17,23-Tri-tert-butyl-5-nitro-25,26,27,28-tetrahydroxy-
calix[4]arene (3b). Purification by flash chromatography with
chloroform and recrystallisation from chloroform–methanol
gave 450 mg (14%) of colourless crystals, mp 300 ЊC; δH (200
MHz; CDCl3) 10.30 (4 H, s, ArOH), 7.99 (2 H, s, ArH), 7.10 (2
H, d, J 2.2, ArH), 7.06 (2 H, d, J 2.2, ArH), 7.02 (2 H, s, ArH),
4.20 (4 H, br s, ArCH2Ar), 3.60 (4 H, br s, ArCH2Ar), 1.22 [18
H, s, C(CH3)3] and 1.18 [9 H, s, C(CH3)3].
11,23-Di-tert-butyl-17-dodecyl-5-nitro-25,26,27,28-tetra-
hydroxycalix[4]arene (3c). Purification by flash chromatography
with chloroform and recrystallisation from chloroform–
methanol gave 450 mg (12%) of colourless crystals, mp 209 ЊC;
δH (200 MHz; CDCl3) 10.23 (4 H, s, ArOH), 7.95 (2 H, s, ArH),
7.09 (4 H, s, ArH), 6.80 (2 H, s, ArH), 4.25 (4 H, br m,
ArCH2Ar), 3.50 (4 H, br m, ArCH2Ar), 2.31 (2 H, t, J 6,
ArCH2), 1.46 (2 H, m, ArCH2CH2), 1.22 [18 H, s, C(CH3)3 and
CH2] and 0.85 (3 H, t, J 6, CH3).
[2H6]DMSO) 8.00 (2 H, s, ArH), 7.64 (2 H, s, ArH), 7.16
(4 H, s, ArH), 4.14 (2 H, q, J 7, CH2CH3), 3.88 (8 H, br s,
ArCH2Ar), 1.25 (3 H, t, J 7, CH2CH3) and 1.16 [18 H, s,
C(CH3)3].
11,23-Di-tert-butyl-5,17-dinitro-25,26,27,28-tetrahydroxy-
calix[4]arene (3i). Purification by flash chromatography with
chloroform–ethyl acetate (3:1) and recrystallisation from
chloroform–methanol gave 590 mg (19%) of light-yellow crys-
tals, mp 305 ЊC (decomp.); δH (200 MHz; [2H6]DMSO) 8.01 (4
H, s, ArH), 7.18 (4 H, s, ArH), 3.90 (8 H, br s, ArCH2Ar) and
1.16 [18 H, s, C(CH3)3].
11,23-Di-tert-butyl-16,17-dimethyl-5-nitro-25,26,27,28-tetra-
hydroxycalix[4]arene (5a). Purification by flash chromato-
graphy with carbon tetrachloride–chloroform (1.5:1) and
recrystallisation from chloroform–methanol gave 200 mg (7%)
of colourless crystals, mp 290–300 ЊC (decomp.) (Found: C,
70.75; H, 6.85; N, 2.0. C38H43NO6 ϩ 0.33 CHCl3 requires C,
70.9; H, 6.7; N, 2.15%); δH (200 MHz; CDCl3) 10.35 (4 H, s,
ArOH), 7.96 (1 H, d, J 2.8, ArH), 7.94 (1 H, d, J 2.8, ArH),
7.27 (1 H, d, J 2.3, ArH), 7.11 (1 H, d, J 2.3, ArH), 7.09 (1 H, d,
J 2.3, ArH), 7.08 (1 H, d, J 2.3, ArH), 6.83 (1 H, s, ArH), 4.00 (8
H, br m, ArCH2Ar), 2.33 (3 H, s, CH3), 2.05 (3 H, s, CH3) and
1.25 [18 H, s, C(CH3)3].
11,23-Di-tert-butyl-17-chloro-16-methyl-5-nitro-25,26,27,28-
tetrahydroxycalix[4]arene (5b). Purification by flash chroma-
tography with carbon tetrachloride–chloroform (1.5:1) and
recrystallisation from chloroform–methanol gave 380 mg (12%)
of colourless crystals, mp 290–300 ЊC (decomp.) (Found: C,
65.35; H, 5.9; N, 1.9. C37H40ClNO6 ϩ 0.5 CHCl3 requires C,
65.3; H, 5.9; N, 2.05%); δH (200 MHz; CDCl3) 10.27 (4 H, s,
ArOH), 7.98 (1 H, d, J 2.7, ArH), 7.95 (1 H, d, J 2.7, ArH),
7.14–7.08 (4 H, m, ArH), 7.06 (1 H, s, ArH), 3.90 (8 H, br m,
ArCH2Ar) and 1.24 [18 H, s, C(CH3)3].
17-Cyclohexyl-11,23-di-tert-butyl-5-nitro-25,26,27,28-tetra-
hydroxycalix[4]arene (3d). Purification by flash chromato-
graphy with chloroform and recrystallisation from chloroform–
methanol gave 720 mg (22%) of colourless crystals, mp 324 ЊC;
δH (200 MHz; CDCl3) 10.27 (4 H, s, ArOH), 7.97 (2 H, s, ArH),
7.10 (2 H, d, J 2.3, ArH), 7.08 (2 H, d, J 2.3, ArH), 6.85 (2 H, s,
ArH), 4.24 (4 H, br m, ArCH2Ar), 3.53 (4 H, br m, ArCH2Ar),
2.24 (1 H, m, ArCH), 1.73 (4 H, m, CH2) and 1.23 [24 H, s, CH2
and C(CH3)3].
11,23-Di-tert-butyl-5-nitro-17-phenyl-25,26,27,28-tetra-
hydroxycalix[4]arene (3e). Purification by flash chromatography
with chloroform and recrystallisation from chloroform–
methanol gave 890 mg (27%) of colourless crystals, mp 278 ЊC
(Found: C, 76.55; H, 6.5; N, 1.95. C42H43NO6 requires C,
76.7; H, 6.6; N, 2.15%); δH (200 MHz; CDCl3) 10.27 (4 H, s,
ArOH), 7.97 (2 H, s, ArH), 7.35–7.22 (5 H, m, ArH), 7.16 (2 H,
d, J 2.1, ArH), 7.12 (2 H, d, J 2.1, ArH), 4.29 (4 H, br m,
ArCH2Ar), 3.64 (4 H, br m, ArCH2Ar) and 1.23 [18 H, s,
C(CH3)3].
11,23-Di-tert-butyl-17-chloro-5-nitro-25,26,27,28-tetra-
hydroxycalix[4]arene (3f). Purification by flash chromatography
with carbon tetrachloride–chloroform (1.5:1) and recrystallis-
ation from chloroform–methanol gave 750 mg (25%) of colour-
less crystals, mp 290–300 ЊC (decomp.) (Found: C, 60.45; H,
5.4; N, 1.8. C36H38ClNO6 ϩ 1 CHCl3 requires C, 60.55; H,
5.35; N, 1.9%); δH (200 MHz; CDCl3) 10.14 (4 H, s, ArOH), 7.97
(2 H, s, ArH), 7.13 (2 H, d, J 2.3, ArH), 7.09 (2 H, d, J 2.3,
ArH), 6.98 (2 H, s, ArH), 3.90 (8 H, br s, ArCH2Ar) and 1.24
[18 H, s, C(CH3)3].
11,23-Di-tert-butyl-17-ethoxycarbonyl-5-nitro-25,26,27,28-
tetrahydroxycalix[4]arene (3g). The crude product was reesteri-
fied by refluxing with ethanol for 3 h prior to the Soxhlet extrac-
tion. Purification by flash chromatography with chloroform–
ethyl acetate (3:1) and recrystallisation from chloroform–
methanol gave 260 mg (8%) of pale yellow crystals, mp 336 ЊC
(decomp.) (Found: C, 71.5; H, 6.45; N, 2.05. C39H43NO8
requires C, 71.65; H, 6.65; N, 2.15%); δH (200 MHz;
11,23-Di-tert-butyl-17-chloro-16,18-dimethyl-5-nitro-
25,26,27,28-tetrahydroxycalix[4]arene (5c). Purification by flash
chromatography with carbon tetrachloride–chloroform (1.5:1)
and recrystallisation from chloroform–methanol gave 250 mg
(8%) of colourless crystals, mp 290–300 ЊC (decomp.) (Found:
C, 68.1; H, 6.2; N, 2.0. C38H42ClNO6 ϩ 0.25 CHCl3 requires C,
68.15; H, 6.3; N, 2.1%); δH (200 MHz; CDCl3) 10.41 (4 H, s,
ArOH), 7.95 (2 H, s, ArH), 7.27 (2 H, d, J 2.3, ArH), 7.12 (2 H,
d, J 2.3, ArH), 4.00 (8 H, br m, ArCH2Ar), 2.53 (6 H, s, CH3)
and 1.24 [18 H, s, C(CH3)3].
12,17,22-Trimethyl-5-nitro-11,23-diisopropyl-5,26,27,28-
tetrahydroxycalix[4]arene (6a). Purification by flash chroma-
tography with carbon tetrachloride–chloroform (1.5:1) and
recrystallisation from chloroform–methanol gave 600 mg (21%)
of colourless crystals, mp 290–300 ЊC (decomp.) (Found: C,
71.6; H, 6.6; N, 2.15. C37H41NO6 ϩ 0.25 CHCl3 requires C,
71.5; H, 6.65; N, 2.25%); δH (200 MHz; CDCl3) 10.49 (4 H, s,
ArOH), 7.98 (2 H, s, ArH), 6.98 (2 H, s, ArH), 6.97 (2 H, s,
ArH), 4.07 (4 H, br s, ArCH2Ar), 3.97 (4 H, br s, ArCH2Ar),
3.03 [2 H, sept., J 6.8, CH(CH3)2], 2.42 (6 H, s, CH3), 2.11 (3 H,
s, CH3) and 1.13 [12 H, d, J 6.8, CH(CH3)2].
12,16,17,22-Tetramethyl-5-nitro-11,23-diisopropyl-
25,26,27,28-tetrahydroxycalix[4]arene (6b). Purification by flash
chromatography with carbon tetrachloride–chloroform (1.5:1)
and recrystallisation from chloroform–methanol gave 100 mg
(3%) of pale yellow crystals, mp 290–300 ЊC (decomp.) (Found:
C, 64.4; H, 6.0; N, 1.8. C38H43NO6 ϩ 1 CHCl3 requires C, 64.35;
H, 6.1; N, 1.9%); δH (200 MHz; CDCl3) 10.03 (4 H, s, ArOH),
8.01 (2 H, s, ArH), 6.98 (1 H, s, ArH), 6.96 (2 H, s, ArH), 3.92 (8
H, m, ArCH2Ar), 2.98 [2 H, sept., J 6.3, CH(CH3)2], 2.33 (3 H, s,
CH3), 2.31 (3 H, s, CH3), 2.21 (3 H, s, CH3), 2.15 (3 H, s, CH3)
and 1.13 [12 H, d, J 6.3, CH(CH3)2].
17-Chloro-12,22-dimethyl-5-nitro-11,23-diisopropyl-
25,26,27,28-tetrahydroxycalix[4]arene (6c). Purification by flash
chromatography with carbon tetrachloride–chloroform
(1.25:1) and recrystallisation from chloroform–methanol gave
1.15 g of a complex mixture (probably of calixarenes different
1198
J. Chem. Soc., Perkin Trans. 2, 1997

View Article Online
ringsize) from which the desired calix[4]arene could be separ-
ated by preparative TLC with chloroform. A final recrystallisa-
tion from chloroform–methanol gave colourless crystals, mp
290–300 ЊC (decomp.) (Found: C, 60.1; H, 5.1; N, 1.8.
C36H38ClNO6 ϩ 0.66 CHCl3 requires C, 60.0; H, 5.05; N, 1.8%);
δH (200 MHz; CDCl3) 10.40 (4 H, s, ArOH), 7.99 (2 H, s, ArH),
7.14 (2 H, s, ArH), 7.02 (2 H, s, ArH), 3.96 (8 H, br s,
ArCH2Ar), 3.01 [2 H, sept., J 6.8, CH(CH3)2], 2.41 (6 H, s, CH3)
and 1.15 [12 H, d, J 6.8, CH(CH3)2].
17-Chloro-12,16,22-trimethyl-5-nitro-11,23-diisopropyl-
25,26,27,28-tetrahydroxycalix[4]arene (6d). Purification by flash
chromatography with carbon tetrachloride–chloroform (1.5:1)
and recrystallisation from chloroform–methanol gave 100 mg
(3%) of colourless crystals, mp 290–300 ЊC (decomp.) (Found:
C, 61.5; H, 5.6; N, 1.85. C37H40ClNO6 ϩ 1 CHCl3 requires
C, 61.0; H, 5.55; N, 1.9%); δH (200 MHz; CDCl3) 9.45 (4 H,
s, ArOH), 8.03 (2 H, s, ArH), 7.22 (1 H, s, ArH), 7.20 (1 H,
s, ArH), 7.09 (1 H, s, ArH), 4.00 (8 H, br s, ArCH2Ar), 2.99
[2 H, sept., J 6.8, CH(CH3)2], 2.47 (3 H, s, CH3), 2.32
(3 H, s, CH3), 2.25 (3 H, s, CH3) and 1.13 [12 H, d, d 6.8,
CH(CH3)2].
amine (freshly distilled) in 450 ml 2-methoxyethanol and 50 ml
water. In a typical experiment 0.8 mg of the nitrocalix[4]arene
was dissolved in 25 ml of this solution (5 × 10Ϫ5 mol lϪ1) using
an ultrasonic bath. 20 ml of this solution were placed in a
thermostatted vessel of an automatic titration apparatus
(Metrohm E510) at 22 ЊC and acidified with conc. HCl.
Titration with NaOH (1 mol lϪ1) was carried out using a glass
electrode, which was calibrated against aqueous buffer stand-
ards. At the desired apparent pH values (2 min elapsed before
the final reading to obtain a constant value) the titration was
stopped, a sample was transferred by a pipette to a 1 cm cuvette
and its UV–VIS spectrum was measured with a Perkin-Elmer
Lambda 17 apparatus connected to a computer.
The sample was afterwards transferred back into the titration
vessel and the titration was continued. The accuracy of this
procedure is demonstrated in Fig. 1, where the volume increase
(usually a total of 100–150 µl until the final pH) is taken into
account for the single spectra.
In this way, usually 15–20 spectra were recorded, including a
spectrum at low pH, before the titration starts and at high pH
(at least pH > pKa ϩ 3). The pKa was obtained from plots
according to eqn. (3). For different wavelengths (usually
between 390–425 nm) deviations less than 0.02 were observed
for a single titration experiment. Usually three titrations were
carried out, including also different or independently calibrated
electrodes. The deviations for these repeated experiments were in
the range of ±0.05.
For the dinitrocalix[4]arenes 3i and 4 the absorbance for the
undissociated form ε(H2A) and for the dianion ε(A2Ϫ) cannot be
obtained exactly at low and at high pH. Therefore plots of eqns.
(4) and (5) were used to determine Ka1 and Ka2. The absorption
17-Chloro-12,16,18,22-tetramethyl-5-nitro-11,23-diiso-
propyl-25,26,27,28-tetrahydroxycalix[4]arene (6e). Purification
by flash chromatography with carbon tetrachloride–chloroform
(1.5:1) and recrystallisation from chloroform–methanol gave
20 mg (0.6%) of colourless crystals, mp 290–300 ЊC (decomp.);
δH (200 MHz; CDCl3) 10.03 (4 H, s, ArOH), 8.08 (2 H, s, ArH),
7.07 (2 H, s, ArH), 3.97 (8 H, br m, ArCH2Ar), 3.05 [2 H, sept.,
J 6.8, CH(CH3)2], 2.47 (6 H, s, CH3), 2.38 (6 H, s, CH3) and 1.15
[12 H, d, J 6.8, CH(CH3)2].
17,23-Di-tert-butyl-5,11-dinitro-25,26,27,28-tetrahydroxy-
calix[4]arene (4). The dimer of p-tert-butylphenol and the bis-
(bromomethyl)ated dimer of p-nitrophenol were reacted as
described above for the 3 ϩ 1-fragment condensations. Purifica-
tion by flash chromatography with dichloromethane and
recrystallisation from hexane gave 1.0 g (32%) of light yellow
crystals, mp 340 ЊC (decomp.) (Found: C, 68.8; H, 6.0; N, 4.4.
C36H38N2O8 requires C, 69.0; H, 6.1; N, 4.5%); δH (200 MHz;
[2H6]DMSO) 8.85 (4 H, br s, ArOH), 8.11 (2 H, d, J 2.5, ArH),
8.05 (2 H, d, J 2.5, ArH), 7.10 (2 H, s, ArH), 7.08 (2 H, s, ArH),
3.90 (8 H, br s, ArCH2Ar) and 1.12 [18 H, s, C(CH3)3].
ε Ϫ ε(HAϪ) ε(H2A)
ε
(4)
(5)
=
Ϫ
[Hϩ]
Ka1
Ka1
[Hϩ][ε Ϫ ε(HAϪ)] = Ka2 ε(A2Ϫ) Ϫ Ka2 ε
coefficient ε(HAϪ) was easily measured at medium pH
(pKa1 ϩ 3 < pH < pKa2 Ϫ 3) and [Hϩ] was calculated from the
apparent pH. Deviations up to 0.1 were observed in this case
for repeated experiments.
Hydrolysis of esters
The calix[4]arene (50 mg) was suspended in a solution of 3 ml
methanol, 1.5 ml water and 50 mg sodium hydroxide and
allowed to stand overnight. The resulting clear solution was
neutralised with acetic acid and dropped into 10 ml water to
form a white precipitate which was filtered and dried.
17-Carboxy-11,23-dimethyl-5-nitro-25,26,27,28-tetra-
hydroxycalix[4]arene (2b). 2b gave 41 mg (87%) of white
powder, mp 355 ЊC; δH (200 MHz; [2H6]DMSO) 8.06 (2 H, s,
ArH), 7.68 (2 H, s, ArH), 6.85 (2 H, s, ArH), 6.81 (2 H, s, ArH),
3.86 (8 H, br s, ArCH2Ar) and 2.01 (6 H, s, CH3).
17-Carboxymethyl-11,23-dimethyl-5-nitro-25,26,27,28-tetra-
hydroxycalix[4]arene (2d). 2d gave 42 mg (89%) of white
powder, mp 346 ЊC; δH (200 MHz; [2H6]DMSO) 8.05 (2 H, s,
ArH), 6.95 (2 H, s, ArH), 6.87 (2 H, s, ArH), 6.81 (2 H, s, ArH),
3.86 (8 H, br s, ArCH2Ar), 3.29 (2 H, s, ArCH2) and 2.02 (6 H,
s, CH3).
X-Ray structure analysis
The data crystal only diffracted weakly [only 6695 (38%) of the
17 726 measured reflections could be labelled as observed]. The
asymmetric unit contains two independent calixarene molecules
and it became obvious early on in the analysis that there were
also molecules of solvation (which could only be chloroform)
occupying what would otherwise have been voids in the
crystal lattice. Difference maps revealed locations of all hydroxy
H atoms (which were then allowed for with the SHELXL AFIX
147 command in the refinement). It was also clear that all tert-
butyl methyl groups were slightly disordered (equally over two
adjacent sites) and this was allowed for by suitable DFIX
restraints in the refinement cycles. It was clear that the chloro-
form molecules of solvation which lay between calixarene mol-
ecules in what would have been a void centred at (0.5, 0.5, 0.5),
were seriously disordered over two distinct sites in the asym-
metric unit and, from the electron-density peak heights, had
only partial occupancy. Attempts at modelling various dis-
ordered chloroform moieties in the two locations were not suc-
cessful and, before the final refinement cycles, the contributions
of the disordered solvent molecules to the observed structure
amplitudes were subtracted using the SQUEEZE option in
PLATON.21 This showed that the combined occupancy of the
two chloroform sites in the asymmetric unit was 0.57. In the
final refinement cycles, all the calixarene non-H atoms were
allowed anisotropic motion.
11,23-Di-tert-butyl-17-carboxy-5-nitro-25,26,27,28-tetra-
hydroxycalix[4]arene (3h). 3h gave 41 mg (85%) of white
powder, mp 300–305 ЊC (decomp.); δH (200 MHz; CDCl3) 10.10
(4 H, s, ArOH), 7.95 (2 H, s, ArH), 7.76 (2 H, s, ArH), 7.15 (2
H, d, J 2.5, ArH), 7.12 (2 H, d, J 2.5, ArH), 4.20 (8 H, br s,
ArCH2Ar) and 1.22 [18 H, s, C(CH3)3].
Determination of pKa values
A buffer solution was made by dissolving 97 mg citric acid
(Aldrich), 53 mg malonic acid (Aldrich) and 69 µl triethyl-
J. Chem. Soc., Perkin Trans. 2, 1997
1199

View Article Online
10 S. Rantsordas, M. Perrin, F. Gharnati, S. Lecocq, W. Vogt, T. Fey
Acknowledgements
and V. Böhmer, J. Incl. Phenom., Mol. Recogn. Chem., 1990, 9,
145.
11 V. Böhmer, L. Merkel and U. Kunz, J. Chem. Soc., Chem. Commun.,
1987, 896.
12 V. Böhmer, F. Marschollek and L. Zetta, J. Org. Chem., 1987, 52,
3200.
These studies were supported in part by the Deutsche For-
schungsgemeinschaft and the European Community. We are
grateful to Frank Marschollek for the synthesis of some
calix[4]arenes.
13 L. Zetta, A. Wolff, W. Vogt, K.-L. Platt and V. Böhmer, Tetrahedron,
1991, 47, 1911.
14 J. de Mendoza, P. M. Nieto, P. Prados and C. Sanchez, Tetrahedron,
1990, 46, 671.
15 G. D. Andreetti, V. Böhmer, J. G. Jordan, M. Tabatabai, F. Ugoz-
zoli, W. Vogt and A. Wolff, J. Org. Chem., 1993, 58, 4023.
16 E. Dahan and S. E. Biali, J. Org. Chem., 1991, 56, 7269.
17 Y. Ueda, T. Fujiwara, K.-I. Tomita, Z. Asfari and J. Vicens, J. Incl.
Phenom., Mol. Recogn., 1993, 15, 341.
References
1 For a recent review see: V. Böhmer, Angew. Chem., 1995, 107, 785;
Angew. Chem., Int. Ed. Engl., 1995, 34, 713.
2 S. Shinkai, K. Araki, H. Koreishi, T. Tsubaki and O. Manabe,
Chem. Lett., 1986, 1351.
3 S. Shinkai, K. Araki, P. D. J. Grootenhuis and D. N. Reinhoudt,
J. Chem. Soc., Perkin Trans. 2, 1991, 1883.
4 P. D. J. Grootenhuis, P. A. Kollman, L. C. Groenen, D. N. Rein-
houdt, G. J. van Hummel, F. Ugozzoli and G. D. Andreetti, J. Am.
Chem. Soc., 1990, 112, 4165.
5 G. Arena, R. Cali, G. G. Lombardo, E. Rizzarelli, D. Sciotto,
R. Ungaro and A. Casnati, Supramol. Chem., 1992, 1, 19.
6 I. Yoshida, N. Yamamoto, F. Sagara, D. Ishii, K. Ueno and
S. Shinkai, Bull. Chem. Soc. Jpn., 1992, 65, 1012.
7 V. Böhmer, E. Schade, C. Antes, J. Pachta, W. Vogt and
H. Kämmerer, Makromol. Chem., 1983, 184, 2361.
18 A. Zinke, R. Kretz, E. Leggewie and K. Hössinger, Monatsh.
Chem., 1952, 83, 1213.
19 F. Arnaud-Neu, V. Böhmer, J.-F. Dozol, C. Grüttner, R. A. Jacobi,
D. Kraft, O. Mauprivez, H. Rouquette, M.-J. Schwing-Weill,
N. Simon and W. Vogt, J. Chem. Soc., Perkin Trans. 2, 1996, 1175.
20 V. Böhmer, W. Lotz, J. Pachta and S. Tütüncü, Makromol. Chem.,
1981, 182, 2671.
21 A. L. Spek, PLATON Molecular Geometry Program, 1994 Version,
University of Utrecht, Utrecht, Holland.
8 V. Böhmer, E. Schade and W. Vogt, Makromol. Chem., Rapid Com-
mun., 1984, 5, 221.
9 T. Fey, Ph.D. Thesis, University of Mainz, 1988; see also V. Böhmer
and J. Vicens, in Calixarenes, a Versatile Class of Macrocyclic
Compounds, eds. J. Vicens and V. Böhmer, Kluwer, Dordrecht, 1991.
Paper 6/07259C
Received 24th November 1996
Accepted 27th February 1997
1200
J. Chem. Soc., Perkin Trans. 2, 1997