Binding of tetramethylammonium to polyether side-chained aromatic hosts. Evaluation of the binding contribution from ether oxygen donors

Article abstract of DOI:10.1021/jo034905h

A set of macrocyclic and open-chain aromatic ligands endowed with polyether side chains has been prepared to assess the contribution of ether oxygen donors to the binding of tetramethylammonium (TMA), a cation believed incapable of interacting with oxygen donors. The open-chain hosts consisted of an aromatic binding site and side chains possessing a variable number of ether oxygen donors; the macrocyclic ligands were based on the structure of a previously investigated host, the dimeric cyclophane 1,4-xylylene-1,4-phenylene diacetate (DXPDA), implemented with polyether-type side chains in the backbone. Association to tetramethylammonium picrate (TMAP) was measured in CDCl ₃ at T = 296 K by ¹H NMR titrations. Results confirm that the main contribution to the binding of TMA comes from the cation - π interaction established with the aromatic binding sites, but they unequivocally show that polyether chains participate with cooperative contributions, although of markedly smaller entity. Water is also bound, but the two guests interact with aromatic rings and oxygen donors in an essentially noncompetitive way. An improved procedure for the preparation of cyclophanic tetraester derivatives has been developed that conveniently recycles the oligomeric ester byproducts formed in the one-pot cyclization reaction. An alternative entry to benzylic diketones has also been provided that makes use of a low-order cyanocuprate reagent to prepare in fair yields a class of compounds otherwise uneasily accessible.

Full text of DOI:10.1021/jo034905h

Bin d in g of Tetr a m eth yla m m on iu m to P olyeth er Sid e-Ch a in ed
Ar om a tic Hosts. Eva lu a tion of th e Bin d in g Con tr ibu tion fr om
Eth er Oxygen Don or s
Sandra Bartoli, Gina De Nicola, and Stefano Roelens*
CNR, Istituto di Chimica dei Composti Organometallici, Dipartimento di Chimica Organica,
Universita` di Firenze, Polo Scientifico, I-50019 Sesto Fiorentino, Firenze, Italy
stefano.roelens@unifi.it
Received J une 26, 2003
A set of macrocyclic and open-chain aromatic ligands endowed with polyether side chains has been
prepared to assess the contribution of ether oxygen donors to the binding of tetramethylammonium
(TMA), a cation believed incapable of interacting with oxygen donors. The open-chain hosts consisted
of an aromatic binding site and side chains possessing a variable number of ether oxygen donors;
the macrocyclic ligands were based on the structure of a previously investigated host, the dimeric
cyclophane 1,4-xylylene-1,4-phenylene diacetate (DXPDA), implemented with polyether-type side
chains in the backbone. Association to tetramethylammonium picrate (TMAP) was measured in
CDCl3 at T ) 296 K by 1H NMR titrations. Results confirm that the main contribution to the binding
of TMA comes from the cation-π interaction established with the aromatic binding sites, but they
unequivocally show that polyether chains participate with cooperative contributions, although of
markedly smaller entity. Water is also bound, but the two guests interact with aromatic rings and
oxygen donors in an essentially noncompetitive way. An improved procedure for the preparation
of cyclophanic tetraester derivatives has been developed that conveniently recycles the oligomeric
ester byproducts formed in the one-pot cyclization reaction. An alternative entry to benzylic
diketones has also been provided that makes use of a low-order cyanocuprate reagent to prepare
in fair yields a class of compounds otherwise uneasily accessible.
In tr od u ction
shown that acetylcholine (ACh) and tetramethyl-
ammonium (TMA) exhibit binding affinity for a family
of flexible, adaptive cyclophanic esters, among which
dimeric 1,4-xylylene-1,4-phenylene diacetate (DXPDA)
showed the best binding properties.5
Due to the coordinative ability of ether oxygen atoms,
polyether chains are the fundamental constituents of
several classes of cyclic and open-chain receptors that are
well-established ligands for metal and ammonium cat-
ions.1 In contrast, polyether chains do not seem to con-
tribute to the binding of quaternary ammonium cations,2
toward which aromatics exert a well-documented attrac-
tion,3 due to the interaction of the aromatic π system with
the cationic species. This lack of affinity is surprising in
consideration of the fact that, in the gas phase, both
experiments and ab initio computations4 indicate that
Me4N+ binds with approximately equal binding enthal-
pies to H2O (9.0 kcal mol-1) and to benzene (9.4 kcal
mol-1) and toluene (9.5 kcal mol-1) and that even stronger
interactions are observed with polyethers (diglyme, 20.6
kcal mol-1; triglyme, 24.2 kcal mol-1).4a In the course of
an investigation on the cation-π interaction, we have
Remarkably, sizable affinity was also detected with
their open-chain counterparts such as, for example, 1,4-
xylylene diphenylacetate (XDPA), caused by the coopera-
tive contribution of the ester groups enhancing the
aromatic binding sites of the host. On the basis of these
results, we thought that by introducing polyether chains
of variable length into the host structure, DXPDA and
related open-chain aromatic esters could be the appropri-
* Corresponding author. Ph:+39-055-457-3546. Fax:+39-055-457-
3570.
(1) Comprehensive Supramolecular Chemistry; Atwood, J . L., Davies,
J . E. D., MacNicol, D. D., Vo¨gtle, F., Eds.; Pergamon: Oxford, 1996;
Vol. 1.
(2) Ru¨diger, V.; Schneider, H.-J .; Solov’ev, V. P.; Kazachenko, V. P.;
Raevsky, O. A. Eur. J . Org. Chem. 1999, 1847-1856.
(3) Ma, J . C.; Dougherty, D. A. Chem. Rev. 1997, 97, 1303-1324.
(4) (a) Deakyne, C. A.; Meot-Ner, M. J . Am. Chem. Soc. 1985, 107,
469-474. (b) Deakyne, C. A.; Meot-Ner, M. J . Am. Chem. Soc. 1985,
107, 474-479. (c) Deakyne, C. A.; Meot-Ner, M. J . Am. Chem. Soc.
1999, 121, 1546-1557.
(5) Roelens, S.; Torriti, R. J . Am. Chem. Soc. 1998, 120, 12443-
12452.
10.1021/jo034905h CCC: $25.00 © 2003 American Chemical Society
Published on Web 09/18/2003
J . Org. Chem. 2003, 68, 8149-8156
8149

Bartoli et al.
CHART 1
SCHEME 1
ate hosts for a quantitative evaluation of the binding
capacities of polyether moieties toward quaternary am-
monium cations in solution; in particular, acyclic and
flexible, sidearmed cyclophane hosts, lacking preorgan-
ized binding sites, would shed light on the inherent
affinity of ether oxygen donors for tetralkylammonium
guests. We report here the results of a systematic study
in chloroform on the binding of TMA to a set of open-
chain hosts containing an aromatic binding site and a
variable number of ether oxygen donors in the side
chains, appropriately positioned to cooperatively interact
with the guest, and of DXPDA-related cyclophanes pos-
sessing polyether-type side chains. Results unequivocally
show that polyether chains participate in the binding of
TMA with cooperative contributions, although the entity
of the contributions is markedly smaller than that of the
cation-π interaction with the aromatic binding sites.
Resu lts a n d Discu ssion
Syn th esis of Op en -Ch a in Liga n d s. The open-chain
hosts were based on the structure of diethyl 1,4-phenyl-
enediacetate (1),6 which displayed conveniently detect-
able binding properties toward ACh and TMA. The hosts
prepared for the present investigation are collected in
Chart 1, together with the commercially available dibenzo-
24-crown-8 (11) and dicyclohexano-24-crown-8 (12), which
were used as reference ligands of appropriate size for the
TMA cation. Host 1 and its homologous polyoxyethylene
congeners 2, 3, and 4 were prepared by esterification of
1,4-phenylenediacetic acid with ethanol and with mono-,
di-, and triethylene glycol monoethyl ethers, respectively.
The adipate 5, the aliphatic analogue of 4, was prepared
in the same way to ascertain the host’s binding ability
in the absence of the aromatic binding site. To investigate
the binding properties of hosts lacking cooperative con-
tributions from ester groups, ethers 67 and 7, analogues
of esters 1 and 4, were prepared by Williamson etheri-
fication of R,R′-dibromo-p-xylene with ethanol and tri-
ethylene glycol monoethyl ether, respectively. The bis-
ortho ester 8, the analogue of 1 possessing geminal ether
oxygen donors somewhat more organized for binding than
6 and 7, was obtained by BF3-catalyzed rearrangement
of the bis-1,4-phenylenediacetic ester (8a ) of the com-
mercially available 3-methyl-3-hydroxymethyl oxetane
(Scheme 1).8 The bis-benzyl ether 97 was obtained by
benzylation of 1,4-benzenedimethanol with benzyl bro-
mide and has been prepared to provide information on
binding contributions arising from additional aromatic
(7) Quelet, R. Bull. Soc. Chim. Fr. 1933, 53, 222-233.
(8) Corey, E. J .; Raju, N. Tetrahedron Lett. 1983, 24, 5571-5574.
(6) Titley, A. F. J . Chem. Soc. 1928, 2576.
8150 J . Org. Chem., Vol. 68, No. 21, 2003

Binding of Me4N+ to Aromatic Hosts
SCHEME 2
rings with respect to 6. The diketone 10 was prepared to
ascertain whether the cooperative contribution of the
ester groups in 1 would depend on the alcoholic oxygen.
For the preparation of 10, common methods for the
synthesis of ketones failed because of the presence of two
benzylic positions.9 In an effort to provide an alternative
entry to benzylic diketones,13 we developed a convenient
procedure relying on the treatment of 1,4-phenylenedi-
acetyl chloride with a lower order cyanocuprate reagent,
prepared by mixing MeLi and CuCN in a 1:1 ratio,14
which afforded 10 in fair yields (=50% unoptimized).
Syn th esis of Sid ea r m ed Cyclop h a n e Liga n d s. The
conveniently large upfield shifts induced on the guests
signals upon binding made the macrocyclic cyclophane
host DXPDA appropriate to evaluate the contribution of
polyether chains in the cyclophane series. Implementa-
tion of the cyclophane structure with polyether chains
may be achieved through functionalization of the back-
bone, provided that structural modifications should not
drastically affect the stereoelectronic and conformational
features of the parent host, to compare the binding
properties of functionalized cyclophanes with those of
DXPDA. To compromise between the above requirements
and the ease of access, cyclophanes 13-17 were prepared
starting from commercially available diethyl 2,5-dihy-
droxyterephthalate 18 (Scheme 2), which provided a
selection of hosts bearing different oxygenated side
chains. The choice of what may appear as a drastic per-
turbation of the parent DXPDA structure, mainly because
of the conversion of two benzene rings into far more
electron-rich aromatic residues, was made on account of
computational work published by Dougherty and co-
workers,15 which predicted a negligible binding energy
difference between benzene and phenol in cation-π
complexes of alkaline cations. Terephthalate bis-ethers
19-23 were obtained from 18 by alkylation of phenolic
hydroxyls with the mesylate of the appropriate alcohol
(19-22) or with benzyl bromide (23), and subsequently
reduced to the corresponding 1,4-benzenedimethanol
derivatives 24-28. The target cyclophanes 13-17 were
obtained from the latter diols by azeotropic dehydration
in the presence of di-n-butyltin oxide, followed by addition
of 1,4-phenylenediacetyl chloride, through a well-estab-
lished organotin-mediated cyclooligomerization reac-
tion.16 In this context, we found that a dramatic improve-
ment in the conversion of diols into the cyclophanic
tetraesters could be achieved by recycling the oligomeric
(9) Treatment of 1,4-phenylenediacetic acid with methyllithium and
chlorotrimethylsilane (ref 10) or treatment of 1,4-phenylenediacetyl
chloride with methylmagnesium bromide (ref 11) gave unreacted
starting material exclusively after workup. On the other hand,
treatment of the Grignard reagent of R,R′-dibromo-p-xylene with acetyl
chloride gave mainly Wurtz coupling products. Yields of diketone not
larger than 5% could be obtained by reacting 1,4-phenylenediacetyl
chloride with Me2Cu(CN)Li2 (ref 12).
(10) Rubottom, G. M.; Kim, C.-w. J . Org. Chem. 1983, 48, 1550-
1552.
byproducts recovered from the crude reaction mixture
after the isolation of the tetraester. The oligomeric
mixture was reequilibrated by refluxing in toluene in the
presence of catalytic amounts (10% mol) of a 1:1 mixture
of di-n-butyltin oxide and di-n-butyltin chloride,17 to
(11) Sato, F.; Inoue, M.; Oguro, K.; Sato, M. Tetrahedron Lett. 1979,
20, 4303-4306.
(12) Lipshutz, B. H.; Wilhelm, R. S.; Kozlowski, J . A.; Parker, D. J .
Org. Chem. 1984, 49, 3928-3938.
(13) Diacetonylbenzene has previously been prepared by arylation
of enol acetate with 1,4-dibromobenzene via tributyltin enolates
catalyzed by dichlorobis(tri-o-tolylphosphine)palladium complex: Ko-
sugi, M.; Hagiwara, I.; Sumiya, T.; Migita, T. Bull. Chem. Soc. J pn.
1984, 57, 242-246.
(14) Acker, R.-D. Tetrahedron Lett. 1977, 18, 3407-3410.
(15) Mecozzi, S.; West, A. P., J r.; Dougherty, D. A. J . Am. Chem.
Soc. 1996, 118, 2307-2308.
(16) (a) Shanzer, A.; Libman, J .; Frolow, F. Acc. Chem. Res. 1983,
16, 60-67. (b) Roelens, S. J . Chem. Soc., Perkin Trans. 2 1988, 1617-
1625. (c) Mandolini, L.; Montaudo, G.; Scamporrino, E.; Roelens, S.;
Vitalini, D. Macromolecules 1989, 22, 3275-3280.
J . Org. Chem, Vol. 68, No. 21, 2003 8151

Bartoli et al.
SCHEME 3
TABLE 1. Associa tion Con sta n ts Ka (M-1), Sta n d a r d F r ee En er gies of Bin d in g -∆G° (k J m ol-1), a n d Gu est Lim itin g
Sh ift Va lu es ∆δ∞ (p p m ) for th e 1:1 Com p lexes of Tetr a m eth yla m m on iu m P icr a te (TMAP ) a n d Wa ter (H2O) w ith
Op en -Ch a in Ar om a tic Liga n d sa
TMAP
H2O
∆δ∞
0.148 (0.22 M)d
0.667
host
Ka (SE)b
∆δ∞
-∆G°
Ka (SE)b
ndc
2.24 (0.02)
3.83 (0.07)
6.43 (0.07)
6.43 (0.05)
ndc
-∆G°
1
2
3
4
5
7.94 (0.07)
8.08 (0.07)
9.3 (0.15)
12.1 (0.27)
ndc
-0.503
5.09(2)
5.14(2)
5.49(4)
6.14(5)
-0.546
1.98(3)
3.30(5)
4.58(3)
4.58(2)
-0.514
0.876
1.149
1.047
-0.464
-0.035 (0.10 M)d
-0.180 (0.30 M)d
-0.506
6
ndc
0.210 (0.30 M)d
1.207
7
8
9
10
11
12
5.5 (0.18)
2.61 (0.05)
3.03 (0.04)
7.3 (0.1)
14.5 (0.65)
ndc
4.20(7)
2.36(5)
2.73(3)
4.89(4)
6.5(1)
7.96 (0.05)
5.11(1)
-0.478
ndc
-0.906
ndc
0.052 (0.13 M)d
0.227 (0.28 M)d
1.112
-0.500
ndc
-0.434
5.18 (0.05)
14.5 (0.12)
4.05(2)
6.58(2)
1.503
a
Measured by 1H NMR (200/300 MHz) at T ) 296 K in CDCl3 on 0.1 mM solutions of TMAP using host concentrations up to 60
mg/mL. Standard error of the nonlinear least-squares fit. c Nondetectable; for Ka e 2, titration curves become indistinguishable from
each other in the investigated concentration range. ∆δ values observed at the indicated titrant concentration.
b
d
regenerate the entire oligomer distribution, which can
be optimized in the tetraester yield by adjusting the
concentration of the reacting mixture.16c,18 Conversion
yields of tetraester achieved by this reequilibration
procedure can be as high as 14% (19% for the unsubsti-
tuted cyclophane DXPDA) in a single step with respect
to the recovered oligomeric mixture. Furthermore, the
recycling procedure could be repeated several times to
progressively upgrade the conversion yield.
While the alcohol reagents used for the introduction
of side chains in 19 and 20 were commercially available
and that for 21 could be readily obtained by monobenzyl-
ation of ethylene glycol, the dibenzyl ether of glycerol
needed for 22 required a slightly more elaborated syn-
thesis that rested on the selective crystallization from
the raw benzylidene acetal mixture of the cis-acetal 30-
cis,19 which could be obtained diastereomerically pure
from diisopropyl ether on a multigram scale (Scheme 3).
Acetal 30-cis was benzylated with benzyl bromide and
reductively cleaved to the desired dibenzyl ether 32.20
Finally, the diphenylacetate 33, the open-chain “half-
molecule” of 17, was prepared from 28 and phenylacetyl
chloride to serve, together with 17, as reference host to
evaluate intrinsic differences in binding ability between
the phenol-type sidearmed ligands and the unsubstituted
parent cyclophane DXPDA and open-chain XDPA and to
check the validity of the assumption on which their
synthesis was based. The choice of the benzyl derivatives
as reference hosts was dictated by the low solubility of
the free hydroxy and of the methyl ether derivatives of
17, as well as that of the corresponding derivatives of
33, none of which was soluble enough in chloroform to
allow for binding measurements.
Bin d in g Stu d ies. The association constants (Ka) for
the 1:1 complexes of tetramethylammonium picrate
(TMAP) with the above set of hosts were measured, as
1
in previous studies,5 in CDCl3 at T ) 296 K by H NMR
titrations, monitoring the time-averaged signal of the free
and complexed TMA with increasing host concentration.
TMAP was selected as the appropriate salt for binding
measurements, because the picrate anion has been shown
to be a convenient partner for TMA, to minimize the
inhibiting contribution of the counterion.21 The details
of the titration procedure and of the analytical treatment
of data have been described elsewhere.22 Results for the
open-chain hosts, together with those for the reference
ligands 11 and 12 and for the sidearmed cyclophane
hosts, along with previously determined data for DXPDA
(17) Roelens, S. J . Chem. Soc., Chem. Commun. 1990, 58-60.
(18) Ercolani, G.; Mandolini, L.; Mencarelli, P.; Roelens, S. J . Am.
Chem. Soc. 1993, 115, 3901-3908.
(19) Carlsen, P. H. J .; Sørbye, K.; Ulven, T.; Aasbø, K. Acta Chem.
Scand. 1996, 50, 185-187.
(20) Unfortunately 29, which is formed in roughly similar amount
in the acetalization, could not be employed for the purpose. Although
methods for the selective cleavage of benzylidene acetals at either the
primary or the secondary oxygen are available (see: Greene, T. W.;
Wuts, P. G. M. Protective Groups in Organic Synthesis, 3rd ed.; Wiley-
Interscience: New York, 1999), our efforts failed with the benzyl ether
of 29, which consistently gave the undesired bis-primary dibenzyl
ether, therefore preventing the use of the whole isomeric mixture of
acetals for the convergent synthesis of 32.
(21) Bartoli, S.; Roelens, S. J . Am. Chem. Soc. 2002, 124, 8307-
8315.
(22) Roelens, S.; Torriti, R. Supramol. Chem. 1999, 10, 225-232.
8152 J . Org. Chem., Vol. 68, No. 21, 2003

Binding of Me4N+ to Aromatic Hosts
TABLE 2. Associa tion Con sta n ts Ka (M-1), Sta n d a r d
F r ee En er gies of Bin d in g -∆G° (k J m ol-1), a n d Gu est
Lim itin g Sh ift Va lu es ∆δ∞ (p p m ) for th e 1:1 Com p lexes of
Tetr a m eth yla m m on iu m P icr a te (TMAP ) w ith Sid ea r m ed
Ma cr ocyclic Ar om a tic Liga n d sa
host
Ka (SE)b
∆δ∞
-∆G°
DXP DA
13
14
15
16
17
33
XDP A
29.7 (0.4)c
18.6 (0.3)
29.0 (1.1)
19.1 (0.9)
35.8 (0.7)
16.1 (0.3)
7.1 (0.2)
-1.479c
-1.290
-0.780
-1.267
-1.255
-1.003
-0.510
-0.464c
8.35(3)c
7.19(4)
8.29(9)
7.3(1)
8.81(4)
6.84(4)
4.82(7)
4.7(1)c
6.7 (0.4)c
a
Measured by 1H NMR (200/300 MHz) at T ) 296 K in CDCl3
on 0.1 mM solutions of TMAP using host concentrations up to 60
mg/mL. Standard error of the nonlinear least-squares fit. c Data
b
from ref 5.
and XDPA, are reported in Tables 1 and 2, respectively.
In addition, in Table 1 association data relative to water
binding are also reported for open-chain hosts possessing
polyoxyethylene units. Indeed, in the titration spectra of
the latter, the water signal showed a downfield shift with
increasing host concentration clearly describing a satura-
tion curve, which could be correctly fitted to a 1:1 binding
isotherm. Binding of water molecules is a known phe-
nomenon for crown ethers,23 and we could indeed deter-
mine the association constants for 11 and 12, but these
became too small to be measured for hosts lacking
polyoxyethylene chains. It must be underlined that
binding of water was determined concomitantly with
TMA, thus under competitive conditions. The general
conclusion that can be drawn from inspection of data in
Table 1 is that polyether chains participate in the binding
of TMA, but only to a minor extent. Variations of binding
free energies are small with respect to the reference
diester 1, showing that the contributions provided by the
ether oxygen donors are of minor entity compared to
those from the aromatics. Nevertheless, since the reason-
ably large chemical shifts induced by the hosts upon
binding allowed us to obtain reliable data, a quantitative
evaluation of contributions from the oxygen donors could
be assessed and some general conclusions could be
outlined. Hosts 1-4, which possess an increasing number
of oxyethylene units in the side chains, exhibit increasing
free energy of binding to TMAP; the same is true for
binding to water, showing that polyether chains interact
with both guests. However, binding behavior is different
for the two guests, in that -∆G° increments are linear
for water but exponential for TMAP. This feature can be
better appreciated from the plots of Figure 1, where
experimental -∆G° values were plotted vs the number
of oxygen donors (n) in the side chains and fitted by linear
and exponential regressions, respectively, with excellent
correlation coefficients. Thus, for water, linear regression
(r2 ) 0.999) gave an increment of binding free energy of
1.3 kJ mol-1 per added oxygen pair, while for TMAP data
were fitted to eq 1 with y0 ) 4.88(4) and a ) 1.419(9).
F IGURE 1. Plot of binding free energies -∆G° in CDCl3 at T
) 296 K vs number of oxygen donors in the polyoxyethylene
side chains for binding of TMAP (A) and H2O (B) to open-chain
hosts 1-4. Symbols are experimental data from Table 1; solid
lines are (A) the best fit curve obtained from eq 1 by nonlinear
regression with y0 ) 4.88(4), a ) 1.419(9) and (B) the best fit
line from linear regression (slope ) 0.650; intercept ) 0.687;
r2 ) 0.999). Plots are displayed on the same scale for visual
comparison.
whereas they bind water with larger but independent
increments. The intercept of the curve for TMA (Figure
1A), i.e., the -∆G° value, for a null number of added
polyoxyethylene units, which corresponds to the binding
free energy of the parent ligand 1, shows that the
contribution of polyoxyethylene side chains to the binding
of TMA, altough cooperative, is much smaller than the
contribution of the aromatic, being 5-fold smaller than
the latter even for six oxyethylene units. Instead, for
water, the -∆G° value (0.687 kJ mol-1) of the linear
regression intercept (Figure 1B) shows that 1 binds to
water as strongly as each oxyethylene unit (slope ) 0.650
kJ mol-1), but the contribution appears to be entirely due
to the two ester groups of 1 rather than to the aromatic
ring, since removing the aromatic from the structure
leaves the free energy of binding to water unaffected (see
Table 1, 4 and 5). On the contrary, removing the aromatic
causes binding of TMA to drop below the detectable limit
for 5, confirming that the main contribution to cation
binding is provided by the aromatic ring. The emerging
picture is consistent with a system in which two guests
-∆G° ) y0 + (log a) an
(1)
The analysis shows that polyoxyethylene side chains
bind TMA with small but cooperative contributions,
(23) Izatt, R. M.; Bradshaw, J . S.; Pawlak, K.; Bruening, R. L.;
Tarbet, B. J . Chem. Rev. 1992, 92, 1261-1354.
J . Org. Chem, Vol. 68, No. 21, 2003 8153

Bartoli et al.
(TMA and water) interact with two independent binding
sites (aromatic and oxygen) in an essentially noncompeti-
tive way.
Removing the carbonyl groups from 1 and 4, to give
ethers 6 and 7, respectively, that is, converting ester
groups into ether moieties, induces predictable varia-
tions: the binding free energy increases for water,
whereas it decreases for TMA. For water, although
figures cannot be evaluated but in the case of 4 and 7,
recalling that the contribution of two ester groups was
the same as that of one single oxyethylene unit, it
appears that the full contribution of two oxygen sites has
been restored in 7. Instead, for TMA, the binding free
energy decrease is worth 2 kJ mol-1 for 7 with respect to
4 and even larger for 6 with respect to 1. On the basis of
models computed at the AM1 level of theory, we have
shown5 that ester groups, which in the binding conforma-
tion lie orthogonal to the plane of the aromatic ring, can
exert a cooperative contribution to the binding of TMA
by polarizing the electrostatic potential of the aromatic
out of plane over one side of the ring, thus rendering the
charge density fully available to interact with the cationic
guest. Such a contribution is lost in the corresponding
ethers, and a plausible rationale is offered by analogous
calculations at the AM1 level (on the dimethyl ether for
simplicity) showing that, in contrast to the ester coun-
terpart, the ether groups depart from the cooperative
orthogonal conformation to lie on the plane of the
aromatic ring in the optimized geometry and that in the
latter the electrostatic potential is polarized in plane with
the aromatic ring and symmetrically distributed over the
two faces (Figure 2). Accordingly, a drop of binding to
TMA is observed for 8, which is the cyclic diortho ester
of 1, due to the loss of two ester groups only partially
compensated by the gain of two pairs of ether-type oxygen
donors. Because of this compensation, the corresponding
binding free energy becomes a measurable quantity,
compared to the simple diether 6, and exhibits a remark-
able additivity of contributions: the value of -∆G° for
TMA can be obtained by adding up the contributions from
one aromatic (2 kJ mol-1)5 and from two oxygen pairs
(0.35 kJ mol-1). The association becomes measurable also
for the dibenzyl ether 9, because of the added phenyl
rings that, however, do not seem to fully contribute to
binding. The larger upfield shift of the complex supports
participation of the phenyl rings, but their contribution
is in fact definitely smaller than that expected for two
aromatics. That the binding contribution of the ester
group is due to the carbonyl rather than to the alcoholic
oxygen is demonstrated by the diketone 10, whose
binding free energy to TMA is nearly the same as that
of 1. Comparison of data for 10, 6, and 1 clearly shows
that the contribution from the alcoholic oxygen of the
ester toward TMA is negligible.
F IGURE 2. AM1-computed model of 1,4-benzenedimethanol
dimethyl ether in the optimized conformation. The optimized
geometry is depicted as a polytube model; the electrostatic
potential isosurface at -10 kcal mol-1 is shown in transparent
body representation: (A) side view and (B) top view with
respect to the aromatic ring plane.
of 1); by adding the value corresponding to two aromatic
rings (4 kJ mol-1) to this figure, -∆G° ) 6.3 kJ mol-1 is
obtained for 11, in excellent agreement with the experi-
mental value, supporting the conclusion that contribu-
tions are additive. On the other hand, with respect to
water, 12 binds better than 11 by 2.5 kJ mol-1; the -∆G°
value calculated from the linear regression for the
corresponding number of oxygen donors (5.2 kJ mol-1
)
is somewhat smaller than that of 12, but larger than that
of 11 (Table 1). For 12, stronger binding than expected
may likely be ascribed to the macrocyclic structure, more
preorganized than that of the open-chain ligands, but this
bonus is lost for 11, for which binding of water may be
hampered by the conformational changes imposed by the
concomitant association with TMA in order to adapt the
aromatic rings to bind the cation, in a sort of allosteric
effect. Eventualy, the literature24 value for the association
of water to 11 gives a binding free energy -∆G° ) 4.25
kJ mol-1, in close agreement with the value reported in
Table 1, thus providing an external check of the reli-
ability of the present data.
From comparison of 1-10 with the two reference crown
ethers 11 and 12, it can immediately be noted that 12,
having the same size and number of oxyethylene units
as 11 but lacking the two aromatic rings, does not display
any binding capacity toward TMA, whereas 11 exhibits
the largest value of the whole set. Thus, as expected,
eight oxyethylene units cannot compensate for the loss
of two aromatic rings. The contribution to -∆G° from
eight oxyethylene units can be calculated by eq 1, to give
2.3 kJ mol-1 (after subtracting the binding free energy
Inspection of results in Table 2 confirms the general
conclusion that also in the marocyclic series oxyethylene
(24) Golovkova, L. P.; Telyatnik, A. I.; Bidzilya, V. A.; Akhmetova,
N. E.; Konovalova, V. I. Teor. Exp. Chem. (Engl. Transl.) 1985, 21,
238-242.
8154 J . Org. Chem., Vol. 68, No. 21, 2003

Binding of Me4N+ to Aromatic Hosts
side chains do not improve cation binding by substantial
amounts. In fact, with one exception, all ligands show
weaker binding than the parent unsubstituted tetraester
DXPDA. Introduction of four phenolic oxygen function-
alities in the xylylene moieties of DXPDA is expected to
raise the electronic density and, consequently, it may
enhance the binding capacity of the host toward the
cationic guest. Likewise, the introduction of benzyl side
chains may enhance the association by incrementing the
number of binding goups. In contrast, it is easily noted
that neither one is the case, since the tetrabenzyloxy-
substituted 17 shows a binding free energy value 1.5 kJ
mol-1 smaller than that of DXPDA. On the other hand,
comparison between the corresponding open-chain “half-
molecules”, the unsubstituted XPDA and the dibenzyloxy-
substituted 33, does not highlight any appreciable dif-
ference in binding to TMA, indicating that neither the
electron-donating phenolic oxygens nor the benzylic side
chains induce any increase in binding capacity. This
evidence confirms the prediction made, in choosing the
dihydroxyterephthalate reagent for the synthesis of the
sidearmed macrocyclic hosts, that binding properties
would not be affected by electronic factors. As expected,
the free energy values for XPDA and 33 are smaller than
those of their macrocyclic counterparts; this reflects the
larger number of binding groups and some degree of
preorganization that the macrocyclic structures, although
flexible, possess compared to their open-chain analogues;
however, the drop in binding strength observed for 17
with respect to DXPDA, a drop lacking for the open-chain
ligands, points to a binding hindrance affecting the
substituted macrocycles, which may be steric and/or
conformational in origin. Taking into account such a
binding inhibition, then side chains possessing an in-
creasing number of oxyethylene units do increase the
binding strength of the macrocyclic hosts as in the open-
chain series. Thus, addition of one oxyethylene unit to
the side chains of 17 raises the binding free energy by
0.46 kJ mol-1 for 15, a value that is nearly unaffected
by replacement of the terminal phenyl groups with
methyl groups as in 13, suggesting that phenyl end
groups cannot easily achieve a cooperative binding
conformation. Instead, three oxyethylene units in 14 raise
the binding free energy by 1.45 kJ mol-1 with respect to
17 and by 1.1 kJ mol-1 with respect to its lower homo-
logue 13, bringing a contribution significantly larger than
that of one single unit. However, while in 13 the four
independent oxyethylene units can probably arrange into
a binding conformation around the cation, thus exerting
their full contribution, this cannot obviously be achieved
by the total 12 oxyethylene units of which 14 is endowed,
necessarily bringing in only a partial contribution. In
terms of additivity of contributions, 13 and 15 show
binding free energy increments very similar to that of
the open-chain counterpart endowed with the same
number of ether units (3), whereas 14 exhibits a binding
increment slightly larger than that of the corresponding
open-chain ligand possessing only half of its oxyethylene
units (4). The glycerol derivative 16 is the only host for
which a net increase in binding free energy is observed
with respect to the parent DXPDA. This may be ascribed
to the branching of the side chains, which may allow for
a better adaptment to the guest, but the improvement
in binding energy, 2 kJ mol-1 with respect to 17 and less
than 0.5 kJ mol-1 with respect to DXPDA, is indeed very
modest compared to the complexity of the structure and
the number of binding groups featured. In conclusion,
polyether chains are intrinsically beneficial as coopera-
tive binding side chains, but they introduce steric/
conformational constraints in the macrocyclic tetraester
that override any advantages gained in binding strength.
Con clu sion . From a systematic investigation of the
binding contribution exerted by polyether side chains on
the interaction of the tetramethylammonium cation with
phanic receptors, the following conclusions have been
drawn: (a) binding of the guest is primarily due to the
aromatic moieties of the receptors; ether groups provide
minor but cooperative contributions whenever they can
achieve a binding arrangement around the cation; (b)
contributions of ether groups are additive for open-chain
hosts, with exponentially growing increments of binding
free energy, whereas participation appears to be strongly
affected by steric and/or conformational constraints for
macrocyclic sidearmed hosts; (c) in contrast to aromatic
rings, polyether chains preferentially bind to water than
to TMA, showing binding free energy increments linearly
additive; (d) for the set of phanic esters investigated,
water and TMA independently interact with polyether
chains and aromatic binding sites respectively in an
essentially noncompetitive fashion; (e) ester groups par-
ticipate in binding of TMA through the carbonyl and of
water through the alcoholic oxygen. In summary, the
contribution from the ether oxygen donors of polyether
chains to the binding of TMA to aromatic hosts could be
reliably assessed on a quantitative basis using a set of
ligands featuring a systematic variation of substituents.
Results confirm that the binding energies expected from
gas-phase data for the interaction between TMA and
oxygen donors vanish to a large extent in solution, in
contrast to the cation-π interaction established with
aromatic donors, which seems to suffer to a minor extent
from the transfer to the condensed phase.
Exp er im en ta l Section
Syn th esis of 1-[4-(2-Oxop r op yl)p h en yl]p r op a n -2-on e,
10. Copper(I) cyanide (582 mg, 6.50 mmol) was suspended in
dry tetrahydrofuran (13 mL) and cooled to -78 °C. A 1.6 M
solution of methyllithium in diethyl ether (3.55 mL, 6.58 mmol)
was added dropwise and the mixture allowed to warm to 0 °C
and cooled again to -78 °C. A solution of (4-chlorocarbonyl-
methylphenyl)acetyl chloride (657 mg, 2.84 mmol) in dry
tetrahydrofuran (7 mL) was added dropwise and the mixture
stirred at 0 °C for further 30 min. Methanol (10 mL) was added
and the mixture poured into water (20 mL). The aqueous layer
was extracted with dichloromethane (3 × 30 mL). The com-
bined organic layer was washed with water (20 mL) and dried
(Na2SO4) and the solvent removed under reduced pressure.
Purification by column chromatography (petroleum ether/ethyl
acetate 3:2), gave 253 mg of 10 as a white solid (47% yield).
Mp: 52-53°C (lit.13 mp: 46-47 °C). Elemental analysis: calcd,
1
C 75.75, H 7.42; found C 76.02, H 7.51. H NMR (200 MHz,
CDCl3): δ 2.15 (s, 6H), 3.68 (s, 4H), 7.17 (s, 4H). 13C NMR (50
MHz, CDCl3): δ 29.3, 50.5, 129.8, 133.06, 206.3. MS-EI: m/z
(%) 191 (11), 190 (12) [M+], 148 (71), 106 (12), 105 (100), 104
(49), 103 (70), 91 (42), 89 (11), 84 (17), 78 (44), 77 (78), 63 (24),
52 (14), 51 (29), 50 (21).
Syn th esis of [2,5-Bis(2-eth oxyeth oxy)-4-h yd r oxym eth -
ylp h en yl]m eth a n ol, 24. Gen er a l Meth od B. Compound 18
(3.46 g, 13.6 mmol), Cs2CO3 (30 g, 92 mmol), and methane-
sulfonic acid 2-ethoxyethyl ester (4.58 g, 27.2 mmol) were
J . Org. Chem, Vol. 68, No. 21, 2003 8155

Bartoli et al.
suspended in acetone (120 mL) and refluxed to disappearance
of the starting material and formation of one single product
(TLC). The mixture was then filtered through Celite and the
solvent evaporated. The residue was dissolved in dichloro-
methane (200 mL), washed with water (3 × 100 mL), and dried
(Na2SO4), and the solvent was removed under reduced pres-
sure to give crude 19 (5.42 g, 100%). A solution of 19 (5.42 g,
13.6 mmol) in tetrahydrofuran (10 mL) was carefully added
to a solution of LiAlH4 (8 mL, 1 M solution in tetrahydrofuran)
while the temperature was kept below 40 °C. After 1 h with
vigorous stirring, the solution was cooled to 0 °C and ethyl
acetate was added carefully to quench the excess of LiAlH4.
The mixture was poured into a 10% aqueous solution of HCl
(300 mL) and the aqueous layer was extracted with dichloro-
metane (3 × 100 mL). The combined organic layer was dried
(Na2SO4) and the solvent removed under reduced pressure to
give the crude product 24 as a yellow oil (3.05 g, 9.70 mmol,
Syn th esis of 15. Gen er a l Meth od D. Compound 15 was
first obtained from 26 according to method C on a 4.76 mmol
scale (155 mg, 7.4% yield) after column chromatography
(dichlorometane/5% ethyl acetate); 2.2 g (3.69 mmol based on
the molecular weight of the monomer, 77% yield) of oligomeric
residue was obtained from column chromatography (dichloro-
methane/methanol 10:1). The oligomeric residue from method
C (2.2 g, 3.69 mmol), dibutyltin oxide (90 mg, 0.362 mmol),
and dibutyltin chloride (110 mg, 0.362 mmol) were refluxed
for 2 days in toluene (600 mL). Dipyridile (59 mg, 0.362 mmol)
was added, the solution filtered to remove the precipitate
(DBTC/dipyridile complex), and the solvent removed under
reduced pressure. The residue was purified by column chroma-
tography (gradient: dichloromethane/6% to 8% ethyl acetate),
to give 300 mg (13.9% yield) of 15 as a white solid and 1.65 g
(69% yield) of oligomeric residue that could be recycled again
as described above. Mp: 127-128 °C. Elemental analysis:
1
71% yield). H NMR (200 MHz, CDCl3): δ 1.22 (t, J ) 7 Hz,
1
calcd, C 72.47, H 6.08; found, C 72.73, H 6.27. H NMR (200
6H), 3.49 (q, J ) 7 Hz, 4H), 3.70-3.80 (m, 4H), 4.12-4.22 (m,
4H), 4.62 (s, 4H), 6.84 (s, 2H). Product 24 was identified by
1H NMR and used as such for the following macrocyclization
reaction.
MHz, CDCl3): δ 3.57 (s, 8H), 3.59-3.64 (m, 8H), 3.84-3.89
(m, 8H), 4.48 (s, 8H), 5.09 (s, 8H), 6.56 (s, 4H), 7.16 (s, 8H),
7.26-7.34 (m, 20H). 13C NMR (50 MHz, CDCl3): δ 41.4, 61.9,
68.6, 68.6, 73.3, 113.1, 125.1, 127.7, 128.5, 129.5, 133.0, 138.0,
150.4, 170.9. MS-EI: m/z (%) no M+, 111 (18), 110 (10), 98
(12), 97 (36), 96 (21), 95 (14), 91 (11), 85 (39), 84 (29), 83 (57),
82 (37), 81 (21), 71 (49), 70 (45), 69 (61), 68 (39), 67 (20), 66
(12), 56 (100), 55 (82), 54 (70).
Syn th esis of 13. Gen er a l Meth od C. Compound 24 (1.50
g, 4.78 mmol) and dibutyltin oxide (1.19 g, 4.78 mmol) were
dissolved in toluene (50 mL) in a flask equipped with a Dean-
Stark apparatus, and the mixture was refluxed for 24 h. A
solution of (4-chlorocarbonylmethylphenyl)acetyl chloride (1.10
g, 4.78 mmol) in CHCl3 (50 mL) was added batchwise and the
mixture refluxed for a further 45 min. The solvent was
removed under reduced pressure and the residue washed with
petroleum ether (3 × 50 mL) and purified by column chroma-
tography (dichloromethane/ethyl acetate 3:1), giving 153 mg
of 13 as a white solid (7% yield). Mp: 141-142 °C. Elemental
analysis: calcd, C 66.09, H 6.83; found, C 65.91, H 6.78. 1H
NMR (200 MHz, CDCl3): δ 1.17 (t, J ) 7 Hz, 12H), 3.47 (q, J
) 7 Hz, 8H), 3.55-3.60 (m, 8H), 3.65 (s, 8H), 3.81-3.86 (m,
8H), 5.08 (s, 8H), 6.54 (s, 4H), 7.22 (s, 8H). 13C NMR (50 MHz,
CDCl3): δ 15.3, 41.5, 61.9, 66.9, 68.8, 69.0, 113.1, 125.1, 129.5,
133.1, 150.4, 170.8. MS-EI: m/z (%) no M+, 111 (34), 97 (12),
88 (11), 86 (61), 85 (22), 84 (100), 83 (38), 82 (32), 81 (79), 71
(16), 70 (11), 69 (16), 57 (27), 55 (20), 54 (11).
Ack n ow led gm en t. We would like to thank the
Consorzio Interuniversitario Risonanze Magnetiche di
Metalloproteine Paramagnetiche (C.I.R.M.M.P.) for a
grant to S.B.
Su p p or tin g In for m a tion Ava ila ble: General methods
and materials; experimental details for the synthesis of
compounds 1-9, 14, 16, 17, 25-28, and 33. This material is
available free of charge via the Internet at http://pubs.acs.org.
J O034905H
8156 J . Org. Chem., Vol. 68, No. 21, 2003