Concave reagents: Part 40 - The copper(ii) complex of a concave reagent as a selective catalyst for ester methanolysis

Article abstract of DOI:10.1002/poc.735

We have shown that the Cu^II complexes of the concave ligand 1 and its model compound 2 are efficient catalysts of ester methanolysis under conditions close to neutrality. Turnover catalysis without product inhibition was demonstrated by the clean first-order release of a greater than stoichiometric amount of product. Compared with background methanolysis, the metal catalysts give greater rate accelerations for methyl acetate methanolysis than for the p-nitrophenyl acetate methanolysis. Analysis of electronic and steric effects on rates of metal-mediated vs metal-free methoxide addition to the esters has provided compelling evidence that transfer of methoxide from the metal to the carbonyl carbon is accompanied by extensive Lewis acid activation of the carbonyl via a four-membered chelate transition state that includes the metal ion. Copyright

Full text of DOI:10.1002/poc.735

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2004; 17: 350–355
Published online 17 February 2004 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/poc.735
Concave reagents: Part 40y —The copper(II) complex
of a concave reagent as a selective catalyst for ester
methanolysis
Roberta Cacciapaglia,1 Stefano Di Stefano,1 Frank Fahrenkrug,2 Ulrich Lu¨ ning2 and Luigi Mandolini1*
1
`
Dipartimento di Chimica and IMC–CNR, Sezione Meccanismi di Reazione, Universita La Sapienza, Box 34–Roma 62, 00185 Rome, Italy
2Institut fu¨ r Organische Chemie, Universita¨t Kiel, Olshausenstr. 40, D-24098 Kiel, Germany
Received 6 October 2003; revised 17 November 2003; accepted 17 November 2003
ABSTRACT: We have shown that the CuII complexes of the concave ligand 1 and its model compound 2 are efficient
catalysts of ester methanolysis under conditions close to neutrality. Turnover catalysis without product inhibition was
demonstrated by the clean first-order release of a greater than stoichiometric amount of product. Compared with
background methanolysis, the metal catalysts give greater rate accelerations for methyl acetate methanolysis than for
the p-nitrophenyl acetate methanolysis.
Analysis of electronic and steric effects on rates of metal-mediated vs metal-free methoxide addition to the esters
has provided compelling evidence that transfer of methoxide from the metal to the carbonyl carbon is accompanied by
extensive Lewis acid activation of the carbonyl via a four-membered chelate transition state that includes the metal
ion. Copyright # 2004 John Wiley & Sons, Ltd.
KEYWORDS: concave reagents; 1,10-phenanthroline ligands; ester methanolysis; mechanism of metal catalysis;
homogeneous catalysis; selectivity
INTRODUCTION
concave ligand on reactivity and selectivity in the CuII-
catalyzed cleavage of esters.
Attempts at mimicking the concave geometry of enzyme
active sites that are crucial for the high substrate selectivity
of their reactions led to the development of concave
reagents whose shape was inspired by a simple model,
the light bulb in a lamp shade.1 Cyclophane structures
incorporating a 2,9-diaryl-1,10-phenanthroline unit (e.g. 1)
constitute an important family of concave reagents, in
which the role of the light bulb is played by a 1,10-
phenanthroline-bound metal ion. Concave 1,10-phenan-
throlines proved to be selectivity-inducing ligands in a
number of synthetically useful reactions, such as transition
metal-catalyzed Diels–Alder reactions,2 PdII-catalyzed
allylations3 and CuI-catalyzed cyclopropanations.4
We report here the results of a kinetic investigation of
methanolysis of esters 3–7 catalyzed by 1[Cu] and by the
model complex 2[Cu]. The choice of methanol as reac-
tion medium was dictated by the insolubility of 1 and of
its CuII complex in water. Neocuproine (2) was used as a
model ligand because the steric effect of the two methyl
groups5d prevents the formation of unreactive dimeric
complexes.5f
The catalytic activity of CuII complexes in the hydro-
lysis of carboxylic and phosphoric acid esters is well
documented,5 and 1,10-phenanthrolines are known to
form strong complexes with CuII salts.6 To widen the
scope of 1,10-phenanthroline-based concave reagents, we
undertook an exploratory investigation aimed at ascer-
taining the influence of a 1,10-phenanthroline-based
RESULTS
*Correspondence to: L. Mandolini, Dipartimento di Chimica and
`
Copper(II) complexes of 1 and 2 were freshly generated
in methanol solution from CuCl2. The UV–visible titra-
tion experiment in Fig. 1 indicates the existence of a
strong (K > 106 Mꢀ1) complexation between CuII and the
IMC–CNR, Sezione Meccanismi di Reazione, Universita La Sapienza,
Box 34–Roma 62, 00185 Rome, Italy.
E-mail: luigi.mandolini@uniroma1.it
yPart 40 of the series Concave Reagents. For part 39, see Storm O,
Lu
¨ning U. Eur. J. Org. Chem. 2003; 3109–3116.
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 350–355

COPPER–CONCAVE REAGENT COMPLEX CATALYST FOR ESTER METHANOLYSIS
351
The acidity-enhancing effect of the concave reagent
1[Cu] is an order of magnitude lower than that of the
corresponding neocuproine species 2[Cu]. This is
most likely related to steric repulsion between the meth-
oxide ligand and the rim of the lamp shade. [X-ray
molecular structures of complexes between CuCl2 and
2,9-disubstituted-1,10-phenanthrolines reveal distorted
tetrahedral species, intermediate between tetrahedral
and square.5d,6b Whereas neutral tetracoordinated CuII
complexes with chelate ligands usually have a planar
coordination, distorted tetrahedral coordination is well
documented in the presence of bulky substituents.9 We
assume that similar distorted CuII occurs in complexes of
Eqn (1).] Interestingly, the increase in acidity of metha-
nol8 on coordination to the neocuproine species 2[Cu]
(ÁpKa ¼ ꢀ 10.6) is nearly two orders of magnitude
higher than the corresponding increase in acidity of water
(ÁpKa ¼ ꢀ8.7).5d
Figure 1. Variations of the UV–visible spectrum of 0.080
mM 1 on addition of increasing amounts of CuCl2 in CH3OH
at 25 ꢂC. A titration plot at 350 nm is shown in the inset
The technique chosen for the kinetics to be conveni-
1
ently monitored was H NMR spectroscopy, which re-
Figure 2. Potentiometric titration of a 1.30 mM 1:1 mixture
of 1 and CuCl2 in CH3OH at 25 ꢂC
quired CD3OD to be used as reaction medium. To carry
out the kinetics around neutrality, a suitable buffer was
sought. Whereas the commonly used N-2-hydroxyethyl-
piperazine-N0-ethanesulfonic acid (HEPES) buffer is
sparingly soluble in methanol, other buffers such as 1-
methylimidazole–perchlorate salt and 4-methylmorpho-
line–perchlorate salt7 proved to be inappropriate, because
of interactions between the buffer components and the
CuII of both 1[Cu] and 2[Cu] as revealed by the UV–
visible spectra. Therefore, catalyst solutions were pre-
pared in situ by mixing stock solutions of phenanthroline
ligand (L), CuCl2 and CD3ONa in CD3OD in the stoichio-
metric ratio 1:1:0.9. A less than stoichiometric amount of
CD3ONa was used to reduce the concentration of free
CD3Oꢀ and, consequently, the rate of background metha-
nolysis. Typical reactant concentrations in the kinetic runs
were 10 mM ester and 4 mM metal catalyst. Kinetic data
for the more reactive substrates 5–7 were obtained from
time-course experiments, whereas an initial-rate techni-
que was conveniently used for the slower substrates 3
and 4. Typical time–concentration profiles are shown in
Figs. 3 and 4.
1,10-phenanthroline unit in 1. Potentiometric titration7
(Fig. 2) reveals that the complex of 1 with CuCl2 behaves
as a weak monoprotic acid with pKa ¼ 8.6 ꢁ 0.1 (the
molar autoprotolysis constant of CH3OH is
Kap ¼ 10ꢀ16.77,8 which implies that in this solvent the
pH value corresponding to neutrality is 8.39 and that the
pKa of CH3OH is 18.16). A pKa value of 7.6 ꢁ 0.1 was
measured in a similar way for the corresponding complex
of neocuproine (2). By analogy of the acid–base beha-
viour of CuII complexes of 1,10-phenanthrolines and
other bidentate nitrogen ligands in aqueous solution,5b–d
the above data are interpreted as the pKa values of a
metal-bound methanol molecule [Eqn (1)].
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 350–355

352
R. CACCIAPAGLIA ET AL.
more acidic phenol product ( p-nitrophenol, pKa in
MeOH ¼ 11.30 at 25 ꢂC10) liberated in the methanolysis
[Eqn (2)] and, consequently, that the concentration of the
base catalyst remains constant throughout the entire
reaction course. Consistently, only the signals of undis-
1
sociated phenol products were found in the H NMR
spectra at the end of reaction and clear first-order beha-
vior was observed in all of the time-course kinetic runs.
In these experiments, the release of a greater than
stoichiometric amount of product without loss of cataly-
tic activity demonstrates turnover catalysis and lack of
product inhibition.
cat
ꢀꢀꢀ!
RCO2R0 þ CD3OD
RCO2CD3 þ R0OH
Figure 3. Product concentration–time profile in the basic
methanolysis of 10.8 mM p-nitrophenyl propanoate (6) in the
presence of 4.00 mM 1[Cu]OCD3 (CD3OD, 25 ꢂC)
cat ¼ L½CuꢃOCD3
ð2Þ
Numerical values of the second-order rate constants (k)
for the reaction of the metal-bound CD3Oꢀ with the ester
substrates [Eqn (3)] were calculated as kobs/[cat], where
kobs is the directly measured pseudo-first-order rate con-
stant [Eqn (4)] corrected for background methanolysis
whenever appropriate (see below).
rate ¼ k½catꢃ ½esterꢃ
kobs ¼ k½catꢃ
ð3Þ
ð4Þ
Rate data are summarized in Table 1. For comparison
purposes, Table 1 also lists second-order rate constants
for reactions of esters with uncomplexed CD3Oꢀ ion
(Naþ salt). Combination of the latter data with the pOMe
values given above afforded approximate rates of back-
ground methanolysis under the conditions of the kinetic
experiments. Methanolyses catalyzed by 2[Cu]OCD3
were in all cases much faster than background, with
ratios of catalyzed to background rates in the range
2600-110, limiting values measured for the reactions of
4 and 5, respectively. However, lower ratios of catalyzed
to background rates were calculated for the reactions
catalyzed by 1[Cu]OCD3, owing to a combination of a
higher basicity of the catalyst solution and a lower
catalytic efficiency compared with 2[Cu]OCD3. Thus,
for ester methanolyses catalysed by 1[Cu]OCD3, back-
ground rates are likely to be a significant fraction of the
Figure 4. Product concentration–time profile in the basic
methanolysis of 10.0 mM phenyl acetate (4) in the presence
of 4.00 mM 2[Cu]OCD3 (CD3OD, 25 ꢂC)
The pH (pOMe) values of CH3OH solutions having the
same composition as those used in the kinetic runs are 9.5
(7.3) and 8.5 (8.3) for the catalyst solutions based on
ligand 1 and 2, respectively. If the above values are taken
as representative of the unknown pD (pOMe) values of
the kinetic solutions, the conclusion is reached that
phenoxide production is negligibly small even from the
Table 1. Second-order rate constants (Mꢀ1 sꢀ1) for the reactions of esters 3–7 with free and CuII-complexed CD3Oꢀ anion in
CD3OD at 25 ꢂC
Compound
CD3Oꢀ
k
1[Cu]OCD3 k1
2[Cu]OCD3 k2
k/k1
k/k2
k2/k1
ꢀ4
ꢀ4
3
4
5
6
7
0.17
4.4a
190
90
1.3 ꢄ 10
4.2 ꢄ 10
1300
2100
ꢅ 35 000
ꢅ 50 000
ꢅ 86 000
400
270
6600
3900
1400
3.2
7.6
ꢅ 5.3
ꢅ 13
ꢅ 60
ꢀ3
ꢀ2
2.1 ꢄ 10
1.6 ꢄ 10
ꢀ3b
ꢀ2
(2.8–5.5) ꢄ 10
(0.4–1.8) ꢄ 10
(0–3.5) ꢄ 10
2.9 ꢄ 10
ꢀ3b
ꢀ5b
ꢀ2
2.3 ꢄ 10
ꢀ3
3.0
2.1 ꢄ 10
a
From Ref. 19.
See text and Table 2.
b
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 350–355

COPPER–CONCAVE REAGENT COMPLEX CATALYST FOR ESTER METHANOLYSIS
353
Table 2. Methanolysis of esters 3–7 catalyzed by 1[Cu]OCD3 in CD3OD at 25 ꢂC: correction for background methanolysis (rate
constants in sꢀ1
)
(KaH/KaD) ¼ 6a
(KaH/KaD) ¼ 1b
c
d
c
kbg
d
Compound
kobs
kbg
ko*bs
ko*bs
ꢀ7
ꢀ 6
ꢀ 5
ꢀ 6
ꢀ7
ꢀ8
ꢀ7
ꢀ5
ꢀ6
ꢀ7
ꢀ7
ꢀ6
ꢀ5
ꢀ6
ꢀ9
ꢀ8
ꢀ6
ꢀ6
ꢀ8
ꢀ7
ꢀ6
ꢀ5
ꢀ6
ꢀ7
3
4
5
6
7
5.2 ꢄ 10
8.4 ꢄ 10
2.4 ꢄ 10
8.0 ꢄ 10
1.8 ꢄ 10
1.2 ꢄ 10
3.2 ꢄ 10
1.3 ꢄ 10
6.5 ꢄ 10
2.2 ꢄ 10
5.1 ꢄ 10
8.1 ꢄ 10
1.1 ꢄ 10
1.5 ꢄ 10
0
2.0 ꢄ 10
5.2 ꢄ 10
2.2 ꢄ 10
1.1 ꢄ 10
3.6 ꢄ 10
5.2 ꢄ 10
8.3 ꢄ 10
2.2 ꢄ 10
6.9 ꢄ 10
1.4 ꢄ 10
a pOMe ¼ 7.14.
b pOMe ¼ 7.92.
c Rate constant of background methanolysis, calculated as kbg ¼ k ꢄ 10ꢀpOMe, with k values from Table 1.
d ko*bs¼ kobs ꢀ kbg
.
overall rates, and corrections are required. Since the exact
pOMe value of the catalyst solution is unavailable, an
approximate treatment was adopted, based on the isotope
effect of KaHp/KaDp ¼ 5 for the autoprotolysis of pure
methanol,11 and on the assumption that 1 ꢆ KaH/KaD ꢆ 6
for the ionization of CuII-bound methanol. This assump-
tion was based on the analogy of the ionization of weak
acids in water, for which the KaH/KaD values are mostly in
the range 3–6, but drop to values very close to 1 for the
aquo complexes of thallium(III) and iron(III).12 Thus, the
two sets of rate constants of background methanolysis
(kbg) listed in Table 2 represent upper and lower esti-
mates, but it seems likely that values based on the
assumption KaH/KaD ¼ 1 are a better approximation.
The neocuproine catalyst 2[Cu]OCD3 is more effective
than the concave catalyst 1[Cu]OCD3 in the cleavage of
all of the esters investigated (Table 1, last column), in
spite of the higher basicity and, presumably, higher
nucleophilicity of the latter. This finding points to the
existence of steric repulsion between the ester under-
going nucleophilic attack and the rim of the concave
catalyst. The resulting adverse effect on catalytic rate is
already apparent in the cleavage of the sterically unhin-
dered acetate esters 3–5, and becomes very significant on
increasing the steric bulk of the acyl portion of p-
nitrophenyl esters in the order 5 < 6 < 7.
DISCUSSION
The catalytic mechanism involves transfer of the
methoxide nucleophile from the metal ion to the carbonyl
carbon either with (A) or without (B) Lewis acid activa-
tion of the carbonyl group.15 The two mechanisms are
kinetically equivalent, but are conceptually different in
that they proceed via valence isomeric transition states,
whose structures are characterized by the presence (A) or
absence (B) of an interaction between the metal ion and
the carbonyl oxygen of the altered substrate in the
transition state.
As shown in Table 1, the reactivity orders 3 < 4 < 5 and
5 > 6 > 7 expected on the basis of electronic effects in the
alkyl (aryl) portion and steric effects in the acyl portion of
the esters, respectively, are experienced not only by free
CD3Oꢀ ion, but also by the CuII-complexed species. Not
surprisingly, the more stable, less basic CuII-complexed
species are in all cases much less reactive than free
CD3Oꢀ ion, which is well in keeping with published
data on the hydrolysis catalyzed by hydroxo complexes
of CuII and ZnII.5b,13
Application of the Hammond postulate16 to the me-
chanism involving nucleophile delivery without electro-
philic assistance (B) leads to the conclusion that the more
stable, less reactive CuII-bound methoxide species should
be more selective than the more reactive free methoxide in
attacking an activated rather than a deactivated substrate.
Our data are markedly at variance with the above conclu-
sion. For example, p-nitrophenyl acetate (5) reacts with
free CD3Oꢀ ion 43 times more rapidly than with phenyl
acetate (4), but with the CuII-bound species the reactivity
ratio drops to factors of about 2. Such an extremely low
sensitivity to the electronic activation of the p-NO2 group
is clearly inconsistent with mechanism B, but is well
It has been pointed out by several workers that cata-
lysts which are highly effective at hydrolyzing activated
esters are sometimes either much less effective or not
effective at all towards unreactive esters.5b,14 This is
clearly not the case with our metal catalysts. Free
methoxide ion reacts with p-nitrophenyl acetate (5)
1100 times more rapidly than with methyl acetate (3),
but the corresponding reactivity ratio drops to 69 in the
reaction with 2[Cu]OCD3 and to a value lying somewhere
in the range 22–42 in the reaction with 1[Cu]OCD3.
Thus, compared with background methanolysis, the me-
tal catalysts give greater rate accelerations for methyl
acetate than for p-nitrophenyl acetate.
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 350–355

354
R. CACCIAPAGLIA ET AL.
EXPERIMENTAL
Instruments and general methods
Reactions of CD3ONa with substrates 4–7 were moni-
tored by UV–visible spectrophotometry. In all other cases
1
kinetics were monitored by H NMR spectroscopy on
either a 300 or 200 MHz Bruker spectrometer in the
presence of cyclohexane as an internal standard.
An initial rate technique was used for reactions of 3,
4 and 7 with L[Cu]OCD3 and of 3 with CD3ONa
(error ꢁ 20%), whereas time-course kinetic experiments
were carried out for reactions of 5 and 6 with L[Cu]OCD3
and 4–7 with CD3ONa (error ꢁ 10%). In the kinetic runs
1
the following H NMR signals were monitored: 3, reac-
tion with CD3Oꢀ and with L[Cu]OCD3, ꢀ ¼ 3.343 (sing-
let, CH3OD); 4, reaction with L[Cu]OCD3, ꢀ ¼ 2.018
(singlet, CH3CO2CD3); 5, reaction with L[Cu]OCD3,
ꢀ ¼ 2.018 (singlet, CH3CO2CD3); 6, reaction with
L[Cu]OCD3, ꢀ ¼ 2.330 (quartet, CH3CH2CO2CD3); 7
reaction with L[Cu]OCD3, ꢀ ¼ 1.181 (singlet,
(CH3)3CO2CD3).
Potentiometric titrations were carried out with a Crison
GLP22 pH meter under an argon atmosphere. Potentio-
metric data were elaborated with Hyperquad 2000 ver-
sion 2.1 (NT).
Figure 5. Logarithmic plot of the second-order rate
constant (k2) for the reaction of 2[Cu]OCD3 with p-nitro-
phenyl esters 5–7 against the second-order rate constant (k)
for the corresponding reaction of uncomplexed CD3Oꢀ ion
in CD3OD, 25 ꢂC (data from Table 1)
accommodated by mechanism A. The p-NO2 substituent
favours nucleophilic attack at the carbonyl carbon, but
decreases the Lewis basicity of the carbonyl oxygen of the
altered substrate in the transition state, with nearly com-
plete cancellation of the two opposing effects.
Materials
Ligand 2 and substrates 3–5 and 7 were commercial
samples. Substrate 6 was prepared by reaction of p-
nitrophenol with propanoic acid in dioxane in the pre-
sence of DCC.17 Ligand 1 was synthesized according to
literature procedures.18 CD3OD and CuCl2 were pur-
chased from Aldrich and used as received. CD3ONa
solutions were prepared from sodium metal and
CD3OD and subsequently titrated using a Normex
0.01 M HCl standard solution.
It is remarkable that the sensitivity to steric effects in
the acyl portion of p-nitrophenyl esters 5–7 is much lower
in the reactions of the bulkier 2[Cu]OCD3 than in those of
free CD3Oꢀ, as shown by the slope of 0.67 of a plot of log
k2 vs log k (Fig. 5). Again, this finding strongly argues
against the operation of mechanism B, which predicts a
higher degree of nucleophile–carbon bond formation in
the transition state of the metal complex reaction and,
consequently, a higher sensitivity to steric effects. Owing
to the chelate structure of the four-membered ring-shaped
transition state that includes the metal ion (mechanism
A), the angle between the nucleophile and the ester
carbonyl should be ꢇ90ꢂ, i.e. much smaller than in the
transition state of the metal-free reaction, in which a
tetrahedral geometry is approached. The resulting reduc-
tion in steric repulsion between the metal-bound nucleo-
phile and the substituent attached to the carbonyl carbon
provides a reasonable explanation for the lower sensitiv-
ity of the reaction of 2[Cu]OCD3 to steric effects.
Acknowledgements
This work was carried out in the frames of COST D11
Action, Supramolecular Chemistry and Progetto MIUR
Dispositivi Supramolecolari. A Short Term Mission of
S.D.S. to Kiel University was supported by COST D11.
REFERENCES
In conclusion, the different sensitivity to electronic and
steric effects exhibited by the CuII-catalyzed reactions
compared with the reactions of free methoxide is fully
consistent with a catalytic mechanism involving con-
certed nucleophile transfer and Lewis acid activation.
1. (a) Lu¨ning U. J. Mater. Chem. 1997; 7: 175–182; (b) Atwood JL,
Steed JW, Lu¨ning U. (eds). In Encyclopedia of Supramolecular
Chemistry. Marcel Dekker: New York, in press.
2. Gelbert M, Lu¨ning U. Supramol. Chem. 2001; 12: 435–444.
¨
3. Meynhardt B, Lu¨ning U, Wolff C, Nather C. Eur. J. Org. Chem.
1999; 2327–2335.
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 350–355

COPPER–CONCAVE REAGENT COMPLEX CATALYST FOR ESTER METHANOLYSIS
355
´
¨
4. (a) Loffler F, Hagen M, Luning U. Synlett 1999; 1826–1828; (b)
Bu¨hl M, Terstegen F, Loffler F, Meynhardt B, Kierse S, Muller M,
¨
10. Rived F, Roses M, Bosch E. Anal. Chim. Acta 1998; 374: 309–324.
11. Gerritzen D, Limbach HH. Ber. Bunsenges. Phys. Chem. 1981;
85: 527–535.
¨
¨
¨
¨
Nather C, Luning U. Eur. J. Org. Chem. 2001; 2151–2160.
5. (a) Nakon R, Angelici RJ. J. Am. Chem. Soc. 1973; 95: 3170–
3174; (b) Chin J, Jubian V. Chem. Commun. 1989; 839–841; (c)
Linkletter B, Chin J. Angew. Chem., Int. Ed. Engl. 1995; 34: 472–
474; (d) Wall M, Linkletter B, Williams D, Lebuis A-M, Hynes
RC, Chin J. J. Am. Chem. Soc. 1999; 121: 4710–4711; (e) Scrimin
P, Tecilla P, Tonellato U. J. Org. Chem. 1991; 56: 161–166; (f)
Burstyn JN, Deal KA. Inorg. Chem. 1993; 32: 3585–3586; (g)
Hay RW, You-Quan C. Polyhedron 1995; 14: 869–872; (h) Liu S,
Luo Z, Hamilton AD. Angew. Chem., Int. Ed. Engl. 1997; 36:
2678–2680; (i) Molenveld P, Engbersen JFJ, Reinhoudt DN.
Chem. Soc. Rev. 2000; 29: 75–86; (l) Fife TH, Pujari MP. Bioorg.
Chem. 2000; 28: 357–373.
12. Laughton PM, Robertson RE. In Solute–Solvent Interactions,
Coetzee JF, Ritchie CD (eds). Marcel Dekker: New York, 1969;
399.
13. (a) Kimura E, Shiota T, Koike T, Shiro M, Kodama M. J. Am.
Chem. Soc. 1990; 112: 5805–5811; (b) Kim DH, Lee SS. Bioorg.
Med. Chem. 2000; 8: 647–652.
14. (a) Kirsch JF, Jencks WP. J. Am. Chem. Soc. 1964; 86: 837–846; (b)
Menger FM, Ladika M. J. Am. Chem. Soc. 1987; 109: 3145–3146.
15. Chin J. Acc. Chem. Res. 1991; 24: 145–152.
16. Lowry TH, Richardson KS. Mechanism and Theory in Organic
Chemistry (3rd edn). Harpen and Row: New York, 1987; 212.
¨
17. Cacciapaglia R, Mandolini L, Arnecke R, Bohmer V, Vogt W.
J. Chem. Soc., Perkin Trans. 2 1998; 419–424.
¨
6. (a) Loffler F, Luning U, Gohar G. New J. Chem. 2000; 24: 935–
938; (b) Gelbert M, Korber C, Friedrich O, Fahrenkrug F,
¨
¨
¨
¨
18. (a) Luning U, Muller M. Chem. Ber. 1990; 123: 643–645; (b)
Lu¨ning U, Mu¨ller M, Gelbert M, Peters K, von Schnering HG,
Keller M. Chem. Ber. 1994; 127: 2297–2306; (c) Lu¨ning U,
Fahrenkrug F, Hagen M. Eur. J. Org. Chem. 2001; 2161–2163;
(d) Lu¨ning U, Abbass M, Fahrenkrug F. Eur. J. Org. Chem. 2002;
3294–3303.
Keller M, Lu¨ning U. Supramol. Chem. 2002; 14:199–210.
7. Neverov AA, McDonald T, Gibson G, Brown RS. Can. J. Chem.
2001; 79: 1704–1710.
´
8. Bosch E, Bou P, Allemann H, Roses M. Anal. Chem. 1996; 68:
3651–3657.
9. Cotton FA, Wilkinson G, Murillo CA, Bochmann M. Advanced
Inorganic Chemistry (6th edn). Wiley: New York, 1999; 866.
19. Mitton CG, Gresser M, Schowen RL. J. Am. Chem. Soc. 1969; 91:
2045–2047.
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 350–355