Design, synthesis and recognition properties of urea-type anion receptors in low polar media

Article abstract of DOI:10.1002/ejoc.200700740

The binding efficiency of simple receptors bearing two NH hydrogen-bond donor groups, of general formula PhCH₂NH-Y-NHPh (Y = C=N-Ts, SO ₂, CS, CO), towards selected anion guests has been evaluated in chloroform solution in order to assess the effect of the nature of different binding sites incorporated into the receptor scaffold. Experimental results together with molecular mechanics and quantum chemical calculations have revealed that the nature of the spacer Y induces important electronic effects and conformational changes that lead to different degrees of preorganisation of the NH binding groups. In addition, the synthesis and binding properties of new lipophilic urea-based receptors, having electron-withdrawing alkylsulfonyl substituents (SO₂C₈H₁₇) and characterised by enhanced NH hydrogen-bond donor ability, is reported. Wiley-VCH Verlag GmbH & Co. KGaA, 2008.

Full text of DOI:10.1002/ejoc.200700740

FULL PAPER
DOI: 10.1002/ejoc.200700740
Design, Synthesis and Recognition Properties of Urea-Type Anion Receptors in
Low Polar Media
Luca Pescatori,[a] Arturo Arduini,[a] Andrea Pochini,[a] Franco Ugozzoli,[b] and
Andrea Secchi*[a]
Keywords: Anion recognition / Lipophilic EWG / Complexation studies / Solvent effects / Urea derivatives / Receptors
The binding efficiency of simple receptors bearing two NH
hydrogen-bond donor groups, of general formula PhCH2NH–
Y–NHPh (Y = C=N–Ts, SO2, CS, CO), towards selected anion
guests has been evaluated in chloroform solution in order to
assess the effect of the nature of different binding sites incor-
porated into the receptor scaffold. Experimental results to-
gether with molecular mechanics and quantum chemical cal-
culations have revealed that the nature of the spacer Y in-
duces important electronic effects and conformational
changes that lead to different degrees of preorganisation of
the NH binding groups. In addition, the synthesis and bind-
ing properties of new lipophilic urea-based receptors, having
electron-withdrawing alkylsulfonyl substituents (SO2C8H17
)
and characterised by enhanced NH hydrogen-bond donor
ability, is reported.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2008)
Introduction
pounds.[5] In contrast, the effect of the spacer Y on the
binding efficiency has been less investigated. The few stud-
ies reported in the literature on this topic mainly focused
on the different binding efficiency of thiourea versus urea
derivatives. These studies were usually carried out for solu-
bility reasons in very polar and competing solvents such as
acetonitrile or DMSO[3a,3c,4b] in which the better complex-
ing properties experienced by the thiourea-based receptors
were generally ascribed to the higher acidity of their NH.[6]
Compounds characterised by the same structural motif
have also been extensively used as non-covalent catalysts in
reactions operating under general acid catalysis condi-
tions.[7] Indeed, these metal-free catalytic systems are par-
ticularly interesting because they mediate the activation of
electrophilic substrates through hydrogen bonding and usu-
ally entail chemoselectivity, higher tolerance to functional
groups and fewer environmental problems than traditional
metal catalysts.
The design of new compounds able to operate with im-
proved properties in both contexts should be based on the
comprehension of the several factors that affect hydrogen
bonding in these systems. It could thus be important to
systematically study the effects induced by the nature of the
spacer Y in a series of strictly related hosts that are soluble
in low polar and less competing solvents.
Herein we report on the recognition properties towards
anions in chloroform solution of a series of acyclic recep-
tors characterised by a different spacer Y and having tosyl-
guanidine (1, Y = C=N–Ts), sulfamide (2, Y = SO2), thio-
urea (3, Y = C=S) and urea (4, Y = C=O) binding units,
Over the past decade, compounds of general formula
RNH–Y–NHR, where R is an alkyl or aryl group, have
attracted considerable attention as neutral receptors for the
recognition of anions.[1] These compounds are charac-
terized by the presence of two NH groups able to donate
two convergent hydrogen bonds to an acceptor species. This
structural motif is found in several compounds charac-
terized by a different spacer Y, such as isophthaldiamide,
pyridine-2,6-dicarboxamide, squaramide and (thio)urea de-
rivatives. In particular amide-based hosts[2] and (thio)urea
derivatives[3] have been extensively employed either as sim-
ple monotopic acyclic receptors or arranged in more com-
plex polytopic hosts for multipoint recognition.[4]
In principle, the binding efficiency of these receptors can
be improved by enhancing the hydrogen-bond donor ability
of their NH groups. Starting from the seminal work of Wil-
cox et al.[3b] who reported the anion-binding properties of
a series of substituted N-aryl-NЈ-alkyl(thio)ureas, most of
the work has usually been directed towards evaluating the
electronic effects of substituents present on the R groups,
when R = aryl, in a series of structurally related com-
[a] Dipartimento di Chimica Organica e Industriale, Università
degli Studi di Parma,
Via G. P. Usberti 17/a, 43100, Parma, Italy
Fax: +39-0521-905472
E-mail: andrea.secchi@unipr.it
[b] Dipartimento di Chimica Generale e Inorganica Chimica
Analitica Chimica Fisica, Università di Parma,
Parco Area delle Scienze 17/A, 43100 Parma, Italy
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Eur. J. Org. Chem. 2008, 109–120
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
109

L. Pescatori, A. Arduini, A. Pochini, F. Ugozzoli, A. Secchi
FULL PAPER
respectively. The synthesis and binding properties of new Binding Studies
and lipophilic urea-based receptors bearing electron-with-
drawing SO2R (R = C8H17) substituents are also reported.
To determine the effect of the spacer Y on the binding
properties of 1–4 in chloroform solution, chloride (Cl–) and
acetate (Ac–) were selected as representative spherical and
planar anions, respectively. The tetrabutylammonium
(TBA) cation was chosen as the counterion for solubility
reasons. However, it is known that TBA salts are present in
solutions of low polar solvents either as solvated ion pairs
or as their aggregates.[9] Moreover, the hydrogen-bond do-
nor and acceptor nature of the –NH–Y–NH– binding unit
of 1–4 might favour extensive host self-association, as found
for instance in derivatives in which both the NH groups are
substituted with phenyl units.[10] Both phenomena could in
principle affect the host’s binding efficiency. Therefore the
complexation of organic salts in solvents of low dielectric
constant requires an initial careful examination of the bind-
ing stoichiometry.
The extent of the host’s self-association was initially
evaluated through 1H NMR dilution experiments. Upon di-
lution of a 2ϫ10–2  solution of each host up to 6ϫ10–4 ,
only 4 showed a not-negligible variation in the chemical
shift of its ureido NH protons (∆δ = +0.3 ppm). However,
attempts to obtain a reliable self-association constant
through the fitting of the NMR dilution data either with a
dimerization isotherm[11] or with an isodesmic model[12] did
not give satisfactory results.
Results and Discussion
Design and Synthesis of Urea-Type Receptors
The design of the acyclic receptors 1–4 was based on the
requirement that the substituents attached to the NH
groups should confer an overall lipophilicity to the resulting
compounds to allow the evaluation of their binding proper-
ties in low polar solvents. With this aim all the receptors
used in the present study are characterised by a central ure-
ido-type group bearing a phenyl unit on one side and a
benzyl unit on the other. The higher conformational flexi-
bility of the latter substituent appreciably enhances the sol-
ubility of these compounds with respect to N,N-diaryl de-
rivatives.
The sulfonylguanidine-based receptor 1 was synthesised
in 30% yield according to the procedure reported by Zhang
and Shi[8] for the preparation of sulfonylguanidinium ana-
logues (see Scheme 1), whereas compounds 2–4 were pre-
pared according to published procedures (see Exp. Sect.).
The stoichiometry of binding was thus verified by 1H
NMR spectroscopy in CDCl3 solution through the use of
continuous variation methods (see Exp. Sect.).[13] The Job
plot obtained during the complexation of 4 with TBACl is
shown in Figure 1. From the plotted data it emerges that,
in the concentration range used in the experiment,
the maximum of the complex formation occurs at x =
[4]0/([4]0+[TBACl]0) = 0.5, which corresponds to a 1:1 stoi-
chiometry of binding.
1
In the H NMR spectrum of 1, taken in CDCl3, the two
non-equivalent NH protons resonate as two broad signals
at δ = 5.1 and 9.1 ppm, respectively (see Supporting Infor-
mation). The large downfield shift of the signal assigned to
the NH proton adjacent to the phenyl unit seems to indicate
that this group is probably involved in intramolecular hy-
drogen bonding with one of the two oxygens of the SO2
moiety.
Figure 1. Job plot obtained by monitoring the chemical shift varia-
tion of the NH proton adjacent to the benzyl group of 4 during its
binding with TBACl. During the experiment the relative concentra-
tion of the two interacting species was varied continuously, but
Scheme 1. Synthesis of the (tolylsulfonyl)guanidine-based receptor 1. their sum was kept constant (1.45ϫ10–3 ).
110
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 109–120

Urea-Type Anion Receptors in Low Polar Media
To determine the binding efficiency of the hosts, 1H
NMR titration experiments were carried out in CDCl3 solu-
tion using methods already published.[14] By adding increas-
ing amounts of the guest solution (1.0–1.3ϫ10–1 ) to a
solution (1.0–1.3ϫ10–2 ) of the host, all receptors but 1
showed an extensive downfield complexation-induced shift
(CIS) of the two chemically different NH protons. As an
example, in Figure 2 has been depicted a stack plot corre-
sponding to the titration of receptor 4 with TBAAc. When
the anion guest is present in solution in a large excess (Fig-
ure 2, a), the two NH protons are downfield shifted to 10.95
and 9.40 ppm, respectively. These findings suggest that
anion recognition occurs mainly through the formation of
hydrogen bonds with the NH protons of the hosts.
Figure 3. Binding isotherms obtained by monitoring the chemical
shift variation of the PhNH–Y–NHCH2Ph proton during the ti-
tration of 3 (o, continuous line; c = 1.2ϫ10–2 ) and 4 (᭹, dashed
line, c = 1.3ϫ10–2 ) with TBAAc in CDCl3, respectively.
The binding data summarised in Table 1 show that the
sulfonylguanidine derivative 1 experiences very weak com-
plexation with all guests, as shown by the negligible down-
field shift of its NHs. In contrast, hosts 2–4 show a rela-
tively strong binding, especially with acetate. The general
trend in the binding efficiency follows the order 4 Ͼ 3 ≈ 2
ϾϾ 1.
A better understanding of the complexing properties of
1–4 can be achieved by considering that, in principle, the
nature of the spacer Y could affect both the hydrogen-bond
donor ability and the preorganisation of the hosts. This lat-
Figure 2. 1H NMR stack plot (300 MHz, T = 300 K, expanded
region) showing the chemical shift variation of the NH protons of
4 upon complexation with TBAAc in CDCl3. Guest/host ratios: a)
4:1; b) 1:1 and c) free 4 (c = 1.0ϫ10–2 ), respectively.
In all the titration experiments the NMR spectra showed ter aspect is particularly important because it is reasonable
time-averaged signals for the free and complexed species. to assume that, due to restricted rotation around the two
The apparent binding constants (K) for host–guest complex pseudoamide NH–Y bonds, these hosts have a different de-
formation were calculated by considering a 1:1 stoichiome- gree of preorganisation. The conformational behaviour of
try using methods previously described based on the non- 1–4 was initially investigated through the analysis of several
linear fitting of the chemical shift variation of the NH pro- solid-state structures of compounds having the general for-
tons (see Exp. Sect.).[15] Best fits of the experimental data mula PhNH–Y–NHR(alkyl) which were retrieved from the
were obtained by using the chemical shift variation of the Cambridge Structural Database (see Exp. Sect.). The orien-
1
NH proton adjacent to the phenyl group. The H NMR tations of the N–H bonds with respect to the spacer Y were
binding isotherms relative to the titrations of TBAAc with evaluated by considering the two dihedral angles H–N–C–
hosts 3 and 4 have been depicted in Figure 3.
O(S,N) along the two pseudoamide bonds.
Table 1. Apparent binding constants (K) for the complexation of hosts of general formula PhNH–Y–NHCH2Ph (1–4) with tetrabutylam-
monium salts and DMSO in CDCl3 solution.[a]
Host[b]
Y
TBAAc
δϱ [ppm] ∆δϱ [ppm]
TBACl
δϱ [ppm] ∆δϱ [ppm] K [–1]
DMSO
δϱ [ppm] ∆δϱ [ppm]
K [–1]
K [–1]
[c]
[d]
[c]
[d]
1
2
3
4
C=N–Ts (1.5Ϯ0.5)ϫ101
9.4Ϯ0.3
0.3
–
4.7
4.7
–
–
(2.2Ϯ0.2)ϫ102
(1.0Ϯ0.1)ϫ102
(1.5Ϯ0.5)ϫ103
9.5Ϯ0.1
11.0Ϯ0.1
9.7Ϯ0.1
3.2
3.3
3.4
45Ϯ8
12Ϯ5
52Ϯ10
7.0Ϯ0.1
8.6Ϯ0.1
7.1Ϯ0.1
0.7
0.6
0.7
[e]
SO2
–
C=S
(5.0Ϯ0.3)ϫ102
(3.0Ϯ0.5)ϫ103
12.4Ϯ0.1
11.0Ϯ0.1
C=O
[a] Determined by 1H NMR in CDCl3 (T = 300 K) by monitoring NH ligand chemical shift variation upon complexation; standard
1
deviations are given in parentheses. [b] C6H5NH–Y–NHCH2C6H5: H NMR signals for the free receptors (δ, ppm): 1 = 9.07, 2 = 6.30,
3 = 7.67, 4 = 6.40. [c] Negligible complexation. [d] No significant chemical shift variation. [e] Extensive broadening of the NH signals
prevented binding constant calculation.
Eur. J. Org. Chem. 2008, 109–120
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
111

L. Pescatori, A. Arduini, A. Pochini, F. Ugozzoli, A. Secchi
FULL PAPER
All the N-phenyl-NЈ-alkylurea derivatives investigated
have their two pseudoamide bonds in the Z,Z form. The
two N–H bonds are thus in an anti conformation with re-
spect to the C=O bond. Interestingly, the most common
geometry found for the corresponding thiourea derivatives
is E,Z. In particular, the N–H bonds adjacent to the phenyl
and alkyl units, respectively, are syn and anti with respect to
the C=S bond. The anti,anti or Z,Z conformation becomes
dominant in the solid state exclusively when both NHs are
involved in intermolecular H-bonding. Unfortunately, no
relevant crystal data were found in the CSD for compounds
with Y = SO2 or C=N–Ts. However, several structures cor-
responding to N,N-dialkylsulfamide derivatives show that
these compounds usually adopt a geometry in which the
two N–H bonds are oriented in opposite directions with
respect to the plane containing the N–S–N atoms. In most
cases this conformation is stabilised through the formation
of intermolecular hydrogen bonding between the NHs and
the oxygen of the SO2 group belonging to different adjacent
molecules.
Further insights into the conformational behaviour of 1–
4 have been gained through computational studies. The
conformational mobility of these compounds is greatly af-
fected by the presence of the benzylic unit. Indeed, the rota-
Figure 4. Selected conformations of compounds 1–4 generated
through a Monte Carlo conformational search (see text for details).
tional degree of freedom ranges from five for 2–4 to eight guest. This geometry could also explain the large downfield
1
for 1. A systematic search for all the possible rotamers was shift of the NH proton found in the H NMR spectrum
carried out by a Monte Carlo method by using the recorded in CDCl3.
MMFF94 force field. Among the several local minima ob-
The Monte Carlo search carried out on 2 yielded, as for
tained, those derived from the different orientation of the 1, several local minima. The three most stable minima
phenyl and benzyl units with respect to the spacer Y were found adopt the gauche,gauche (2-g,g), anti,gauche (2-a,g)
selected and further minimised either by molecular mechan- and syn,syn (2-s,s) geometries. The second arrangement (see
ics or by quantum chemical calculations using DFT meth- Figure 4, e) is similar to that usually found in the solid-state
ods (see Exp. Sect.). Such an approach yielded rotamers structures, whereas in the third both N–H bonds are almost
whose structures and relative energies have been summa- parallel and oriented on the same side of the molecule (Fig-
rised in Figure 4 and in Table 2, respectively.
ure 4, f). These minima all have very similar energies but,
Several local minima with very similar energies were albeit in the gas phase, the 2-g,g rotamer seems to be the
found for 1, although not one corresponds to the Z,Z ge- least strained one (see Table 2).
ometry. A common motif of these rotamers is the presence
The conformational search performed on compounds 3
of an intramolecular hydrogen bond between one of the and 4 was carried out by using an MMFF94 force field
NH groups and one of the oxygens of the tosyl moiety. In modified with new torsional parameters developed by Hay
the less strained conformer, such an intramolecular interac- and co-workers for (thio)urea derivatives (MMFF94+).[16]
tion forces 1 to adopt an E,Z geometry (see Figure 4, a). In This approach yielded the E,Z (Figure 4, h) and the Z,Z
this arrangement only the less acidic NH adjacent to the (Figure 4, m) conformers as the most stable rotamers in the
benzyl unit is potentially able to donate a H-bond to a gas phase for 3 and 4, respectively (see Table 2).[17] The Z,Z,
112
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 109–120

Urea-Type Anion Receptors in Low Polar Media
Table 2. Computed relative energies for selected conformers of 1–4.
pattern, characterised by several minima. However, by
using the simpler PhNHSO2NHMe molecule as a model it
was possible to estimate an interconversion energy of
around 9 kcalmol–1. For 3 and 4 both the (Ph)C–N–C–O(S)
(see Supporting Information) and the (Bn)C–N–C–O(S) di-
hedrals were varied from –180 to 180°, affording rotational
Conformer
∆E [kcalmol–1]
Dihedral angle [°]
(MMFF94)[a,b]
H–NPh–C–N
H–NBn–C–N
1-E,Z
1-E,E
1-Z,E
1-Z,Z
0
0.3
58
–52.1
98
151.3
–17
–4
barriers that are generally higher for
3
(12.5–
1.9
13.4 kcalmol–1, depending on the path chosen) than for 4
(11–11.3 kcalmol–1, depending on the path chosen). These
results are consistent with those determined at a higher
level of theory (MP2/aug-ccpVDZ) for N-phenyl-NЈ-alkyl-
thiourea (9.8 kcalmol–1) and N-phenyl-NЈ-alkylurea
(9.1 kcalmol–1) derivatives.[16] The higher interconversion
energy for thiourea derivatives relative to the urea ones has
mainly been ascribed to an increase in the partial double
bond character of the corresponding C–N bond.[19]
[c]
–
–
H–NPh–S–O
H–NBn–S–O
2-g,g
2-a,g
2-s,s
0
1.4
3.1
76.9
–176.8
–33.7
–46.5
44.5
27.7
H–NPh–C–S
H–NBn–C–S
3-E,Z
3-E,E
3-Z,E
3-Z,Z
0
3.8
9.1
164.6
–171.9
172.2
4.7
–1.1
–177
5.7
4.5
3.1
The tendency of each host to interact with the anion spe-
cies should also be correlated to the electronic effects ex-
erted on the NH groups by substituents incorporated into
the host scaffold. As previously mentioned, these effects
have been extensively discussed in the literature (see Intro-
duction) when the substituents are located on the aromatic
unit directly linked to the hydrogen-bond donor group. By
H–NPh–C–O
H–NBn–C–O
4-E,Z
4-E,E
4-Z,E
4-Z,Z
2.8
–13
–
166.8
180
158.2
–
–1.3
–180
[c]
3.3
0
[a] Local minima were obtained through a Monte Carlo conforma-
tional search by using the MMFF94 force field. [b] For 3 and 4 using a similar approach, the electronic effects of the spacer
different torsional parameters were used (see ref.[16]). [c] Local
minimum not found.
Y were initially extrapolated by considering either the Ham-
mett substituent constants or the corresponding resonance
and field parameters of substituents structurally related to
Z,E and E,Z conformers obtained from the previous molec- Y such as CONH2, CSNH2 and SO2NH2. Unfortunately,
ular mechanics studies on 3 and 4 were further minimised no data were found in the literature corresponding to the
by quantum chemical calculations using DFT methods at electronic effects of substituents similar to Y = C=N–Ts.
the B3LYP/6-31+G* level both in the gas phase and by ap- Regardless of the parameter chosen, it appears that the elec-
plying the polarisable solvent continuum model (PCM)[18] tron-withdrawing strength follows the order SO2NH2 ϾϾ
to simulate the effect of the solvent molecules (CHCl3). CONH2 Ͼ CSNH2.[20] This trend was supported by the
These preliminary and more accurate calculations did not analysis of the molecular electrostatic potential (MEP) cal-
substantially change the energy order found in the molecu- culated at the B3LYP/6-31+G* level for the structures of 2–
lar mechanics studies, although the single-point calculation 4 in the Z,Z conformation (syn,syn for 2). MEPs can be
carried out using the PCM model revealed that for 4 the considered as reliable descriptors of intermolecular interac-
E,Z conformer is more stable than the Z,Z by around tions that have an important electrostatic component such
1.4 kcalmol–1.
as hydrogen bonding.[21] The MEP plotted either onto
The equilibrium structures of the complexes of 1–4 with isodensity (0.002 eau–13) or van der Waals surfaces (see
the Cl– anion were calculated in the gas phase by a molecu- Supporting Information) indeed showed evidence that the
lar mechanics method using the MMFF94 force field. The highest positive electrostatic potential calculated at the
outcome of these simulations showed that the complex be- point at which an NH direction crosses the molecular
tween 1 and Cl– is created only through the formation of surface is found for 2 (+39 kcalmol–1), followed by 4
one hydrogen bond with the host in the E,Z conformation (+38 kcalmol–1) and then 3 (+29 kcalmol–1).
(see Figure 4, c) whereas for 2–4 the complexation is gen-
erally assisted by the formation of two hydrogen bonds with titration experiments, the findings of the molecular mechan-
the hosts in the Z,Z conformation (syn,syn for 2). ics and quantum chemical calculations can be used with
Considering the low polarity of the media used in the
Considering the relative stability of the free rotamers good reliability to rationalise the binding properties of the
found for each host in the previous Monte Carlo search, it hosts. The poor complexing properties evidenced by 1 and
thus appears that, except for 4, the binding of the anion 2 have mainly been ascribed to their large conformational
needs, as a prerequisite, a significant conformational re- flexibility. The entropic loss that these hosts must endure to
arrangement of the hosts which implies rotation of the NH adopt an appropriate geometrical arrangement to interact
groups around the corresponding pseudoamide bonds. with the charged species is probably not fully compensated
Such rotations are strongly dependent on the corresponding by the consequent enthalpic gain derived by the hydrogen-
interconversion energies which were determined for 2–4 by bonding interactions. This compensation is particularly dis-
using the MMFF94+ force field. For 2 variation of the favoured in the case of 1 since this host cannot coopera-
(Ph)C–N–S–O dihedral afforded a very complicated energy tively use its NH groups to bind charged guests. The impor-
Eur. J. Org. Chem. 2008, 109–120
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
113

L. Pescatori, A. Arduini, A. Pochini, F. Ugozzoli, A. Secchi
FULL PAPER
tance of the role played by the host preorganisation is also of its binding properties through NMR spectroscopic ti-
evidenced by comparison of the results obtained for 2–4. trations.
Indeed, the trend in the electron-withdrawing effect exerted
A more promising alternative to the aforementioned ap-
by the spacer Y, as deduced both by the MEP analysis and proach is based on the use of EWGs that are themselves
resonance field parameters, can only partly account for the quite lipophilic; the alkylsulfonyl group (RSO2-) possesses
better recognition properties of 4 compared with 3.
both features. In fact its electron-withdrawing strength is
The experimental results obtained in CDCl3 for thiourea comparable to that of the nitro group[23] and the solubility
3 and urea 4 deserve further comment. These findings con- of the corresponding host can be modulated by varying the
trast with those reported in the studies carried out, for solu- nature of the R group.
bility reasons, in more polar solvents such as DMSO or
The introduction of alkylsulfonyl groups onto the phenyl
CH3CN in which the better properties of the thiourea deriv- unit of 4 cannot, however, be carried out by direct synthetic
atives have been explained on the basis of the higher acid- methods. Therefore the new 4-(octylsulfonyl)aniline (9) was
ity[22] of their NH groups (see Introduction). To verify our prepared in 32% overall yield following the synthetic path-
results, two series of experiments were designed. Initially, way described in Scheme 2. The new lipophilic urea-based
the affinity of 1–4 for DMSO was evaluated in CDCl3 solu- receptor 11 (X = SO2C8H17) was synthesised in 38% yield
tion by means of 1H NMR titrations. Analysis of the calcu- by reaction of amine 9 with benzyl isocyanate in CH2Cl2.
lated binding constants (see Table 1) shows a trend in the Compounds 12 (X = H) and 13 (X = NO2) were hence
host-binding efficiency similar to that found for the com- synthesised in 40 and 70% yields, respectively, by reaction
plexation of the anion species. The titration experiments of the corresponding isocyanates with 4-(octyloxy)benzyl-
with TBACl and TBAAc were then repeated in [D6]DMSO amine (10) (see Scheme 3). These new hosts, which experi-
solution. In this very polar solvent, 1 and 2 experienced ence good solubility in chloroform solution due to the pres-
negligible complexation, whereas 3 and 4 showed a similar ence of an octyloxy chain on their benzyl units, were pre-
binding efficiency (see Table 3). Considering that 4 is a bet- pared for comparison with 11 in order to assess the effect
ter host than 3 for DMSO in CDCl3, comparison of the of the electron-withdrawing nature of the SO2R group on
binding constants calculated for urea 4 and thiourea 3 with the binding properties.
acetate and chloride in CDCl3 and [D6]DMSO, respectively,
indicates how the data obtained in the latter solvent are
the result of competition between DMSO molecules, which
through their S=O groups can behave as hydrogen-bond ac-
ceptor species, and the anions for the coordination site of
the hosts.
Table 3. Binding constants (K) for the complexation of 3 and 4 with
tetrabutylammonium acetate and chloride in DMSO solution.[a]
Host[b]
TBAAc
TBACl
K [–1]
δϱ [ppm] ∆δϱ [ppm] K [–1] δϱ [ppm] ∆δϱ [ppm]
3
4
570Ϯ120 12.8Ϯ0.1
360Ϯ40 11.5Ϯ0.1
3.2
3.0
24Ϯ6 11.6Ϯ0.1
26Ϯ10 10.0Ϯ0.1
2.0
1.5
1
[a] Determined by H NMR in [D6]DMSO (T = 300 K) by moni-
toring NH ligand chemical shift variation upon complexation;
standard deviations are given in parentheses. [b] C6H5NH–Y–
NHCH2C6H5 signal of the free receptors (δ, ppm): 3: 9.60, 4: 8.52.
Scheme 2. Synthesis of the lipophilic amino derivative 9.
Design and Synthesis of Lipophilic Urea-Based Receptors
From the binding studies it emerges that the develop-
ment of new and more efficient acyclic synthetic anion re-
ceptors able to operate in low polar solvents depends on
two related requisites that need to be addressed in the de-
sign of the host: the host solubility and the hydrogen-bond
donor ability of the binding unit. It can, however, be fore-
seen that an improvement of the latter property in 4,
through the incorporation of strong electron-withdrawing
groups (EWG) onto the host aromatic scaffold, negatively
affects the overall solubility of this compound in solvents
of low dielectric constant. Indeed, the low solubility experi-
enced in CDCl3 solution by 1-benzyl-3-(4-nitrophenyl)urea
(5), in which the para position of the phenyl unit of 4 is
substituted with a nitro group, prevented the evaluation Scheme 3. Synthesis of the new urea derivatives 11–13.
114 Eur. J. Org. Chem. 2008, 109–120
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Urea-Type Anion Receptors in Low Polar Media
Binding Studies
The first attempts to evaluate the binding properties of
11–13 towards TBAAc and TBACl in CDCl3 solution by
1
using H NMR titrations were successful only for host 12
(see Table 4). In contrast, very steep isotherms were always
obtained for binding when 11 and 13 were used as hosts
regardless of the concentration of the interacting species
used in the titrations. The non-linear fitting of the experi-
mental data with a 1:1 binding isotherm was in fact very
poor since the conditions of the Weber parameter p (see
Exp. Sect.) were never entirely satisfied during the experi-
ments[15] due to the high efficiency of binding experienced
by these receptors.
Scheme 4. Synthesis of the new lipophilic urea derivatives 14 and
15.
The recognition behaviour of 11–13 was thus evaluated
by UV/Vis spectroscopic titrations. A solution of the host titrations were carried out by using the same experimental
(2–5ϫ10–5 ) in CHCl3 was titrated with a solution of the conditions as employed for the binding studies of 11–13.
TBA salt (2–5ϫ10–4 ) in CHCl3. Upon complexation all During the titrations a bathochromic shift of the receptors’
hosts except for 12 showed a bathochromic shift of the main main absorbing band was always observed upon addition
absorbing band with the formation of an isosbestic point. of aliquots of solutions of the salts with the formation of
All the spectroscopic titration curves were fitted with the an isosbestic point. It should, however, be pointed out that
Specfit/32 software.[24] A non-linear least-squares treatment the addition of an excess of the guest solution usually in-
of the spectroscopic data applied to a 1:1 stoichiometry duced a supplementary redshift of the maximum of the
complexation model gave the binding constants reported in band corresponding to the complexed species. This effect
Table 4.
was negligible for 14, whereas it was very noticeable during
The marked effect of the substituent X on the binding the titration of 15, especially with TBAAc and TBAF. In
efficiency is well evidenced by comparison of the binding particular, the spectra reported in Figure 5 (a) shows that
constants calculated for 11 (X = SO2C8H17) and 13 (X = the addition of 1 equivalent of the acetate solution deter-
NO2) with that calculated for 12 (X = H). The introduction mines a redshift of the main absorbing band of 15 from λ
of electron-withdrawing substituents leads to stronger bind- = 282 to 298 nm with the formation of a definite isosbestic
ing with both anions. Although this effect was expected for point (λ = 286 nm). The addition of an excess of the acetate
13 on the basis of previously reported results,[25] the binding solution causes a further drift of the maximum to λ =
efficiency experienced by 11 confirms the hypothesis that 302 nm.
the electron-withdrawing character of the SO2C8H17 group
can be successfully exploited for the preparation of more titration of 15 could be ascribed to several phenomena such
lipophilic urea-based receptors. as dilution effects, variation of the solution ionic strength
In principle, this additional redshift observed during the
The good solubility properties of 11 prompted us to ex- and the formation of adducts with different stoichiometries.
plore the possibility of employing the amine 9 in the prepa- Considering that this final effect could substantially affect
ration of new lipophilic diphenylurea derivatives. Com- the correct determination of binding constants, the stoichi-
pound 9 was converted by reaction with phenyl isocyanate ometry of the binding was verified by the application of
and triphosgene into the unsymmetrical (14) and symmetri- continuous variation methods. These experiments evidenced
cal (15) urea derivatives in 60 and 50% yields, respectively a change in the maxima of the Job plots as a function of
(see Scheme 4).
the wavelength used for its determination. This behaviour
The binding properties of the new diphenylurea deriva- could be ascribed either to a change in the binding stoichi-
tives 14 and 15 towards acetate, chloride and fluoride, as ometry during the titration or to self-association phenom-
TBA salts, were analysed in chloroform solution. UV/Vis ena.[26] The self-association of both hosts was thus evalu-
Table 4. Absorption spectral parameters and binding constants (logK) for the complexation of 11–13 with TBA acetate and chloride in
CHCl3.[a,b]
Host λH(max.) [nm]
εH [mol–1Lcm–1
]
TBAAc
TBACl
[c]
[c]
λHG(max.) [nm]
εHG [mol–1Lcm–1
]
logK11
λHG(max.) [nm]
εHG [mol–1Lcm–1
]
logK11
11
12[d]
13
259
277
325
2.4ϫ105
2.1ϫ103
1.1ϫ104
280
277
365
2.3ϫ105
4.2ϫ103
1.7ϫ104
4.73(4)
3.46(2)
4.88(3)
270
277
353
2.6ϫ105
4.8ϫ103
1.4ϫ104
4.4(1)
2.91(5)
4.22(2)
[a] Determined by UV/Vis spectroscopy in CHCl3 (T = 297 K). [b] λH(max.) and λHG(max.) represent the wavelength of the main
absorbing band of the free host and of the complex, respectively. [c] The uncertainty in the last figure is given in parentheses. [d] The
binding constants for host 12 were also calculated by 1H NMR spectroscopic measurements in CDCl3 (T = 300 K), affording the following
results: logK(TBAAc) = 3.5(6) and logK(TBACl) = 2.8(5).
Eur. J. Org. Chem. 2008, 109–120
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
115

L. Pescatori, A. Arduini, A. Pochini, F. Ugozzoli, A. Secchi
FULL PAPER
spectroscopic data by considering the host dimerization
process along with 1:1 complexation did not give satisfac-
tory results. The spectroscopic data collected during the ti-
tration experiments of both receptors were thus analysed by
using the evolving factor analysis (EFA)[27] tool im-
plemented in Specfit/32.[24]
The factor analysis applied to the titrations of 14 always
revealed the presence of only two chromophoric species in
solution, that is, the free receptor and the corresponding
1:1 adduct. For this receptor the spectroscopic data were
thus fitted by applying a pure 1:1 binding model. The calcu-
lated binding constants (logK11) were all characterised by
satisfactory experimental errors (see Table 5). The EFA for
15 revealed a possible deviation from a pure 1:1 complex-
ation model. With both TBACl and TBAAc more than two
absorbing species might be simultaneously present in solu-
tion. The spectroscopic data were thus treated by using dif-
ferent binding models, as described by Equations (1), (2)
and (3), where RH is the urea receptor and A– the anion
guest.
RH + A– h [RH···A]–
(1)
(2)
(3)
[RH···A]– + RH h [RH···A···RH]–
[RH···A]– + A– h R– + [HA2]–
With TBACl comparable results were obtained by fitting
the data either with a simple 1:1 binding model (logK11
= 5.9Ϯ0.1) or by considering two simultaneous equilibria,
Equation (1) and Equation (2), with the formation of a 2:1
host–guest adduct along with the 1:1 adduct (see Table 5).
The 2:1 adduct is probably present in solution at the begin-
ning of the titration when 15 is present in large excess with
respect to Cl–. Such an adduct then dissociates to yield the
1:1 complex as the concentration of the salt increases in
solution due to the large binding constant governing the
first equilibrium (see Supporting Information for the distri-
bution diagram).
In the case of TBAAc, application of the 1:1 binding
model afforded a logK11 = 5.34Ϯ0.08 with a satisfactory
experimental error. However, bearing in mind the previous
EFA results, good fittings of the spectroscopic data were
also obtained that indicated the possible formation of,
along with the 1:1 complex, an adduct with a 1:2 host–guest
stoichiometry. This adduct started to appear in solution
when more than 1 equivalent of TBAAc was added (see
Figure 5, b). Application of this binding model afforded a
logK11 = 6.7Ϯ0.1 and a logK12 = 6.0Ϯ0.1 and the maxima
Figure 5. a) UV/Vis spectra for the titration of a solution of 15
[1.9ϫ10–5 ] with a solution of TBAAc [2.0ϫ10–4 ] in CHCl3; b)
distribution diagram for the chromophoric species present in solu-
tion during the titration of 15 with TBAAc; c) UV/Vis spectra for
the titration of 15 (1.9ϫ10–5 ) with TBAF (2.0ϫ10–4 ) in
DMSO.
1
ated by H NMR dilution experiments. The chemical shift of the bands corresponding to the 1:1 and 1:2 adducts were
variation of the NH singlet was monitored in CDCl3 upon found at λ = 298 (ε = 49200 –1 cm–1) and 302 nm (ε =
dilution of the host solution from 5ϫ10–2 to 9.8ϫ10–4 . 53700 –1 cm–1), respectively.
In this concentration range the variation of the NH chemi-
The nature of the possible 1:2 adduct formed between 15
cal shift of 15 was ∆δ = 0.35 ppm. Application of the iso- and acetate in chloroform solution is, however, difficult to
desmic model[12] gave reasonable results by considering the rationalise. Fabbrizzi and co-workers recently showed that
formation of dimers to have a self-association constant of in the more polar acetonitrile urea-based receptors, charac-
180Ϯ30 –1. Although not negligible, the self-association terised by very acidic NH groups, can give rise either to
of 15 cannot fully account for the deviation recorded in the oxoanions or fluoride complexes having a 1:2 host–guest
UV titration experiments. In fact, any attempt to fit the stoichiometry.[3f–3i,28] The formation of these unusual ad-
116
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 109–120

Urea-Type Anion Receptors in Low Polar Media
Table 5. Absorption spectral parameters and binding constants for the complexation of receptors 14 and 15 with tetrabutylammonium
acetate, chloride and fluoride, as calculated by UV/Vis titration experiments in CHCl3 and DMSO (T = 297 K).
Guest
Solvent
14
15
λH(max.) [nm][a] λHG(max.) [nm][a]
logK11
λH(max.) [nm][a] λHG(max.) [nm][a] logK11
logK12
6.0(1)
logK21
[b]
[b]
[b]
[b]
TBAAc CHCl3
DMSO
TBAF
266
282
266
283
267
293
292
286
4.63(9)
3.77(3)
5.10(3)
2.9(1)
282
292
283
292
284
298
303
297
6.7(1)[d]
3.22(5)
6.4(4)
CHCl3
DMSO
295; 347[c]
287
303; 356[c]
296
3.2(1)
TBACl CHCl3
4.74(3)
6.02(8)[d]
4.6(1)
[a] λH(max.) and λHG(max.) indicate the wavelength of the main absorbing band of the free host and of the complex, respectively. [b]
The uncertainty of the last figure is given in parentheses. [c] Wavelength of the new absorbing band that develops after the addition of
more than 1 equivalent of the salt. [d] The binding constants (logK) calculated assuming only the formation of the 1:1 host–guest adduct
were 5.9(1) and 5.34(8) for TBAAc and TBACl, respectively.
ducts was explained on the basis of the two-step equilibria roform solution towards the anion guests chosen as repre-
described by Equation (1) and Equation (3). In the second sentative of spherical and planar anions. The complexing
step the hydrogen-bonded complex undergoes proton ex- behaviour of hosts 1–4 has been explained also with the aid
change with a second anion leaving a deprotonated species of molecular mechanics and DFT studies. These simula-
that is responsible for the development of a new band, usu- tions showed that the urea 4 is the most preorganised
ally at a longer wavelength, in the corresponding UV spec- among the hosts examined. The better complexing proper-
trum. The occurrence of the second equilibrium is related ties of thiourea 3 compared with urea 4 in more polar sol-
to several factors such as a) the acidity of the urea deriva- vents such as DMSO and acetonitrile are mostly a result of
tive, b) the basicity of the anion, c) the stabilisation of the the competing effect of the solvent molecules for the bind-
charged host through delocalisation and d) the stability of ing site inserted in 4.
the [HA2]– species.[3h] It is also reasonable to foresee that
Moving on from these results, new urea-based receptors
the polarity of the media can sensibly affect the stability of 14 and 15, characterised by lipophilic electron-withdrawing
the products of the second equilibria. Indeed, over the groups on their phenyl units, have been synthesised. The
course of the titrations of 14 and 15 with both acetate and interesting binding results obtained with these hosts indi-
fluoride in chloroform solution there was no evidence of cate that alkylsulfonyl substituents are useful groups for en-
host deprotonation. In contrast, when the titration experi- hancing the hydrogen-bond donor ability of these com-
ments were carried out in the more polar DMSO, in spite pounds without depressing their solubility properties in or-
of a general decrease in the binding efficiency (see Table 5), ganic solvents, as found with nitro substituents. These find-
weak absorption bands start to appear in the UV spectra ings lead to the possibility of employing such derivatives as
after the addition of more than 1 equivalent of TBAF (see non-covalent catalysts in reactions operating in low polar
Figure 5, c). The occurrence of these new bands exclusively solvents under general acid catalysis conditions.
in the titration experiments with TBAF and not with
TBAAc is consistent with the findings of Fabbrizzi and co-
workers.[3h] The acetate anion can in fact form stable bifur-
Experimental Section
cated 1:1 hydrogen-bonded complexes with 14 and 15 and
All reactions were carried out under nitrogen. All solvents were
the host deprotonation becomes less favoured than with
freshly distilled under nitrogen and stored over molecular sieves for
fluoride.
at least 3 h prior to use. Column chromatography was performed
These findings can also be useful in part to explain the
on silica gel 63–200 mesh. NMR spectra were recorded in CDCl3
redshift of the 1:1 complex’s main absorbing band observed
in chloroform when more than 1 equivalent of TBAAc was
added to a solution of 15. This shift can be ascribed to an
incipient host–guest proton exchange induced by a neigh-
bouring acetate anion that changes the overall polarity of
the 1:1 adduct.
unless otherwise indicated. Melting points are uncorrected. Com-
pounds 2,[29] 3,[30] 4,[31] 5,[32] and 10[33] were synthesised according
to reported procedures. All other reagents were of reagent grade
quality as obtained from commercial suppliers and were used with-
out further purification.
N-(1-Benzylamino-1-phenylaminomethylidene)-4-methylbenzenesul-
fonamide (1): NaH (0.35 g, 14.6 mmol) was added to a solution of
4-methylbenzenesulfonamide (2.5 g, 14.6 mmol) in dry DMF
(100 mL), maintained at 0 °C through an external ice bath. After
stirring for 15 min, phenyl isothiocyanate (2 g, 14.8 mmol) was
added. The resulting solution was stirred for 30 min at 60 °C, co-
oled to room temperature and then benzylamine (1.6 g, 14.6 mmol)
was slowly added. Upon addition of HgCl2 (4 g, 14.6 mmol) a
black solid precipitated from the solution which was filtered off
through a plug of Celite. The collected filtrate was evaporated to
dryness under vacuum and the residue taken up with ethyl acetate
Conclusions
From the data reported in the present study it appears
that the recognition of anions in solvents of low polarity
by a series of structurally related urea-type acyclic hosts
is governed both by conformational and electronic effects
induced by the nature of the spacer Y bridging the hydro-
gen-bond donor NH groups. Among the compounds
studied, the urea host 4 showed better recognition in chlo- (100 mL) and with a saturated aqueous solution of Na2CO3
Eur. J. Org. Chem. 2008, 109–120
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
117

L. Pescatori, A. Arduini, A. Pochini, F. Ugozzoli, A. Secchi
(100 mL). The separated organic layer was washed with water up C 62.42, H 8.61, N 5.20, S 11.90; found C 62.70, H 8.57, N 5.11,
FULL PAPER
to neutrality, dried with Na2SO4 and the solvents evaporated to
dryness under reduced pressure. Purification of the solid residue by
column chromatography (hexane/ethyl acetate = 2:1) afforded 1.7 g
(30%) of 1 as a white solid (m.p. 147–149 °C). 1H NMR
(300 MHz): δ = 2.41 (s, 3 H), 4.48 (d, J = 5.7 Hz, 2 H), 5.14 (br. s,
1 H), 7.1–7.2 (m, 3 H), 7.2–7.3 (2 m, 6 H), 7.76 (d, J = 8.2 Hz, 2
H), 9.1 (br. s, 1 H) ppm. 13C NMR (75 MHz): δ = 21.4, 45.2, 126.0,
127.4, 127.6, 128.6, 129.1, 130.1, 135.1, 140.7, 141.9, 153.9 ppm.
MS [CI(+)]: m/z = 381 [MH + 1]+. C21H21N3O2S (379.48): calcd.
C 66.47, H 5.58, N 11.07, S 8.45; found C 66.20, H 6.01, N 10.95,
S 8.32.
S 11.91.
General Procedure for the Synthesis of Urea Derivatives 11–13: A
solution of the appropriate isocyanate (15 mmol) in dry CH2Cl2
(30 mL) was added dropwise to a solution of the amine 9 or 10
(15 mmol) in dry CH2Cl2 (30 mL), maintained at 0 °C through an
external ice bath. The resulting mixture was stirred at room tem-
perature for 48–72 h and then the solvent was evaporated to dry-
ness under reduced pressure. The residue was taken up with a 10%
aqueous solution of HCl (50 mL) and ethyl acetate (50 mL). The
organic layer was separated, washed with water up to neutrality,
dried with Na2SO4 and the solvent evaporated to dryness under
reduced pressure.
1-Nitro-4-(octylsulfanyl)benzene
(7):
1-Octanethiol
(8.8 g,
60.2 mmol) and KOH (3.4 g, 60.5 mmol) were added to a solution
of 1-chloro-4-nitrobenzene (6) (10 g, 63.5 mmol) in dry DMF
(70 mL). The resulting reaction mixture was stirred at 100 °C for
6 h and then the solvent was evaporated to dryness under reduced
pressure. The solid residue was taken up with a 10% aqueous solu-
tion of HCl (100 mL) and CH2Cl2 (100 mL). The organic layer was
separated, washed with water up to neutrality, dried with Na2SO4
and the solvent completely evaporated under reduced pressure. The
oily residue was taken up with cold methanol to afford 11.2 g
(70%) of 7 as a yellowish solid which was collected by filtration
1-Benzyl-3-[4-(octylsulfonyl)phenyl]urea (11): Amine 9 and benzyl
isocyanate were used as reagents. Purification of the residue by col-
umn chromatography (hexane/ethyl acetate = 7:3) afforded 2.3 g
(38%) of 11 as a white solid (m.p. 76.0–76.8 °C). 1H NMR
(300 MHz): δ = 0.88 (t, J = 6.9 Hz, 3 H), 1.2–1.4 and 1.5–1.7 (2 m,
12 H), 3.02 (br. t, 2 H), 4.36 (d, J = 5.7 Hz, 2 H), 6.11 (t, J =
5.7 Hz, 1 H), 7.1–7.3 (m, 5 H), 7.47 (d, J = 9.0 Hz, 2 H), 7.62 (d,
J = 9.0 Hz, 2 H), 7.96 (s, 1 H) ppm. 13C NMR (75 MHz): δ = 14.0,
22.4, 22.6, 28.1, 28.8, 28.9, 31.6, 44.0, 56.4, 118.1, 127.3, 128.6,
129.1, 138.5, 144.8, 154.9 ppm. MS [ESI(+)]: m/z (%) = 425 (100)
[M + Na]+. C22H30N2O3S (402.55): calcd. C 65.64, H 7.51, N 6.96,
S 7.97; found C 65.50, H 7.38, N 7.15, S 7.62.
1
(m.p. 31–32 °C; ref.[34] 29.5–31.0). H NMR (300 MHz): δ = 0.88
(br. t, 3 H), 1.2–1.4, 1.4–1.5 and 1.6–1.8 (3 m, 12 H), 3.00 (t, J =
7.2 Hz, 2 H), 7.29 (d, J = 8.9 Hz, 2 H), 8.10 (d, J = 8.9 Hz, 2
H) ppm. 13C NMR (75 MHz): δ = 14.0, 22.5, 28.4, 28.8, 29.0,
29.05, 31.7, 31.9, 123.8, 125.8, 144.7, 148.1 ppm. MS [EI(+)]: m/z
(%) = 267 (100) [M]+. C14H21NO2S (267.39): calcd. C 62.89, H
7.92, N 5.24, S 11.99; found C 63.08, H 7.92, N 5.40, S 11.62.
1-[4-(Octyloxy)benzyl]-3-phenylurea (12): Amine 10 and phenyl iso-
cyanate were used as reagents. Purification of the residue by
recrystallisation from methanol gave 2.1 g (40%) of pure 12 (m.p.
101–102 °C). 1H NMR (300 MHz): δ = 0.88 (br. t, 3 H), 1.3–1.5
(m, 10 H), 1.7–1.8 (m, 2 H), 3.92 (t, J = 6.6 Hz, 2 H), 4.79 (d, J =
5.4 Hz, 2 H), 6.2 (br. t, 1 H), 6.84 (d, J = 8.7 Hz, 2 H), 7.1–7.2 (m,
J = 5 Hz), 7.39 (t, J = 7.5 Hz, 2 H), 7.69 (br. s, 1 H) ppm. 13C
NMR (75 MHz): δ = 14.0, 22.6, 26.0, 29.2, 29.3, 31.8, 43.5, 68.0,
114.5, 120.6, 123.4, 125.2, 128.6, 129.0, 130.6, 138.6, 156.1,
158.3 ppm. MS [CI(+)]: m/z = 355 [MH]+. C22H30N2O2 (354.49):
calcd. C 74.54, H 8.53, N 7.90; found C 74.61, H 8.42, N 7.96.
1-Nitro-4-(octylsulfonyl)benzene (8): A solution of 3-chloroperben-
zoic acid (3.8 g, 22.4 mmol) in CH2Cl2 (150 mL) was added drop-
wise to a solution of 7 (5 g, 18.7 mmol) in CH2Cl2 (250 mL), main-
tained at 0 °C through an external ice bath. The resulting reaction
mixture was stirred at room temperature for 4 h, then washed in
turn with a 38–40% w/v aqueous solution of NaHSO3 and with a
saturated aqueous solution NaHCO3. The separated organic layer
was dried with Na2SO4 and the solvent was evaporated under re-
duced pressure. Purification of the oily residue by column
chromatography (hexane/ethyl acetate = 4:1), afforded 3.65 g (65%)
of 8 as a pale yellow fluffy solid (m.p. 49.5–50 °C). 1H NMR
(300 MHz): δ = 0.88 (br. t, 3 H), 1.2–1.4, 1.4–1.5 and 1.6–1.8 (3 m,
12 H), 3.15 (br. t, 2 H), 8.14 (d, J = 8.8 Hz, 2 H), 8.44 (d, J =
8.8 Hz, 2 H) ppm. 13C NMR (75 MHz): δ = 13.9, 22.4, 28.1, 28.7,
28.8, 31.5, 56.1, 124.4, 129.5, 144.8, 151.2 ppm. MS [EI(+)]: m/z
(%) = 299 (30) [M]+. C14H21NO4S (299.39): calcd. C 56.17, H 7.07,
N 4.68, S 10.71; found C 56.01, H 7.01, N 4.69, S 10.47.
1-(4-Nitrophenyl)-3-[4-(octyloxy)benzyl]urea (13): Amine 10 and 4-
nitrophenyl isocyanate were used as reagents. Purification of the
residue by recrystallisation from methanol gave 4.2 g (70%) of pure
1
13 (m.p. 162–163 °C). H NMR (300 MHz, CDCl3/[D6]DMSO): δ
= 0.63 (br. t, 3 H), 1.0–1.2 (m, 10 H), 1.4–1.6 (m, 2 H), 3.68 (t, J
= 6.5 Hz, 2 H), 4.08 (d, J = 5.4 Hz, 2 H), 6.2 (br. t, 1 H), 6.61 (d,
J = 8.4 Hz, 2 H), 6.98 (d, J = 8.4 Hz, 2 H), 7.33 (d, J = 9.0 Hz, 2
H), 7.85 (d, J = 9.0 Hz, 2 H), 8.6 (br. s, 1 H) ppm.13C NMR: δ =
(75 MHz, CDCl3/[D6]DMSO): δ = 13.9, 22.3, 25.7, 28.9, 29.0, 31.5,
43.0, 67.8, 114.3, 116.7, 124.8, 128.7, 130.7, 139.9, 146.7, 154.7,
158.2 ppm. MS [CI(+)]: m/z = 400 [MH]+. C22H29N3O4 (399.48):
calcd. C 66.14, H 7.32, N 10.52; found C 65.81, H 7.28, N 10.30.
4-(Octylsulfonyl)aniline (9): SnCl2·2H2O (7.5 g, 33.2 mmol) was
added to a solution of 8 (2 g, 6.7 mmol) in ethanol (150 mL). The
resulting heterogeneous solution was refluxed whilst stirring for 4 h
and then the solvent was completely removed under reduced pres-
sure. The residue was taken up with 1  aqueous solution of NaOH
(100 mL) and extracted with ethyl acetate. The separated organic
phase was dried with Na2SO4 and the solvents evaporated to dry-
ness under reduced pressure. Purification of the solid residue by
column chromatography (hexane/ethyl acetate = 3:2) afforded 1.3 g
(70%) of 9 as a white solid (m.p. 97.8–98.5 °C). 1H NMR
(300 MHz): δ = 0.88 (t, J = 7.2 Hz, 3 H), 1.2–1.4 and 1.6–1.7 (2 m,
12 H), 3.03 (br. t, 2 H), 4.2 (br. s, 2 H), 6.73 (d, J = 8.7 Hz, 2 H),
1-[4-(Octylsulfonyl)phenyl]-3-phenylurea (14): Phenyl isocyanate
(0.12 g, 1.0 mmol) was added to a solution of 9 (0.2 g, 0.7 mmol)
in dry THF (50 mL). The resulting mixture was stirred at room
temperature for 48 h and then the solvent was evaporated to dry-
ness under reduced pressure. Purification of the solid residue by
column chromatography (hexane/ethyl acetate = 3:2) followed by
recrystallisation with methanol afforded 0.14 g (60%) of 14 as a
1
white solid (m.p. 174–175 °C). H NMR (300 MHz): δ = 0.88 (br.
t, 3 H), 1.2–1.4 and 1.5–1.8 (2 m, 12 H), 3.11 (br. t, 2 H), 7.09 (t,
J = 7.5 Hz, 1 H), 7.59 (t, J = 7.4 Hz, 2 H), 7.72 (t, J = 7.4 Hz, 1
H), 7.82 (d, J = 8.7 Hz, 2 H), 7.92 (d, J = 8.7 Hz, 2 H), 7.99 (d, J
7.67 (d, J = 8.9 Hz, 2 H) ppm. 13C NMR (75 MHz): δ = 14.0, 22.5, = 8.7 Hz, 2 H), 9.0 (br. s, 1 H), 11.3 (br. s, 1 H) ppm. 13C NMR
22.8, 28.2, 28.8, 28.9, 31.6, 56.7, 114.0, 127.3, 130.1, 151.1 ppm.
(75 MHz): δ = 14.0, 22.5, 22.7, 28.2, 28.9, 29.0, 31.6, 56.5, 118.7,
120.6, 124.3, 129.2, 131.4, 144.4, 152.5 ppm. MS [ESI(+)]: m/z (%)
MS [EI(+)]: m/z (%) = 269 (20) [M]+. C14H23NO2S (269.40): calcd.
118
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 109–120

Urea-Type Anion Receptors in Low Polar Media
= 411 (100) [M + Na]+. C21H28N2O3S (388.52): calcd. C 64.92, H
7.26, N 7.21, S 8.25; found C 64.66, H 7.18, N 6.93, S 8.09.
the MEP plotted on van der Waals molecular surfaces were ob-
tained with Molekel.[41]
UV/Vis Titrations: Spectroscopic titrations with 11–15 were carried
out in a quartz cuvette (path length = 1 cm) maintained at 293 K
through an external thermostat by adding small aliquots of solu-
tions of the guests (usually 2ϫ10–4 ) in chloroform to a solution
of the hosts in chloroform (usually 2ϫ10–5 ). The spectroscopic
data were collected in the 250–650 nm wavelength range. The bind-
ing constants were calculated by selecting different binding models
with the Specfit/32[24] software. The fitting of the spectroscopic data
was carried out by considering the optical variation for the 250–
380 nm wavelength range. For each titration experiment the Weber
parameter p = [concentration of complex]/[maximum possible con-
centration of complex] was determined from the calculated binding
constant. If necessary the concentrations of the two reactants were
adjusted and the UV/Vis titration experiment repeated to explore
the proper p range (0.2–0.8).
1,3-Bis[4-(octylsulfonyl)phenyl]urea (15): Triphosgene (0.25 g,
0.8 mmol) was added to a solution of 9 (0.5 g, 1.9 mmol) in dry THF
(50 mL). The resulting mixture was stirred at room temperature for
48 h and then the solvent was evaporated to dryness under reduced
pressure. Purification of the solid residue by column chromatography
(hexane/ethyl acetate = 1:1) followed by recrystallisation with meth-
anol afforded 0.52 g (50%) of 15 as a white solid (m.p. 158.5–
1
159.5 °C). H NMR (300 MHz): δ = 0.85 (br. t, 6 H), 1.2–1.4 and
1.6–1.8 (2 m, 24 H), 3.14 (br. t, 4 H), 7.66 (d, J = 8.7 Hz, 4 H),
7.81 (d, J = 8.7 Hz, 4 H), 8.0 (s, 2 H) ppm. 13C NMR (75 MHz):
δ = 14.0, 22.5, 22.7, 28.2, 28.8, 28.9, 31.5, 56.5, 118.7, 129.2, 131.5,
144.2, 151.4 ppm. MS [ESI(+)]: m/z (%) = 587 (100) [M + Na]+.
C29H44N2O5S2 (564.80): calcd. C 61.67, H 7.85, N 4.96, S 11.35;
found C 61.40, H 8.18, N 5.13, S 11.11.
Continuous Variation 1H NMR Methods (Job Plot): Aliquots of
stock solutions in CDCl3 of 4 and of each of the TBA salts were
added to several 5-mm NMR tubes in different ratios. The relative
concentrations of the two interacting species were varied continu-
ously in the tubes but their sum was kept constant (1.45ϫ10–3 ).
In this way, 14 samples were prepared in which the mole fraction
(x) of 4 was varied from 0.1 to 1. A 1H NMR spectrum was re-
corded for each sample. The corresponding Job plot was obtained
by plotting a property proportional to the complex concentration,
(δobs – δfree)[4]0, vs. the mole fraction of 4. δobs and δfree represent
the chemical shifts of the host NH proton in the complex and in
the free ligand, respectively, and [4]0 is the total concentration of 4
in each tube. The stoichiometry of binding was obtained from the
value of the mole fraction in the abscissa corresponding to the
maximum of the curve. For a host–guest ratio of 1:1, x = 0.5.
Supporting Information (see also the footnote on the first page of
this article): Spectral characterization (1H and 13C NMR) of new
compounds 1 and 11–15, calculated interconversion barriers for the
rotation around the (Ph)C–N–C–O(S) dihedral from –180 to 180°
in 3 and 4, molecular electrostatic potential (MEP) plotted on the
van der Waals surfaces of 2–4 and UV spectra with the species dis-
tribution diagram for the titration of 15 with TBACl (16 pages).
Acknowledgments
This work was supported by the Ministero dell’Università e della
Ricerca (MIUR) (Sistemi supramolecolari per la costruzione di
macchine molecolari, conversione dell’energia, sensing e catalisi).
The authors thank the CIM (Centro Interdipartimentale di Misure)
“G. Casnati” of the University of Parma for NMR and mass mea-
surements and Dr. Matteo Tegoni of the University of Parma for
fruitful discussions on the UV/Vis measurements.
1H NMR Titrations: Solutions (500 µL) of hosts 1–4 and 12 in
CDCl3 (usually c = 1.2ϫ10–2 ) were prepared in a 5-mm NMR
tube and small aliquots of solutions of tetrabutylammonium salts
(usually c = 1.2ϫ10–1 ) in CDCl3 were added. A spectrum was
recorded after each addition. The binding constants were calcu-
lated by using the Wilcox equation[15] implemented in the Sigma-
plot package[35] by the non-linear fitting of the chemical shift varia-
tion of the host’s NH protons. For each titration experiment the
Weber parameter p = [concentration of complex]/[maximum pos-
sible concentration of complex] was determined according to the
calculated binding constant. If necessary the concentrations of the
two reactants were adjusted and the NMR titration experiment re-
peated to explore the proper p range (0.2–0.8).
[1] For books and reviews on anion receptors, see: a) P. D. Beer,
P. A. Gale, Angew. Chem. Int. Ed. 2001, 40, 486–516; b) C.
Suskai, T. Tuntulani, Chem. Soc. Rev. 2003, 32, 192–202; c) V.
McKee, J. Nelson, R. M. Town, Chem. Soc. Rev. 2003, 32, 309–
325; d) V. Amendola, M. Bonizzoni, D. Esteban-Gómez, L.
Fabbrizzi, M. Licchelli, F. Sancenón, A. Taglietti, Coord.
Chem. Rev. 2006, 250, 1451–1470; e) J. L. Sessler, P. A. Gale,
W. S. Cho, Anion Receptor Chemistry, The Royal Society of
Chemistry, Cambridge, 2006; f) J. W. Steed, Chem. Commun.
2006, 2637–2649; g) P. A. Gale, Acc. Chem. Res. 2006, 39, 465–
475.
Cambridge Structural Database: X-ray diffraction crystal structures
of compounds having the general formula PhNH–Y–NHR(alkyl)
were retrieved from the Cambridge Structural Database (CSD) and
analysed with Mercury.[36] Searches yielded 12 structures for Y =
O, eight structures for Y = S, one structure for Y = SO2 although
39 structures were found for N,N-dialkylsulfamide derivatives.
[2] a) C. R. Bondy, S. J. Loeb, Coord. Chem. Rev. 2003, 240, 77–
99; b) K. Choi, A. D. Hamilton, Coord. Chem. Rev. 2003, 240,
101–110; c) K. Bowman-James, Acc. Chem. Res. 2005, 38, 671–
678, and references cited therein.
[3] For papers on anion recognition using (thio)urea derivatives,
see: a) E. Fan, S. A. Van Arman, S. Kincaid, A. D. Hamilton,
J. Am. Chem. Soc. 1993, 115, 369–370; b) C. S. Wilcox, E. Kim,
D. Romano, L. H. Kuo, A. L. Burt, P. D. Curran, Tetrahedron
1995, 51, 621–634; c) S. Nishizawa, P. Bühlmann, M. Iwao, Y.
Umezawa, Tetrahedron Lett. 1995, 36, 6483–6486; d) T.
Gunnlaugsson, A. P. Davis, G. M. Hussey, J. Tierney, M.
Glynn, Org. Biomol. Chem. 2004, 2, 1856–1863; e) T.
Gunnlaugsson, P. E. Kruger, P. Jensen, J. Tierney, H. D. P. Ali,
G. M. Hussey, J. Org. Chem. 2005, 70, 10875–10878; f) M.
Boiocchi, L. Del Boca, D. Esteban-Gómez, L. Fabbrizzi, M.
Licchelli, E. Monzani, J. Am. Chem. Soc. 2004, 126, 16507–
16514; g) M. Boiocchi, L. Del Boca, D. Esteban-Gómez, L.
Fabbrizzi, M. Licchelli, E. Monzani, Chem. Eur. J. 2005, 11,
3097–3104; h) D. Esteban-Gómez, L. Fabbrizzi, M. Licchelli,
Structure Calculations: The Monte Carlo search and determi-
nations of the rotational barriers were performed by using the
MMFF94 force field[37] implemented in the Spartan 4.0 molecular
modelling software.[38] For 3 and 4, a modified force field
(MMFF94+) was implemented in the Spartan package, in accord
with the torsional parameters reported in ref.[16]. The figures of
the most representative conformers of 1–4 and the corresponding
complexes with chloride were created with PLATON for Win-
dows.[39] Further structure minimisation and molecular electro-
static potential (MEP) calculations on 2–4 were performed with PC
GAMESS[40] by using ab initio and DFT methods. Illustrations of
Eur. J. Org. Chem. 2008, 109–120
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
119

L. Pescatori, A. Arduini, A. Pochini, F. Ugozzoli, A. Secchi
FULL PAPER
E. Monzani, Org. Biomol. Chem. 2005, 3, 1495–1500; i) V.
Amendola, D. Esteban-Gómez, L. Fabbrizzi, M. Licchelli, Acc.
Chem. Res. 2006, 39, 343–353.
Bharatam, R. Moudgil, D. Kaur, J. Phys. Chem. A 2003, 107,
1627–1634.
The reported σp values for SO2NH2, CONH2 and CSNH2 are
0.49, 0.26 and 0.24, respectively, see, for example: C. Hansch,
A. Leo, R. W. Taft, Chem. Rev. 1991, 91, 165–195, and refer-
ences cited therein.
J. S. Murray, K. Sen (Eds.), Molecular Electrostatic Potentials:
Concepts and Applications in Theoretical and Computational
Chemistry, Elsevier, 1996, vol. 3 .
The pKa values determined in DMSO for N,N-diphenylthio-
urea and N,N-diphenylurea are 13.5 and 19.5, respectively, see:
F. G. Bordwell, D. J. Algrim, J. A. Harrelson, J. Am. Chem.
Soc. 1998, 110, 5903–5904.
[20]
[4] a) B. H. M. Snellink-Ruël, M. M. G. Antonisse, J. F. J.
Engbersen, P. Timmerman, D. N. Reinhoudt, Eur. J. Org.
Chem. 2000, 165–170; b) B. R. Linton, M. S. Goodman, E.
Fan, S. A. Van Arman, A. D. Hamilton, J. Org. Chem. 2001,
66, 7313–7319; c) J. P. Clare, A. J. Ayling, J.-B. Joos, A. L. Sis-
son, G. Magro, M. N. Pérez-Payán, T. N. Lambert, R. Shukla,
B. D. Smith, A. P. Davis, J. Am. Chem. Soc. 2005, 127, 10739–
10746; d) D. R. Turner, M. J. Paterson, J. W. Steed, J. Org.
Chem. 2006, 71, 1598–1608; e) F. Otón, A. Tárraga, A. Espi-
nosa, M. D. Velasco, P. Molina, J. Org. Chem. 2006, 71, 4590–
4598; f) B. Schazmann, N. Alhashimy, D. Diamond, J. Am.
Chem. Soc. 2006, 128, 8607–8614.
[5] a) G. Cooke, V. M. Rotello, Chem. Soc. Rev. 2002, 31, 275–286;
b) L. Nie, Z. Li, J. Han, X. Zhang, R. Yang, X.-W. Liu, F.-Y.
Wu, J.-W. Xie, Y.-F. Zhao, Y.-B. Jiang, J. Org. Chem. 2004, 69,
6449–6454; c) F.-W. Wu, Z. Li, L. Guo, X. Wang, M.-H. Lin,
Y.-F. Zhao, Y.-B. Jiang, Org. Biomol. Chem. 2006, 4, 624–630,
and references cited therein.
[6] For a review on the acidity scale in DMSO, see: F. G. Bordwell,
Acc. Chem. Res. 1988, 21, 456–463, and references cited
therein.
[7] a) A. Wittkopp, P. R. Schreiner, Chem. Eur. J. 2003, 9, 407–
414; b) P. R. Schreiner, Chem. Soc. Rev. 2003, 32, 289–296; c)
S. J. Connon, Chem. Eur. J. 2006, 12, 5418–5427, and references
cited therein.
[8] J. Zhang, Y. Shi, Tetrahedron Lett. 2000, 41, 8075–8078.
[9] a) S. Bartoli, S. Roelens, J. Am. Chem. Soc. 2002, 124, 8307–
8315; b) J. W. Jones, H. W. Gibson, J. Am. Chem. Soc. 2003,
125, 7001–7004, and references cited therein.
[21]
[22]
[23]
[24]
The values of σp found for NO2 and SO2CH2CH3 are 0.78 and
0.77, respectively, see ref.[20]
.
®
SPECFIT/32 Global Analysis System for Microsoft Win-
dows®.
See
http://www.bio-logic.info/rapid-kinetics/spec-
fit.html.
[25]
Wilcox et al. have shown that the introduction of a nitro sub-
stituent into the para position of an N-phenyl-NЈ-alkylthiourea
derivative leads to a strong increment of the corresponding
binding constant in chloroform solution; see ref.[3b]
.
[26]
[27]
[28]
See, for example: W. C. Vosburgh, G. R. Cooper, J. Am. Chem.
Soc. 1941, 63, 437–442.
E. R. Malinowski, Factor Analysis in Chemistry, 2nd ed., 1991,
Wiley, New York, 1991.
Very recently this behaviour was also observed by other au-
thors, see, for example: a) L. S. Evans, P. A. Gale, M. E. Light,
R. Quesada, Chem. Commun. 2006, 965–967; b) Y.-J. Kim, H.
Kwak, S. J. Lee, J. S. Lee, H. J. Kwon, S. H. Nam, K. Lee, C.
Kim, Tetrahedron 2006, 62, 9635–9640.
P. O. Burke, S. D. McDermott, T. J. Hannigan, W. J. Spillane,
J. Chem. Soc. Perkin Trans. 2 1984, 1851–1854.
K. Ramadas, N. Srinivasan, N. Janarthanan, Tetrahedron Lett.
1993, 40, 6447–6450.
J. R. Dimmock, S. C. Vashishtha, J. P. Stables, Pharmazie 2000,
55, 490–494.
C. Naegeli, A. Tyabji, L. Conrad, Helv. Chim. Acta 1938, 21,
1127–1143.
R. F. Collins, B. J. Large, J. Pharm. Pharmacol. 1962, 14, 48–
58.
G. Braunerová, V. Buchta, L. Silva, J. Kunesˇ, K. Palát Jr, Il
Farmaco 2004, 59, 443–450.
SigmaPlot, Graphing software from Systat Software, Inc. See
http://www.systat.com/products/sigmaplot/.
Mercury: for visualization and analysis of crystal structures,
see: C. F. Macrae, P. R. Edgington, P. McCabe, E. Pidcock,
G. P. Shields, R. Taylor, M. Towler, J. van de Streek, J. Appl.
Crystallogr. 2006, 39, 453–457.
a) T. A. Halgren, J. Comput. Chem. 1996, 17, 490–519; b) T. A.
Halgren, J. Comput. Chem. 1996, 17, 520–552; c) T. A.
Halgren, J. Comput. Chem. 1996, 17, 553–586; d) T. A.
Halgren, R. B. Nachbar, J. Comput. Chem. 1996, 17, 587–615;
e) T. A. Halgren, J. Comput. Chem. 1996, 17, 615–641; f) T. A.
Halgren, J. Comput. Chem. 1999, 20, 720–729; g) T. A.
Halgren, J. Comput. Chem. 1999, 20, 730–748.
Wavefunction, Irvine, CA, 2001.
A. L. Speck, Acta Crystallogr., Sect. A 1990, 46, C34.
A. A. Granovsky, PC GAMESS, version 7.0.
See http://classic.chem.msu.su/gran/gamess/index.html.
Molekel 5.2, Swiss National Supercomputing Centre (CSCS),
2007. See http://www.cscs.ch.
[10] N,N-Diphenylurea experiences extensive self-association in
1
CH2Cl2 solution, as shown by H NMR dilution studies, see:
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
Y. Ge, L. Miller, T. Ouimet, K. D. Smith, J. Org. Chem. 2000,
65, 8831–8838.
[11] a) C. S. Wilcox, J. C. Adrian Jr, T. H. Webb, F. J. Zawacki, J.
Am. Chem. Soc. 1992, 114, 10189–10197; b) R. Deans, G. Co-
oke, V. M. Rotello, J. Org. Chem. 1997, 62, 836–839.
[12] a) For the oligomerisation of N,N-diphenylurea derivatives, see:
F. Lortie, S. Boileau, L. Bouteiller, Chem. Eur. J. 2003, 9, 3008–
3014; b) for a review on the application of the isodesmic model,
see: R. B. Martin, Chem. Rev. 1996, 96, 3043–3064, and refer-
ences cited therein.
[13] a) K. A. Connors, Binding Constants, Wiley, New York, 1987;
b) K. Hirose, J. Inclusion Phenom. Macrocyclic Chem. 2001, 39,
193–209.
[14] J. Scheerder, M. Fochi, J. F. J. Engbersen, D. N. Reinhoudt, J.
Org. Chem. 1994, 59, 7815–7820.
[15] C. S. Wilcox in Frontiers of Supramolecular Organic Chemistry
and Photochemistry (Eds.: H.-J. Schneider, H. Dürr), VCH,
Weinheim, 1991, pp. 123–143.
[16] a) V. S. Bryantsev, T. K. Firman, B. P. Hay, J. Phys. Chem. A
2005, 109, 832–842; b) V. S. Bryantsev, B. P. Hay, J. Phys.
Chem. A 2006, 110, 4678–4688.
[17] The higher stability of the Z conformer with respect the E con-
former in phenylurea derivatives has also been ascribed to the
formation of a weak hydrogen bond between the C=O group
and the neighbouring ortho hydrogen of the phenyl ring. This
interaction is not possible when the phenylurea adopts the E
conformation, see: R. Emery, N. A. Macleod, L. C. Snoek, J. P.
Simons, Phys. Chem. Chem. Phys. 2004, 6, 2816–2820.
[18] a) J. Tomasi, R. Cammi, B. Mennucci, Int. J. Quantum Chem.
1999, 75, 783–803; b) J. Tomasi, R. Cammi, B. Mennucci, C.
Cappelli, S. Corni, Phys. Chem. Chem. Phys. 2002, 4, 5697–
5712; c) J. Tomasi, B. Mennucci, R. Cammi, Chem. Rev. 2005,
105, 2999–3093.
[37]
[38]
[39]
[40]
[41]
[19] The different character of the C–N bond in urea and thiourea
has been explained on the basis of orbital interactions, see: P. V.
Received: August 10, 2007
Published Online: November 12, 2007
120
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 109–120