Complexation of anions including nucleotide anions by open-chain host compounds with amide, urea, and aryl functions

Article abstract of DOI:10.1002/(SICI)1522-2675(20000216)83:2<465::AID-HLCA465>3.0.CO;2-F

A systematical evaluation of association constants between halide, phosphate, and carboxylate anions with N-methylformamide (1) and the related bidentate receptors 2-6 (derived from, e.g., phthalic acid or ethylenediamine) in CDCl₃ as solvent y

Full text of DOI:10.1002/(SICI)1522-2675(20000216)83:2<465::AID-HLCA465>3.0.CO;2-F

Helvetica Chimica Acta ± Vol. 83 (2000)
465
Complexation of Anions Including Nucleotide Anions by Open-Chain Host
Compounds with Amide, Urea, and Aryl Functions
by Frank Werner and Hans-Jörg Schneider*
FR Organische Chemie der Universität des Saarlandes, D-66041 Saarbrücken (Fax: #681 302 4105/3209; e-mail:
ch12hs@rz.uni-sb.de)
Dedicated to Prof. Gunther Wulff on the occasion of his 65th birthday
A systematical evaluation of association constants between halide, phosphate, and carboxylate anions with
N-methylformamide (1) and the related bidentate receptors 2 ± 6 (derived from, e.g., phthalic acid or
ethylenediamine) in CDCl3 as solvent yielded increments of complexation free-energy DDG for each single H-
bond, which varied like, e.g., 5.1 kJ/mol (for Cl ), 4.0 kJ/mol (for Br ), 4.0 kJ/mol (for I ) (with values taken
from Tables 1 and 2), in line with expected H-bond strength. The observed complexation induced
NH-NMR shift (CIS) values also showed a regular change, in the case of 1, e.g., from 5.0 to 2.8 to 2.1 ppm
(Table 1), with about half of these values with the bidentate ligands (Tables 2 and 3). Tridentate hosts led to a
substantial binding increase, if strain-free convergence of all NH donor functions towards the anion was
possible. The tris[urea] ligand 10 yielded, even in the polar solvent DMSO, with Cl a DG of 21.5 kJ/mol and
with Br of 10 ´ 5 kJ/mol, whereas with I , no association was detectable. The results demonstrated that small,
inexpensive, and conformationally mobile host compounds can exhibit high affinities as well as descrimination
with anions, as much as more preorganized receptors do which require multistep synthesis. The corresponding
1
adamantyl derivative 13 allowed measurements also in CDCl3, with K ˆ 4.3 ´ 104 m for chloride (Table 7).
Complexes with nucleotide anions were again particularly strong with the tridentate urea-based ligands, the
latter being optimal ligands for chloride complexation. For the association of 10 with AMP2 and GMP2 in
(D6)DMSO, the association constants were 3 ´ 104 m 1 (Table 8) and almost the same as with Cl . In the case of
the urea derivatives 17, 18, and 21, containing only one phenyl or pyrenyl substituent, however, the DG values
decreased in the order A > C > T> G (e.g. 13.6, 11.6, 7.6, 10.5 kJ/mol in the case of 17, resp.; Table 8). In
H2O, the pyrenyl-substituted urea derivatives allow measurements with fluorescence, and, unexpectedly, show
1
only smaller nucleobase discrimination, with constants around 3 ´ 103 m
.
Introduction. ± Anion complexation has often been deplored as being a less
developed area of supramolecular chemistry: an excellent recent review by Antonisse
and Reinhoudt [1], which allows us to restrict ourselves here to few references,
illustrates the current high interest in the field and its rapid expansion. The binding of
anions to electroneutral receptors is of special significance, also in view of the absence
of competing counterion complexes, which are usually present if one uses cationic host
compounds. Recent investigations, particularly by Reinhoudt et al. (see [1]), have
already shown the potential of amides for complexation of anions. A so-called tuneable
amide based on a calixarene has been synthesized by Cameron and Loeb [2], although
the anion-association constants reported were low. Coordination to amide functions is
also essential for orthophosphate [3] or sulfate-ion [4] binding in proteins. One aim of
the present work was to establish free-energy increments for binding of different anions
by a systematic experimental investigation to the ligands 1 ± 22 (see Fig. 1), and another

466
Helvetica Chimica Acta ± Vol. 83 (2000)
Fig. 1. Structures of the ligands

Helvetica Chimica Acta ± Vol. 83 (2000)
467
to explore the use of substituted amide-type derivatives for the biologically particularly
interesting complexation of nucleotide anions [5] in water.
Complexation of Simple Anions. ± First, we tested the affinity of the amide function
for the binding of different anions with the simple ligand N-methylacetamide (1) which,
in contrast to earlier assumptions [6], could well be measured in CDCl3 as solvent and
presents a cornerstone for understanding anion-peptide interactions (Table 1).
1H-NMR-Shift titrations of 1 with the anions usually showed a nearly perfect nonlinear
least-squares fit for a 1:1 stoichiometric model of amide/anion. The observed higher
affinity of phosphate vs. benzoate likely reflects on the one hand the polydentate
advantage of the dihydrogenphosphate anion, and on the other hand the increased
charge at the O-atom as confirmed by ab initio calculations (preliminary charge
calculations with the 6-31G** basis set indicate 0.56 e units at the O-atoms of
benzoate, and 1.1 e at the O-atoms of dihydrogenphosphate). The sequence Cl >
Br > I reflects the change of the H-bond-acceptor strength of the halides, which is
noticeably also found in the case of complexes between hydroxy compounds and the
halides [7]; another factor could be the change in anion radii of the halides. The
1
observed complexation-induced H-NMR shifts (CIS) show typical values for amide
protons in CDCl3 involved in H-bonding, again in the Cl > Br > I sequence with
decreasing H-bond strength.
Table 1. Titration of N-Methylacetamide (1) with Different Anions
Proton Aniona)
Dihydrogenphosphate
Benzoate
Chloride
Kb)
Bromide
CISc) Kb)
Iodide
Kb)
Kb)
CISc)
Kb)
CISc)
CISc)
CISc)
NH
MeN
MeCO
DGÆ
26
35
4.49
14
16
14
3.99
0.1
0.015
10
4.99
8
5
2.84
5
3
2.11
0.10
0.06
0.084
±e) ±e)
±e) ±e)
5.7
0.09
±e)
(0.07f))
±e)
4.6
(0.08f))
8g)
4.1
d
)
8.4
6.6
a
) Measured as Bu4N salts in CDCl3. b) In l/mol; calculated from d(H) for a 1 : 1 model; error< 10% from Dd
(NH); <15% from Dd (CH). c) Complexation-induced NMR shifts (CIS) in ppm relative to SiMe4 in CCl4
(external); positive sign, deshielding; negative sign, shielding. d) Average value from evaluation of several
‡
e
f
‡
proton shifts, in kJ/mol at 298 K. ) Not observable, due to overlapping Bu4N ion signals. ) Calculated with the
K value obtained from other proton shifts. g) Error> 20%.
The results of the diamide titrations with 1,2-phthalodiamide (2) or N,N'-(ethane-
1,2-diyl)bis[acetamide] (3) (Table 2) clearly establish a chelate effect, with more or less
additive DG increments. Even the CIS values are smaller by a rather constant factor of
nearly 50% in comparison to the monofunctional ligand 1, in line with the charge
distribution over two NH bonds instead of one. The increased affinity of 2 in
comparison to 3 is due to the preorganization of the two amide functions (see Fig. 2 for
the structure of benzoate ´ 2), although preliminary molecular-mechanics calculations
show a rather unfavorable geometry in the ground state of 2. Crabtree and co-workers
[8] have obtained K values (l/mol) of 6 ´ 104 for Cl , of 7 ´ 103 for Br , and of 5 ´ 102 for
I with isophthalamides, however, in the particularly unpolar solvent CH2Cl2. Energetic
disadvantages are even more pronounced for the diamides 4 ± 6, which must adopt

468
Helvetica Chimica Acta ± Vol. 83 (2000)
Table 2. Titrations of Hosts 2 and 3 with Different Anions
Proton
Aniona)
Benzoate
Chloride
Bromide
Iodide
Kb)
CISc)
2.750
0.095
0.175
Kb)
CISc)
Kb)
CISc)
Kb)
CISc)
2
3
NH
MeN
H
DGÆ
NH
54
17
12
9.9
328
±e)
14.4
21
±e)
±e)
7.5
108
±e)
11.6
2.001
±e)
16
1.270
±e)
9
±e)
11
5
39
37
9
0.802
±e)
0.071
±e)
±e)
6.9
18
C(3)/H C(6)
)
±e)
±e)
d
3.45
±e)
2.82
2.25
±e)
1.510
0.076
MeCO
(0.13f))
±e)
7.2
d
DGÆ
)
a) ± f) See Table 1.
partial gauche conformations for the anion complexation and, therefore, exhibit
decreasing affinities (Table 3). In another context, it has been shown that the decrease
in affinity of H-bond complexes for open-chain ligands like 3, 5, and 6 is, with only ca.
1.3 kJ/mol disadvantage per additional single bond, relatively small [9], as long as there
is no built-up of steric strain by complexation. Therefore, open-chain combinations of
receptor sites offer an attractive alternative to synthetically difficult to obtain and
expensive preorganized receptors. Derivatives of tris(aminoethyl)amin (tren) offer a
convenient entry to tridentate ligands such as 7, or to ligands with more binding sites
such as 8. The increase of association constants by introduction of a third function,
however, is only moderate (Table 4), in agreement with related literature data [10].
Significantly improved affinities can be achieved with urea derivatives [6][11 ± 13]
which are easily prepared from isocyanates and the corresponding amines. The
1H-NMR spectra of the N,N'-disubstituted ureas show only the (E,E)-isomer (> 99%)
around the CO N pseudo-double bond, which is due to the high strain energy of the
Table 3. Titrations of Hosts 4 ± 6 with Different Anions
Proton
Aniona)
Benzoate
Kb)
Chloride
Kb)
Bromide
Kb)
Iodide
Kb)
CISc)
CISc)
CISc)
CISc)
4
NH
CH2
H
H
DGÆ
NH
MeCO
DGÆ
NH
53
52
56
48
3.12
0.22
0.33
0.11
< 10
±f)
±f)
±f)
±f)
±f)
< 5
±f)
±f)
±f)
±f)
±f)
±f)
±f)
< 5
±f)
±f)
±f)
±f)
±f)
C(2)/H C(5)
±f)
±f)
±f)
4
27
40
8.2
26
21
18
C(3)/H C(6)
±f)
d
)
9.8
78
>
5.7
>
4
>
5
6
3.03
±e)
37
3.22
±e)
21
2.03
±e)
1.39
0.03
±e)
10.8
48
±e)
8.9
±e)
d
)
7.5
28
18
±e)
8.3
2.65
±e)
±e)
29
15
±e)
8.3
2.38
0.09
±e)
1.77
0.08
±e)
1.12
0.06
0.09
MeCO
CH2(2)/CH2(3)
DGÆ
±e)
±e)
9.6
d
)
8.1
a) ± f) See Table 1.

Helvetica Chimica Acta ± Vol. 83 (2000)
469
Table 4. Titrations of Triamide 7 and Hexamide 8 as Hosts
Proton
Aniona)
Benzoate
Chloride
Bromide
Iodide
Kb)
CISc)
Kb)
CISc)
Kb)
CISc)
Kb)
CISc)
7
8
NH
CH2(2)
CH2(1)
MeCO
DGÆ
NH (imino)
NH (carbam.) 348
PhCH2
COCH2
195
±e)
2.90
±e)
0.03
0.02
307
±e)
2.45
±e)
±e)
±e)
125
94
46
81
12.0
30
175
150
±e)
12.6
1.81
0.05
0.05
0.143
52
±e)
20
39
9.8
17
31
38
40
8.8
1.17
±e)
0.02
0.08
(436)f)
(390)f)
13.1
±e)
±e)
d
)
14.2
395
1504
1537
±e)
213
1.26
1.89
±e)
0.30
1.68
0.19
±e)
0.36
1.09
0.183
±e)
0.30
0.53
0.10
0.08
±e)
±e)
14
±e)
d
DGÆ
)
18.2
a) ± f) See Table 1.
Fig. 2. Structure of the benzoate ´ 3 complex (from CHARMm energy minimization)
(Z,Z)- and (E,Z)-conformers; thus they provide two NH protons pointing towards the
anion. As a result, even the simple N,N'-diphenylurea (9) allows measurement of anion
complexation in the highly competitive solvent DMSO with chloride and bromide
(Table 5). If one makes such urea derivatives tridentate (see ligands 10 and 11),
affinities with a DG of up to 21.3 kJ/mol are observed (Table 6), which, to the best of
our knowledge, are the highest observed with electroneutral ligands in such a polar
solvent (see Fig. 3 for the structure of chloride ´ 10). At the same time, these easily
accessible ligands discriminate the different halides quite well. Besides this, Alcazar
and co-workers [13] described ureas bound to chromane skeletons which associate the

470
Helvetica Chimica Acta ± Vol. 83 (2000)
Table 5. Titrations of N,N'-Diphenylurea (9) as Host
Proton
Aniona)
Chloride
Kb)
Bromide
Kb)
Iodide
Kb)
Iodidef)
Kb)
CISc)
CISc)
1.51
CISc)
CISc)
e)
NH
Ho
Hm
Hp
DGÆ
40
28
47
50
9.2
1.52
4
< 1
±
±
±
±
±
±
1070
1380
±g)
1290
20.4
1.21
e)
e)
e)
e)
0.029
0.029
0.044
±
±
±
±
0.322
±g)
e)
e)
e)
e)
e)
e)
e)
e)
±
±
±
0.192
d
)
3.4
> 0
a) ± g) See Table 1, except that (D6)DMSO was used as solvent. h) In CDCl3.
Table 6. Titrations of Tren-Based Urea Derivatives 10 and 11
Proton
Aniona)
Chloride
Kb)
Bromide
Kb)
Iodide
Kb)
ca. 0
ca. 0
ca. 0
ca. 0
ca. 0
CISc)
CISc)
CISc)
10 NHAr
1300
0.63
0.41
0.05
0.03
±
48
47
85
43
85
10.2
9
0.30
0.45
0.01
0.07
0.07
±e)
±e)
±e)
±e)
±e)
NHCH2 1800
Ho
1050
1300
±
21.3
88
Hm
Hp
DGÆ
d
)
> 0
11 NHAr
NHCH2
Ho
0.60
0.41
0.08
±e)
0.51
0.26
0.06
±e)
ca. 1
±e)
±e)
±e)
±e)
±e)
77
40
11
9
±e)
±e)
Hm
±e)
±e)
10.5
±e)
±e)
5.6
±e)
Hp
DGÆ
±e)
±e)
±e)
d
)
ca. 0
a) ± f) See Table 1, except that (D6)DMSO was used as solvent.
di- or trianion sulfate or phosphate, respectively, in MeOH with association constants of
up to 6 ´ 104 l/mol. A report of Reinhoudt and co-workers [14] citing comparable
constants suffers from unsuitable measurement conditions (the anion concentration
was at least 5 ´ 10 4 mol/l, being more than 30-fold too high following general
considerations (see [15a]; in addition, Bu4NF was used, which is not stable without
crystal water according to [15b]).
The ligand 10, derived from tris(2-aminoethyl)amine, and the ligand 11, derived
from tris(3-aminopropyl)amine, differ substantially in affinity towards anions, in spite
of their quite similar structure. Computer-aided molecular modelling shows that ligand
10 can form a complex with chloride without substantial strain. Preliminary force-field
calculations in the gas phase indicate that a closed conformation of 10 is favored by ca.
90 kJ/mol in comparison to an open conformation, similar to that shown in Fig. 8,
whereas for 11, the open form seems to be favored by ca. 5 kJ/mol. These predictions
are confirmed by the observation of vicinal coupling constants 3J, which remain
constant during the 1H-NMR titration of 10 with chloride, but change from 7.5 to 6.8 Hz

Helvetica Chimica Acta ± Vol. 83 (2000)
471
Fig. 3. Structure of the chloride ´ 10 complex (from CHARMm energy minimization)
in the case of 11. The closed conformation of 10 is also in agreement with the starting
NH shift which is 0.2 ppm higher for 10 than for 11, indicating a stabilization due to
intramolecular H-bonding.
To study the influence of different substituents on the ꢀisocyanate sideꢁ, we
synthesized compounds 12 ± 16. The results of the chloride titrations (Table 7) show
that the nature of the ꢀisocyanateꢁ substituent has little influence on the association
constants. The recently reported K value of 5 ´ 104 l/mol in DMSO for the 16 ´ chloride
complex [16] could not be reproduced in our hands. Deviations due to different water
content can be excluded, as addition of even 5 vol-% of water reduced the constant only
from 360 l/mol to 110 l/mol. The adamantyl derivative 13 was also titrated with chloride
in the solvent CDCl3: the DG value of 26.9 kJ/mol is at the measuring limit of NMR
titrations and showed the expected increase.
Complexation of Nucleotide Anions. ± Ligands containing aromatic units besides
amide-type functions for anion complexation can be expected to provide additional
Table 7. Chloridea) Titrations of Urea Derivatives 12 ± 22
d
Ligand
NH (A)
NH (B)
DGÆ
)
Kb)
CISc)
Kb)
CISc)
12
13
1191
811
43123
1859
1979
363
0.805
0.756
0.825
0.567
0.371
0.646
0.693
971
868
43890
1931
1596
363
0.460
0.512
0.709
0.427
0.642
0.571
0.584
17.6
16.9
26.9
19.0
18.9
14.8
16.9
13 (CDCl3)
14
15
16
22
819
805
a) ± d) See Table 1, except that (D6)DMSO was used instead of CDCl3.

472
Helvetica Chimica Acta ± Vol. 83 (2000)
stacking interactions. Similarly to the halides, the tridentate 10 carrying aryl
substituents showed in (D6)DMSO with nucleotide anions promising K values of ca.
1
3 ´ 104 m (Table 8), with negligible difference between nucleotides since stacking in
solvents like (D6)DMSO can contribute little. With the analogous ligand 11, different
protons in the titration yielded association isotherms which could not be evaluated,
1
estimated Kꢁs derived from the NH shifts of 11 ranging from 1 ´ 102 ± 1 ´ 103 m , these
characteristics being likely due to the formation of several complexes. In contrast to 11,
ligand 10 titrated with AMD2 showed a maximum near x ˆ 0.5 for all three protons in
a Jobꢁs plot (Fig. 4,a; modification of methods developed by Ostromisslensky [17] and
Denison [18]); also a modified ꢀone-species testꢁ, as proposed for spectrophotometric
measurements [19], plotted with a common origin for all lines, clearly demonstrated for
10 a predominant 1:1 complex with the nucleotide anions, with higher complexes
contributing only to a minor extent (Fig. 4,b).
The introduction of pyrene units into urea-derived host compounds 17 ± 20 with the
simultaneous implementation of hydrophilic amino groups was initiated in the hope of
allowing larger stacking effects, measuring also in D2O with better discrimination
between the nucleobases, and of allowing of the highly sensitive fluorescence method
for detection. Ligands like 19 or 20 which were only sparingly soluble in D2O due to the
aromatic units and the absence of an; additional hydrophilic amino group (like in 17
and 18), showed in (D6)DMSO much weaker complexes with the nucleotides than, e.g.,
10. By comparison with NH shifts of other ligands, we could estimate the K values to be
below 10 l/mol.
Table 8. Titrations with Nucleotide Anions
Ligand
Solvent
AMPa)
CMPa)
GMPa)
TMPa)
Kb)
DGd)
Kb)
DGd)
Kb)
DGd)
Kb)
DGd)
NMRa)
20
19
21
17
18
10
12
22
(D6)DMSO
(D6)DMSO
(D6)DMSO
(D6)DMSO
(D6)DMSO
< 10e)
< 10e)
180
220
78
>
>
5.8e)
5.8e)
13.0
< 10e)
< 10e)
80
100
±
>
>
5.8e)
5.8e)
11.0
< 10e)
< 10e)
18
20
±
>
>
5.8e)
< 10e)
>
>
5.8e)
5.8e)
7.2
7.6
±
< 10e)
5.8e)
9.8
50
64
13.6
11
11.6
10.5
±
±
±
±
±
±
36000
±
(D6)DMSO 30000
26.0d) 42000
29000
25.9
(D6)DMSO 25000f)
25.5
±
22000f)
25.2
±
g
g
g
g
g
g
g
g
(D6)DMSO
)
)
)
)
)
)
)
)
Fluorescenceh)
20
H2O
<10i)
<10i)
2680
2310
1205
52
>
>
5.8
5.8
19.9
19.5
17.8
9.9
<10i)
<10i)
2857
2540
396
>
>
5.8
5.8
20.0
19.8
15.1
6.4
<10i)
<10i)
3300
2960
1250
45
>
>
5.8
5.8
20.4
20.1
17.9
9.6
<10i)
<10i)
4120
3750
572
>
>
5.8
5.8
19
17
18
H2O
H2O
H2O
H2O
21.0
20.7
16.0
6.6
Pyrene
Pyrenej) H2O
13
14
a)b)d)e) See Table 1. f) Error >20% due to high association constant. g) No consistent association constants;
values varied more than one order of magnitude for different protons, K ranged from 103 ± 104 l/mol. h) At
pH 6.2 Æ 0.07; no added buffer, no additional salts, no organic solvents; nucleotides as disodium salts; lEx
320 nm, lEm ˆ 373, 384, 393 nm; slit width 10 nm; measurements in triplicate, association constants differed by
less than 10%. i) Calculated from estimated fluorescence changes taken from other measurements. j) From [20],
in H2O with 5 vol-% MeOH; pH 7; 1 mm sodium cacodylate.

Helvetica Chimica Acta ± Vol. 83 (2000)
473
a
b
2
&
^
~
^
Fig. 4. a) Jobꢁs plot ( : NH , : NH , : CH(arom.)), and b) one-species test for the 10 ´ AMP complex (
:
a
b
a
b
&
~
NH vs. NH , : NH vs. CH(arom.), Â : NH vs. CH(arom.), : CH(arom.) vs. CH'(arom.))

474
Helvetica Chimica Acta ± Vol. 83 (2000)
In contrast to these results, the ligands 17 and 18 with only one pyrene unit showed
fairly good complexation in D2O with K values around 3 ´ 103 (Table 8), slightly
increasing in the order A < C < G < T (see Fig. 5 for the structure of AMP2 ´ 17).
Larger differences, remarkably however, in the reverse order A > C > T> G were
observed with ligands 17 and 21 in (D6)DMSO as solvent. While these differences hold
promise for applications, the elucidation of their origin requires further studies. In view
of the rather uniform affinities observed with the pyrene-substituted ligands 17 and 18
in D2O, association with unsubstituted pyrene as ligand1) was also investigated, yielding
free association energies DG which are 2 ± 5 kJ/mol smaller than those obtained with 17
and 18. The relatively small contribution of the urea anion-binding centers in 17 and 18
to the association can be a consequence of the geometric disposition preventing
simultaneous full contact at the stacking and anion-binding sites, in analogy to the
situation found with cyclophane complexes [21]. The distinct upfield CIS of 0.07 ppm
for pyrene signals, as seen in the complex of 17 with AMP2 , indicate that face-to-face
stacking interactions play a major role here.
Fig. 5. Structure of the AMP2 ´ 17 complex (from CHARMm energy minimization)
Experimental Part
General. Unless stated otherwise, all reagents were obtained from commercial sources and used without
further purification. Tris(3-aminopropyl)amine was a kind gift from Dr. R. Hettich [22]. Solvents for reactions
were purified and dried as described in [23], except DMF, which was bought from Fluka (> 99.5%, over 3-Š
molecular sieves). All synthetic operations were carried out in a fume hood under a protective layer of dry N2.
Care should be taken in the case of the isocyanates as they are extremely toxic. Column chromatography (CC):
silica gel. 1H-NMR Measurements: 99.8% deuterated solvents, dried with small amounts of 3-Š molecular
sieves that had been pre-dried by prolonged heating to 480 K at 8 ´ 10 3 mbar; to CDCl3, a tiny amount of anh.
K2CO3 was added to remove traces of HCl; Bruker-400 or -500-MHz spectrometers, FID editing with either the
WINNMR or XWINNMR program provided by Bruker Instruments. Fluorescence measurements: Hitachi-F-
2000 spectrometer. Combustion analyses: Carlo-Erba-Analyzer, model 1106.
1
)
DG Values for association of pyrene and nucleotide anions in water with only 5% of MeOH added are
lower by 8 ± 10 kJ/mol [20].

Helvetica Chimica Acta ± Vol. 83 (2000)
475
Optimized concentrations and steps for NMR titrations were calculated by known equations [24]. An
EXCEL spreadsheet comprising them can be found on our website. Evaluations of isotherms were done with
Sigmaplot Version 4.0. Molecular modelling was done with the CHARMm force field (Version 6.02) and with
the MM ‡ force field in the ꢀbond dipoleꢁ option (Hyperchem 5.02 modified version of MM2 from 1992).
Phthalodiamide (ˆ Benzene-1,2-dicarboxamide; 2) was synthesized according to [25]. N,N'-(Ethane-1,2-
diyl)-N,N'-(Propane-1,3-diyl), and N,N'(Butane-1,4-diyl)bis[acetamide] (3, 5, and, resp. 6) were synthesized
from the corresponding diamines and acetic acid anhydride (50% excess) according to [26][27][29], resp.
N,N',N''-(Nitrilotriethone-2,1-diyl)tris[acetamide] (7) was obtained according to [28].
N,N'-Methylenebis[benzamide] (4) was synthesized according to [29], with trioxane instead of form-
aldehyde. M.p. 2198 ([29]: 2188).1H-NMR ((D6)DMSO): 4.891 (t, 2 H); 9.078 (t, 4 H); 7.915 (dd, 4 H); 7.486 ±
7.449 (m, 4 H); 7.534 (t, 2 H). 13C-NMR ((D6)DMSO): 45.19; 166.49; 133.96; 128.21; 127.39; 131.36.
N,N',N''-{Nitrilotris[ethane-2,1-diylimino(2-oxoethane-2,1-diyl)]}tris[carbamic Acid] Tribenzyl Ester (8).
To a soln. of N-[(benzyloxy)carbonyl]glycine (6.28 g, 30 mmol) in DMF (100 ml), Bu3N (5.56 g, 30 mmol;
stored over solid KOH) was added. After 15 min, the soln. was cooled to 108 and 3.26 g of carbonochloridic acid
ethyl ester in DMF (10 ml) was carefully added at 108. After 30 min stirring at 108, a soln. of tren (ˆtris(2-
aminoethyl)amine; 1.46 g) in DMF (10 ml) was added at once. After stirring overnight at r.t., the cloudy soln.
was filtered through a layer of Celite and was washed with an additional portion of DMF (20 ml). The combined
solns. were evaporated at r.t. The residue was taken up with CHCl3, washed with H2O, dried (MgSO4), and
purified by CC (CHCl3/MeOH/NH3 100 :10 :1): 3.74 g (90%) of 8. Colorless viscous oil which was used without
further purification. 1H-NMR (CDCl3): 2.438 (br., 6 H); 3.178 (br. 6 H); 3.798 (d, 6 H); 5.04187 (s); 7.244 ± 7.313
(m, 15 H); 6.237 (d, 3 H). 13C-NMR (CDCl3): 37.720; 44.456; 157.023; 54.494; 170.165; 67.004; 136.255; 127.926 ±
128.452.
Tris[ureas] 10 ± 16: General Procedure. The tris[ureas] were obtained from the corresponding amine and
the iso(thio)cyanate as described for 10. To a soln. of tren (1.46 g; 10 mmol) in CH2Cl2 (100 ml), a soln. of 4-
chlorophenyl isocyanate (4.61 g, 30 mmol) in CH2Cl2 (100 ml) was added slowly and dropwise. After some
seconds (R ˆ 4-Cl-C6H4) to some hours (R ˆ Bu), the corresponding urea precipitated. The mixture was stirred
overnight. Then the mother liquor was removed and the residue crystallized from the appropriate solvents.
N',N''',N'''''-Tris(4-chlorophenyl)-N,N'',N''''-(nitrilotriethane-2,1-diyl)tris[urea] (10): 4.5 g (90%). M.p. 2408
(AcOH/MeOH). 1H-NMR ((D6)DMSO): 2.583 (t, 6 H); 3.175 (m, 6 H); 6.187 (t, 3 H); 8.632 (s, 3 H); 7.386
(d, 6 H); 7.217 (dd, H). 13C-NMR ((D6)DMSO): 37.011; 53.284; 154.582; 138.898; 127.800; 118.659; 123.933.
Anal. calc. for C27H30N7O3 (500.580): C 53.43, H 4.98, N 16.15; found: C 53.36, H 5.09, N 15.94.
N',N''',N'''''-Tris(4-chlorophenyl)-N,N'',N''''-(nitrilotripropane-3,1-diyl)tris[urea] (11): 4.5 g (90%). M.p.
2208 (AcOH/i-PrOH). 1H-NMR ((D6)DMSO): 2.402 (t, 6 H); 1.573 (t, 6 H); 3.108 (t, 6 H); 6.185 (t, 3 H); 8.463
(s, 3 H); 7.373 (dd, 6 H); 7.211 (dd, 6 H). 13C-NMR ((D6)DMSO): 37.340; 27.177; 50.938; 154.962; 39.332;
128.158; 119.144; 124.339. Anal. calc. for C30H36N7O3 (542.66): C 55.52, H 5.59, N 15.11; found: C 55.15, H 5.71,
N 15.08.
N',N''',N'''''-Tri(naphthalen-1-yl)-N,N'',N''''-(nitrilotriethane-2,1-diyl)tris[urea] (12): 5.88 g (90%). M.p.
2428 (AcOH/i-PrOH). 1H-NMR ((D6)DMSO): 6.699 (t, 3 H); 8.611 (s, 3 H); 2.698 (t, 6 H); 3.309 (dt, 6 H);
8.056 (d, 3 H); 7.929 (d, 3 H); 7.838 (d, 3 H); 7.313 ± 7.519 (m, 9 H). 13C-NMR ((D6)DMSO): 37.911; 54.385;
155.994; 117.224; 135.055; 133.723; 121.514 ± 128.240. Anal. calc. for C39H39N7O3 (653.783): C 71.65, H 7.48,
N 12.17; found: C 72.93, H 7.36, N 12.47.
N',N''',N'''''-Tri(adamantan-1-yl)-N,N'',N''''-(nitrilotriethane-2,1-diyl)tris[urea] (13): 3.63 g (60%). M.p.
1
1838 (CHCl3/cyclohexane), after CC (CHCl3/MeOH 95 :5). H-NMR (CDCl3): 2.448 (t, 6 H); 3.099 (m, 6 H);
5.085 (s, 3 H); 5.903 (br., 3 H); 1.979 (br., 18 H); 2.046 (br., 9 H); 1.665 (br., 18 H). 13C-NMR (CDCl3): 56.290;
50.684; 158.576; 38.594; 42.692; 36.557; 29.667. Anal. calc. for C33H63N7O3 (605.907): C 69.09, H 9.37, N 14.46;
found: C 68.80, H 9.37, N 14.24.
N',N''',N'''''-Tris(4-nitrophenyl)-N,N'',N''''-(nitrilotriethane-2,1-diyl)tris[urea] (14): 5.74 g (90%). M.p.
>1838 (dec.; acetone/MeCN). 1H-NMR ((D6)DMSO): 2.646 (t, 6 H); 3.236 (m, 6 H); 6.426 (t, 3 H); 7.589
(dd, 6 H); 8.088 (dd, 6 H); 9.333 (s, 3 H). 13C-NMR ((D6)DMSO): 37.561; 53.467; 116.727; 124.973; 140.279;
147.087; 154.427. Anal. calc. for C27H30N10O9 (638.589): C 50.78, H 4.74, N 21.93; found: C 50.80, H 4.57, N 21.70.
N',N''',N'''''-Tris(4-nitrophenyl)-N,N'',N''''-(nitrilotriethane-2,1-diyl)tris[thiourea] (15): 6.18 g (90%). M.p.
1
>1858 (dec.; acetone/MeCN). H-NMR ((D6)DMSO): 2.815 (t, 6 H); 3.2756 (m, 6 H); 7.775 (dd, 6 H); 8.130
(dd, 6 H); 10.1581 (s, 3 H). 13C-NMR ((D6)DMSO): 41.802; 51.522; 120.282; 124.450; 141.681; 146.160; 179.844.
Anal. calc. for C27H30N10O6S3 (686.789): C 47.22, H 4.40, N 20.39; found: C 47.80, H 4.57, N 20.50.
N',N''',N'''''-Tributyl-N,N'',N''''-(nitrilotriethane-2,1-diyl)tris[urea] (16): 3.99 g (90%). M.p. 1628 (nitro-
methane). 1H-NMR ((D6)DMSO): 0.851 (t, 9 H); 1.251 (m, 6 H); 1.323 (m, 6 H); 1.255 (t, 6 H); 2.411 (t, 3 H);

476
Helvetica Chimica Acta ± Vol. 83 (2000)
2.966 (m, 6 H); 2.991 (m, 6 H). 13C-NMR ((D6)DMSO): 13.652; 19.523; 32.124; 37.737; 38.933; 54.409; 158.211.
Anal. calc. for C21H45N7O3 (443.627): C 56.86, H 10.22, N 22.10; found: C 56.76, H 10.57, N 22.40.
g-Oxopyrene-1-butanoic Acid. According to [30], with necessary modifications: At 08, AlCl3 (46.2 g,
228 mmol) was slowly added to a well-stirred soln. of succinic anhydride (17.5 g, 227 mmol) and pyrene (46.2 g,
228 mmol) in dry nitromethane (300 ml). The soln. turned red, and a precipitate was formed slowly. One hour
after all AlCl3 had been added, the mixture was allowed to reach r.t. until HCl evolution ceased. The resulting
red soln. was quenched with ice/HCl (1 l). The slurry was extracted with CH2Cl2 (4 Â 200 ml) and the extract
dried (MgSO4) and evaporated. The resulting acid was transformed to the sodium salt by means of a 10%
ethanolic NaOH soln. The precipitating salt was crystallized from H2O (3Â) and then hydrolyzed in the warmth
with HCl soln. and the acid recrystallized from AcOH: 18 g (26%). M.p. 1848 ([30]: 1728; [31]: 1848). 1H-NMR
((D6)DMSO): 2.758 (t, 2 H); 3.472 (t, 2 H); 8.090 ± 8.761 (m, 9 H); 12.23 (s, 1 H); 13C-NMR ((D6)DMSO):
28.462; 36.809; 123.373 ± 132.923; 173.844; 203.046.
Pyrene-1-butanoic Acid. According to [30], but with necessary modifications: A soln. of g-oxopyrene-1-
butanoic acid (12.09 g, 0.04 mol) in triethylene glycol (150 ml) and hydrazine hydrate (100 ml) was slowly
heated to 1008 and kept at 1008 for 1 h. Then, the soln. was cooled to r.t., and KOH (20 g) was added. The
mixture was heated 3 h to 1408 (oil bath) until gas evolution ceased. Then, the H2O/hydrazine mixture was
distilled until the boiling temp. reached 1508. Then, the mixture was cooled to r.t. and carefully acidified. The
precipitate was recrystallized from AcOH/AcOEt: 10 g (87%). M.p. 189 ± 1908 ([30]: 1668; [32]: 1908).
1H-NMR ((D6)DMSO): 2.009 (m, 2 H); 2.384 (t, 2 H); 3.3201 (t, 2 H); 7.890 ± 8.376 (m, 9 H); 11.93 (s, 1 H).
13C-NMR ((D6)DMSO): 26.795; 31.980; 123.379 ± 136.290; 174.355.
2
Acyl Azides for Curtius Rearrangement: General Procedure ). A soln. of Et3N (1.44 g, 14 mmol) in acetone
(10 ml) was added dropwise to a soln. of the corresponding carboxylic acid (12 mmol) in acetone (20 ml) at 08.
After 10 min at 08, a soln. of carbonochloridic acid ethyl ester (1.67 g, 15.4 mmol) was added at once. Then, the
mixture was stirred at 08 for another 30 min. Then, a soln. of NaN3 (1.18 g, 18 mmol) in H2O (5 ml) was added at
once. In case of pyrene-containing azides, the product precipitated within 15 min and was dried over P4O10
overnight at 9 mbar at r.t., the dried azides having a sufficient purity for synthesis. In case of benzenebutanoic
acid [34] the mixture was extracted after 15 min with Et2O (3 Â 30 ml), dried (MgSO4) and evaporated at r.t.
The residue was mixed with toluene (20 ml) and heated to reflux until N2 evolution ceased (caution: foaming
may occur). The isocyanate was distilled (1568/10 ± 12 mbar) and could be stored for prolonged time under N2.
Aryl-Containing Ureas. The acyl azides were suspended in toluene and heated to reflux until N2 evolution
ceased. The ureas were obtained by addition of stoichiometric amounts of the corresponding amines to this
isocyanate soln. at r.t.
N-[2-(Dimethylamino)ethyl]-N'-[3-(pyren-3-yl)propyl]urea (17): 3.59 g (80%). M.p. >1858 (dec.; tol-
uene). 1H-NMR ((D6)DMSO): 1.881 (m, 2 H); 2.171 (s, 6 H); 2.302 (t, 2 H); 3.125 (m, 4 H); 3.323 (m, 2 H);
5.764 (t, 3 H); 6.183 (t, 3 H); 7.953 ± 8.359 (m, 27 H). 13C-NMR ((D6)DMSO): 29.922; 32.151; 37.147; 39.039;
44.905; 58.823; 123.183 ± 127.235; 157.972. Anal. calc. for C27H30N10O6S3 (373.491): C 77.18, H 7.29, N 11.25;
found: C 77.28, H 7.47, N 11.40.
N-[2-(Dimethylamino)ethyl]-N'-[3-oxo-3-(pyren-3-yl)propyl]urea (18): 3.71 g (80%). M.p. 1608 (toluene).
1H-NMR ((D6)DMSO): 2.072 (s, 2 H); 2.201 (t, 2 H); 3.074 (m, 2 H); 3.396 (t, 2 H); 3.503 (m, 3 H); 5.911
(t, 3 H); 6.234 (t, 3 H); 8.099 ± 8.829 (m, 27 H). 13C-NMR ((D6)DMSO): 35.378; 37.254; 42.788; 45.046; 58.949;
123.415 ± 133.109; 158.029; 203.610. Anal. calc. for C24H25N3O2: C 74.39, H 6.50, N 10.84; found: C 74.60, H 6.57,
N 10.70.
N',N''',N'''''-Tris[3-oxo-3-(pyren-1-yl)propyl]-N,N'',N''''-(nitrilotriethan-2,1-diyl)tris[urea] (19): 2.92 g
(70%). M.p. 1798 (dec.; DMF/toluene). 1H-NMR ((D6)DMSO): 2.449 (m, 6 H); 3.091 (m, 6 H); 3.391
(m, 6 H); 3.536 (m, 6 H); 6.147 (t, 3 H); 6.293 (t, 3 H); 7.974 ± 8.778 (m, 27 H). 13C-NMR ((D6)DMSO): 30.683;
35.353; 35.690; 38.943; 42.667; 54.309; 123.265 ± 132.973; 158.234; 162.235; 203.453. Anal. calc. for C66H57N7O6
(1044.202): C 75.91, H 5.50, N 9.39; found: C 75.80, H 5.57, N 9.45.
N',N''',N'''''-Tris[3-(pyren-1-yl)propyl]-N,N'',N''''-(nitrilotriethan-2,1-diyl)tris[urea] (20): 2.8 g (70%). M.p.
1388 (methylcellosolve). 1H-NMR ((D6)DMSO): 1.779 (m, 6 H); 2.471 (m, 6 H); 3.072 (m, 6 H); 3.169 (t, 6 H);
5.909 (t, 3 H); 6.094 (t, 3 H); 7.746 ± 8.172 (m, 27 H). 13C-NMR ((D6)DMSO): 30.060; 32.288; 37.921; 40.413;
54.568; 123.260 ± 136.449; 158.430. Anal. calc. for C66H63N7O3 (1002.252): C 79.09, H 6.34, N 9.78; found:
C 79.08, H 6.57, N 9.70.
2
)
Analogously to a procedure described in [33a]. The mixed-anhydride reaction gave purer compounds than
the diphenylphosphoryl-azide method for the preparation of the pyren-1-yl isocyanate described in [33b].

Helvetica Chimica Acta ± Vol. 83 (2000)
477
N-[2-(Dimethylamino)ethyl]-N'-(3-phenylpropyl)urea (21): 2.10 g (70%). M.p. 1068 (acetone). 1H-NMR
((D6)DMSO): 1.797 (m, 2 H); 1.993 (t, 2 H); 2.272 (s, 6 H); 2.504 (t, 2 H); 2.589 (t, 2 H); 2.841 (t, 2 H); 7.136 ±
7.257 (m, 5 H); ca. 7.8 (br., 1 H); ca. 10.7 (br., 1 H). 13C-NMR (CD3OD): 32.955; 34.116; 36.783; 40.925; 43.914;
60.030; 72.596; 126.924; 129.436; 143.074; 161.527. Anal. calc. for C14H23N3O (249.352): C 67.43, H 9.30, N 16.85;
found: C 67.63, H 9.57, N 16.70.
N',N''',N'''''-Tris(3-phenylpropyl)-N,N'',N''''-(nitrilotriethan-2,1-diyl)tris[urea] (22): 1.76 g (70%). M.p. 1368
(acetone). 1H-NMR ((D6)DMSO): 1.637 (m, 6 H); 2.436 (m, 6 H); 2.531 (t, 6 H); 2.976 (m, 6 H); 3.021
(m, 3 H); 5.831 (t, 3 H); 5.971 (t, 3 H); 7.128 ± 7.256 (m, 15 H). 13C-NMR ((D6)DMSO): 31.810; 32.481; 37.764;
38.882; 54.415; 125.610; 128.197; 141.759; 158.211. Anal. calc. for C66H63N7O3 (629.835): C 68.65, H 8.16,
N 15.57; found: C 68.70, H 8.17, N 15.70.
This work was supported by the Deutsche Forschungsgemeinschaft, Bonn, and the Fonds der Chemischen
Industrie, Frankfurt.
This paper is Part 95 of our series ꢀSupramolecular Chemistryꢁ. Part 94: [35].
REFERENCES
[1] M. M. G. Antonisse, D. N. Reinhoudt, Chem. Commun. 1998, 443.
[2] R. C. Cameron, S. J. Loeb, Chem. Commun. 1997, 573.
[3] H. Luecke, F. A. Quiocho, Nature (London) 1990, 347, 402, J. J. He, F. A. Quiocho, Science (Washington,
D.C.) 1991, 251, 1497.
[4] F. A. Quiocho, J. W. Pflugrath, Nature (London) 1985, 314, 257; J. Mol. Biol. 1988, 200, 163.
[5] P. Schwinte, R. Darcy, F. OꢁKeeffe, J. Chem. Soc., Perkin Trans. 2 1998, 805; Y. Shi Youheng, H.-J.
Schneider, J. Chem. Soc., Perkin Trans. 2 1999, in press; D. A. Nation, Q. Lu, A. E. Martell, Inorg. Chim.
Acta 1997, 263, 209; J. A. Aguilar, E. Garcia-Espana, J. A. Guerrero, S. V. Luis, Inorg. Chim. Acta 1996, 246,
287; K. G. Ragunathan, H.-J. Schneider, J. Chem. Soc., Perkin Trans. 2 1996, 2597; V. Kral, A. Andrievsky,
J. L. Sessler, Chem. Commun. 1995, 2349; S. M. Lacy, D. M. Rudkevich, W. Verboom, D. N. Reinhoudt,
Tetrahedron Lett. 1994, 35, 5953; A. V. Eliseev, H.-J. Schneider, J. Am. Chem. Soc. 1994, 116, 6081; J.
Rebek, Jr., Acc. Chem. Res. 1990, 23, 399; A. D. Hamilton, in ꢀBioorg. Chem. Frontiersꢁ, Ed. H. Dugas,
Springer, Berlin-Heidelberg, 1991, Vol. 2, p. 115.
[6] D. N. Reinhoudt, J. Scheerder, J. F. J. Engbersen, Rec. Trav. Chim. Pays-Bas 1996, 115, 307.
Â
[7] J. M. Coteron, F. Hacket, H.-J. Schneider, J. Org. Chem. 1996, 61, 1429.
[8] K. Vavallieratos, S. R. de Gala, D. J. Austin, R. H. Crabtree, J. Am. Chem. Soc. 1997, 119, 2325.
[9] F. Eblinger, H.-J. Schneider, Angew. Chem. 1998, 110, 821; Angew. Chem., Int. Ed. 1998, 37, 826.
[10] D. N. Reinhoudt, S. Valiyaveettil, J. F. J. Engbersen, W. Verboom, Angew. Chem. 1993, 105, 942; D. N.
Reinhoudt, J. Scheerder, M. Fochi, J. F. J. Engbersen, J. Org. Chem. 1994, 59, 7815.
[11] R. Jairam, P. G. Potvin, J. Org. Chem. 1992, 57, 4136.
[12] D. Hamilton, J. Am. Chem. Soc. 1993, 115, 369, and lit. cit. therein.
[13] V. Alcazar, Chem. Lett. 1995, 759.
[14] H. Baerrigter, L. Grave, J. W. M. Nissink, L. A. J. Chrisstoffels, J. H. van der Maas, W. Verboom, F. de Jong,
D. N. Reinhoudt, J. Org. Chem. 1998, 63, 4174.
[15] a) G. Carta, G. Crisponi, V. Nurchi, Tetrahedron 1981, 37, 2115; D. A. Deranleau, J. Am. Chem. Soc. 1969,
91, 4044; b) J. L. Fry, R. K. Sharma, J. Org. Chem. 1983, 48, 2112.
[16] Kyu-Sung Jeon, Kyu-Myoung Hahn, Young Lag Cho, Tetrahedron Lett. 1998, 39, 3779.
[17] I. Osromisslensky, Chem. Ber. 1911, 44, 268.
[18] R. B. Denison, Trans Farad. Soc. 1921, 8, 20.
[19] F. R. Hartley, C. A. Burgess, R. M. Alcock, ꢀSolution Equilibriaꢁ, Ellis Horwood Ltd., Chichester 1980,
p. 36 ff.
[20] P. Lianos, S. Georghiou, Photochem. Photobiol. 1979, 29, 13.
[21] H.-J. Schneider, I. Theis, Angew. Chem. 1989, 101, 757; Angew. Chem. Int. Ed. 1989, 28, 753.
[22] R. Hettich, Dissertation, University of the Saarland, 1997.
[23] D. D. Perrin, W. L. F. Armarego, D. R. Perrin, ꢀPurification of Laboratory Chemicalsꢁ, Elsevier Science,
1980, and lit. cit. therein.
[24] J. Granot, J. Magn. Res. 1983, 55, 216.
[25] L. J. Spiessens, M. Anteunis, Bull Soc. Chim. Belg. 1983, 92, 1; S. Wolfe, K. S. Hasan, Can. J. Chem. 1970, 48,
3577.

478
Helvetica Chimica Acta ± Vol. 83 (2000)
[26] B. Strache, Chem. Ber. 1923, 21, 2359.
[27] M. Haga, W. Wesser, Chem. Ber. 1938, 36, 338.
[28] J. S. Bradshaw, K. E. Krakowiak, R. M. Izatt, J. Org. Chem. 1990, 55, 3364.
[29] A. W. Hofmann, Chem. Ber. 1923, 21, 2332.
[30] O. Perin-Roussel, N. P. Buu-Hoi, Do-Phuoc-Hien, Bull. Soc. Chim. Fr. 1971, 6, 2160.
[31] E. Buchta, P. Vincke, Chem. Ber. 1965, 98, 208.
[32] E. Fieser, A. Fieser, G. Hershberg, J. Am. Chem. Soc. 1936, 58, 1463.
[33] a) J. Weinstock, J. Am. Chem. Soc. 1961, 26, 3511; b) G. Read, N. R. Richardson, J. Chem. Soc., Perkin
Trans. 1 1996, 167.
[34] P. F. Dicks, S. A. Glover, A. Goosen, C. W. McCleland, Tetrahedron 1987, 923.
[35] M. Sirish, H.-J. Schneider, Chem. Commun. 2000, 23.
Received September 30, 1999