Selective synthesis of non-symmetrical bis-ureas and their self-assembly

Article abstract of DOI:10.1039/b316913h

Symmetrical bis-ureas composed of two urea functions linked together by a toluene ring were previously shown to form long supramolecular polymers and thus highly visco-elastic solutions, thanks to cooperative self-assembly involving four hydrogen bonds. I

Full text of DOI:10.1039/b316913h

View Article Online / Journal Homepage / Table of Contents for this issue
P A P E R
Selective synthesis of non-symmetrical bis-ureas and their
self-assemblyw
Olivier Colombani and Laurent Bouteiller*
er
`
´
´
Laboratoire de Chimie des Polymeres, Universite Paris VI, Tour 44-54, 1 etage, case courrier
185, 4, Place Jussieu, 75252 Paris cedex 05, France. E-mail: bouteil@ccr.jussieu.fr;
Fax: +33 (0)1 44 27 70 89; Tel: +33 (0)1 44 27 73 78
Received (in Toulouse) 23rd December 2003, Accepted 25th June 2004
First published as an Advance Article on the web 19th October 2004
Symmetrical bis-ureas composed of two urea functions linked together by a toluene ring were previously
shown to form long supramolecular polymers and thus highly visco-elastic solutions, thanks to cooperative
self-assembly involving four hydrogen bonds. In this paper, we report the direct and selective synthesis of
non-symmetrical bis-ureas. Mono-isocyanate/mono-ureas were first obtained from a one-step selective
reaction between one aromatic amine and one isocyanate function of 2,4-toluene diisocyanate. Then, non-
symmetrical bis-ureas, tetra-ureas, and bis-urea functional polydimethylsiloxanes (PDMS) were obtained by
reacting the mono-isocyanate/mono-ureas with well chosen amines. The chloroform solutions of these
compounds were characterised by quantitative FTIR spectroscopy and viscosimetry. It was shown that non-
symmetrical bis-ureas substituted on one side by an aromatic moiety and on the other by an aliphatic group
combine the solubility of aliphatic bis-ureas and the strong association of aromatic ones. Moreover, the
association of bis-ureas grafted onto polydimethylsiloxanes is efficient and leads to the physical cross-linking
of these polymers, even in chloroform.
According to the literature, the ortho isocyanate function of
Introduction
2,4-TDI is slightly less reactive than the para one.7–9 Moreover,
when one of the two isocyanates is converted to a urea or a
urethane function, the reactivity of the remaining isocyanate
significantly decreases.9,10 Thus, mono-isocyanate/mono-ur-
ethane may be obtained by reacting an alcohol with an excess
of 2,4-TDI.10 However, amines being more reactive than
alcohols, the selectivity of the reaction with 2,4-TDI is expected
to decrease. In fact, no detailed selective synthesis of mono-
isocyanate/mono-urea based on 2,4-TDI has been reported yet,
even though clues have been given by some authors.8,11 This
paper presents such a selective synthesis with aromatic amines,
which are less reactive than aliphatic ones.7
Supramolecular polymers result from the self-assembly of
small molecules, which are held together by weak interactions
and lead to polymer-like structures.1–4 These systems are
increasingly studied because of the reversible character of the
process through various stimuli, including temperature, solvent
and concentration.
Among these systems, we have focused on bis-ureas (Scheme
1), whose two urea functions are linked together by a rigid
spacer. Some of these molecules dissolve spontaneously at
room temperature in apolar solvents and lead to visco-elastic
solutions at concentrations as low as 1% by weight. This is due
to their self-assembly into long and rigid wires, by four strong
cooperative hydrogen bonds.5 A range of symmetrical bis-
ureasz has previously been synthesised by reacting 1 equiv. of
2,4-toluene diisocyanate (2,4-TDI) with 2 equiv. of a primary
amine.6 Up to now, no dissymetrical bis-ureas based on 2,4-
TDI have been obtained.
Results and discussion
1. Synthesis of mono-isocyanate/mono-ureas
Analysis. The unselective reaction of equimolar amounts of
2,4-TDI and 4-n-butylaniline (BuA) leads to a mixture of 4
products (Scheme 6). The moisture sensitivity of these pro-
ducts, caused by the remaining isocyanate function, compli-
cates their analysis. Thus, the product of the first step was
reacted with an excess of 2-ethylhexylamine (EHA). After this
On the other hand, bis-urea grafted polymers may form self-
assembled materials with unique rheological properties, both
in bulk and solution. Furthermore, multifunctional com-
pounds such as tetra-ureas can be expected to show extremely
efficient association.
To obtain these new molecules, the selective reaction of an
amine with only one isocyanate function of 2,4-TDI is required
(Scheme 2). The previously cited products may then be ob-
tained by reacting the remaining isocyanate function of the
mono-adduct with a well-chosen amino-functional molecule
(Schemes 3–5).
w Electronic supplementary information (ESI) available: NMR spectra
in the 2 ppm region for the products, before and after recrystallisation,
from runs OC105 and OC072, having different product distributions.
See http://www.rsc.org/suppdata/nj/b3/b316913h/
z The word symmetrical refers to R1 and R2 substituents and neglects
the presence of the methyl group on the central ring.
Scheme 1 General formula of bis-ureas.
T h i s j o u r n a l i s & T h e R o y a l S o c i e t y o f C h e m i s t r y a n d t h e
C e n t r e N a t i o n a l d e l a R e c h e r c h e S c i e n t i f i q u e 2 0 0 4
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 3 7 3 – 1 3 8 2
1373

View Article Online
Scheme 2 Synthesis of mono-isocyanate/mono-ureas.
second step, the final mixture is free of isocyanate functions
1
and can be analysed directly by H NMR.
first step was investigated. Three solvents were tested (THF,
heptane, dichloromethane) with 2,4-TDI : BuA ratios varying
from 1 : 1 to 5 : 1 (Fig. 2). In heptane or dichloromethane, 2
precipitates during the first step, which makes it possible to
remove the excess of 2,4-TDI by filtration. In THF however,
the mono-isocyanate/mono-ureas are soluble and the excess of
2,4-TDI was not removed. The composition of the final
product was evaluated for all runs (Table 1).
According to Table 1 and Fig. 2, the reaction is not selective
in stoichiometric conditions since small amounts of 16 are
detected at the end of the second step for OC103, OC134 and
OC108. This is caused by the fact that the 2 : 2,4-TDI ratio
becomes very high at the end of the reaction, and thus the
probability for BuA to react with 2 and form 16 becomes
significant.10
The composition of the mixture after the first step can then
be deduced from the analysis of the crude product of the
second step. Indeed, according to Scheme 6, the proportions
of 5a, 5b, 1 and 16 in the second step correspond to the
amounts of 2a, 2b, 2,4-TDI, and 16, respectively, in the first
step. Moreover, the four bis-ureas potentially obtained at the
end of the second step are easily identified by 1H NMR (spectra
a–d in Fig. 1) even for proportions of symmetrical bis-ureas as
low as 5 mol % (spectra e and f in Fig. 1). Finally, deconvolu-
tion of the 4 peaks, corresponding to the four different bis-
ureas, gives an accurate determination of the amount of each
bis-urea. Thus, the selective formation of mono-isocyanate/
mono-urea from 2,4-TDI and BuA can be accurately and easily
evaluated by this two-step method.
Moreover, several differences are observed when the solvent
is changed. First, almost no 1 can be detected when the first
step is conducted in heptane or dichloromethane. Indeed, 2 is
not soluble in these solvents, unlike 2,4-TDI. Thus, the pro-
ducts of the first step are filtered at the end of the reaction, and
Influence of solvent and stoichiometry. The influence of the
solvent and of the 2,4-TDI : BuA ratio on the selectivity of the
Scheme 3 Synthesis of non-symmetrical bis-ureas.
1374
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 3 7 3 – 1 3 8 2

View Article Online
Scheme 4 Synthesis of tetra-ureas.
the remaining 2,4-TDI is eliminated before adding EHA. On
the contrary, 2 is soluble in THF and no purification is done
before the second step. The remaining 2,4-TDI yields 1 during
the second step.
Furthermore, the selectivity is slightly lower in dichloro-
methane than in THF or heptane (see the amount of 16 for
each run). This observation may be attributed to the formation
of a more viscous reaction mixture in dichloromethane than in
THF or heptane, due to the formation of a thick precipitate.
Thus, the poorer selectivity could result from inhomogeneity of
the reaction medium.
Finally, the 5a : 5b ratio is much greater than 1, which
confirms the difference of reactivity between the para and
ortho isocyanate functions.8 This ratio depends on the solvent
in the order dichloromethane4heptane4THF. This is prob-
ably caused by a difference of solubility between 2b and 2a
rather than by a difference of reactivity of the two isocyanate
functions of 2,4-TDI induced by the solvent. Indeed, 2b may be
Scheme 5 Synthesis of PDMS grafted bis-ureas.
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 3 7 3 – 1 3 8 2
1375

View Article Online
Scheme 6 The four possible products of the reaction of 2,4-TDI with BuA in the first step and EHA in the second step.
more soluble than 2a and some of it may be eliminated during
the filtration in heptane or dichloromethane, but not in THF.
This hypothesis is in agreement with the lower yield of the first
step in dichloromethane than in heptane.
With an excess of 2,4-TDI, the selectivity of the reaction is
improved according to the decreasing amount of 16. Indeed,
the excess of 2,4-TDI keeps the 2 : 2,4-TDI ratio low through-
Fig. 1 1H NMR spectra of the possible products of the reaction
described in Scheme 6. Non-selective synthesis as run OC108 (a);
symmetrical bis-ureas 16 (b) and 1 (c); non-symmetrical bis-urea 5a
(d); intentional mixtures of 5a + 5% 16 (e), and 5a + 5% of 1 (f). The
peak at ca. 2.1. ppm corresponds to the methyl group on the aromatic
ring.
Fig. 2 1H NMR spectra of the crude product of the second step
(Scheme 6), for different reaction conditions of the first step. See Table
1 for the experimental conditions.
1376
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 3 7 3 – 1 3 8 2

View Article Online
Table 1 Evaluation of the selectivity of the synthesis of 2
Run
Solvent
2,4-TDI : BuA
% Yielda
% 16b
% 5bb
% 5ab
% 1b
5a/5b
OC103
OC105
OC116
OC127
OC084
OC134
OC135
OC072
OC108
OC109
a
Heptane
Heptane
Heptane
Heptane
Heptane
CH2Cl2
CH2Cl2
CH2Cl2
THF
1
92
94
6
1
11
15
14
8
83
83
o1
1
8
6
1.5
3
95
92
o1
o1
o1
14
5
86
92
o1
o1
o1
o1
2
6
3
12
8
5
97
67
11
2
89
84
1
35
26
497
3
1.5
5
70
3
90
73
o1
5
o1
19
14
497
59
o1
17
1
1.5
n.a.c
n.a.c
THF
1
48
37
3
b
Product yield for the first step (obtained weight divided by weight of 2 expected if the reaction were fully selective). The amounts of 16, 5b, 5a
c
and 1 at the end of the second step correspond to those of 16, 2b, 2a and 2,4-TDI, respectively, at the end of the first step. Not available because the
product of the first step was not purified.
out the reaction and thus favours the selectivity. Here again,
the lower selectivity in dichloromethane is confirmed. Indeed,
when 1.5 equiv. of 2,4-TDI is used in the first step, almost no
16 is detected in heptane (OC105) or THF (OC109), whereas a
significant proportion is formed in dichloromethane (OC135).
Nevertheless, full selectivity is reached in heptane and dichlor-
omethane with 3 or 5 equiv. of 2,4-TDI (OC116, OC084,
OC072).y
In conclusion, selectivity of the first step is easily reached in
either solvent with an excess of 2,4-TDI. Thus, 2 free of 16 can
be obtained. Moreover, the use of dichloromethane or heptane
is recommended to get rid of the unreacted 2,4-TDI at the end
of the reaction. The choice between heptane or dichloro-
methane depends on the desired mono-isocyanate/mono-urea.
Indeed, heptane should be preferred to obtain 2 (mixture of
isomers) in yields higher than 90%, whereas dichloromethane
is more indicated to obtain almost pure 2a, but with slightly
lower yields (B70%).
reaction of this new mono-adduct with EHA and recrystallisa-
tion of the final product yielded pure 6a (Scheme 3).
Finally, N-methylaniline, a secondary amine, was used to
synthesise a new type of mono-adduct 4 (Scheme 2). The
synthesis was performed with 5 equiv. of 2,4-TDI in heptane
at 0 1C. Then 4 was either reacted with EHA to obtain 7 or
with an amino-functional PDMS to yield the PDMS-g-bis-urea
15 (Schemes 3 and 5). As the purification of 4 was more difficult
than that of the other mono-adducts, a small quantity of 2,4-
TDI remained at the end of the first step. Pure 7 was obtained
after column chromatography, precipitation, and recrystallisa-
tion. The polymer grafted with 4 underwent limited chain
extension reactions because of the presence of traces of 2,4-
TDI,12 but its molecular weight increased only moderately and
it was not chemically cross-linked.
4. Characterisation in solution
The solubility of all synthesised molecules was first tested
(Table 2).
2. Synthesis of non-symmetrical bis-ureas and bis-urea
functional molecules
The aliphatic tetra-ureas 8 to 11 are hardly soluble, even in
polar solvents such as THF. The presence of branching in the
spacer of 10 does not improve the solubility significantly.
Consequently, these compounds were not studied further.
The amino-functional PDMS are viscous oils, whereas the
bis-urea modified PDMS 12 to 15 are rubbery solids. However,
this dramatic change of behaviour is only caused by a physical
cross-linking of the material. Indeed, the PDMS modified by
bis-ureas are soluble in THF (a solvent of the polymer back-
bone and a very good hydrogen-bond competitor as well),
which confirms that they are not chemically cross-linked. On
the contrary, 12 and 14, which contain a large amount of bis-
urea, swell in heptane, toluene or chloroform, as if they were
cross-linked. Indeed, these solvents of the polymer backbone
are too weak hydrogen-bond competitors to prevent the self-
assembly of bis-ureas into a physical network. As already
reported,13 the physical cross-linking of PDMS by bis-ureas
improves dramatically the tensile properties of these materials.
The properties of solutions of 13 in chloroform are described
below.
Finally, 5a and 7 are very soluble at room temperature.
Moreover, 5a forms a gel in toluene at only 1 wt %. The self-
assembly of these molecules was characteriszed by FTIR
spectroscopy and viscosimetry and compared to that of a
model symmetrical bis-urea (1) whose properties were already
reported.5,6 Bis-urea 1 is a good comparative model since it
differs from 5a by only one moiety (Schemes 1 and 3).
Compound 6 has poor solubility and was not studied further.
As already reported,6 FTIR spectroscopy in chloroform is a
powerful tool to quantify the self-association of bis-ureas.
Indeed, the vibration frequency of the NH function, n(NH),
depends on whether it is involved in a hydrogen bond or not:
Several bis-urea based molecules were synthesised from 2.
First, 2, synthesised in dichloromethane and containing low
amounts of 2b, was reacted with EHA to obtain almost pure 5a
(Scheme 3). The remaining 5b was eliminated by recrystallisa-
tion in ethyl acetate. It should be highlighted that when 2
containing large amounts of 2b is used for this reaction, the
quantity of 5b is too large to be eliminated by recrystallisation
in ethyl acetate (see Fig. S1 in Electronic supplementary
information). Similarly, amino-functional polydimethylsilox-
anes were transformed into bis-urea-functional ones (PDMS-
g-bis-urea) (Scheme 5). Moreover, tetra-ureas (i.e., bis-ureas
linked together by an aliphatic or a polymeric spacer) were
synthesised from diamines and 2 (Scheme 4). The difficult
recrystallisation of tetra-ureas containing an aliphatic spacer
explains the poor yields of their synthesis. The characteristics
of these new compounds are discussed in Section 4 below.
3. Synthesis of other mono-isocyanates/mono-ureas and non-
symmetrical bis-ureas
The synthesis of mono-isocyanate/mono-urea from 2,4-TDI is
not limited to the use of BuA. With the analytical technique
described above, it was proven that almost pure 3a is obtained
by reacting 1 equiv. of 2,6-diethylaniline with 5 equiv. of 2,4-
TDI in dichloromethane at room temperature (Scheme 2). The
y In dichloromethane, the elimination of 2,4-TDI is more difficult
because of the viscosity of the reaction mixture when 1.5 equiv. are
used: a small amount of 1 is formed in the second step of this run. With
a larger excess of 2,4-TDI though, more solvent is used and the
filtration eliminates all the remaining 2,4-TDI.
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 3 7 3 – 1 3 8 2
1377

View Article Online
Table 2 Solubility tests at room temperaturea
Name
Compound type
Heptane
Toluene
Chloroform
THF
5a
6a
7
Bis-urea
Bis-urea
Bis-urea
I(10)
I(2)
G(10)
I(2)
S
S
S
I(2)
S
I(2)
S
I(10)
I(5)
8
Aliphatic tetra-urea
Aliphatic tetra-urea
Aliphatic tetra-urea
Aliphatic tetra-urea
PDMS tetra-urea
PDMS tetra-urea
PDMS-g-bis-urea
PDMS-g-bis-urea
Bis-urea
I(5)
I(10)
I(1)
I(1)
SW
S/G(50)
SW
S
I(5)
I(10)
I(1)
I(1)
SW
S
I(5)
I(10)
I(1)
I(1)
S
9
I(10)
I(1)
10
11
12
13
14
15
16
a
I(1)
I(10)
S/G(50)
SW
S
SW
S
S
S
I(1)
S
I(1)
I(1)
I(1)
I(X): not soluble at X g Lꢀ1; G(X): physical gel at X g Lꢀ1; SW: swollen (the polymer is in equilibrium with excess solvent); S: soluble at 10 g Lꢀ1
.
n(bonded NH) B3350 cmꢀ1 and n(free NH) B3430 cmꢀ1 (Fig.
3). The quantitative measurement of these bands provides the
fraction of chain ends and thus the length of the supramole-
cular chains.
ing since these molecules are composed of very similar bis-
ureas. Nevertheless, the cooperativity of the self-assembly of 13
is much lower than that of 5a. This may be explained by the
bulkiness of the macromolecular chain of 13, which probably
hinders the formation of very long supramolecular chains.17
Moreover, the fact that bis-ureas are linked pair-wise by a
macromolecular spacer may add an additional entropic cost to
the self-assembly.18 However, the self-assembly of bis-ureas
grafted on a PDMS chain is still very good.
The evolution of the fraction of free NH functions was
plotted versus bis-urea concentration in chloroform for 5a, 7,
and 13 (Fig. 4). The curves reveal that the self-assembly of bis-
ureas increases with their concentration in chloroform, as
expected. Moreover, the self-assembly of 1, 5a and 13 is much
stronger than that of a reference mono-urea (N-2-ethylhexyl-
N0-2-methylphenylurea). This was already reported for 16,14
and is due to a pre-organisation of the urea functions by the
toluene spacer. Indeed, when one of the urea functions is
hydrogen-bonded, the association of the second one is fa-
voured, increasing the overall association strength. Further-
more, the sharp transition between dissociated molecules and
almost fully associated ones, observed for the bis-ureas, is
characteristic of a strongly cooperative self-assembly.6
The comparison of 5a and 1 shows that the non-symmetrical
bis-urea self-assembles much better than the symmetrical one
(Fig. 4). This is certainly caused by the presence of the
additional aromatic moiety in 5a, which increases the acidity
of the nearest NH,15,16 without hindering too much the asso-
ciation. However, the cooperativity of the association is not
significantly affected by the increased strength of the hydrogen
bonds. It is worth mentioning that the symmetrical bis-urea
with two aromatic substituents 16 should self-assemble even
more strongly than 5a, but 16 is not soluble in the solvents
considered (Table 2). This emphasises the usefulness of non-
symmetrical bis-urea 5a, which combines the solubility of
aliphatic bis-ureas with the strong association of aromatic
ones.
Finally, the behaviour of the bis-urea with only 3 NH
functions, 7, is intermediate between that of the mono-urea
and the other bis-ureas.
Beside these qualitative results, a theoretical model can be
used to describe the association of bis-ureas (Scheme 7).6 In
this model, K2 represents the equilibrium constant for the
association between two free bis-ureas, whereas K corresponds
to the association between a free bis-urea and a supramolecular
oligomer containing at least two bis-ureas. With this model, the
different bis-ureas can be compared. Indeed, K/K2 is an
evaluation of the cooperativity of the system, whereas K2/K2
is the association constant between two oligomers and char-
acterises the strength of the self-assembly.6
The values of K2 and K were determined by fitting this
theoretical model to the experimental points for 1, 5a, 13, and
the mono-urea (Table 3). The experimental points were not
accurate enough to determine precisely K and K2 for 7. The
values of K/K2 and K2/K2 confirm the previous qualitative
observations. In particular, K2/K2 is much lower for the mono-
The PDMS tetra-urea 13 begins to associate at almost the
same concentration (10ꢀ4 mol Lꢀ1) as 5a, which is not surpris-
Fig. 4 Evolution of the fraction of free NH function versus bis-urea
concentration in chloroform. Full curves are obtained by curve fitting
with the model described in Scheme 7. See Schemes 1, 3 and 4 for
molecular structures. 5a (-B-), 1 (-m-), 13 (-&-), 7 (n), N-2-ethylhexyl-
N0-2-methylphenylurea (K).
Fig. 3 FTIR spectra of 7 at 4.5 ꢁ 10ꢀ2 mol Lꢀ1 (–), 13 at 1.0 ꢁ 10ꢀ3
mol Lꢀ1 (. . .), and 5a at 2.5 ꢁ 10ꢀ4 mol Lꢀ1 (—) in CDCl3. These
solutions exhibit a free NH fraction of B60%.
1378
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 3 7 3 – 1 3 8 2

View Article Online
Scheme 7 Association equilibria of bis-ureas.
Fig. 6 Relative viscosity of a 7 + 1 solution in toluene versus the
weight fraction of 7, at a constant total weight concentration of 0.5
mg gꢀ1 and at 25 ꢂ 0.1 1C.
Table 3 Association constants (L molꢀ1) in chloroform, at 20 1C,
determined by curve fitting6
Compound K2
K
K/K2
K2/K2
Finally, the bis-urea 7 forms solutions of very low viscosity.
This is shown by monitoring the viscosity of a solution of 1 +
7 in toluene as a function of the ratio of 7, for a constant total
concentration (Fig. 6). This curve shows that the viscosity of
the solution decreases rapidly when even small amounts of 7
are added. This bis-urea, possessing only 3 NH functions, one
of which is sterically hindered, acts as a supramolecular chain
stopper.
5a
13
1
110 ꢂ 50 7400 ꢂ 500 70 ꢂ 45 5.1 ꢂ 3.7 ꢁ 105
470 ꢂ 20 2200 ꢂ 300 4 ꢂ 1
9.9 ꢂ 2.2 ꢁ 103
21 ꢂ 3
1400 ꢂ 200 70 ꢂ 20 1.0 ꢂ 0.5 ꢁ 105
Mono-urea 2 ꢂ 1
8 ꢂ 1
4 ꢂ 2
36 ꢂ 10
urea than for the bis-ureas. Moreover, the self-assembly of 13 is
much less cooperative than that of the low molecular weight
compounds 1 and 5a.
Conclusion
As the bis-urea concentration in chloroform increases, the
self-assembly leads to larger and larger structures. Thus, the
measurement of the resulting increase of viscosity of the
solution is a good way to estimate the strength of the self-
assembly. Viscosimetry measurements are reported on Fig. 5
and show that the viscosity of 5a is much higher than that of 1
at a given concentration. This difference is in agreement with
the much stronger self-assembly of 5a. The PDMS bis-urea 13
is also more viscous than 1 at a given mass concentration in
spite of its low bis-urea content. This may be explained first by
the bifunctionality of 13, which probably leads to the forma-
tion of a reversible network, because each bis-urea function is
expected to take part in the formation of a multifunctional
supramolecular chain. Moreover, the contribution of the
macromolecular backbone of 13 is not negligible, as hinted
by the evolution of the viscosity of the amino-functional
precursor of 13. Thus, 13 possesses very promising rheological
properties, even for low bis-urea concentrations, thanks to its
macromolecular structure and its bifunctionnality.
The selective reaction of an aromatic amine with an excess of
2,4-TDI yielded pure mono-isocyanate/mono-ureas for three
different amines. 1H NMR was used to identify the four
possible products of the reaction and determine their propor-
tion. With 4-n-butylaniline as the first amine, the first step is
fully selective in dichloromethane or heptane when an appro-
priate excess of 2,4-TDI is used. Moreover, the excess of 2,4-
TDI can be eliminated easily by filtration. Almost pure 2a can
be obtained in dichloromethane with a yield of about 70%,
whereas a 2a : 2b mixture (85 : 15 molar ratio) is collected with a
higher yield (490%) in heptane.
The selective synthesis of these mono-isocyanate/mono-
ureas is of great interest since it enables the development of
interesting new compounds. First, the non-symmetrical bis-
urea 5a combines the solubility of aliphatic bis-ureas with the
strong self-assembly of aromatic ones. Consequently, it self-
assembles in chloroform more strongly than 1. On the con-
trary, 7 self-assembles to a lesser extent than classical bis-ureas,
because one of its NH functions is missing. Thus, this com-
pound can be used as a supramolecular chain stopper of bis-
ureas.
Moreover, liquid amino-functional PDMS are changed into
rubbery solids when modified by bis-ureas, thanks to physical
cross-linking.13 More details on the mechanical properties of
these materials will be given shortly.19
Furthermore, compound 13, which consists of two bis-ureas
grafted at each end of a PDMS chain, presents very interesting
solution properties according to viscosity measurements in
chloroform. These properties are related to the very good
self-assembly of bis-ureas grafted on this compound, in spite
of the bulkiness of the PDMS polymeric chain.
Experimental
Spectroscopic methods
1H NMR spectra were recorded on a Bruker AC200r 200
¨
MHz spectrometer at 20 1C. FTIR spectra were obtained with
a Nicolet Avatar 320r spectrometer at 20 1C. Routine spectra
were recorded from solutions evaporated on KBr disks. Quan-
Fig. 5 Relative viscosity of chloroform solutions of 5a (-B-), 13 (-&-),
1 (-n-), and of the amino-functional precursor of 13 (-K-) versus
concentration of the bis-urea at 25 ꢂ 0.1 1C.
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 3 7 3 – 1 3 8 2
1379

View Article Online
titative FTIR was done in solution in KBr cells of 0.1–2.5 cm
path length.
Mono-isocyanate/mono-urea 3. The reaction procedure was
identical to that of 2 and conducted at room temperature in dry
dichloromethane. A white solid (89%) was collected by filtra-
tion. 1H NMR (200 MHz, [D6]DMSO dried over molecular
sieve, 22 1C): d = 8.88/7.76 (s, 2H, ArNH), 7.42 (s, 1H, ArH),
7.13 (m, 5H, ArH), 2.54 [q, 3J(H,H) = 7 Hz, 4H, ArCH2], 2.21
Quantitative FTIR spectroscopy
Mother chloroform solutions were prepared and diluted to
reach low concentrations. They were prepared at room tem-
perature with stirring for at least one night. After dilution of
the mother solutions, daughter solutions were obtained, and
stirred for 1 h or longer. Deuterated chloroform (CDCl3) was
used instead of hydrogenated chloroform (CHCl3) as the latter
absorbs too much in the NH region for diluted solutions.
CDCl3 was dried over molecular sieves (4 A) 2 days before
use. The solutions themselves were not dried because bis-ureas
are absorbed on molecular sieves.
3
(s, 3H, ArCH3), 1.12 [t, J(H,H) = 7 Hz, 6H, ArCH2CH3].
Mono-isocyanate/mono-urea 4. A solution of N-methylani-
line (6.0 mL, 0.056 mol) in dry heptane (200 mL) was added 2.5
h, at 0 1C and under nitrogen, to a stirred solution of 2,4-TDI
(40 mL, 0.279 mol) in dry heptane (270 mL). The reaction
mixture was stirred for several hours at 0 1C and allowed to
warm up to room temperature overnight. The white precipitate
formed was then quickly filtered on a fritted funnel, rinsed with
dry heptane (3 ꢁ 50 mL), and dried under vacuum to afford
10.1 g of a white solid (64%). No further purification was
performed because of the moisture sensitivity of the product.
The purity of the product was only checked by the two-step 1H
NMR method described in Section 1.
Capillary viscosimetry
Measurements were performed at 25 ꢂ 0.1 1C with a Cannon–
Manning (M108 n 125) semimicro viscometer. Solutions in
chloroform (stabilised with amylenes) were filtered on a Millex
PVDF filter (0.45 mm) before viscosity measurements, whereas
toluene solutions were not filtered because of their high
viscosity. No significant difference were observed when CDCl3
was used instead of CHCl3.6 The toluene mother solution of 1
was prepared 1 week before use and heated 1 day at 50 1C to
improve its homogeneity. The toluene mother solution of 7 was
prepared 1 day before use at room temperature. Solutions of 1
and 7 at the same weight concentration were mixed to obtain
the daughter solutions at different 1 : 7 ratios and constant
concentration.
Non-symmetrical bis-ureas: evaluation of the purity of the
mono-isocyanate/mono-urea. Pure EHA (0.27 mL, 1.6 ꢁ 10ꢀ3
mol) was added to a stirred solution of 2 (0.5 g, 1.5 ꢁ 10ꢀ3 mol
of mono-isocyanate/mono-urea expected) in dry THF (20 mL)
in order to convert the remaining isocyanate functions of this
product. After standing for one night, the end of the reaction
was checked by the absence of the isocyanate band by FTIR
(2270 cmꢀ1) and the crude product was collected after elimina-
tion of the solvent. 1H NMR of this crude product was realised
in order to identify the bis-ureas it contains and deduce the
selectivity of the first step. 1H NMR (200 MHz, [D6]DMSO, 22
1C) between 1.1 and 2.6 ppm: d = 2.5 (t, 2H, ArCH2 and
DMSO), 2.18 (s, 3H, ArCH3 of 16), 2.15 (s, 3H, ArCH3 of 5b),
2.11 (s, 3H, ArCH3 of 5a), 2.08 (s, 3H, ArCH3 of 1), 1.52 (m, 2H,
ArCH2CH2), 1.27 (m, 11H, CH + CH2 + ArCH2CH2CH2).
Syntheses
Analytical grade solvents were used and could be dried by
refluxing over calcium hydride (dichloromethane, chloroform),
or sodium (THF, heptane, toluene, dioxane) for several hours
and then distilling.
2,4-Toluene diisocyanate (96% 2,4-TDI + 4% 2,6-TDI),
2-ethylhexylamine (98%), 2-butyl-2-ethyl-1,5-diaminopentane
(98%), and 1,10-diaminodecane (97%), from Aldrich, as well
as AMS-162, from ABCR, were used as received. 4-nButylani-
line (97%), 2,6-diethylaniline (98%), N-methylaniline (98%),
1,2-diaminoethane (99%), and 1,3-diaminopropane (99%),
were purchased from Aldrich, and distilled before use. The
synthesis of 1 was previously reported.5
Non-symmetrical bis-urea 5a. A solution of EHA (1.06 mL,
6.5 ꢁ 10ꢀ3 mol) in dry dioxane (20 mL) was added over 10 min,
at room temperature and under nitrogen, to a stirred solution
of 2 (run OC072: dichloromethane, 2,4-TDI : BuA = 5 : 1; 2.0
g, 6.2 ꢁ 10ꢀ3 mol) in dry dioxane (150 mL). A white precipitate
appeared rapidly. After stirring overnight, the precipitate was
filtered on a fritted funnel, rinsed 3 times with small quantities
of dry dioxane, and dried under vacuum to afford 2.1 g of a
white solid. The product was then recrystallised in ethyl acetate
1
(final yield = 56%) to get rid of the remaining 5b. H NMR
(200 MHz, [D6]DMSO, 22 1C): d = 8.55/8.38 (s, 2H, ArNH),
7.87 [d, J(H,H) = 2 Hz, 1H, ArH], 7.54 (s, 1H, ArNH), 7.32
Mono-isocyanate/mono-urea 2. A solution of BuA (13.2 mL,
0.084 mol) in dry dichloromethane (400 mL) was added over 8
h, at room temperature and under nitrogen, to a stirred
solution of 2,4-TDI (60 mL, 0.419 mol) in dry dichloromethane
(400 mL). The reaction mixture was stirred overnight. The
white precipitate formed was then quickly filtered on a fritted
funnel, rinsed with dry dichloromethane (3 ꢁ 50 mL), and
dried under vacuum to afford 19.7 g of a white solid (73%). No
further purification was done because of the moisture sensitiv-
ity of the product. The purity of the product was checked by
1H NMR and by the two-step method described in Section 1.
1H NMR (200 MHz, [D6]DMSO dried over molecular sieve,
22 1C): d = 8.73/8.64 (s, 2H, ArNH), 7.42 (s, 1H, ArH), 7.34/
7.12 [d, 3J(H,H) = 8 Hz, 4H, ArH], 7.13/7.06 (s, 2H, ArH), 2.5
(t, 2H, ArCH2), 2.22 (s, 3H, ArCH3), 1.52 (m, 2H, ArCH2C
4
3
3
4
[d, J(H,H) = 8 Hz, 2H, ArH], 7.16 [dd, J(H,H) = 8 Hz,
J(H,H) = 2 Hz, 1H, ArH], 7.07/6.97 [d, 3J(H,H) = 8 Hz, 3H,
ArH], 6.53 [t, 3J(H,H) = 6 Hz, 1H, CH2NH], 3.06 (m, 2H, NC
H2), 2.5 (t, 2H, ArCH2), 2.11 (s, 3H, ArCH3), 1.52 (m, 2H,
ArCH2CH2), 1.27 (m, 11H, CH + CH2 + ArCH2CH2CH2),
0.88 [t, 3J(H,H) = 7 Hz, 9H, CH3]; 13C-NMR (50 MHz,
[D ]DMSO, 22 1C): d = 155.5/152.6 (C O), 138.6/137.9/
Q
6
137.5/135.5/130.1/ 128.5/119.2/118.1/111.4/109.9 (Ar), 41.5 (N
CH2), 39.8 (CH), 34.3/33.4/30.6/28.6/23.8/22.6/21.8 (CH2),
17.3 (ArCH3), 14.0/13.8/10.9 (CH3); IR(KBr): n = 3333/3281
(N–H), 1641 (C O) cmꢀ1; anal. calcd (%) for C27H40N4O2
Q
(452.6): C 71.65, H 8.91, N 12.38, O 7.07; found: C 71.22, H
8.98, N 12.29, O 7.62. 5a contains a little water (B1% by
weight), which increases the percentage of O compared to C
and N.
3
H2), 1.31 (m, 2H, ArCH2CH2CH2), 0.88 [t, J(H,H) = 7 Hz,
3H, CH3].
When the reaction is performed in dry heptane, the experi-
mental conditions are the same. No purification is done when
the reaction is conducted in dry THF. The 2,4-TDI : BuA ratio
is adapted for each run, but the concentration of each reagent
is unchanged.
Non-symmetrical bis-urea 6. The synthesis is derived from
that of 5a except that the reaction was conducted in dry
chloroform and with the mono-isocyanate/mono-urea 3 at a
1380
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 3 7 3 – 1 3 8 2

View Article Online
concentration of 2.3 g Lꢀ1. At the end of the reaction, the
precipitate was too thin to be filtered. After elimination of most
of the solvent, the product was precipitated in heptane, filtered,
washed thoroughly with heptane, dried under vacuum, and
C40H50N8O4 (706.9): C 67.96, H 7.13, N 15.85, O 9.05; found:
C 67.41, H 7.08, N 16.08, O 9.44.
Compound 9. 9 was recrystallised in DMF–AcOEt (50/ : 50
v/v) at 80 1C (10%). 1H NMR (200 MHz, [D6]DMSO, 22 1C):
d = 8.51/8.39 (s, 4H, ArNH), 7.85 [d, 4J(H,H) = 2 Hz, 2H, Ar
1
recrystallised in ethanol (yield = 63%). H NMR (200 MHz,
[D6]DMSO, 22 1C): d = 8.62 (s, 1H, ArNH), 7.83 [d, 4J(H,H)
= 2 Hz, 1H, Ar-H), 7.53/7.47 (s, 2H, ArNH), 7.17 [dd,
3
3
H], 7.61 (s, 2H, ArNH), 7.32 [d, J(H,H) = 8 Hz, 4H, ArH],
3
7.14 [dd, J(H,H) = 8 Hz, J(H,H) = 2 Hz, 2H, ArH], 7.07/
4
4
J(H,H) = 8 Hz, J(H,H) = 2 Hz, 1H, ArH], 7.10 (m, 3H, Ar
3
3
3
3
H), 6.95 [d, J(H,H) = 8 Hz, 1H, ArH], 6.50 [t, J(H,H) = 6
6.98 [d, J(H,H) = 8 Hz, 6H, ArH], 6.61 [t, J(H,H) = 6 Hz,
2H, CH2NH], 3.16 (m, 4H, NCH2), 2.5 (t, 4H, ArCH2), 2.12
(s, 6H, ArCH3), 1.59 (m, 2H, CH2CH2CH2), 1.48 (m,
4H, ArCH2CH2), 1.31 (m, 4H, ArCH2CH2CH2), 0.88 [t,
3J(H,H) = 7 Hz, 6H, CH3]; anal. calcd (%) for C41H52N8O4
(720.9): C 68.31, H 7.27, N 15.54, O 8.88; found: C 67.83,
H 7.08, N 15.44, O 9.07.
3
Hz, 1H, CH2NH], 3.06 (m, 2H, NCH2), 2.59 [q, J(H,H) = 7
Hz, 4H, ArCH2], 2.10 (s, 3H, ArCH3), 1.27 (m, 9H, CH + C
H2), 1.13 [t, 3J(H,H) = 7 Hz, 6H, ArCH2CH3], 0.88 (m, 6H, C
H3); anal. calcd (%) for C27H40N4O2 (452.6): C 71.65, H 8.91,
N 12.38, O 7.07; found: C 71.40, H 8.97, N 12.32, O 7.70. 6
contains a little water (B1% by weight), which increases the
percentage of O compared to C and N.
Compound 10. 10 was collected by evaporation of THF
instead of filtration because the precipitate was too thin to be
filtered. Moreover, the reaction took 1 week instead of over-
night. The compound was recrystallised in DMF–AcOEt
(17 : 83 v/v) at 80 1C (yield 21%). 1H NMR (200 MHz,
[D6]DMSO, 22 1C): d = 8.51/8.37 (s, 4H, ArNH), 7.86/7.83
Non-symmetrical bis-urea 7. The synthesis is derived from
that of 5a, except that the mono-isocyanate/mono-urea used
was 4 and that the reaction was conducted in dry THF. The
crude product was collected by evaporation of the solvent, and
then purified by silica gel column chromatography with chloro-
form–ethanol (96 : 4 v/v) as the eluent, precipitated from a
concentrated chloroform solution in heptane, and recrystal-
lised in an heptane–cyclohexane (78 : 22 v/v) mixture. A white
product was obtained with a final yield of 11%. 1H NMR (200
4
[d, J(H,H) = 2 Hz, 2H, ArH], 7.62/7.52 (s, 2H, ArNH), 7.32
[d, 3J(H,H) = 8 Hz, 4H, ArH], 7.18/7.15 [dd, 3J(H,H) = 8 Hz,
3
4J(H,H) = 2.0 Hz, 2H, ArH], 7.06/6.98 [d, J(H,H) = 8 Hz,
3
6H, ArH], 6.58/6.40 [t, J(H,H) = 6 Hz, 2H, CH2NH], 3.07/
3.00 (m, 4H, NCH2), 2.5 (t, 4H, ArCH2), 2.13/2.10 (s, 6H, ArC
4
H3), 1.55 (m, 4H, ArCH2CH2), 1.31 (m, 16H, CH2
+
MHz, [D6]DMSO, 22 1C): d = 7.97 (s, 1H, ArNH), 7.79 [d,
J(H,H) = 2 Hz, 1H, ArH], 7.50 (s, 1H, ArNH), 7.32 (m, 5H,
ArCH2CH2CH2), 0.88/0.80 [t, 3J(H,H) = 7 Hz, 12H, CH3];
anal. calcd (%) for C49H68N8O4 (833.1): C 70.64, H 8.23, N
13.45, O 7.68; found: C 70.59, H 8.23, N 13.67, O 7.92.
3
ArH), 7.07 [dd, J(H,H) = 8 Hz, 4J(H,H) = 2 Hz, 1H, ArH],
3
3
6.94 [d, J(H,H) = 8 Hz, 1H, ArH], 6.47 [t, J(H,H) = 6 Hz,
1H, CH2NH], 3.25 (s, 3H, NCH3), 3.03 (m, 2H, NCH2), 2.10
(s, 3H, ArCH3), 1.25 (m, 9H, CH + CH2), 0.86 (m, 6H, CH3);
Bis-urea grafted PDMS 14. The content of amine functions
of PDMS-g-NH2 was determined by 1H NMR (200 MHz,
CDCl3, 22 1C). Indeed, the peak at 2.65 ppm corresponds to 2n
protons, whereas the peak at 0.06 ppm corresponds to 6m + 3n
protons if the chain ends are neglected (Scheme 5). Therefore,
the ratio m/n can easily be determined.
13C NMR (50 MHz, [D ]DMSO, 22 1C): d = 155.3/154.8 (C
Q
6
O), 144.3/138.0/129.4/129.2/126.2/ 125.6/120.3/113.9/112.6
(Ar), 41.6 (NCH2), 39.3 (CH), 37.4 (NCH3), 30.5/28.5/23.7/
22.5 (CH2), 17.3 (ArCH3), 14.0/10.8 (CH3); anal. calcd (%) for
C24H34N4O2 (410.6): C 70.21, H 8.35, N 13.65, O 7.79; found:
C 69.82, H 8.35, N 14.05, O 8.03.
A solution of the amino-functional PDMS precursor (AMS-
162, 35.0 g, 19.8 ꢁ 10ꢀ3 molar equivalent of NH2 functions) in
dry THF (300 mL) was slowly added, at room temperature and
under nitrogen, to a stirred solution of 2 (run OC127: heptane,
2,4-TDI : BuA ratio = 3 : 1;7.66 g, 23.7 ꢁ 10ꢀ3 mol) in dry
THF (300 mL). After 1 week, the reaction mixture was
concentrated down to about 200 mL by evaporation of the
solvent and precipitated in 2 L of methanol. The precipitate
was then filtered on a n1 2 fritted funnel, washed thoroughly
with methanol and dried under vacuum to give 31.5 g (76%
yield) of an off-white rubbery solid. 1H NMR {200 MHz,
CDCl3–[D6]DMSO (85 : 15 v/v), 22 1C}: d = 8.14/8.03 (s, 2H,
ArNH), 7.66 (s, 1H, ArH), 7.26 (s, 1H, ArNH), 7.22 (m, 3H,
ArH), 6.97 [d, 3J(H,H) = 8 Hz, 2H, ArH], 6.90 [d, 3J(H,H) =
8 Hz, 1H, ArH], 6.10 (t, 1H, CH2NH), 3.08 (m, 2H, NCH2),
2.46 (t, 2H, ArCH2), 2.08 (s, 3H, ArCH3), 1.49 (m, 4H, CH2C
H2CH2 + ArCH2CH2), 1.24 (m, 2H, ArCH2CH2CH2), 0.84 [t,
3J(H,H) = 7 Hz, 3H, CH3], 0.48 (m, 2H, SiCH2), 0.01 (s,
Tetra-urea 11. A solution of 1,10-diaminodecane (1.20 g,
6.98 ꢁ 10ꢀ3 mol) in dry THF (240 mL) was slowly added,
at room temperature and under nitrogen, to a stirred solution
of 2 (run OC084: heptane, 2,4-TDI : BuA ratio = 5 : 1; 4.51g,
13.95 ꢁ 10ꢀ3 mol) in dry THF (180 mL). An off-white solid
precipitated quickly. After one night, FTIR spectroscopy
revealed the absence of remaining isocyanate functions (2270
cmꢀ1). The precipitate was then filtered on a Buchner, rinsed
with THF and dried under vacuum to give 5.4 g of a yellowish
solid. The recrystallisation of this crude product in DMF at
80 1C led to a white product (17%). 1H NMR (200 MHz,
[D6]DMSO, 22 1C): d = 8.52/8.38 (s, 4H, ArNH), 7.85 (s, 2H,
ArH), 7.50 (s, 2H, ArNH), 7.32 [d, 3J(H,H) = 8 Hz, 4H, ArH],
7.11 [dd, J(H,H) = 8 Hz, J(H,H) = 2 Hz, 2H, ArH], 7.07/
6,97 [d, 3J(H,H) = 8 Hz, 6H, ArH], 6.53 (m, 2H, CH2NH), 3.1
(m, 4H, NCH2), 2.5 (t, 4H, ArCH2), 2.10 (s, 6H, ArCH3), 1.48
(m, 4H, ArCH2CH2), 1.29 (m, 20H, CH2 + ArCH2CH2CH2),
0.88 [t, 3J(H,H) = 7 Hz, 6H, CH3); anal. calcd (%) for
C48H66N8O4 (819.1): C 70.38, H 8.12, N 13.68, O 7.81; found:
C 70.01, H 8.20, N 13.61, O 8.54. 11 contains a little water
(B1% by weight), which increases the percentage of O com-
pared to C and N.
3
4
139H, SiCH ); IR(KBr): n = 3318 (N–H), 1640 (C O) cmꢀ1
;
anal. calcd (%): C 38.9, H 7.9, N 2.7; found: C 38.53, H 7.92, N
Q
3
2.72.
PDMS-spaced tetra-ureas (12, 13) and other bis-urea grafted
PDMS (15). They were synthesised and purified in the same
way as 14.
1
12. Molar ratio 2 : diamine = 4.4 (yield = 54%), H NMR
{200 MHz, CDCl3–[D6]DMSO (85 : 15 v/v), 22 1C}: d =
Tetra-ureas 8, 9, 10. The synthesis of these tetra-ureas is
derived from that of 11.
8.09/7.99 (s, 2H, ArNH), 7.63 [d, 4J(H,H) = 2 Hz, 1H, ArH],
3
7.26 (s, 1H, ArNH), 7.22/7.14 (m, 3H, ArH), 6.97 [d, J(H,H)
3
= 8 Hz, 2H, ArH], 6.90 [d, J(H,H) = 8 Hz, 1H, ArH], 6.05
Compound 8. 8 was recrystallised in DMF–AcOEt (50 : 50
v/v) at 80 1C (yield = 10%). No 1H NMR was done because of
the poor solubility of this product. Anal. calcd (%) for
3
[t, J(H,H) = 6 Hz, 1H, CH2NH], 3.10 (m, 2H, NCH2), 2.46
3
[t, J(H,H) = 7 Hz, 2H, ArCH2], 2.07 (s, 3H, ArCH3), 1.48
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 3 7 3 – 1 3 8 2
1381

View Article Online
(m, 4H, CH2CH2CH2
+
ArCH2CH2), 1.26 (m, 2H,
2.5 (t, 4H, ArCH2), 2.17 (s, 3H, ArCH3), 1.52 (m, 4H, ArCH2C
3
H2), 1.29 (m, 4H, ArCH2CH2CH2), 0.89 [t, J(H,H) = 7 Hz,
6H, CH3].
ArCH2CH2CH2), 0.84 [t, 3J(H,H) = 7 Hz, 3H, CH3], 0.49
(m, 2H, SiCH2), 0.00 (s, 146H, SiCH3); IR(KBr): n = 3304 (N–
H), 1638 (C O) cmꢀ1; anal. calcd (%): C 38.5, H 7.9, N 2.6;
found: C 39.11, H 8.14, N 2.73.
Q
Acknowledgements
1
13. Molar ratio 2 : diamine = 6.6 (yield = 86%), H NMR
{200 MHz, CDCl3–[D6]DMSO (85 : 15 v/v), 22 1C}: d = 8.07/
7.94 (s, 2H, ArNH), 7.62 [d, J(H,H) = 2 Hz, 1H, ArH], 7.35
Sylvaine Ragout and Matthieu Jalabert are thanked for their
technical assistance. Nexans Company is acknowledged for
financial support.
4
3
(s, 1H, ArNH), 7.27/6.97 (m, 6H, ArH), 5.92 [t, J(H,H) = 6
Hz, 1H, CH2NH], 3.09 (m, 2H, NCH2), 2.45 (t, 2H, ArCH2),
2.09 (s, 3H, ArCH3), 1.47 (m, 4H, CH2CH2CH2 + ArCH2C
3
H2), 1.25 (m, 2H, ArCH2CH2CH2), 0.82 [t, J(H,H) = 7 Hz,
References
3H, CH3], 0.48 (m, 2H, SiCH2), ꢀ0.01 (s, 1160H, SiCH3);
1
2
3
4
5
J. -M. Lehn, Supramolecular Chemistry:Concepts and Perspectives,
VCH, 1995.
(a) A. Ciferri.Supramolecular polymers: New York, 2000; (b) A.
Ciferri, J. Macromol. Sci., Polym. Rev., 2003, C43(2), 271.
L. Brunsveld, B. B. Folmer, E. W. Meijer and R. P. Sijbesma,
Chem. Rev., 2001, 101(12), 4071.
N. Zimmerman, J. S. Moore and S. C. Zimmerman, Chem. Ind.,
1998, 604.
´
F. Lortie, S. Boileau, L. Bouteiller, C. Chassenieux, B. Deme, G.
Ducouret, M. Jalabert, F. Laupretre and P. Terech, Langmuir,
2002, 18, 7218.
V. Simic, L. Bouteiller and M. Jalabert, J. Am. Chem. Soc., 2003,
125, 13148.
J. H. Saunders and R. J. Slocombe, Chem. Rev., 1948, 43, 203.
R. G. Arnold, J. A. Nelson and J. J. Verbanc, Chem. Rev., 1957,
57, 47.
IR(KBr): n = 3303 (N–H), 1638 (C O) cmꢀ1; anal. calcd
(%): C 33.3, H 8.1, N 0.4; found: C 33.28, H 8.15, N 0.43.
Q
1
15. Molar ratio 2 : amine = 1.2 (yield = 69%), H NMR
{200 MHz, CDCl3–[D6]DMSO (85 : 15 v/v), 22 1C}: d = 7.39/
7.24 (m, 8H, ArH + ArNH), 6.89 [d, 3J(H,H) = 8 Hz, 1H, Ar
H], 6.32 (s, 1H, ArNH), 6.10 (t, 1H, CH2NH), 3.22 (m, 3H, NC
H3), 3.04 (m, 2H, NCH2), 2.18 (s, 3H, ArCH3), 1.43 (m, 2H,
CH2CH2CH2), 0.44 (m, 2H, SiCH2), 0.00 (s, 141H, SiCH3);
IR(KBr): n = 3429/3346/3290 (N–H), 1637 (C O) cmꢀ1
;
anal. calcd (%): C 37.7, H 7.8, N 2.7; found: C 37.33, H
Q
6
7.92, N 2.56.
7
8
Symmetrical bis-urea 16. This symmetrical bis-urea could
have been obtained directly by reacting 2,4-TDI with 2 equiv.
of BuA; but the synthesis described here consists in reacting 2
with BuA. A solution of BuA (0.22 mL, 1.4 ꢁ 10ꢀ3 mol) in dry
THF (10 mL) was added over 10 min, at room temperature and
under nitrogen, to a stirred solution of 2 (run OC072: dichlor-
omethane, 2,4-TDI : BuA = 5 : 1; 0.220 g, 6.8 ꢁ 10ꢀ4 mol) in
dry THF (20 mL). A white precipitate appeared more than 3 h
after addition of the amine. After 11 days at room temperature,
the reaction was complete according to the absence of iso-
cyanate functions (2270 cmꢀ1 in FTIR spectroscopy). The
precipitate was filtered on a fritted funnel, rinsed 3 times with
small quantities of THF, and dried under vacuum to afford a
white solid (70%). It was not recrystallised because of its poor
solubility. 1H NMR (200 MHz, [D6]DMSO, 22 1C): d =
9
B. Grepinet, F. Pla, P. Hobbes, P. Swaels and T. Monge, J. Appl.
Polym. Sci., 2000, 75, 705.
¨
and M. Moller, J. Polym. Sci. Part A: Polym. Chem., 2000, 38(14),
10 C. F. Bartelink, M. De Pooter, H. J. M. Grunbauer, U. Beginn
¨
2555.
11 C. W. Carpenter, T. G. Savino and A. L. Steinmetz, EP 465 805,
2000.
12 O. Colombani, PhD Thesis, Universite
13 O. Colombani, L. Bouteiller, C. Barioz, J. Bonetti-Riffaud and
L. Fomperie, Polymer Preprints, 2003, 44(1), 620.
14 S. Boileau, L. Bouteiller, F. Laupretre and F. Lortie, New J.
Chem., 2000, 24, 845.
15 T. V. Kozlova and V. V. Zharkov, Zh. Prikl. Spektrosk., 1981,
35(2), 303.
16 H. Lee, J. Park, C. Yoon and Y. S. Choi, Bull. Korean Chem. Soc.,
1988, 19(1), 110.
17 L. de Lucca Freitas, C. Auschra, V. Abetz and R. Stadler, Colloid
Polym. Sci., 1991, 269, 566.
18 R. Stadler, Macromolecules, 1988, 21, 121.
19 O. Colombani, C. Barioz, L. Bouteiller, C. Chane
F. Lortie and H. Montes, submitted.
´
Paris VI, (France), 2003.
´
4
8,96/8.61/8.42 (s, 3H, ArNH), 7.92 [d, J(H,H) = 2 Hz, 1H,
3
ArH], 7.84 (s, 1H, ArNH), 7.37/7.32 [d, J(H,H) = 8 Hz, 4H,
3
4
ArH], 7.19 [dd, J(H,H) = 8 Hz, J(H,H) = 2 Hz, 1H, ArH],
ac, L. Fomperie,
´ ´
3
7.09/7.07 [d, J(H,H) = 8 Hz, 4H, ArH], 7.03 (m, 1H, ArH),
1382
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 3 7 3 – 1 3 8 2