Synthesis and structural characterization of new stiff rod oligomeric domains by X-ray crystallography and NMR

Article abstract of DOI:10.1021/jo025980f

The monomers N,N′-dibenzylbenzene-1,4-diamine (1), N,N′-dibenzylnaphthalene-1,5-diamine (2), and N,N′-dibenzylanthracene-1,9-diamine (3) were reacted with phosgene in the presence of a base to produce the corresponding N,N′-dibenzyl-1,4-bis(chlorocarbonylamino)benzene (4), N,N′-dibenzyl-1,5-bis(chlorocarbonylamino)naphthalene (5), and N,N′-dibenzyl-9,10-bis(chlorocarbonylamino)anthracene (6). These monomers were used to create zigzag type stacks, in a stepwise fashion, of trimers and 9-mers of either 1,4-diureidobenzenes (^PheK(3) and ^PheK(9)) or 1,5-diureidonaphthalenes (^NapK(3) and ^NapK(9)). A byproduct in the formation of ^PheK(9) was a cyclic hexamer ^PheK(6). NMR gave evidence of the structure in solution while X-ray crystallographic information was obtained for 5, 6, ^NapK(3), and the cyclic ^PheK(6).

Full text of DOI:10.1021/jo025980f

Syn th esis a n d Str u ctu r a l Ch a r a cter iza tion of New Stiff Rod
Oligom er ic Dom a in s by X-r a y Cr ysta llogr a p h y a n d NMR
Frederik C. Krebs and Mikkel J ørgensen*
The Danish Polymer Center, Risø National Laboratory, P.O. Box 49, Roskilde, Denmark
mikkel.joergensen@risoe.dk
Received May 28, 2002
The monomers N,N′-dibenzylbenzene-1,4-diamine (1), N,N′-dibenzylnaphthalene-1,5-diamine (2),
and N,N′-dibenzylanthracene-1,9-diamine (3) were reacted with phosgene in the presence of a base
to produce the corresponding N,N′-dibenzyl-1,4-bis(chlorocarbonylamino)benzene (4), N,N′-dibenzyl-
1,5-bis(chlorocarbonylamino)naphthalene (5), and N,N′-dibenzyl-9,10-bis(chlorocarbonylamino)-
anthracene (6). These monomers were used to create zigzag type stacks, in a stepwise fashion, of
trimers and 9-mers of either 1,4-diureidobenzenes (P h eK(3) and P h eK(9)) or 1,5-diureidonaphthalenes
(
Na p K(3) and Na p K(9)). A byproduct in the formation of P h eK(9) was a cyclic hexamer P h eK(6). NMR
gave evidence of the structure in solution while X-ray crystallographic information was obtained
for 5, 6, Na p K(3), and the cyclic P h eK(6).
In tr od u ction
features be maintained with other N-alkyl groups? And
what would happen if the central benzene groups were
substituted for larger groups such as naphthalene and
anthracene? We chose to use N-benzyl groups instead of
the N-methyl groups to demonstrate that the effect on
the structure was maintained but more importantly to
create an easy route to large amounts of the monomer
units with either a central 1,4-phenyl or 1,5-naphthyl
unit (see Scheme 1).
N-Methylation of benzanilides and N,N′-diarylureas
has a dramatic effect on their structure, as shown first
by Yamaguchi and Matsumura.1 Amides and ureas
almost invariably have a flat transoid structure when the
nitrogen atoms still have one hydrogen atom attached.
Substitution with a methyl group changes the geometry
to a cisoid form with the aryl groups placed almost face-
to-face.
For easy reference to these compounds we have in this
paper chosen a new shorthand nomenclature. The whole
class of compounds we call K domains (from one of the
authors). A prefix (Phe or Nap) designates the central
aromatic unit being either a benzene or a naphthalene.
A number in parentheses denotes the number of sub-
units. Thus P h eK(9) is meant to identify the 9-mer of 1,4-
dibenzylphenylene diamine units joined via urea link-
ages.
The corresponding diamines were condensed with
benzaldehyde to the diimines, which could be reduced
quantitatively with sodium borohydride in tetrahydro-
furan-acetic acid to N,N′-dibenzylamines. 1,4-dibenzyl-
aminobenzene (1) and 1,5-dibenzylamino-naphthalene (2)
have been prepared previously from the corresponding
diamines and benzyl alcohol.5 The monomer 9,10-diben-
zylaminoanthracene (3) was less easily obtained in pure
form since it oxidizes readily. Anthraquinone was reacted
with formamide at reflux for several hours to prepare
9,10-diformylaminoanthracene as described by Schiedt.6
This compound could be N-alkylated with benzyl chloride
and potassium tert-butoxide in tetrahydrofuran to pro-
duce the rather insoluble N,N-dibenzyl-9,10-diformyl-
aminoanthracene. Hydrolysis in ethanol with a large
excess of potassium hydroxide gave the monomer 9,10-
This characteristic structural feature, called a molec-
ular splint, or the N-methyl amide effect, has been
exploited to direct the structure of compounds such as
calixarenes2 and metal-binding receptors.3
Yamaguchi and Matsumura used the effect to construct
a stiff rod of five benzene rings joined in a zigzag fashion
by N,N′-dimethylurea moieties. Their brief and purely
structural note inspired us to undertake a further
synthetic and structural exploration of this phenomenon.
Our motivation was that such rod-like compounds with
a central stack type aromatic structure might find several
interesting applications. From a purely structural view
they could be utilized as a new type of building block to
construct very large molecules from domains. The central
aromatic moieties have geometry reminiscent of the
pancake-like discotic phases of, e.g., electronically con-
ducting alkyl phthalocyanines.4
Resu lts a n d Discu ssion
From the considerations above several questions im-
mediately came to mind. Could the overall structural
(1) Yamaguchi, K.; Matsumura, G. J . Am. Chem. Soc. 1991, 113,
5474.
(2) Krebs, F. C.; Larsen, M.; J ørgensen, M.; J ensen, P. R.; Bielecki,
M.; Schaumburg, K. J . Org. Chem. 1998, 63, 9872-9879.
(3) J ørgensen, M.; Krebs, F. C. Tettrahedron Lett. 2001, 42, 4717-
4720.
(4) Schouten, P. G.; Warman, J . M.; de Haas, M. P. J . Phys. Chem.
1993, 97, 9863-9870.
(5) Sprinzak, Y. J . Am. Chem. Soc. 1956, 78, 3207.
(6) Schiedt, B. J . Prakt. Chem. 1941, 157, 203.
10.1021/jo025980f CCC: $22.00 © 2002 American Chemical Society
Published on Web 09/26/2002
J . Org. Chem. 2002, 67, 7511-7518
7511

Krebs and J ørgensen
SCHEME 1. Syn th esis of Mon om er Un its a n d Oligom er iza tion Rea ction s to th e P h eK a n d Na p K Oligom er
Dom a in s
dibenzylaminoanthracene in low yield. Bis-carbamoyl
chlorides (4, 5, and 6) were then prepared from the three
dibenzylamine monomers by reaction with phosgene.
The bis-carbamoyl chlorides could either be isolated as
stable crystalline solids or alternatively reacted in situ
with an excess of the N,N′dibenzylamine. It was found
that this reaction was rather sluggish at the usual
conditions for a reaction between an acid chloride and
an amine. Much effort was spent in determining the best
reaction conditions. We found that reaction between the
bis-carbamoyl chlorides and amines was best carried out
in freshly purified collidine (2,4,6-trimethylpyridine) at
reflux using dry cesium carbonate as the base. Other
combinations of solvents and bases gave poorer yields and
a significant amount of decomposition. Thus, pure tri-
mers of the phenyl and naphthalene could be obtained
in large scale (50-100 g) after suitable workup (see
Figure 2). In the case of the anthracene monomer no
trimer could be obtained in our hands. Even prolonged
reaction times did not yield any of the desired products.
F IGURE 1. Oligomeric aryl-ureido stiff rod domains. On the right is shown a schematic view of the relatively easily prepared
9-mer showing the zigzag type structure with good π-π overlap of the central aromatic residues.
7512 J . Org. Chem., Vol. 67, No. 21, 2002

Characterization of New Stiff Rod Oligomeric Domains
of X-ray crystallographic quality and for further synthetic
elaboration. Instead a fractionated crystallization scheme
was developed that gave gram scale quantities of the
oligomers in reasonably pure form (see Figure 2). The
purity was checked both by SEC and also by a mass
spectroscopic technique for large molecules (MALDI-
TOF).
A typical oligomerization of the Na p K(3) compound to
produce the 9-mer can be illustrated from the mass-
spectroscopic data obtained from MALDI-TOF shown in
Figure 3. A mixture of possible oligomers is formed
progressing toward higher molecular mass with time. A
sample was drawn and analyzed to give peaks belonging
to oligomers Na p K(3) to Na p K(15). The trimer can be seen
to dominate (MW 1067.32) with the desired 9-mer as the
second major component (assuming similar ionization
cross sections for the different oligomers). The above-
mentioned purification procedure then produced the
compound giving the MALDI-TOF spectrum shown in
Figure 3 (right) having peaks at 3254 and 3357 (m/z)
which can be ascribed to the 9-mer and a complex with
silver ion added to the matrix.
F IGURE 2. Simplified scheme of the workup of a reaction
mixture from a reaction between P h eK(3) and the correspond-
ing bis-carbamoyl chloride. A cyclic hexamer P h eK(6) could be
obtained from the hexane washings of the toluene fraction. A
pure sample of the less soluble P h eK(9) oligomer could be
obtained from mesitylene. The same procedure can be used in
the naphthalene series where Na p K(9) is obtained by repeated
crystallization from much mesitylene. A difference is that a
cyclic 6-mer is not produced in the naphthalene series.
In Solu tion . Some information on the structure of
these compounds in solution can be obtained from their
NMR spectra. When aromatic rings are placed in close
proximity with fixed orientations dramatic shielding
This is probably due to the instability of the 9,10-
dibenzylaminoanthracene that leads to extensive decom-
position. Thus we conclude that larger central aromatic
moieties than phenyl and naphthyl are not feasible with
the present route.
1
effects can be observed in the H spectra. Protons that
are placed above the ring plane of an aromatic nucleus
will thus have a signal that is shifted toward higher field
while the opposite is observed for protons in the aromatic
plane. Comparison of the 1H spectra of P h eK(3) and P h eK-
(9) with reference compounds 1,4-dibenzylaminobenzene
(1) and N,N′-diacetyl-1,4-dibenzylaminobenzene shows
that the protons attached to the central 1,4-diaminoben-
zene rings of P h eK(3) and P h eK(9) are indeed shifted from
about 7.0 ppm to about 6.15 and 6.0-6.3 ppm, respec-
tively. This shift is in accordance with the zigzag type
structure shown in Figure 1. The benzylic methylene
groups show resonances at ca. 4.0 and 4.5-4.8 ppm. The
first group is assigned to the two benzyl groups at the
ends of the molecules while the other is assigned to the
benzyl groups in the urea linkages. A smaller upfield
shifting of these resonances is also observed.
In the phenylene and the naphthylene series of oligo-
mers further extension of the trimers could be carried
out by first reaction with phosgene to introduce the
carbamoyl groups at each end of the trimers then reaction
with excess trimer in collidine with cesium carbonate,
at reflux for extended periods of time. The extent of the
reaction and the composition of oligomers varied slowly
with time and could be conveniently followed by size
exclusion chromatography (SEC). Using this approach it
was possible to obtain mixtures of oligomers where, e.g.,
the 9-mers predominated. Pure fractions could be ob-
tained by preparative SEC. This however was not an
appealing option since it was our intention to obtain large
enough amounts of the substances for growing crystals
F IGURE 3. Mass spectra (MALDI-TOF) of the crude mixture of oligomers from the synthesis of Na p K(9) on the left and the
purified 9-mer on the right. Silver triflate was added to the matrix facilitating ionization so most peaks are due to a molecular
ion + Ag (107.87 g/mol).
J . Org. Chem, Vol. 67, No. 21, 2002 7513

Krebs and J ørgensen
the locked conformation around the C-N bond also seen
in the X-ray structure (see Figure 5). This is in contrast
to the spectrum of the 1,5-dibenzylaminonaphthalene (2)
where only one resonance at 4.54 ppm is observed. It can
therefore be concluded that the introduction of the
carbamoyl group restricts the geometry around the C-N
bond so that only torsion angles around 90° are allowed.
In the 13C spectrum of compound 5 some of the reso-
nances are split in two. This indicates that two atropi-
somers are present with the groups on the two nitrogen
atoms arranged in either cis or trans configuration.
Similar atropisomerism was observed by J ulia et al. for
a series of N-aryl-N-benzyl alkyl carbamates.7 When the
aryl group was a substituted phenyl group the NMR
coalescence temperature was in the range 300-400 K,
and with naphthyl derivatives it was above 400 K. This
is in accordance with the values obtained in this study.
F IGURE 4. Schematic representation of the three conformers
in the Na p K(3) obtained by rotation around the C-N bond
between the naphthalene and urea units. Benzyl groups and
end groups have been omitted for clarity. The conformers on
the left and right are both present in the crystal structure of
the compound. Dynamic 1H NMR studies show that intercon-
version between the structures becomes rapid in solution above
400 K.
The cyclic hexamer P h eK(6) has a much simpler 1H
spectrum than the linear trimer and 9-mer. Singlets are
observed for the protons on the central 1,4-diaminoben-
zene rings and for the benzyl groups, in accordance with
the symmetry of the structure.
Th e Solid Sta te. As documented by the NMR studies
in solution the geometry of the simple chlorocarbamoyl
monomers is a locked conformation that exhibits hin-
dered rotation. The solution of the crystal structures for
5 and 6 confirmed this expectation. Extensive studies of
the effect on the molecular geometry upon alkylation of
N-arylarylamides and N,N′-dialkyl-N,N′-diarylureas have
shown the geometry to be very different from that of the
corresponding NH-amides and NH-ureas. There is only
one example of an N-benzyl-N-arylarylamide8 in the
CSD9 (Cambridge Structure Database) that can serve for
comparison with the structures presented here (there is
one additional cyclic structure with a locked geometry10
and one additional structure not found in the database).11
There are no examples of N,N′-dibenzyldiarylureas. Most
studies of N-alkylation on these systems have been based
on N-methylation.12 In 5 the chlorine atoms of the two
substituents are on opposite sides of the plane of the
naphthalene ring as shown in Figure 5. This is in direct
opposition to the geometry observed when the aromatic
system is a 9,10-disubstituted anthracene molecule. It
is interesting to note that two positions of the benzyl
groups, which correspond to energy minima, exist and
In the case of Na p K(3), the protons from the naphtha-
lene units give rise to rather broad resonances at 300 K
while the benzyl group signals are partly resolved.
Spectra obtained with 10-deg intervals from 300 to 420
K show that the naphthalene signals first broaden even
more then become sharper until a well-defined pattern
that can be assigned is established above 400 K. This
behavior is typical of a dynamical system and we
interpret this temperature dependence being due to
rotations around the C-N bonds between the naphtha-
lene and urea units. From the X-ray crystallographic
structure of this compound, which will be discussed later,
it is evident that two types of face-to-face geometries
between the naphthalene units are possible. They are
either fully or partially eclipsed, leading to three different
conformers for this system as shown in Figure 4.
The 1H spectrum of 1,5-bischlorocarbamoyl-N,N′-
dibenzylaminonaphthalene (5) shows two doublets at 4.23
and 5.44 ppm that can be assigned to each of the protons
in the benzyl-methylene groups. This splitting is due to
F IGURE 5. Stereoviews of the 5 (above) and 6 (below) systems. While the two benzyl substituents are on opposite sides of the
plane of the naphthalene system in 5 they are on the same side of the plane of the anthracene system in 6 when both of the
molecular geometries are observed in the asymmetric unit.
7514 J . Org. Chem., Vol. 67, No. 21, 2002

Characterization of New Stiff Rod Oligomeric Domains
F IGURE 6. On the left is a stereoview showing the geometry of the P h eK(6) cyclic hexamer. On the right benzyl groups have
been omitted to clearly show the backbone.
F IGURE 7. On the left are stereoviews showing the geometry of the two possible conformers observed in the asymmetric unit
of Na p K(3). On the right monoviews show the cisoid (top) and transoid (bottom) configurations of the backbone with benzyl groups
and hydrogens omitted for clarity.
both are observed in the solid state. This is also the
reason that there are two molecules in the asymmetric
unit, and while this is normally considered rare or
indicative of overseen symmetry in the structure solution
procedure, this is an example of two geometrically
distinct molecular entities in the asymmetric unit that
are not related by symmetry. It is also noteworthy that
the benzyl groups and the chlorine atoms prefer to be
situated at the same side of the plane of the anthracene
ring in the crystal.
flexible at least during the conditions of the synthesis
(boiling collidine at 180 °C). This is supported by the
formation of some cyclic oligomers for the P h eK system.
These oligomers were observed on the SEC (Solvent
Exclusion Chromatography) trace and while it was clear
from the molecular mass obtained this way during the
synthesis of P h eK(9) from P h eK(3) and 4 that it would
have to be the P h eK(6) it was not until a tiny amount of
this byproduct was isolated by preparative SEC and
The advantage of the P h eK-oligomers is that there is
no obvious possibility of having conformers. This is
supported by the NMR studies of both the P h eK(3) and
P h eK(9) system where all the signals are well resolved.
The system must however be considered to be moderately
(12) Itai, A.; Toriumi, Y.; Saito, S.; Kagechika, H.; Shudo, K. J . Am.
Chem. Soc. 1992, 114, 10649. Azumaya, I.; Kagechika, H.; Yamaguchi,
K.; Shudo, K. Tetrahedron 1995, 51, 5277. Yamaguchi, K.; Matsumura,
G.; Kagechika, H.; Azumaya, I.; Ito, Y.; Itai, A.; Shudo, K. J . Am. Chem.
Soc. 1991, 113, 5474. Azumaya, I.; Yamaguchi, K.; Okamoto, I.;
Kagechika, H.; Shudo, K. J . Am. Chem. Soc. 1995, 117, 9083. Hamuro,
Y.; Geib, S. J .; Hamilton, A. D. J . Am. Chem. Soc. 1996, 118, 7529.
Kashino, S.; Matsusita, T.; Iwamoto, T.; Yamaguchi, K.; Haisa, M. Acta
Crystallogr. 1986, C42, 457. Azumaya, I.; Yamaguchi, K.; Kagechika,
H.; Saito, S.; Itai, A.; Shudo, K. J . Pharm. Soc. J pn. 1994, 114, 414.
Williams, D. J . J . Chem. Soc., Chem. Commun. 1977, 170. Harkema,
S.; Gaymans, R. J .; van Hummel, G. J .; Zylberlicht, D. Acta Crystallogr.
1979, B35, 506. Adams, W. W.; Fratini, A. V.; Wiff, D. R. Acta
Crystallogr. 1978, B34, 954. Du Plessis, M. P.; Modro, T. A.; Nassim-
beni, L. R. J . Cryst. Spectrosc. Res. 1983, 13, 179. Bocelli, G.; Rizzoli,
C.; Ori, O. Z. Kristallogr. 1989, 189, 301. Smith, G.; Kennard, C. H.
L.; Katekar, G. F. Aust. J . Chem. 1983, 36, 2455. Northolt, M. G. Eur.
Polym. J . 1974, 10, 799.
(7) J ulia`, S.; Ginebreda, A.; Sala, P.; Sancho, M.; Annunziata, R.;
Cozzi, F. Org. Magn. Reson. 1983, 21, 573.
(8) Rangappa, K. S.; Mallesha, H.; Anilkurnar, N. V.; Yathirajan,
H. S.; Sridhar, M. A.; Lokanath, N. K.; Prasad, J . S. J . Chem. Cryst.
2000, 30, 255.
(9) ConQuest version 1.4; Cambridge Crystallographic Data Centre
(CCDC): 12 Union Road, Cambridge CB2 1EZ, England.
(10) Williams, D. J . Chem. Commun. 1977, 170.
(11) Ganis, P.; Avitabile, G.; Benedetti, E.; Pedone, C.; Goodman,
M. Proc. Natl. Acad. Sci. 1970, 67, 426.
J . Org. Chem, Vol. 67, No. 21, 2002 7515

Krebs and J ørgensen
TABLE 1. Cr ysta llogr a p h ic Da ta for Com p ou n d s 5, 6, Na p K(3), a n d P h eK(6)cyclic
compd
formula
5
C
6
C
Na p K(3)
74H62N6O2
P h eK(6)cyclic
C150H164N12O6
33H28N2O2Cl2
30H22N2O2Cl2
C
formula wt
crystal system
space group
Z
555.47
513.40
1067.30
triclinic
P1h
2230.93
tetragonal
I41/a
triclinic
P1h
1
triclinic
P1h
4
2
8
a, Å
b, Å
c, Å
8.7748(9)
9.2451(9)
9.9577(10)
70.518(2)
84.189(2)
65.367(2)
691.58(12)
1.334
12.186(2)
13.548(3)
16.948(3)
69.354(14)
74.221(14)
89.874(16)
2505.9(8)
1.361
11.7054(14)
15.5692(19)
15.9976(19)
74.834(3)
85.802(2)
81.981(2)
2784.4(6)
1.273
37.842(14)
37.842(14)
16.208(13)
90
90
90
23209(22)
1.277
0.75 × 0.37 × 0.37
Mo KR
0.078
R, deg
â, deg
γ, deg
V, Å3
F, g cm3
cryst dimens, mm3
type of radiation
µ, cm-1
0.50 × 0.25 × 0.13
0.75 × 0.50 × 0.25
0.50 × 0.27 × 0.15
Mo KR
Mo KR
Mo KR
0.269
0.290
0.077
T, K
120(2)
120(2)
120(2)
120(2)
no. of reflns
no. of unique reflns
R(F), Rw(F2) all data
9007
3709
0.0383, 0.0925
32293
13301
0.0827, 0.2595
36320
14899
0.0593, 0.1419
150010
16722
0.2405, 0.6444
crystallized that MALDI-TOF showed it to have a mass
that was P h eK(6) + 26 (+CO - 2H). This was fully
appreciated to be the cyclic hexamer through crystalliza-
tion and crystal structure solution. While the physical
quality of the crystals was very poor, which gave poor
crystallographic data, it was possible to solve the struc-
ture and show that the overall geometry of the molecule
is cyclic. Figure 6 is a stereoview of the cyclic hexamer
obtained from the crystal structure. It is interesting to
compare the geometry of the cyclic P h eK(6) molecule with
that of many of the well-studied macrocycles such as the
cyclophanes (calixarenes, recorsinarenes, etc.) and the
cyclodextrins.
some structural flexibility is allowed, they have a well-
defined predictable structure with amine groups at both
ends suitable for further elaboration. Such entities are
reminiscent of domains in protein structures and we
could envisage a similar role for large building blocks
such as those created in this work.
Exp er im en ta l Section
N,N′-Diben zylben zen e-1,4-d ia m in e (1).4 1,4-Diaminoben-
zene (108 g, 1 mol) and benzaldehyde (210 g) were mixed in
absolute ethanol (250 mL). The mixture became warm and a
thick mass formed. It was then suspended in THF (1 L) and
glacial acetic acid was added (500 mL). The imine was then
reduced by careful addition of sodium borohydride (70 g) until
the yellow/green coloration of the imine disappeared. Water
was then added carefully and the phases were separated. The
dark organic phases were concentrated in a vacuum. The dark
product was recrystallized from toluene/heptane (5 L). Yield
288 g (100%) as colorless plates. 1H NMR (CDCl3, 250.1 MHz,
300 K) δ 3.69 (br s, 2H), 4.32 (s, 4H), 6.64 (s, 4H), 7.29-7.46
(m, 10H). Anal. Calcd for C20H20N2: C, 83.30; H, 6.99; N, 9.71.
Found: C, 82.77; H, 6.92; N, 9.66.
N,N′-Dib en zyl-1,4-bis(ch lor oca r bon yla m in o)b en zen e
(4). Compound 1 (2.9 g, 10 mmol) was added to a phosgene
solution (20 mL, 20% in toluene) and triethylamine (5 mL).
The mixture was heated under reflux for 30 min and then
evaporated to dryness. The salts were removed by trituration
with water and the tan compound recrystallized twice from
CCl4. Yield (3.1 g, 75%); mp 126-7 °C; 1H NMR (CDCl3, 250.1
MHz, 330 K) δ 4.91 (s, 4H), 7.02 (s, 4H), 7.1-7.2 (m, 4H), 7.28-
7.33 (m, 6H). Anal. Calcd for C22H18Cl2N2O2: C, 63.93; H, 4.39;
N, 6.78. Found: C, 63.98; H, 4.13; N, 6.63.
While the P h eK system was considered flexible the Na p
K
system was considered to be less flexible because of the
steric demand and rigidity of the naphthalene unit. The
fact that the naphthalene is 1,5-disubstututed also hosts
the possibility of different conformers as shown in Figure
4. Crystal structure solution of single crystals of Na p K(3)
showed both possible conformers in the asymmetric unit.
The reason for this is ascribed to serendipity and it is
difficult to say whether one conformer will predominate
when the oligomers become larger. Figure 7 shows the
geometry of the two conformers in the asymmetric unit.
Again while there are two molecules in the asymmetric
unit this is understandable as they are crystallographi-
cally distinct molecules that are not related by symmetry.
Con clu sion
Two series of oligomers based on 1,4-dibenzylami-
nobenzene and 1,5-dibenzylaminonaphthalene have been
prepared and characterized both in solution and in the
solid state. N-Alkylation with benzyl groups has been
demonstrated to have the same type of structure gener-
ating effect as methyl groups. This effect changes the
transoid geometry of substituted ureas with remaining
hydrogens on both nitrogen atoms into a structure with
the aryl groups overlapping in a face-to-face arrange-
ment. Interesting dynamic behavior was observed espe-
cially in the case of the 1,5-diaminonaphthalene deriva-
tives, leading to two different types of stack-like orienta-
tions. When several of these substituted urea fragments
are coupled together zigzag stacks are obtained. Although
P h eK(3). Compound 1 (25 g, 86.7 mmol) and Cs2CO3 (60 g,
0.184 mol) were mixed with phosgene (125 mL, 20% in
toluene). A slight pink coloration was observed. Dichlo-
romethane (200 mL) was added and the mixture was stirred
for 0.5 h. The mixture was evaporated to dryness and
compound 1 (50 g, 0.173 mol) was added along with distilled
collidine (250 mL). The reaction was followed with SEC and
stopped after 3 h where complete conversion was observed.
The collidine was evaporated and the residue slurried with
water and filtered and washed with light petroleum. The
material was recrystallized from toluene (1.6 L). Yield 38.65
1
g (49%); mp >260 °C; H NMR (CDCl3, 250.1 MHz, 300 K) δ
3.61 (t, 2H, J ) 4 Hz), 4.04 (d, 4H, J ) 4 Hz), 4.51 (s, 4H),
4.53 (s, 4H), 5.88 (d, 4H, J ) 9 Hz), 6.09 (d, 4H, J ) 9 Hz),
6.15 (s, 4H), 7.1-7.25 (m, 30 H); 13C NMR (CDCl3, 62.9 MHz,
7516 J . Org. Chem., Vol. 67, No. 21, 2002

Characterization of New Stiff Rod Oligomeric Domains
300 K) δ 48.6, 55.7, 56.1, 112.8, 127.1, 127.4, 127.8, 128.5,
128.6, 128.7, 129.0, 129.3, 134.5, 139.0, 139.4, 139.5, 142.1,
146.0, 161.8. Anal. Calcd for C62H56N6O2: C, 81.19; H, 6.15;
N, 9.16. Found: C, 80.99; H, 6.03; N, 8.97.
135.8), (150.8, 150.9), numbers in parentheses are due to
atropisomerism. Anal. Calcd for C26H20Cl2N2O2: C, 67.40; H,
4.35; N, 6.05. Found: C, 67.37; H, 4.26; N, 6.09.
Na p K(3). Cesium carbonate (50 g, 155 mmol) was dried in
an oven at 150 °C for 4 h to remove water. In a dried flask
cesium carbonate and collidine (500 mL, freshly distilled) were
mixed together with compound 2 (75 g, 220 mmol) and
compound 5 (50 g, 110 mmol). The reaction mixture was heated
to reflux under argon for 18 h and then left to cool overnight.
Water (0.5 L) was added with stirring and the slurry was
filtered. The product was triturated with more water, filtered,
and washed with petroleum ether. Finally the product was
recrystallized from toluene (2 L). Yield 57.5 g (49%); mp >260
°C; 1H NMR (CDCl3, 250.1 MHz, 300 K) δ 4.45 (s, 2H), 4.58 (s,
4H), 5.10 (s, 4H), 5.17 (s, 4H), 6.24 (d, 2H, J ) 8 Hz), 6.51 (d,
2H, J ) 7 Hz), 6.68 (t, 4H, J ) 8 Hz), 6.80 (t, 2H, J ) 4 Hz),
7.18-7.55 (m, 32H), 7.64 (d, 6H, J ) 7 Hz). Anal. Calcd for
P h eK(9) a n d th e Cyclic P h eK(6). P h eK(3) (5 g, 5.5 mmol)
was placed in a dried flask containing Cs2CO3 (16 g, 49 mmol)
and methylene dichloride (50 mL). Phosgene (20 mL, 20% in
toluene) was added and the mixture was stirred at 25 °C for
30 min. The mixture was then evaporated to dryness. P h eK(3)
(10 g, 11 mmol) was mixed with collidine (100 mL distilled
from CaH2, must be very dry). The mixture was heated and a
few milliliters of collidine to remove the last traces of moisture.
This solution was then added to the acid chloride and heated
to reflux. The reaction was stopped after 4.5 h where P h eK(9)
was the major product. The mixture was cooled and poured
into water (600 mL) and washed out of the flask with toluene
(100 mL). After being stirred for 10 min the mixture was
filtered and the white powder washed with ether and light
petroleum. The raw product was then purified in the following
way: The solid was stirred in boiling toluene (600 mL) and
left at ambient temperature until the next day where the
product was filtered and washed with toluene and then
petroleum ether. A tiny amount of crystals later formed in the
washings. These were isolated by filtration and dried (0.2 g).
X-ray crystallography and mass spectrometry (MALDI-TOF)
showed this product to be the pure cyclic P h eK(6). Anal. Calcd
for C126H108N12O6‚(H2O)2: C, 78.73; H, 5.87; N, 8.74. Found:
C, 78.92; H, 5.84; N, 8.92. 1H NMR (CDCl3, 250.1 MHz, 300
K) δ 4.46 (s, 24H), 6.46 (s, 24H), 7.0-7.1 (m, 24H), 7.15-7.22
(m, 36H); 13C NMR (CDCl3, 62.9 MHz, 300 K) 55.6, 121.6,
126.8, 127.7, 138.2, 141.8, 161.1.
C
74H62N6O2: C, 83.27; H, 5.86; N, 7.87. Found: C, 83.25; H,
5.83; N, 7.87.
Na p K(9). Na p K(3) (6 g, 5.6 mmol) was placed in a dried flask
containing Cs2CO3 (10 g, 31 mmol) and commercial collidine
(100 mL). Phosgene (6.0 mL, 20% in toluene) was added and
the mixture was stirred at 100 °C for 15 min when Na p K(3)
(12 g, 11.2 mmol) was added and the first sample for SEC
drawn. The mixture was then heated to reflux. After 24 h the
reaction has stopped and the major component was starting
material Na p K(3). The mixture was left to cool. Acetone (100
mL) was added, the mixture was stirred and then poured into
water (1 L), and the crystalline compound was filtered and
washed with water (2 × 250 mL), ether (3 × 200 mL), and
light petroleum (100 mL). The product was then boiled in
toluene (1 L) for 30 min and filtered. Yield 14 g, 77%. An
analytical sample was obtained by repeated fractional crystal-
lization from mesitylene. Mp >260 °C. Anal. Calcd for
The solid product (6.42 g) filtered from the trituration with
toluene was then crystallized from mesitylene to give the pure
P h eK(9). Yield 0.51 g; mp >260 °C. The mother liquor still
contained most of the product. Anal. Calcd for C188H164N18O8:
C, 80.54; H, 5.90; N, 8.99. Found: C, 80.25; H, 5.68; N, 8.93.
1H NMR (CDCl3, 250.1 MHz, 300 K) δ 3.77 (s, 2H), 4.06 (s,
4H), 4.63 (br s, 24H), 4.74 (s, 4H), 4.82 (s, 4H), 5.99 (br s, 24H),
6.16 (d, 4H, J ) 8 Hz), 6.28 (d, 4H, J ) 8 Hz), 6.35 (d, 4H, J
) 8 Hz), 7.2-7.3 (m, 90H).
C
224H182N18O8‚(H2O)3: C, 81.33; H, 5.73; N, 7.62. Found: C,
81.32; H, 5.61; N, 7.31.
9,10-Diben zyla m in oa n th r a cen e (3). 9,10-Bis(N-benzyl-
formamido)anthracene was prepared from 9,10-diformylami-
noanthracene6 by alkylation with benzyl chloride and an excess
of tBuOK in THF. A satisfactory elemental analysis could not
be obtained for this compound, but it was pure enough to use
directly in the next step. 9,10-Bis(N-benzylformamido)an-
thracene (30 g, 67.5 mmol), KOH (85%, 60 g), and ethanol (120
mL) were refluxed under argon; the mixture becomes dark red
and after 20 min a thick homogeneous solution had formed.
It was difficult to isolate the product and many attempts were
made. It was found that (probably) 9,10-bisbenzalimine was
formed (2.6 g obtained upon recrystalization from toluene as
bright red needles). When the toluene liquor was reduced with
a little NaBH4 the color turned dark green (probably due to
the reduction of the alleged acceptor imine). When this was
filtered through silica a bright yellow compound was obtained.
N,N′-Diben zyln a p h th a len e-1,5-d ia m in e (2).4 Naphtha-
lene-1,5-diamine (100 g, 0.632 mol) and benzaldehyde (74 g,
10% excess) were mixed neat with a little acetic acid to prepare
the imide. THF (1 L) was added together with acetic acid (180
mL). The mixture was cooled on an ice-bath and stirred
vigorously while NaBH4 (56 g, 1.48 mol) was added in small
portions. (Ca u tion ! The reaction is exothermic and hydrogen
is evolved.) After 1 h water was added (2 L) slowly to avoid
excessive foaming. Separation of the raw product by filtration
followed by recrystalization from toluene (5 L) gave the product
3 as a white powder. Yield 375 g (80%); mp 187-8 °C; 1H NMR
(CDCl3, 250.1 MHz, 300 K) δ 4.54 (s, 4H), 4.78 (s, 2H), 6.68
(d, 2H, J ) 7 Hz), 7.27-7.52 (m, 14H). Anal. Calcd. for
1
Yield 7 g, 27%; mp 120-2 °C; H NMR (CDCl3, 250.1 MHz,
C
24H22N2: C, 85.17; H, 6.55; N, 8.28. Found: C, 84.89; H, 6.46;
N, 8.29.
N,N′-Diben zyl-1,5-bis(ch lor oca r bon yla m in o)n a p h th a -
300 K) δ 3.8 (br s, 2H), 4.49 (s, 4H), 7.4-7.52 (m, 10H), 7.61
(d, 4H, J ) 7 Hz), 8.3-8.37 (m, 4H); 13C NMR (CDCl3, 62.9
MHz, 300 K) δ 56.6, 124.0, 125.4, 126.3, 127.0, 127.6, 127.8,
128.4, 129.1, 134.5, 137.4, 140.7.
len e (5). 1,5-Dibenzylaminonaphthalene (2) (34 g, 0.10 mol)
was placed in a flask containing chloroform/toluene (1:1) (200
mL) and triethylamine (40 mL, 0.54 mol). Phosgene (150 mL,
20% in toluene) was added. The mixture became warm and
the color changed to a deeper yellow. A precipitate of the
triethylammonium hydrochloride formed. After 1h the mixture
was washed with water and dried (MgSO4). The phase was
evaporated to dryness and the product recrystallized from
toluene (400 mL). It was left in the freezer overnight. After
the product was filtered and washed with light petroleum it
was dried in the vacuum oven at 80 °C for 4 h. Yield 46 g (99%);
N,N′-Diben zyl-9,10-bis(ch lor oca r bon yla m in o)a n th r a -
cen e (6). The amine 3 (1 g, 2.6 mmol) was mixed with
phosgene (10 mL, 20% in toluene) and triethylamine (5 mL)
in methylene chloride (100 mL). The mixture was stirred for
1 h. A precipitate had formed. The mixture was poured into
water (100 mL) and the organic phase was separated, dried
(MgSO4), and evaporated to give a dark oil (toluene and
triethylamine); light petroleum (50 mL) was added giving a
precipitate that was filtered (0.5 g), which contained the
product and some impurities. When light petroleum wss added
to the liquor yellow crystals formed slowly, whcih was found
1
mp 178-9 °C; H NMR (CDCl3, 250.1 MHz, 300 K) δ 4.53 (d,
1
2H, J ) 14 Hz), 5.44 (d, 2H, J ) 14 Hz), 7.10 (d, 2H, J ) 8
Hz), 7.2-7.33 (m, 10H), 7.46 (t, 2H, J ) 8 Hz), 7.81 (d, 2H, J
) 8 Hz); 13C NMR (CDCl3, 62.9 MHz, 300 K) δ 56.7, 124.0,
127.3, (128.4, 128.5), 128.8, (129.0, 129.1), 129.9, 131.7, (135.7,
to be the pure product. Yield 0.4 g (30%); mp 139-40 °C; H
NMR (CDCl3, 250.1 MHz, 300 K) δ 5.09 (s, 4H), 6.94 (d, 4H, J
) 7 Hz), 7.10 (t, 4H, J ) 7 Hz), 7.21 (d, 2H, J ) 7 Hz), 7.38-
7.42 (m, 4H), 7.65-7.71 (m, 4H); 13C NMR (CDCl3, 62.9 MHz,
J . Org. Chem, Vol. 67, No. 21, 2002 7517

Krebs and J ørgensen
300 K) δ 57.3, 123.7, 127.7, 128.7, 128.8, 129.5, 130.8, 134.3,
134.9, 151.5. Anal. Calcd for C30H22Cl2N2O2‚H2O: C, 68.97; H,
4.44; N, 5.36. Found: C, 68.97; H, 4.42; N, 5.65.
An almost complete sphere of reciprocal space was covered by
a combination of several sets of exposure frames: each set with
a different æ angle for the crystal and each frame covering a
scan of 0.3° in ω. Data collection, integration of frame data,
and conversion to intensities corrected for Lorenz, polarization,
and absorption effects were performed with the programs
SMART,15 SAINT,15 and SADABS.16 Structure solution, refine-
ment of the structures, structure analysis, and production of
crystallographic illustrations were carried out with the pro-
grams SHELXS97,17 SHELXL97,17 and SHELXTL.18 The
structures were checked for higher symmetry and none was
found.19 In all of the structures H atoms were included and
their calculated positions. Crystallographic data (excluding
structure factors) for the structures reported in this paper have
been deposited with the Cambridge Crystallographic Data
Centre as supplementary publication no. CCDC-186137, CCDC-
186138, CCDC-186139, and CCDC-186140. Copies of the data
can be obtained free of charge on application to CCDC, 12
Union Road, Cambridge CB2 1EZ, UK (fax (+44)1223-336-
033; e-mail deposit@ccdc.cam.ac.uk).
Cr ysta llogr a p h y. General crystallographic data can be
found in Table 1. The crystals were drawn directly from the
mother liquor, coated with a thin layer of protecting oil, and
mounted on glass fibers with use of Apiezon grease and
transferred quickly to the cold stream of nitrogen (Oxford
Cryostream) on the diffractometer (Siemens SMART CCD
Platform). Crystals of 5 and cyclic P h eK(6) lost solvent upon
exposure to air and this impaired the quality of the data due
to a degradation of the crystalline properties. In compound 5
the solvent toluene molecule was situated close to the center
of symmetry, which was modeled as two toluene solvent
molecules with a sof of 0.5. The cyclic P h eK(6) crystallizes in
the tetragonal spacegroup I41/a. This gives rise to large
channels extending throughout the crystal along the unique
axis that contains solvent hexane molecules. Similar problems
with poor data quality due to crystal degradation have been
observed in similar crystallographic problems with solvent-
containing channels in calixarenes.13 Further the problems
associated with larger observed R-factors for crystals with
many weakly scattering atoms in the asymmetric unit have
been shown.14 The solution to this structure was particularly
problematic, and while the quality of the data does not allow
any detailed crystallographic detail to be extracted, the con-
nectivity, molecular conformation, and packing properties
could be determined. The benzene rings were constrained to
approach a hexagon and the symmetry constraints on the
solvent hexane molecules positioned close to a center of
symmetry were modeled by neglecting the symmetry con-
straints and keeping the sof at a value of 1. In the case of 6
and Na p K(3) two molecules were found in the asymmetric unit,
and while this is normally considered rare the molecular
conformation of the two different molecules was distinct,
making the consideration of overseen symmetry unnecessary.
Ack n ow led gm en t. This work was supported by the
Danish Technical Science Foundation of Denmark
(STVF).
Su p p or tin g In for m a tion Ava ila ble: Tables of fractional
coordinates, equivalent isotropic and anisotropic thermal
parameters, bond lengths, and bond angles. This material is
available free of charge via the Internet at http://pubs.acs.org.
J O025980F
(15) Siemens, 1995; SMART and SAINT, Area-Detector Control and
Integration Software; Siemens Analytical X-ray Instruments Inc.:
Madison, WI, USA.
(16) Sheldrick, G. M. SADABS; Empirical absorption program
written for the Siemens SMART platform.
(17) Sheldrick, G. M. SHELX-97, Program for structure solution and
refinement; 1997.
(13) Krebs, F. C.; J ørgensen, M. J . Chem. Soc., Perkin Trans. 2 2000,
1935-1941.
(14) Krebs, F. C. J . Appl. Crystallogr. 2000, 33, 392-393.
(18) Sheldrick, G. M. SHELXTL95; Siemens Analytical X-ray
Instruments Inc., Madison WI, 1995..
(19) Spek, A. L. Acta Crystallogr. 1990, A46, C-31.
7518 J . Org. Chem., Vol. 67, No. 21, 2002