New liquid crystalline materials based on two generations of dendronised cyclophosphazenes

Article abstract of DOI:10.1002/chem.201202748

A divergent approach was used for the synthesis of dendritic structures based on a cyclotriphosphazene core with 12 or 24 hydroxyl groups, by starting from [N₃P₃(OC₆H₄OH-4)₆] and using an acetal-prote

Full text of DOI:10.1002/chem.201202748

FULL PAPER
DOI: 10.1002/chem.201202748
New Liquid Crystalline Materials Based on Two Generations of Dendronised
Cyclophosphazenes
Josefina Jimꢀnez,*[a] Antonio Laguna,[a] Elena Gascꢁn,[a] Josꢀ Antonio Sanz,[a]
Josꢀ Luis Serrano,[b] Joaquꢂn Barberꢃ,[c] and Luis Oriol*[c]
1
Abstract: A divergent approach was
used for the synthesis of dendritic
structures based on a cyclotriphospha-
zene core with 12 or 24 hydroxyl
by using 31P{1H} and H NMR spectro-
scopy and the chemical structure of the
resulting materials was confirmed by
using different spectroscopic tech-
The dendrimer with 12 4-pentylbi-
AHCTUNGTRENGNpUN henyl mesogenic units gives rise to
columnar rectangular organisation,
whereas the one with 24 pentylbi-
groups,
[N3P3(OC6H4OH-4)6] and using an
acetal-protected 2,2-di(hydroxy-
methyl)propionic anhydride as the
acylating agent. Hydroxyl groups in
these first- and second-generation den-
drimers, G1-(OH)12 or G2-(OH)24,
were then condensed in turn with
mono- or polycatenar pro-mesogenic
acids to study their ability to promote
self-assembly into liquid crystalline
structures. Reactions were monitored
by
starting
from
G
and
mass
spectrometry
ACHTUNGTRENNpUNG henyl units does not exhibit mesomor-
phic behaviour. In the case of materials
that contain polycatenar pro-mesogenic
units with two aromatic rings (A4 vs.
A5), the incorporation of a short flexi-
ble spacer connected to the periphery
of the dendron (acid A5) was needed
to achieve mesomorphic organisation.
In this case, both dendrimer genera-
tions G1A5 and G2A5 exhibit a hex-
agonal columnar mesophase.
G
cyclotriphosphazene
G
Introduction
tailored by the choice of appropriate functional groups.[1]
Phosphazenes have attracted interest for such applications
as flame-retardant and thermally stable macromolecules,
low-temperature elastomers, biomaterials, photo-sensitive
and/or photo-inert substrates, solid-state electrolytes, varia-
ble types of hybrid membranes, solid-state fuel cells, optical
and electro-optical polymers.[1c,2] We have also recently con-
tributed to this field, reporting the first fully carborane-sub-
stituted cyclo- and polyphosphazenes,[3] gold or silver
Phosphazenes comprise a broad class of molecules based on
the repeating unit [NPR2], which includes cyclic or linear
oligomers and polymers. The most striking characteristic of
this type of compound is its associated synthetic versatility,
which enables the introduction of almost any substituent
group R at the phosphorus and allows the properties to be
metalloACHTNUTRGNEpNUG hosphazenes that are promising for catalytic appli-
cations,[4] and cyclophosphazenes with liquid-crystalline
properties.[5,6] These monodisperse liquid-crystal (LC) cyclo-
triphosphazenes have been less explored than the linear
[a] Dr. J. Jimꢀnez, Prof. A. Laguna, E. Gascꢁn, J. A. Sanz
Departamento de Quꢂmica Inorgꢃnica
Facultad de Ciencias
polyACHTNUGRTNEUNG
meric analogues.[7] In this regard, we have reported a
Instituto de Sꢂntesis Quꢂmica y Catꢃlisis Homogꢀnea (ISQCH)
Universidad de Zaragoza-C.S.I.C.
series of these non-conventional mesogens, which have the
pro-mesogenic units directly attached to the inorganic ring.
All of them were synthesised from a six-armed cyclotriphos-
phazene core, [N3P3(OC6H4OH-4)6], by esterification of the
peripheral hydroxyl groups with mono- or polycatenar ben-
zoic acid derivatives, and their mesomorphic properties
depend on a suitable volume balance between the rigid core
and the alkyl chains in the pro-mesogenic group, which
ranges from calamitic, discotic, to cubic mesomorphism
when the number of alkyl chains increases.[5,6] Calamitic
mesophases were promoted by monocatenar units,[5a,8]
whereas columnar or cubic phases resulted from the incor-
poration of polycatenar units.[5,6]
Pedro Cerbuna 12, 50009 Zaragoza (Spain)
Fax: (+34)976761187
E-mail: jjimvil@unizar.es
[b] Prof. J. L. Serrano
Departamento de Quꢂmica Orgꢃnica
Facultad de Ciencias
Instituto Universitario de Nanociencia de Aragꢁn
Universidad de Zaragoza
Mariano Esquillor Edif. I+D, 50018, Zaragoza (Spain)
[c] Prof. J. Barberꢃ, Dr. L. Oriol
Instituto de Ciencia de Materiales de Aragꢁn (ICMA)
Universidad de Zaragoza-C.S.I.C
Departamento de Quꢂmica Orgꢃnica, Facultad de Ciencias
Pedro Cerbuna 12, 50009 Zaragoza (Spain)
Fax : (+34)976762276
E-mail: loriol@unizar.es
Chem. Eur. J. 2012, 18, 16801 – 16814
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
16801

Cyclophosphazenes are also very appropriate to use as
the core of dendrimers due to the fact that this inorganic
core can be easily functionalised at the phosphorus atoms,
as mentioned above, and thus give rise to a wide variety of
branched molecules.[9,10] Indeed, they lead to an increase in
the number of end groups for an equal number of synthetic
steps, in contrast to the more commonly used trisubstituted
cores, which saves time and is a very valuable property for
the synthesis of dendrimers. Majoral et al. have demonstrat-
ed that this inorganic core can be used to obtain dendritic
compounds with very original architectures.[9,11] Other inter-
esting characteristics of cyclophosphazenes as cores, which
have also been indicated by Majoral, are that they induce
the appearance of internal cavities close to the core more
easily than trifunctional cores and that they should easily
give spherical structures as a consequence of the highly
functionalised core.[9] Therefore, various interesting applica-
tions of these dendrimers have been proposed in the fields
of catalysis,[9,12] medicinal chemistry[13] and materials sci-
ence,[14] even if none of these applications appears related to
the nature of the core. Moreover, it has been emphasised
that applications of dendrimers are still in their infancy and
the structural diversity offered by varying the nature of the
core should help to develop forthcoming applications.[9]
One of the most active lines of dendrimer research is the
synthesis and study of liquid crystals. Despite the fact that
the first works in this field appeared in the late 80s and
early 90s and used self-assembling dendrons, dendronised
polymers and dendrimers based on conformational isomer-
ism,[15] the most significant results in this topic have only re-
cently appeared, mainly motivated by their applications.[16]
The most common approach to obtain LC dendrimers is the
attachment of pro-mesogenic units to polyfunctional den-
dritic platforms. In these derivatives, the focal point of the
dendrimer is substituted by a complex structure, such as
poly
AHCTUNGTRENNUNG
ACHTUNGTRENaNGNU romatic scaffolds, macrocycles, polymers and oligo-
mers.[17] According to this strategy, the structure of cyclo-
phosphazene seems to be especially useful as a central core
because of its singular branched geometry. We should also
highlight that the most commonly used dendrimers for these
kinds of materials have been poly(propylene imine), poly-
AHCTUNGTREG(NNUN amidoamine), and polyester-based and silicon-containing
dendrimers (siloxanes, carbosilanes, carbosilazanes), but no
cyclophosphazene-based dendrimers have been used to date
to our knowledge.[16]
Herein we present some examples of this approach to
new mesogenic dendrimeric derivatives based on cyclophos-
phazene as the central core. First and second generations of
dendrimers with this central core have been synthesised by
attaching dendrons derived from 2,2-di(hydroxymethyl)pro-
pionic acid described by Hult[18] and Frꢀchet.[19] They have
been further functionalised at their terminal positions with
different mono- and polycatenar pro-mesogenic units (see
Figure 1) and the ability of the resulting materials to pro-
mote self-assembly into liquid-crystalline structures have
been studied by different techniques, such as polarising opti-
cal microscopy, differential scanning calorimetry (DSC) and
X-ray diffraction measurements.
Abstract in Spanish: En este trabajo se ha usado un mꢀtodo
divergente para obtener estructuras dendrꢁticas no convencio-
nales basadas en un nfflcleo de ciclotrifosfazeno con doce o
veinticuatro grupos hidroxilo, partiendo de [N3P3(OC6H4OH-
4)6] y utilizando como agente acilante un anhꢁdrido protegi-
do derivado del µcido 2,2-bis(hidroximetil)propiónico. Los
grupos hidroxilo en estos dendrꢁmeros de primera y de segun-
da generación, G1-(OH)12 or G2-(OH)24, se condensaron
con µcidos promesógenos mono o policatenares para estudiar
su capacidad para promover estructuras cristal lꢁquido. Las
reacciones se monitorizaron por espectroscopia de RMN de
Results and Discussion
Synthesis and structural characterisation: The approach
used for the two hydroxyl-terminated cyclotriphosphazene
dendrimers, G1-(OH)12 and G2-(OH)24, is shown in
Scheme 1. It involves a divergent strategy starting from the
hexa(4-hydroxyphenoxy)cyclotriphosphazene
core,
1
31P{1H} y H, y la estructura quꢁmica de los materiales resul-
[N3P3(OC6H4OH-4)6], and using the benzylidene-protected
2,2-di(hydroxymethyl)propionic anhydride as the building
block to achieve the different generations by following a
previously described method.[19] This Frꢀchetꢅs synthetic
method is remarkable because, unlike other methods, it only
requires a small excess of reagent to achieve a quantitative
growth of the dendrimer generation. Coupling of this anhy-
dride with the phenolic cyclotriphosphazene core is readily
accomplished by using 1.6 mol of the anhydride per mol of
hydroxyphenyl group in a 1:3 pyridine/CH2Cl2 mixture (or
THF/CH2Cl2) with 20 to 25 mol% of DMAP as the acylat-
ing catalyst. As in the Frꢀchetꢅs method with other phenolic
core molecules,[19] it requires no means of purification other
than a simple solvent extraction or precipitation (see Exper-
imental Section). Thus, after obtaining the first-generation
dendrimer, G1-(O2Bn)6, the next step of the synthesis in-
volved a Pd-catalysed removal of the benzylidene protecting
tantes se confirmó mediante diferentes tꢀcnicas espectroscópi-
cas y espectrometrꢁa de masas (MALDI-TOF). Los resulta-
dos estµn de acuerdo con ciclofosfazenos monodispersos y
completamente funcionalizados. Las propiedades tꢀrmicas y
mesomorfas se estudiaron por microscopꢁa óptica, calorime-
trꢁa diferencial de barrido y difracción de rayos-X. El dendrꢁ-
mero que tiene 12 unidades mesógenas 4-pentilbifenilo da
lugar a organizaciones columnares, mientras que aquel con
24 unidades de este tipo es un material isótropo. En el caso
de las unidades promesógenas policatenares con dos anillos
aromµticos se necesitó incorporar un corto espaciador flexi-
ble en la periferia del dendrón para conseguir organizaciones
mesomorfas en ambos dendrꢁmeros, de primera y segunda
generación. En este caso, ambos dendrꢁmeros exhiben una
mesofase columnar hexagonal.
16802
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16801 – 16814

New Liquid Crystalline Materials
FULL PAPER
Figure 1. Chemical structure and nomenclature of the cyclophosphazene-based dendrimers.
groups under medium pressure of H2, which gives rise to the
corresponding first-generation hydroxyl-functional dendrim-
er, G1-(OH)12, in 97% yield after purification by precipita-
tion.
groups in these first- and second-generation dendrimers
were condensed with mono and polycatenar acids A1–A5,
as shown in Scheme 2 by following a similar method to the
one used previously to prepare liquid crystals by direct es-
terification of the N3P3(OC6H4OH-4)6] central core with
pro-mesogenic acids.[6] This was carried out in the presence
of N,N’-dicyclohexylcarbodiimide (DCC) and 4-(dimethyl-
The sequence of the anhydride coupling step and the sub-
sequent deprotection by hydrogenolysis is repeated to pre-
pare the corresponding monodisperse second-generation
dendrimers, G2-(O2Bn)12 and G2-(OH)24 (Scheme 1), which
were also obtained in high yields under similar reaction and
purification conditions (see the Experimental Section). All
the reactions were easily monitored by using 31P{1H} and
1H NMR spectroscopy with (CD3)2CO as the solvent.
31P{1H} NMR spectroscopy was invaluable for checking that
the complete reaction had taken place by monitoring the ap-
pearance of a single resonance in the spectrum, and
1H NMR spectroscopy was needed to monitor the complete
removal of the benzylidene protective groups. We also ex-
plored an alternative approach, involving cyclohexene in
ethanol under reflux and the presence of Pd(OH)2 as a cata-
lyst, to remove the benzylidene protecting groups. This
method led to lower conversions and longer reaction times.
In addition, decomposition products were frequently ob-
served and the mixture could not be purified by convention-
al methods (extraction, precipitation or chromatographic
techniques).
AHCTUNGTRENNaGUN mino)pyridinium 4-toluenesulfonate (DPTS), with CH2Cl2
as the solvent, to give new dendrimers G1-An and G2-An
(n=1–5, see Scheme 2), respectively. Complete esterifica-
tion of all hydroxyl groups in G1-(OH)12 or G2-(OH)24 re-
quired a greater excess of carboxylic acid (2.6 mol of car-
boxylic acid per mol of OH group) than in the similar reac-
tion we performed with [N3P3(OC6H4OH-4)6].[6]
Reactions were also monitored by using 31P{1H} NMR
spectroscopy to check that complete esterification of the hy-
droxyl groups of the starting cyclotriphosphazene dendrimer
took place, which needed long reaction times (see the Ex-
perimental Section). Compounds G1-An and G2-An (n=1–
5) were also easily purified from the remaining N,N'-dicyclo-
hexylurea (DCU), excess DCC and the corresponding acid
by precipitation of the CH2Cl2 solution into DMSO and
ethanol under vigorous stirring.
All of these dendrimers were characterised by using ele-
1
mental analysis, IR, H and 13C{1H} NMR spectroscopy and
Dendrimers G1-(OH)12 and G2(OH)24 can be used as
platforms to prepare functional materials by functionalisa-
tion of the peripheral hydroxyl groups. In this case, hydroxyl
MALDI-TOF mass spectrometry. All these data are given
in the Experimental Section and confirm the expected struc-
ture. The IR spectra of all these compounds feature peaks
Chem. Eur. J. 2012, 18, 16801 – 16814
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16803

J. Jimꢀnez, L. Oriol et al.
13C{1H} NMR spectra are also
consistent with the formulae in-
dicated. For example, in addi-
tion to signals for the protons
in the aromatic ring connected
to
phosphazene
(OC6H4O,
AA’BB’ spin system), the
1H NMR spectrum of G1-
(O2Bn)6 contains resonances for
the benzylidene-2,2-bis(oxyme-
thyl)propionate group at d=
1.11 (s, CH3), at 4.71 (d) and
3.66 (d; AB system, OCH2),
and at 7.45–7.33 (m, C6H5) and
5.47 ppm (s, CHPh). These last
signals of the benzylidene unit
were replaced by a signal at d=
4.93 ppm (t) due to OH protons
after conversion to G1-(OH)12.
The 1H NMR spectra of first-
generation dendrimers G1-An
(n=1–5) also showed resonan-
ces corresponding to the other
aromatic protons and the pro-
tons of the alkyl chains (see
Table 1). The different types of
protons in these compounds are
shown in Figure 2 to make the
data in Table 1 clearer.
The 1H NMR spectroscopic
data of the second-generation
dendrimers, G2-(O2Bn)12, G2-
(OH)24 and G2-An (n=1–5).
are collected in Table 2. The
different type of protons in
these compounds is also shown
in Figure 3 to make the data in
Table 2 clearer. The 31P{1H} and
1H NMR spectra of G2-A1 are
shown in Figure 4. The signals
in the 13C{1H} NMR spectra
were assigned by comparison
with those of the corresponding
carboxylic acids and with those
observed for other similar com-
pounds.[5]
Scheme 1.
A MALDI-TOF mass spec-
trometry study of all dendrim-
at around 1200 cmꢀ1 (s; P=N),[20] which provide evidence for
the presence of the six-membered phosphazene ring. The
31P{1H} NMR spectra consist of a singlet at around d=8 to
10 ppm, similar to those observed for other (phenoxy)cyclo-
triphosphazenes.[5,6,21] In the second-generation dendrimers
this signal is displaced toward a higher field than that of
first-generation dendrimers, a decrease of d=0.6 to 1 ppm
being observed in the chemical shift with increasing genera-
tion (see the Experimental Section). The 1H and
ers was carried out by using dithranol (or dithranol-NaTFA)
as the matrix. In all cases a peak for the molecular mass of
the 12- or 24-functionalised cyclotriphosphazene was ob-
served, which confirmed the expected chemical structure
(see Figure 5). Furthermore, no peaks corresponding to de-
rivatives with incomplete functionalisation were detected,
which is in accordance with the spectroscopic results and
confirms the monodisperse nature of these dendrimers.
16804
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16801 – 16814

New Liquid Crystalline Materials
FULL PAPER
Scheme 2.
Thermal and liquid-crystalline properties: The thermal tran-
sitions and liquid-crystalline properties of these dendrimers
were first studied by polarising optical microscopy (POM)
and differential scanning calorimetry (DSC), and the nature
of the detected mesophases was corroborated by X-ray dif-
fraction measurements (XRD). Table 3 summarises the ther-
mal transitions and mesophases detected in these materials.
Acids A1 and A2 were previously used in the preparation
of mesomorphic cyclotriphosphazenes, in which were they
directly joined to the [N3P3(OC6H4OH-4)6] central rigid
core.[5] When dendrimers G1-(OH)12 and G2-(OH)24 were
modified with these acids, the resulting materials did not ex-
hibit mesomorphic properties, probably due to the low an-
slightly higher for the higher-generation dendrimer. Con-
versely, both dendrimers derived from acid A2 are viscous
liquids at room temperature and no transition was detected
in the DSC study under the experimental conditions.
The monocatenar and commercial A3 acid has been
widely used in the preparation of calamitic liquid crystals,
and was selected as a functional group with a high length/di-
ameter ratio to modify the periphery of dendrimers and
induce mesomorphic organisations. The first-generation den-
drimer obtained by using this acid, G1-A3, is solid at room
temperature and progressively softens at above 608C, which
gives rise to a fluid phase that exhibits birefringence and
shows textures typical of a liquid-crystal phase, although
that is difficult to assign. Isotropisation is detected at above
1108C. When the sample is then cooled and annealed close
to the transition from the isotropic to the mesomorphic
phase, a pseudo-fan shaped develops (see Figure 6). This
ACHTUNGTRENNUNGisotropy of these acids, which contain only one aromatic
ring. Both G1-A1 and G2-A1 are amorphous materials (no
evidence of crystallisation was detected by either POM or
DSC) with a Tg (glass transition) at around 08C, which is
Chem. Eur. J. 2012, 18, 16801 – 16814
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16805

J. Jimꢀnez, L. Oriol et al.
Table 1. 1H NMR spectroscopic data for first generation dendrimers and corresponding carboxylic acids for comparison.[a]
d
(OC6H4O) [ppm]
d
E
d(C
(CH3)) [ppm]
d
N
R
d
N
(C-C6H2) [ppm]
(Ha)
(Hb)
(Hc)
(Hd)
(He)
G1-(O2Bn)6
G1-(OH)12
G1-A1
7.01, 6.95
(AB system, 4H)
7.03, 6.91
(AB system, 4H)
6.90, 6.85
(AB system, 4H)
–
4.71 (d, 2H)
3.66 (d, 2H)
3.66, 3.55
(ABX system, 4H)
4.66, 4.63
1.11 (s, 3H)
–
–
1.20
–
–
–
–
–
ACHTUNGTRNE(NUNG s, 3H)
[b]
7.91 (d, 4H)
6.83 (d, 4H)
8.05 (d 2H)
6.93 (d 2H)
7.19 (s, 4H)
(AB system, 4H)
–
A1
–
[b]
G1-A2
6.97, 6.92
4.72, 4.62
(AB system, 4H)
(AB system, 4H)
A2
G1-A3
–
–
–
–
7.33 (s, 2H)
7.45 (d, 4H)
7.21 (d, 4H)
7.66 (d, 2H)
7.34 (d, 2H)
7.36 (s, 4H)
[b]
6.93, 6.84
(AB system, 4H)
–
4.74, 4.71
(AB system, 4H)
–
8.00 (d, 4H)
7.54 (d, 4H)
8.11 (d, 2H)
7.79 (d, 2H)
8.06 (d, 4H)
7.24 (d, 4H)
8.21 (d, 2H)
7.33 (d, 2H)
7.07 (d, 4H)
6.89 (d, 4H)
7.09 (d, 2H)
6.92 (d, 2H)
A3
–
[b]
G1-A4
A4
6.98, 6.92
(AB system, 4H)
–
4.78, 4.69
(AB system, 4H)
–
–
7.42 (s, 2H)
7.38 (s, 4H)
7.40 (s, 2H)
[b]
G1-A5
A5
6.99
4.45, 4.38
(AB system, 4H)
–
(AB not solved, 4H)
–
–
[a] Data recorded in CDCl3 as the solvent except for G1-(OH)12 ((CD3)2SO) and for A3 ((CD3)2CO). See Figure 2 for the proton assignments. [b] The
singlet assigned to C(CH3) in G1-An (n=1–5) is hidden by signals corresponding to alkyl terminal chains (see also Figure 2). The additional resonances
ACHTUNGTRENNUNG
for benzylidene in G1-(O2Bn)6, hydroxide in G1-(OH)12 and alkyl protons in G1-An (n=1–5) are omitted for clarity; they are given in the Experimental
Section.
1
Figure 2. Nomenclature of proton signals detected in the H NMR spectra of G1 dendrimers (see Table 1).
texture is retained as a consequence of vitrification on cool-
ing the sample to room temperature, and seems to corre-
spond to a columnar organisation (see below), despite the
fact that a monocatenar calamitic pro-mesogenic unit was
used to functionalise the periphery of the dendrimer. These
optical observations were corroborated by using DSC. In
fact, this material only exhibits two thermal transitions. On
heating, a glass transition at approximately 608C is ob-
served, followed by an endothermic peak at about 1108C
that corresponds to the mesophase–isotropic phase transi-
tion. On cooling, scans of both transitions were also detect-
ed and no evidence of crystallisation is observed, which con-
firms the mesomorphic glassy nature of this dendrimer at
room temperature (see Figure 7).
birefringent textures when heated from room temperature
and observed by using POM. Accordingly, only a glass tran-
sition was observed at around 408C in successive heating
and cooling DSC scans, and no additional peaks were de-
tected. These results suggest that this dendrimer is a non-
mesomorphic, amorphous material.
As mentioned above, in previous papers we introduced
polycatenar aromatic acids directly linked to the hexa(4-hy-
droxyphenyloxy)phosphazene central core to obtain materi-
als exhibiting columnar or cubic phases depending on the
number of terminal alkyl chains. In those cases, the meso-
genic units joined to the central core had two aromatic
rings.[5,6] According to this strategy and, in concordance with
the mesogenic units previously used, we selected acid A4 as
a functional group to generate liquid crystallinity in G1-
(OH)12 and G2-(OH)24. However, resulting dendrimers G1-
A4 and G2-A4 were found to be optically isotropic by using
POM. Furthermore, at around room temperature, the DSC
Given the results for G1-A3, it should be expected that
the incorporation of 24 peripheral mesogenic units derived
from A3 (dendrimer G2-A3) would result in a new meso-
morphic material. However, this dendrimer did not exhibit
16806
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16801 – 16814

New Liquid Crystalline Materials
FULL PAPER
Table 2. 1H NMR spectroscopic data for second-generation dendrimers and the corresponding carboxylic acids for comparison.[a]
d
(OC6H4O) [ppm]
d
E
d(C
(CH3)) [ppm]
d
N
R
d
N
(C-C6H2) [ppm]
(Ha)
(Hb and Hd)
(Hc and He)
(Hf)
(Hg)
G2-(O2Bn)12
G2-(OH)24
G2-A1
6.89, 6.77
(AB system,4H)
4.57 (d, 4H),
4.50, 4.47
(AB system, 4H),
3.58 (d, 4H)
4.43, 4.37
(AB system, 4H)
3.66, 3.58
(AB system, 8H)
4.49, 4.44
1.38(s, 3H)
0.91 (s, 6H)
ACHTUNGTRENNUNG
–
–
7.06, 6.91
(AB system, 4H)
1.44 (s, 3H)
1.12 (s, 6H)
–
–
–
[b]
6.95
7.86 (d, 8H)
6.80 (d, 8H)
(AB not solved, 4H)
(AB system, 8H)
4.43, 4.38
(AB system, 4H)
–
A1
–
–
–
–
8.05 (d, 2H)
6.93 (d, 2H)
7.14 (s, 8H)
[b]
G2-A2
7.06, 6.99
(AB system, 4H)
–
6.95
4.63 (d, 4H)
4.38 (m, 8H)
–
4.59, 4.52
(AB system, 8H)
4.51, 4.45
(AB system, 4H)
–
A2
G2-A3
–
–
7.33 (s, 2H)
7.42 (d, 8H)
7.17 (d, 8H)
[b]
7.96 (d, 8H)
7.51 (d, 8H)
(AB not solved, 4H)
A3
–
–
8.11 (d, 2H)
7.79 (d, 2H)
7.99 (brd, 8H)
7.19 (brd, 8H)
8.21 (d, 2H)
7.33 (d, 2H)
7.04 (d, 8H)
6.86 (d, 8H)
7.66 (d, 2H)
7.34 (d, 2H)
7.32 (s, 8H)
[b]
G2-A4
A4
7.04 (brm, 4H)
–
4.69–4.39 (m, 12H)
–
–
7.42 (s, 2H)
7.35 (s, 8H)
[b]
G2-A5
7.09 (m, 4H)
4.46, 4.39
(AB system, 4H)
4.30, 4.22
(AB system, 8H)
–
A5
–
–
7.09 (d, 2H)
6.92 (d, 2H)
7.40 (s, 2H)
[a] Data using CDCl3 as solvent except for G2-(OH)12 (MeOD) and for A3 ((CD3)2CO). See Figure 3 for the proton assignments. [b] The singlet assigned
to C(CH3) for G2-An (n=1–5) is hidden by signals corresponding to alkyl terminal chains (see also Figure 3). The additional resonances for benzylidene
ACHTUNGTRENNUNG
in G2-(O2Bn)12, hydroxide in G2-(OH)24 and alkyl protons in G2-An (n=1–5) are omitted for clarity; they are given in the Experimental Section.
1
Figure 3. Nomenclature of proton signals detected in the H NMR spectra of G2 dendrimers (see Table 2).
traces only exhibited a baseline jump that corresponded to
the glass transition but no peaks due to crystallisation,
mesoACHTUNGTRENNUNGphase–isotropic phase or any other endothermic or
exothermic transition were observed. These facts suggest
that these dendrimers, like G2-A3, are non-mesomorphic
amorphous materials.
In an attempt to induce liquid crystallinity, acid A5, which
is very similar to A4, was synthesised and incorporated into
the periphery of dendrimers G1-(OH)12 and G2-(OH)24
through a short flexible spacer with three methylene groups.
The resulting dendrimers, G1-A5 and G2-A5, are highly vis-
cous liquids at room temperature and exhibit birefringent
textures due to a liquid-crystalline phase that becomes iso-
tropic at temperatures above 70 or 808C for G1-A5 and G2-
A5, respectively. When they were cooled from the isotropic
state, the liquid-crystalline phase was observed again and
they exhibited typical textures of a hexagonal columnar
phase (see Figure 8), which was retained at room tempera-
ture. The DSC study of these materials confirmed these op-
tical observations and, accordingly, only a glass transition at
Chem. Eur. J. 2012, 18, 16801 – 16814
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16807

J. Jimꢀnez, L. Oriol et al.
Table 3. Thermal transitions and mesomorphic properties of the synthes-
ised dendrimers.[a]
Thermal transition [8C]
DH (DH)[b] [kJmolꢀ1
]
G1-A1
G2-A1
G1-A2
G2-A2
G1-A3
G2-A3
G1-A4
G2-A4
G1-A5
G2-A5
g ꢀ3 I
g 5 I
I[c]
I[c]
g 62 Colr 110 I
g 40 I
g 24 I
g 27 I
g 5 Colh 70 I
g 7 Colh 83 I
33.6, 2.8
14.3, 1.2
38.6, 1.6
[a] Thermal transition corresponding to the second DSC heating scan.
g=glassy state (either amorphous or mesomorphic glass), I=Isotropic
fluid phase; Colr =rectangular columnar mesophase; Colh =hexagonal
columnar mesophase. [b] DH of the mesophase–isotropic phase transi-
tion. The first value corresponds to the enthalpy per mol of dendrimer
and the second value corresponds to the enthalpy per mol of mesogenic
unit. [c] Isotropic at RT. Glass transition is far below RT and not detected
under the experimental conditions of the DSC measurements.
Figure 4. a) 31P{1H} and b) 1H NMR spectra of G2-A1.
Figure 6. Microphotograph of G1-A3 annealed at 1008C for 12 h.
similar to but different from G1-A3. This fact indicates simi-
lar interactions between mesogenic units derived from A5,
regardless of the generation.[22] The difference with respect
to G1-A3 is due to the different mesogenic unit and symme-
try of the columnar organisation.
Structural study by X-ray diffraction: The three compounds
that showed birefringent textures were studied by X-ray dif-
fraction to verify the nature of the mesophase and measure
the structural parameters. The optically isotropic materials,
G2-A3 and G1-A4, were also tested to confirm their non-
mesomorphic, amorphous nature and rule out the presence
of an optically isotropic cubic mesophase (although no addi-
tional transitions to Tg were detected by DSC, as mentioned
above). The studies were performed in all cases at room
temperature after slow cooling of the sample from the iso-
tropic liquid. The results are gathered in Table 4.
Figure 5. MALDI-TOF mass spectrum of G1-A3 (matrix: dithranol+
NaTFA). The observed peak corresponds to [M+Na]+. Inset: Experi-
mental isotopic pattern (top) and theoretical isotopic pattern (bottom).
around 58C and a peak corresponding to the reversible
mesoACHTUNGTRENNUNGphase–isotropic phase transition were observed in the
DSC traces. The isotropisation temperatures were similar to
those observed by using POM. When the isotropisation en-
thalpies per mesogenic unit are compared, it can be ob-
served that G1-A5 and G2-A5 have values that are roughly
Compound G1-A3 gave X-ray diffractograms that con-
tained a set of four small- and middle-angle sharp maxima
16808
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16801 – 16814

New Liquid Crystalline Materials
FULL PAPER
Table 4. Structural data for the mesomorphic compounds measured by
using X-ray diffraction at room temperature.
[a]
[b]
Mesophase Parameters [ꢆ] dexptl
dcalcd
h, k[c] Z[d]
G1-A3 Colr
G1-A5 Colh
G2-A5 Colh
a=45.9
b=49.6
33.7
24.8
16.9
10.3
4.5 diff
47.4
27.1
23.8
18.3
4.5 diff
48.6
33.7
24.8
16.8
10.4
1, 1
0, 2
2, 2
4, 2
2
1
1
a=54.9
a=55.8
47.5
27.4
23.8
18.0
1, 0
1, 1
2, 0
2, 1
48.3
27.9
24.2
18.3
1, 0
1, 1
2, 0
2, 1
27.6
23.9
18.6
4.5 diff
[a] Experimentally measured spacing (diff=diffuse scattering). [b] Calcu-
lated spacing. [c] Miller indexation. [d] Number of molecules per 2D lat-
tice.
Figure 7. DSC scans (108Cminꢀ1) of the first cooling (c) and second
heating run (c) of G1-A3.
Figure 8. Microphotograph of G2-A5 taken at 758C on cooling from the
isotropic phase.
and a large-angle broad diffuse band (Figure 9). The sharp
maxima can be indexed to a two-dimensional rectangular
structure with lattice constants a=45.9, b=49.6 ꢆ. The dif-
fuse character of the large-angle intensity confirms the
liquid-crystal nature and thus G1-A3 exhibits a vitrified Colr
mesophase at room temperature.
Figure 9. X-ray diffractograms and indexation of the Colr phase of G1-A3
recorded at RT. a) Small-angle X-ray scattering pattern; b) wide-angle X-
ray scattering pattern.
stant deduced from the X-ray measurements is close to 55–
56 ꢆ and is practically independent of the generation, with a
slightly larger value for G2-A5.
To get insight into the packing of these dendritic mole-
cules in the columnar mesophases, some simple calculations
can be carried out based on density estimations. For a par-
ticular unit cell, the density 1 in gcmꢀ3 can be calculated di-
viding the mass of the cell by its volume:
Dendrimers G1-A5 and G2-A5 gave X-ray diffractograms
that were unambiguously characteristic of a two-dimensional
hexagonal structure in both cases, which is in accordance
with the optical study by using POM. This is revealed by the
presence of a set of four sharp maxima with a reciprocal
p
p p
spacing ratio 1: 3: 4: 7 in the small-angle region (see
Figure 10). As for G1-A3, the diffuse character of the large-
angle intensity confirms the liquid crystallinity of the den-
drimers and thus G1-A5 and G2-A5 exhibit a vitrified Colh
mesophase at room temperature. The hexagonal lattice con-
1 ¼ ðMZ ꢁ 1024Þ=ðVNAÞ
Chem. Eur. J. 2012, 18, 16801 – 16814
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16809

J. Jimꢀnez, L. Oriol et al.
that is similar to the first-generation compound with the
same mesogenic unit (G1-A5). Therefore, the molecules
need twice the previous height to make room for double the
number of mesogenic units and approximately double the
molecular mass (10499.28 for G1-A5 and 20905.70 for G2-
A5). To achieve this conformation, the central core must
extend mainly in the direction of the column axis to accom-
modate the increasing number of mesogenic units with no
significant modification of the column diameter.
The similar molecule height (6.6–6.7 ꢆ) calculated for the
two dendrimers of the same generation (G1-A3 and G1-A5)
is not surprising because they have the same core, which is
expected to adopt the same conformation. In this case, the
difference between these two dendrimers is the nature of
the mesogenic units. As a consequence of this, the dimen-
sions of the two-dimensional lattice are very different, apart
from the difference in symmetry. It is interesting to note
that the above-mentioned height calculations have been
made assuming Z=1 for G1-A5 and G2-A5 (as usual in
Colh mesophases) and Z=2 for G1-A3 (as usual in Colr
mesophases). The area S of the two-dimensional cell is in
the interval of about 2600–2770 ꢆ2 for the Colh mesophase
of G1-A5 and G2-A5, and 2278 ꢆ2 for the Colr mesophase
of G1-A3. For the Colh mesophase, S is the cross-sectional
area of the column, whereas for the Colr mesophase of G1-
A3 with Z=2 the cross-sectional area of the column is 2278/
2=1134 ꢆ2. This relatively low value is not unexpected con-
sidering the smaller size of the A3 mesogenic units com-
pared with A5, and the much smaller molecular mass of G1-
A3 (4490.26) compared with G1-A5 (10499.28).
Figure 10. X-ray pattern of G2-A5 recorded at RT.
in which M is the molar mass in g, Z is the number of mole-
cules per unit cell, V is the unit cell volume and NA is Avo-
gadroꢅs number. 1024 is the dividing factor to transform the
cell volume in ꢆ3 to cm3. The unit cell volume is deduced
from V=Sc, in which c is the height of the cell and S is the
area of the two-dimensional cell in ꢆ2 calculated as S=ab
p
for a rectangular lattice and S=a2 3/2 for a hexagonal lat-
tice.
Assuming that the density 1 of organic compounds must
be close to 1 gcmꢀ3, the above-mentioned equation allows
prediction of the height (c) of the unit cell. Because the co-
A
In the case of the optically isotropic materials, the X-ray
patterns of compounds G2-A3 and G1-A4 were recorded as
representative examples. The patterns of these compounds
contain only broad, diffuse bands both at small and large
angles, which is consistent with an amorphous, isotropic
phase (also corroborated by POM and DSC) and the diffuse
scattering is related to the short-range interactions between
the disordered molecules located in close proximity.
The X-ray results indicate that, for the three dendrimers
of this study that adopt columnar mesomorphism, the sym-
metry and the dimensions of the two-dimensional lattice de-
pends on the nature of the mesogenic units but not on the
generation; different mesogenic units (A3 and A5) produce
different symmetries but the same mesogenic units (G1-A5
and G2-A5) produce the same symmetry and dimensions in
the plane of the two-dimensional lattice. However, the
length of column needed to accommodate one molecule de-
pends on the generation and not on the nature of the meso-
genic units, that is, the height is the same for G1-A3 and
G1-A5, but it twice as big when the generation is increased
to G2-A5.
columns and, in fact, there is no periodicity along the
column axis, the real meaning of c is the height of the
column filled by each molecule. From the above-mentioned
equation it was deduced that the height can be estimated as
c=(MZꢇ1024)/
ACHTUNGTRENNUNG(SNA). Following this method, it was con-
cluded that a column length of about 6.6 ꢆ is needed to ac-
commodate one molecule of G1-A3, about 6.7 ꢆ for G1-A5
and about 13 ꢆ for G2-A5.
It can be considered that, when packed in the columnar
mesophase, the molecules adopt a more or less cylindrical
shape and the columns are generated by the stacking of
these relatively flat cylinders. These compounds contain a
central cyclophosphazene ring to which six dendrons are at-
tached. A number of mesogenic units are attached to the pe-
riphery of these dendrons, which surround the central core
(12 for G1 or 24 for G2). It is reasonable to think that 12
mesogenic units (in G1-A3 and G1-A5) are located at the
periphery of the molecule and the whole molecule is able to
adopt the shape of a cylinder with a height of 6.6 to 6.7 ꢆ.
The central part of the molecules, particularly the flexible
dendrons, adapts the conformation to this situation so as to
accommodate the 12 mesogenic units. In the case of G2-A5,
which contains 24 mesogenic units identical to those of G1-
A5, the mesogenic units need more space and the portion of
column deduced to accommodate one molecule is about
13 ꢆ high. In this case, the columns maintain a diameter
Conclusion
Hexa(4-hydroxyphenoxy)cyclotriphosphazene is a rigid cen-
tral core that can be easily modified to prepare dendritic
16810
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16801 – 16814

New Liquid Crystalline Materials
FULL PAPER
CuKa beam. The samples were held in Lindemann glass capillary tubes
(0.9 mm diameter) and heated when necessary by using a variable-tem-
perature oven. The patterns were collected on flat photographic films.
The capillary axis and the film were perpendicular to the X-ray beam.
Spacings were obtained by using Braggꢅs law.
The starting cyclotriphosphazene, [N3P3(OC6H4OH-4)6],[5a] benzylidene-
2,2-di(oxymethyl)propionic anhydride,[19] carboxylic acids, and DPTS (4-
(dimethylamino)pyridinium 4-toluenesulfonate)[25] were prepared by
using literature methods.
materials. The divergent growth of dendrimers by using this
rigid platform through an iterative process of esterification
with benzylidene-protected 2,2-di(hydroxymethyl)propionic
anhydride and subsequent hydrogenolysis gives rise to den-
drimers G1-(OH)12 and G2-(OH)24. These dendrimers have
been completely functionalised with different acids, as con-
firmed by different spectroscopic techniques. The mesomor-
phic properties strongly depend on the anisotropic geometry
of the aromatic acids. Acids that contain only one aromatic
ring do not promote mesomorphism. However, the incorpo-
ration of a monocatenar acid, such as A3, which is typically
used in the preparation of calamitic liquid crystals and
which has a long L/D aspect ratio, results in a liquid-crystal-
line dendrimer for the first generation (G1-A3), which ex-
hibits a rectangular columnar phase, as observed by using
XRD. Nevertheless, the analogous dendrimer of second-gen-
eration G2-A3 is a non-mesomorphic material. This might
well be related to the tendency of dendrimers based on
Synthesis of G1-(O2Bn)6: [N3P3(OC6H4OH-4)6] (355 mg, 0.45 mmol) and
4-(dimethylamino)pyridine (DMAP; 111 mg, 0.90 mmol) were dissolved
in pyridine or THF (2 mL) and then diluted with CH2Cl2 (6 mL). Benzyl-
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
31P{1H} NMR spectroscopy, excess benzylidene-2,2-bis(oxymethyl) pro-
pionic anhydride was quenched by stirring the reaction mixture overnight
with a 1:1 pyridine/H2O solution (2 mL). The organic phase was diluted
with CH2Cl2 (100 mL) and extracted with NaHSO4 (1m, 2ꢇ40 mL),
Na2CO3 (10%, 2ꢇ40 mL) and finally with brine (40 mL) to remove the
pyridine, DMAP and the benzylidene-protected acid that formed as a by-
product. The organic phase was dried by using Na2SO4 and evaporated
under vacuum to give G1-(O2Bn)6 as a white solid (yield: 822 mg,
cycloACHTUNGTRENNUNGphosphazenes to adopt a spherical conformation, which
is not compatible with a columnar organisation.[9] Moreover,
the presence of only one chain in each pro-mesogenic unit
produces inefficient filling of the peripheral space of the
molecules, which is necessary to promote columnar meso-
phases. In first-generation compound G1-A3 these problems
are overcome because the number of pro-mesogenic units is
low enough for all of them to lie with their main axes ap-
proximately in the plane perpendicular to the column axis.
However, for the second-generation analogue this kind of
arrangement is hindered by the higher number of pro-meso-
genic units.
90.7%). 1H NMR (CDCl3): d=7.45 (m, 2H; C6H5), 7.33 (m, 3H; C6H5),
3
7.01, 6.95 (AB spin system, J
G
CHPh), 4.71 (d, 2J(H,H)=11.4 Hz, 2H; OCH2), 3.66 (d, 2J
ACTHNUGTRENNUGN ACHTUNGTRENNUNG
11.4 Hz, 2H; OCH2), 1.11 ppm (s, 3H; CH3); 13C{1H} NMR (CDCl3): d=
172.41 (1C; C(O)O), 147.86, 147.75, 137.80, 129.01, 128.23, 126.22,
122.49, 121.80 (12C; aromatic carbons), 101.93 (1C; CHPh), 73.44 (2C;
OCH2), 42.76 (1C;
CACTHNGUTERNNUG
(CH3)), 17.64 ppm (1C; CH3); 31P{1H} NMR
(CDCl3): d=9.31 (s, 3P; N3P3 ring); IR (ATR): n˜ =1751 (s) (C=O), 1212
(m), 1162 (vs, br), 1142 (vs, br) (P=N), 949 cmꢀ1 (vs, br) (P O); MALDI-
ꢀ
TOF (dithranol): m/z (%): 2015.6 (100) [M+H]+; elemental analysis
calcd (%) for C108H102O30N3P3 (2014.89): C 64.38, H 5.10, N 2.09; found:
C 64.40, H 5.08, N 2.07.
Synthesis of G1-(OH)12: Compound G1-(O2Bn)6 (806 mg, 0.4 mmol) was
dissolved in 1:1 CH2Cl2/MeOH (4 mL) and Pd/C (5%, 400 mg) was
added to the solution. The apparatus for catalytic hydrogenolysis was
evacuated and filled three times with H2. After vigorous stirring over-
night, the catalyst was filtered off and carefully washed with methanol.
The filtrate was evaporated under vacuum and subsequent addition of
chloroform (20 mL) gave G1-(OH)12 as a white solid (yield: 574 mg,
In the case of polycatenar acids A4 and A5, which have a
similar structure, it was concluded that the presence of a
short flexible spacer plays a key role in the mesomorphic or-
ganisation of the resulting dendrimers. The positive effect of
the spacer on the liquid-crystal properties may be related to
its role in making the pro-mesogenic units independent of
the central core, which thus releases these units from any
conformational restrictions or other perturbing effects due
to the cyclophosphazene. This decoupling effect of the
spacer allows for independent movement and arrangement
of the mesogenic units and has been shown for both low-[23]
and high-molecular-weight[24] liquid crystals.
96.6%). 1H NMR (DMSO): d=7.03, 6.91 (AB spin system, 3J
9.0 Hz, 4H; OC6H4O), 4.93 (t, 3J
(H,H)=5.6 Hz, 2H; OH), 3.66, 3.55
(ABX spin system, 2J(H,H)=10.4 Hz, 3J
(H,H)=5.6 Hz, 4H; OCH2),
1.20 ppm (s, 3H; CH3); 1H NMR ((CD3)2CO): d=6.98, 6.91 (AB spin
system, 3J(H,H)=8.8 Hz, 4H; OC6H4O), 4.22 (t, 3J
(H,H)=5.6 Hz, 2H;
(H,H)=10.8 Hz, 3J
(H,H)=5.6 Hz,
ACHTUNGTRENNUNG(H,H)=
AHCTUNGTRENNUNG
A
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
OH), 3.88, 3.81 (ABX spin system, 2J
E
ACHTUNGTRENNUNG
4H; OCH2), 1.32 ppm (s, 3H; CH3); 13C{1H} NMR (DMSO): d=173.53
(1C; C(O)O), 147.82, 146.92, 123.11, 121.31 (6C; C6H4), 63.79 (2C;
OCH2), 50.73 (1C;
CACTHNGUTERNNUG
(CH3)), 16.79 ppm (1C; CH3); 31P{1H} NMR
(DMSO): d=9.18 ppm (s, 3P; N3P3 ring); 31P{1H} NMR ((CD3)2CO): d=
10.45 ppm (s, 3P; N3P3 ring); IR (ATR): n˜ =3365 (m, br), 3268 (m, br)
(OH), 1751 (s), 1734 (s) (C=O), 1204 (m), 1167 (s), 1147 (vs) (P=N), 958
Experimental Section
(vs), 945 cmꢀ1 (s) (P O); MALDI-TOF (dithranol): m/z (%): 1486.4
ꢀ
(100) [M]+, 1508.4 (6.5) [M+Na]+; elemental analysis calcd (%) for
C66H78O30N3P3 (1486.25): C 53.34, H 5.29, N 2.83; found: C 53.40, H 5.35,
N 2.80.
General data: Instrumentation and general experimental techniques (ele-
mental analysis, IR, thermogravimetry and optical microscopy) were
used or performed as described in the literature.[3,5] NMR spectra were
recorded by using a Brucker AV 400 spectrometer. Chemical shifts are
quoted relative to SiMe4 (TMS, 1H and 13C, external) and 85% H3PO4
(31P, external). MALDI-TOF mass spectrometry was carried out by using
a REFLEX III Bruker instrument with dithranol as the matrix. DSC was
performed by using a DSC Q-2000 instrument from TA Instruments with
Synthesis of G1-An ACHTUNGRTNEN(GU n=1–4): A mixture of G1-(OH)12 (45 mg,
0.03 mmol), the corresponding carboxylic acid (0.936 mmol; 261 mg for
G1-A1; 553 mg for G1-A2; 251 mg for G1-A3, 665 mg for G1-A4) and
DPTS (58 mg, 0.187 mmol) was added to dry CH2Cl2 (2 mL) and cooled
to 08C, then N,N’-dicyclohexylcarbodiimide (DCC; 3 mL, 215 mg,
1.03 mmol) in dry CH2Cl2 was added. The reaction mixture was vigorous
stirred under an argon atmosphere for 9 d. After reaction completion, as
monitored by using 31P{1H} NMR spectroscopy, the precipitated DCU
was filtered off and washed with CH2Cl2 (3ꢇ3 mL). For G1-An (n=1 or
2), the filtrate was evaporated to dryness and subsequent addition of
samples (2–5 mg) sealed in aluminum pans and
a scanning rate of
108Cminꢀ1 under a nitrogen atmosphere. In general the peaks obtained
were broad and the transition temperatures were, therefore, read at the
peak maxima. The X-ray diffraction patterns were obtained by using a
pinhole camera (Anton–Paar) operating with a point-focussed Ni-filtered
Chem. Eur. J. 2012, 18, 16801 – 16814
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16811

J. Jimꢀnez, L. Oriol et al.
hexane (30 mL) gave DPTS as a white solid that was filtered off. The re-
sulting filtrate was evaporated to dryness to give G1-A1 or G1-A2 as a
colourless oil that was purified by chromatography using silica gel
column and hexane/ethyl acetate (20:1) as the eluent. For G1-An (n=3
or 4), the CH2Cl2 filtrate was concentrated to a viscous liquid and then
poured into ethanol to give G1-A3 or G1-A4 as a white solid that was
washed with ethanol and dissolved in CH2Cl2. The solution was concen-
trated again to a viscous liquid and poured into ethanol twice. All com-
pounds were dried in vacuum at 408C for 48 h.
matic carbons), 73.55, 69.22, 65.64 (8C; OCH2), 47.07 (1C; CCH3), 31.91,
30.36, 29.70, 29.60, 29.41, 29.32, 26.08, 22.68 (48C; CH2), 18.02 (1C; C-
AHCTUNGTRENNUNG
(CH3)), 14.09 ppm (6C; CH3); 31P{1H} NMR (CDCl3): d=9.00 ppm (s,
3P; N3P3 ring). IR (ATR): n˜ =2955 (sh, w), 2921 (m), 2870 (sh, w), 2852
ꢀ1
ꢀ
(m) (C H), 1761 (sh, w), 1729 (m) (C=O), 1187 (s), (P=N), 953 cm (m)
ꢀ
(P O); MALDI-TOF (dithranol+NaTFA): m/z (%): 9819.0 (100)
[M+H2O]+, 9802 (5) [M]+; elemental analysis calcd (%) for
C594H894O102N3P3 (9802.33): C 72.78, H 9.19, N 0.43; found: C 71.99, H
8.56, N 0.49.
Data for G1-A1: Yield: 78 mg, 56.4%. 1H NMR (CDCl3): d=7.91 (d, 3J-
Synthesis of G1-A5: A mixture of G1-(OH)12 (29 mg, 0.02 mmol), the
ACHTUNGTRENNUNG ACHTUNGTRENNUNG
(H,H)=8.9 Hz, 4H; C6H4), 6.90, 6.85 (AB spin system, 3J
corresponding
acid
HOC(O)ACTHUNGTREN(UNG CH2)3OC6H4ACHTUNGTRENN{UGN 4-OC(O)(3,4,5-C6H2-
4H; OC6H4O), 6.83 (d, 3J
system, 2J
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
CH2Cl2 (4 mL) was cooled to 08C and N,N’-dicyclohexylcarbodiimide
(DCC; 143 mg, 0.69 mmol) in dry CH2Cl2 (2 mL) was added. The reac-
tion mixture was vigorous stirred under an argon atmosphere for 9 d.
After reaction completion, as monitored by using 31P{1H} NMR spectro-
scopy, the precipitated DCU was filtered off and washed with CH2Cl2
(3ꢇ3 mL). The filtrate was concentrated to a viscous liquid that was
poured into DMSO to give a highly viscous oil, which was washed with
DMSO and dissolved in CH2Cl2. The solution was again concentrated to
a viscous liquid and poured into ethanol twice. Compound G1-A5 was
obtained as a highly viscous oil that was dried in vacuum at 408C for
N
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
165.63 (3C; C(O)O), 163.18, 147.86, 147.38, 131.61, 122.42, 121.80,
121.55, 114.15 (18 C; C6H4), 68.19, 65.59 (4C; C6H4OCH2 +C(O)OCH2),
47.20 (1C; C
ACHTUNGTRENNUNG
CH2), 17.92 (1C; C
ACHTUNGTRENNUNG
d=8.94 ppm (s, 3P; N3P3 ring); IR (ATR): n˜ =2952 (sh, w), 2924 (m),
ꢀ
2870 (sh, w), 2854 (m) (C H), 1755 (m), 1716 (s) (C=O), 1163 (vs, br)
(P=N), 948 cmꢀ1 (s) (P,C->O); MALDI-TOF (dithranol): m/z (%):
4633.7 (100) [M+Na]+, 4256.4 (11) [MꢀC6H4OC10H21ꢀC10H21+Na]+ ele-
mental analysis calcd (%) for C270H366O54N3P3 (4610.78): C 70.33, H 8.00,
N 0.91; found: C 70.45, H 7.95, N 0.89.
1
48 h (yield: 146 mg, 69%). H NMR (CDCl3): d=7.38 (s, 4H, C6H2), 7.07
(d, 3J
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
Data for G1-A2: Yield: 180 mg, 71.8%. 1H NMR (CDCl3): d=7.19 (s,
A
ACHTUNGTRENNUNG
4H; C6H2), 6.97, 6.92 (AB spin system, 3J
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
4.72, 4.62 (AB spin system, 2J
3J
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
N
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
C6H2OCH2), 1.81–1.16 (m, 99H; alkyl chains+CCAHTUNGTRENNNUG
AHCTUNGTRENNUNG
18H; CH2CH3); 13C{1H} NMR (CDCl3): d=171.36, 165.61 (3 C, CO),
152.86, 147.97, 147.34, 142.70, 123.96, 122.42, 121.78, 107.95 (18C; C6H4),
171.24, 165.27 (5C; C(O)O), 156.39, 152.88, 148.01, 147.37, 144.52,
142.83, 123.95, 122.48, 122.39, 121.89, 115.00, 108.43 (30C; aromatic car-
73.47, 69.08 (8C; C6H4OCH2 +C(O)OCH2), 47.11 (1C; C
ACHTUNGTRENUN(GN CH3)), 32.00,
bons), 73.50, 69.18, 66.92, 64.96 (10C; OCH2), 46.69 (1C; C
30.32, 29.70, 29.60, 29.55, 29.37, 29.28, 26.07, 24.52, 22.64 (52C; CH2),
17.82 (1C; C
(CH3)), 14.07 ppm (6C; CH3); 31P{1H} NMR (CDCl3): d=
ACHTUNGTRENUN(GN CH3)), 31.87,
31.97, 31.91, 30.51, 30.37, 29.79, 29.75, 29.69, 29.66, 29.61, 29.47, 29.40,
29.37, 29.34, 26.15, 26.08, 22.73, 22.68 (48C; CH2), 18.12 (1C; CACHTUNRGTNEUNG(CH3)),
AHCTUNGTRENNUNG
14.10 ppm (6C; CH3); 31P{1H} NMR (CDCl3): d=8.73 ppm (s, 3P; N3P3
8.84 ppm (s, 3P; N3P3 ring); IR (ATR): n˜ =2955 (sh, w), 2921 (m), 2874
ꢀ
ring); IR (ATR): n˜ =2955 (sh, m), 2920 (vs), 2871 (sh, m), 2853 (s) (C
ꢀ
(sh, w), 2852 (m) (C H), 1751 (sh, w), 1732 (m) (C=O), 1184 (vs), 1160
H), 1753 (m), 1718 (s) (C=O), 1207 (s), 1169 (s) (P=N), 952 cmꢀ1 (s) (P
ꢀ
(sh, s) (P=N), 953 cmꢀ1 (m) (P O); MALDI-TOF (dithranol): m/z (%):
ꢀ
O); MALDI-TOF (dithranol): m/z (%): 8383.4 (100) [M+Na]+, 7695.8
10489.0 (100) [M]+ and peaks derived from sequential loss of 3,4,5-C6H2-
(7) [MꢀC6H
(OC10H21)3ꢀC10H21+Na]+; elemental analysis calcd (%) for
ACHNUTTGERG(NUNN OC10H21)3 and OC10H21; elemental analysis calcd (%) for C630H966
N3O114P3 (10499.28): C 72.07, H 9.27, N 0.40; found: C 72.64, H 9.89, N
0.47.
C
510H846O78N3P3 (8361.22): C 73.26, H 10.20, N 0.50; found: C 73.35, H
10.22, N 0.48.
Data for G1A3: Yield: 84 mg, 62%. 1H NMR (CDCl3): d=8.00 (d, 3J-
Synthesis of G2-(O2Bn)12: A mixture of G1-(OH)12 (37 mg, 0.025 mmol)
and 4-(dimethylamino)pyridine (DMAP; 12 mg, 0.096 mmol) were dis-
solved in pyridine (1 mL) and then diluted with CH2Cl2 (2 mL). Benzyl-
3
(H,H)=8.4 Hz, 4H; C6H4), 7.54 (d, J
E
ACHTUNGTRENNUNG ACHTUNGTRENNUNG
(H,H)=8.4 Hz, 4H; C6H4), 7.21 (d, 3J
AHCTUNGTRENNUNG
AHCTUNGERTGiNNUN dene-2,2-bis(oxymethyl) propionic anhydride (205 mg, 0.48 mmol) was
A
ACHTUNGTRENNUNG
added and the reaction mixture was vigorously stirred under an argon at-
mosphere for 4 d. After reaction completion, as monitored by using
AHCTUNGTRENNUNG
(m, 8H, CH2), 0.89 ppm (t, 3J
ACHTNUTRGNEUNG(H,H)=6.9 Hz, 6H; CH2CH3);
31P{1H} NMR spectroscopy, the excess benzyl
ACTHNUTRGNEiUNG dene-2,2-bis(oxymethyl)
13C{1H} NMR (CDCl3): d=171.34, 165.81 (3C, C(O)O), 147.87, 147.36,
propionic anhydride was quenched by stirring the reaction mixture over-
night with a 1:1 pyridine/H2O solution (2 mL). The organic phase was di-
luted with CH2Cl2 (50 mL) and extracted with NaHSO4 (1m, 2ꢇ20 mL),
Na2CO3 (10%, 2ꢇ20 mL) and brine (1ꢇ20 mL). The organic phase was
dried by using Na2SO4 and evaporated under vacuum to give G2-
145.85, 143.17, 137.01, 130.09, 128.95, 127.84, 127.04, 126.83, 122.42,
121.89 (30C; C6H4), 65.91 (2C; C(O)OCH2), 47.23 (1C; CACHTUNTRGENUG(N CH3)), 35.58,
31.52, 31.09, 22.53 (8C; CH2), 18.00 (1C; CACTHNURGTNEG(UN CH3)), 14.02 ppm (2C; CH3);
31P{1H} NMR (CDCl3): d=9.43 ppm (s, 3P; N3P3 ring); IR (ATR): n˜ =
ꢀ
2954 (m), 2925 (m), 2871 (m), 2854 (m) (C H), 1750 (m), 1717 (vs) (C=
1
(O2Bn)12 as a white solid (yield: 88 mg, 89%). H NMR (CDCl3): d=7.38
O), 1205 (w), 1165 (vs), 1151 (sh, w) (P=N), 953 cmꢀ1 (s) (P O);
ꢀ
(m, 4H; C6H5), 7.24 (m, 6H; C6H5), 6.89, 6.77 (AB spin system, 3J-
MALDI-TOF (dithranol): m/z (%):4513.0 (100) [M+Na]+; elemental
analysis calcd (%) for C282H294O42N3P3 (4490.26): C 75.43, H 6.60, N 0.93;
found: C 75.43, H 6.55, N 0.97.
ACHUTNGRENNUG ACHTUNGTRENNUNG
(H,H)=9.2 Hz, 4H; OC6H4O), 5.40 (s, 2H; CHPh), 4.57 (d, 2J
10.8 Hz, 4H; OCH2), 4.50, 4.47 (AB spin system, 2J
OCH2), 3.58 (d, 2J
ACTHNUGTNREUN(G H,H)=10.8 Hz, 4H; OCH2), 1.38 (s, 3H; CH3),
ACHTUNGTRENNUNG
Data for G1-A4: Yield: 194 mg, 66%. 1H NMR (CDCl3): d=8.06 (d, 3J-
0.91 ppm (s, 6H; CH3); 13C{1H} NMR (CDCl3): d=173.23, 171.18 (3C;
C(O)O), 147.92, 147.15, 137.69, 128.83, 128.07, 126.09, 122.47, 121.66
(18C; aromatic carbons), 101.66 (2C; CHPh), 73.35 (4C; OCH2), 65.35
ACHTUNGTRENNUNG
(H,H)=8.4 Hz, 4H; OC(O)C6H4O), 7.36 (s, 4H, C6H2), 7.24 (d, 3J-
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
(2C; OCH2), 47.02 (1C; CACTHNURGTNNEUG(CH3)), 42.59 (2C; CCAHTUGNTRNEN(UGN CH3)), 17.69 ppm (3C;
CH3); 31P{1H} NMR (CDCl3): d=8.39 ppm (s, 3P; N3P3 ring); IR (ATR):
AHCTUNGTRENNUNG
n˜ =1736 (s) (C=O), 1216 (m), 1155 (vs, br) (P=N), 950 cmꢀ1 (vs, br) (P
ꢀ
ACHTUNGTRENNUNG
O); MALDI-TOF (dithranol+NaTFA): m/z (%): 3959.5 (100) [M+Na]+;
elemental analysis calcd (%) for C210H222O66N3P3 (3936.91): C 64.07, H
5.68, N 1.07; found: C 64.13, H 5.68, N 1.25.
ACHTUNGTRENNUNG
d=171.25, 165.13, 164.32 (5C; C(O)O), 155.02, 152.95, 148.01, 147.29,
143.19, 131.22, 126.97, 123.31, 122.41, 122.05, 121.95, 108.55 (30C; aro-
16812
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16801 – 16814

New Liquid Crystalline Materials
FULL PAPER
Synthesis of G2-(OH)24: Compound G2-(O2Bn)12 (394 mg, 0.1 mmol) was
dissolved in 1:1 CH2Cl2/MeOH (4 mL) and Pd/C (5%, 200 mg) was
added to the solution. The apparatus for catalytic hydrogenolysis was
evacuated and filled three times with H2. After vigorous stirring over-
night, the catalyst was filtered off and carefully washed with methanol.
The filtrate was evaporated and subsequent addition of CHCl3 (20 mL)
(CH3)), 14.08 ppm (12C; CH3); 31P{1H} NMR
18.16, 17.94 (3C; CACHTUNTGNRUEGN
(CDCl3): d=8.02 ppm (s, 3P; N3P3 ring); IR (ATR): n˜ =2952 (sh, m),
ꢀ
2924 (s), 2871 (sh, m), 2852 (s) (C H), 1739 (m), 1719 (sh, s) (C=O),
1204 (s), 1171 (m), 1152 (s) (P=N), 953 cmꢀ1 (m) (P O); MALDI-TOF
ꢀ
(ditranol): m/z (%): 16628.5 (100) [M]+; elemental analysis calcd (%) for
C1014H1710O162N3P3 (16629.58): C 73.24, H 10.36, N 0.25; found C 73.22, H
10.40, N 0.22.
gave G2-(OH)24 as
a
white solid (yield: 275 mg, 95.5%). 1H NMR
(H,H)=8.4 Hz, 4H;
(H,H)=
(H,H)=10.8 Hz, 8H;
(MeOD): d=7.06, 6.91 (AB spin system, 3J
AHCTUNGTRENNUNG
Data for G2-A3: Yield: 97 mg, 73%. 1H NMR (CDCl3): d=7.96 (d, 3J-
OC6H4O), 4.87 (brs, 4H; OH), 4,43, 4,37 (AB spin system, 2J
ACHTUNGTRENNUNG
3
G
N
10.8 Hz, 4H; OCH2), 3.66, 3.58 (AB spin system, 2J
N
3J(H,H)=7.6 Hz, 8H; C6H4), 7.17 (d, 3J
ACTHNUGTRENNUGN ACHTUNGTRENNNUG
1
OCH2), 1.44 (s, 3H; CH3), 1.12 ppm (s, 6H; CH3); H NMR ((CD3)2CO):
d=7.18, 7.04 (AB spin system, 3J
4.41 (AB spin system, 2J(H,H)=11.2 Hz, 4H; OCH2), 3.99 (t, 3J
5.8 Hz, 4H; OH), 3.72, 3.67 (ABX spin system, 2J(H,H)=10.8 Hz, 3J-
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
G
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
(m, 33H, alkyl chains+CACTHNUTRGNEU(GN CH3)), 0.89 ppm (br^t, 2H; CH2CH3);
13C{1H} NMR (CDCl3): d=172.00, 170.69, 165.69 (7C; C(O)O), 148.05,
147.09, 145.68, 143.04, 137.00, 130.08, 128.87, 127.85, 127.00, 126.74,
122.28, 121.84 (54C; aromatic carbons), 65.77, 65.31 (6C ; OCH2), 46.84
ACHTUNGTRENNUNG(H,H)=5.8 Hz, 8H; OCH2), 1.49 (s, 3H; CH3), 1.16 ppm (s, 6H; CH3);
13C{1H} NMR (MeOD): d=176.00 (2C; C(O)O), 173.23 (1C; C(O)O),
149.29, 149.09, 124.07, 123.05 (6C; C6H4), 66.37 (2C; OCH2), 65.92 (4C;
OCH2), 51.90 (2C; C
N
N
(3C; C
AHCTUNGTRENNUNG
ACHTUGNTRENUN(NG CH3)), 35.54, 31.51, 31.05, 22.51 (16C; CH2), 17.94, 17.71 (3C; C-
(CH3)), 14.01 ppm (4C; CH3); 31P{1H} NMR (CDCl3): d=8.49 (s, 3P;
N3P3 ring); IR (ATR): n˜ =2954 (sh, w), 2926 (m), 2869 (sh, w), 2855 (w)
ꢀ
(C H), 1750 (sh, m), 1717 (vs) (C=O), 1205 (sh, m), 1175 (s), 1157 (m)
(P=N), 961 cmꢀ1 (s) (P O); MALDI-TOF (dithranol+NaTFA): m/z (%):
br) (P=N), 954 cmꢀ1 (vs, br) (P O); MALDI-TOF (dithranol): m/z (%):
ꢀ
8905.8 (100) [M+H2O]+ and peaks derived from sequential loss of
2879.5 (100) [M]+, 2901.4 (68) [M+Na]+; elemental analysis calcd (%)
for C126H174O66N3P3 (2879.63): C 52.55, H 6.09, N 1.46; found: C 52.85, H
6.19, N 1.50.
C6H4C6H4ACHTUNGTRNNE(UG C5H11) and C6H4ACHTUNGTRENN(UGN C5H11); elemental analysis calcd (%) for
C558H606O90N3P3 (8887.67): C 75.41, H 6.87, N 0.47; found: C 74.53, H
6.66, N 0.52.
Synthesis of G2-An ACHTUNGTRENNUNG(n=1–4): A mixture of G2-(OH)24 (43 mg,
Data for G2-A4: Yield: 121 mg, 41%. 1H NMR (CDCl3): d=7.99 (br^d,
8H; OC(O)C6H4O), 7.32 (s, 8H, C6H2), 7.19 (br^d, 8H; OC(O)C6H4O),
7.04 (br^m, 4H; OC6H4O), 4.69–4.39 (m, 12H; C(O)OCH2), 4.10–3.91
0.015 mmol), the corresponding carboxylic acid (0.936 mmol; 261 mg for
G2-A1; 553 mg for G2-A2; 251 mg for G2-A3; 665 mg for G2-A4;
720 mg for G2-A5) and DPTS (58 mg, 0.187 mmol) in dry CH2Cl2 (2 mL)
was cooled to 08C and N,N’-dicyclohexylcarbodiimide (DCC; 215 mg,
1.03 mmol) in dry CH2Cl2 (3 mL) was added. The reaction mixture was
stirred vigorously under an argon atmosphere for 9 d. After completion,
as monitored by using 31P{1H} and 1H NMR spectroscopy, the precipitat-
ed DCU was filtered off and washed with CH2Cl2 (3ꢇ5 mL). For G2-An
(n=1 or 2), the filtrate was evaporated to dryness to obtain a white
solid, which was washed with ethanol to give a colourless oil of G2-A1 or
G2-A2. For G2-An (n=3–5), the CH2Cl2 filtrate was concentrated to a
viscous liquid that was poured into DMSO to give a white solid of G2-
A3 or a colourless highly viscous oil of G2-A4 or G2-A5, which was
washed with DMSO and dissolved in CH2Cl2. The solution was concen-
trated again to a viscous liquid and poured into ethanol twice. All com-
pounds were dried in vacuum at 408C for 48 h.
(m, 24H, C6H2OCH2), 1.87–1.11 (m, 201H, alkyl chains+CACHTUNGTRENNUNG(CH3)),
0.86 ppm (br^t, 36H; CH2CH3); 13C{1H} NMR (CDCl3): d=171.90,
170.74, 165.06, 164.22 (11C; C(O)O), 154.93, 152.92, 148.14, 147.19,
143.11, 131.19, 126.98, 123.37, 122.33, 122.19, 121.97, 108.50 (54C; aro-
matic carbons), 73.51, 69.17, 65.65 (18C; OCH2), 46.76 (3C; C
31.89, 30.38, 29.68, 29.59, 29.43, 29.34, 26.12, 22.66 (96C; CH2), 18.03,
17.74 (3C; C
(CH3)), 14.09 ppm (12C; CH3); 31P{1H} NMR (CDCl3): d=
ACHTUNGTRENNUNG(CH3)),
AHCTUNGTRENNUNG
8.43 ppm (s, 3P; N3P3 ring); IR (ATR): n˜ =2955 (sh, w), 2921 (m), 2878
ꢀ
(sh, w), 2852 (m) (C H), 1733 (m), 1720 (sh, w) (C=O), 1187 (vs), 1159
(vs) (P=N), 953 cmꢀ1 (m) (P O); MALDI-TOF (dithranol+NaTFA): m/
ꢀ
z (%): 19524 (100) [M]+ and peaks derived from sequential loss of
OC(O)C6H
(OC10H21)3;
elemental
analysis
calcd
(%)
for
C1182H1806O210N3P3 (19511.80): C 72.76, H 9.33, N 0.21; found: C 70.87, H
8.93, N 0.22.
Data for G2-A1: Yield: 112 mg, 82%. 1H NMR (CDCl3): d=7.86 (d, 3J-
ACHTUNGTRENNUNG
1
Data for G2-A5: Yield: 141 mg, 45%. H NMR (CDCl3): d=7.35 (s, 8H,
C6H2), 7.09 (m, 4H; OC6H4O), 7.04 (d, 3J
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
OC(O)C6H4O), 6.86 (d, 3J
(AB spin system, 3J
spin system, 3J
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
A
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
170.72, 165.55 (7C; C(O)O), 163.12, 148.09, 147.12, 131.61, 122.31,
121.79, 121.61, 114.11 (30C; C6H4), 68.17, 65.44, 65.23 (10C;
C6H4OCH2 +COOCH2), 46.85 (3C; CACHTUNTRGENNUG(CH3)), 31.88, 29.56, 29.40, 29.31,
AHCTUNGTRENNUNG
(CDCl3): d=172.52, 172.01, 165.24 (11C; C(O)O), 156.43, 152.90, 148.27,
147.22, 144.51, 142.86, 123.98, 122.49, 121.89, 115.0, 108.43 (54C; aromat-
ic carbons), 73.52, 69.19, 66.95, 64.93 (22C; OCH2), 46.91, 46.43 (3C; C-
29.10, 25.97, 22.61 (32C; CH2), 17.93, 17.65 (3C; CACHTUNTRGNE(NUG CH3)), 14.10 ppm
(4C; CH3); 31P{1H} NMR (CDCl3): d=8.34 ppm (s, 3P; N3P3 ring); IR
AHCTUNGTRENNUNG
ꢀ
(ATR): n˜ =2953 (sh, w), 2922 (m), 2869 (sh, w), 2853 (m) (C H), 1736
AHCTUNGTRENNUNG
(m), 1715 (s) (C=O), 1164 (vs) (P=N), 953 cmꢀ1 (m) (P O); MALDI-
ꢀ
(CDCl3): d=8.27 ppm (s, 3P; N3P3 ring); IR (ATR): n˜ =2955 (sh, w),
TOF (dithranol): m/z (%): 9153.5 (100) [M+Na]+; elemental analysis
calcd (%) for C534H750O114N3P3 (9128.54): C 70.26, H 8.28, N 0.46; found
C 70.30, H, 8.30, N 0.43.
ꢀ
2921 (m), 2872 (sh, w), 2852 (m) (C H), 1750 (sh, w), 1732 (m) (C=O),
1205 (m), 1184 (vs), 1157 (sh, s) (P=N), 953 cmꢀ1 (m) (P O); MALDI-
ꢀ
TOF (dithranol): m/z (%): 20822.8 (86) [M]+ and peaks derived from se-
1
Data for G2-A2: Yield: 180 mg, 72%. H NMR (CDCl3): d=7.14 (s, 8H;
quential loss of C(O)C6H2ACHTUNGTRNEN(UG OC10H21)3; elemental analysis calcd (%) for
C1254H1950O234N3P3 (20905.70): C 72.04, H 9.40, N 0.20; found: C 72.47, H
9.61, N 0.24.
C6H2), 7.06, 6.99 (AB spin system, 3J
N
2
(d, J
N
3.88 (m, 24H; C6H2OCH2), 1.80–1.16 (m, 201H; alkyl chains+CACTHNUTRGNENUG(CH3)),
0.86 ppm (m, 36H; CH2CH3); 13C{1H} NMR (CDCl3): d=171.95, 170.71,
165.49 (7C, C(O)O), 152.81, 148.33, 147.13, 142.68, 124.04, 122.31, 121.65,
107.98 (30C; aromatic carbons), 73.41, 69.07, 65.27, 64.79 (18C;
C6H2OCH2 +C(O)OCH2), 46.75 (3C; C
ACHTUNGTNER(NUGN CH3)), 31.94, 31.92, 30.40, 29.77,
29.71, 29.68, 29.64, 29.52, 29.40, 29.38, 26.20, 26.11, 22.67 (96C; CH2),
Chem. Eur. J. 2012, 18, 16801 – 16814
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16813

J. Jimꢀnez, L. Oriol et al.
therein; c) A. Hameau, S. Fuchs, R. Laurent, J. P. Majoral, A. M.
Caminade, Beilstein J. Org. Chem. 2011, 7, 1577–1583; d) K. Milow-
ska, T. Gabryelak, M. Bryszewska, A. M. Caminade, J. P. Majoral,
Int. J. Biol. Macromol. 2012, 50, 1138–1143 and references therein.
[14] a) C. Marmillon, F. Gauffre, T. Gulik-Krzywicki, C. Loup, A. M. Ca-
minade, J. P. Majoral, J. P. Vors, E. Rump, Angew. Chem. 2001, 113,
2696–2699; Angew. Chem. Int. Ed. 2001, 40, 2626–2629; b) A. M.
Caminade, J. P. Majoral, Acc. Chem. Res. 2004, 37, 341–348 and ref-
erences therein; c) K. Ciepluch, N. Katir, A. El Kadib, M. Weber,
A. M. Caminade, M. Bousmina, J. P. Majoral, M. Bryszewska, J.
Lumin. 2012, 132, 1553–1563.
[15] a) V. Percec, M. Kawasumi, Macromolecules 1992, 25, 3843–3850;
b) V. Percec, J. Heck, Am. Chem. Soc. Div. Polym. Chem. 1989, 30,
450–451; c) V. Percec, J. Heck, Polym. Bull. 1990, 24, 255–262;
d) V. Percec, J. Heck, J. Polym. Sci. Part A 1991, 29, 591–597; e) V.
Percec, M. Lee, J. Heck, H. E. Blackwell, G. Ungar, A. Alvarez-Cas-
tillo, J. Mater. Chem. 1992, 2, 931–938; f) V. Percec, J. Heck, M.
Lee, G. Ungar, A. Alvarez-Castillo, J. Mater. Chem. 1992, 2, 1033–
1039; g) V. Percec, J. Heck, G. Ungar, Macromolecules 1991, 24,
4957–4962; h) V. Percec, G. Johansson, J. Heck, G. Ungar, S. V.
Battyb, J. Chem. Soc. Perkin Trans. 1 1993, 13, 1411–1420; see also
several reviews covering the early developments of the field: i) V.
Percec, J. Heck, G. Johansson, D. Tomazos, M. Kawasumi, G.
Acknowledgements
This work was supported by the MICINN, Spain, under the projects
MAT2011-27978-C02-01, CTQ2009-09030 and CTQ2011-22589, FEDER
funding, Fondo Social Europeo, and Gobierno de Aragꢁn (Dpto. de
Ciencia, Tecnologꢂa y Universidad).
[1] a) J. E. Mark, H. R. Allcock, R. West, Inorganic Polymers, Prentice
Hall, Englewood Clifts, 1992, pp. 61–141; b) for a number of refer-
ences, see also: C. H. Honeyman, I. Manners, C. T. Morrissey, H. R.
Allcock, J. Am. Chem. Soc. 1995, 117, 7035–7036; c) H. R. Allcock,
Chemistry and Applications of Polyphosphazenes, Wiley, Hoboken,
N.J. , 2003.
[2] M. Gleria, R. De Jaeger, Phosphazenes: A Worldwide Insight, Nova
Science Publishers, New York, 2004.
[3] a) D. Abizanda, O. Crespo, M. C. Gimeno, J. Jimꢀnez, A. Laguna,
Chem. Eur. J. 2003, 9, 3310–3319; b) C. Dꢂaz, D. Abizanda, J. Jimꢀ-
nez, A. Laguna, M. L. Valenzuela, J. Inorg. Organomet. Polym.
Mater. 2006, 16, 211–218.
[4] a) J. Jimꢀnez, A. Laguna, M. Benouazzane, J. A. Sanz, C. Dꢂaz,
M. L. Valenzuela, P. G. Jones, Chem. Eur. J. 2009, 15, 13509–13520;
b) C. Dꢂaz, M. L. Valenzuela, A. Laguna, V. Lavayen, J. Jimꢀnez,
L. A. Power, C. OꢅDwyer, Langmuir 2010, 26, 10223–10233.
[5] a) J. Barberꢃ, M. Bardajꢂ, J. Jimꢀnez, A. Laguna, M. P. Martinez, L.
Oriol, J.L. Serrano, I. Zaragozano, J. Am. Chem. Soc. 2005, 127,
8994–9002; b) J. Barberꢃ, J. Jimꢀnez, A. Laguna, L. Oriol, S. Pꢀrez,
J. L. Serrano, Chem. Mater. 2006, 18, 5437–5445.
Ungar, J. Macromol. Sci., Part
A 1994, A31, 1031–1070; j) V.
Percec, J. Heck, G. Johansson, D. Tomazos, G. Ungar, Macromol.
Symp. 1994, 77, 237–265; k) V. Percec, J. Heck, G. Johansson, D.
Tomazos, M. Kawasumi, G. Ungar, J. Macromol. Sci., Part A 1994,
31, 1719–1758.
[16] a) M. Marcos, R. Martꢂn-Rapffln, A. Omenat, J. L. Serrano, Chem.
Soc. Rev. 2007, 36, 1889–1901; b) B. Donnio, S. Buathong, I. Bury,
D. Guillon, Chem. Soc. Rev. 2007, 36, 1495–1513.
[17] B. M. Rosen, C. J. Wilson, D. A. Wilson, M. Peterca, M. R. Imam, V.
Percec, Chem. Rev. 2009, 109, 6275–6540.
[18] H. Ihre, A. Hult, E. Sçderling, J. Am. Chem. Soc. 1996, 118, 6388–
6395.
[19] a) H. Ihre, O. L. Padilla De Jesffls, J. M. J. Frꢀchet, J. Am. Chem. Soc.
2001, 123, 5908–5917; b) S. L. Goh, M. B. Francis, J. M. J. Frꢀchet,
Chem. Commun. 2002, 2954–2955.
[20] H. R. Allcock, J. Am. Chem. Soc. 1964, 86, 2591–2595.
[21] H. R. Allcock, M. S. Connolly, J. T. Sisko, S. Al-Shali, Macromole-
cules 1988, 21, 323–334.
[22] a) R. Martꢂn-Rapffln, M. Marcos, A. Omenat, J. L. Serrano, G. R.
Luckhurst, A. Mainal, Chem. Mater. 2004, 16, 44969–44971; b) J. del
Barrio, L. Oriol, R. Alcalꢃ, C. Sꢃnchez, Macromolecules 2009, 42,
5752–5760.
[23] M. Ince, M. V. Martꢂnez-Dꢂaz, J. Barberꢃ, T. Torres, J. Mater. Chem.
2011, 21, 1531–1535.
[24] R. C. Tandel, J. Gohil, N. K. Patel, Res. J. Recent Sci. 2012, 1, 122–
127.
[6] J. Jimꢀnez, A. Laguna, A. M. Molter, J. L. Serrano, J. Barberꢃ, L.
Oriol, Chem. Eur. J. 2011, 17, 1029–1039.
[7] a) C. Kim, H. R. Allcock, Macromolecules 1987, 20, 1726–1727;
b) R. E. Singler, R. A. Willingham, R. W. Lenz, A. Furukawa, H.
Finkelmann, Macromolecules 1987, 20, 1727–1728; c) V. Percec, D.
Tomazos, R. A. Willingham, Polym. Bull. 1989, 22, 199–206; d) A. J.
Jaglowski, R. E. Singler, E. Atkins, Macromolecules 1995, 28, 1668–
1672, and references therein.
[8] a) K. Moriya, T. Suzuki, Y. Kawanishi, T. Masuda, H. Mizusaki, S.
Nakagawa, H. Ikematsu, K. Mizuno, S. Yano, M. Kajiwara, Appl.
Organomet. Chem. 1998, 12, 771–779; b) K. Moriya, Y. Kawanishi,
S. Yano, M. Kajiwara, Chem. Commun. 2000, 1111–1112.
[9] A. M. Caminade, J. P. Majoral in Applicative Aspect of Cyclophos-
phazenes (Eds.: M. Gleria, R. De Jaeger), Nova Science Publishers,
New York, 2004, pp. 45–69, and references therein.
[10] K. Inoue, T. Itaya, Bull. Chem. Soc. Jpn. 2001, 74, 1381–1395.
[11] a) F. Sournies, F. Crasnier, M. Graffeuil, J. P. Faucher, R. Lahana,
M. C. Labarre, J. F. Labarre, Angew. Chem. 1995, 107, 610–613;
Angew. Chem. Int. Ed. Engl. 1995, 34, 578–581; b) C. Larrꢀ, B. Don-
nadieu, A. M. Caminade, J. P. Majoral, Chem. Eur. J. 1998, 4, 2031–
2036; c) J. P. Majoral, C. Larrꢀ, R. Laurent, A. M. Caminade, Coord.
Chem. Rev. 1999, 190–192, 3–18.
[25] J. S. Moore, S. I. Stupp, Macromolecules 1990, 23, 65–70.
[12] a) R. Schneider, C. Kçllner, I. Weber, A. Togni, Chem. Commun.
1999, 2415–2416.
[13] a) C. Loup, M. A. Zanta, A. M. Caminade, J. P. Majoral, B. Meunier,
Chem. Eur. J. 1999, 5, 3644–3650; b) A. M. Caminade, C. O. Turrin,
J. P. Majoral, New J. Chem. 2010, 34, 1512–1524 and references
Received: July 31, 2012
Revised: September 4, 2012
Published online: November 7, 2012
16814
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16801 – 16814