ZnO/liquid crystalline nanohybrids: From properties in solution to anisotropic growth

Article abstract of DOI:10.1002/chem.201200233

Branched thermotropic liquid crystals (LCs) were successfully used as stabilizers for the synthesis of isotropic photoluminescent ZnO nanoparticles (NPs) in solution. ¹H NMR spectroscopy and differential scanning calorimetry experiments demonstrate the existence of specific interactions between the LC and both the ZnO precursor and ZnO NPs. This offers the possibility for some branched molecules to act as structurally ordered hosts for the anisotropic growth of ZnO NPs. Copyright

Full text of DOI:10.1002/chem.201200233

DOI: 10.1002/chem.201200233
ZnO/Liquid Crystalline Nanohybrids: From Properties in Solution to
Anisotropic Growth
Sarmenio Saliba,[a, b] Yannick Coppel,[b] Christophe Mingotaud,[a]
Jean-Daniel Marty,*[a] and Myrtil L. Kahn*[b]
Abstract: Branched thermotropic liquid crystals (LCs) were successfully used as
stabilizers for the synthesis of isotropic photoluminescent ZnO nanoparticles
(NPs) in solution. H NMR spectroscopy and differential scanning calorimetry ex-
Keywords: anisotropic growth · liq-
uid crystals · morphology control ·
1
periments demonstrate the existence of specific interactions between the LC and
nanoparticles · photoluminescence ·
both the ZnO precursor and ZnO NPs. This offers the possibility for some
zinc oxide
branched molecules to act as structurally ordered hosts for the anisotropic growth
of ZnO NPs.
Over the past decade, composites based on the dispersion
of nanoparticles (NPs) in liquid crystals have been of grow-
ing interest.[1] The driving force of research in this field is
the possible development of new materials or devices with
multiple or improved properties. Indeed, NPs may influence
the state[2] or the surface anchoring of doped LCs.[3] The
electro-optical behavior of LC/NP systems may then be ad-
justed by changing the NP content.[4,5] Conversely, the prop-
erties of the nanomaterial may be modified and even con-
trolled by the LC matrix.[6,7]
Achieving such a goal often requires functionalization of
NPs with LC moieties to improve compatibility between
NPs and LC or to obtain new LC nanohybrids. Indeed, LC
hybrids involving gold,[8–10] cobalt,[11] cadmium selenide or
cadmium telluride,[12] and iron oxide NPs[13,14] have been de-
scribed. In a previous article we reported the first successful
hybridization of small-sized ZnO NPs with a mesogenic
molecule.[15] Due to the presence of fluxional ligands, the hy-
bridized material presented mesomorphic behavior even
with a high content of inorganic material. However, the
fluxionality of these ligands at the surface of the NP may
hinder their application due to the presence of excess free li-
gands present in solution. In addition, in most cases these
hybrids were obtained in a two-step process. Recently,
branched capping agents have attracted special interest in
the field of nanohybrids due to their effective stabilization
properties for ZnO nanoparticles in polar and non-polar sol-
vents.[16–18] These stabilizers are characterized by their three-
dimensional molecular shapes; they also have low mobility
at the NP surface and present a large number of functional
end groups, which makes them easy to tailor for end-user
purposes.
Here, we report the elaboration of ZnO NP hybrids, syn-
thesized by using an organometallic method and using mac-
romolecules (i.e., Tren-LCs) as the sole stabilizing agents
(see Scheme 1). This approach involves the hydrolysis of di-
cyclohexylzinc, [ZnACHTNUGTRNEUNG(Cy2)] in air to give ZnO NPs with vola-
tile cyclohexane as the unique side product.[19] This process
may avoid disruption of the liquid crystalline mesophase
during NP formation, and the use of branched LC molecules
with amino groups may favor the interactions with both the
NP precursor and the resultant NP, which thus allows for
a one-step synthesis of ZnO NPs.[18,20] Moreover, the use of
highly viscous LCs should prevent fast disruption of the LC
structural organization during the NP formation. Depending
on the functionality, two liquid crystals designated as Tren-
OMeLC and Tren-CNLC (see below) were chosen. To
assess the key interactions involved in the formation of ZnO
NPs, we investigated three different systems. First, the prep-
aration of ZnO with simple LC acrylates (OMeLC and
CNLC) in THF was studied. The second system involved
the synthesis of ZnO NPs in THF containing a Tren-LC.
The scope of the latter was the formation of LC colloidal
solutions of ZnO NPs and the investigation (in solution) of
the interactions between 1) the LC and the zinc precursor
and 2) the LC and the synthesized ZnO NPs. Lastly, we
have used this information to control the growth of ZnO
NPs by using the bulk state of thermotropic LC.
[a] Dr. S. Saliba, Dr. C. Mingotaud, Dr. J.-D. Marty
Laboratoire IMRCP, CNRS UMR 5623
University of Toulouse
118, route de Narbonne, 31062 Toulouse (France)
Fax : (+33)561-55-30-03
E-mail: marty@chimie.ups-tlse.fr
[b] Dr. S. Saliba, Dr. Y. Coppel, Dr. M. L. Kahn
Laboratoire de Chimie de Coordination, CNRS UPR 8241
University of Toulouse
205, route de Narbonne, 31062 Toulouse (France)
E-mail: kahn@lcc-toulouse.fr
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201200233.
8084
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 8084 – 8091

FULL PAPER
Table 1. DSC data for LCs and LC/ZnO hybrids.
[b]
Sample
Tg[a] [8C] TLCÀI [8C] DH [JgÀ1
]
Mesophase
type[c]
OMeLC
Tren-OMeLC
–
À22
À5.3
40
10
29
1.0
0.9
N[d]
N
N
Tren-OMeLC/ZnO
(synthesized in solvent)
Tren-OMeLC/ZnO[e]
(synthesized in bulk)
CNLC
<0.2[g]
À1
37
0.4
N
–
8
35
58
29
0.5
0.8
N[d]
N
N
Tren-CNLC
Tren-CNLC/ZnO
(synthesized in solvent)
Tren-CNLC/ZnO[f]
(synthesized in bulk)
À4.7
<0.2[g]
4
49
0.6
N
[a] The glass transition temperatures, Tg, were recorded in heating cycles
at 58CminÀ1. [b] The transition temperatures, TLCÀI, were taken to be the
average of the onset and endset values of peaks recorded during cooling
cycles at À58CminÀ1. [c] N: nematic phase. [d] Nematic monotropic.
[e] Containing ZnO nanowires synthesized in the nematic bulk state of
Tren-OMeLC with
a molar ratio of 0.4:1.0 Tren-OMeLC/[Zn(Cy)2].
[f] Containing isotropic ZnO nanoparticles synthesized in the nematic
bulk state of Tren-CNLC with a molar of 0.45:1.0 Tren-CNLC/[Zn(Cy)2].
[g] Approximate value due to low peak intensity.
scattering pattern that only displays a diffuse ring at high
angles (d=4.5 ꢁ) (Figure 1d and e) confirm the nematic (N)
organization of the LC below the clearing temperature.
Tren-CNLC presents a nematic phase below 588C and a Tg
at 8.18C (see S1 in the Supporting Information).
Scheme 1. General synthesis for Tren-based liquid crystals: Tren-OMeLC
and Tren-CNLC. Tren=tris(2-aminoethyl)amine, OMeLC=4-(acryloyl-
oxybutyloxy)phenyl-4’-(methoxy)benzoate, CNLC=4-(acryloyloxybutyl-
oxy)phenyl-4’-(cyano)benzoate.
Synthesis of hybrid zinc oxide nanoparticles in solution: For-
mation of ZnO NPs decorated with hyperbranched thermo-
tropic LC polymers was performed in THF by using a one-
step hydrolysis of the organometallic precursor, dicyclohexyl
Results and Discussion
Synthesis of liquid crystals: Two branched thermotropic
liquid crystals were synthesized by a Michael addition reac-
tion between Tren and OMeLC or CNLC, as depicted in
Scheme 1.[21] Briefly, a solution of Tren in chloroform was
added dropwise at 08C to a reaction flask containing an
excess of mesogenic acrylate in the same solvent. The mix-
ture was then stirred at room temperature. The reaction was
zinc [ZnACTHUNGTERNN(UG C6H11)2] ([Zn(Cy)2]). This method generally in-
volves the use of amine ligands with long alkyl chains as sta-
bilizers and takes advantage of the exothermic reaction
commonly displayed by organometallic complexes in air.[19]
Using [Zn(Cy)2] as a precursor provides a safer synthetic
method than other organometallic complexes, such as dieth-
yl zinc, which was used by Richter and co-workers[17] and is
a precursor with more easily controlled hydrolysis.[24] In
a typical experiment, a given quantity of either Tren-
OMeLC or TREN-CNLC was dissolved in a volume of dry
THF. One equivalent of [Zn(Cy)2] was then added and the
mixture was allowed to stand for 1 h under an argon atmos-
phere. For both Tren-OMeLC and Tren-CNLC, three differ-
ent mixtures with molar ratios of 0.2, 0.5, and 1.0 Tren-LC/
[Zn(Cy)2] were prepared. The vial containing the mixture
was then removed from the inert conditions and opened to
the atmosphere. At this point the organozinc precursor was
exposed to moisture and formed zinc oxide.
1
monitored by H NMR through the decrease of alkene sig-
nals of the acrylate reagent at d=6 ppm upon completion.
The excess acrylate was then removed by washing the result-
ing product in hot ethanol to give the final products as Tren-
OMeLC and Tren-CNLC in the form of viscous gums.
Although both OMeLC and CNLC precursors exhibit
a monotropic mesophase, differential scanning calorimetry
(DSC) demonstrates that both branched polymers present
a similar behavior with a stable liquid crystalline state (see
Table 1). Figure 1a shows the thermogram for Tren-
OMeLC. It presents a glass temperature at À228C. This be-
havior is normally observed for larger polymers, but has al-
ready been reported for low-molar-mass compounds super-
cooled to a glassy state.[22,23] An isotropic/LC phase transi-
tion was also recorded at around 158C. The texture ob-
served between crossed polarizers by polarized optical mi-
croscopy (POM; Figure 1b and c) together with the X-ray
A turbid yellow luminescent suspension was formed when
the Tren-LC/[Zn(Cy)2] ratio was equal to 0.2, whereas the
higher ratios (0.5 and 1.0) led to the formation of stable lu-
minescent solutions of ZnO NPs (Figure 2b). In the first
case, the amount of LC was not enough to stabilize the ZnO
NPs, which thus leads to the formation of both aggregates
Chem. Eur. J. 2012, 18, 8084 – 8091
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8085

J.-D. Marty, M. L. Kahn et al.
and dispersed NPs as observed in TEM micrographs (see S2
in the Supporting Information). In the latter two cases,
transparent colloidal solutions were obtained with no pre-
cipitation over several weeks. The formation of isotropic
regular ZnO NPs by using Tren-based LC solutions has
been further confirmed by TEM. Thus, in a system in which
the synthetic molar ratio of Tren-OMeLC/[Zn(Cy)2] is set to
0.5:1, NPs of 3.4ACTHNUTRGNEUNG(0.3) nm were obtained (see Figure 2a). This
is in agreement with DLS measurements of a colloidal solu-
tion of Tren-OMeLC/ZnO hybrid in THF (5.5 mm) in which
Tren-OMeLC alone presents a Stokes radius of less than
2 nm and the Tren-OMeLC/ZnO hybrid has a radius of
around 4.3
ACHTUNGTRENNUNG
CNLC/ZnO presenting an average diameter of 6.0ACHTUNGTRENNUNG
(see S3 in the Supporting Information). The difference in
size may come from the difference in the kinetics of the re-
action. In addition, the systems also conserved their yellow
luminescent properties, which originate from the ZnO NPs
(Figure 2c). However, when the ratio Tren-LC/[Zn(Cy)2]
was equal to 1.0, the yellow luminescence was masked due
to the high quantity of organic material (see S4 in the Sup-
porting Information). These results prove that an interaction
takes place between the LC and ZnO NPs and is an efficient
one as confirmed by the stabilization of NPs in solution. To
better understand the precise nature of these interactions
1
and the role of the stabilizerꢂs architecture, further H NMR
spectroscopy experiments were performed in solution, first
with LC moieties only (OMeLC and CNLC) and then with
the branched LCs (Tren-OMeLC and Tren-CNLC).
Figure 1. a) Thermogram for Tren-OMeLC during a heating/cooling cycle
at a rate of 58CminÀ1. Blue and red arrows correspond to cooling and
heating processes, respectively. b,c) Thread-like texture of Tren-OMeLC
and Tren-CNLC, respectively, observed between crossed polarizers on
cooling from the isotropic state. d) X-ray scattering pattern and e) the
corresponding diffraction intensity profile for TREN-OMeLC. The diffu-
sion at small angles close to the center of the 2D X-ray image is related
to reflections from the sample holder. The increased intensity (orange
color) shows that the mesogens were partially aligned.
Interactions between Tren-LC and [Zn(Cy)2] or ZnO NPs,
an NMR spectroscopic study: The ability of LC acrylate
moieties alone to induce the stabilization of ZnO NPs was
first evaluated by forming ZnO NPs in a solution of LC ac-
rylate (OMeLC or CNLC) in THF. Similar results were ob-
tained for both types of LCs. In a typical experiment,
[Zn(Cy)2] was added to a solution of LC acrylate with an
LC/[Zn(Cy)2] ratio of 3:1. This ratio was chosen to corre-
spond to the amount of LC present in the Tren-LC/
[Zn(Cy)2] mixture (0.5:1.0). To understand any possible in-
teractions occurring between the organozinc precursor and
the LC, NMR spectroscopy of these mixtures was carried
out in [D8]THF. No clear modifications of the NMR signal
of the LC molecules upon introduction of [Zn(Cy)2] were
observed (see Figure S5A and S5B in the Supporting Infor-
mation). When the NMR tube containing the LC/[Zn(Cy)2]
mixture was then opened to the air (hydrolysis of the pre-
cursor to form zinc oxide), the LC NMR signals stay un-
modified but a second set of NMR signals slowly start to
appear. These new peaks are much broader than the original
LC NMR signal (see Figure S5A and S5B in the Supporting
Information). Furthermore, DOSY experiments showed a re-
duced self-diffusion coefficient for these new species (2.0ꢃ
10À10 and 2.6ꢃ10À10 m2 sÀ1 for OMeLC and CNLC solutions,
respectively) compared with that of the LC molecule alone
(9.2ꢃ10À10 and 9.9ꢃ10À10 m2 sÀ1 for OMeLC and CNLC, re-
spectively). The slow-diffusing species can be clearly identi-
Figure 2. a) TEM micrograph and histogram from a Tren-OMeLC/ZnO
solution showing ZnO NPs of 3.4ACHTNUTRGNE(NUG 0.3) nm (ca. 300 nanoparticles ana-
lyzed) obtained from a synthetic ratio of Tren-OMeLC/[Zn(Cy)2] 0.5:1.
b) The colloidal solution in THF and c) the same solution showing lumi-
nescence under a UV lamp at l=365 nm.
8086
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 8084 – 8091

ZnO/Liquid Crystalline Nanohybrids
FULL PAPER
1
fied by a diffusion-filtered H NMR experiment that shows
no presence of alkene protons from the acrylate (see Fig-
ure S5C and S5D in the Supporting Information). An inte-
1
gration of H NMR signals in the diffusion-filtered spectrum
resulted in an unexpected increase in the alkane proton
number. Two possible explanations for this observation can
be suggested. The first possibility is that the conjugate addi-
tion of dicyclohexyl zinc to the acrylate forms the 1,4-addi-
tion product, a reaction previously reported in the litera-
ture.[25,26] The second possibility involves an acrylate poly-
merization, which would form larger, slower-diffusing spe-
cies. Whatever the nature of this reaction may be, the NMR
spectroscopy study does not allow us to precisely identify
the nature of the interaction with the NPs. The observed re-
sults may come from either interactions of the modified LC
with the NP surface or the formation of polymeric species
(interacting or not with the NPs). The type of interaction, if
any, did not occur with specific atoms because all the proton
signals experienced a change in electronic environment.
This may lead us to conclude that the type of interaction
was of a steric nature due to the vicinity of ZnO NPs. In any
case, this was not sufficient to stabilize most of the formed
NPs, as confirmed by the observed precipitation in the reac-
tion vial and the formation of ZnO aggregates clearly ob-
served in TEM micrographs (see Figure S6 in the Support-
ing Information). Moreover, the remaining solution presents
no additional luminescent properties.
Figure 3. 1H NMR spectra at 291 K for Tren-CNLC only in [D8]THF
(2.2 mm; bottom), after addition of 8 equiv of [Zn(Cy)2] before exposure
to air (middle) and after exposure to air for 4 h (top). Proton signal as-
signment: a) NCH2CH2COO-, b) -NCH2CH2N-, and c) -NCH2CH2COO-.
*
#: [Zn(Cy)2] signals, : THF).
other hand, after hydrolysis of [Zn(Cy)2], a global sharpen-
ing of peaks was observed, which indicates the generation of
a more homogeneous environment. Thus, the presence of
the NPs induces structural rearrangements within the mac-
romolecules. However, interaction between the Tren-LC
molecules and the ZnO NPs was not clearly observed.
DOSY experiments were then performed on these systems
and analyzed by using the CONTIN method (see Figure S7E
and F in the Supporting Information). The average Tren-LC
diffusion coefficients are only slightly affected by the pres-
ence of ZnO NPs. The major change can be observed in the
diffusion coefficient distribution profiles, which show an in-
creasing population of slower diffusing species in the pres-
ence of ZnO NPs. This can be clearly observed for the
decay of the Tren-LC signals during the diffusion measure-
ments, in which some Tren-LC molecules show slightly
slower diffusion in the presence of ZnO NPs. We also con-
firmed that, in the absence of ZnO NPs, the Tren-LC mole-
cules have a more uniform signal decay profile (see Fig-
ure S7G and H in the Supporting Information). No trans-
ferred NOE could be detected in a NOESY experiment on
the post-hydrolysis state (see Figure S7I in the Supporting
Information), which excludes a fast exchange of Tren-LC
molecules between the NP surfaces and the solution.
In conclusion, the NMR studies demonstrate the existence
of specific interactions between [Zn(Cy)2] precursor and
Tren-LC through nitrogen atoms. The extent of this interac-
tion depends strongly on the nature of the LC moiety. After
hydrolysis and NP formation, there is probably no specific
chemical interaction between Tren-LC molecules and the
ZnO NPs. However, the nanoparticles do affect the structur-
al arrangement of the Tren-LC molecules in solution to
a certain extent, which indicates steric interactions between
the NPs and the macromolecules.
Although LC moieties alone were not able to lead to the
formation of stable ZnO NPs, the previously mentioned re-
sults showed that stabilization of ZnO NPs by Tren-based
LCs was possible. Interactions between the zinc precursor
and the branched polymer host were assessed by using
1H NMR spectroscopy. NMR experiments were carried out
in [D8]THF for Tren-LC alone, Tren-LC/[Zn(Cy)2] (pre-hy-
drolysis), and Tren-LC/ZnO (post-hydrolysis). Similar re-
sults were obtained for both Tren-OMeLC and Tren-CNLC.
1
The H NMR spectra show a broadening of the LC NMR
signals in the presence of [Zn(Cy)2] (Figure 3 and Fig-
ure S7A and B in the Supporting Information). For Tren-
CNLC, the broadening was only observed for protons in the
a-position with respect to amine functionalities in the poly-
meric core (d(1H)=2.81, 2.57, and 2.47 ppm). This suggests
that in the pre-hydrolysis state the zinc precursor interacts
with the nitrogen atoms present in the LC. This phenomen-
on has been previously reported for nitrogen-containing
molecules, such as polyamidoamine-based dendritic macro-
molecules[18] or amines with long alkyl chains.[27] For Tren-
OMeLC, the broadening was more uniformly observed for
all the Tren-LC resonance signals and it is possible that car-
bonyl functions are also involved in the interaction between
Tren-OMeLC and [Zn(Cy)2]. In general, the Tren-LC NMR
signals have a complex line shape that indicates a certain
degree of spatial inhomogeneity in the sample, which is un-
affected by thermal agitation, even in the case of LC only
(see Figure S7C and D in the Supporting Information). Ad-
dition of [Zn(Cy)2] to Tren-LC (pre-hydrolysis state) in-
duced only slight changes in this inhomogeneity. On the
Chem. Eur. J. 2012, 18, 8084 – 8091
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8087

J.-D. Marty, M. L. Kahn et al.
this excitation wavelength a competition exists between the
two previously discussed photoemissions.
Liquid crystal behavior in bulk: We study here the meso-
morphic behavior of the synthesized bulk Tren-LC/ZnO hy-
brids by using DSC, POM, and X-ray scattering. Between
crossed polarizers the Tren-OMeLC/ZnO hybrid displays
a birefringent texture at room temperature, owing to the
anisotropic nature of the liquid crystal. This texture was ob-
served up to around 298C, after which the material starts
losing its birefringence. Thermograms for Tren-OMeLC/
ZnO confirm these observations. In fact, a transition peak
occurs at 298C and a change in heat capacity linked to a Tg
could be observed at À58C (see Table 1). Between these
two extremes (À5 and 298C), SAXS measurements showed
diffused rings (d=4.5 ꢁ) characteristic of the nematic LC
phase. Because pure Tren-OMeLC exhibits the LC state be-
tween À22.0 and 108C, we can clearly conclude that the
presence of ZnO NPs shifted the mesophase temperature
window towards higher temperatures. Interestingly, the op-
posite phenomenon was observed for Tren-CNLC/ZnO NPs
(see Table 1). In that case, the introduction of ZnO NPs was
accompanied by a decrease in the glass transition tempera-
ture (destabilization) of LC phases, as generally reported in
the literature.[30] This might suggest that Tren-OMeLC inter-
acts with ZnO NPs in a very specific way that favors the ex-
istence of stable mesophases, as we observed previously
with more fluxional systems.[15]
Figure 4. Emission spectra for a solution of Tren-OMeLC/ZnO nanocom-
posite in THF (0.011m with respect to LC) when excited at wavelengths
Synthesis of Tren-LC/ZnO nanocomposites in the absence
of a solvent: To take advantage of the interactions between
LC moieties and ZnO NPs, we decided to perform the for-
mation of ZnO NPs directly inside the bulk LC mesophase.
Tren-CNLC and Tren-OMeLC exhibit a nematic LC phase
at temperatures close to ambient. This avoids any organo-
zinc precursor thermal breakdown before NPs are produced.
In a typical LC/ZnO nanocomposite synthesis, a given
amount of Tren-LC was dissolved in a minimum amount of
dry THF under an argon atmosphere. To this solution one
equivalent of [Zn(Cy)2] was then added (Tren-LC/[Zn(Cy)2]
ratio 0.4:1). The prerequisite for the chosen quantities of
products is that the expected ZnO NP content in the final
product does not exceed 10% of the total weight of the
system. This is done as a precaution so as not to disrupt any
LC order in the material. Still under inert conditions, the
vial was left open and THF was allowed to evaporate from
the mixture. Some solvent was mixed in at the pre-hydroly-
sis state to ensure good material homogeneity. Once the
mixture was dried it was placed in an oil bath at 208C and
opened to the atmosphere. Moisture in the air breaks down
the precursor and nucleation and growth of ZnO NPs fol-
lows. The material was then investigated by TEM. Interest-
ingly, by using Tren-OMeLC in a nematic state, it was found
that the ZnO NPs grew in a certain preferred direction. The
NPs seem to adopt a worm-like morphology and present
a dense population. This anisotropy can be clearly seen in
Figure 5. The nanoworms have an average width of 2.5-
~
*
*
of 340 ( ), 360 ( ), and 380 nm ( ). A)–C) Images of the solution when
irradiated at wavelengths of 380, 360, and 340 nm, respectively.
Photoluminescent properties of Tren-LC/ZnO nanohybrids:
As commonly found in the literature, zinc oxide presents in-
teresting luminescent properties in the visible region of the
spectrum. Absorption spectroscopy on a given NP colloidal
solution of Tren-OMeLC/ZnO (Tren-OMeLC/[Zn(Cy)2]
0.5:1), shows an absorption up to l=355 nm (ꢀ3.49 eV).
This is typical physical behavior for nanometric zinc oxide
NPs with a bandgap at approximately l=365 nm (3.4 eV).
The photoluminescent properties of this solution are pre-
sented in Figure 4. Three emissions were observed depend-
ing on the excitation wavelength. A yellow emission cen-
tered at l=565 nm was observed when the Tren-OMeLC/
ZnO solution was irradiated at a wavelength of l=340 nm,
which is known for ZnO NPs. The origin of this emission is
attributed to surface defects, such as oxygen vacancies in the
structure.[28] Interestingly, when the sample was subject to an
excitation wavelength of l=380 nm, an emission band cen-
tered at l=460 nm was recorded. This blue emission has
been previously reported by our group for ZnO NPs pre-
pared following this organometallic synthesis.[29] When the
system was irradiated at l=360 nm, a broad white-colored
emission band was observed. This is due to the fact that at
8088
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 8084 – 8091

ZnO/Liquid Crystalline Nanohybrids
FULL PAPER
does not affect the shape of the NPs for amine ligands with
long alkyl chains.
We studied the mesomorphic behavior of the synthesized
bulk Tren-LC/ZnO hybrids by using DSC (see Figure S10 in
the Supporting Information), POM (see Figure 6) and X-ray
Figure 6. POM images for a) Tren-OMeLC/ZnO and b) Tren-CNLC/ZnO
at RT after cooling from the isotropic state.
Figure 5. a,b) TEM micrographs obtained for Tren-OMeLC/ZnO nano-
composite synthesized in the liquid crystalline state (Tren-LC/[Zn(Cy)2]
ratio 0.4:1) at 08C, showing anisotropic wire-like nanoparticles with
scattering. A nematic mesophase was still observed by
SAXS between À1 and 378C (see Figure S11 in the Support-
ing Information). Interestingly, as observed for ZnO NPs
generated in solution, the presence of ZnO NPs shifts the
mesophase temperature window towards higher tempera-
tures. When ZnO NPs were synthesized in the isotropic
phase of the LC, DSC results showed a near-complete disap-
pearance of the N–I transition peak and poor birefringence
under a POM. This confirmed that the LC state was some-
how perturbed by the inorganic content introduced. In this
regard, we suggest that the growth of ZnO NPs outside the
mesophase leads to a lack of morphological control and the
formation of large ZnO aggregates (observed in TEM).
Consequently this leads to a disruption of the LCꢂs intrinsic
N phase. These results demonstrate that preparation of ZnO
NPs inside the LC phase is crucial to control the properties
of the LC/NP composite.
lengths that range from 5 to 100 nm and a width of 2.5ACHTNUTRGNE(NUG 0.2) nm; scale
bars are 100 nm. c,d) The two HRTEM micrographs represent perpen-
dicular and parallel orientations to the z plane, respectively. Insets: The
corresponding numerical FT-diffractograms for the given crystalline
region.
ACHTUNGTRENNUNG(0.2) nm and various lengths up to around 100 nm. The crys-
talline nature of the ZnO nanoworms was then demonstrat-
ed by using high-resolution transmission electron microsco-
py (HRTEM). HRTEM micrographs and a typical numeri-
cal diffractogram (Fourier transform) are shown in Figure 5.
The measurement of the radial distances of the different
spots corresponds to the hexagonal wurtzite structure of
ZnO crystals (a=b=3.2 ꢁ and c=5.2 ꢁ, space group
P6mc). This experiment also demonstrates that the aniso-
tropic growth proceeds along the c axis. To prove that the
anisotropic growth of ZnO nanoparticles comes about due
to the order existing in the LC state, we have performed ex-
actly the same synthesis but at a temperature of 258C. At
this temperature, the LC alone is in the isotropic state ac-
cording to its DSC thermogram (see Figure 1). Results from
TEM showed ZnO NPs with no sort of organization or ani-
sotropy (see Figure S8 in the Supporting Information).
Conclusion
Herein we reported the use of thermotropic liquid crystal
macromolecules as efficient stabilizing agents for the direct
synthesis of isotropic ZnO NPs in solution. Moreover, we
demonstrated that the structural order present in these
branched LCs can be exploited for anisotropic growth of
ZnO nanostructures. Tren-OMeLC allowed the formation of
ZnO nanoworms, thus proving the direct effect of the organ-
ized nematic phase on the morphology of the nanoparticles.
The control of shape and size at the nanoscale level is desir-
able due to its influence on potential applications. There-
fore, we suggest that LCs in which the mesophase only de-
pends on temperature can be ideal candidates as organiza-
tion agents at the nanoscale level.
The in situ synthesis of ZnO NPs within Tren-CNLC in
the absence of solvent was then performed. The same syn-
thetic procedure as for Tren-OMeLC was adopted; however,
the reaction temperature was set at 208C (mesophase tem-
perature for Tren-CNLC; see Table 1). TEM micrographs
for the resultant product showed the formation of isotropic
NPs of an average diameter of 3.2ACTHUNTGRNEUNG(0.3) nm (see Figure S9 in
the Supporting Information). In this case, the LC state did
not have an important effect on the growth of ZnO NPs
unlike the case for Tren-OMeLC, this may be mainly due to
a difference in the interactions with [Zn(Cy)2] or ZnO NPs,
as demonstrated previously. The effect of hydrolysis temper-
ature could not be excluded. However because the branched
LCs are not liquid crystalline in the same temperature
range, no experiment can confirm this effect. Nevertheless,
we have shown that the cooling of the reaction temperature
Chem. Eur. J. 2012, 18, 8084 – 8091
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8089

J.-D. Marty, M. L. Kahn et al.
),4.02 (m, 12H; -COO-CH2-), 4.11 (m, 12H; -CH2-O-Ar), 7.00 (d, J=
9 Hz, 12H; Ar-H-1), 7.61 (d, J=9 Hz, 12H; Ar-H-2), 7.73 ppm (m, 24H;
Ar-H-3,4); MS (MALDI-TOF): m/z calcd for C126H132N10O18: 2074.0;
found: 2074.0.
Experimental Section
Materials: Unless otherwise stated, all chemicals were used as provided
by suppliers (Acros, Fluka, or Aldrich) without any further modifications.
Oxygen- and moisture-sensitive substances and reactions were handled
either in an MBraun Inert Gas System or under an argon atmosphere in
carefully dried glassware, by using standard Schlenk techniques. THF
used for sensitive compounds was obtained from an MBraun Purificator
followed by degassing. The residual water content was systematically
measured by Karl Fischer coulometric titration using a Metrohm instru-
ment. Dicyclohexyl zinc precursor was purchased from NANOMEPS.
Synthesis of OMeLC/ZnO and CNLC/ZnO (in solvent): In a screw-top
vial, the LC (24 mg, 6.6ꢃ10À2 mmol) was dissolved in dried and <de-
gassed THF (6 mL) under an inert atmosphere inside a glovebox. Dicy-
clohexyl zinc precursor (5 mg, 2.2ꢃ10À2 mmol) was then added to the so-
lution. The reaction vial was closed and left to stand for 30 min. The solu-
tion was then taken out of the glovebox and opened to the atmosphere
to decompose the precursor and form ZnO nanoparticles.
Synthesis of Tren-OMeLC/ZnO or Tren-CNLC/ZnO (in solvent): In
a screw-top vial, Tren-LC (1.1ꢃ10À2 mmol) was dissolved in dried and
degassed THF (6 mL) under an inert atmosphere inside a glovebox. Di-
cyclohexyl zinc precursor (5 mg, 2.2ꢃ10À2 mmol) was then added to the
solution. The reaction vial was closed and left to stand for 30 min. The
solution was then taken out of the glove box and opened to the atmos-
phere in order to decompose the precursor and form ZnO nanoparticles.
4’-(4-Acryloyloxybutyloxy)phenyl-4-(methoxy)benzoate (OMeLC): The
synthesis of OMeLC was adapted from a literature procedure.[31] Diethy-
lazodicarboxylate (5.6 g, 32.0 mmol) was added dropwise to a solution of
4-(methoxy)phenyl-4’-(hydroxy)benzoate (6.4 g, 34.2 mmol), 4-hydroxy-
butyl acrylate (7.8 g, 32 mmol), and triphenylphosphine (TPP, 8.4 g,
32.0 mmol) in THF (150 mL). The reaction mixture was stirred for 24 h
at RT, after which the volatile components were removed under vacuo.
To precipitate phosphine the residue was dissolved in cyclohexane/ethyl
acetate (25 mL, 20:5 v/v). It was then recrystallized from ethanol to give
a white crystalline product (yield: 60%). 1H NMR (500 MHz, [D8]THF):
d=1.82–1.90 (m, 4H; -CH2-CH2-CH2-CH2-), 3.86 (s, 3H; -OCH3), 4.00 (t,
J=6.5 Hz, 2H; -COO-CH2-), 4.21 (t, J=6.5 Hz, 2H; -CH2-O-Ar), 5.81
(m, 1H; -OCO-CH=CH2), 6.05 (m, 1H; -OCO-CH=CH2), 6.32 (m, 1H;
-OCO-CH=CH2), 6.93 (d, J=9.0 Hz, 2H; Ar-H-1), 7.02 (d, J=9.0 Hz,
2H; Ar-H-4), 7.10 (d, J=9.0 Hz, 2H; Ar-H-2), 8.10 ppm (d, J=9.0 Hz,
2H; Ar-H-3).
Synthesis of Tren-OMeLC/ZnO (no solvent): Tren-OMeLC (16 mg, 6.7ꢃ
10À3 mmol) or Tren-CNLC (16 mg, 7.7ꢃ10À3 mmol) was transferred to
a Schlenk tube and dissolved in dry THF (300 mL) under inert conditions
inside a glovebox. Dicyclohexyl zinc precursor (4 mg, 1.7ꢃ10À2 mmol)
was then added to the solution containing the LC. THF in the reaction
tube was then evaporated under vacuo (still inside the glovebox). When
the mixture was dry, it was placed in an ice bath at the desired tempera-
ture. After which the Schlenk tube was opened to the atmosphere.
Characterization: To determine the structural characteristics of obtained
products, NMR spectra were recorded by using a Bruker Avance 500
spectrometer equipped with a 5 mm triple-resonance inverse Z-gradient
probe. Unless otherwise mentioned, measurements were performed in
deuterated THF or chloroform at RT. The chemical shifts, d, are given in
ppm. The 2D NOESY measurements were performed with a mixing time
of 100 ms. All diffusion measurements were made by using the stimulated
echo pulse sequence with bipolar gradient pulses and were processed
with the Laplace inversion routine CONTIN (Topspin software). The
temperatures at which glass transitions (Tg) and crystallizations (Tc)
occur were determined by using DSC by using a Mettler Toledo DSC
1 STARe System thermal analysis calorimeter equipped with a gas con-
troller GC200. Temperatures at which phase transitions occurred were
taken to be the temperature at the top of the DSC peaks as the tempera-
ture decreased at different rates: 20, 10, 5, and 18CminÀ1, and finally ex-
trapolated to 08CminÀ1. Tg was measured as the temperature increased
at a rate of 108CminÀ1. Characterization of liquid crystalline phases was
performed by using a hot-stage FP 82HT (Mettler Toledo) under a polar-
ized light optical microscope BX50 from Olympus. For X-ray scattering
measurements, a microfocus rotating anode X-ray source (Rigaku Micro
Max-007 HF) combined with high-performance multi-layer optics and
three-pinholes collimation provided intense X-ray radiation on the
sample. The sample holder was mounted on an X-Y stage. A two-dimen-
sional detector (image plate from Mar Research) collected the scattered
radiation. The sample–detector distances were varied to obtain wide and
small angles and calibration of this distance was performed by using
silver behenate as a reference. DLS measurements were carried out by
using a Malvern Instruments Nano-ZS equipped with a He–Ne laser (l=
633 nm). The correlation function was analyzed by using the general pur-
pose method (NNLS) to obtain the distribution of diffusion coefficients
(D) of the solutes. The apparent diameter was then determined by using
the Stokes–Einstein equation. Mean diameter values were obtained from
three different runs. Standard deviations were evaluated from diameter
distribution. Besides scattering measurements, the morphology of the ob-
tained nanohybrids was studied by using TEM. Samples for TEM were
prepared by slow evaporation of droplets of colloidal solution deposited
on a carbon-coated 200 mesh copper TEM grid (Ted Pella). The samples
were then carefully dried overnight under a pressure of 5ꢃ10À5 mbar by
using a BOC Edward turbomolecular pump. The TEM experiments were
performed at the microscopy service of University Paul Sabatier (TEMS-
CAN) by using a JEOL JEM1011 electron microscope operating at
100 kV with a resolution point of 0.45 nm. The nanoparticle size-distribu-
4-Cyano-4’-(4-acryloyloxybutyloxy)biphenyl (CNLC): Diethylazodicar-
boxylate (4.4 g, 25.5 mmol) was added dropwise to a solution of 4-hy-
droxy-4’-cyano biphenyl (3.6 g, 18.6 mmol), 4-hydroxybutyl acrylate
(3.7 g, 25.5 mmol), and TPP (6.7 g, 25.5 mmol) in THF (50 mL). The re-
action mixture was stirred for 24 h at RT, after which the volatile compo-
nents were removed under vacuo. To precipitate phosphine, the residue
was dissolved in cyclohexane/ethyl acetate (25 mL, 20:5 v/v). It was then
recrystallized from ethanol to give a white crystalline product (yield:
60%). 1H NMR (500 MHz, [D8]THF): d=1.84–1.90 (m, 4H; -CH2-CH2-
CH2-CH2-), 4.06 (t, J=6.5 Hz, 2H; -COO-CH2-), 4.21 (t, J=6.5 Hz, 2H;
-CH2-O-Ar), 5.75 (m, 1H; -OCO-CH=CH2), 6.12 (m, 1H; -OCO-CH=
CH2), 6.33 (m, 1H; -OCO-CH=CH2), 7.01 (d, J=9 Hz, 2H; Ar-H-1),
7.62 (d, J=9 Hz, 2H; Ar-H-2), 7.61–7.77 ppm (m, 4H; Ar-H-3,4).
Tren-OMeLC: Tren-OMeLC was synthesized as previously described.[20]
Briefly, 4-(acryloyloxybutyloxy)phenyl-4’-(methoxy)benzoate (882 mg,
2.38 mmol) was transferred to a round-bottomed flask and dissolved in
chloroform (8 mL) with stirring. Tris(2-aminoethyl)amine (50 mg,
0.34 mmol) was then dissolved in chloroform (1 mL) and added dropwise
to the reaction flask over an ice bath. The flask was then left to stir at
RT under an argon atmosphere for 4 d, during which the initially color-
less reaction mixture turned pale pink. Any volatile solvents were then
removed under vacuo and the residue was stirred in hot ethanol for
10 min. This latter solvent was removed while hot by using a syringe.
This was repeated three consecutive times. Residual solvent was removed
under reduced pressure to give a pink viscous product (yield: 21%).
1H NMR (500 MHz, [D8]THF): d=1.73–1.85 (m, 24H; -CH2-CH2-), 2.45
(m, 12H; -N-CH2-CH2-COO-), 2.55 (m, 12H; -NCH2CH2N-), 2.79 (m,
12H; -N-CH2-CH2-COO-), 3.85 (m, 18H; -OCH3), 3.98 (m, 12H; -COO-
CH2-), 4.11 (m, 12H; -CH2-O-Ar), 6.91 (m, 12H; Ar-H-1), 7.01 (m, 12H;
Ar-H-4), 7.08 (m, 12H; Ar-H-2), 8.07 ppm (m, 12H; Ar-H-3); MS
(MALDI-TOF): m/z calcd for C132H150N4O36: 2368.0; found: 2368.4.
Tren-CNLC:
4-Cyano-4’-(4-acryloyloxybutyloxy)biphenyl
(300 mg,
0.93 mmol) was transferred to a round-bottomed flask and dissolved in
chloroform (8 mL) with stirring. Tris(2-aminoethyl)amine (20 mg,
0.13 mmol) was then dissolved in chloroform (1 mL) and added dropwise
to the reaction flask over an ice bath. The rest of procedure is as de-
scribed for Tren-OMeLC. The final product was obtained as a pale
yellow gum (yield: 13%). 1H NMR data is available in the Supporting In-
formation section. 1H NMR (500 MHz, [D8]THF): d=1.80–1.90 (brs,
24H; -CH2-CH2-), 2.44 (t, J=7.5 Hz, 12H; -N-CH2-CH2-COO-), 2.54
(brs, 12H; -NCH2CH2N-), 2.77 (t, J=7.5 Hz, 12H; -N-CH2-CH2-COO-
8090
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 8084 – 8091

ZnO/Liquid Crystalline Nanohybrids
FULL PAPER
tion histograms were determined by using magnified TEM images. The
size distribution of the particles was determined by measuring a minimum
of 200 particles of each sample. The observed size distributions were ana-
lyzed in terms of Gaussian statistics. For photoluminescent analysis emis-
sion, spectra were recorded by using a Horiba–Jobin–Yvon Fluoromax-4
spectrofluorometer equipped with a xenon lamp.
[14] K. Kanie, A. Muramatsu, J. Am. Chem. Soc. 2005, 127, 11578–
11579.
[15] S. Saliba, Y. Coppel, P. Davidson, C. Mingotaud, B. Chaudret, M. L.
Kahn, J. D. Marty, J. Mater. Chem. 2011, 21, 6821–6823.
[16] R. O. Moussodia, L. Balan, C. Merlin, C. Mustin, R. Schneider, J.
Mater. Chem. 2010, 20, 1147–1155.
[17] T. V. Richter, F. Schuler, R. Thomann, R. Mulhaupt, S. Ludwigs,
Macromol. Rapid Commun. 2009, 30, 579–583.
[18] S. Saliba, C. V. Serrano, J. Keilitz, M. L. Kahn, C. Mingotaud, R.
Haag, J. D. Marty, Chem. Mater. 2010, 22, 6301–6309.
[19] M. Monge, M. L. Kahn, A. Maisonnat, B. Chaudret, Angew. Chem.
2003, 115, 5479–5482; Angew. Chem. Int. Ed. 2003, 42, 5321–5324.
[20] S. Saliba, Y. Coppel, M.-F. Achard, C. Mingotaud, J. D. Marty, M. L.
Kahn, Angew. Chem. 2011, 123, 12238–12241; Angew. Chem. Int.
Ed. 2011, 50, 12032–12035.
Acknowledgements
This work has been supported and financed by the Centre National de la
Recherche Scientifique (CNRS) together with the European Marie Curie
Project NANOTOOL (MEST-CT-2005-020195). We also acknowledge
the EU for financial support (FEDER-35477 “Nano-objets pour la bio-
technologie”). We thank M.-F. Achard, CRPP laboratory, for SAXS ex-
periments and fruitful discussions.
[21] H. J. Liu, Y. Chen, D. D. Zhu, Z. Shen, S. E. Stiriba, React. Funct.
Polym. 2007, 67, 383–395.
[22] Y. Gonzꢇlez, M. B. Ros, J. L. Serrano, B. Palacios, M. R. de La
Fuente, M. A. P. Jubindo, Chem. Mater. 1997, 9, 2786–2797.
[23] G. S. Attard, C. T. Imrie, F. E. Karasz, Chem. Mater. 1992, 4, 1246–
1253.
[24] Comprehensive Organometallic Chemistry II, Vol. 3 (Eds.: E. W.
Abel, F. G. A. Stone, G. Wilkinson), Elsevier, Oxford, 1995, 175–
203.
[25] C. Petrier, J. L. Luche, C. Dupuy, Tetrahedron Lett. 1984, 25, 3463–
3466.
[26] C. K. Reddy, A. Devasagayaraj, P. Knochel, Tetrahedron Lett. 1996,
37, 4495–4498.
[27] M. L. Kahn, M. Monge, V. Colliere, F. Senocq, A. Maisonnat, B.
Chaudret, Adv. Funct. Mater. 2005, 15, 458–468.
[28] A. van Dijken, E. A. Meulenkamp, D. Vanmaekelbergh, A. Meijer-
ink, J. Lumin. 2000, 90, 123–128.
[29] M. L. Kahn, T. Cardinal, B. Bousquet, M. Monge, V. Jubera, B.
Chaudret, Chemphyschem 2006, 7, 2392–2397.
[1] H. K. Bisoyi, S. Kumar, Chem. Soc. Rev. 2011, 40, 306–319.
[2] P. K. S. Antharjanam, E. Prasad, New J. Chem. 2010, 34, 420–425.
[3] L. J. Martꢄnez-Miranda, K. M. Traister, I. Melꢅndez-Rodrꢄguez, L.
Salamanca-Riba, Appl. Phys. Lett. 2010, 97.
[4] M. Akimoto, S. Kundu, K. Isomura, I. Hirayama, S. Kobayashi, K.
Takatoh, Mol. Cryst. Liq. Cryst. 2009, 508, 1/ACHTNUGTRENNUG[363]–13/CAHTUNGTRENN[UGN 375].
[5] J. D. Busbee, A. T. Juhl, L. V. Natarajan, V. P. Tongdilia, T. J. Bun-
ning, R. A. Vaia, P. V. Braun, Adv. Mater. 2009, 21, 3659-.
[6] P. A. Kossyrev, A. J. Yin, S. G. Cloutier, D. A. Cardimona, D. H.
Huang, P. M. Alsing, J. M. Xu, Nano Lett. 2005, 5, 1978–1981.
[7] S. Khatua, P. Manna, W. S. Chang, A. Tcherniak, E. Friedlander,
E. R. Zubarev, S. Link, J. Phys. Chem. C 2010, 114, 7251–7257.
[8] L. Cseh, G. H. Mehl, J. Am. Chem. Soc. 2006, 128, 13376–13377.
[9] I. Gascon, J. D. Marty, T. Gharsa, C. Mingotaud, Chem. Mater. 2005,
17, 5228–5230.
[10] I. In, Y. W. Jun, Y. J. Kim, S. Y. Kim, Chem. Commun. 2005, 800–
801.
[30] M. Mitov, C. Portet, C. Bourgerette, E. Snoeck, M. Verelst, Nat.
Mater. 2002, 1, 229–231.
[11] L. Zadoina, B. Lonetti, K. Soulantica, A. F. Mingotaud, M. Respaud,
B. Chaudret, M. Mauzac, J. Mater. Chem. 2009, 19, 8075–8078.
[12] B. Kinkead, T. Hegmann, J. Mater. Chem. 2010, 20, 448–458.
[13] A. Demortiꢆre, S. Buathong, B. P. Pichon, P. Panissod, D. Guillon, S.
Begin-Colin, B. Donnio, Small 2010, 6, 1341–1346.
[31] J. D. Marty, M. Mauzac, C. Fournier, I. Rico-Lattes, A. Lattes, Liq.
Cryst. 2002, 29, 529–536.
Received: January 21, 2012
Published online: May 23, 2012
Chem. Eur. J. 2012, 18, 8084 – 8091
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8091