A modulated helical nanofilament phase

Article abstract of DOI:10.1002/anie.201209453

A new liquid crystal phase, denoted modulated helical nanofilament (HNF_(mod)), is formed from a very simple class of biphenyl carboxylates lacking the benzylidene aniline moieties typically found in HNF mesogens. The HNF_(mod) phase represents a novel kind of nanoparticle possessing stacked aromatic rings, with potential applications in organic electronics. Copyright

Full text of DOI:10.1002/anie.201209453

.
Angewandte
Communications
DOI: 10.1002/anie.201209453
Liquid Crystals
A Modulated Helical Nanofilament Phase**
Ethan Tsai, Jacqueline M. Richardson, Eva Korblova, Michi Nakata, Dong Chen,
Yongqiang Shen, Renfan Shao, Noel A. Clark, and David M. Walba*
The modern study of the liquid crystal (LC) phases formed by
bent-core molecules[1] has led to many interesting and
unexpected phenomena, including the first example of
spontaneous reflection symmetry breaking and conglomerate
formation in a bulk fluid phase.[2] Indeed, exploration of the
unique behavior of bent-core materials is currently one of the
most productive frontier research areas in soft matter
science.[3] Perhaps the most complex of the known bent-core
phases, the helical nanofilament (HNF) phase (also known as
the B4 banana phase) has been under serious investigation
since 1997,[1c,d] and continues to be a focus of interest in the
bent-core materials constellation. Here, we report full char-
acterization of the first example of a new phase in this family,
HNF(mod), composed of simple alkoxybiphenylcarboxylate
units and lacking the Schiff base groups found in previously
known HNF mesogens.
The classic HNF phase possesses a unique hierarchical
nanostructure: An assembly of twisted layers stacked to form
well-defined chiral nanorods (individual HNFs ca. 40 nm
diameter), with a structure driven by intra-layer frustration
leading to spontaneous saddle splay, and formation of layers
with negative curvature.[4–6] Solid state NMR data suggest that
within individual HNF layers the structure is crystalline,[7]
though electron diffraction shows that no interlayer positional
correlation exists.[4] The HNFs in turn form a kind of hexatic
LC phase, which freezes into a glassy state at ca. 1108C. When
the mesogens are achiral or racemic, an LC conglomerate of
large heterochiral domains is easily seen in LC cells by
polarized optical microscopy. The bulk HNF phase is
porous,[4,6] and when grown in the presence of other materials,
can produce nanostructured composites.[8] Potential applica-
tions of the HNF phase and composites include nonlinear
optics, organic electronics, photovoltaics, and chiral separa-
tions.
ble benzylideneaniline (Schiff base) moiety. This is problem-
atic for some potential applications. Also, the rarity of the
known HNF phases in the bent-core “structure space”
motivates discovery of new HNF phases in order to better
understand the molecular structural factors leading to their
formation, and to allow future design of functional HNF
materials.
Here we report characterization of seven homologues of
biphenyl-3,4’-diyl bis-(4’-alkoxybiphenyl-4-carboxylate),[10]
(diesters 1(n) (n = 9–15), Figure 1a). All of these exhibit
a new HNF phase possessing in-layer modulation in addition
to the characteristic negative curvature of the layers.
Bent-core phases with in-layer modulation are known,
including for example a tilted polar smectic (SmCP) phase
possessing undulated smectic layers (also known as the B7
banana phase).[11] Here, we designate the new modulated
HNF phase HNF(mod), with the proposed hierarchical struc-
ture for individual HNFs illustrated in Figure 1. As for the
classic HNF phase, we propose that the aromatic cores in the
new phase are tilted, while the tails are extended almost
normal to the local layer plane.[4]
However, for the HNF(mod), in-layer modulation with
a spacing of about 40 ꢀ, not seen in the HNF phase, is
suggested by experimental data. We propose this secondary
modulation results from a structural periodicity with wave
vector parallel to the layers and normal to the polar axis (P),
as indicated in Figure 1c. The precise nature of this periodic
defect structure is not known, though the lack of two-fold
symmetry for rotation about b in diesters 1(n) could
reasonably produce such a periodic structural change by
simple 1808 rotation about b, mediating in-layer “stripes”
about eight molecules wide, as indicated (Figure 1c). Finally,
as for the HNF phase, the twisted layers stack to form helical
nanofilaments (Figure 1d).
Interestingly, of the great many bent-core structures
reported to date, with the exception of a single reported
outlier[9] all HNF mesogens include the hydrolytically unsta-
Evidence for this structural picture of the HNF(mod) phase
formed by diesters 1(n) derives from: 1) polarized optical
microscopy (POM); 2) differential scanning calorimetry
(DSC); 3) X-ray diffraction (XRD) at small angle; and 4)
transmission electron microscopy (TEM). The phase assign-
ments, transition temperatures and enthalpies on cooling, and
HNF(mod) layer spacing (d1) and in-layer modulation dimen-
sions (d2) for diesters 1(n), are given in Table 1. A brief
discussion of the key observations and interpretations leading
to the model illustrated in Figure 1 follows.
As indicated in Table 1, all of the homologues are very
similar in their basic behavior. All show a transition from
isotropic to a more conventional bent-core phase, tentatively
assigned as either a B1 (columnar) phase, or a B2 (tilted
smectic) phase. For the lower homologues, (n = 9–12) the high
[*] Dr. E. Tsai, Dr. J. M. Richardson, Dr. E. Korblova,
Prof. Dr. D. M. Walba
Department of Chemistry and Biochemistry
University of Colorado, 215 UCB
Boulder, CO 80309-0215 (USA)
E-mail: walba@colorado.edu
Dr. M. Nakata, Dr. D. Chen, Y. Shen, R. Shao, Prof. Dr. N. A. Clark
Department of Physics, University of Colorado, 390 UCB
Boulder, CO 80309-0390 (USA)
[**] We thank the Liquid Crystal Materials Research Center (NSF MRSEC
award no. DMR-0820579) for financial support of this work.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201209453.
temperature phases are monotropic with respect to HNF(mod)
.
5254
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 5254 –5257

Angewandte
Chemie
Figure 1. Hierarchical structure of the HNF(mod) phase, with metrics specific to diester 1(9). a) Chemical structure of diesters 1(n). b) Space-filling
model of a reasonable conformation of 1(9), illustrating the proposed orientation of molecules in the smectic layers. c) Sketch of a single helical
nanofilament layer bounded by doubly-ruled saddle surfaces (red and green) representing the HNF layer interfaces. Within the layer the molecules
are organized into “stripes” which extend the entire length of the layer—indicated schematically by the blocks of tilted molecules at the bottom
left of the graphic. d) Twisted layers stack to form HNFs. The layer width and the number of layers in a stack are limited by the free energy cost of
less than ideal curvature, leading to nearly cylindrical HNFs.
Table 1: Phases, transition metrics, and HNF metrics for 1(n).
[b]
[c]
n
Phases, transition temperatures,
and enthalpies[a]
d1
d2
HNF(mod)–178!Iso (exotherm at 172)
9
46.2 39.8
48.3 41.9
50.3 41.9
52.8 41.9
54.6 42.5
55.6 41.9
57.6 42.7
!
!
HNF(mod) (28.4) 166–B1 (18.5) 175–Iso
HNF(mod)–172!Iso (exotherm at 164)
10
11
12
13
14
15
!
!
HNF(mod) (27.4) 164–B1 (15.4) 167–Iso
HNF(mod)–171!Iso (exotherm at 166)
!
!
HNF(mod) (30.0) 164–B2 (19.6) 171–Iso
HNF(mod)–174!Iso (exotherm at 166)
!
!
HNF(mod) (29.2) 160–B1 (22.8) 172–Iso
HNF(mod)–169!B1–172!Iso
!
!
HNF(mod) (31.7) 160–B1 (20.8) 170–Iso
HNF(mod)–167!B1–172!Iso
!
!
HNF(mod) (33.6) 159–B1 (21.8) 170–Iso
HNF(mod)–160!B1–168!Iso
!
!
HNF(mod) (31.2) 151–B1 (20.4) 165–Iso
[a] Transition temperatures [8C] and enthalpies [kJmolꢀ1] are taken from
differential scanning calorimetry data.[12] [b] Layer spacing d1 (2p/q1, ꢀ).
[c] In-layer stripe spacing d2 (2p/q2, ꢀ), measured from small-angle X-ray
scattering (Figure 2d).
Figure 2. Data is for diester 1(9): a) Photomicrograph of the B1 texture
seen by POM with crossed polarizer and analyzer (2078C, 4 mm glass
cell, at 50ꢁ magnification). b) Log plot of the small-angle X-ray
diffraction intensity (I) vs. scattering angle q for the B1 phase.
c) Nearly optically isotropic texture observed for the HNF(mod) phase by
POM at room temperature. d) Small-angle XRD log plot for the
HNF(mod) phase showing three well-resolved peaks with sequentially
lower scattering intensity with increasing q (q1 =0.136, q2 =0.158, and
q3 =0.198 ꢀꢀ1). The very broad and weak fourth peak centered around
q=0.24 ꢀꢀ1 is interpreted as resulting from a small amount of crystal
phase, while the relatively large peak at q=0.273 ꢀꢀ1 is the second
harmonic of q1. e) Photomicrographs of the crystal phase of 1(9) with
analyzer and polarizer uncrossed, showing weak circular birefringence
in heterochiral domains (room temperature).
Elucidation of the detailed structure of these higher temper-
ature phases has proven difficult, and is currently on-going.
Strikingly, for the HNF(mod), while the layer spacing increases
with increasing molecular length as expected, the in-layer
modulation is essentially the same for all homologues.
The behavior of diester 1(9) is illustrative (Figure 2).[12]
On cooling from the isotropic phase a clear B1 texture can be
seen by POM (Figure 2a). For this phase small-angle XRD
(rotating anode) from “powder” samples in glass capillaries
Angew. Chem. Int. Ed. 2013, 52, 5254 –5257
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5255

.
Angewandte
Communications
shows two scattering vectors, supporting a tentative B1 phase
assignment (Figure 2b). The major layer spacing indicated for
this B1 phase is 33 ꢀ (q = 0.193).
Further cooling into the HNF(mod) phase produces a tex-
ture almost perfectly isotropic in the plane (Figure 2c). The
layer spacing of the this phase is 46.2 ꢀ (q1 = 0.136). Such
layer expansion at the B1-HNF transition is fully consistent
with the classic HNF behavior. However, for previously
studied HNF phases only one peak is observed for the layers,
while for all of the 1(n) homologues the HNF(mod) phase
exhibits three peaks (Figure 2d). The scattering vectors (q) of
peaks 1, 2, and 3 (on increasing q) are related as q3 = SQRT
2
(q12 + q2 ), suggesting two different electron density waves in
a rectangular lattice.
In the above XRD experiments the peak half width at half
height (HWHH) for the major layer peaks is instrument
resolution limited for both the B1 and HNF(mod) phases.
However, high-resolution XRD experiments (Brookhaven
synchrotron) run on powders of diester 1(14) show an
instrument resolution-limited HWHH in the B1 phase,
consistent with the expected long-range layer ordering
(correlation length = HWHHꢀ1 ꢁ 600 nm), and a much
broader peak (correlation length ꢁ 60 nm) for the HNF(mod)
phase.[12] The relatively short-range order observed for the
HNF(mod) layers is due to the limited diameter of the HNFs,
and is consistent with high resolution XRD behavior shown
by known HNF phases.
Interestingly, while the HNF(mod) phases of 1(n) seem
indefinitely stable once established, at least for some homo-
logues a bulk crystalline conglomerate phase, difficult to
obtain as a single pure phase, is nevertheless the thermody-
namically most stable phase of the system (Figure 2e).[13]
The proposed HNF(mod) structure is confirmed by textural
analysis of images obtained by transmission electron micros-
copy (TEM, Figure 3). For 1(9), cooling from an isotropic film
on a glass substrate, followed by TEM imaging of the free
surface by oblique evaporation of a thin film of metal, clearly
shows the signature HNFs for the HNF(mod) phase (Figure 3a).
Measurements taken from several individual HNFs in such
images provide the HNF diameter and helix pitch given in
Figure 1.[14]
The classic HNF phase of the double-benzylideneaniline
NOBOW (P-9-O-PIMB: 1,3-phenylene bis[4-(4-nonyloxy-
phenyl-iminomethyl)benzoate),[1,2] shows a variety of inter-
esting textures close to a glass surface, including flat lamellar
textures, and “toric focal conics” possessing negative curva-
ture, in addition to HNFs.[15] Imaging of the HNF(mod) phase of
1(10) shows a lamellar texture close to the glass surface, but
with clear topographical modulation normal to the layer
plane (Figure 3b), with a periodicity 40 ꢀ, providing strong
support for the structure proposed in Figure 1.
Interestingly, in our hands it has proven impossible to
grow long HNFs from neat samples of homologues of 1 other
than 1(9). TEM investigation of 1(10), 1(12), and 1(15) show
a “broken HNF” texture, clearly exhibiting HNF fragments
with negative layer curvature (Figure 3c), as is often seen for
the NOBOW HNF phase. However, long HNFs were seen for
1(10) and 1(15) (the only homologues studied in this way)
when the HNF(mod) phase was grown from 8-cyanobiphenyl
Figure 3. TEM images: a) Helical nanofilaments of 1(9) grown at a free
surface with air (scale bar 400 nm). b) Lamellar texture obtained from
1(10) on a glass surface. The layers are parallel to the image plane,
and the proposed in-layer stripes are clearly visible (scale bar 200 nm).
c) Broken HNF texture obtained from neat 1(10) (scale bar 200 nm).
d) Helical nanofilaments obtained by growing the HNF(mod) phase from
solution in 8-CB (75% 8CB by weight) (scale bar 100 nm).
5256
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 5254 –5257

Angewandte
Chemie
[3] a) G. Pelzl, S. Diele, W. Weissflog, Adv. Mater. 1999, 11, 707;
(8CB) solution (Figure 3d). Growing the HNF phase of
NOBOW from 8CB produces similar textures.[8d]
b) L. Perez-Garcia, D. B. Amabilino, Chem. Soc. Rev. 2002, 31,
342; c) R. A. Reddy, C. Tschierske, J. Mater. Chem. 2006, 16, 907;
d) H. Takezoe, Y. Takanishi, Jpn. J. Appl. Phys. Part 1 2006, 45,
597; e) J. W. Goodby, I. M. Saez, S. J. Cowling, V. Gortz, M.
Draper, A. W. Hall, S. Sia, G. Cosquer, S.-E. Lee, E. P. Raynes,
Angew. Chem. 2008, 120, 2794; Angew. Chem. Int. Ed. 2008, 47,
2754; f) J. Etxebarria, M. B. Ros, J. Mater. Chem. 2008, 18, 2919.
[4] L. E. Hough, H. T. Jung, D. Krꢁerke, M. S. Heberling, M.
Nakata, C. D. Jones, D. Chen, D. R. Link, J. Zasadzinski, G.
Heppke, J. P. Rabe, W. Stocker, E. Kçrblova, D. M. Walba, M. A.
Glaser, N. A. Clark, Science 2009, 325, 456.
[5] a) D. B. Amabilino, Science 2009, 325, 402; b) I. Dierking,
Angew. Chem. 2010, 122, 30; Angew. Chem. Int. Ed. 2010, 49, 29.
[6] E. A. Matsumoto, G. P. Alexander, R. D. Kamien, Phys. Rev.
Lett. 2009, 103, 257804.
[7] D. M. Walba, L. Eshdat, E. Kçrblova, R. K. Shoemaker, Cryst.
Growth Des. 2005, 5, 2091.
[8] a) Y. Takanishi, G. J. Shin, J. C. Jung, S. W. Choi, K. Ishikawa, J.
Watanabe, H. Takezoe, P. Toledano, J. Mater. Chem. 2005, 15,
4020; b) N. A. Clark, et al., Presented at the 3rd Banana
Workshop, Tokyo, Japan, 2007; c) T. Otani, F. Araoka, K.
Ishikawa, H. Takezoe, J. Am. Chem. Soc. 2009, 131, 12368; d) C.
Zhu, D. Chen, Y. Shen, C. D. Jones, M. A. Glaser, J. E.
Maclennan, N. A. Clark, Phys. Rev. E 2010, 81, 011704; e) D.
Chen, C. Zhu, R. K. Shoemaker, E. Korblova, D. M. Walba,
M. A. Glaser, J. E. Maclennan, N. A. Clark, Langmuir 2010, 26,
15541; f) D. Chen, J. E. Maclennan, R. Shao, D. K. Yoon, H.
Wang, E. Korblova, D. M. Walba, M. A. Glaser, N. A. Clark, J.
Am. Chem. Soc. 2011, 133, 12656; g) H. Nagayama, Y. Sasaki, F.
Araoka, K. Ema, K. Ishikawa, H. Takezoe, Soft Matter 2011, 7,
8766; h) Y. Sasaki, Y. Setoguchi, H. Nagayama, H. Yao, H.
Takezoe, K. Ema, Physica E 2011, 43, 779.
As mentioned above, a signature property of the classic
HNF phase is formation of a conglomerate with large
heterochiral domains easily seen by de-crossing polarizer
and analyzer in POM experiments.[1b,c,4] Interestingly, these
are not seen in neat HNF(mod) phases of diesters 1 by POM.
The fact that the POM texture of 1(9), which forms very long
HNFs (ca. 25 mm), is extremely dark with crossed polarizer
and analyzer, suggests very little if any circular birefringence.
This could be simply due to low rotatory power for this type of
HNF. However, inability to observe large chiral domains by
POM could be due to rapid nucleation of the phase relative to
HNF growth, leading to uniformly chiral domains, which are
small on the length scale of the wavelength of visible light.
The latter interpretation gains support from the observation
that HNF(mod) grown from solution does exhibit easily
observable rotatory power in heterochiral domains.[16]
In conclusion, evidence for the formation of a new HNF
phase, possessing in-layer modulation and lacking Schiff
bases, formed by a homologous series of simple diesters is
presented. This expands the demonstrated molecular “struc-
ture space” capable of providing the unique and potentially
useful HNF phases—which represent a novel type of organic
nanoparticle system. The HNF(mod) phase is very stable over
a broad temperature range, though at least in some cases
seems to be metastable with respect to a crystal phase. The
geometry of the layer structure is similar to that of the classic
HNF phase: Spontaneous negative curvature of the layers,
limited in width to about 10 times their thickness. In addition,
a structural supramolecular modulation normal to the helix
axis is also present, as evidenced by XRD and TEM imaging
data. The periodicity of the latter modulation is similar for all
homologues, suggesting that this aspect of the HNF(mod)
structure results from a similar packing of the cores. Addi-
tional studies designed to explore potential applications of the
HNF(mod) phase of 1(n), and to extend the HNF mesogen
structure space, which currently seems quite limited, are
underway.
[9] E. Bialecka-Florjanczyk, I. Sledzinska, E. Gorecka, J. Przed-
mojski, Liq. Cryst. 2008, 35, 401.
[10] The 3,4’-biphenol core unit was first exploited by Tschierske
et al. in the preparation of the first bent-core mesogens lacking
the benzylideneaniline moiety: D. Shen, S. Diele, I. Wirt, C.
Tschierske, Chem. Commun. 1998, 2573. Synthesis of 1(n) is
described in the Supporting Information (Scheme SI-1 and
detailed experimental).
[11] a) D. A. Coleman, J. Fernsler, N. Chattham, M. Nakata, Y.
Takanishi, E. Korblova, D. R. Link, R. F. Shao, W. G. Jang, J. E.
Maclennan, O. Mondainn-Monval, C. Boyer, W. Weissflog, G.
Pelzl, L. C. Chien, J. Zasadzinski, J. Watanabe, D. M. Walba, H.
Takezoe, N. A. Clark, Science 2003, 301, 1204; b) R. Amarana-
tha Reddy, M. W. Schroder, M. Bodyagin, H. Kresse, S. Diele, G.
Pelzl, W. Weissflog, Angew. Chem. 2005, 117, 784; Angew. Chem.
Int. Ed. 2005, 44, 774.
Received: November 26, 2012
Revised: January 26, 2013
Published online: April 19, 2013
[12] POM data for 1(10)–1(15) is presented in Section 2.2 (Figure SI-
1) in the Supporting Information. DSC data for 1(9), 1(10),
1(13), and 1(14) is presented in Section 2.3 (Figure SI-2) in the
Supporting Information. XRD data for all 1(n) is presented in
Section 2.4 (Figure SI-3) in the Supporting Information.
[13] A brief discussion of the metastability of the HNF phase of
compound 1(9), and the formation of the crystal phase
illustrated in the photomicrographs in Figure 2e, is presented
in Section 2.5 of the Supporting Information.
[14] Measurements of HNF thickness and half-pitch for 1(9) and
1(15) are given in Figure SI-4.
[15] D. Chen, M.-S. Heberling, M. Nakata, L. E. Hough, J. E.
Maclennan, M. A. Glaser, E. Korblova, D. M. Walba, J. Wata-
nabe, N. A. Clark, ChemPhysChem 2012, 13, 155.
Keywords: bent-core liquid crystals ·
ferroelectric liquid crystals · helical nanofilament phase
.
[1] a) T. Akutagawa, Y. Matsunaga, K. Yasuhara, Liq. Cryst. 1994,
17, 659; b) T. Niori, T. Sekine, J. Watanabe, T. Furukawa, H.
Takezoe, J. Mater. Chem. 1996, 6, 1231; c) T. Sekine, T. Niori, M.
Sone, J. Watanabe, S.-W. Choi, Y. Takanishi, H. Takezoe, Jpn. J.
Appl. Phys. 1997, 36, 6455; d) T. Sekine, T. Niori, J. Watanabe, T.
Furukawa, S. W. Choi, H. Takezoe, J. Mater. Chem. 1997, 7, 1307.
[2] a) D. R. Link, G. Natale, R. Shao, J. E. Maclennan, N. A. Clark,
E. Korblova, D. M. Walba, Science 1997, 278, 1924; b) D. M.
Walba, E. Kçrblova, R. Shao, J. E. Maclennan, D. R. Link, M. A.
Glaser, N. A. Clark, J. Phys. Org. Chem. 2000, 13, 830; c) D. M.
Walba, Top. Stereochem. Mater. Chirality 2003, 24, 457.
[16] POM images of large conglomerate domains of 1(10) grown
from 8-cyanobiphenyl solution are given in section 2.6 (Figure
SI-5) in the Supporting Information.
Angew. Chem. Int. Ed. 2013, 52, 5254 –5257
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5257