The electrical Freedericksz transition characteristics of planar aligned liquid crystal cells doped with harvested single ferroelectric domain 9 nm nanoparticles of BaTiO3 have been measured. We demonstrate for the first time that the electrical pre-history of the cells imparts significant polarity sensitivity to the Freedericksz characteristics. The presence of harvested single domain ferroelectric nanoparticles enables cells to be programmably semi-permanently polarized. This reduces or increases the Freedericksz transition threshold by 0.8 V, depending on the polarity of the applied voltage, giving a net 1.6 V Freedericksz threshold asymmetry for 8 μm thick cells filled with TL205 liquid crystal.
© 2010 OSA
There has been much interest in the physical effects of adding ferroelectric materials in nanoparticulate form to liquid crystals [1–6]. These materials have variously been reported, for example, to influence the dielectric anisotropy , moderate the LC phase transition temperatures , to increase optical diffraction or beam coupling efficiencies [3,4], and to affect the Freedericksz transition characteristics . The latter effect is of prime importance to many liquid crystal systems since the Freedericksz transition threshold characteristics play a crucial role in determining the operating voltage and power requirements for liquid crystal displays and devices. In a recent development, methods to selectively harvest single ferroelectric domain nanoparticles as small as 9 nm using strong electric field gradients have been reported . The use of single ferroelectric domain nanoparticles in liquid crystals has been shown to have a profound effect on the electrical Freedericksz transition threshold, as well as increasing the optical two-beam coupling gain in hybrid photorefractive devices .
In this paper we describe the electrical Freedericksz transition characteristics of planar aligned liquid crystal cells doped with harvested single ferroelectric domain 9 nm nanoparticles of BaTiO3. We find that the Freedericksz transition threshold can be either increased or decreased, depending intimately on the electrical pre-history of the cells. We show that the presence of harvested single domain ferroelectric nanoparticles enables cells to be programmably semi-permanently polarized. In nanoparticle doped cells of TL205 (Merck), this pre-polarization process reduces or increases the Freedericksz transition threshold by 0.8 V, depending on the polarity of the pre-polarization voltage, giving a net 1.6 V Freedericksz threshold asymmetry for 8 μm thick cells filled with TL205 liquid crystal.
Liquid crystal cells were each prepared using two Indium Tin Oxide (ITO) coated glass windows. The interior ITO coated surface of each window was spin coated at 4000 RPM for 30 seconds with a nylon multipolymer (Elvamide® 8023R, supplied by DuPont, prepared as a 0.125 weight % solution in anhydrous methanol), followed by air drying for one hour. After drying, the spin-coated surfaces were uni-axially rubbed with a low-speed nylon roller to induce a planar alignment of the liquid crystal media. The two windows were arranged such that the rubbing directions were anti-parallel with respect to each other in order to minimize any splay alignment of the liquid crystal molecules arising from any residual pre-tilt at the glass window surfaces. Separate measurements of these cells indicated that pre-tilt angles between the liquid crystal director and the glass surface were typically less than 1 degree.
Ferroelectric 9 nm ( ± 1.5 nm) nanoparticles of BaTiO3 were prepared by mechanically milling bulk material , from which single ferroelectric domain nanoparticles were harvested using field gradient methods . The harvested nanoparticles were added to the liquid crystal TL205 (Merck) at a concentration of 0.5 weight %, together with a small quantity of 8 μm ( ± 0.2 μm) glass spacer beads used to define the cell spacing. The nanoparticles and glass spacer beads were uniformly dispersed by ultrasonic mixing. Each cell was then assembled by adding a drop of the liquid crystal mixture, in the nematic phase, to one window and then placing a second window on top. The lower window of each cell during the assembly process was marked so that the order of assembly could be identified later and used for correlation assessments during the later Freedericksz transition measurements (see below). Simple spring clips were used to squeeze the windows together to expel any excess liquid crystal and to ensure the cell spacing was defined by the low concentration of dispersed 8 μm glass rods. Interferrometric and polarization methods were used to confirm the cell spacing (and uniformity) and planar alignment of the liquid crystal media, respectively.
The DC Freedericksz transition characteristics for each cell were measured using the arrangement shown in Fig. 1 . The measurements were taken by manually incrementally increasing the applied voltage in steps of 0.1 V between readings. Each cell was placed at normal incidence between two crossed calcite polarizers with the cell rubbing direction at 45 degrees with respect to the two polarization directions. A low power 594 nm laser (~2.5 mW yellow helium neon laser) was used to measure the linear transmission through the system as a function of the DC voltage applied between the two cell windows. The previously marked “lower” window during the cell fabrication process enabled us to determine whether the sign of the applied voltage with respect to this window influenced the results.
Figure 2 shows typical results obtained from TL205 cells doped with 0.5 weight % of harvested 9 nm nanoparticles of BaTiO3. There is a very surprising difference of approximately 1.6 V between the Freedericksz transition thresholds for voltages applied in one direction compared with voltages applied in the opposite direction [similar to that reported in . The nomination of “positive” and “negative” directions here is entirely arbitrary and we found no correlation between the sign of the voltage characteristics and the polarity of lower window during cell assembly. The asymmetry in the Freedericksz characteristics did not exist in identical cells filled with pure TL205 liquid crystal, where the Freedericksz thresholds were independent of the sign of the applied field. A small asymmetry did exist in similar cells fabricated with raw (not harvested) 9 nm BaTiO3 nanoparticles, but the asymmetry here was less than 0.2 V. By comparison, the use of harvested nanoparticles had a dramatic effect on the Freedericksz asymmetry.
The asymmetric Freedericksz characteristics were initially baffling, especially as no correlation existed between the Freedericksz threshold and the polarity in respect of the assembly order of the windows. However, we noted that common practice was to test new cells, before making Freedericksz measurements, by applying a DC potential to ensure that switching occurred reliably. We postulated that the voltage pre-history of the cells might be the cause of the Freedericksz asymmetry. To test this hypothesis, we measured the Freedericksz characteristics of a virgin cell, which had previously not had any voltages applied to it. Figure 3 shows the Freedericksz characteristics for this cell, measured as the voltage was increased from zero to + 10 V. Two characteristics are immediately apparent; the Freedericksz threshold is approximately + 1.3 V, exactly mid way between the + 0.5 V or + 2.1 V for the two polarities shown in Fig. 2. Higher voltages (≥ + 4 V) create a series of distortions and jumps in the Freedericksz transmission curve, indicating that the Freedericksz threshold is being increased by the presence of the higher voltages. These characteristics were a feature of all virgin cells and had no correlation with which window was made positive or negative with respect to the window assembly order. After applying the ascending voltage, the Freedericksz transmission characteristics were monitored as the voltage was decreased from + 10 V to zero. Figure 3 compares the descending voltage Freedericksz characteristics with the ascending (virgin) voltage characteristics. The descending voltage characteristics are smooth and lack any of the ascending (virgin) jumps and distortions and the descending voltage Freedericksz threshold is now higher at approximately + 2.1 V, the same magnitude as the higher Freedericksz threshold observed with other pre-tested cells (see Fig. 2).
After completing the sequential ascending and descending voltage Freedericksz threshold measurements on the virgin cell, the cell was grounded (to short out any capacitive charge) and the measurements repeated with the polarity reversed with respect to the preceding measurements. Figure 4 compares the ascending (reversed) voltage and descending (reversed) voltage Freedericksz characteristics for the same cell. The ascending reversed voltage Freedericksz threshold is now much lower (−0.5 V, the same as the lower Freedericksz threshold in Fig. 2) and higher ascending reversed voltages create distortions and discontinuities which are reminiscent of those found at higher voltages in the virgin situation. Subsequent decreasing of the reversed voltage from −10 V to zero generates a smooth Freedericksz transition curve with a greatly increased −2.1 V Freedericksz threshold, identical in magnitude to the higher threshold observed in Fig. 2. Continuing this sequence by applying a subsequent ascending positive voltage generates an identical curve to that created by the previously applied ascending negative voltage. A clear hysteresis effect is therefore omnipresent from the sign and magnitude of the voltage pre-history. The first application of an external voltage of greater than approximately 4 V to a virgin cell creates a semi-permanent electrical polarization. Thereafter, the cell follows a repeatable, programmable Freedericksz hysteresis sequence as the voltage is sequentially cycled from positive to negative polarity. Figure 5 illustrates this clearly by comparing the four voltage cycles after the cell has been pre-poled by the first application of voltage.
Repeated testing of different cells, all made to the same nominal specifications, revealed that most cells can be re-polarized by the application of greater than 4 to 5 Volts. A few cells required somewhat larger voltage (≥8 V) and one cell required a much larger voltage (≈18 V) to be re-polarized. Similarly, the induced polarization could be erased in most cases by heating the cell to just above the isotropic phase transition temperature (the clearing point), although one cell was able to retain the polarization when heated to just above the clearing temperature, providing the thermal exposure was brief (less than a few seconds).
The field induced polarization characteristics of liquid crystal cells doped with nanoparticles of harvested BaTiO3 are striking. Even though the bulk of the medium comprises nematic liquid crystal, the polarization hysteresis indicates the cell is behaving as a ferroelectric material. The presence of a low concentration of harvested BaTiO3 nanoparticles is clearly creating an internal bias field which either adds or subtracts from the externally applied field, leading to an asymmetry in the external Freedericksz transition threshold.
Recent work has demonstrated that nanoparticles of BaTiO3 as small as 9 nm can retain bulk ferroelectric properties . Harvesting with high electric field gradients (i.e. using strong electric field gradients to impart a translational force and subsequent collection to nanoparticles with strong dipole moments) ensures that the majority of collected nanoparticles have a single ferroelectric domain and therefore have the largest possible dipole moment . From the results obtained in this paper, we surmise that the initial alignment state of the nanoparticle dipole moments within the liquid crystal may be random. The Freedericksz transition threshold therefore adopts a nominal value as shown in the ascending voltage curve of Fig. 3. However, the application of an electric field above a given threshold (typically about 4 V in most of our cells) may cause the nanoparticle dipole moments to align parallel with the external electric field. Once aligned, dipole to dipole interaction between neighboring nanoparticles seems to have sufficient integrity to allow the nanoparticles to remain in alignment after the external field has been removed. The low concentration of nanoparticles makes direct coupling between adjacent nanoparticles unlikely, but the coupling may be mediated by the liquid crystal medium owing to strong intermolecular elastic coupling between the liquid crystal molecules. Similarly, the presence of the strong dipole moments from the dispersed and aligned nanoparticles may help to re-enforce the alignment integrity of the liquid crystal molecules. The aligned nanoparticles and the surrounding liquid crystal molecules then become mutually coupled systems.
After application of an external field of sufficient magnitude to ferroelectrically “pole” the cell (similar to the coercive field for bulk ferroelectrics), there exists an internal spontaneous polarization with an opposite field direction to that used to align the nanoparticles. Note that the term “pole” is used here to refer to the creation of a semi-permanent spontaneous polarization arising from the applied field induced macroscopic alignment of the nanoparticles with respect to each other (i.e. chain or ordering formation). There is no intrinsic re-poling of the individual nanoparticles, only an initial ordering followed by subsequent reorientation/reordering in response to the electric fields. The apparent Freedericksz transition threshold for the same polarity used to pole the cell is therefore increased by the magnitude of the net internal dipole moment for the whole cell. However, on reversal of the external field, the net internal dipole field adds to the external field, therefore creating a reduction of the external Freedericksz transition threshold. This condition persists until the reversed field magnitude becomes sufficiently large to re-pole the entire cell. The same mechanism accounts for why the DC field required to completely homeotropically align nanoparticle doped liquid crystal cells is usually higher than that required for pure liquid crystal cells, even though the addition of nanoparticles usually slightly reduces the Freedericksz transition threshold in virgin cells. The mechanism shown in the ascending voltage case of Fig. 3 causes the highest voltage peak transmission to shift to larger voltages.
The observations of thermal and high voltage erasures appears to rule out permanent surface charge effects from aggregation of nanoparticles at the cell windows as a possible mechanism to explain the observed Freedericksz asymmetry. We assume that fabrication variability of our simple cell construction method is the root cause of the cell to cell voltage and temperature magnitude variations for re-polarizing and depolarizing the cells, respectively.
The demonstration of electrically programmable ferroelectric poling of a nominally nematic phase liquid crystal medium has widespread potential applications. The reduction in the Freedericksz transition threshold has the potential to significantly lower the operating voltage of liquid crystal systems and displays. We have already noted that some cells have demonstrated a very large re-poling voltage, indicating that the internal dipole field has the potential to be sufficient to retain an image in a liquid crystal display in the event of power failure. The use of alternative alignment surfaces, liquid crystal media and ferroelectric nanoparticles, offers great potential for future device development and power saving.
We are very grateful to A. Glushchenko for providing the raw BaTiO3 nanoparticle material. This work was partly supported by EOARD Grant 078001 (VYR).
References and links
1. Y. Reznikov, O. Buchnev, O. Tereshchenko, V. Reshetnyak, A. Glushchenko, and J. West, ““Ferroelectric nematic suspension,” Appl. Phys. Lett. 82(12), 1917 (2003). [CrossRef]
2. F. Li, O. Buchnev, C. I. Cheon, A. Glushchenko, V. Reshetnyak, Y. Reznikov, T. J. Sluckin, and J. L. West, “Orientational coupling amplification in ferroelectric nematic colloids,” Phys. Rev. Lett. 97(14), 147801 (2006). [CrossRef] [PubMed]
3. O. Buchnev, A. Dyadyusha, M. Kaczmarek, V. Reshetnyak, and Y. Reznikov, “Enhanced two-beam coupling in colloids of ferroelectric nanoparticles in liquid crystals,” J. Opt. Soc. Am. B 24(7), 1512 (2007). [CrossRef]
4. G. Cook, A. V. Glushchenko, V. Reshetnyak, A. T. Griffith, M. A. Saleh, and D. R. Evans, “Nanoparticle doped organic-inorganic hybrid photorefractives,” Opt. Express 16(6), 4015–4022 (2008). [CrossRef] [PubMed]
5. P. Mikulko, P. Arora, A. Glushchenko, A. Lapanik, and W. Haase, “Complementary studies of BaTiO3 nanoparticles suspended in a ferroelectric liquid-crystalline mixture,” Europhys. Lett. 87(2), 27009 (2009). [CrossRef]
7. G. Cook, J. L. Barnes, R. F. Ziolo, A. Ponce, V. Yu. Reshetnyak, A. Glushchenko, S. A. Basun, P. P. Banerjee, and D. R. Evans, “Harvesting Single Domain Stressed Ferroelectric Nanoparticles for Disparate Optical and Ferroic Applications,” J. Appl. Phys. (to be published). [PubMed]
8. G. Cook, J. L. Barnes, V. Yu. Reshetnyak, A. Glushchenko, R. F. Ziolo, A. Ponce, S. A. Basun, P. P. Banerjee, and D. R. Evans, “Harvesting single ferroelectric domain nanoparticles and their use in hybrid organic-inorganic photorefractive media,” in 13th Topical Meeting on the Optics of Liquid Crystals, Erice, Italy, September 28th - October 2nd (2009)