Article of PDLC
Article of PDLC
Article of PDLC
This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.
Naila Nasir, Hyeryeon Hong, Malik Abdul Rehman, Sunil Kumar and Yongho Seo *
Polymer-dispersed liquid crystals (PDLCs) exhibiting transmittance switching are utilized for preserving
energy and protecting privacy. Here we prepared a typical PDLC mixture from a commercially available
polymer and liquid crystal with nano-beads. A wire-bar coater was used to coat the PDLC mixture on
indium-tin-oxide-coated glass, and a PDLC cell was assembled by coupling another glass in a vacuum,
instead of a polymer film. Uniform glass-based PDLCs were fabricated successfully with an area of
15 15 cm2, while an injection process with capillary action was not available for this large-scale device
fabrication. The switching behavior of the cells was characterized by ramping the AC voltage, and
a transmittance change of 70% was measured. In a typical roll-to-roll process, only flexible polymer
films have been used in lamination, in which deterioration in transparency occurs over the course of
Received 7th July 2020
Accepted 17th August 2020
time, reducing the efficiency. In this study, to improve the optical properties, PDLC switchable glazing is
fabricated directly onto glass substrates instead of onto plastic polymers. This PDLC switchable glazing,
DOI: 10.1039/d0ra05911k
exhibiting low haziness and wide-angle vision, can be fabricated at a large scale by a vacuum-coupling
rsc.li/rsc-advances process, with potential use as glass windows for energy-efficient buildings.
This journal is © The Royal Society of Chemistry 2020 RSC Adv., 2020, 10, 32225–32231 | 32225
View Article Online
source, as in the PIPS process, and the LC molecules formed mixture to improve the uniformity of the lm thickness in the
droplets within the polymer matrix. The LC droplets grew until coating process.27–29,31 Moreover, NPs speeded up the curing
the polymerization was completed, and the PDLC lm was process, as NPs play a role of a scattering center for UV due to the
solidied. The size of the LC droplets depends on the concen- refractive index mismatch between the NP and the monomer.
This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.
chosen aer many trials. The composition with the best results
typical roll-to-roll process. In this process, PDLC lm is lami-
contained LC, NOA65, and SiO2 with a weight% of 60%, 35%,
nated onto plastic sheet, which is then attached to a rigid glass
and 5%, respectively. The mixture was stirred at room temper-
forming the PDLC window. However, plastic degrades slowly
ature for 1 hour using an homogenizer (Ultra Turrax, IKA Ltd) at
with the passage of time due to atmospheric affects. The
30 000 rpm. To prepare a homogeneous mixture without
ultraviolet rays make it yellowish, shattering the physical
undissolved particles, stirring was done at a higher temperature
appearance of the glazing. Moreover, the plastic sheet placed
(130 C) as well. Aer heating, the mixture was also kept in
between the glass substrates also reduces the transmittance,
a vacuum for half an hour so that the SiO2 particles were
reducing the efficiency of the glazing. Compared to plastic, glass
completely dissolved in LC and NOA65, resulting in the
is much more durable, as it is chemically inert and able to
formation of a uniform and homogeneous PDLC mixture. The
withstand high temperature. However, it is impossible to
mixture was coated on 15 cm 15 cm ITO-coated glass (sheet
laminate PDLC lm directly onto glass substrates using a roll-
resistance: 7 U sq1, Wooyang GMS, South Korea). Before
to-roll process because the glass is not exible. Thus, we
coating, the ITO-coated glass was carefully cleaned to remove
employed a coating technique to fabricate homogeneous and
impurities and dust particles by sonication with acetone, iso-
uniform PDLC lm directly onto an indium-tin-oxide (ITO)-
propyl alcohol, and distilled water. The PDLC mixture was
coated glass substrate to enhance the transmittance of the
coated onto ITO-coated glass using a wire-bar coater, as shown
PDLC glazing. Although small-sized PDLC glass with high
in Fig. 1a. The wire-bar coater consists of a rectangular board
transmittance has been prepared by other groups using an
and a metallic bar to roll over the glass substrate. A metal wire is
injection process or capillary action, these methods are limited
wound around the bar, thus making grooves. The diameter of
to a small-sized cell and are not scalable.27–29 We demonstrated
the grooves determines the amount of PDLC passing through
a PDLC glazing with an area of 15 15 cm2 using wire-bar
the grooves. The glass substrate is xed onto the rectangular
coating and coupling in a vacuum, preventing the formation
board. The bar rolls over, fabricating a uniform PDLC lm on
of air bubbles in the PDLC device.
the glass substrate. To attach the top glass to the PDLC-coated
bottom glass under vacuum, we designed a special coupling
2. Experimental device, as shown in Fig. 1b. It consists of a rectangular plate
PDLC lm composition with four hinges at the corners. The PDLC-coated glass is placed
at the bottom of the plate, while the top glass is placed on the
In this work, commercially prepared liquid crystal, E7, was used to
hinges separating the top and bottom glass (1 cm). The hinges
fabricate the PDLC lm. E7 is extensively used in making PDLC
are connected to rotating shas, where permanent magnets are
lm as it sustains anisotropic behavior over a wide temperature
attached. This structure is placed inside a vacuum desiccator. A
range and has high birefringence. E7 is a mixture of 4-pentyl-40 -
vacuum was created in the vacuum chamber up to 0.1 torr for
cyanobiphenyl (51%), 4-heptyl-40 -cyanobiphenyl (25%), 4-octyloxy-
1 hour. By bringing some external magnets close to the internal
40 -cyanobiphenyl (16%), and 4-pentyl-40 -cyanobiphenyl (8%) (see
magnets, the rotating shas inside the chamber were rotated
ESI Fig. S1†) with ordinary and extraordinary refractive indexes of
using the magnetic force between the magnets. As a result, the
1.521 and 1.746, respectively.30 A UV-curable adhesive (NOA65,
hinges were moved away, dropping the top glass onto the
Norland Optical Adhesive) was used as a starting material for the
bottom glass. Aerwards the chamber was purged for the top
polymer with a refractive index of 1.524. NOA65 cures rapidly when
and bottom glasses to form a PDLC cell, as shown in Fig. 1c (see
exposed to UV light with a maximum absorption range of 350–
ESI Fig. S3† for the apparatus used in the fabrication of the
380 nm wavelength. The thickness of the polymer and the UV light
PDLC device). The PDLC cell was then cured for 15 minutes
intensity affect the curing speed. Beside LC and polymer, silicon
using an ultraviolet light source with a mercury lamp (1 kW
dioxide (SiO2) nanoparticles in white powder form were also
power) at 23 cm distance from the lamp, with an intensity of 95
added. The SiO2 nanoparticles, purchased from Nanogra, have
mW cm2.
a purity of 97.3 wt% and average size 16 nm, and are coated with
2 wt% silane. The particle size was conrmed by the scanning
electron microscopy (SEM) image obtained by dissolving SiO2 Dispersion of nanoparticles in the mixture
nanoparticles in ethanol (see ESI Fig. S2†) (More details regarding The SiO2 nanoparticles should be completely dissolved in the
the nanoparticles are in the ESI Table S1†). The incorporation of PDLC mixture with LC and NOA65. Otherwise, the undissolved
SiO2 nanoparticles (NPs) increases the viscosity of the PDLC particles will result in non-uniform lms causing
32226 | RSC Adv., 2020, 10, 32225–32231 This journal is © The Royal Society of Chemistry 2020
View Article Online
Fig. 1 (a) A wire-bar coater was used to coat the PDLC mixture. (b) In a vacuum chamber the vacuum assembly holds the top and bottom glasses
for vacuum coupling. (c) The structure of the PDLC cell is shown.
inhomogeneous switchable behavior, and not fullling the wire-bar #10 was used to coat the lm at a thickness of 23.0
criteria required for a commercial product. Thus, we examined mm. Aer UV curing, the lm thickness uniformity was
the mixture at different stages using SEM in order to ensure its measured using a three-dimensional (3D) coordinate
homogeneity. SEM images were obtained aer coating the measuring machine (600XP, VIEW). The thickness of the PDLC
mixture on a silicon substrate. Fig. 2a shows the SEM image of was measured at 15 points on the lm. Then, the average value
the mixture status aer it was stirred at room temperature for 1 of the thickness was obtained as 22.0 mm, which is close to the
hour. Some undissolved particles were observed in the mixture, target value with a standard deviation of 0.95 mm, thus con-
as shown in the inset. Fig. 2b shows the state when the mixture rming the uniformity.
was stirred on a hot plate at 130 C for 1 hour. It is observed that
the homogeneity of mixture was improved, with less undis-
solved particles compared to the image in Fig. 2a. Then, the
3. Results and discussions
PDLC mixture was stirred at 130 C and kept in vacuum, and the Switching behavior of PDLC
stirring-vacuum process was repeated for 2 hours. Fig. 2c shows PDLC shows a switching behavior between the opaque state and
the result, exhibiting no undissolved particles le behind. This the transparent state, as illustrated in Fig. 3. The directors
completely dissolved PDLC mixture was used for the subse- (average direction of the LC molecules) inside the LC droplets
quent PDLC cell fabrication. are randomly distributed when no eld is applied (the off state).
In this state, the effective refractive index of the LC and the
refractive index of the polymer are mismatched, scattering the
Uniformity of lm thickness incident light and rendering the PDLC a milky white in
The PDLC mixture coated onto glass substrates using a wire-bar appearance. When an external eld is applied, the directors
coater was shown to be a highly uniform lm. The standard within the LC droplets reorient along the eld, and the incident
Fig. 2 SEM images show states of the PDLC mixtures: (a) after stirring at room temperature, (b) after stirring at the hot plate, and (c) after stirring
at the hot plate and under vacuum. The insets are the magnified images showing any undissolved clusters of nanoparticles.
This journal is © The Royal Society of Chemistry 2020 RSC Adv., 2020, 10, 32225–32231 | 32227
View Article Online
Fig. 3 (a) Liquid-crystal droplets are dispersed in the polymer matrix. (b) The switching mechanism of PDLC is illustrated dependent on the
applied electric field.
light passes through it without scattering, since the ordinary poly(ethylene terephthalate) (PET) as a reference sample, as
refractive index of the LC matches the refractive index of the shown in Fig. 5a. Transmittance versus voltage curves for the
polymer, causing PDLC to have a transparent appearance. The glass-based PDLC (g-PDLC) were measured in comparison to
strength of the electric eld required for switching from opaque the PET-based PDLC (p-PDLC). These curves show the typical
to transparent state depends upon the shape, size, and dielec- behavior of general PDLCs, which are similar to the other
tric anisotropy of the LC droplets.32 samples fabricated under different conditions, in broad outline,
The switching behavior of the PDLC devices was character- as shown in ESI Fig. S4.† In detail, the on-state transmittance
ized at different voltages using an AC voltage driver at 60 Hz. for g-PDLC at 0 V was 6% greater than the reference sample,
The PDLC device appeared opaque in the off state, scattering which is interpreted as being due to the larger LC droplets in the
light, as shown in Fig. 4a. The PDLC cell was completely g-PDLC. In the vacuum-coupling process, microbubbles trap-
homogeneous over the whole area, and no hole or blister was ped in the mixture could be removed, and phase separation of
observed, which are commonly generated if the mixture is LC might be facilitated, thus forming larger droplets. Further-
injected by capillary action, or if the top layer is attached in air. more, the increased LC droplet size decreased the threshold
On the other hand, the LC domains were aligned in the direc- voltage (10 V) to turn-on. In general, the threshold voltage
tion of the applied eld in the on state, and the PDLC became depends on the droplet size, droplet shape, and anchoring
transparent, as shown in Fig. 4b. The change in transmittance condition of the LC. If the size of the LC is larger, a lower
DT ¼ (Ton Toff) between the off and on states was estimated as threshold voltage is required, because more LC molecules
70% using a laser source, a photodiode, and an oscilloscope. inside the droplet are less inuenced by the anchoring force at
the interface between the LC droplet and the polymer.
Conversely, the small size of the LC droplet increases the
Transmittance depending on incident angle threshold voltage.7,33,34 The enhanced transmittance was almost
The optical properties of the PDLC cells fabricated on ITO glass saturated with an applied voltage of 60 V, which was similar to
were compared with a commercial PDLC lm made of the reference sample. As a result, while DT of g-PDLC was
Fig. 4 (a) The PDLC device was opaque in the off state. (b) The PDLC device became transparent in the on state.
32228 | RSC Adv., 2020, 10, 32225–32231 This journal is © The Royal Society of Chemistry 2020
View Article Online
Fig. 5 (a) Transmittance versus voltage curves for g-PDLC and p-PDLC. (b) Ton versus angle curves for g-PDLC and p-PDLC. Transmittance in
the on state was measured at different angles q from 0 to 51 . (c) The schematic shows the tilt-angle-dependent measurement setup.
similar to that of p-PDLC, the whole transmittance of g-PDLC p-PDLC on state is attributed to the lower transparency with
was higher due to the high transparency of the glass. milky appearance of the polymer compared with the glass.
An angle-dependent transmittance measurement was per-
formed to test the transparency in the tilt-angle view, as shown
in Fig. 5c. The incident angle (q) between the normal direction Response time measurement
of the PDLC plane and the laser beam was varied while a driving A response time measurement was performed as shown in
voltage of 60 V was applied to the PDLC. The transparency of g- Fig. 7. The measurement was done using a laser source and
PDLC was compared with p-PDLC by rotating the PDLC, as a photodiode with a pinhole. Fig. 7a and b show, respectively,
shown in Fig. 5b. While the Ton values for the samples were the transmission response as a function of time when the PDLC
similar at normal incident (q ¼ 0) beam, a rapid decrease in Ton device was turned on and turned off. The data show the typical
was observed in the p-PDLC. From the tilt-angle results, it was behavior for a short turn-on time (0.3 ms) and a long turn-off
conrmed that g-PDLC is superior, with a wide-angle view, time (0.1 s). During turn-on the applied electric eld aligns the
compared to p-PDLC. This effect can be explained by Fresnel liquid crystals more rapidly because energy is supplied from the
equations depending on the incident angle, q. While the external source. On the other hand, during turn-off the trans-
refractive index of PET is 1.575, the refractive index of glass is mittance approaches zero slowly, implying that the liquid
1.52, close to that of NOA65. From the Fresnel equations for crystals take longer to return to the disordered state (long turn-
unpolarized light, higher reectance is expected at the interface off time), as the turn-off process is an energy-dissipating
of a transparent medium with higher refractive index, including process via surface anchoring of LC molecules.7 These turn-on
air-to-medium and medium-to-air incident events. The rapid and turn-off times are typical values, similar to those of
decrease in Ton near 50 corresponds to the total internal
reection for medium-to-air incident light. The g-PDLC
exhibited much higher transmittance than p-PDLC near the
total reection angle, in particular.
Measurement of haziness
The haziness of the PDLC cells fabricated on ITO glass were
measured with a commercial PDLC lm made of PET as
a reference sample, as shown in Fig. 6. Haze factor versus
voltage curves for g-PDLC were plotted in comparison to the p-
PDLC at 0, 50, 60 and 70 V. We observed that the haziness
reduces as the applied voltage increases for both g-PDLC and p-
PDLC. In the off state (0 V), a maximum haze factor was
measured, as the incident light was scattered, giving the PDLC
a milky appearance. As voltage is applied, the PDLC becomes
transparent, allowing the incident light to pass through, and
reducing the haze factor. The off-state haze factor is 90% for
both the g-PDLC and p-PDLC, while the on-state haze factor for
g-PDLC is 10% lower than that for p-PDLC, conrming that the
g-PDLC fabricated via vacuum glass coupling is superior to p- Fig. 6 Haze factor versus voltage curve for glass-based PDLC and
PDLC in terms of transparency. The high haze factor for the PET-based PDLC.
This journal is © The Royal Society of Chemistry 2020 RSC Adv., 2020, 10, 32225–32231 | 32229
View Article Online
Fig. 7 Time evolution transmittances were recorded in (a) the turn-on process and (b) the turn-off process to estimate the response times.
conventional PDLCs. It can be seen that the vacuum-coupling Korea (NRF) grant funded by the Korea government (MSIT)
process does not inuence the switching speed of the PDLC. (NRF-2020R1A6A1A03043435).
4. Conclusion References
We demonstrated a facile and cost-effective method, so-called 1 P.-G. De Gennes and J. Prost, The physics of liquid crystals,
“vacuum glass coupling”, to fabricate a glass-based PDLC Oxford university press1993.
device, as an alternative method for the conventional roll-to-roll 2 D.-K. Yang and S.-T. Wu, Fundamentals of liquid crystal
process that is available only for exible substrates. This fabri- devices, John Wiley & Sons2014.
cation method can be applied widely for general solid 3 A. Jákli and A. Saupe, One-and two-dimensional uids:
substrates without limitation of size. This g-PDLC device, properties of smectic, lamellar and columnar liquid crystals,
fabricated using this simple method, exhibited a high trans- CRC Press, 2006.
mittance change with a low driving voltage of 60 V. Further- 4 P. d. Gennes, J. Prost and R. Pelcovits, The physics of liquid
more, the high transmittance was sustained at a wide range of crystals, Oxford University, Press1995.
tilt angles. This glass-based smart window can be utilized not 5 J. W. Doane, A. Golemme, J. L. West, J. B. Whitehead and
only for purposes of indoor privacy, but also for energy saving B. G. Wu, Polymer Dispersed Liquid-Crystals for Display
purposes under direct sunlight, as the glass is durable for UV Application, Mol. Cryst. Liq. Cryst., 1988, 165, 511–532.
irradiation. Thus, this switchable glazing has signicant 6 S. Kumar, H. Hong, W. Choi, I. Akhtar, M. A. Rehman and
potential for numerous commercial and industrial Y. Seo, Acrylate-assisted fractal nanostructured polymer
applications. dispersed liquid crystal droplet based vibrant colored
smart-windows, RSC Adv., 2019, 9(22), 12645–12655.
7 Y. Kim, K. Kim, K. B. Kim, J. Y. Park, N. Lee and Y. Seo,
Author contributions Flexible polymer dispersed liquid crystal lm with
Research design, N. N. and Y. S.; Experiments and data graphene transparent electrodes, Curr. Appl. Phys., 2016,
analysis, N. N., H. H., S. K., M. A. R., and Y. S; Paper writing, S. 16(3), 409–414.
K., and Y. S. 8 Y. Kim, D. Jung, S. Jeong, K. Kim, W. Choi and Y. Seo, Optical
properties and optimized conditions for polymer dispersed
liquid crystal containing UV curable polymer and nematic
Conflicts of interest liquid crystal, Curr. Appl. Phys., 2015, 15(3), 292–297.
9 M. Mucha, Polymer as an important component of blends
The authors declare no conicts of interest.
and composites with liquid crystals, Prog. Polym. Sci., 2003,
28(5), 837–873.
Acknowledgements 10 Y. J. Jeon, Y. Bingzhu, J. T. Rhee, D. L. Cheung and M. Jamil,
Application and New Developments in Polymer-Dispersed
This research was supported by the World Class 300 Project Liquid Crystal Simulation Studies, Macromol. Theory
(S2561932) of the SMBA, the Korea Institute of Energy Tech- Simul., 2007, 16(7), 643–659.
nology Evaluation and Planning (KETEP) and the Ministry of 11 Y. Liu and X. Sun, Holographic polymer-dispersed liquid
Trade, Industry & Energy of the Republic of Korea crystals: materials, formation, and applications, Adv.
(20184030202260), and the National Research Foundation of Optoelectron., 2008, (2008), 684349.
32230 | RSC Adv., 2020, 10, 32225–32231 This journal is © The Royal Society of Chemistry 2020
View Article Online
12 C. M. Lampert, Progress in switching windows, Solar and 24 A. Golemme, S. Zumer, J. Doane and M. Neubert, Deuterium
Switching Materials, International Society for Optics and NMR of polymer dispersed liquid crystals, Phys. Rev. A, 1988,
Photonics, 2001, pp. 95–103. 37(2), 559.
13 C. M. Lampert, Large-area smart glass and integrated 25 E. R. Soule and A. D. Rey, Modelling complex liquid crystal
This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.
photovoltaics, Sol. Energy Mater. Sol. Cell., 2003, 76(4), mixtures: from polymer dispersed mesophase to nematic
489–499. nanocolloids, Mol. Simul., 2012, 38(8–9), 735–750.
14 G. P. Crawford, Flexible at panel display technology, Wiley 26 I. Dierking, A. Rey, L.-C. Chien, U. Maschke, T. Bunning,
Online Library, 2005. S. Morris, D.-K. Yang, H. Furue, E. Soto-Bustamante and
15 T. J. White, L. V. Natarajan, T. J. Bunning and C. Allan P. Rudquist, Polymer-modied liquid crystals, Royal Society
Guymon, Contribution of monomer functionality and of Chemistry, 2019.
Open Access Article. Published on 01 September 2020. Downloaded on 1/19/2021 8:09:17 AM.
additives to polymerization kinetics and liquid crystal 27 J. Heo, J.-W. Huh and T.-H. Yoon, Fast-switching initially-
phase separation in acrylate-based polymer-dispersed transparent liquid crystal light shutter with crossed
liquid crystals (PDLCs), Liq. Cryst., 2007, 34(12), 1377–1385. patterned electrodes, AIP Adv., 2015, 5(4), 047118.
16 A. Elqidrea, R. Meziane and U. Maschke, Kinetics and 28 V. N. John, S. N. Varanakkottu and S. Varghese, Flexible,
molecular weight characterization of poly (2-ethylhexyl ferroelectric nanoparticle doped polymer dispersed liquid
acrylate) and liquid crystal (5CB) composites, Mol. Cryst. crystal devices for lower switching voltage and nanoenergy
Liq. Cryst., 2011, 547(1), 128/[1818]–134/[1824]. generation, Opt. Mater., 2018, 80, 233–240.
17 A. Seeboth, J. Schneider and A. Patzak, Materials for 29 A. Kumari and A. Sinha, Role of BaTiO3 nanoparticles on
intelligent sun protecting glazing, Sol. Energy Mater. Sol. electro-optic performance of epoxy-resin-based PDLC
Cell., 2000, 60(3), 263–277. devices, Liq. Cryst., 2020, 1–12.
18 H. Watanabe, Intelligent window using a hydrogel layer for 30 P. Oswald and P. Pieranski, Nematic and cholesteric liquid
energy efficiency, Sol. Energy Mater. Sol. Cell., 1998, 54(1– crystals: concepts and physical properties illustrated by
4), 203–211. experiments, CRC press, 2005.
19 G.-S. Kim and S.-H. Hyun, Synthesis of window glazing 31 X. Yan, Y. Zhou, W. Liu, S. Liu, X. Hu, W. Zhao, G. Zhou and
coated with silica aerogel lms via ambient drying, J. Non- D. Yuan, Effects of silver nanoparticle doping on the electro-
Cryst. Solids, 2003, 320(1–3), 125–132. optical properties of polymer stabilized liquid crystal
20 A. Matsuyama, Phase separations in mixtures of a liquid devices, Liq. Cryst., 2019, 1–8.
crystal and a nanocolloidal particle, J. Chem. Phys., 2009, 32 J. Kelly, W. Wu and P. Palffy-muhoray, Wavelength
131(20), 204904. dependence of scattering in PDLC lms: droplet size
21 S. A. Khadem and A. D. Rey, Theoretical Platform for Liquid- effects, Molecular Crystals Liquid Crystals Science
Crystalline Self-Assembly of Collagen-Based Biomaterials, Technology Section A, Mol. Cryst. Liq. Cryst., 1992, 223(1),
Front. Phys., 2019, 7, 88. 251–261.
22 J. L. West, Phase separation of liquid crystals in polymers, 33 W. B. Li, Y. B. Cao, H. Cao, M. Kashima, L. J. Kong and
Mol. Cryst. Liq. Cryst., 1988, 157(1), 427–441. H. Yang, Effects of the structures of polymerizable
23 N. A. Vaz, G. W. Smith and G. P. Montgomery Jr, A light monomers on the electro-optical properties of UV cured
control lm composed of liquid crystal droplets dispersed polymer dispersed liquid crystal lms, J. Polym. Sci., Part B:
in an epoxy matrix, Mol. Cryst. Liq. Cryst., 1987, 146(1), 17– Polym. Phys., 2008, 46(13), 1369–1375.
34. 34 S. X. Cheng, R. K. Bai, Y. F. Zou and C. Y. Pan, Electro-optical
properties of polymer dispersed liquid crystal materials, J.
Appl. Phys., 1996, 80(4), 1991–1995.
This journal is © The Royal Society of Chemistry 2020 RSC Adv., 2020, 10, 32225–32231 | 32231