Negative dispersion of birefringence of smectic ... - OSA Publishing

Negative dispersion of birefringence of smectic liquid crystal-polymer composite: dependence on the constituent molecules and temperature Seungbin Yang, Hyojin Lee, and Ji-Hoon Lee* Division of Electronics Engineering, Chonbuk National University, Jeonju, Jeonbuk 561-756, South Korea * [email protected]

Abstract: We investigated the dependence of the negative dispersion of birefringence of smectic liquid crystal-polymer composites on the constituent molecules and temperature. The dispersion of birefringence was significantly varied from positive dispersion to negative dispersion by the change of the relative fraction of the constituent monomers. For the temperature dependence of the dispersion, a composite with more fraction of monomers located at the inter-layer space showed a wider temperature range of the negative dispersion of birefringence. ©2015 Optical Society of America OCIS codes: (160.1190) Anisotropic optical materials; (160.3710) Liquid crystals; (310.6860) Thin films, optical properites.

References and links 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.

S. Pancharatnam, “Achromatic combinations of birefringent plates. Part I. An achromatic circular polarizer,” Proc. Ind. Acad. Sci. A 41, 130 (1955). D. Clarke, “Achromatic halfwave plates and linear polarization rotators,” Opt. Acta (Lond.) 14(4), 343–350 (1967). P. Hariharan, “Achromatic and apochromatic halfwave and quarterwave retarders,” Opt. Eng. 35(11), 3335 (1996). Y.-C. Yang and D.-K. Yang, “Achromatic reduction of off-axis light leakage in LCDs by self-compensated phase retardation (SPR) film,” Proc. SID Digest 1955 (2008). S.-W. Oh, B. Wok Park, J.-H. Lee, and T.-H. Yoon, “Achromatic optical compensation using dispersion of uniaxial films for elimination of off-axis light leakage in a liquid crystal cell,” Appl. Opt. 52(32), 7785–7790 (2013). S. Shen, J. She, and T. Tao, “Optimal design of achromatic true zero-order waveplates using twisted nematic liquid crystal,” J. Opt. Soc. Am. A 22(5), 961–965 (2005). R. K. Komanduri, K. F. Lawler, and M. J. Escuti, “Multi-twist retarders: broadband retardation control using self-aligning reactive liquid crystal layers,” Opt. Express 21(1), 404–420 (2013). A. Uchiyama and T. Yatabe, “Control of wavelength dispersion of birefringence for oriented copolycarbonate films containing positive and negative birefreingent units,” Jpn. J. Appl. Phys. 42(11), 6941–6945 (2003). A. Uchiyama, Y. Ono, Y. Ikeda, H. Shuto, and K. Yahata, “Copolycarbonate optical films developed using birefringence dispersion control,” Polym. J. 44(10), 995–1008 (2012). K. Kuboyama, T. Kuroda, and T. Ougizawa, “Control of wavelength dispersion of birefringence by miscible polymer blends,” Macromol. Symp. 249(1), 641–646 (2007). O. Parri, G. Smith, R. Harding, H.-J. Yoon, I. Gardiner, J. Sargent, and K. Skjonnemand, “Patterned retarder films using reactive mesogen technology,” Proc. SPIE 7956, 79560W (2011). H. Lee and J.-H. Lee, “Negative dispersion of birefringence in two-dimensionally self-organized smectic liquid crystal and monomer thin film,” Opt. Lett. 39(17), 5146–5149 (2014). H. Lee, S. Yang, and J.-H. Lee, “Electro-optical properties of smectic liquid crystal-polymer composite with negative dispersion of birefringence,” Curr. Appl. Phys. submitted. C. A. Guymon, E. N. Hoggan, N. A. Clark, T. P. Rieker, D. M. Walba, and C. N. Bowman, “Effects of monomer structure on their organization and polymerization in a smectic liquid crystal,” Science 275(5296), 57–59 (1997). J.-H. Lee and T.-K. Lim, “Inducing an antiferroelectric phase by segregating the layers of an intrinsically ferroelectric phase-only liquid crystal with linear-shaped molecules,” J. Appl. Phys. 98(9), 094110 (2005). J.-H. Lee, T.-K. Lim, Y.-W. Kwon, and J.-I. Jin, “Memory effects in polymer stabilized ferroelectric liquid crystals, and their dependence on the morphology of the constituent molecules,” J. Appl. Phys. 97(8), 084907 (2005). T. Takahashi, T. Umeda, H. Furue, and S. Kobayashi, “Modeling and computer simulation of the electrooptic response of polymer-stabilized ferroelectric liquid crystal cells,” Jpn. J. Appl. Phys. 38(10), 5991–5995 (1999).

#230615 - $15.00 USD © 2015 OSA

Received 11 Dec 2014; revised 22 Jan 2015; accepted 23 Jan 2015; published 28 Jan 2015 9 Feb 2015 | Vol. 23, No. 3 | DOI:10.1364/OE.23.002466 | OPTICS EXPRESS 2466

1. Introduction Generally, optically anisotropic materials show a positive dispersion (PD) of birefringence Δn decreasing with a longer wavelength of light λ. PD of Δn inevitably results in different phase retardation Г = 2πΔnd/λ at different λ, where d is the thickness of the medium. There have been many efforts to fabricate an achromatic retarder with a negative dispersion (ND) of Δn increasing with λ. For instance, stacking multi-layers with orthogonal optic axes orientation [1–5] or coating multi-layers of twisted liquid crystalline polymers [6,7] were suggested. Meanwhile, a ND retarder using a single layer has been rarely reported using copolymers [8– 10] and reactive mesogen [11–13]. To obtain ND using a single layer, the dispersion of the extraordinary refractive index ne should be smaller than that of the ordinary refractive index no. Recently, we reported another type of the single layer ND retarder using a twodimensionally self-organized smectic liquid crystal (LC) and polymer composite [12–17]. The polymers absorbing a longer wavelength of light were located at the inter-layer space of the LC host and the dispersion of Δn was gradually converted from PD to ND with more concentration of the polymers [12,13]. In this paper, we investigated the dependence of ND of the smectic LC-polymer composite on the constituent molecules and temperature (T). We varied the relative fraction of the constituent monomers and measured the change of the dispersion property. It was found that the dispersion of Δn was significantly varied from PD to ND by the change of the relative fraction of the constituent monomers triallyl-1,3,5-triazine-2,4,6(1h,3h,5h)-trione (triallyl) and 1,6-hexanediol diacrylate (HDDA), even with a constant total concentration of the monomers. For the temperature dependence, the LC-polymer composite with a relative fraction of trially and HDDA at 7:3 showed the widest ND range from 21 to 71 °C. 2. Experimental procedure

Fig. 1. Molecular structure and composition of the reactive monomer mixtures.

A commercial smectic LC (FELIX018-100, Clariant) was mixed with UV-reactive monomers at a weight concentration of 66:34. The phase sequence of pure LC is Cr −8 °C SmC 69 °C SmA 83 °C N 89 °C I. The monomers were composed of triallyl-1,3,5-triazine2,4,6(1h,3h,5h)-trione (triallyl, Aldrich), 1,6-hexanediol diacrylate (HDDA, Aldrich), and photoinitiator (Irgacure651, Ciba Chem) [Fig. 1]. Triallyl was mixed with HDDA at weight ratio of 30:70, 47:53, 70:30, and 90:10. The corresponding monomer mixtures were referred as T3H7, T5H5, T7H3, and T9H1, respectively, hereinafter. The concentration of the photointiator was kept 5 wt% of the total weight of the monomers in all mixtures. The LCmonomer mixture was injected whose substrates were coated with a planar alignment polyimide (PIA-X189-KU1, JNC). The polyimide-coated substrates were baked at 200 °C for 1 h, and then rubbed with a cotton cloth. Both substrates were assembled in an antiparallel fashion and the cell gap d was 2.2 μm. The samples were then exposed to a 30 mW/cm2 UV light for 1 m. The retardation α = Δnd was measured with a commercial retardation measurement system (Axo Scan-OPMF2, Axo Metrics). The layer spacing l of the mixture was measured by a small angle x-ray scattering (SAXS) experiment (D8-Discover, Bruker). For the SAXS measurement, the LC-monomer mixture was injected into a quartz capillary tube with a diameter of 0.5 mm (Charles Supper Company). For the IR dichroism experiment, CaF2 substrates (Edmund Optics) were used with the same fabrication conditions. The absorption intensity of polarized IR light was measured with a Fourier-transform (FT)-IR spectrometer (FTS7000, Varian).

#230615 - $15.00 USD © 2015 OSA

Received 11 Dec 2014; revised 22 Jan 2015; accepted 23 Jan 2015; published 28 Jan 2015 9 Feb 2015 | Vol. 23, No. 3 | DOI:10.1364/OE.23.002466 | OPTICS EXPRESS 2467

3. Results and discussion

Fig. 2. (a) Retardation α(λ) of the various LC-polymer composites; pure LC, LC-T3H7 (Triallyl 30 wt%, HDDA 70 wt%), LC-T5H5 (Triallyl 47 wt%, HDDA 53 wt%), LC-T7H3 (Triallyl 70 wt%, HDDA 30 wt%), and LC-T9H1 (Triallyl 90 wt%, HDDA 10 wt%). (b) α(λ) of the samples normalized to α(550 nm).

Figure 2(a) shows the dispersion property of α(λ) of the various LC-polymer composite samples at 21 °C. The pure LC and the LC-T3H7 mixtures showed PD of α(λ). Meanwhile, the LC-T5H5, LC-T7H3, and LC-T9H1 samples showed ND at 21 °C. Figure 2(b) shows α(λ) normalized to α(550 nm). The slope of α(λ) was gradually increased with more fraction of triallyl up to 70 wt%, i.e., ND was promoted. On the other hand, the slope of α(λ) of the 90 wt% triallyl (T9H1) mixture was smaller than that of the 70 wt% triallyl mixture (T7H3). We found that the nematic-isotropic phase transition temperature (TNI) of the LC-T9H1 mixture was 69 °C which was lower than TNI of the LC-T7H3 mixture 76.6 °C. Thus, the triallyl molecule certainly promoted ND, but excessive fraction of triallyl suppressed the LC phase range, diminishing ND of α(λ). For this ideal achromatic dispersion, α(450)/α(550) and α(650)/α(550) should be 0.82 and 1.18, respectively. The α(450)/α(550) and α(650)/α(550) values of the LC-T7H3 sample were 0.85 and 1.08, repectively, while the corresponding values of the previously reported ND reactive mesogen were 0.91 and 1.03, respectively [11]. Thus, the LC-polymer composite which was suggested in this study showed a better dispersion property close to the ideal one. The optical transmittance (TR) of the LC-T7H3 composite sample at 450, 550, 650, and 750 nm were 89.2, 91.1, 92.3, and 92.1%. Thus, the sample showed TR comparable to the commercial retarder films ~90%. Figure 3 shows α(λ) of the samples at different temperature. Figure 3(a)-3(e) show α(λ) at various temperature, while Fig. 3(f)-3(j) show α(λ) normalized to α(550 nm). α(λ) of the pure LC [Fig. 3(a), 3(f)] and LC-T3H7 [Fig. 3(b), 3(g)] samples always showed ND through the LC phase. On the other hand, α(λ) of the LC-T5H5 [Fig. 3(c), 3(h)], LC-T7H3 [Fig. 3(d), 3(i)], and LC-T9H1 [Fig. 3(e), 3(j)] samples showed ND at 21 °C, but converted to PD at 41, 76, and 61 °C, respectively. TNI of the corresponding samples were 79, 76.6, and 69 °C, respectively. Thus, the LC-T7H3 mixture certainly showed the widest ND temperature region as well as the steepest ND property.

#230615 - $15.00 USD © 2015 OSA

Received 11 Dec 2014; revised 22 Jan 2015; accepted 23 Jan 2015; published 28 Jan 2015 9 Feb 2015 | Vol. 23, No. 3 | DOI:10.1364/OE.23.002466 | OPTICS EXPRESS 2468

Fig. 3. Retardation α(λ) of the LC-polymer composites at various temperatures. (a) pure LC, (b) LC-T3H7, (c) LC-T5H5, (d) LC-T7H3, and (e) LC-T9H1 mixtures. (f)-(j) correspond to α(λ) of the corresponding samples normalized to α(550 nm).

For the better understanding of the dependence of ND on the temperature, we deduced the derivative dα/dλ value of the various samples at λ = 550 nm [Fig. 4]. As described above, the

#230615 - $15.00 USD © 2015 OSA

Received 11 Dec 2014; revised 22 Jan 2015; accepted 23 Jan 2015; published 28 Jan 2015 9 Feb 2015 | Vol. 23, No. 3 | DOI:10.1364/OE.23.002466 | OPTICS EXPRESS 2469

pure LC and LC-T3H7 mixtures showed negative value of dα/dλ, i.e., PD, through the LC phase. dα/dλ of the pure LC was nearly similar when T < 66 °C and increased when T > 66 °C. dα/dλ of the LC-T5H5, LC-T7H3, and LC-T9H1 mixture samples were continuously decreased with increasing temperature. The change of dα/dλ was most prominent in the LCT7H3 mixture which showed the steepest ND in Fig. 3(d) and 3(i). On the other hand, the LCT3H7 and LC-T5H5 samples showed smaller change of dα/dλ than the other samples.

Pure LC LC+T3H7 LC+T5H5 LC+T7H3 LC+T9H1

0.4

-1

dα/dλ (nm )

0.6

0.2 0.0 -0.2 -0.4 20

30

40

50

60

70

80

90

Temperature (°C) Fig. 4. dα/dλ value of the various LC-polymer composites at λ = 550 nm. Lines are guide to the eyes.

Fig. 5. (a) SAXS spectra of the pure LC and the LC-polymer mixtures vs. 2θ at 21 °C. (b) Smectic layer spacing l of the mixtures vs. temperature. Lines are guide to the eyes.

To examine the structural change of the composites with increasing temperature, we also investigated the smectic layer spacing l of the pure LC and the LC-polymer mixtures using the SAXS experiment [Fig. 5]. Figure 5(a) shows the SAXS spectra of the mixtures at 21 °C. The pure LC showed a sharp resonance peak, while the LC-polymer composites showed a relatively broad spectrum. In addition, 2θ of the LC-polymer composites showing the resonance peak was smaller than the pure LC. The reduction of the resonance peak angle means that l of the LC-polymer composites was increased compared to pure LC [also see the l data at 21 °C in Fig. 5(b)]. This can be explained by the polymers located at the inter-layer space [14]. Figure 5(b) shows the temperature dependence of l of the pure LC and the LCpolymer composites. The pure LC showed a gradual increase of l with increasing temperature. This is certainly due to the reduction of the LC tilt angle in the layer. The LC-T3H7 and LCT9H1 composites also showed increase of l. On the other hand, the LC-T5H5 and LC-T7H3 composites showed relatively smaller change of l. Thus, it is confirmed that the smectic layer structure of the LC-T5H5 and LC-T7H3 mixtures remained stable at higher temperatures. The stable layer spacing of the LC-T5H5 and LC-T7H3 mixtures might be related to the higher polymerization rate between triallyl and HDDA due to the similar molecular fraction.

#230615 - $15.00 USD © 2015 OSA

Received 11 Dec 2014; revised 22 Jan 2015; accepted 23 Jan 2015; published 28 Jan 2015 9 Feb 2015 | Vol. 23, No. 3 | DOI:10.1364/OE.23.002466 | OPTICS EXPRESS 2470

In Fig. 5(b), l of the LC-polymer composites was certainly larger than pure LC. On the other hand, the difference of l between the LC-polymer composites was not significant. Nevertheless, the dispersion of Δn between these samples was quite different as shown in Fig. 3 and Fig. 4. This means that the change of l by the polymers is not directly related to the dispersion of Δn. Instead, we conjectured that the orientation of the constituent monomers might be related to the dispersion property. 90 120

Absorption (arb. units)

1.0

0.9

60

30

150

0.8 180

0.9

1.0

-1

1690 cm , Triallyl -1 1620 cm , HDDA

0 Rubbing direction =Layer normal 330

210

240

300 270

Fig. 6. FT-IR dichroism data of the constituent monomers in the LC-T5H5 composite. The absorption intensity of triallyl and HDDA was measured at 1690 and 1620 cm−1, respectively.

Figure 6 shows the FT-IR dichroism data of the triallyl and HDDA in the LC-T5H5 sample. The absorption intensity of triallyl and HDDA was measured at 1690 and 1620 cm−1, respectively. The triallyl molecule showed a stronger absorption to the layer parallel direction, while the HDDA showed no absorption anisotropy. With the SAXS data in Fig. 5, the IR dichroism data also indicates that the triallyl molecules were aligned to the layer parallel direction at the inter-layer space, whereas the HDDA molecules have no preferential orientation. Thus, we can expect that more fractions of the triallyl molecules were located at the inter-layer space in the order of T3H7, T5H5, T7H3, and T9H1. Comparing the retardation data in Fig. 3 and Fig. 4, we think that more triallyl molecules at the inter-layer space promoted the ND of Δn and the effect of trially was more prominent than HDDA. As described above, the T9H1 mixture is an exception. The ND property of the T9H1 composite was smaller than the T7H3 composite and it is due to the suppressed LC phase range. We also investigated the dynamic properties of the LC-T7H3 mixture sample at 25 °C. The switching voltage showing the maximum contrast ratio (CR) was saturated at 9 V. In the presence of 10 V square voltage, the CR was 135:1 and the rising and falling time were 275 and 281 μs, respectively. The rising and falling time was defined as the elapse time from 10% TR to 90% TR and vice versa. 4. Conclusion To summarize, we investigated the dependence of the dispersion of Δn on the constituent monomers and temperature. It was found that the ND property was promoted with more fraction of the triallyl molecules. The triallyl molecules were located at the inter-layer space parallel to the layer plane, while HDDA showed no preferential orientation. The LC-polymer composite system showed the widest temperature range of ND from 21 to 71 °C when the relative fraction between triallyl and HDDA was 7:3. The change of the dispersion slope with temperature was decreased when the fraction of triallyl was small. Acknowledgments This research was supported by the Brain Korea 21 PLUS project and the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the ministry of Science, ICT & Future Planning (NRF-2013R1A1A1058681).

#230615 - $15.00 USD © 2015 OSA

Received 11 Dec 2014; revised 22 Jan 2015; accepted 23 Jan 2015; published 28 Jan 2015 9 Feb 2015 | Vol. 23, No. 3 | DOI:10.1364/OE.23.002466 | OPTICS EXPRESS 2471