- Email: [email protected]
Tunable microlens arrays using polymer network liquid crystal
Optics Communications 230 (2004) 267–271 www.elsevier.com/locate/optcom
Tunable microlens arrays using polymer network liquid crystal Hongwen Ren *, Yun-Hsing Fan, Sebastian Gauza, Shin-Tson Wu School of Optics/CREOL, University of Central Florida, 4000 Central Florida Blvd., Orlando, FL 32816, USA Received 22 September 2003; received in revised form 10 November 2003; accepted 13 November 2003
Abstract A tunable-focus microlens array based on polymer network liquid crystal (PNLC) is demonstrated. The PNLC was prepared using an ultraviolet (UV) light exposure through a patterned photomask. The photocurable monomer in each of the UV exposed spot forms an inhomogeneous centro-symmetrical polymer network which acts as a lens when a homogeneous electric field is applied to the cell. The focal length of the microlens arrays is tunable with the applied voltage. 2003 Elsevier B.V. All rights reserved. PACS: 61.30.)v; 42.70.Df; 78.20.Jq Keywords: Liquid crystals; Microlens; Tunable focal length
1. Introduction The electrically-tunable liquid crystal (LC) microlens array is useful for information processing, optoelectronics, integrated optics and optical communications. Various attempts, such as zone patterned structure [1], hole or hybrid-patterned electrode [2–4], and surface relief profile [5,6], have been proposed to develop the microlens array. However, most of the lens fabrication processes are rather complicated. Another challenge in microlens design is to control the focal length. The
* Corresponding author. Tel.: +407-823-4965; fax: +407-8236880. E-mail addresses: [email protected] (H. Ren), [email protected] (S.-T. Wu).
focal length is determined by the radius of the microlens aperture, LC cell gap and refractive index profile. For practical applications, the focal length has to be larger than the thickness of the glass substrate, which is about 0.5–1.1 mm. It is also highly desirable to have fast switching time during focus change. There is an urgent need to develop high performance tunable microlens arrays with simple fabrication process. Recently, we have demonstrated a large aperture LC lens using a gradient refractive index polymer network liquid crystal (PNLC) [7,8]. The lens was fabricated by exposing ultraviolet (UV) light onto LC/monomer mixture through a patterned photomask. The major advantages of such a PNLC lens are in its simplicity, low operating voltage, scalable aperture size, and fast switching speed. However, due to the small refractive index
0030-4018/$ - see front matter 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.optcom.2003.11.033
268
H. Ren et al. / Optics Communications 230 (2004) 267–271
profile, a large aperture LC lens would produce a long focal length which is inconvenient for common applications. In this paper, we demonstrate a tunable-focus microlens array using PNLC cell. The unique feature of this approach is that, we apply a uniform electric field to a homogeneous LC cell which, in turn, generates the centro-symmetric refractive index profile for the focusing effect to occur. The UV exposure through a patterned photomask gives rise to a spatially varying polymer network concentration. In the spot center, polymerization rate is faster due to the higher UV intensity so that, polymer networks are denser. This polymer-rich region exhibits a higher threshold voltage than that of the border LC-rich region. When a uniform voltage is applied to the lens arrays, the border LC molecules are reoriented first. This gradient refractive index makes the PNLC behave like a positive lens. On the other hand, if the photomask pattern is reversed the microlens will function as a negative lens.
tensity. The diameter of the hole is 30 lm and the distance from the CCD camera to the pinhole is 2 cm. To prepare the PNLC microlens array, we mixed 3 wt% of rod-like photocurable monomer BAB6 together with a small amount of photoinitiator in the Merck E48 LC host (Dn ¼ 0:231 at k ¼ 589 nm and T ¼ 20 C). The mixture was injected into a homogeneous empty cell composed of indium–tin–oxide (ITO) glass substrates. The inner surfaces of the ITO glass substrates were coated with a thin polyimide layer and buffed in antiparallel directions. The cell gap and substrate thickness is 15 lm and 1.1 mm, respectively. A chromium layer with circular aperture array deposited on a glass substrate is then placed on the top of the cell. The diameter of each pinhole is 25 lm and the pitch of the aperture arrays is 110 lm. The LC/monomer cell was exposed to UV light (40 mW/cm2 ) from the photomask side. The curing time is 45 min. After exposure, the photomask was removed form the sample cell.
2. Diffraction theory and experiments
3. Results and discussion
To make microlens arrays, our photomask consists of arrays of circular apertures. When the UV light passes through the pinholes, diffraction takes place. This diffraction effect is helpful for generating the gradient refractive index profile. Based on the diffraction theory, when a laser beam transverses a tiny circular aperture the diffracted light intensity profile can be expressed as: [9] 2 2 IðhÞ ¼ I0 ½2J1 ðxÞ=x , where I0 / ðpaÞ =k2 is the light intensity at the central point, x ¼ ð2pa=kÞ sin h, h is the diffraction angle, a is the radius of the pinhole, k is the incident wavelength and J1 is the Bessel function of the first order. The formed diffraction pattern consists of a series of rings, but most of light energy concentrates on the 0th order ring (called Airy disk) which presents a paraboliclike profile. Similarly, when an incoherent UV light passes through a pinhole, it is diffracted too. To characterize the output intensity profile, an expanded UV light (k ¼ 365 nm, Loctite Wand System) was normally incident to a pinhole and a CCD camera was used to detect the output in-
Fig. 1 shows the measured three-dimensional UV intensity profile. As expected, the intensity distribution presents a parabolic-like profile and no higher-order diffraction rings appear. If the LC/ monomer mixture is cured using such a UV light, the stronger intensity in the central region will
Fig. 1. The intensity profile recorded by a CCD camera when a UV light passing through a pinhole. Pinhole diameter is 30 lm and k ¼ 365 nm.
H. Ren et al. / Optics Communications 230 (2004) 267–271
accelerate the polymerization rate and form a denser polymer network. This is known as the polymer-rich region. At the borders, the weaker light will have a slower polymerization rate resulting in a looser polymer network (i.e. LC-rich region). Therefore, a centro-symmetric inhomogeneous polymer network is formed. If there is a need to change the spot diameter, we could either modify the photomask patterns or adjust the distance between the LC sample and the photomask. The latter is simpler as it does not need to redesign the photomask. We used a polarizing optical microscope to inspect the PNLC microlens arrays at different applied voltages. The rubbing direction of the cell was oriented at 45 with respect to the axis of the linear polarizer and the analyzer was crossed to the polarizer. Three photographs at V ¼ 0, 5 Vrms and 20 Vrms were taken, as shown in Figs. 2(a), (b) and (c), respectively. At V ¼ 0, the UV exposed area is almost homogeneous and no focusing effect occurs. When an electric field is applied to the cell and increased slowly, the UV cured spots become noticeable. In the mean time, the color change starts from the borders and gradually moves to the center. Fig. 2(b) shows such an example. This implies that, the threshold voltage at the spot center is higher than that at the borders. Such results confirm the expected gradient phase retardation. When the LC sample was rotated such that, its rubbing direction is parallel to the polarizerÕs axis, the array spots appear dark. This means the PNLC microlens is polarization sensitive. As the applied voltage is sufficiently higher than the threshold, most of the
ND
269
P
L2 BE
He-Ne
L1
LC µ-lens
CCD PC
Fig. 3. The experimental setup for characterizing PNLC microlens arrays. ND, neutral density filter; P, linear polarizer; L1 and L2 , lenses; BE, beam expander.
bulk LC directors are reoriented perpendicular to the substrates. Under such a circumstance, the refractive index profile is flattened and focusing effect is vague, as shown in Fig. 2(c). A CCD camera (SBIG Model ST-2000XM) was used to characterize the light focusing properties of the microlens arrays. The experimental setup is sketched in Fig. 3. A linearly polarized He–Ne laser beam (k ¼ 633 nm) was expanded to 5 mm diameter before illuminating the LC sample. A 10· beam expander was positioned between the CCD and the LC sample in order to clearly resolve the output intensity profiles. The recorded data was analyzed by a computer. Fig. 4 shows the images of the focusing properties (left) of the microlens array as well as the corresponding CCD intensity profiles (right) at different voltages. The sample cell was set at 2 cm in front of the beam expander. Fig. 4(a) shows the sample microlens arrays at V ¼ 0. Since there is no focusing effect, the output He–Ne laser beam is relatively uniform and its intensity reaches 4000 arbitrary units. As the voltage increases to 5 Vrms , the focusing effect manifests. Arrays of bright spots are clearly observed, as shown in the left side
Fig. 2. Photographs of the PNLC sample cell at different operating voltages: (a) V ¼ 0, (b) V ¼ 5 Vrms and (c) V ¼ 20 Vrms . Polarizers are crossed. LC used is E48 and cell gap d ¼ 15 lm.
270
H. Ren et al. / Optics Communications 230 (2004) 267–271
Fig. 4. Focusing properties and intensity profiles of the PNLC microlens arrays recorded by the CCD camera at different operating voltages. (a) V ¼ 0, (b) V ¼ 5 Vrms and (c) V ¼ 20 Vrms . The cell gap d ¼ 15 lm and k ¼ 633 nm.
4 Focal Length, cm
of Fig. 4(b). The measured intensity (at the CCD focal plane) of each microlens exceeds 12 000 arbitrary units and the average focal spot size of each microlens is 30 lm, as shown in the right side of Fig. 4(b). At V ¼ 20 Vrms , the LC directors in the polymer-rich and LC-rich regions are all reoriented toward the electric field. The curvature of the refractive index profile is gradually flattened. As a result, the focal length of the microlens arrays increases and the measured light intensity at the CCD focal plane decreases, as shown in Fig. 4(c). The voltage-dependent focal length of the microlens arrays was investigated and results are plotted in Fig. 5. At V ¼ 0, the PNLC exhibits a
3 2 1 0
0
5
10 15 Voltage, Vrms
20
25
Fig. 5. Voltage-dependent focal length of the microlens arrays. LC used is E48, d ¼ 15 m and k ¼ 633 nm.
homogeneous alignment due to the surface anchoring effect from the substrates. Thus, no focusing effect occurs. As the voltage increases, the
H. Ren et al. / Optics Communications 230 (2004) 267–271
focal length decreases first, reaches a minimum at 5 Vrms and then increases again. These results are consistent to those shown in Fig. 4. The focal length of an LC lens is determined by three parameters as f ¼ a2 =ð2ddnÞ;
ð1Þ
where a is the radius of the lens, d is the cell gap, and dn is the refractive index difference between the lens center and border. Although the pinhole radius of our photomask is 12.5 lm, the actual spot radius is about 50 lm due to diffraction effect. Plugging a ¼ 50 lm, d ¼ 15 lm, and f 2 cm to Eq. (1), we find that dn 4:2 103 . As the voltage exceeds 5 V, the focal length starts to increase. Theoretically, a sufficiently high electric field would reorient the bulk LC director perpendicular to the substrates. Under such a circumstance, dn ! 0 and f ! 1 However, in our microlens arrays the polymer networks exert the weakest stabilization on LC molecules at the border of the microlens, thus the radius of the microlens has the tendency to shrink at high voltage. In this condition, the focal length can not be lengthened, it will gradually saturate as the voltage exceeds 10 Vrms , as shown in Fig. 5. While comparing with other tunable microlens technologies, the major PNLC advantages are in the uniform cell gap and using single electrode on both substrates. Besides, the PNLC lens exhibits a reasonably fast switching speed (<20 ms), wide range tunable focal length (after having optimized the cell structure and UV curing conditions), and relatively low operating voltage. The focal length (a few centimeter) of the PNLC microlens arrays is in a convenient range for optical fiber switches and incoherent correlator applications. Long-term stability of the microlens arrays is an important aspect for practical applications. We have not studied the lifetime of our microlens thoroughly. However, we expect that high voltage operation and completeness of polymerization would affect the device long-term stability. Especially, if the applied voltage exceeds the saturation voltage, then the polymer networks may be
271
unwound to certain extent and become irreversible. If the monomers are not completely cured, then the residual monomers would interact with ambient UV light resulting in polymer concentration change. The electro-optic effect of the PNLC cell would alter with time.
4. Conclusion We have demonstrated an electrically tunable microlens array using polymer stabilized nematic liquid crystal. By exposing the LC/monomer cell with UV light through a patterned photomask, a centro-symmetrical polymer network is formed. With the application of a homogeneous electric field across the LC cell, a gradient refractive index profile is generated at the exposed spots. Using this fabrication method, the LC microlens or lens arrays with tunable focal length can be easily fabricated.
Acknowledgements The authors are indebted to Mr. Yi-Pai Huang for designing the photomask. This work is supported by DARPA under Contract No. DAAD1902-1-0208.
References [1] J.S. Patel, K. Rastani, Opt. Lett. 16 (1991) 532. [2] T. Nose, S. Masuda, S. Sato, J. Li, L.C. Chien, P.J. Bos, Opt. Lett. 22 (1997) 351. [3] S. Masuda, T. Nose, S. Sato, Appl. Opt. 37 (1998) 2067. [4] W. Klaus, M. Ide, Y. Hayano, S. Morokawa, Y. Arimoto, Proc. SPIE 3635 (1999) 66. [5] L.G. Commander, S.E. Day, D.R. Selviah, Opt. Commun. 177 (2000) 157. [6] Y. Choi, J.H. Park, J.H. Kim, S.D. Lee, Opt. Mater. 21 (2002) 643. [7] H. Ren, S.T. Wu, Appl. Phys. Lett. 82 (2003) 22. [8] H. Ren, Y.H. Fan, S.T. Wu, Appl. Phys. Lett. 83 (2003) 1515. [9] E. Hecht, Optics, second ed., Addison-Wesley, Reading, MA, 1987 (Chapter 10).
Tunable microlens arrays using polymer network liquid crystal Hongwen Ren *, Yun-Hsing Fan, Sebastian Gauza, Shin-Tson Wu School of Optics/CREOL, University of Central Florida, 4000 Central Florida Blvd., Orlando, FL 32816, USA Received 22 September 2003; received in revised form 10 November 2003; accepted 13 November 2003
Abstract A tunable-focus microlens array based on polymer network liquid crystal (PNLC) is demonstrated. The PNLC was prepared using an ultraviolet (UV) light exposure through a patterned photomask. The photocurable monomer in each of the UV exposed spot forms an inhomogeneous centro-symmetrical polymer network which acts as a lens when a homogeneous electric field is applied to the cell. The focal length of the microlens arrays is tunable with the applied voltage. 2003 Elsevier B.V. All rights reserved. PACS: 61.30.)v; 42.70.Df; 78.20.Jq Keywords: Liquid crystals; Microlens; Tunable focal length
1. Introduction The electrically-tunable liquid crystal (LC) microlens array is useful for information processing, optoelectronics, integrated optics and optical communications. Various attempts, such as zone patterned structure [1], hole or hybrid-patterned electrode [2–4], and surface relief profile [5,6], have been proposed to develop the microlens array. However, most of the lens fabrication processes are rather complicated. Another challenge in microlens design is to control the focal length. The
* Corresponding author. Tel.: +407-823-4965; fax: +407-8236880. E-mail addresses: [email protected] (H. Ren), [email protected] (S.-T. Wu).
focal length is determined by the radius of the microlens aperture, LC cell gap and refractive index profile. For practical applications, the focal length has to be larger than the thickness of the glass substrate, which is about 0.5–1.1 mm. It is also highly desirable to have fast switching time during focus change. There is an urgent need to develop high performance tunable microlens arrays with simple fabrication process. Recently, we have demonstrated a large aperture LC lens using a gradient refractive index polymer network liquid crystal (PNLC) [7,8]. The lens was fabricated by exposing ultraviolet (UV) light onto LC/monomer mixture through a patterned photomask. The major advantages of such a PNLC lens are in its simplicity, low operating voltage, scalable aperture size, and fast switching speed. However, due to the small refractive index
0030-4018/$ - see front matter 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.optcom.2003.11.033
268
H. Ren et al. / Optics Communications 230 (2004) 267–271
profile, a large aperture LC lens would produce a long focal length which is inconvenient for common applications. In this paper, we demonstrate a tunable-focus microlens array using PNLC cell. The unique feature of this approach is that, we apply a uniform electric field to a homogeneous LC cell which, in turn, generates the centro-symmetric refractive index profile for the focusing effect to occur. The UV exposure through a patterned photomask gives rise to a spatially varying polymer network concentration. In the spot center, polymerization rate is faster due to the higher UV intensity so that, polymer networks are denser. This polymer-rich region exhibits a higher threshold voltage than that of the border LC-rich region. When a uniform voltage is applied to the lens arrays, the border LC molecules are reoriented first. This gradient refractive index makes the PNLC behave like a positive lens. On the other hand, if the photomask pattern is reversed the microlens will function as a negative lens.
tensity. The diameter of the hole is 30 lm and the distance from the CCD camera to the pinhole is 2 cm. To prepare the PNLC microlens array, we mixed 3 wt% of rod-like photocurable monomer BAB6 together with a small amount of photoinitiator in the Merck E48 LC host (Dn ¼ 0:231 at k ¼ 589 nm and T ¼ 20 C). The mixture was injected into a homogeneous empty cell composed of indium–tin–oxide (ITO) glass substrates. The inner surfaces of the ITO glass substrates were coated with a thin polyimide layer and buffed in antiparallel directions. The cell gap and substrate thickness is 15 lm and 1.1 mm, respectively. A chromium layer with circular aperture array deposited on a glass substrate is then placed on the top of the cell. The diameter of each pinhole is 25 lm and the pitch of the aperture arrays is 110 lm. The LC/monomer cell was exposed to UV light (40 mW/cm2 ) from the photomask side. The curing time is 45 min. After exposure, the photomask was removed form the sample cell.
2. Diffraction theory and experiments
3. Results and discussion
To make microlens arrays, our photomask consists of arrays of circular apertures. When the UV light passes through the pinholes, diffraction takes place. This diffraction effect is helpful for generating the gradient refractive index profile. Based on the diffraction theory, when a laser beam transverses a tiny circular aperture the diffracted light intensity profile can be expressed as: [9] 2 2 IðhÞ ¼ I0 ½2J1 ðxÞ=x , where I0 / ðpaÞ =k2 is the light intensity at the central point, x ¼ ð2pa=kÞ sin h, h is the diffraction angle, a is the radius of the pinhole, k is the incident wavelength and J1 is the Bessel function of the first order. The formed diffraction pattern consists of a series of rings, but most of light energy concentrates on the 0th order ring (called Airy disk) which presents a paraboliclike profile. Similarly, when an incoherent UV light passes through a pinhole, it is diffracted too. To characterize the output intensity profile, an expanded UV light (k ¼ 365 nm, Loctite Wand System) was normally incident to a pinhole and a CCD camera was used to detect the output in-
Fig. 1 shows the measured three-dimensional UV intensity profile. As expected, the intensity distribution presents a parabolic-like profile and no higher-order diffraction rings appear. If the LC/ monomer mixture is cured using such a UV light, the stronger intensity in the central region will
Fig. 1. The intensity profile recorded by a CCD camera when a UV light passing through a pinhole. Pinhole diameter is 30 lm and k ¼ 365 nm.
H. Ren et al. / Optics Communications 230 (2004) 267–271
accelerate the polymerization rate and form a denser polymer network. This is known as the polymer-rich region. At the borders, the weaker light will have a slower polymerization rate resulting in a looser polymer network (i.e. LC-rich region). Therefore, a centro-symmetric inhomogeneous polymer network is formed. If there is a need to change the spot diameter, we could either modify the photomask patterns or adjust the distance between the LC sample and the photomask. The latter is simpler as it does not need to redesign the photomask. We used a polarizing optical microscope to inspect the PNLC microlens arrays at different applied voltages. The rubbing direction of the cell was oriented at 45 with respect to the axis of the linear polarizer and the analyzer was crossed to the polarizer. Three photographs at V ¼ 0, 5 Vrms and 20 Vrms were taken, as shown in Figs. 2(a), (b) and (c), respectively. At V ¼ 0, the UV exposed area is almost homogeneous and no focusing effect occurs. When an electric field is applied to the cell and increased slowly, the UV cured spots become noticeable. In the mean time, the color change starts from the borders and gradually moves to the center. Fig. 2(b) shows such an example. This implies that, the threshold voltage at the spot center is higher than that at the borders. Such results confirm the expected gradient phase retardation. When the LC sample was rotated such that, its rubbing direction is parallel to the polarizerÕs axis, the array spots appear dark. This means the PNLC microlens is polarization sensitive. As the applied voltage is sufficiently higher than the threshold, most of the
ND
269
P
L2 BE
He-Ne
L1
LC µ-lens
CCD PC
Fig. 3. The experimental setup for characterizing PNLC microlens arrays. ND, neutral density filter; P, linear polarizer; L1 and L2 , lenses; BE, beam expander.
bulk LC directors are reoriented perpendicular to the substrates. Under such a circumstance, the refractive index profile is flattened and focusing effect is vague, as shown in Fig. 2(c). A CCD camera (SBIG Model ST-2000XM) was used to characterize the light focusing properties of the microlens arrays. The experimental setup is sketched in Fig. 3. A linearly polarized He–Ne laser beam (k ¼ 633 nm) was expanded to 5 mm diameter before illuminating the LC sample. A 10· beam expander was positioned between the CCD and the LC sample in order to clearly resolve the output intensity profiles. The recorded data was analyzed by a computer. Fig. 4 shows the images of the focusing properties (left) of the microlens array as well as the corresponding CCD intensity profiles (right) at different voltages. The sample cell was set at 2 cm in front of the beam expander. Fig. 4(a) shows the sample microlens arrays at V ¼ 0. Since there is no focusing effect, the output He–Ne laser beam is relatively uniform and its intensity reaches 4000 arbitrary units. As the voltage increases to 5 Vrms , the focusing effect manifests. Arrays of bright spots are clearly observed, as shown in the left side
Fig. 2. Photographs of the PNLC sample cell at different operating voltages: (a) V ¼ 0, (b) V ¼ 5 Vrms and (c) V ¼ 20 Vrms . Polarizers are crossed. LC used is E48 and cell gap d ¼ 15 lm.
270
H. Ren et al. / Optics Communications 230 (2004) 267–271
Fig. 4. Focusing properties and intensity profiles of the PNLC microlens arrays recorded by the CCD camera at different operating voltages. (a) V ¼ 0, (b) V ¼ 5 Vrms and (c) V ¼ 20 Vrms . The cell gap d ¼ 15 lm and k ¼ 633 nm.
4 Focal Length, cm
of Fig. 4(b). The measured intensity (at the CCD focal plane) of each microlens exceeds 12 000 arbitrary units and the average focal spot size of each microlens is 30 lm, as shown in the right side of Fig. 4(b). At V ¼ 20 Vrms , the LC directors in the polymer-rich and LC-rich regions are all reoriented toward the electric field. The curvature of the refractive index profile is gradually flattened. As a result, the focal length of the microlens arrays increases and the measured light intensity at the CCD focal plane decreases, as shown in Fig. 4(c). The voltage-dependent focal length of the microlens arrays was investigated and results are plotted in Fig. 5. At V ¼ 0, the PNLC exhibits a
3 2 1 0
0
5
10 15 Voltage, Vrms
20
25
Fig. 5. Voltage-dependent focal length of the microlens arrays. LC used is E48, d ¼ 15 m and k ¼ 633 nm.
homogeneous alignment due to the surface anchoring effect from the substrates. Thus, no focusing effect occurs. As the voltage increases, the
H. Ren et al. / Optics Communications 230 (2004) 267–271
focal length decreases first, reaches a minimum at 5 Vrms and then increases again. These results are consistent to those shown in Fig. 4. The focal length of an LC lens is determined by three parameters as f ¼ a2 =ð2ddnÞ;
ð1Þ
where a is the radius of the lens, d is the cell gap, and dn is the refractive index difference between the lens center and border. Although the pinhole radius of our photomask is 12.5 lm, the actual spot radius is about 50 lm due to diffraction effect. Plugging a ¼ 50 lm, d ¼ 15 lm, and f 2 cm to Eq. (1), we find that dn 4:2 103 . As the voltage exceeds 5 V, the focal length starts to increase. Theoretically, a sufficiently high electric field would reorient the bulk LC director perpendicular to the substrates. Under such a circumstance, dn ! 0 and f ! 1 However, in our microlens arrays the polymer networks exert the weakest stabilization on LC molecules at the border of the microlens, thus the radius of the microlens has the tendency to shrink at high voltage. In this condition, the focal length can not be lengthened, it will gradually saturate as the voltage exceeds 10 Vrms , as shown in Fig. 5. While comparing with other tunable microlens technologies, the major PNLC advantages are in the uniform cell gap and using single electrode on both substrates. Besides, the PNLC lens exhibits a reasonably fast switching speed (<20 ms), wide range tunable focal length (after having optimized the cell structure and UV curing conditions), and relatively low operating voltage. The focal length (a few centimeter) of the PNLC microlens arrays is in a convenient range for optical fiber switches and incoherent correlator applications. Long-term stability of the microlens arrays is an important aspect for practical applications. We have not studied the lifetime of our microlens thoroughly. However, we expect that high voltage operation and completeness of polymerization would affect the device long-term stability. Especially, if the applied voltage exceeds the saturation voltage, then the polymer networks may be
271
unwound to certain extent and become irreversible. If the monomers are not completely cured, then the residual monomers would interact with ambient UV light resulting in polymer concentration change. The electro-optic effect of the PNLC cell would alter with time.
4. Conclusion We have demonstrated an electrically tunable microlens array using polymer stabilized nematic liquid crystal. By exposing the LC/monomer cell with UV light through a patterned photomask, a centro-symmetrical polymer network is formed. With the application of a homogeneous electric field across the LC cell, a gradient refractive index profile is generated at the exposed spots. Using this fabrication method, the LC microlens or lens arrays with tunable focal length can be easily fabricated.
Acknowledgements The authors are indebted to Mr. Yi-Pai Huang for designing the photomask. This work is supported by DARPA under Contract No. DAAD1902-1-0208.
References [1] J.S. Patel, K. Rastani, Opt. Lett. 16 (1991) 532. [2] T. Nose, S. Masuda, S. Sato, J. Li, L.C. Chien, P.J. Bos, Opt. Lett. 22 (1997) 351. [3] S. Masuda, T. Nose, S. Sato, Appl. Opt. 37 (1998) 2067. [4] W. Klaus, M. Ide, Y. Hayano, S. Morokawa, Y. Arimoto, Proc. SPIE 3635 (1999) 66. [5] L.G. Commander, S.E. Day, D.R. Selviah, Opt. Commun. 177 (2000) 157. [6] Y. Choi, J.H. Park, J.H. Kim, S.D. Lee, Opt. Mater. 21 (2002) 643. [7] H. Ren, S.T. Wu, Appl. Phys. Lett. 82 (2003) 22. [8] H. Ren, Y.H. Fan, S.T. Wu, Appl. Phys. Lett. 83 (2003) 1515. [9] E. Hecht, Optics, second ed., Addison-Wesley, Reading, MA, 1987 (Chapter 10).