This paper presents an optical film that enhances cosmetic appearance as well as brightness in a liquid crystal display (LCD) through microprisms that have a variable pitch. The optical film utilizes Fourier transformation to optimize the arrangement of microprisms for improving the cosmetic appearance in the display. The optical film has an improved light-collimating feature that redirects light more effectively, resulting in higher brightness. This paper shows details of the design procedure, but more importantly, presents optical measurement results of an actual optical film prototype to confirm the performance improvement.
© 2007 Optical Society of America
Liquid crystal display (LCD) has become a mainstream in the flat panel display (FPD) industry because of its versatility, ever- improving cost performance, and much-enhanced image quality. In particular, mobile electronics such as the MP3 player, laptop, and PDA are indebted to the rapid development of LCD for their popularity. It is no exaggeration that they could not survive in the consumer market without LCD as a main display unit.
It is noticeable that, because LCD is not a self-luminous display device, it needs a separate illumination system, the so-called backlight unit, as shown in Fig. 1. Most backlight units consist of a light source, wave- guiding plate, and very thin optical film such as a prism sheet. While cold cathode fluorescent lamps (CCFL) or LEDs are most popularly utilized because they have long lifetimes, are reliable, and are cost-effective, they are also spatially localized. That is, they are linear or point-like, so it is necessary to have additional non-imaging optical devices such as wave-guiding plates to distribute the light uniformly over the display area. However, the luminance distribution from a light-guiding plate (LGP) is not in a usable form because it has low brightness in the on-axis direction, the normal displaying area, while the peak brightness occurs in the off-axis direction. A prism sheet transforms the luminance distribution into a more desirable shape, for instance, directing the higher brightness in the viewing direction rather than the other direction.
The design of an illumination system in a mobile display focuses on a high on-axis brightness for the single user, as well as on cosmetic appearance, thinness, and mechanical robustness. An optical film has been developed to satisfy these optical and physical requirements. Optical films such as Brightness Enhancement Film from 3M and DIAART prism from Mitsubishi Rayon have been utilized to improve the viewing attributes of display [1,2]. To improve viewing attributes even more, there are numerous research efforts being made, such as creating micro-patterns or adding scattering particles on a wave-guiding plate [3,4,5,6,7,8]. However, while these research projects are mainly focused on the wave-guiding plate, very little improvement has been made in optical film since its introduction to LCD. Mainly, the LCD industry uses an optical film that has an array of linear prisms arranged in a parallel. For example, most of optical films used in LCD have 90 degree or 60 degree linear prism array with a little shape change such as a prism height variation to minimize wet-out. It is noticeable LCD industry keeps using same type of optical film for more than a decade. However, it is not unthinkable that there is room for design improvement in optical film to achieve a higher brightness and a better cosmetic appearance.
In this paper, we present an optical film that delivers improved on-axis brightness, while being less prone to create the Moiré appearance that has been problematic in many LCD. The principle behind improving Moiré appearance is presented in section 2. The following section covers a design procedure of how to optimize a prism shape under multiple constraints. In section 4, we discuss prototyping the design chosen in the previous section.
2. Improving Moiré appearance
To improve Moiré appearance, it is important to understand how a Moiré appearance occurs in LCD. Because a Moiré appearance is nothing more than the interference between two periodic structures, in this example, a thin-film transistor liquid crystal (TFT-LC) and an optical film, the Moiré appearance can be nullified by weakening the periodicity of the patterns . Let us say there are two periodic patterns, a film ffilm(x,y) and a TFT-LC fTFT-LC(x,y). After a Fourier transformation of them, their Fourier counterparts are represented by:
The (ξ,η) is a coordinate of spatial frequency in Fourier space. Generally FTFT-LC(ξ,η) is a series of δ-functions located at a harmonic frequency because TFT-LC can be approximated as an array of rectangular apertures. To generate Moiré appearance, two patterns are superimposed, which is equivalent to a multiplication of the equations ffilm(x,y) and fTFT-LC(x,y). Mathematically, the final pattern in Fourier space is:
where * means convolution. Notice that a multiplication in a real space is converted to a convolution in Fourier space . Assuming that the pitches of TFT-LC are PTFT-LC in x-direction and QTFT-LC in y-direction, FTFT-LC (ξ,η) can be re-written as:
where an,m is amplitude of harmonics in TFT-LC. It represents how strong a periodic pattern is at given periodicity. Likewise, Ffilm(ξ,η) is represented by
where δ(x - a)* δ(x - b)= δ(x - a + b). The first two terms in Eq. (6) are original Fourier space representations of the patterns, that is, TFT-LC and film, while the third term represents the Moiré appearance . Given that we are only interested in Moiré appearance, we could separate the third term from the others. Considering the resolution of the human eye, it is not necessary to consider a high order of spatial frequency because an,m and bn,m decrease rapidly as n and m increase. In practice, we could set up upper criteria of n, m without losing any accuracy.
According to Eq. (7), Moiré appearance only occurs when the following condition is satisfied:
The amplitude of Moiré appearance is:
Equation (9) means that the visibility of Moiré appearance is proportional to the amplitude of the two periodic patterns, that is, the strength of periodicity. In other words, Moiré appearance could be weakened by reducing the strength of periodicity of the two patterns. This is our approach to resolving Moiré appearance.
Let us explain how to suppress the periodicity of a film. Fig. 2 shows a 3D rendering of the proposed optical film. We will present actually prototyped sample of the proposed optical film in section 4.The microstructure on the optical film is an array of elongated microprisms that are arranged to have a lateral waviness while microprisms are randomly staggered. As a result, the pitches between two microprisms randomly vary over a two-dimensional (2D) space.
This concept is illustrated in Fig. 3. The wiggly lines indicate apexes of microprisms while the triangles at the bottom of the figures illustrate a profile of the microprisms. Lateral waviness alone does not create the variation in pitch. As shown on the left in the figure, the pitch in every location is constant (P1=P2=P3=P4). The random staggering on the right shows each microprism arranged in a non-parallel manner. This makes the pitch at different locations vary (P1≠P2≠P3≠P4). The lateral waviness along with the random staggering is effective to suppress the periodicity of microstructure that is directly related to the strength of Moiré appearance. While this approach helps resolve Moiré appearance, there is a chance that it might impact optical performance, that is, the on-axis brightness, in an undesirable manner. The reason is that this type of optical film is designed to accept a very well-defined incidence angle of beam emitted from a well-defined area. The varying pitch may disturb these well-defined angles and areas. This paper will address this issue in more detail in the next section.
To verify the concept that varying pitches reduce a strength of periodicity in optical film, a spatial frequency of the example shown in Fig. 2 is calculated through Fast Fourier Transformation (FFT). We use a commercial software package, Mathematica. The result is presented in Fig. 4 in periodicity amplitude vs. spatial frequency. A higher peak means that a pattern has a stronger periodicity at the same spatial frequency. The solid line is a spatial frequency of one pattern with no lateral waviness and no random staggering, and the dotted line is a spatial frequency of one pattern that does have lateral waviness and random staggering. The peak amplitude in the solid line is generally higher than the one in the dotted line. This implies that the periodicity in the pattern with no lateral waviness and no random staggering is stronger than the one in the pattern that does have lateral waviness and random staggering. Simply put, bń,ḿ in Eq. (9) decreases, which results in the reduction of the amplitude of Moiré appearance. Therefore, Moiré appearance could be suppressed because of weaker periodicity in the optical film with lateral waviness and random staggering. The Moiré analysis using FFT is recommended in actual applications because FFT calculation shows numerical strength of periodicity that is related to Moiré appearance.
3. Design of optical film
Although lateral waviness helps to suppress Moiré appearance, it may cause the reduction of on-axis brightness, which is not desirable in a mobile display. In order to compensate for this undesirable outcome, the microprism is newly designed to maintain the same brightness or to enhance brightness further.
Because this type of optical film uses reflection on one surface of the microprism, the surface shape and its slope might be important factors in an optical design. The conventional optical film uses a flat surface tilted at some proper angle to simply turn incident light in the viewing direction. However, it is well known that some curved geometry can be optimized to improve the collimation effect, i.e., it sends light within the smaller cone of the angle to improve on-axis brightness . The geometry shown in Fig. 5(b) is a profile of the microprism, consisting of one flat surface on which light is incident, and one curved surface on which light is reflected, while the conventional optical film in Fig. 5(a) has an array of prisms with two flat surfaces. The curved shape is optimized to create the highest on-axis brightness. The black lines in the figures are geometrical rays in commercially available ray-tracing code. The geometry is modeled in the code to optimize a curved geometry.
The brightness of the curved microprism is simulated in ASAP from the Breault research organization. Although a prism is seemingly simple to design, there are multiple design parameters that should be carefully considered because they interact with each other to affect the final result, that is, the brightness. The design parameters include prism height, a radius of the curved surface, a tilt angle of a flat surface, and a tilt angle of a curved surface. In addition, one should consider the manufacturability of the microprism. Because of the dimension of a microprism on a micrometer scale, there is often a severe constraint on what could be manufacturable. For example, a simple conic shape that is manufacturable in macroscale is painstaking and expensive to fabricate although it has the potential of a higher brightness. Because of this, only the spherical shape is considered in this design practice. Because the design space is hyperdimensional, each parameter is adjusted within some manufacturable range, and its optical performance, i.e., its on-axis brightness, is plotted to select the best performer [see Fig. 6]. For example, the radius of the curved side ranges from 255—320 μm. It is found that 290 μm is a best performance value as well as a manufacturable one. The graphs in Fig. 6 share the same trend—a relatively slow variation around the peak performance. The range of parameter variations results in an acceptable head-on performance. This is particularly good for manufacturability because it means that this design has a relatively low sensitivity to parameter change, that is, loose tolerance. Through numerous series of modeling by tweaking design parameters, the final design is selected as shown in Table 1.
After finishing the entire design practice, the best performer is chosen in consideration of optical performance, manufacturability, and tolerance, i.e., the variation of performance distribution.
Fig. 7 shows a SEM picture of the proposed optical film sample. It shows the apex of the microstructures pointing upward, but in actual application, the apex needs to point downward to a wave-guiding plate such that light from the wave-guiding plate is totally, internally reflected in the viewing direction. As explained in section 2, the microstructure on the optical film is the array of elongated microprisms that are created to have a lateral waviness and are staggered carefully such that pitches between two microprisms are randomly varied over 2D space. The nominal pitch in Fig. 7 is approximately 50 μm and the amplitude of lateral waviness is ±2 μm. Therefore the pitch ranges from 46–54 μm.
This lateral waviness creates the variation of pitch because one column of microprisms is staggered longitudinally with the adjacent one. The example has equal amplitude of the lateral waviness over every microprism, but the phase of the lateral waviness on each microprism is randomly chosen through a random number generator. When the optical film is dissected along the line a, as shown in Fig. 7, the pitch, that is, the spacing between apexes, varies along the same line.
The final design is prototyped in polycarbonate by using well-known mechanical recombination technologies . The creation of a micro-optical sample via the mechanical recombination process involves several steps including the creation of the master mold, creation of an electroform replica, and pattern transfer via embossing. The first step of the process involves the creation of the master mold containing the optical-quality microprism. The microprism within the master mold may be created via chemical or mechanical micromachining methods, including photoresist patterning, silicon etching, ion-beam machining, computer numerical control (CNC) micromachining, and/or electrical discharge machining (EDM). Upon successful creation of the master mold, an electroform replica is typically generated to provide a metallic substrate for subsequent process steps. The electroforming step itself employs a metal deposition method (either electroless or electrolytic) to continuously build up a layer of metal onto the master mold at an atomic scale producing an extremely accurate replica of the master mold microstructure. The resulting metallic build-up forms a solid structure of controlled thickness that can be separated from the master. The final step of mechanical recombination employs an embossing method to transfer patterns and/or microstructures from the electroform via heat and pressure to a polymeric substrate.
5. Optical performance and cosmetic appearance
A prototyped sample is compared with the commercially available optical film from a laptop backlit unit manufactured by Yuka Electronics (Tokyo, Japan, http://www.yukadenshi.co.jp). As shown in Fig. 8, the proposed optical film demonstrates superior optical performance. The peak brightness is 5553 nits vs. 4397 nits from the conventional optical film, which is equal to a 27% improvement. The two iso-luminance plots show better collimation effect and higher peak brightness (see numbers in blue ovals) in the proposed optical film. Both of the films are placed on the same wave-guiding plate with the same CCFL for direct comparison. Brightness is measured under a 2D brightness meter, EZContrast from ELDIM (Caen, France http://www.eldim.fr).
The same sets of data are presented with a luminance profile in Fig. 9. The solid line in is a measured brightness profile of the proposed optical film while the dotted line in the same graph is same data of the conventional optical film.
As claimed before, the proposed optical film is effective to suppress Moiré appearance. To verify Moiré appearance improvement, both optical films are superimposed with a grating whose periodicity is 147 μm, which is one of the common periodicities in mobile display. Fig. 10 shows a much weaker Moiré appearance in the proposed optical film, while Moiré appearance is more visible in a conventional optical film.
In summary, the proposed optical film shows superior optical performance, i.e., higher on-axis brightness along with much weaker Moiré appearance, which is desirable for a mobile display. To achieve both effects, this paper proposes a novel microprism arrangement of the lateral waviness combined with random staggering to improve the viewing experience of LCD. It also reports an optimization process to design a curved geometry on a microprism. The curved geometry is optimized to maximize the beam collimation resulting in higher on-axis brightness while still being manufacturable.
The concept of a novel microprism arrangement, the combination of the lateral waviness with random staggering, is a simple but effective method to suppress the Moiré appearance that has been a problem in LCD because it addresses the fundamental cause of Moiré appearance–the strength of periodicity. The mathematical development behind it supports the argument that the same approach can be applied to a wider range of display devices. The effect of the novel microprism arrangement is demonstrated through the data showing weaker Moiré appearance compared to a conventional system using a simple linear arrangement.
The optimization of the curved geometry provides the ideal design to have higher on-axis luminance as well as maintain a reasonable manufacturability. The microprism has a shallow curvature on the reflective side so as to reflect the incident beam in the on-axis direction more effectively. The prototype confirms the proposed design to be superior to the conventional system in terms of on-axis brightness.
Considering all the improvements made here, the proposed optical film must satisfy current demands for a better display from viewers and the display industry.
The authors would like to thank Display Sciences and Technology Center in Eastman Kodak Company for its financial support on this study.
References and links
3. T. Okumura, A. Tagaya, and Y. Koike, “Highly-efficient backlight for liquid crystal display having no optical films,” Appl. Phys. Lett. 83, 2515–2517 (2003). [CrossRef]
4. T. Idé, H. Mizuta, H. Numata, Y. Taira, M. Suzuki, M. Noguchi, and Y. Katsu, “Dot pattern generation technique using molecular dynamics,” J. Opt. Soc. Am. A 20, 248–255 (2003). [CrossRef]
5. K. Käläntär, “Functional Light-Guide Plate for Backlight Unit,” in Society for Information Display 1999 International Symposium, J. Morreale, ed., (San Jose, CA, 1999) 30, pp. 764–767.
6. K. Käläntär, S. Matsumoto, T. Katoh, and T. Mizuno, “Double-Side Emissive Backlight Unit for Transmissive LCD using a Single Functional Light-Guide Plate,” in The Tenth International Display Workshops, T. Uchida, ed., (Fukuoka, Japan, 2003), pp. 661–664.
7. K. Käläntär, S. Matsumoto, T. Mizuno, and T. Katoh, “Backlight Unit With Double Surface Light Emission using a Single Micro-Structured Light-Guide Plate,” in Society for Information Display 2004 International Symposium, J. Morreale, ed., (San Jose, CA, 2004), pp. 1182–1185.
8. K. Oki, “Novel Backlight with High Luminance and Low Power Consumption by Prism-on-Light-Pipe Technology,” in Society for Information Display 1998 International Symposium, Morreale J., ed., (Anaheim, CA, 1998) 29, pp. 157–160.
9. I. Amidror and R. D. Hersch, “Fourier-based analysis of phase shifts in the superposition of periodic layers and their moiré effects,” J. Opt. Soc. Am. A 13, 974–987 (1996). [CrossRef]
10. Jack D. Gaskill, Linear Systems, Fourier Transforms, and Optics(John Wiley & Sons, New York, 1978).
11. K. Creath and J. C. Wyant, “Moire and Fringe Projection Techniques,” in Optical Shop Testing, D. Malacara, ed. (John Wiley & Sons, New York, 1992), pp. 653–655.
12. John E. Greivenkamp, Field Guide to Geometrical Optics(SPIE, Washington, 2004). [CrossRef]
13. Marc Madou, Fundamentals of Microfabrication(CRC Press, New York, 1997).