Open Access
Add to BibSonomy
Abstract
All-glass, micro-groove arrays were etched into a glass light guide plate surface, transforming edge-lit incident light into uniform area output from its surface. All micro-grooves in the array shared similar shape and extraction efficiency. They were spray etched into Corning alkali-borosilicate glass (Iris) effectively controlling the degree of light extraction uniformity along the array’s normal direction by varying the inter-groove pitch from 300–1000 µm. Luminance uniformities in excess of 80% for 9-point, or 75% for 455-point measurements were achieved with 1.2% extraction efficiency per groove. Enhanced light redirection from individual micro-grooves towards normal viewing reduced number of additional beam steering films required in light guide plate, backlight units.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Large, bright, and colorful liquid crystal thin displays (LCD) rely on a spatially uniform and pure spectral distribution of surface light, a function produced for displays typically by a back-light unit (BLU) using light guide plates (LGP) [1–3]. The left side of Fig. 1 shows an edge-lit conventional BLU for LCD displays utilizing a light emitting diode (LED) array source, a light guide plate, a backlight reflector, a diffusive film (sometimes two), and a couple of brightness enhancement films (BEF-vertical horizontal, and dual brightness enhancement film (DBEF) [4–5].
Fig. 1. Left - Illustration of a typical LCD – LED-LGP system layout. Right - The corresponding angular luminance distribution simulation results obtained at locations corresponding to the adjacent system layout component. The colored diagrams represent the locally scattered light intensity towards the viewer, with the center of the diagram representing the direction toward the viewer. Monotonically decreasing light intensity is represented with pseudo-colors from red, green, to blue respectively.
A light extraction pattern is printed or molded on the back side of the LGP to uniformly extract light from the entire substrate [3]. The right side of Fig. 1 graphically illustrates the angular distribution associated with the individual optical macro-elements shown to the left and offers insight into the task of the optical engineer to fashion a collection of sequential optical film behaviors to optimize the overall system design efficiency. The right side of Fig. 1 also shows the angular distribution of the extracted light from the LGP, by simulation, which is detected from (a) the front surface of the LGP, (b) after the diffusive film, (c) and after the BEF. A diffusive film is typically placed on top of the LGP to smoothen the angular and spatial distributions of the output light of the LGP (see pseudo-color Diffuser output), while BEFs are used when one needs to substantially redirect light (see pseudo-color BEF output). Dual brightness enhancement film (DBEF) provides additional light output by reflecting light of the “wrong” polarization instead of absorbing it, subjecting it to further reflections, effectively randomizing its polarization with each reflection, so some of the light passes back through the DBEF.
From an efficiency standpoint, current refinements to the conventional BLU described above consider new films that combine or eliminate the film stack’s individual optical functions. From a material perspective, highly transmissive glass substrates, with negligible color shift, provide dimensional stability required for ultra-slim displays against variations in heat or humidity that might otherwise warp or expand, compromising it’s opto-mechanical performance. An integrated backlight module, composed of a compound optical film (COF) and LGP, was reported in [6–7]. In this backlight module, the COF not only adjusts the distribution of the output light by varying the density of the microstructure on the COF but can also collimate the LGP cross-directional light trajectories into the normal direction. We believe applications requiring dimensional stability under high temperature operation would profit from thin, all-glass substrates prepared using low cost procedures with minimal process steps.
Optical designs were recently developed for new LGP glass substrates with different micro-sized groove surface topologies (see Fig. 2), with specified wall angles and depths, [8] based on previous work etching all-glass lenticular light guide plates [9]. These features enable “turning” the light output angle in the normal direction (direction of an on-axis viewer) thus enabling a significantly simplified edge-lit BLU by eliminating BEF(s).
Fig. 2. (top) Cross-section view of one micro-size groove design where two groove morphologies (b1 – left, and b2 - right) exhibit a concave wall shape with an α wall angle and depth H. These specific shapes arose from the use of wet etch processes under different conditions. (bottom) Schematic 3D view of LGP showing the orientation of microgrooves aligned with the LED input array.
Ray tracing simulation of the optical designs indicate that they could also serve the function of light extraction (See Supplemental Document Section 1) and thereby eliminate the need to print or mold an extraction dot pattern . The following was adapted from Ref. [8] and discussed in the Supplemental Document Sections 1 and 2 describing some of the key micro-groove surface features influencing the individual micro-groove’s extraction efficiency. While the modeling entailed a variety of different micro-groove morphologies (convex walls, concave walls, flat walls), the key dependencies are described below using simple flat wall micro-groove surfaces since they adequately capture the general observed trends.
Figure 3 illustrates some of the key influences on micro-groove extraction efficiency [8] by modeling in-coupled TIR ray interactions with different micro-groove surface features. The top row in Fig. 3 illustrates the micro-groove wall angle effect on extraction efficiency. A representative ray is shown (ΘTIR = 85°) from the range of in-coupled TIR rays (42 < ΘTIR ≤ 90°) interacting with two micro-groove surfaces differing in wall angle: left α = 20°, right α = 45°. The ray either refracts off the micro-groove wall surface if the wall angle α is less than ∼ 30° or reflects off it if α exceeds ∼ 30°. No extraction occurs when the ray simply refracts at the low wall angle surface, typically passing through the air gap, refracting back into the LGP and retaining the TIR condition. When the wall angle exceeds ∼ 30°, the refraction is locally disrupted, instead reflecting off the groove wall by a local TIR condition towards the bottom interface and emerging from the LGP by refraction, thus increasing the extraction efficiency. The right panel in the top row (a) shows the extraction efficiency per micro-groove variation with wall angle. Micro-groove wall angles beyond ∼ 50° limit the amount of light re-coupling back into the LGP because the larger the groove width or wall groove bottom, the less extracted light hits the back-groove wall (See Supplemental Document Section 2). The middle row (b) left panel illustrates the micro-groove depth’s influence on extraction efficiency using an array of rays, with fixed offset, impacting two micro-groove structures varying only in groove depth: left structure depth is shallower than the right. The right panel of the middle row describes the extraction efficiency’s linear scaling dependence on depth. The ray traces on the left panel graphically illustrate the number of additional TIR rays that are additionally reflected and ultimately extracted with the increasing micro-groove depth. The bottom row (c) illustrates the behavior of increasing the micro-groove width from two perspectives. It first shows the behavior of light re-coupling back into the LGP from reflections off the back-groove wall. It also shows the extraction efficiency behavior with increasing micro-groove width due to refraction from the front-groove wall. The in-coupled ray trajectory of the top-left cartoon describes the loss of re-coupling rays when the bottom width is too large, while the in-coupled ray trajectories in the bottom-left cartoon show negligible influence of groove width on extracted rays from front-groove wall interactions. The percentage of light that is recycled back into the LGP as the groove width increases is shown in the bottom row right panel.
Fig. 3. In-coupled, total-internal-reflected (TIR) ray interaction with micro-groove surface features were modeled with ray tracing software (Light Tools, Mountain View, CA) showing key influences on micro-groove extraction efficiency – (a) Top row illustrates the micro-groove wall angle α effect on extraction efficiency. (b) Middle row illustrates the micro-groove depth’s influence on extraction efficiency. (c) Bottom row illustrates the BEF degradation behavior with increasing micro-groove width.
Efforts to produce uniform area output from edge-lit substrates using previous screen-print glass-etching procedures [9] failed because the etching duration required to achieve appropriate wall angles greater than ∼ 30° resulted in deeply etched grooves with unsuitably large extraction efficiencies (see Supplement Document Section 3 and Figures S4, S6). Much shallower etch depths were required to make uniform light extraction possible over the large distances required for display applications. Our objective and challenge were therefore to determine fabrication conditions enabling shallow micro-groove formation with wall angles greater than ∼ 30° to make light extraction possible over relevant display-length distances. We now describe our approach to fabricating the microgrooves directly in an alkali-borosilicate glass based on screen printing and wet etch processes, with luminance, angular distribution measurements and performance characterization.
2. Concept overview
Uniform light extraction can be achieved by using different patterned microgroove designs on an LGP. Three typical design patterns are as follows. The first design involves varying microgroove density (or spacing) along the light propagating direction (from light coupling edge to output edge) while keeping microgroove shape and size the same. The second design involves varying microgroove shape and/or size along the light propagating direction while keeping the microgroove spacing constant. The third design simultaneously varies microgroove spacing and microgroove shape and/or size along the light propagating direction. In this paper, the first design is used. The top view of Fig. 4 illustrates the basic arrangement of microgrooves distributed along the y-direction normal to the LED array length. The inter-groove spacing varies along the y direction in a unique monotonic sequence S(y), initially with a sparse density, tending towards denser inter-groove spacing to redirect ever more of the diminishing LED-injected TIR light, providing uniform illumination to the viewer. In practice, we typically performed several iterations in varying S(y) for a given microgroove shape and size to achieve uniform surface illumination greater than 80% with 9-point spatial uniformity measurements, or 75% using 455-point.
Fig. 4. Schematic model of LGP light extraction with micro-grooves. (Top view) Cartoon illustrates the micro-groove spatial variation, S(y), used to confer uniform surface output illumination towards the viewer/reader. The inset at the right, illustrates the essential micro-groove surface attributes (angle α, depth, width) used to redirect light. (Side view) Cross-sectional view of LGP portion along the ray propagation direction, for one specific micro-size groove design. Injected LED light Po, into the LGP, propagates as total internally reflected (TIR) light, and is refracted or reflected at each micro-groove with a characteristic efficiency factor ε ≡ (Pout /Pin) = (Pf + Pb)/Pin, describing the ratio of light lost per micro-groove per unit mm, regardless of mechanism: Pf, Pb, etc.
The inset in the top view of Fig. 4 again illustrates the basic individual microgroove structural elements that refract the propagating totally internally reflected light. Simulations [8], Fig. 3 and the Supplemental Document Sections 1-2 described the key micro-groove structural features influencing the overall extraction behavior: wall angle α, depth H, and width w.
The side view in Fig. 4 illustrates our use of the individual micro-grooves in the LGP. While the spacing S(y) between adjacent micro-grooves varies, the extraction efficiency does not since they share an equivalent topology. The power extracting fraction at each individual micro-groove is assumed constant and characterized as a ratio of the average amount of power lost at a given micro-groove, to the average power incident on it. We formulate this assumption by first characterizing the average power lost at the nth microgroove as Pn-out, the incident power on it as Pn-1, and the power propagating to the next micro-groove as Pn. The assumption of constant extraction efficiency ɛ per micro-groove is then described as
3. Mask and etch processes
Figure 5 provides a broad overview of the key individual steps used in the overall screen print mask and spray etch procedures that made the micro-groove arrays. Cleaned iris glass parts were first subjected to an atmospheric plasma step to oxidize physisorbed organics on the glass and promote efficient adhesion of potential adhesion promoters.
Fig. 5. Overview of “Mask and “Etch” process individual steps - Corning Iris glass substrate was cleaned, plasma treated to oxidize organic residue on the surface, prior to deposition of the chrome adhesion promoter. A stainless steel 360 mesh screen with optimized emulsion pattern was used to screen print the microgrooves, followed by a post bake thermal cure. The cured substrate was exposed to a chrome etchant to remove chromium layers not masked by the screen-print ink and followed by the spray etching process. The final step is to remove the screen print mask and chromium residue.
The process closely follows previous mask and etch procedures used to prepare optically functional glass surfaces [9], with the key addition of a sputtered thin chromium film (40 nm) introduced as an adhesion promoter to simultaneously provide micro-grooves with suitably low wall angle and shallow etch depth (see Supplemental Document Section 3). Screen-print ink (CGSN-4011; Sun Chemical, Parsippany, NJ-USA) was directly printed on the chromium coated glass, without dilution, to define the pattern that achieved the microgroove features after the etching step. The chromium’s superior adhesion, along with the screen-print ink chemistry, was sufficient to withstand the harsh HF etching conditions required to achieve shallow microgroove depths in the relatively short etch duration. The screen-printed surface was then exposed to a chrome etchant (Sigma -Aldrich: Prod. No. 651826) to remove chrome from the non-masked region to expose the glass surface for etching. Based on our previous studies [9] with spray etching, we chose a 10 vol% HF- 30 vol% H2SO4 mixture at 38°C with a 90° nozzle angle, a spray pressure of 7-25 psi, for spray etching due to its fast and uniform etch rate (5-6 µm/min). The etch time was varied to achieve a shallow etch depth of 3-20 µm required for the extraction pattern. The final step was the removal of both the screen print ink and the chrome layer using both 5-10% potassium hydroxide (diluted Fisher: SP2261 with distilled water) for 5-10 minutes followed by chrome etchant for 3 minutes.
4. Results and discussion
An SEM image in Fig. 6 shows a typical etched microgroove in an Iris glass substrate 1.1 mm thick. While wall angles α, groove depth H and top and bottom groove widths w were initially calculated both by SEM image analysis and profilometer measurements, SEM image analysis measurements were ultimately used for much of this work. Optical modeling [8] suggested wall angles in the range: 30° < α < 50° were suitable for extracting light.
Fig. 6. SEM image of a typical microgroove is shown illustrating the features from which wall angle α, groove depth H, and top widths w were derived.
Some examples of the microgrooves fabricated are shown in Table 1 with the corresponding spray etch condition, wall angle, groove width and depth. Samples M-1, 2, 4, 6 used a chromium layer, while sample KIW03, using an epoxy-silane additive as a solution-based adhesion promoter, and sample CGSN-1,2, used without additives, were prepared without chromium film and yielded etch structures with significant groove widths. The spray etch experiments used either 53° or 90° cone nozzle angles at pressures between 7 or 25 psi. Chromium coated samples clearly showed promise for producing micro-rooves with large α and shallow depth. Wider and deeper channels were obtained using higher pressures irrespective of the nozzle angle. In practice, spray parameter optimization was performed once the etch depth was selected.
Table 1. Compilation of experimental screen mask and etch parameters, and micro-groove surface features described in the text to fabricate an extracting all-glass LGP micro-groove array.
Figure 7 shows angular radiance distribution results for microgroove samples that were edge lit with an LED array parallel to the microgrooves, with the grooves facing the reflector in a Dell monitor tear-down (Dell-S2718D 27-inch-LED-edge-lit-monitor). Measurements were carried out first without any optical films, then adding 1 diffuser film, or 2 diffuser films. No BEFs were used in this measurement. The image on the left (without diffuser film) shows broad angle-based light distribution with significant amount of light at 90° from normal and less light at 0°. Because of the wide flat bottom groove, part of the low angle light that cannot hit the microgroove back wall is not recoupled into the LGP, resulting in the light propagating towards the output edge. This was improved upon by addition of a single diffuser film, where most of the light is directed to 0° toward the viewer, as shown on the image on right in Fig. 7. This behavior suggests the diffuser film served to recollimate the locally extracted light towards the viewer in a manner reminiscent of integrated backlight modules (IBLM), composed of compound optical films (COF) used to provide collimated and uniform planar light output [6]. The diffuser sheets we employed were taken from the Dell monitor tear-down and appeared to be a surface diffuser prepared by coating spherical particles dispersed on a flat polymer surface. Further reduction of the groove flat bottom width may promote more efficient light extraction towards the viewer in the normal direction, as discussed in Supplemental Document Section 2, and described in Fig. 3. Based on these experiments, screen printing on the thin chrome adhesion layer followed by spray etching was sufficient in producing microgrooves with wall angles and widths suitable for extracting TIR light in a uniform manner.
Fig. 7. Angular radiance distribution data (from ELDIM tool) is shown for a micro-groove sample with and without a diffuser film in an edge lit configuration as described. (Left) Image shows the sample without diffuser where broad angle based light distribution is observed with significant amount of light at 90° from normal, with a lesser amount of light remaining at 0°. (Right) Image shows the result with addition of a single diffuser film, where all the light is observed shifted to the diagram center, toward the viewer.
Microgroove substrates were prepared using the procedures described above and their micro-groove structural attributes, extraction performance, and uniformity characterized with luminance output images shown in Fig. 8 below. LED-array illumination was launched along the sample’s lower edge [9], as shown in Fig. 8 – column 2, with reflecting tape (Avery-Denison, FT 5255) placed along the sample’s top edge for efficient light operation [8]. The absence of reflecting tape manifests as a linear luminance spike along the top edge, as shown with sample WS3694. The luminance output images were made using a scientific grade CCD camera (Prometric 16i - Radiant Vision Systems; Redmond, Washington), from which profilometric traces were taken to characterize our output uniformity.
Fig. 8. Experimental micro-groove luminance output data of etched Iris glass substrates 1.1 mm thick, illustrating the luminance uniformity output performance with varying micro-groove parameters: width, depth, wall-angle, and extraction efficiency ɛ. Column 1 - sample name, Column 2 – luminance surface area image grayscale, Column 3 – typical profilometer trace of luminance (nits), Column 4 – SEM image, Column 5 – top and bottom widths, Column 6 – groove depth, Column 7 – wall angle, Column 8 - extraction efficiency per micro-groove ɛ measured from luminance data. The source from which the micro-groove parameters were obtained are indicated in the top of columns 5, 6, and 7.
Figure 8 – columns 2 and 3 show the variation of the display luminance spatial uniformity to compare against the individual micro-groove’s surface attributes (columns - 4 through 7) and extraction efficiency per micro-groove (column 8). The luminance uniformity data was fit using the extraction efficiency determined by the experimental method described above in Concept Overview and an example shown below in Fig. 9(a).
Fig. 9. (a) Experimental fit of sample WS3727 spatial luminance data with extraction efficiency value determined by procedure described in Concept Overview. (b) Experimental and modeled extraction efficiency per microgroove of 0.7 mm and 1.1 mm thick glass LGPs with microgrooves. The orange dot data set (●) represents the experimental data of extraction efficiency vs etch depth for a 0.7 mm thick glass LGP. The solid orange line is the modeled curve of extraction efficiency per microgroove as a function of groove depth for a 0.7 mm thick LGP with microgrooves designed with a 40° wall angle α and 60 µm bottom width. The blue dot data set (●) represents experimental data with an extraction efficiency per microgroove vs etch depth for a 1.1 mm thick glass LGP. The solid blue line is the modeled curve of extraction efficiency per microgroove as a function of microgroove depth for 1.1 mm thick LGP with microgrooves designed with a 40° wall angle α and 60 µm bottom width.
Simulation of extraction efficiency vs groove depth was performed with the Light Tool LGP model described in [8] and characterized in Supplemental Document – sections 1 and 2. All micro-grooves assumed a 40° wall angle and 60 µm bottom width, with Iris glass substrate thicknesses either 1.1 mm (blue data) or 0.7 mm (orange data). The results are shown in Fig. 9(b). The experimentally determined extraction efficiencies for the 1.1 mm LGP compared well with simulation results indicating extraction efficiency of the order ∼ 0.4% per micro-groove were suitable for extraction patterns providing uniform light output with > 80% extraction efficiency over ∼200 mm long LGPs (9-point measurement [8]). It also shows the light extraction efficiency correlates well with micro-groove depth for the thicker substrates. Simulations also showed the general increase in extraction efficiency with the thinner 0.7 mm substrates, and yet a greater discrepancy between the experimental and simulated extraction efficiencies was observed with the thinner LGP substrates. This is due to efficient sampling of the microgrooves when the ray trajectories reflect at the glass-air interface more often with the thinner substrates. The difference between experimental and modeling results are mainly caused by scattering properties of etched microgroove surface roughness and shape that are unaccounted for. In the model, microgroove surfaces were all assumed to be smooth with no extraneous scattering introduced. In the experiments however, the etched microgroove surface always exhibited a finite roughness which could cause more light extracted out of the LGP. This deviation from our model was more pronounced with thinner substrates due to a small but finite LGP surface roughness and more ray-substrate wall interactions.
Finally, we designed a micro-groove extraction pattern by varying the inter-groove spacing from 300 to 966 µm along the distance between input and output edges based on an extraction efficiency (∼ 1.2%) of 15 µm deep micro-grooves, using the methods outlined in [8]. We then prepared a 1.1 mm thick 115 mm x 305 mm LGP by the mask and etch process. Figure 10 shows the image of the resulting 115 mm x 305 mm BLU using the 1.1 mm thick all-glass LGP etched with microgrooves. An average luminance of 19956 nits with 73% uniformity (455-points) was achieved by this BLU. It confirmed the feasibility of the mask and etch process in fabricating all-glass LGPs.
Fig. 10. A 115 mm x 305 mm BLU is shown without diffuser plate, using a 1.1 mm thick all-glass LGP prepared with microgrooves using the mask and etch procedure described above. Incident light was coupled along the image’s bottom edge, from an LED array in the Dell monitor tear-down towards the top edge of the image. Reflecting tape was placed along the top edge. A pseudo-color spectrum is shown illustrating the scale associated with the luminance (NIT) imaging measurement.
5. Summary
In summary, all glass, micro-groove, LGP substrates were fabricated in a mask and etch procedure transforming edge lit incident light into highly uniform area output from the its surface. The micro-groove array (width: 70–90 µm, depth: 5 - 18 µm) was etched into the glass surface, parallel to the LED array along the length of the thin-slab glass edge, with pitch variation in the range 300 - 966 µm using an etchant optimized for the Iris Glass. A thin 40 nm chromium coating applied to the LGP prior to screen-printing made production of shallow depth micro-grooves with large wall angles between ∼ 30° < α < 50° possible. The trench-like, negative-relief etched groove structures exhibited an extraction efficiency per micro-groove between 0.45–1.4%, with a varying pitch effectively controlling the degree of light extraction uniformity (luminance uniformity: > 80% - 9 points, and > 75% - 455 points) along the micro-grooves’ varying pitch direction. The range of microgroove wall angles, depths, and widths effectively controlled the light output angle towards the LCD display viewer, using a single diffuser film without additional need for brightness enhancing films. The all-glass construction enabled high brightness operation and applications requiring dimensional stability under high temperature operation.
Acknowledgments
Authors gratefully acknowledge etching by Lou Stempin, Tracie Carleton and Chemcut Corporation (State College, PA), screen printing assistance from David L Baker and Mandakini Kanungo, ELDIM imaging assistance by Tomohiro Ishikawa and Tina Proulx, SEM images from Dave E. Baker, Ron Davis, and input from Jim West, Steve Rosenblum, and Nat Venkataraman. We are also thankful to Todd St Clair, Robert Lee, Aleksandra Boskovic and Adam Ellison for their support.
Disclosures
The authors declare that there are no conflicts of interest related to this article.
Supplemental document
See Supplement 1 for supporting content.
References
1. Yi, “Display Light Guide Plate (LGP) Report – 2013,” IHS Electronics & Media market, technology and supply chain analysis, (2013).
2. B.-Y. Joo and D.-H. Shin, “Design guidance of backlight optic for improvement of the brightness in the conventional edge-lit LCD backlight,” Displays 31(2), 87–92 (2010). [CrossRef]
3. B.-L. Huang and T.-L. Guo, “Fast optimization method based on the diffuser dot density for uniformity of the backlight module,” Appl. Opt. 55(6), 1494–1499 (2016). [CrossRef]
4. B.-Y. Joo, J. J. Kang, and J.-P. Hong, “Analysis of the light-scattering power of patterned dot material printed on the light guide plate in liquid crystal display,” Displays 33(4-5), 178–185 (2012). [CrossRef]
5. B. W. Lee, M. Y. Yu, and J. H. Ko, “Dependence of the gain factor of the reflective polarizer on the configuration of optical sheets,” J. Inf. Disp. 10(1), 28–32 (2009). [CrossRef]
6. B.-L. Huang and T.-L. Guo, “Integrated backlight module to provide a collimated and uniform planar light source,” Appl. Opt. 55(26), 7307–7313 (2016). [CrossRef]
7. B.-L. Huang, J.-T. Lin, Y. Ye, S. Xu, E.-G. Chen, and T.-L. Guo, “Pattern optimization of compound optical film for uniformity improvement in liquid-crystal displays,” Opt. Laser Technol. 97(6), 254–259 (2017). [CrossRef]
8. S. Li, “Light apparatus comprising a light guide plate with grooves and methods for using the same to direct light,” US Prov. AppNo.62/713,614,2018.
9. M. Quesada, S. Li, W. Senaratne, M. Kanungo, X.-D. Mi, L. Stempin, W. Walczak, T. Carleton, P. Maurey, L. Liu, H. Tadesse, and L. Dabich, “All-Glass, Lenticular Lens, Light Guide Plate by Mask and Etch,” Opt. Mater. Express 9(3), 1180–1190 (2019). [CrossRef]
