GLASS SUBSTRATES PLAY A vital role in nearly all electronic displays. For TFT-LCDs, the two sheets of glass provide electronic, optical, and mechanical functionality. One sheet of glass serves as the substrate for the creation of the active matrix – an array of thousands of TFTs, typically silicon-based semiconductor devices, in addition to transparent conductors, typically fabricated from indium tin oxide, to connect these transistors to drive circuitry. These semiconductor processes require pristine surface quality, high temperature resistance, and specific chemistry. Color filters are deposited on the other piece of glass, which also requires a high-quality surface. From a mechanical standpoint, the glass sheets form a space that maintains the liquid-crystal material at a fixed thickness (cell gap), serve as the base for polarizer sheets and other optical films, and provide the mechanical stability for the display panel.

For active-matrix TFT-LCDs, non-alkali glass is used for both layers rather than the alkali substrates found in passive-matrix LCDs since the sodium ions and heavy metals found in alkali glasses can cause problems in the fabrication of TFTs. Such problems include unstable gate voltages and increased current leakage, which affects TFT performance and causes defects. This non-alkali glass must be produced cost effectively at large sizes in thin sheets, with low density to minimize weight, an ability to withstand high temperatures with minimal thermal shrinkage, resistance to harsh chemicals and lateral cracking during scribe and break processes, freedom from bubbles, and a low coefficient of thermal expansion. The non-alkali glass-substrate market was pioneered by Corning, which developed the first a-Si TFT boro-silicate glass substrates. There are four major methods for fabricating non-alkali glass substrates.

Fabrication Methodologies

The float method, pioneered by Pilkington, has been used for architectural and other applications that require lower-quality glass and has been modified for LCDs by Asahi Glass. In the float method, glass is melted and then flowed into a chamber featuring an underlying tin layer. Under the correct pressure and temperature, the glass effectively floats on the surface of the tin. After the glass reaches the desired thickness, it goes into the cooling chamber and is then pulled out from the other end, cleaned, dried, and scribed. This approach is scalable and economical for large substrates but requires polishing (as the process results in planar non-uniformities that adversely affect image quality), which offsets some of the benefits.

An approach that can be implemented separately or in conjunction with the float process is the redraw method. In this approach, the molded glass is reheated, redrawn, and remolded into the desired thickness. This approach is quite flexible because the material properties can easily be reconfigured. It also requires polishing to enhance surface planarity.

A third approach is the down or slot-draw approach in which molten glass is poured into an agitator. The glass is then drawn downward through a platinum orifice or slot with a specified stretching force, then drawn through rollers into a cooling chamber. By changing the orifice or slot, different substrate thicknesses can be produced. This method causes minute surface irregularities derived from the uneven nature of the slot surface edges. As a result, polishing is required. This approach is not as economical as the float method. Nippon Electric Glass has adopted this approach.

A fourth approach, pioneered by Corning, is the fusion method. It is a variant of the down-draw, but bypasses the harmful effects of the orifice. In this approach, over-flowing molten glass flows into a trough-shaped fusion pipe from both sides and is joined into a single sheet as the glass is pulled downward. The substrate surface does not contact metal rollers or guides during film formation. As a result, an improved surface is produced and polishing can be omitted, reducing costs. Thinner glass increases throughput and lowers cost. Fusion tanks do not require as large an investment as do float tanks, resulting in higher utilization and increased flexibility. On the other hand, float tanks can produce substrates that are significantly wider than those from fusion tanks, giving this method an advantage for larger substrates.

The market for glass substrates is closely related to the overall TFT-LCD market, particularly the applications that dominate area production of TFT-LCD panels: TVs, desktop monitors, and mobile PCs. Slowing demand for TFT-LCDs in 2011, combined with a sizeable glass inventory left over from 2010, caused a significant reduction in the growth of total demand for glass in 2011(Fig. 1).

Fig. 1: Demand for glass substrates grew rapidly over the past few years as TFT-LCD manufacturers expanded TV panel production, but market growth slowed in 2011 and is expected to continue to do so. Source: DisplaySearch Quarterly LCD Glass Substrate Report.

The challenge for glass makers is to anticipate TFT-LCD production investments and make appropriate investments in melting tanks and to moderate production as TFT-LCD factory utilization changes. This must be done for each generation of TFT-LCD fabs, since each requires a different substrate size. While TFT-LCD area production has been growing by roughly 20% per year over the past few years, the financial crisis of 2008 caused a significant disruption in production as TFT-LCD makers temporarily closed some fabs. This led glass makers to shut down their production, and as a result there was a slowdown in capacity growth in 2009 (Fig. 2).

Fig. 2: After slow growth in capacity in response to the global recession, glass production grew rapidly in 2010 and grew faster than demand in 2011, leading to oversupply. Source: DisplaySearch Quarterly LCD Glass Substrate Report.

Because re-starting production requires many weeks of increasing the temperature of the melting tanks, a period of glass shortage followed. Glass makers increased capacity dramatically in 2010, and by 2011 were running ahead of demand.

The glass market is characterized by a high degree of concentration, with four companies (including Corning and its joint venture with Samsung) accounting for 95% of production (Fig. 3).

Fig. 3: Corning, along with its joint production venture Samsung Corning Precision (SCP), leads the production of glass substrates (figures are for Q2 '11). Source: DisplaySearch Quarterly LCD Glass Substrate Report.

Asahi Glass Company (AGC) has expanded its market share by expanding production in Korea and Taiwan, and Nippon Electric Glass (NEG) has invested heavily in Japan, as well as in Korea via a joint production venture with LG. AvanStrate (formerly NH Techno) is a smaller producer, and LG Chemical, which has licensed technology from Schott, is also developing production. Glass is typically produced close to or even in conjunction with the TFT-LCD fabrication facility that will use it. However, due to the history of the evolution of TFT-LCD production in Japan, combined with the strong presence of Japanese companies in the glass market, there is a higher level of glass production than consumption in Japan; while 37% of glass capacity is in Japan, only about 12% of TFT-LCD capacity is there. Some glass is exported to Korea, which also accounts for 37% of glass production but 48% of TFT-LCD production, and Taiwan, with 25% and 35%, respectively.

The expected growth of TFT-LCD production in China is leading the established glass manufacturers to develop production in that country, starting with finishing or polishing facilities. This growth is also opening up the possibility of Chinese companies entering the market. Three such companies – Irico, Xufei, and CNBM – have installed glass tanks for Gen 4 and 5 substrates; another, a joint venture of the Baoshi group and Dongxu, is called Xuxin.

New Directions for Glass: Thinner and Stronger

There have been many efforts to develop substrates with materials other than glass, including plastic, metal, and other materials. While the goal of developing flexible displays is often mentioned as a key motivator for moving away from glass, there is uncertainty regarding the market for flexible displays. In the near term, lighter weight and resistance to breakage are key benefits offered by such materials.¹ Despite being a mature technology, the form and features of glass continue to evolve. Two of the most important aspects of these developments are thin glass substrates and rugged cover glass.

Using thinner glass substrates can enable reduction in the weight of the display and eliminate expensive panel-thinning techniques used for displays in mobile applications. Of equal importance, producing thinner glass allows glass makers to increase output without making investments in new tanks. While most glass substrates produced are 0.7 mm thick for Gen 6 and larger and 0.5 mm for smaller substrates, glass makers have begun producing thinner glass, 0.5 and 0.4 mm, respectively, with 0.3 mm a possibility. One factor limiting adoption of thin glass is that panel makers need to adopt new glass-handling equipment because thinner substrates have less stability.

The trend in thinner glass production has led to the possibility of "flexible" glass. This is glass that is 0.1 mm or even thinner, which allows for a small enough bend radius that it can be shipped in roll form and potentially used in roll-to-roll production. While there is no commercial production of displays using such thin glass, it can also be used in applications such as touch screens and cover glass for OLED encapsulation.

The development of strengthened cover glass has been an important source of growth for the glass industry. Since the introduction of the original iPhone, device manufacturers and consumers have appreciated the benefits of this technology, which is scratch resistant, optically clear, and capable of withstanding high stress levels. In addition, it can integrate a touch sensor and also serve as part of the case and industrial design.

There are different methods for strengthening glass, including physical strengthening, which involves tempering the glass by heating and then rapidly cooling. For the thin glass used in displays, chemical strengthening is typically used. In chemical strengthening, a piece of display-quality glass (typically aluminosilicate, although soda-lime is also used) is immersed in a bath of liquid KNO₃, which has the effect of replacing sodium ions in the glass with larger potassium ions, creating stress in the glass, which needs to be exceeded in order for cracks to form.

Since chemical strengthening makes the glass impossible to cut, it must be cut to shape before strengthening, typically by mechanical or laser-based scribing. Because distinctive shapes and holes are often required, computer numerical control (CNC) is used to control the tool and grind edges smooth, removing flaws, cracks, or chips that can cause breakage under stress. Going beyond flat surfaces, cover-glass makers are developing curved and shaped products. So-called "2.5-D" and "3-D" shapes introduce curvature in one or both axes of the glass surface, enabled through techniques such as hot forming and mold pressing.

In addition to the design and durability improvements enabled by cover glass, it is in most cases required for the use of projected-capacitive touch sensors. Since this is the most prevalent touch technology, cover glass is enjoying rapid growth in smartphones and other mobile devices. In addition, the explosion of tablet PCs, all of which use cover glass – and have much larger areas than phones – has also provided a boost. In 2011, 755 million devices will ship with cover glass, for a Y/Y growth of 146% (Fig. 4). Mobile phones account for 84.4% of the shipments.

Fig. 4: Growth in smartphones and tablet PCs has provided a boost to the market for strengthened cover glass, which will be in 755 million devices in 2011. Source: DisplaySearch 2011 Cover Glass Technology and Market Forecast Report.

The Future of Display Glass

The year 2011 has been a challenging one for glass makers, as demand slowed significantly, and combined with falling prices, led to a drop in revenues in the display glass market. While demand is expected to recover in 2012, it is likely that there will be continued excess capacity. With some of the demand growth coming from new TFT-LCD fabs in China, glass makers will need to make investments in production and/or finishing facilities in China. At the same time, the move towards thinner glass means that more area can be produced from existing tanks. With demand for strengthened cover glass growing, glass makers will need to skillfully manage the timing, location, and product type of their production capacity.

Glass makers also face potential shifts in technology. With the growth of AMOLED displays, the potential exists for elimination of one sheet of glass if AMOLED makers are able to perfect thin-film encapsulation, or even both sheets of glass if this can be combined with the production of AMOLEDs on metal or plastic substrates. Other types of flexible displays are also in production (such as electrophoretic) or envisioned. And there are many types of plastic materials that attempt to compete with cover glass.

Can any of these developments unseat the central role of glass in flat-panel displays? While it may be possible in the long run, it is unlikely during this decade. It is difficult to replicate the combination of optical, electrical, and mechanical properties offered by glass. In the case of backplanes, few materials have the ability to be mass-produced with the pristine surface quality and chemistry that allows high-yield TFT production. For applications such as OLED displays, glass also provides a very high barrier to oxygen and water vapor along with high transmission. Other objections to glass – weight and rigidity – are being addressed by the development of thin glass, though there will be intense competition between glass and other materials if the flexible display (or lighting) markets were to take off. For now, there is no other material that can be produced in the hundreds of millions of square meters per year, at the specifications required, to meet the needs of the flat-panel-display industry.