Perfect substrate within reach for wide-bandgap materials
The lack of a native substrate has led to a variety of choices for those developing GaN-based devices. Jon Newey looks at what is currently in use and what the future developments might be in this fast-moving area.
Any attempts to advance semiconductor technology rely heavily on the availability of a material that can be used as the platform upon which the end product is manufactured. The success of the silicon and GaAs industries is largely due to the availability of high-quality native substrates that can be successfully scaled to larger wafer diameters, providing economies of scale. Wide-bandgap semiconductors, especially GaN, have been the subject of intensive research for several decades now. The lack of a suitable GaN substrate was a factor in delaying the development of blue and UV LEDs and lasers, and high-power, high-frequency electronic devices. It has also forced those wishing to exploit the benefits of GaN to look elsewhere for a substrate.
Currently, sapphire and SiC are the most popular substrate choices, with work on the use of Si progressing well. However, while each of these materials has its advantages, it has taken much effort to grow GaN layers of the quality necessary for commercially viable devices. There have been some notable commercial successes, such as GaN-based blue LEDs grown on sapphire and SiC, but such successes have not stopped the search for bulk GaN substrates. These are now becoming available, albeit at rather small sizes, in low volumes and at high costs.
SiC substrates
After sapphire, SiC is the most commonly used and well developed substrate for GaN epitaxy and of course SiC homoepitaxy. It forms the basis for high-power, high-frequency electronic devices and high-brightness blue and green LEDs. Single-crystal bulk SiC is generally produced using physical vapor transport (PVT) techniques, where the starting materials sublime and crystallize on a seed. Each manufacturer using PVT has its own proprietary variation on the technique. However, there are exceptions, for example Okmetic in Finland uses a high-temperature CVD method to grow bulk SiC directly from gas-phase precursors.
The 6H and 4H polytypes of SiC are available in 50 and 75 mm diameter wafers in production volumes. The need to lower costs by increasing wafer diameters is as pressing for the wide-bandgap semiconductor industry as it is for the Si and III-V areas. The diameter of SiC wafers is increasing at a higher rate than that of Si and GaAs wafers. In fact, SiC has moved from 25 to 100 mm wafers in half the time that it took Si or GaAs to make the same leap.
A roadmap for SiC wafer development was discussed at the 2001 ICSCRM meeting (see Compound Semiconductor January/February 2002, p59). Here, the development of 150 and even 200 mm wafers was discussed, with the expectation that they would be available in 10 years' time. However, such a roadmap has a number of boundary conditions. First, does the capability to make larger wafers exist? Second, can the larger wafers actually be used? And finally, is there a market for them? Although MOCVD reactors can take large wafers for GaN deposition, manufacturers of tools for SiC epitaxy will need to develop reactors that can accommodate larger wafers and successfully grow on them. As Rob Glass, general manager and VP of Cree's SiC materials area points out, "Fewer people will buy larger wafers if they can't put a SiC epilayer on it. Three inch material is out there now being tested and prototyped with 100 mm a year, perhaps two, behind."
What the market needs it will get
Nippon Steel announced in 2001 that it had developed 100 mm SiC wafers and was expecting to have these in volume production by 2005 (see Compound Semiconductor November 2001, p11). It seems to be the case that companies involved in SiC wafer manufacture are being patient, and waiting for a true market to develop for larger diameter products rather than pushing a technology onto a market that is not ready.
However, supply helps to create demand. "If somebody comes out with a cost-effective, high-quality 100 mm substrate today, I think that would help push the market ahead. Nobody needs to wait until 2005," said Glass. "Cree demonstrated 100 mm wafers at the 1999 ICSCRM meeting, and market growth will help to justify volume production of these wafers. As always, those willing to do the early work with devices on large-area substrates will be in a better position when the market develops."
Micropipes: the killer defect
Possibly the most actively researched area of SiC development has been the reduction of micropipe defect density. Micropipes are voids that lie at the core of screw dislocations, reaching the surface and appearing as holes. Any device fabricated on an area in which a micropipe breaks the surface will be unlikely to work. It is this effect on yield that is driving the work to reduce micropipe density. The specifications in commercially available material are in the 10-100 micropipes per cm2 range, depending on wafer size, grade and how much the customer is willing to pay.
While efforts continue to eradicate micropipe formation during bulk growth, techniques have also been developed to fill the pipes. Subsequent SiC epilayers demonstrate greatly reduced or zero micropipe density, significantly enhancing device yields. However, such additional steps add costs to the production, and so the quest to eliminate micropipes during bulk growth will continue.
The user has to establish an acceptable balance between yield and cost per unit area of the wafer for their particular needs. If the yield reduction from the micropipes is minimal, then getting to the zero level may be unnecessary. However, in future some very-high-power devices will be of such a large area that a single device will occupy an entire wafer. Here, a single micropipe may prove fatal.
High-power, high-frequency devices
Essential to the operation of RF devices is the use of semi-insulating substrates. This adds another level of complexity to substrate manufacture as the material is usually made semi-insulating by doping with vanadium, which has limited solubility in SiC. Semi-insulating SiC is generally only available as 50 mm substrates. Although 75 and 100 mm semi-insulating material has been demonstrated, there is little market pull for this at present.
In the absence of III-N substrates, SiC is practically the only choice as a substrate for GaN-based power devices. It is no coincidence that the best-performing GaN power devices are those grown on SiC (see Compound Semiconductor November 2001, p69). SiC provides a smaller lattice mismatch with GaN than sapphire (3.3% for GaN versus 14.8% for sapphire) and has a higher thermal conductivity than sapphire and even GaN itself. However, this performance gain comes at a high price compared with sapphire substrates, and has led to some novel attempts to retain sapphire as the substrate, including thinning of the sapphire and bonding to a high thermal conductivity heat sink (see Compound Semiconductor December 2001, p43).
Gallium nitride
In an ideal world, those growing GaN epilayers would have a high-quality GaN substrate to grow on. No complicated buffer layers would be needed to account for the lattice mismatch between the GaN epilayer and a foreign substrate, and no differences in the mechanical and thermal properties of the substrate and epilayers would need to be taken into account for device processing purposes or the end application. With this in mind, there is much activity around the world aimed at providing the GaN community with GaN wafers.
Bulk GaN material can be divided into two categories. The first is GaN produced by depositing a thick layer of GaN onto a foreign substrate such as sapphire or GaAs. The substrate is then removed, leaving behind a free-standing GaN wafer. The second category of GaN is produced by the growth of GaN boules. To some degree, the two go hand in hand. To grow a boule, a seed crystal is required and wafers produced by substrate removal can provide this seed.
"We see free-standing GaN substrates as a step in the boule growth process, but we also sell the free-standing GaN substrates for GaN epitaxy," explained George Brandes, director of GaN operations at ATMI in Danbury, CT. Although a number of companies are sampling GaN wafers produced through boule growth, the production technology is relatively immature, making yields and production volumes low. As a result, such wafers are expensive and diameters are small compared with the free-standing wafers or sapphire and SiC.
Despite the ability to produce free-standing GaN wafers, Brandes and his counter-parts in other companies recognize that in the longer term it will be more cost effective to go to a boule growth process that will also yield better quality material. TDI of Silver Spring, MD, is also developing GaN boule growth techniques. "We think that for scalability and the lowest defect density, our approach [boule growth] has a much better future than the free-standing approach," said Vladimir Dmitriev, TDI's president and CEO.
GaN wafers certainly have a way to go if they are to become accepted in the commercial world. Commercially successful GaN-based devices such as blue LEDs are being manufactured in large numbers on relatively cheap sapphire substrates, and GaN-on-Si LEDs have also been demonstrated. While low- to medium-brightness LEDs are remarkably tolerant of crystal defects, blue and UV lasers are less so.
To provide epitaxial GaN of the quality necessary for lasers, researchers have adopted techniques such as epitaxial layer overgrowth (ELO) on sapphire substrates. ELO typically involves growing through openings in a silicon nitride mask. Growth takes place vertically through the openings and then laterally across the masked area. The laterally grown material has a greatly reduced level of defects. However, this adds to the complexity of laser manufacture, and laser grade material is only present in the overgrown regions of the wafer. If large-area GaN substrates were available, achieving high-quality material for laser manufacture would be much simpler.
"We believe that single-crystal, low-defect-density GaN will be the substrate of choice for laser diodes, and maybe UV and white LEDs, once the price approaches that of SiC," said Dmitriev.
Pushing the use of GaN substrates into applications other than lasers, such as LEDs and RF power devices, will require further reductions in price and an increase in wafer size. Those working on bulk GaN seem confident that this is achievable, though getting anybody to put numbers and times to their goals is understandably difficult. "There is nothing in the intrinsic costs that prevent us getting to SiC-like price levels in a reasonable time," said ATMI's Brandes. "It's a matter of getting the process streamlined enough and yields high enough. I find it hard to believe that we won't be there in four years and maybe in as little as two years."
Aluminum nitride
Many of the companies working on GaN boule growth also have programs for AlN bulk growth. AlN is highly resistive, has a higher thermal conductivity than GaN (but less than SiC) and provides a lattice match to high Al content AlGaN. These properties make it an ideal substrate for high-power, high-frequency devices, and UV emitters and detectors. However, bulk growth of AlN is at a similar stage to GaN in its development. Because of the possible military applications of devices grown on AlN and GaN substrates, the US Department of Defense has provided funding to companies such as Kyma (Raleigh, NC), Cermet (Atlanta, GA) and Crystal IS (Latham, NY) to accelerate the development of larger substrates
Zinc oxide for blue LEDs?
Unlike GaN, conducting ZnO 50 mm wafers are available and provide a good lattice match to GaN with a mismatch of about 2%. However, it is not just its compatibility with GaN that has stimulated interest in ZnO, but also the possibility of using ZnO itself as a material for blue and UV emitters. The long-running patent disputes in the GaN-based LED world have provided a strong motivating force for those pursuing ZnO as a possible alternative to GaN for short-wavelength emitters. ZnO is naturally n-type, and growing epitaxial p-type layers to form pn junctions has proved to be a stumbling block. However, with the announcement earlier this year by Eagle-Picher that it had successfully grown p-type ZnO layers by MBE, actual devices are a step nearer.
Cermet has development activities in a number of wide-bandgap substrate materials, and the company is growing ZnO using a melt growth process. Jeff Nause, Cermet's president, sees this growth method as the main reason for ZnO's advanced state of development compared with bulk GaN growth. "Zinc oxide growth will scale more rapidly than any of the bulk nitrides, including our own GaN growth," said Nause. "GaN uses a different growth technique that is inherently slower than the ZnO growth process. GaN growth will scale, but it will take longer to do so."
Despite its apparent suitability for GaN epitaxy, there appears to be some inertia to using ZnO as a substrate. Nause suspects that much of this reluctance comes from articles in the literature that suggest good-quality GaN can't be grown or is difficult to grow on ZnO. This results in the desire to wait for larger GaN substrates.
Silicon muscles in
In common with other compound semiconductor material systems, GaN is also being grown onto Si. The development of GaN growth on Si is not just confined to university research departments. A number of companies are pursuing this, including Nitronex of Raleigh, NC, which has demonstrated GaN-based electronic and optoelectronic devices grown on Si wafers (see Compound Semiconductor November 2001, p63).
Silicon wafers are cheap (as little as 10% of the cost of sapphire), widely available in a range of diameters, compatible with standard semiconductor processing tools, and the use of Si raises the possibility of monolithic integration with Si microelectronics. Si also has some specific advantages when the types of devices to which GaN can be applied are considered. Si wafers are available in conducting and high-resistivity forms, making them suitable for LEDs and high-frequency electronic devices, respectively. Si has a similar thermal conductivity to GaN, making heat dissipation less of a problem than with sapphire substrates.
Unfortunately, the properties of Si mean it is harder to grow GaN on Si than on sapphire or SiC. The lattice mismatch to GaN is 17% and the difference in the coefficients of thermal expansion is 100%. This results in the cracking of GaN layers that are thicker than about 1 µm. Si and Ga also react at high temperatures. The literature is full of exotic buffer-layer materials deposited using a range of methods and temperatures. These are designed to engineer out the strain that causes cracking, and work with varying degrees of success. Also, AlN seed layers, AlN interlayers, patterned substrates and Si nitride mask layers to promote ELO are being investigated for the growth of thick, high-quality GaN layers. However, recent reports indicate that such efforts are starting to meet with some success and device results are being published, though some of these methods, such as Nitronex's Sigantic growth technology, remain proprietary.
The big question is, does the extra effort required to successfully grow on Si make the process cost-effective despite the low-cost starting material? The answer will depend on the application and if market demand is sufficient to warrant the use of large-area substrates. Any performance trade-off in going to Si must also be considered. Si will not remove heat from a power device as well as SiC, but it will perform sufficiently for medium-power devices, and also lends itself to substrate thinning processes. Si absorbs blue, green and UV wavelengths, a point of concern to those wanting to manufacture GaN-based LEDs at these wavelengths. Si process technology has been refined and perfected by the microelectronics industry, meaning that the material can, with relative ease, be subjected to additional process steps to improve light extraction.
Room for all
The substrate world is evolving quickly. While the absence of a native substrate for GaN may have hampered progress, it has certainly resulted in the development of a variety of substrate options. With GaN substrates starting to appear, the landscape will continue to change. Different types of GaN devices will suit the use of different substrates. The eventual widespread availability of a bulk GaN substrate will not necessarily mean that all GaN-based devices will be grown on GaN substrates. "GaN, sapphire and SiC will always share the market for GaN-based devices," said TDI's Dmitriev. "For low- to mid-brightness LEDs, sapphire is good enough. SiC is a good thermal conductor, so it may continue to occupy a niche in high-power devices even when high-quality GaN and AlN substrates are widely available."
About the author
Jon Newey is the technology editor of Compound Semiconductor.
compoundsemiconductor.net |
