GaN on Si Makes Opto Integration Possible

Let's face it. Even though silicon (Si) is a great material for logic, it is not the material of choice for many of the other parts of an embedded system. This is particularly true for systems involving high power, high frequency, or optical functionality.

For those applications, gallium nitride (GaN) is a much better material. It has a wider bandgap, higher breakdown voltage, a larger critical electric field, and higher thermal conductivity. Bandgap is the energy required for an electron to jump from the top of the valence band to the bottom of the conduction band within the semiconductor. Stated another way, it is the amount of energy needed to free an electron so that it can become mobile. It is typically on the order of a few electron volts (eV). It is 3.4eV for GaN and 1.1eV for Si.

This means that GaN devices can operate at higher voltages and higher switching frequencies, handle higher power density, and (perhaps most importantly) provide higher power efficiency than Si devices. GaN is also being used more often for optical components such as LEDs, laser diodes, and optocouplers. This is because GaN has brighter emission characteristics than silicon and other materials. A TechNavio report (purchase required) predicts that the GaN semiconductor device market will grow at a compound annual rate of 18 percent from 2012 to 2016, versus 3 percent for silicon. One area of disagreement about GaN is which substrate to use. The most common ones are sapphire and silicon carbide, but they come with problems, including high cost, poor thermal conductivity, and small wafer size. For example, a six-inch sapphire wafer costs about $400, compared with $15 for a silicon wafer of that size.

High levels of integration are required these days. There are companies that would like to get all the benefits of GaN on Si, but this is not easy. Thermal mismatch and lattice mismatch have limited the effectiveness of growing GaN directly on silicon, because they cause issues such as bowing and cracking, which can lead to device failure. However, whenever a problem presents itself in this industry, an army of people will look for a solution.

Most of the research in this area has centered on applying a seed layer on to the silicon before the GaN is deposited. This seed layer could be aluminum nitride (AlN), which is created by depositing aluminum and then subjecting it to ammonia (NH3 ). During epilayer growth, compressive stresses are generated. It has been shown that the residual stress is dependent on the impurity levels, and that doping can enhance the epilayer's tensile strength. If the tensile strength is comparable to the compressive strength, it is possible to make the GaN epitaxy crack free. (Epitaxy is where one crystalline structure is deposited on to another. In some cases, there is a specific, defined orientation of the two structures.) Patterning substrates by masking or etching the substrate or buffer layer has also been shown to be effective in reducing cracking.

Yield is more of an issue in the power domain than for optoelectronics. The reason is simple. GaN is a defect-laden material. With an LED covering 0.1mm2 and a 10A transistor spanning several square millimeters, it is clear that yields will be higher where GaN use can be minimized. Also, for an LED of this size, redundancy could be built in easily.

Another area of investment is in wafer size. Six-inch wafers are common today, but larger ones are coming and are necessary for further cost reductions. Still, production is beginning to ramp up for LEDs, and the knowledge learned from this could easily make other optoelectronics devices available within a few years. Electronic design automation companies are already gearing up for this.

Are LEDs such a big market opportunity that most of the investment dollars will go into this narrow field, or is this a necessary step to optimize the process?

Related posts:

25 comments on “GaN on Si Makes Opto Integration Possible

  1. eafpres
    May 29, 2013

    By coincidence, today on All LED Lighting (sibling site to Planet Analog) a blog was posted of a new process growing GaN “pillars” on Si and using them as LEDs.  The blog is here:

    Aledia's 3D GaN-on-Si Microwire LEDs

    In the comments I link some other work on this as well.

  2. amrutah
    May 29, 2013

    Brian,  Thanks for the post and covering in brief about the process.

       Si is an element and GaN is a compound.  Does it have a extrinsic or intrinsic semiconductor characteristics?

      Sapphire or silicon carbide are again compounds, how similar are they when compared to the monocrystalline wafer substrates??

  3. amrutah
    May 29, 2013

    Increasing the wafer areas and reducing the die sizes will help improve the Yield.

    “Also, for an LED of this size, redundancy could be built in easily.”

       Adding Redundancy here refers to having some dummy LEDs??

  4. SunitaT
    May 30, 2013

    LED manufactured with GaN on silicon have advantage of low price, large possible substrate. But LED efficiency is measured by multiplying internal quantum efficiency and external light extraction efficiency. Lattice mismatch limits the efficiency by raising the difficulty in achieving high brightness.

  5. SunitaT
    May 30, 2013

    GaN on GaN will give zero lattice mismatch and will have the strength of homogenous substrates. Is there any development happening on GaN wafer generation?

  6. BrianBailey
    May 30, 2013

    Exactly. Given their small size, they could be replicated and then if one is found to not work the other could be used. So for a small area penalty you have doubled the chance of a successful die. This assumes that the failure rate of the GaN part is considerably higher than for the rest of the silicon area otherwise the economics may not work as well.

  7. BrianBailey
    May 30, 2013

    Agreed, and silicon absorbs light. This is another advantage of the AlN seed layer as it provides a barrier to absorption.

  8. BrianBailey
    May 30, 2013

    I don't know of any, but would be interested to hear if any other readers have information about this.

  9. amrutah
    May 30, 2013


       As I understand, GaN is a compound having a semiconductor properties.  We don't get a perfect seed of GaN to start with so that there are large chances of dislocations. On the contrary if we start with the Silicon seed we can grow a uniform substrate and then use epitaxy to grow the higher Bandgap GaN substrate.

    There are still many questions like, 

    Homonogeneous is the bonding? Do Sapphire and Silicon  carbide form good substrates to begin with?

    May 30, 2013

    I gather with the recent years of high-bright LEDs, the basis is GaN technology? We have a project at work where we do not need the high bright LEDs and finding it more difficult to find regular brightness LEDs. Application is in the instrumentation panel for night-vision compatibility.

    I gather that with the high-bright GaN LEDs, the die size shrinks as well to obtain similar brightness as a regular Si LED?

  11. Brad Albing
    May 30, 2013

    @DK – even tho' you don't need the high-brightness devices, you will probably do well to use them. Just operate them at much lower current than you thought you'd need (that can only help overall power draw + heating). And, since everyone uses the high-brightness devices now, they should be priced quite attractively.

  12. RedDerek
    May 30, 2013

    That is what we do. However, at the low currents we need to operate, there is unevenness in illumination between the LEDs. We are driving down to 0.1mA to get the dim lighting, but not flood out the night-vis equipment. We are actually buying LEDs that have been screened for a certain range of illumination at a low current value.

  13. eafpres
    May 30, 2013

    Take a look at the work in my earlier comment (Aledia).  They do not cite efficacy figures but imply the cost of 25% of comparable HB LEDs and due to no phosphor the energy conversion to light is higher.  Also, by having vertical pillars of GaN closely spaced more light per unit area of the Si wafer is produced and leaves the structure than a planar LED.

    I don't know how this will play out but they are commercializing the process on 8 inch silicon wafers, so we should know soon.

  14. Brad Albing
    May 30, 2013

    @eafpres – thanks – good tie-in with Mr. Bailey's blog.

  15. Brad Albing
    May 30, 2013

    @eafpres >>due to no phosphor the energy conversion to light is higher . I like the approach they are taking by doping the microwires with various amounts of indium so that the hopping around of the electronics in their respective shells is occurring right there where it counts, rather than in a separate fluorescing layer of phosphors bombarded by photons. Very sensible.

  16. Brad Albing
    May 30, 2013

    @RedDerek >> buying LEDs that have been screened for a certain range of illumination at a low current value . Ah – did know you were finding a lack of conformity with the LEDs. Sadly, you have to get them screened – so there goes any cost savings you might have realized.

    Or sort them yourself – pro'ly tough to do with the tiny sur-mnt devices.

  17. Netcrawl
    May 31, 2013

    @Brian thanks for the share, GaN is getting more attention these days because its has something unique, this unique material properties possesed by GaN translate into the ability to deliver high power at high frequencies, and best of all it could withstand high temperature. Excellent thermal conductivity also makes GaN a perfect stuff for very high power applications.   

  18. Netcrawl
    May 31, 2013

    @Amrutah you're right there's so many question need to answer here like where are the market opportunities for GaN? I think the possible commercial market for GaN would be the wireless infrastructures, the massive 4G rollout would definitely drive demand for more smaller base station. 

  19. Netcrawl
    May 31, 2013

    @Brad we seen GaN's performance and properties, we have seen it in action, so what makes GaN Good or ideal? GaN is a very hard material with crystal structure, it's bandgap of 3.45 eV suit it in high power and high frequency devices, GaN's properties give it an unmatched porperties and ideal for extreme environments. I believe that GaN's wider bandgap properties give it a great advantages.  

  20. BrianBailey
    May 31, 2013

    In related news out today – NXP has just been awarded some money from the UK government for GaN power semiconductor development. IN the press release they say: NXP is putting its efforts into commercializing 650-V diodes and switches based on a proprietary GaN-on-Si process technology. NXP has stated that it plans to release a portfolio of GaN discrete devices made at the Hazel Grove PowerMOS wafer fab in 2H14. The products will be targeted at applications where energy conversion efficiency is key, such as power factor correction power supplies, solar energy, motor control and automotive electronics.

  21. Brad Albing
    May 31, 2013

    @Brian >>NXP has just been awarded some money from the UK government for GaN power semiconductor development. Interesting – we will all have to keep an eye on NXP – probably lots of info will be coming from them on what they are up to – so lots of blogs to be written.

  22. Brad Albing
    May 31, 2013

    @Netcrawl – That is a good summary of what Brian discussed in his blog, so thanks.

  23. bjcoppa
    June 1, 2013

    Glad to see an abundance of comments on this topic. One of my recent articles on this blog was related to this post. GaN on Si was considered once a dream and a myth. Naysayers are like armchair QBs on a Monday morning. It's easy to scoff at an immature technology 5-10 yrs before it is mainstream as many did in the case of GaN-on-Si. One that come from an electrical engineering perspective often quickly dismiss new potential materials systems for devices. However, materials scientists like myself simply smile and drum up solutions in the lab that surprise and amaze.

  24. bjcoppa
    June 1, 2013

    GaN on Si has apps for both LEDs and high-power, high-frequency devices. However, most companies focus on one or the other. Cree on the other hand covers both devices. It started off as a SiC wafer vendor and eventually became vertically-integrated for GaN HEMTs and GaN LEDs on SiC wafers. Most LED companies use sapphire and most GaN on Si is dedicated to power ICs. The LED focus is more recent including companies such as Azzurro. 3D LEDs via the formation of micropillars on Si have been developed by Aledia. Nanoscale GaN structures on Si reduce the stress associated with the lattice mismatch.

  25. bjcoppa
    June 3, 2013

    Nitronex and International Rectifier have commercial RF power devices based on GaN-on-Si. SiC substrates are many times higher in cost over Si as SiC has been the substrate of choice for GaN over the last 10yrs. However, buffer layers and template transfer methods are being implemented to bridge GaN to Si. The lower cost of Si can be sold in some applications as the benefit for any lower performance  in using this substrate over SiC.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.