A few weeks ago, I had a chance to attend the HiTEN (high temperature Electronics Network) conference in Oxford, UK. While the temperature there never quite made it above 65F the whole time I was there, there were plenty of talks about things working at much higher temperatures 400-600F (These types of temperatures are more appropriate for roasting and broiling than they are for running electronics).
In earlier posts I’ve talked about some of the challenges with running electronics at such high temperatures. Picture a leaky water faucet where the amount of water leaking increases with temperature. As long as the leakage is small, while you may be wasting water in the process, by turning the tap on or off, you still have plenty of control on the amount of water coming out. At some point, if the leak gets big enough, while you can still turn the tap, its effect is swamped by the amount of water flowing through regardless of the tap position. At temperature of 400-600F, conventional electronics tend to be quite leaky that the amount of control one has on the operation of electronics is rather poor. While the operation of the electronics by itself is a challenge at high temperature, that isn’t the only problem. Packaging and interconnect pose major difficulties at these temperatures as well. My Colleague Tan has been looking at packaging issues at high temperature for some time, and she also presented a paper at the conference on some of her work in that area. Here is Tan talking about her work and the conference.
As Emad mentioned, him and I recently returned from 2011 High Temperature Electronics Network (HiTEN) held in UK. This is the 9th symposium organized by International Microelectronics Assembly and Packaging Society (IMAPS), combined with High Temperature Electronics Conference (HiTEC) series held in US in even years. First experiences are always exciting, and this was my first time attending this conference series, my first time meeting this group of experts, and my first time joining into this unique technology group. I find this technology group to be especially exciting, as the goal is set clearly, aiming for high temperature electronics.
The need for exploration and monitoring of down-hole activities in petroleum and geothermal wells has had some interesting outcomes. One is technology that has applications in the automotive and aviation industries and another is that it has driven research in high temperature electronics technologies. Geologists theorize that the formation of the universe 15 billion years ago released an immense amount of energy. Geothermal energy is one of the residual energy sources. In the geothermal applications, deep wells are formed below the crust of the earth and down to upper mantle layer of the earth (Figure-1). As you get deeper into the layers of the earth, the temperature increases by approximately 20-25°C per kilometer (25°C degrees is approximately the difference between a nice warm summer evening in the Northeast and freezing temperatures.)
In our current Department of Energy (DoE) funded geothermal programs, we are working to develop enhanced geothermal sensors and telemetry electronics for operating at 300°C. We are using a wide bandgap material Silicon Carbide (SiC) for this development because at 300C, the leakage in SiC is close to 1 billion times smaller than for conventional silicon.To build functional circuits, the substrate and electronics packaging technologies must also be demonstrated to provide interconnectivity between the SiC devices developed here in our clean room [link to blogs], passive components and the circuit board. We are exploring three different types of substrate as we work to develop these circuits. The three different types are alumina, low temperature cofired ceramic (LTCC) and aluminum nitride. As a warning, I am going to get a bit technical as a talk more about each of these different options. This may not be interesting to all, but I wanted to share the details for those who may also be interested in this, as it is what I spoke about at the recent conference.
Alumina is a widely used substrate with many options for compatible conductor and dielectrics materials. However, it lacks materials screening, process development and reliability evaluation for applications at 300°C. However, Professor R. Wayne Johnson and research assistant Rui Zhang have carried out extensive studies based on the alumina substrate. They have modified the firing temperature profile to improve the metal and dielectric adhesion; they invented cross-over structure (light blue pad in Figure-2) to enable the multilayer capability; and they have characterized electrical performance to ensure minimum leakage from conductor pattern and multilayer dielectric structure. Figure-2 shows the first generation alumina board with SiC device and passive components integrated.
The LTCC effort is carried out at GE Global Research with a supplier’s assistance in fabricating multilayer substrates. The LTCC multilayer substrate fabrication process can be summarized as: via drill, conductor printing, tape lamination and stacks firing. Gold filled vias are used to connect multiple layers, which is different from the cross-over strategy in alumina substrate. The layer numbers can be increased to accommodate more complex circuitry.
Another way to realize the complex circuit substrate is to introduce thin film metallization technology into ceramic substrate. Opposite of thick film (10-30µm thick) conductors and dielectrics in alumina and LTCC, thin film can range from Angstrom to a couple of microns. A human hair is generally 70 µm thick on average, so both thick and thin film processes are dealing with thicknesses much smaller than that of a human hair. The ongoing study of thin film on AlN includes the e-beam metallization and PECVD dielectrics to form multilayer structure. Thin film metallization enable the fine and accurate substrate features of 75µm, which is hard to reach in thick film alumina and LTCC.
So far I’ve described techniques related to making the high temperature circuit board, the next challenge comes with attaching the electronics to the board. There are a few things we’re looking at including gold tin die attach, gold wire bonding and gold flip chip, as shown in Figure-4. Combined with the substrate technology, the Au bonding system provides the capability and stability for continuous operation at 300°C, but we’re also looking at cheaper options.
If you’d like to like to learn more about this work, please take a look at papers listed below.
Reference:
1. R. Zhang, R.W. Johnson, D. Shaddock, T. Zhang and V. Tilak,, Characterization of Thick Film Technology for 300°C Packaging, 2010 High Temperature Electronics Conference, Albuquerque, NM, May 11-13, 2010
2. K. Fang, R. Zhang, R.W. Johnson, E. Andarawis and A. Vert, Thin Film Multichip Packaging for High Temperature Digital Electronics, 2011 High Temperature Electronics Network (HiTEN), Oxford, UK, July 18-20, 2011
3. T. Zhang, D. Shaddock, A, Vert, R. Zhang and R.W. Johnson, Characterization of LTCC-Thick Film Technology for 300°C Packaging, 2011 HiTEN, Oxford, UK, July 18-20, 2011
Acknowledgment:
This material is based upon work supported by the Department of Energy Geothermal Technologies Programs DE-FG36-08GO18181 and DE-EE0002755.
