The future of Printed Electronics (in a short tetralogy!) or The Internet of Things (IoT) and how the proliferation of low cost, low profile microelectronics is coming to an item near you….part 3.
In our first two blogs in this series on printed electronics we argued that plays which envisaged replacing complex silicon-based integrated circuits (ICs), such as RFID chips, were doomed. But one area where printed electronics offers real promise is when the required function is very simple. Functionality drives transistor or gate count; typically the greater the functionality then the higher the number of gates. So, for example, a simple HF RFID chip with no security protocols and write once (perhaps a unique number) capability requires about 10,000 gates. If you want a high security HF chip for contactless payment then you’re into the hundreds of thousands of gates. And if you’re interested in a core chip for an iPad then you’re looking at something with 3B gates which today can be squashed into an area that is just 1 cm square! An issue for printed electronics is that with feature sizes are on the scale of 10’s of microns (versus nanometers for silicon) everything requires ~3 orders of magnitude more material and therefore many orders more of space (because additionally it’s hard to stack features when you’re printing). If you planning on printing Apple’s chip then you’ll need to make sure that you have a football field available - that’ll make for an iPad which might be better called an iStadium! But silicon does have a couple of potential weaknesses which offer some hope to those looking to print logic. The first is that if the functionality is low, as in the case of a backplane for a display or a very simple RF license plate or a very low memory device, then silicon’s feature size is not an advantage because shrinking the size of device below ca. 250 micron square would make it too hard to handle….so although a chip could be much much smaller its size and cost are artificially higher to allow downstream processing - this means that printed approaches can compete. The other area of weakness is if the silicon device demands many connections i.e. a high input-output or pin count. Now the silicon size and therefore cost is driven artificially higher just to make room for lines coming in. Printed electronics is conversely relatively easy and cheap to distribute. A good example of where this may an advantage is for a smart blister pack (see image) which requires many connections (to each blister) to record an opening event - expensive if each connections runs to silicon. But each event could be recorded with a simple printed device and this could make for both a cheaper and smarter package.
Replacing silicon will happen but only in very specific use cases - Flex R&D can help you figure out when and where this will happen.
In our last blog in this series we will look at where printed electronics is getting most traction and some of the possibilities it opens up in the next year or so.
I'm busy working on my blog posts. Watch this space!