A team at the University of California, Riverside Bourns College of Engineering has developed a method to build the next generation of memory storage devices for portable electronic devices including smart phones, tablets, laptops and digital cameras.
The device is based on the principles of resistive memory, which can be used to create memory cells that are smaller, operate at a higher speed and offer Terabytes of storage capacity over the current flash memory cells (Gigabytes).
The key advance is the creation of a zinc oxide nano-island on silicon. It eliminates the need for a second element called a selector device, which is often a diode.
Jianlin Liu, a professor of electrical engineering at UC Riverside said, "This is a significant step, as the electronics industry is considering wide-scale adoption of resistive memory as an alternative for flash memory. It really simplifies the process and lowers the fabrication cost."
Resistive memory has the potential to lower cost, improve performance and offer greater storage. There is no rare earth element dependence and the materials used are readily available. I had the opportunity to learn more about resistive memory in an exchange below with Jianlin:
Please explain how the costs in fabrication are reduced? Is it because the selector device (diode) is eliminated?
Yes, the cost that is otherwise used to fabricate additional selector diode is saved.
Is the time to manufacture reduced?
It is correct that the time to fabricate one memory cell would be saved due the fact that the time to fabricate an additional diode is not necessary. For actual chip fabrication, all devices are fabricated simultaneously. We used a special approach to demonstrate the concept that ZnO [zinc oxide] on silicon would work; however, we have not used industry-standard process, therefore I am afraid that I could not comment on the actual time to fabricate a chip. One thing is sure: the time to fabricate one device and all devices on a chip is the same if the industry-standard lithographic and etching process is used.
Are any rare earth elements eliminated in the fabrication process?
Yes and that's because no rare earth elements are used. Zinc oxide and silicon are abundant in nature, environmentally friendly and cost effective.
What kind of anticipated cost reductions--per device, per 100, 1,000 devices etc. can be expected?
Again, the claim in the report regarding the cost is a general true claim, and I acknowledge that the actual cost reductions are not clear. However I agree this is a good question--in particular, it has to be answered very clearly to a venture capital guy.
What are the vulnerabilities to the resistive memory devices? At first glance I would think heat, static and power.
Sure, there shall be some vulnerabilities to this type of memory, I should say that heat is a problem for most memories, so it is not a particular problem for resistive memory, which uses redox reaction as its operational mechanism. If resistive switching uses phase change as its mechanism, then excess heat may very much be a vulnerability for that kind of memory. In terms of power, for resistive memory, writing voltage can be scaled to be very small amplitude; therefore writing energy can be very small, for example, less than 10 to the 17th joules per bit. So power can been saved, which is a good thing. But then, yes, read voltage shall be also small, so static might be posing a problem on this technology--so I would think that small read voltage for such kind of memory might indeed be a vulnerability.
As for simplification of the "process," can you explain?
The process would be just putting zinc oxide on silicon. There are multiple ways to do it. In our case, we self-assembled zinc oxide nano-island on silicon substrate.
What does the power savings really mean? Does it mean longer battery life or longer battery life and reduced energy consumption?
Sure, if we no longer need to use high voltage (that would be needed in other memory technologies such as flash) to write and read the resistive memories, it certainly means the energy in an external battery will be consumed [over a] much longer [period of time].
Conclusion
There is a cost associated with these devices and the concern I have is, will consumers continue the purchasing cycles of smartphones? Costs need to come down and not stay the same or increase. While mobility is highly regarded, I think there is no regard towards reasonable budgets, and without subsidies to the consumers where would we be today?
Resistive memory if successful can prolong battery life and offer more storage at a cheaper cost. But for the overall cost of the smartphone or other devices to lower, elimination or minimizing the use of rare earth elements and reducing manufacturing costs remain key. The other consideration is life cycle--how long do smart phones really last? Consider the churn rate and then the true cost of owning one, two or three or more smart phones along with contracts for cell minutes, data plans and texting; are we getting value?
Lastly, has the PC market eroded and been displaced with tablets, smart phones and iPads? I ponder over these questions because I can't help but wonder whether we are on a sustainable path or not, and then given all the planning of resources around BYOD, are we over planning, under planning or on target?
Follow Matt Brunk on Twitter and Google+!
@telecomworx
Matt Brunk on Google+
