«

»

Jan 03

Happy New Year and Some Crystal Ball Gazing


Admittedly rather dusty and rusty after not being used since early 1999 for a panel discussion looking at the prospects for 25nm lithography at the SPIE Microlithography* conference, I found it at the back of my wardrobe. So I thought maybe it would work better this time if I restricted the search to the next 12 months…. Well, I peered and peered and slowly the smoke (and grime) cleared and the following appeared. So here are the www.ReRAM-Forum .com predictions for 2013!

  • No high density ReRAM/CBRAM based products in 2013
    There’s life still in NAND Flash memory and given the problems that PCRAM (Phase Change RAM) has had in gaining a foothold in the market place, I feel fairly safe with this prediction. However, I fully expect that products with a low density embedded ReRAM/CBRAM memory to appear, at least for sampling, from one or more companies.
  • Cells based on metal oxide bilayers will become the structure of choice for ReRAM
    There have been some impressive progressive recently in ReRAM cell performance showing highly non-linear IV characteristics in both high and low resistance states. I’d point to this as the most important advance of 2012. The elimination of the need for a separate select device in each cell will have a major impact on manufacturability. Similarly, the simpler the metal oxide(s), the better.
  • Consolidation in the ReRAM based next generation memory field
    When I started this Blog I did a survey of the small companies in the field. I was surprised by the number and the market place cannot support the growing pains of so many start-ups. I fear some will fall by the way-side but I also expect that some will be purchased and merged successfully into larger organizations.
  • EUV lithography will be delayed
    Again, I feel fairly safe with this one. In fact I was beaten to it by Luc van Hove from IMEC at the recent IEDM conference in San Fran. As described in previous Blogs, the technical challenges of EUV are immense specifically at present with the source power. If/when a solution is found, I fear that there will be system integration type issues that will need to be debugged related to the operation of the source, masks, optics and vacuum under volume manufacturing conditions.
  • www.ReRAM-Forum.com will celebrate its first birthday
    Well I certainly hope so!
  • Christie Marrian, www-ReRAM-Forum Moderator.

    *Emerging Lithographic Technologies III, Yuli Vladimirsky, Chair, part of SPIE Microlithography 1999, Santa Clara, CA, 3/14/1999. The panel session was lively but alas was ‘off-the record’ and did not make it into the Conference proceedings, proceedings.spiedigitallibrary.org/volume.aspx?conferenceid=1717&volumeid=9269

    1 comment

    1. TinyTechIP

      1) No high density ReRAM/CBRAM based products in 2013

      Agree

      2) Cells based on metal oxide bilayers will become the structure of choice for ReRAM

      The start-up Crossbar Inc. seems to be supporting using amorphous silicon as a ReRAM material and I have also seen some work on SiO2 ReRAM. These materials have more obvious advantages for large scale manufacturing and integration with CMOS. There has been a lot of work on bilayer metal oxides resistance switching over the past ten years starting with work by Alex Ignatiev at the University of Houston but the manufacturing is more difficult and sneak path problems may persist in larger arrays. However, if the bilayer metal oxides do become standard Samsung has a very basic patent (US Patent 7417271). So if this prediction is correct it could give Samsung some control over the ReRAM industry.

      3) Consolidation in the ReRAM based next generation memory field

      This is definitely something that is desirable. Unity was already aquired by Rambus but I am unsure any of the other startups have the growth potential to be attractive in 2013 (later perhaps yes). It is also notable that most of the major industry players (e.g. Samsung, Sharp, Micron, Sandisk, Panasonic) already have substantial R&D and patents covering various versions of ReRAM. It seems unlikely they would purchase a start-up unless it provided something necessary that they lacked.

      Great blog by the way.

    Leave a Reply

    Your email address will not be published. Required fields are marked *

    You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>