Breaking News
Blog

Phase Change Memory: The Discontinuity & Melting Question

NO RATINGS
Page 1 / 2 Next >
View Comments: Newest First | Oldest First | Threaded View
Ron Neale
User Rank
Author
Re: Un-scientific opinions PCM and ReRAMs
Ron Neale   3/3/2016 12:00:50 PM
NO RATINGS
rbtbob: With respect to the ECD hearing and your comment on the universe of memory. Couple of points, amongst all the original 3DXPoint (TM) now Optane(TM) announcements I think it was Intel who specifically stated they had an Ovonic switch in series with the memory device, I suppose that will have a link back to Ovonyx.

On the matter of dealing with the entire universe of alternate memory technologies and linking them together. For one of my lectures I did provide the students with a family tree that linked all the NV memory technologies back to PCM (It did not include Flash). I suppose I could submit a copy of that to the judge as a submission. If you want a copy email me it is a very large chart, we (EETimes) did discuss the possibility of producing it as a poster for one of the exhibitions.

 

rbtbob
User Rank
Author
Re: Un-scientific opinions PCM and ReRAMs
rbtbob   3/3/2016 11:08:28 AM
NO RATINGS
Thanks for the notes on those two articles.  I am very curious about the outcome of the ECD Judge's hearing regarding Ovonyx on March 30th although any info will likely be repressed as being legally proprietary.  Maybe a hint will come from any ruling. Just the fact of the questioning and hearing is a curious thing.  I don't see how the Patent Office is going to be able to make an informed judgement on the entire universe of the alternate memory technologies.

Ron Neale
User Rank
Author
Re: Un-scientific opinions PCM and ReRAMs
Ron Neale   3/2/2016 4:35:44 AM
NO RATINGS
Rbtbob: Both of the articles you cite offer innovative new small-volume analysis techniques. The first article appears to be looking at single crystal. It is quite well established that there is a minimum sized amorphous dot that can exist in a crystal matrix. I will try and dig out the references for you. From memory I think a small minimum-sized (of the order 1nm) amorphous dot in a GST crystal matrix spontaneously reverts to the crystal state after a short period of time. My understanding of the paper is they appear to be observing a crystal to a meta-crystal state. Although at the resolution they are claiming the second state could be the short range order of an amorphous state. In my article I tried to leave open the door that there might be a co-ordination number change in relation to Ge that gets buried in the thermal noise, some work on super-lattices did suggest this for the Topological-switching RAM, the TRAM, at IEDM 2014.

http://www.eetimes.com/author.asp?section_id=36&doc_id=1325017&page_number=2

The second paper on moving oxygen to change bonding and/or create oxygen vacancies to change the electrical conductivity  of a material is the basis of many ReRAMs.

http://www.eetimes.com/author.asp?section_id=36&doc_id=1328077

We have another one in the EETimes editorial review hopper that should be out in a week or so. Finding a new material like strontium cobaltite, or SrCoOx. might be an important step forward.

rbtbob
User Rank
Author
Re: Un-scientific opinions
rbtbob   3/1/2016 7:34:02 PM
NO RATINGS
Google News has coughed up a couple of recent articles of interest:

http://phys.org/news/2016-02-successful-real-time-atomic-motion-sub-nanometer.html

http://nextbigfuture.com/2016/02/switchable-material-could-enable-new.html

Ron Neale
User Rank
Author
Re: Un-scientific opinions
Ron Neale   2/25/2016 11:42:44 AM
NO RATINGS
Rbtbob For their devices the IBM evidence points to a thermal initiator to the transition, they did choose "thermal assisted" in the title of their paper. The switching time is the sum of the delay time and the transition time.
If you apply pulses of the order one or two nano seconds then I think you will find you enter a constant voltage Zener-like region of conduction as I indicated in my figure leading to destructive breakdown.
It is possible that hidden amongst the dominant thermal effects an electronic effect like the change in coordination number of germanium is occurring. I like the model where a discontinuity in conductivity on melting of a local hotspot is the cause of the transition. If you believe in electro-crystallization you can throw that into the ring as the reason for a sudden transition.
With chalcogenides there is always the keep-hope-alive get out that other material compositions or device structures might have electronic initiators. I think the test for electronic switching is the free running oscillator based on the rapid negative resistance transition, show me one of those and that will make your case for electronic switching. It must oscillate at frequencies above the reciprocal of the thermal time constant.
On neurons and brain-like functions can we leave that until next week and my next article. I think you will find something of interest along the lines you suggest.

rbtbob
User Rank
Author
Un-scientific opinions
rbtbob   2/25/2016 10:50:24 AM
NO RATINGS
As one who drank the Ovshinsky Koolaid long ago, I still subscribe to the theory of an electronic mechanism.  Descriptions in a patent for a threshold switch talked of additive electric pulses building to the last pulse causing the switch to trip. (much like a neuron)  That seems to me to require a capacitance mechanism.  If I remember correctly the tripping was far too fast to be thermal.  This last bit is my unsupportable belief; the best makeup of memory cell material will not technically be a chalcogenide, and it will be doped with a rare earth.  Most likely Terbium.

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)
Like Us on Facebook
EE Times on Twitter
EE Times Twitter Feed