REGISTER | LOGIN
Breaking News
Blog

For PCM Threshold Switching Read Thermistor?

NO RATINGS
Page 1 / 3 Next >
View Comments: Newest First | Oldest First | Threaded View
Ron Neale
User Rank
Author
Re: threshold switching Update
Ron Neale   1/20/2016 3:47:49 AM
NO RATINGS
Resistion: Dr Luca Crespi of the Politecnico di Milano, Italy, the lead author of the paper I referenced in my post below has just advised me that his paper deals with a discontinuity in electrical conductivity at the crystal-to -melting phase change and not the amorphous-to- melting phase change. If IBM are right and there is not a discontinuity  for the amorphous to melt transition that leads to the questions at what temperatures are the values of electrical conductivity for amorphous and crystalline material the same or  on crystallization does the conductivity  of the amorphous GST material get reduced to  join the curve of electrical conductivity as a function of temperature for crystalline material.

Ron Neale
User Rank
Author
Re: threshold switching
Ron Neale   1/19/2016 3:54:05 PM
NO RATINGS
Resistion; I do not think the matter is settled, I am sure we are getting much closer. I think we have to eliminate the possibility that there might be a post-threshold switching state before crystallization occurs that is not just hot amorphous material. Other than melting, in some devices, I did think that the co-ordination number change in Ge was a possibility that needed to be isolated, in a normal device it gets swamped by thermal effects.

On melting, after reading my article in EETimes one of my colleagues drew my attention to the paper " Electrical Conductivity Discontinuity at Melt in Phase Change Memory, by Luca Crespi et al.IEEE Electron Device Letters Vol 35 No7 July 2014, pp747-749.
I hope I can get IBM to respond for you and the readers to what at first site might be a differing view with respect to discontinuities on melting.

On you comment re filament formation and ReRAMs. For me "S" shaped threshold switching and negative resistance is the signature of filament formation or at least a localized hotspot, while "N" shaped negative resistance (Gunn device) is a planar domain. In some ReRAMs an initial filament is formed during the "Forming" process and subsequent switching events occur within that filament at a necked down region. While the "S" shaped negative resistance would suggest a even smaller filament might be forming we have to allow for the possibility that a bulk switching or interface effect could be occurring, even a Mott/Hubbard transition. If a narrow sub-filament is forming it might be possible to see evidence of quantum effects.

Ron Neale
User Rank
Author
Re: Answer to author Historical facts
Ron Neale   1/19/2016 3:05:00 PM
NO RATINGS
 Semyon Thank you for the good guy reference. I needed to check that I was not including for EETimes readers an unreliable source. I think Stan Ovshinsky always acknowledged Boris T. Kolomiet's important contribution to the science of amorphous materials. I think the discovery of the switching effects is usually jointly credited to Bell labs, USA, Boris T. Kolomiets in Russia and Ovshinsky at ECD, USA.
I know Russian is your native language and it may be a problem as you appear to be suggesting that Ovshinsky only learned about these materials in 1967. While he may have learned of the work Kolomiet' had been carrying out in 1967, he (Ovshinsky ) was already well aware of the properties of the materials.
I know that because I visited Ovshinsky's lab in Troy, Michigan in 1961 as a technical emissary of Close Brothers the London based Merchant Bank to investigate the memory and switching devices and was given demonstrations of the crude devices he was able to show. On my advice Close Brothers were one of the early investors in ECD, and a licence was secured for Electronic Machine Company (EMC)  and Vacwell Engineering in the UK. That is when I first started work on amorphous devices.
Sort of relevant to the debate of the present paper on the thermal or electronic nature of switching I now include in the history part of my lecture series, the micrographs of my early unsuccessful attempts at thin film devices, circa 1962, switching and memory, made with my colleague Brian Smith. One planar and one a gap device. Because at that time we believed we were dealing with an electronic effect and not a thermal effect we used gold electrodes and polished stainless steel as an electrodes, a double diaster. That latter to keep things cool and a choice because we owned a company "Ellis Optical" who had expertise in that area.  If you want the slides I can send them to you I will try and post them here later, if I cannot any reader wanting a copy can contact me.

Semyon Savransky
User Rank
Author
Answer to author
Semyon Savransky   1/19/2016 11:48:45 AM
NO RATINGS
Ron Neale asked me "In the preamble I referenced a Russian paper with respect to the large number of electronic theories on threshold switching. Have you heard of
the author, is he reliable source?"

Yes, Prof. Konstantin Damdinovich Tsendin (Константин Дамдинович Цэндин) from A.F. Ioffe Institute (the effect of threshold switching in chalcogenides was co-discovered there in 1962 in Boris T. Kolomiets laboratory in which Stan Ovshinsky learned about this class of materials during his visit in 1967) is THE MOST reliable and knowledgeable source of theories on threshold switching and some other phenomena in chalcogenide glasses he have developed from ~ 1970.

Semyon D. Savransky

 

IBM_Research
User Rank
Author
Re: threshold switching
IBM_Research   1/19/2016 10:26:00 AM
NO RATINGS
Thanks for the comment. In the main paper (Ref. [1]) we did measurements of threshold voltage drift over 6 decades in time at different temperatures (see figure 7).

resistion
User Rank
Author
threshold switching
resistion   1/18/2016 7:59:54 PM
NO RATINGS
Thanks Ron for a very interesting article on threshold switching of PCM. So it seems the explanation is not "settled" yet? I think the circumstances under which the threshold voltage varies or drifts would be very relevant. Also, filamentary RRAMs fundamentally undergo threshold switching during their SET. I wonder if threshold switching is fundamentally filamentary.

Most Recent Comments
David Ashton
 
Smartalix
 
Sang-Yun
 
Sang-Yun
 
realjjj
 
DMcCunney
 
David Ashton
 
rick merritt
 
rick merritt
Like Us on Facebook
EE Times on Twitter
EE Times Twitter Feed