Breaking News
Design How-To

PCM scalability--Myth or realistic device projection

8/19/2010 01:24 AM EDT
10 comments
NO RATINGS
More Related Links
View Comments: Newest First | Oldest First | Threaded View
JanineLove
User Rank
Author
re: PCM scalability--Myth or realistic device projection
JanineLove   12/2/2010 5:55:04 PM
NO RATINGS
See part two here: http://www.eetimes.com/design/memory-design/4210054/PCM-Scalability-The-Myth--Part-2-?Ecosystem=memory-design

Allen Wallace
User Rank
Author
re: PCM scalability--Myth or realistic device projection
Allen Wallace   10/20/2010 2:55:53 PM
NO RATINGS
Ron Neale, Can you provide more details on your proposed low-cost experiments for student regarding PCM? I mentor High school students with advance science fair projects and I'd like to suggest PCM to one of my students. Allen Wallace

zman_tekinsil
User Rank
Author
re: PCM scalability--Myth or realistic device projection
zman_tekinsil   9/6/2010 2:49:08 PM
NO RATINGS
I appreciated the content and details in this article. Some Numonyx people, now part of Micron took on the PCM doubters: http://www.eetimes.com/electronics-news/4084947/Numonyx-CEO-takes-on-PCM-doubters. Is this a narrow-minded objection or a marketing fluff!

R G.Neale
User Rank
Author
re: PCM scalability--Myth or realistic device projection
R G.Neale   9/6/2010 12:24:39 PM
NO RATINGS
Contd from previous post/ For a memory device based on structural change to succeed what appears to be needed is some electrically “assisted” structural change that when removed has a highly improbable chance of reforming in an exactly ordered manner, thermally or electrically. For read the resonance of a particular structural shape is detected not general crystallization. I have seen the term “vacancy ordering” described to account for the memory effect in one of the new resistive memory devices. If that does not involve structural ordering, i.e. a nano-phase change then finding the structural equivalent would be one direction to pursue. In that regard it might be worth pursuing nano-filaments and revisiting the Work of Prof LeCoomber, (1980s) amorphous silicon, University, Scotland and Ovshinsky (1970s)chalcogenides. Both demonstrated the step-like resonant effects in the V-I characteristics of in two terminal amorphous silicon and chalcogenides devices respectively. This author, and many others, saw the Ovshinsky demonstrations. At the time they were dismissed as contact art effects, although accounting for the equal spacing and exact multiples of amplitude of the “resonance” steps was a little more than difficult. This writer had many conversations with Prof LeCoomber both then and in his paper he was firmly convinced the steps in the observed V-I characteristics were caused by resonance in micro or nano-filaments in both cases.

R G.Neale
User Rank
Author
re: PCM scalability--Myth or realistic device projection
R G.Neale   9/6/2010 11:47:18 AM
NO RATINGS
What is the next memory technology? If really knew I think I would be working on it or investing in it. I think the research and development on NVM or universal memory must continue on many fronts. That includes all of: electrically manipulated structural change (including PCM), resistance effects, magnetic effects, ferro-electric effects, optical and nano structures. Contrary to the impression that my paper above may have given to some, I think the electrical manipulated structural change may have a role to play, without specifying the materials. My purpose was to use existing published and peer group reviewed data as a prediction tool to expose the problems with PCM devices (sub 30nm) that could not at the moment be fabricated. Furthermore it was also to bring current density and electro-migration to a position high on the list of serious PCM problems to be discussed. I wanted to replace unreal promises with real problems.

Helicopter0
User Rank
Author
re: PCM scalability--Myth or realistic device projection
Helicopter0   9/2/2010 5:36:11 AM
NO RATINGS
excellent technical analysis in your view, what is the next memory technology?

Most Recent Comments
michigan0
 
SteveHarris0
 
realjjj
 
SteveHarris0
 
SteveHarris0
 
VicVat
 
Les_Slater
 
SSDWEM
 
witeken
Most Recent Messages
9/25/2016
4:48:30 PM
michigan0 Sang Kim First, 28nm bulk is in volume manufacturing for several years by the major semiconductor companies but not 28nm FDSOI today yet. Why not? Simply because unlike 28nm bulk the LDD(Lightly Doped Drain) to minimize hot carrier generation can't be implemented in 28nm FDSOI. Furthermore, hot carrier reliability becomes worse with scaling, That is the major reason why 28nm FDSOI is not manufacturable today and will not be. Second, how can you suppress the leakage currents from such ultra short 7nm due to the short channel effects? How thin SOI thickness is required to prevent punch-through of un-dopped 7nm FDSOI? Possibly less than 4nm. Depositing such an ultra thin film less then 4nm filum uniformly and reliably over 12" wafers at the manufacturing line is extremely difficult or not even manufacturable. If not manufacturable, the 7nm FDSOI debate is over!Third, what happens when hot carriers are generated near the drain at normal operation of 7nm FDSOI? Electrons go to the positively biased drain with no harm but where the holes to go? The holes can't go to the substrate because of the thin BOX layer. Some holes may become trapped at the BOX layer causing Vt shift. However, the vast majority of holes drift through the the un-dopped SOI channel toward the N+Source,...

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