Breaking News
Design How-To

FinFET structure design and variability analysis enabled by TCAD

NO RATINGS
1 saves
Page 1 / 3 Next >
More Related Links
View Comments: Newest First | Oldest First | Threaded View
Adele.Hars
User Rank
Author
re: FinFET structure design and variability analysis enabled by TCAD
Adele.Hars   10/12/2012 12:26:26 PM
NO RATINGS
Hmm. In a presentation at the SOI Conference last week, IBM painted a very different picture. It would be interesting to get their take here.

Victor Moroz
User Rank
Author
re: FinFET structure design and variability analysis enabled by TCAD
Victor Moroz   10/11/2012 4:49:54 PM
NO RATINGS
Adele, the industry is making bulk FinFETs due to the lower cost. There are no serious advantages in SOI FinFETs as long as you can keep consistent fin height in bulk FinFETs, and it was reported to be +-1nm at VLSI Technology conference this year. The channel-stop doping is below the fin channel and has a negligible impact on the overall variability.

Adele.Hars
User Rank
Author
re: FinFET structure design and variability analysis enabled by TCAD
Adele.Hars   10/11/2012 2:51:56 PM
NO RATINGS
Victor, have you looked at FinFET on SOI? Re: fin shape, the GSS guys (and others) are saying FinFETs would be better for power, perf & mfg on SOI (see http://www.advancedsubstratenews.com/2012/08/power-performance-gss-sees-soi-advantages-for-finfets/ ) And with respect to variability, with an SOI FinFET, you don't dope, so you don't have the RDF problem. Modeling guru Jerry Fossum very eloquently explained it a five (!) years ago (see http://www.advancedsubstratenews.com/2007/05/a-perspective-on-multi-gate-mosfets/ ). Your thoughts?

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