|
My response to the Carl Johnson article "NO Guts, No Glory"
e-mail to: feedback@cmpnetmail.com {Att: Carl Johnson]
I take exception with several points in Carl Johnson's February 3, 1998 article "No Guts, No Glory" excerpted below.
(1) The article states that Intel does not plan extending production to deep sub-micron when, in fact, there are several authoritative published articles saying that Intel is already using SVGI Micrascan-2 in producing sub 0.25 micron critical dimensioned ICs and plans using Micrascan-3 DUV scanner/steppers for sub 0.18 micron printed wafers.
(2) One large US based device manufacturer is alleged to have reallocated resources back to i-line after experiencing problems with DUV steppers. The report wrongly implies that the DUV stepper failed to do the intended job and is no longer used. As I understand it most IC makers are using combined i-line and DUV tools in mix and match applications. That is, DUV tools are used for printing sub 0.35 micron features and i-line is used for 0.35-and-larger dimensions that are printed on the same wafer. While it's true that all manufacturers have undergone a difficult learning curve in transitioning to DUV, I challenge the accuracy of the implication made in the article that there are DUV tools sitting idle.
(3) Far from being a migration from DUV to i-line as the article states there have been numerous authoritative reports showing that the ratio of i-line to DUV tools has diminished from 3:1 in Sept 97 to 2:1 in Jan 98 , and that ratio is projected to diminish further to 1:1 by the end of 1998.
I am taking the liberty of posting this e-mail on the SI Cymer thread. I will post your reply there also, if you are agreeable to my doing so.
Sincerely,
Tulvio Durand
***********************
techweb.com
EXCERPT
>>>Speaking of the front end, last week we attended presentations by many members of this segment at the Nationsbanc-Montgomery Technology Conference. Several device manufacturers suggested their existing front-end tools will allow them to extend production to deep sub-micron feature sizes. This does not hold true for all manufacturers and is certainly not the case at Intel. From what we heard at the conference, Intel views the current dislocation as an opportunity to bury some of their competition -- hardly surprising. Those who suggested the extension approach have plans to continue using their I-line lithography tools or, take presently installed DUV lithography systems to feature sizes below 0.25 micron.
Recent news we have received from a large U.S.-based device manufacturer regarding their DUV efforts is very discouraging. This
manufacturer has decided to allocate resources back to I-line lithography because of problems they are having with DUV steppers. Yes, there are a few companies implementing DUV successfully, but reassessment and migration back to I-line lithography indicates that the DUV transition will be more difficult than many anticipate. Of course, stretching I-line to the limits will produce a whole new set of problems. There has also been talk about the production of devices with 0.18-micron feature sizes. We are not in the camp that believes a smooth transition will occur to this new generation. In a nutshell, there are still many technical issues to overcome and very few players will be able to make the necessary investments facilitating the production of devices at this level.>>>>>
|
