SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : ASML Holding NV -- Ignore unavailable to you. Want to Upgrade?


To: SemiBull who wrote (911)2/9/2004 8:25:07 PM
From: SemiBull  Read Replies (4) | Respond to of 43397
 
ASML, Canon, Nikon vie for AMD's 300-mm litho biz (Updated)

Mark LaPedus
02/09/2004 10:50 AM EST
URL: siliconstrategies.com

SUNNYVALE, Calif.--ASML Holding NV, Canon Inc., and Nikon Corp. are competing for a major 300-mm, lithography-tool order at Advanced Micro Devices Inc.'s proposed fab in Germany, according to officials from the supplier of microprocessors and flash memory devices.

AMD is also evaluating immersion lithography tools from various vendors, which could be inserted in the later part of the 65-nm node, said David Bennett, director of strategic equipment technology and alliances for the Sunnyvale-based chip maker.

Bennett said AMD has yet to order an immersion tool, but he dropped hints that the technology could eliminate 157-nm lithography, perhaps even from the company's own roadmap. "Immersion may be the death of 157-nm," he said in an interview with Silicon Strategies..

(Editors note: Originally, Silicon Strategies reported that Bennett said immersion is the death of 157-nm lithography. Silicon Strategies reported this in error and Bennett said immersion may be the death of 157-nm lithography.)

Initially, AMD is looking to order an unclosed number of "dry" 193-nm lithography tools for its 300-mm wafer fab, which is under construction in Dresden, Germany. The fab will be used to make next-generation processors. In November, AMD broke ground on a manufacturing facility for 300-mm diameter wafers, named Fab 36, in Dresden. The fab, to cost $2.4 billion, is a 65- to 45-nm plant that is expected to move into production in 2006 (see November 20, 2003 story).

The leading contenders for the lithography order at AMD's Dresden fab are ASML, Canon, and Nikon, although no decision has been reached, according to Bennett. "We are looking at all vendors," he said.

At present, AMD basically uses two lithography vendors for the critical layers for its chip designs--ASML of the Netherlands and Canon of Japan. Generally, the chip maker has a large installed base of ASML's 248- and 193-nm lithography tools, which are used to process the wafers for its microprocessors, he said.

The main lithography vendor for its flash-memory business is Canon, he said. AMD and Fujitsu Ltd. have a joint flash-memory venture, dubbed FASL LLC, and Canon is the primary scanner vendor in that operation, he said.

Some observers believe the final decision for the lithography order may come down to both AMD and its technology partner, IBM Corp. IBM and AMD have a joint technology deal to devise process technology for the 65-nm and 45-nm nodes on 300-mm substrates.

Bennett indicated that AMD may "align" its tool choices with IBM, which could reportedly benefit at least one vendor--ASML. Sources believe that IBM mainly uses ASML's 193-nm scanners within its new 300-mm fab in East Fishkill, N.Y.

"Relative alignment with the development tool set is one of those considerations," he said.

ASML is not a slam-dunk to win the business within AMD's Dresden plant, however. "Cost is a major component," Bennett said.

Indeed, analysts believe that ASML's 193-nm tools cost $12-to-$20 million each. In contrast, tools from Canon and Nikon sell for below that, reportedly $8-to-$15 million, according to analysts.

Tool price is not the only consideration. "We all know that ASML carries a premium on a stepper-by-stepper basis, and for that reason, AMD looks at total cost of ownership to guide its decision and accounts for things like throughput, technical capability and other factors," he said.

Meanwhile, Canon is making inroads as well. Besides being the dominate supplier of scanners to FASL, Canon has also installed a leading-edge tool within an undisclosed development center. The tool is believed to be installed within Advanced Mask Technology Center GmbH & Co. KG (AMTC), a joint photomask venture in Dresden between AMD, Infineon, and DuPont Photomasks. AMTC is developing masks for the 90- and 65-nm nodes.

Nikon is also not out of the running for the business. "Nikon is also way out there with immersion," according to Bennett. Like most leading-edge chip makers, AMD is evaluating immersion lithography and 157-nm despite the problems with latter technology.

"We're looking at all the technologies," he said. "Immersion seems to be far closer for the 65-nm node. (Immersion makes sense) for the latter part of the 65-nm node."



To: SemiBull who wrote (911)2/20/2004 10:34:23 AM
From: Proud_Infidel  Read Replies (1) | Respond to of 43397
 
Optical maskless litho needs three years, customers, says ASML
By Peter Clarke
Silicon Strategies
02/20/2004, 9:40 AM ET

PARIS -- Optical maskless lithography, where an addressable digital spatial light modulator is updated on the fly to give different masking patterns, could be shipped by ASML Holding NV within three years according Doug Dunn, chairman and chief executive officer of the lithograpy equipment maker.

However, doubts remain as to whether the technique is commercially viable the executive said during an invited keynote speech Wednesday (February 18, 2004) for a session on lithography within the International Forum on Semiconductor Technology held here.

Maskless lithography is a possible solution to escalating mask costs but has the disadvantage over conventional lithography of reduced throughput. Companies that make high volumes of standard products, such as Intel Corp. and Samsung Electronics Co. Ltd. would have less need for maskless lithography but also have a significant influence on the commercialization of lithography techniques and extensions.

ASML finally announced the formation of a joint venture to develop an optical maskless lithography system, with Micronic Laser Systems AB about two years after the original intention was announced back in 2001 (see July 18 2003 story).

During the talk Dunn gave on Wednesday he showed photographs of the Micronic spatial light modulator, which has an active area of 8-mm by 33-mm. But he also indicated that the throughput was limited to about five to 10 wafers per hour.

"Are there enough customers to make this worthwhile? I am not sure. It's still a research program," said Dunn. "Optical maskless lithography will fit into the same platform (as conventional lithography). We could ship within three years," Dunn said.

"The technique applies to low volume, high cycle runs. Up to 100 wafers it has advantages. But at 500 wafers or more it does not," Dunn told the IFST delegates.

Other companies believed to be working on maskless lithography tools, include Canon Inc., E-Beam Corp., Mapper Lithography NV and Leica Microsystems Inc.



To: SemiBull who wrote (911)2/23/2004 9:31:58 PM
From: Proud_Infidel  Read Replies (1) | Respond to of 43397
 
193-nm immersion simulations extend down to 22-nm node
By Mark LaPedus
Silicon Strategies
02/23/2004, 7:45 PM ET

SANTA CLARA, Calif.--During the SPIE Microlithography conference here this week, researchers from the University of New Mexico will claim that they have simulated 22-nm half-pitch lines, by deploying 193-nm immersion lithography technology.

The disclosure confirms that 193-nm immersion technology could be a viable lithography solution beyond the 45-nm node. It also implies that the technology could push out the need for 157-nm or extreme ultraviolet (EUV) technologies, according to analysts.

"Taking advantage of a water (n=1.44 at 193-nm) immersion lens with NA equal up to 1.3, the 45-nm half-pitch node should be accessible," according to a paper, which will be presented by the University of New Mexico at SPIE. "However, extension of optical lithography to the 22-nm node poses a bigger challenge."

To solve this issue, the university and others have proposed imaging interferometric lithography (IIL). This is a new imaging concept that provides an approach for accessing the fundamental limits of optics. With IIL, the University of New Mexico enabled 22-nm simulations with immersion via KLA-Tencor Inc.'s Prolith simulation tools.

"To get the highest available spatial frequencies, light has to interfere at large angles in the resist and the image contrast become highly polarization dependent," according to the paper. "The choice of polarization in different exposures dramatically improves the resist patterns."

IIL also eases many of the constraints in optical lithography. Besides the University of New Mexico, others are moving down a similar path.

Pushing the theoretical limits of optics, Rochester Institute of Technology (RIT) last week disclose that it has demonstrated image resolutions down to 38-nm with a prototype 193-nm immersion lithography tool (see January 23 story).

Also last month, Taiwan Semiconductor Manufacturing Company Ltd. said it plans to introduce immersion lithography for critical layers of the 65-nm node, starting in the middle of 2005. TSMC also said that 193-nm "wet" lithography could extend to the 22-nm manufacturing process node (see January 28 story).