Online information source for semiconductor professionals

Spansion has officially announced that it has started production at its first 300mm fab, SP1. This is an important milestone for the NOR flash memory producer within its first year of becoming a public company. 

In the latest edition of Semiconductor Fabtech’s (#35 - just gone to print), I had the chance to interview Jim Doran, Spansion’s EVP & COO about its manufacturing strategies, which included the initial ramp of SP1.

It was very clear that the future success of the company rests on its ability to both shift NOR production to 300mm wafers and step up node migrations.

I have also been impressed by the way Spansion is executing these migrations and I realised that there is an interesting story that could play out very well in 2008 for the company.

Without giving the game away from the interview with Doran, an area that I didn’t cover in the article concerned what the company’s competitors are up to in response to Spansion 300mm migration and one-year node migration plans.

Currently, Numonyx, the Intel/STMicroelectronics spin-off, and Samsung are its main competitors and neither produces NOR flash on 300mm wafers. With the move by Spansion to the 65nm node with the initial ramp at SP1, Spansion has caught up with its rivals in that department.

However, in around the third quarter next year (2008), Spansion will migrate to its 45nm process just a year after starting 65nm production, which should put Spansion ahead of its rivals both in technology and wafer size.

With NOR remaining a very competitive market, Spansion should become the lowest cost NOR manufacturer and leverage the significant cost savings of smaller die and the larger wafer size.

Numonyx has already stated that it has no plans to start 300mm production in 2008. It has a shuttered 300mm fab in Catania, Italy, courtesy of STMicroelectronics. Being debt-laden and with no IPO yet announced, Numonyx is cash strapped and unlikely to spend any money on equipping that fab until the business is operating properly and new sources of funding are found.

It also has to juggle when and if it migrates to phase-change technology to enable further scaling and then bring up the 300mm fab. A lot is up in the air at Numonyx right now.

Samsung is in a slightly different position as it can leverage fully depreciated 200mm fabs for NOR production and is in need of finding uses for these facilities as DRAM and NAND production on 200mm wafers rapidly becomes obsolescent from a cost and technology perspective.

With DRAM and NAND of prime importance for 300mm production, it is unlikely Samsung will rush to fabricate NOR in 300mm fabs.

Basically, Spansion has at least a year of being the main NOR producer solely using 300mm wafers. Don’t forget that although the initial ramp is only 15,000wspm, it is using TSMC’s Fab 14 for NOR production at the 90nm node and should migrate to 65nm at TSMC when SP1 migrates to 45nm.

This all puts Spansion in the best position it has ever been in, and the company really should be able to capitalize on its lead in the next few years. With the boost in capacity and lowest manufacturing costs compared to its rivals, I would expect Spansion to really ramp up the release of new products so that it can offer NOR to new markets and actually drive the expansion of NOR rather than see it suffering within the dominant mobile phone market.

It will be interesting to see what its competitors will do to combat this move to 300mm and faster node migrations.

Spansion should be congratulated for making the NOR space interesting again. Indeed, NOR is no bore anymore!