The semiconductor industry cannot afford the R&D costs for both shrinking transistors and moving to 450mm wafers according to a report from Equipment Productivity Working Group (EPWG) of SEMI, the trade body of the semiconductor production equipment manufacturing industry.
”450-mm wafer scale-up represents a low-return, high-risk investment opportunity for the entire semiconductor ecosystem 450-mm should, therefore, be an extremely low priority area for industry investment,” said the SEMI report.
The only people calling for 450mm wafers are Intel, Samsung and TSMC. TSMC, it is said, was reluctantly dragooned by Intel into adding its name to a call for 450mm.
Apart from these three, the semiconductor industry in general is lukewarm about 450mm and is not prepared pay for it, but will use 450mm wafers if someone else pays for it. According to the SEMI report, the cost of developing 450mm equipment could be $25bn.
The SEMI report continues: “The research and development dollars for shrinks, new processes, and new materials are highly constrained, and must be invested where they will offer a demonstrable positive return. Somewhat surprisingly, the change to a larger wafer size in itself does not lead to significantly reduced costs. Intel saw this in the changeover from 150 mm to 200 mm, and the same will be true for a change from 300 mm to 450 mm. “
The report continues: “The true driver of the overall reduced cost for 300 mm wafers was the use of factory automation (that is, advanced materials handling systems, or AMHS), advanced process control systems (APC), mini-environments (FOUPs), and stopping work on 200 mm node advancement.”
The SEMI report concludes: “Shrinks, new materials, and new processes will continue to advance the industry on Moore’s Law, but there are simply not enough R&D resources available to continue such advancement in nodes and processes AND to work on a 450 mm wafer size transition.”
Very much in agreement with that view is Dr Mark Pinto, CTO of the world’s largest manufacturer of semiconductor manufacturing equipment, Applied Materials, “if we’re going to go to 32nm and 22nm and beyond, and do 450mm as well, the costs are just huge,” Pinto told EW recently.
Asked why Intel, Samsung and TSMC were pushing for it. Pinto replied: “Because no one else in the industry wants 450mm wafers.” Asked what might justify 450mm in the future, Pinto replied: “If scaling ends”.
Applied Materials recently confirmed that it is doing no work on 450mm development despite the wishes of Intel, Samsung and TSMC.
Stan Myers, president and CEO of SEMI, equates 450-mm technology to 157nm lithography on which billions were spent but nothing was developed and Intel, having encouraged the effort, pulled out of the programme. 450mm could suffer a similar fate.
There is a question about whether silicon wafers can be made sufficiently robust to be capable of being handled at 450mm.
Agreeing with Pinto pointed out the huge difficulties in producing a 450mm wafer sufficiently robust to withstand the stresses of processing.
“In the future we may find a material which is capable of producing a 450mm wafer,” Dr Susumu Kohyama, president and CEO of materials specialist Covalent Materials, recently told EW, adding, “but not now.”