The CHIPS and Science Act, signed into law by President Biden on Aug. 9, will not only boost U.S. semiconductor manufacturing and packaging, but testing as well, Sen. Mark Kelly (D-Ariz.), the chairman of the Senate Armed Services Committee’s emerging threats and capabilities panel, said last week at an Arizona State University (ASU) forum on semiconductors and national security.
“Over the last 20 years, we’ve seen China innovate, develop technologies in some cases even beyond the capabilities that we have,” Kelly said. “Things that come to mind are things like artificial intelligence, hypersonic missile systems. Those things require the best semiconductor chips in the world, and DoD has to have the ability to do the necessary research and development. But often…some of the DoD chips that go in our weapons systems are, during this process, routed through China to be tested. We don’t have this capability here.”
Beyond incentives for U.S. chip manufacturing, the new law “is going to improve the [chip] research, development [R&D] and testing as well,” Kelly said.
The CHIPS and Science Act authorizes a new National Semiconductor Technology Center under DoD to spearhead such R&D and prototyping.
“We could see testing capacity that currently exists in Europe, or China, in some cases, [move to the U.S.],” Kelly said. “When an entrepreneur is looking to test his newly designed semiconductor chip, they don’t have a lot of money. Where are they gonna go? They’re often gonna go to China because it’s cheaper to do it there. But through this program, a couple of billion dollars to set up this capability, that will now be here in the United States.”
The U.S. share of semiconductor production has fallen from 37 percent in 1990 to 12 percent, while the domestic share of chip testing is even below that. In addition, foreign competitors have been able to rely on government subsidies to construct $10 billion chip fabrication plants–“fabs”–and build chips for at least 30 percent less than U.S. companies, according to U.S. semiconductor company officials.
The Semiconductor Industry Association said that 75 percent of global chip manufacturing is in East Asia and that China is slated to hold the lead by 2030 “due to its government’s massive investments in this sector.”
The landscape may be shifting, however. For example, Intel Corp. [INTC] is investing $20 billion to build two fabs in Arizona, while Taiwan-based Taiwan Semiconductor Manufacturing Company (TSMC) just built a $12 billion chip plant in the state.
Semiconductors are critical for modern weapons. The Lockheed Martin [LMT] F-35 has more than 3,000 computer chips, while the company’s CH-53K helicopter for the U.S. Marine Corps has more than 2,000, and the Javelin anti-tank missile has more than 250.
Artificial intelligence (AI) relies on semiconductors smaller than 14 nanometers (nm), and just two companies–South Korea’s Samsung Electronics and TSMC–have built chips below 10 nm.
The $280 billion CHIPS and Science Act includes $52.7 billion to bolster domestic semiconductor manufacturing capacity and $200 billion for science and technology efforts to beat Chinese competition (Defense Daily, July 28). Section 103 of the law prohibits “the recipients of federal [semiconductor] incentive funds from expanding or building new manufacturing capacity for certain advanced semiconductors in specific countries that present a national security threat to the United States.”
“To ensure that these restrictions remain current with the status of semiconductor technology and with U.S. export control regulations, the Secretary of Commerce, in coordination with the Secretary of Defense and the Director of National Intelligence, would be required to regularly reconsider, with industry input, which technologies are subject to this prohibition,” per the law.
Of the $52.7 billion in the CHIPS portion, $39 billion is to expand and improve domestic semiconductor manufacturing capacity, covering $19 billion in fiscal year 2022 of which $2 billion would go toward legacy chip production, followed by $5 billion annually from FY ‘23 to ‘26.
The bill also includes $2 billion for chip production related to DoD-unique requirements, $11 billion for microelectronics research and development and steps to bolster the workforce and a 25 percent investment tax credit covering manufacturing of semiconductors and associated equipment in the U.S.
“The CHIPS and Science Act is probably the biggest piece of competitiveness legislation that’s been passed in the United States since World War II,” Bruce Andrews, Intel’s vice president and chief of government affairs, said at last week’s ASU forum.
Through the Defense Microelectronics Activity’s (DMEA) Trusted Foundry Program, which certifies U.S.-based computer chip suppliers for varying levels of secure access, DoD has had partnerships with IBM [IBM] and GlobalFoundries, Inc. [GFS], which bought IBM’s chip business in 2014, to build semiconductors domestically. In May, 2020, Mark Lewis, the Pentagon’s then director of defense research and engineering for modernization, said that the Trusted Foundry model has “failed from a business standpoint,” and must change its approach to take advantage of commercially available technology (Defense Daily, May 19 2020).
A RAND study earlier this year advocates that DoD undertake supply chain risk management to minimize supply chain disruptions.
“As my CEO says, ‘We’re moving from a world of just in time to a world of just in case’ because we realize we can’t rely on being overconsolidated in one region,” Intel’s Andrews said last week. “We have to have diversification and resilience. We just can’t be successful without it.”
Last year, Heidi Shyu, the under secretary of defense for research and engineering, said that her plan for microelectronics would take a new approach to the Trusted Foundry model by having DoD leverage the latest commercial chip designs while working with industry to ensure that the DoD chips are secure (Defense Daily, Dec. 6, 2021).