Some have more than 50 billion transistors that are less than 10,000 times the width of a human hair. They are made on the huge, state-of-the-art factory floor that can seven stories high and runs the length of four football fields.
In many ways, microchips are the lifeblood of the modern economy. They power computers, smartphones, cars, home appliances, and scores of other electronic devices. But the world’s demand for them has skyrocketed since the pandemic, which has also caused supply chain disruptions, leading to a global shortage.
That, in turn, is fueling inflation and raising alarms that the United States is becoming too dependent on chips made abroad. The United States accounts for only about 12 percent of global semiconductor manufacturing capacity; More than 90% of the most advanced chips come from Taiwan.
Intel, a Silicon Valley giant looking to restore its longtime leadership in chip technology, is betting $20 billion that it can help alleviate the chip shortage. It is building two plants at its chip manufacturing complex in Chandler, Ariz., that will take three years to complete, and recently announced plans for a potentially larger expansion, with new locations in New Albany, Ohio and Magdeburg, Germany.
Why does creating millions of these tiny components mean building – and spending – so big? A look inside Intel’s manufacturing plants in Chandler and Hillsboro, Ore., provides some answers.
What do chips do?
Chips, or integrated circuits, began to replace bulky individual transistors in the late 1950s. Many of these small components were manufactured on a single piece of silicon and connected to work together. The resulting chips store data, amplify radio signals, and perform other operations; Intel is best known for its variety of so-called microprocessors, which perform most of the computing functions of a computer.
Intel has managed to shrink the transistors on its microprocessors to a staggering size. But rival Taiwan Semiconductor Manufacturing Company can make components even smaller, a major reason Apple chose the company to make chips for its latest iPhones.
Such victories by a company based in Taiwan, an island that China claims as its own, add to the signs of how a growing gap in technology could make advances in the field of technology. The computer, consumer equipment and military hardware sectors are at risk from both China’s ambitions and natural threats in Taiwan such as earthquakes and droughts. And it has put a stamp on Intel’s efforts to regain its technological leadership.
How chips are made
Chipmakers are packing more and more transistors onto each piece of silicon, which is why technology is making more and more every year. That’s also why new chip factories cost billions of dollars, and fewer and fewer companies can afford to build them.
In addition to paying for buildings and machines, companies have to spend a lot to develop the complex processing steps used to manufacture chips from sheet-sized silicon wafers – that’s why Star factories are called “fabs”.
Giant machines project designs for the chips on each wafer, then lay and etch layers of material to make their transistors and connect them. Up to 25 panels at a time move between those systems in special housings on overhead automated rails.
Processing a wafer takes thousands of steps and up to two months. TSMC has set the pace for output in recent years, operating “gigafabs” sites with four or more production lines. Dan Hutcheson, vice president of market research firm TechInsights, estimates that each site can process more than 100,000 wafers per month. He said the capacity of Intel’s two $10 billion planned facilities in Arizona is about 40,000 wafers per month.
How to pack chips
After processing, the wafers are cut into thin slices. They are tested and wrapped in plastic packages to connect them to circuit boards or system components.
That move has become a new battleground, because even smaller transistors are harder to make. Companies are now stacking multiple chips or putting them side by side in a package, connecting them to act as a single piece of silicon.
As it is now common practice to pack a small number of chips together, Intel has developed an advanced product that uses new technology to package 47 notable individual chips, including some made by TSMC. and other companies manufacture as well as chips manufactured in Intel fabs.
What makes chip factories different?
Intel chips often sell for hundreds to thousands of dollars each. For example, in March, Intel released its fastest desktop processor with a starting price of $739. A single speck of dust not visible to the naked eye can damage a dust layer. Therefore, factories must be cleaner than hospital operating rooms and require complex systems to filter the air and regulate temperature and humidity.
The Fab must also be impervious to any vibrations, which can cause expensive equipment to fail. So the fab clean room is built on giant concrete slabs on a special damping system.
Also very important is the ability to move large amounts of liquids and gases. The top floor of Intel’s factories, about 70 feet high, has giant fans to help circulate air to the clean room directly below. Beneath the clean room are thousands of pumps, transformers, electrical cabinets, utility pipes and chillers connected to the production machines.
The necessity of water
Fab is a water-intensive activity. That’s because water is needed to clean wafers at many stages of the manufacturing process.
The two Intel facilities in Chandler collectively collect approximately 11 million gallons of water per day from the local utility. Intel’s future expansion will require significantly more, a challenge that seems to be for a drought-stricken state like Arizona, which has already cut water allocations to farmers. But agriculture actually consumes more water than a chip factory.
Intel says its Chandler sites, which rely on supplies from three rivers and well systems, recover about 82% of the freshwater they use through filtration, settling ponds and other equipment. That water is sent back to the city, which operates Intel-funded processing facilities, and redistributed for irrigation and other uses.
Intel hopes to help increase water supply in Arizona and other states by 2030, by working with environmental groups and others on water conservation and restoration projects for local communities.
How are fabs built?
To build the factories of the future, Intel will need about 5,000 skilled construction workers over three years.
They have a lot of work to do. Dan Doron, Intel’s chief construction officer, said excavation of the foundation is expected to remove 890,000 cubic meters of soil, which is transported at a rate of one dump truck per minute.
The company is expected to pour more than 445,000 cubic meters of concrete and use 100,000 tons of reinforcing steel for the foundation – more than building the tallest building in the world, the Burj Khalifa in Dubai.
Some of the construction cranes are so large that more than 100 trucks are needed to bring the parts to assembly, Mr. Doron said. The cranes will lift, among other things, the 55-ton chiller for the new equipment.
Patrick Gelsinger, who became CEO of Intel a year ago, is lobbying Congress to provide grants for plant construction and tax credits for equipment investment. To manage Intel’s spending risk, he plans to emphasize building the right “shells” that can be outfitted with equipment in response to market changes.
To solve the chip shortage, Mr. Gelsinger will have to make good on plans to produce chips designed by other companies. But a company can only do so much; products like phones and cars require components from multiple suppliers, as do older chips. And no country can stand alone in the field of semiconductors. While boosting domestic production may reduce supply risks somewhat, the chip industry will continue to rely on a complex global network of players for raw materials, manufacturing equipment, hardware and software. Design software, talent and specialized production.
Made by Alana Celii