Some have more than 50 billion tiny transistors that are 10,000 times smaller than the width of a human hair. They can be made on the gigantic, ultraclean factory room floor Seven storeys high and four football fields in length.
Microchips are the lifeblood of the modern economy in many ways. They power the scores of computers, smartphones, cars, devices and other electronics. But global demand for them has increased since the epidemic, which has also led to supply-chain disruptions, resulting in global shortages.
This, in turn, is fueling inflation and raising the alarm that the United States is becoming increasingly dependent on overseas chips. The United States accounts for only 12 percent of global semiconductor production capacity; More than 90 percent of the advanced chips come from Taiwan.
Intel, the Silicon Valley Titan that seeks to restore its long-standing lead in chip manufacturing technology, is betting $ 20 billion that could help alleviate chip shortages. It is building two factories in its chip-manufacturing complex in Chandler, Ariz., Which will take three years to complete, and recently announced plans for a potentially large expansion with new sites in New Albany, Ohio and Magdeburg, Germany.
Making millions out of these small components means why building – and costs – are so big? A look inside the Intel production plants in Chandler and Hillsboro, Ore., Provides some answers.
What chips do
Chips, or integrated circuits, began to replace large individual transistors in the late 1950s. Many of these small components are produced on a piece of silicone and joined together to work together. The resulting chips store data, amplify radio signals and perform other operations; Intel is famous for its variety of microprocessors, which perform most of the computational tasks of a computer.
Intel has managed to shrink the transistor on its microprocessor to mind-bending size. But rival Taiwanese semiconductor manufacturing company could make even smaller components, mainly because Apple chose it to make chips for its latest iPhones.
Such a victory by a Taiwan-based company, an island that China claims, adds signs of growing technological differences that could jeopardize advances in computing, consumer devices and military hardware from both China’s ambitions and natural hazards in Taiwan. As earthquakes and famines. And it points to Intel’s efforts to recapture the technology lead.
How chips are made
Chip makers are increasingly packing transistors on every piece of silicon, which is why technology is doing more every year. This is the reason why new chip factories cost billions and fewer companies can manufacture them.
In addition to paying for buildings and machinery, companies will have to spend heavily on developing complex process measures used to make chips from plate-sized silicon wafers – which is why factories are called “fabs”.
Huge machines design projects for chips in each wafer, and then deposit and remove layers of material to make their transistors and connect them. Up to 25 wafers are moved at a time between those systems in special pods on automated overhead tracks.
Wafers take thousands of steps and up to two months to process. TSMC has set the pace for output in recent years, operating “gigafabs” sites with four or more product lines. Dan Hutcheson, vice chair of market research firm TechInsights, estimates that each site can process more than 100,000 wafers a month. It puts the capacity of Intel’s two planned $ 10 billion facilities in Arizona at about 40,000 wafers per month.
How chips are packed
After processing, the wafers are cut into individual chips. These are tested for attachment to circuit board or system parts and wrapped in a plastic package.
That move has become a new battleground, as it is even more difficult to make the transistor smaller. Companies are now stacking multiple chips or putting them side by side in one package, connecting them to work as one piece of silicone.
Where packaging a handful of chips together is now routine, Intel has developed a state-of-the-art product that uses new technology to bundle a significant 47 individual chips, some manufactured by TSMC and others as well as Intel Fabs.
What sets chip factories apart
Intel chips typically sell for hundreds to thousands of dollars. Intel introduced its fastest microprocessor for desktop computers in March, with a starting price of $ 739, for example. A piece of dust invisible to the human eye can spoil one. Fabs should therefore be cleaner than a hospital operating room and require complex systems to filter air and control temperature and humidity.
Fabs should also be impervious to any vibration, which can make expensive equipment defective. So the Fab Clean Room is built on a massive concrete slab on a special shock absorber.
The ability to move large amounts of liquids and gases is also important. The top level of Intel’s factories, which is about 70 feet high, has huge fans that help circulate air directly into the clean room below. Below the clean room are thousands of pumps, transformers, power cabinets, utility pipes and chillers that are connected to production machines.
Fabs are water-intensive operations. This is because many stages of the manufacturing process require water to clean the wafer.
Intel’s two sites in Chandler receive about 11 million gallons of water a day from a local utility. Significantly more will be needed for Intel’s future expansion, an apparent challenge for a drought-stricken state like Arizona, which has reduced its allocation of water to farmers. But farming actually uses more water than a chip plant.
Intel says its Chandler sites, which rely on three river supply and well system, reclaim about 82 percent of the fresh water they use through purification systems, ponds and other equipment. That water is sent back to the city, which operates treatment facilities funded by Intel, and redistributes it for irrigation and other non-potable uses.
Intel hopes to help increase water supply in Arizona and other states by 2030, working with environmental groups and others on projects to save and restore water for local communities.
How fabs are built
To build its future factories, Intel will need about 5,000 skilled construction workers for three years.
They have a lot to do. The excavation of the foundation is expected to remove 890,000 cubic yards of dirt, which is being removed at the rate of one dump truck per minute, said Dan Dorone, Intel’s construction chief.
The company expects to pour more than 445,000 cubic yards of concrete and use 100,000 tons of reinforced steel for the foundation – more than the world’s tallest building in Dubai, the Burj Khalifa.
Some of the cranes for construction are so large that it requires more than 100 trucks to bring the pieces to be assembled, Mr. Said Dorothy. The cranes, among other things, will lift 55-ton chillers for new fabs.
Patrick Gelsinger, who became Intel’s chief executive a year ago, has been lobbying Congress for grants for fab construction and tax credits for equipment investments. To manage Intel’s cost risk, it plans to focus on building fab “shells” that can be equipped with tools to respond to market changes.
To overcome the chip shortage, Mr. Gelsinger has to make good his plans to produce chips designed by other companies. But that is all a company can do; Products like phones and cars require components from many suppliers as well as older chips. And no country can stand alone in semiconductors. While boosting domestic production can reduce supply risks to some extent, the chip industry will rely on companies’ complex global web of raw materials, manufacturing equipment, design software, talent and specialized production.
Produced by Alana Sally