Every smartphone, car, and data center runs on a supply chain most people never think about, until it breaks.

According to the Washington Post, a worldwide chip shortage shuttered auto plants and held up electronics shipments in 2021, forcing the word “semiconductor” into the lexicon of everyday life. For anyone in international trade, logistics, or manufacturing, it’s no longer optional to know how chips are actually made, and how many countries and companies handle a single chip before it is turned into a finished product.

Read also: Customer-Centric Resilience: A New Focus for the Semiconductor Supply Chain Industry 

The semiconductor supply chain is among the most intricate in modern manufacturing. One chip might cross a dozen borders as it is manufactured, with each step taking place in a different facility specialized in one small part of an extremely long process. From raw material to finished product, this is how it works.

It Starts With Sand, But Not Just Any Sand

The chips are made from silicon, and silicon is made from quartz sand. But the sand that is used in the manufacture of semiconductors is not the kind you find on a beach. It has to be incredibly pure, with a silicon dioxide level above 99 percent, because even a tiny impurity can mess with a chip’s performance.

That high-purity quartz is then made into metallurgical-grade silicon, and subsequently converted into electronic-grade polysilicon, which is sufficiently pure for chip manufacturing. This refining stage is capital intensive and a handful of producers worldwide concentrate the activity, hence disturbances in one milling plant can affect the global industry.

Growing and Slicing the Silicon Wafer

Once the polysilicon is prepared, the producers melt it and grow it into a single crystal which is cylindrical in shape with the help of the Czochralski method. The resulting ingot is cut into thin, round slices called wafers, which are then polished until they are mirror-finished.

Wafers are the platform upon which everything else rests. The size (typically 200 mm or 300 mm in diameter) can influence the quantity of chips that can be manufactured from each wafer, which affects cost and yield. 

The production of wafers is concentrated among a few specialized companies, and similar to polysilicon refining, it is a market with minimal redundancy. When a major wafer provider is hit with an outage, chipmakers the world over feel the pinch nearly immediately.

Specialty Materials: The Overlooked Layer of the Supply Chain

Silicon wafers don’t just go into making a chip. Manufacturing involves dozens of specialized products and chemicals, such as photoresists, etching gases and metals used in building the miniature circuitry on the surface of the wafers.

One class of materials that normally needs to be paid more attention to than it does in the sputtering targets. They are precision-engineered discs or squares of metal (copper, tantalum, titanium etc.) which are employed in a process known as sputtering where ultra thin metal films are applied to the wafer. Those films eventually form the electrical connections inside the chip. 

Since the purity and uniformity of these targets has a direct impact on chip performance, sputtering target manufacturers are subject to very stringent quality control regimes, and there are not many companies worldwide that can manufacture them at the purity levels required by advanced fabrication. That concentration makes them a quiet but important pressure point in the broader supply chain, somewhat akin to the risk posed by chokepoints in polysilicon or wafers.

Chip Fabrication: Where the Real Complexity Lives

Fabrication or “fab” is the phase that most people envision when they think of chipmaking and is by far the most complex part of the process. Within a fab, wafers are subjected to hundreds of steps, sometimes more than a thousand, of lithography, etching, deposition, and doping that build layers of circuitry.

Lithography is light used to project patterns on the wafer surface, at the nanometer scale. Etching strips away unwanted material to sculpt these patterns. Deposition, such as sputtering, deposits layers of material, including the metal films above. Doping is the process of adding controlled amounts of impurities to the silicon so that electricity behaves in silicon in a certain way.

Constructing a state-of-the-art fab costs billions of dollars and can take years to get up and running, which is a big part of why the leading edge of chip fabrication is concentrated in a handful of countries. Taiwan and South Korea have some of the world’s most sophisticated fabs, and the United States, Japan, and to a growing extent parts of Europe and India, are funneling significant investment to develop more local capacity.

Packaging and Testing: Turning a Chip Into a Usable Product

After fabrication is complete, the wafer is cut into individual chips, and each one is packaged. It shields the fragile silicon die and provides the electrical pins or contacts that allow the semiconductor to be connected to other components on a printed circuit board.

This stage has traditionally been centered in Malaysia, the Philippines, China and Vietnam, where labor costs are low and packaging know-how has been honed for decades. After packaging, chips are subjected to stringent testing to identify defects before they are released to a customer.

A single bad chip in a finished product – whether it’s a car or industrial equipment – can cost tens or hundreds of thousands of dollars to track down and fix retrospectively, so testing is considered a mandatory step, and not an afterthought.

Logistics: Moving Chips Around the World, More Than Once

One fact that surprises many people outside the industry is the number of times a chip crosses international borders before it is completed. The materials could be mined in one country, refined in another, processed into wafers in a third, fabricated into chips in a fourth and then shipped once again for packaging and testing before eventually getting to the company that turns it into a finished product.

This constant ping-ponging between countries also makes semiconductor supply chains particularly vulnerable to shipping delays, bottlenecks at ports and export controls. Chips are tiny, and they weigh next to nothing when compared to their value, so air freight is used for finished products, while raw materials and equipment required to make the chips often travel by sea, introducing its own timing risks.

Why This Supply Chain Is So Vulnerable

There are a few structural aspects that make semiconductors more susceptible to disruption than other manufactured goods.

The main one is geographic concentration. Chip processing is centralized in a small number of facilities in a few countries, so a natural disaster, political turmoil, or a regional power outage in one of these countries can disrupt the world chip supply. Twi­ter In particular, Taiwan accounts for an outsized proportion of the world’s most advanced chips, which is why the region’s political stability is monitored closely by manufacturers and trade analysts well beyond the semiconductor industry itself.

The availability of raw materials also adds kipple of risk. High-purity quartz and some rare earth elements and specialty metals used in sputtering targets and other products are unevenly distributed around the world, and the mining or refining of certain of these materials is restricted to a handful of countries.

Geopolitical tensions have been rising too. Over the past several years, export controls, tariffs, and trade regulations on chip making equipment and finished semiconductors have redefined sourcing strategies, forcing companies to diversify suppliers and, in some cases, to establish redundant capacity in multiple regions ­even if doing so means higher costs.

The capacity to manufacture is rigid, itself to short-term alters. The construction or expansion of a fab takes years, so the industry can’t just ramp up capacity to meet demand when it suddenly surges, as it did during the pandemic. That imbalance between demand and capacity is what transformed a short-term disruption into a multi-year shortage.

What This Means for Global Trade Professionals

If you work in supply chain, purchasing, or international trade, the semiconductor industry offers a valuable lesson in how concentrated, specialized production can generate outsized risk. A disruption at a single wafer maker, a single advanced fab, or even a single supplier of sputtering targets or other specialty materials can throw production schedules for companies that don’t even have a direct relationship with that supplier into chaos.

That’s why more companies, and more governments, are looking closely at the origins of their components, not only the finished chip, but all the materials and processing steps behind it. Substitute suppliers, create strategic inventory cushions, and invest in domestic or regional manufacturing capacity: those are among the now more familiar solutions to the brittleness that this supply chain has revealed.

Knowing the entire path from silico­n to finished chip isn’t just about quenching your curiosity. It’s a practical business reality for any business that depends, directly or indirectly, on having a reliable supply of semiconductors, which is to say virtually every industry in global commerce.