With the global economy reeling from a shortage in semiconductor chips, policy makers have turned their attention to strengthening the resilience of the supply chain, recognizing the centrality of this technology to economic growth and national security. This supply chain incorporates the extremely complex and costly processes of harvesting raw materials, designing, manufacturing, packaging, and shipping that are now carried out across the world to produce the variety of semiconductors that go on to live in our toasters, smartphones, computers, buildings, and cars.
To better understand the vulnerabilities in the semiconductor supply chain, we take a closer look at each step in a chip’s life cycle.
What Are Semiconductors and What Materials Are Required for Their Manufacture?
A semiconductor is a physical substance designed to manage and control the flow of current in electronic devices and equipment. The name “semiconductor” comes from the fact that a semiconductor chip is made from material that is neither entirely conductive of electricity nor fully insulating. They are typically created by adding impurities to, or “doping”, elements such as silicon, germanium, or other pure elements to alter their conductivity. A “wafer” of silicon or another semiconductor material is then edited to create complex circuits, which are capable of completing computing tasks. Examples of conventional devices and components built by using semiconductors include computer memory, integrated circuits, diodes, and transistors.
Semiconductors are typically made from one of two elements whose molecular structure when crystalized is secure enough to facilitate and regulate an electrical current: germanium and silicon. Germanium, the element
on which the first transistor was developed, is a relatively rare and, therefore, expensive semiconductor. Currently, the U.S. maintains
greater than 50% reliance on imported germanium from Belgium and China. U.S.
germanium reserves are estimated to be near 2,500 tons, significantly behind China, which leads annual production at 85,000 tons. Silicon, however, is the
second-most abundant element on Earth, accounting for roughly
28% of the Earth’s crust. While pure silicon, an ideal semiconductor, does not occur naturally on Earth, it can be synthesized by
superheating silicon dioxide with carbon materials.
How are Semiconductors Designed?
While the science that undergirds the logic of a semiconductor’s basic functions is relatively simple, the mass-manufacturing of such small and delicate electrical components requires a complex design process. According to Synopsys, a chip design and verification firm, the process can be broken down into
five steps.
The first of these is the
architectural design of the chip, wherein the parameters of the chip are determined including its size, desired function, level of power consumption, and preferred cost.
Next is the
logic and circuit design. After the parameters are outlined, engineers begin translating the required functions into circuit logic. Today, this process is done on automated logic simulators to verify that everything is in order before production.
Third is the
physical design phase. Here, the circuit logic is mapped onto a silicon wafer. Essentially, this is a plan of where each transistor, diode, or other component will sit on the chip.
Finally, the
verification and sign-off phases are used to verify whether the designed chip is manufacturable and whether it can withstand the physical stresses of its assigned function. Specifically, added resistance from wiring, signal crosstalk, and variability are all factors to be considered.
How are Semiconductors Manufactured, Packaged, and Shipped?
The process used to print circuits onto silicon-crystal wafers is called “
photolithography.” The silicon wafer is coated with a layer of light-resistant material called the “photoresist.” Then, using photolithography, the photoresist is weakened or hardened in certain pre-determined regions by exposing it to UV radiation (light). During a step called “etching” the weakened sections of photoresist are removed. The exposed silicon crystal is then “doped” with impurities to alter its conductivity and create microelectronic components like transistors and diodes. Thousands of these circuits can be printed onto a single wafer side by side, and the wafer will go through a
series of other complex steps before it is completed. Finally, each “die” of semiconductor is sliced from the wafer using precision sawing or laser technology.
Silicon chips are
extremely fragile microelectronics that can be irrevocably damaged by excessive vibration, temperature fluctuations, or even static electricity. This has spurred the inception of an entire new industry adjacent to semiconductor manufacturing: chip packaging. Packages are meant to protect the semiconductor and facilitate its connection to a larger circuit or board. While packaging innovations and production used to be an entirely separate process, chip manufacturers themselves have begun developing expertise and integrating it into the manufacturing process. The fragility of the semiconductors also affects their shipping process, which must be done by specialized logistics firms.
An Interdependent and International Supply Chain
Because semiconductors are extremely complex products to design and manufacture, a highly specialized global supply chain has developed over the past few decades. The supply chain has four main components: sourcing of raw materials, design, manufacture, and packaging. Complicating matters, different countries
specialize in each component.
In terms of raw materials, China is the world’s leading supplier of silicon, accounting for an estimated
64% of total silicon materials in 2019 according to the U.S. Geological Survey.
With regards to the design phase, the United States is still the world’s leader. Seven of the top ten integrated circuit design companies by annual revenue are
headquartered in the United States. Often, these firms operate via a “fabless” business model, whereby they design the chips,
then license the intellectual property (IP) to firms around the world to produce them. The name “Silicon Valley” is therefore a holdover from the days that the U.S. dominated this global industry.
Yet, for a few decades now, Taiwan has been the undisputed global leader in terms of market share for semiconductor manufacture. Taiwan Semiconductor Manufacturing Co. (TSMC) alone accounts for roughly
54% of all global foundry revenue. South Korea’s Samsung trails TSMC at 17%, while Global Foundries, the largest U.S.-based manufacturing firm, controls 7% of the market. Adjacent to manufacturing, the Dutch firm Advanced Semiconductor Materials Lithography (ASML) is currently the
only company in the world that builds lithographic machines powerful enough for the most sophisticated chips. Each machine is comprised of over 100,000 individual parts and costs roughly $150 million. The firm ASML is expected to hit
$28-$35 billion in annual revenue by 2025. This makes the Netherlands a key node in the global semiconductor supply chain.
According to the Center for Security and Emerging Technology, the semiconductor
assembly and packaging market is extremely diverse, with firms from the United States, Japan, China, South Korea, Singapore, and the Netherlands specializing in inspecting wafers, “dicing” them into individual chips, packaging them, and integrating them into larger electronic components.
These realities mean that each node on the semiconductor supply chain is interdependent on one another. States rely on international trade to move materials, equipment, and products around the world to facilitate the manufacture of this key ingredient in the global high-tech economy.
Full self-sufficiency, therefore, is a difficult and expensive goal to achieve. According to a
study conducted by the Boston Consulting Group, if regional supply chains (U.S., East Asia, China, Europe, and others) wanted to reach total self-sufficiency, it would require $1 trillion in incremental up-front investment to meet current levels of semiconductor consumption. It would also result in a
35% to 65% overall increase in semiconductor prices and higher costs of electronic devices for end users. The study also concluded that even to meet projected semiconductor demand in today’s globally connected market, the industry will need to invest at least $3 trillion over the next ten years in R&D and capital expenditure alone.
Recent Strains and Efforts to Alleviate Them
While disruptions in the semiconductor supply chain caused by COVID-19 brought awareness of the global chip shortage into the public domain, the issue predates the pandemic. Given their ubiquity in the devices that power the modern digital economy, demand for chips
has been skyrocketing for decades, and will continue to increase along with demand for technologies like mobile phones and electric vehicles according to a recent report by Accenture. As the industry struggles to keep pace, ‘black swan’ events like earthquakes, floods, and fires (such as
the blaze at the Renesas chip manufacturing plant in Japan) have had disastrous cascading effects. Pandemic-related lockdowns and border closures only exposed and aggravated the supply chain’s inherent fragility caused by its international nature and intensifying global demand.
In response to these developments, policy makers around the world have unveiled plans to bolster their domestic manufacture of semiconductors to mitigate the worst effects of supply chain breakdowns. In Europe, the European Commission has
drafted legislation to mobilize over €43 billion in public and private funds to double its share of the global semiconductor manufacturing market by 2030. Meanwhile, in the United States, lawmakers continue to debate the
CHIPS for America Act and the
FABS Act, which provide lump-sum and tax-based incentives for chip manufacturers to “onshore” their operations. These efforts, while they have yet to take effect, will be the first steps in strengthening regional and national resiliency against future crises plaguing the supply chain of this critical technology.
Gabrielle Athanasia is a Program Coordinator and Research Assistant with the Renewing American Innovation Project at the Center for Strategic and International Studies in Washington, DC.
Gregory Arcuri is a research intern with the Renewing American Innovation Project at the Center for Strategic and International Studies in Washington, DC.
The Perspectives on Innovation Blog is produced by the Renewing American Innovation Project at the Center for Strategic and International Studies (CSIS), a private, tax-exempt institution focusing on international public policy issues. Its research is nonpartisan and nonproprietary. CSIS does not take specific policy positions. Accordingly, all views, positions, and conclusions expressed in this publication should be understood to be solely those of the author(s).
