Industries around the world have been hugely affected by global events over the past several years. Wars, famine and water shortages, political unrest and pandemics have all contributed to persistent supply constraints for some of the modern world’s most important commodities. One product that is particularly central to almost every part of our daily lives is the semiconductor, making up parts of our phones, TVs, washing machines, fridges, and cars (to name but a few).
Here we’ll take a closer look at semiconductors: how they work, why they are important, and why semiconductor manufacturing – particularly in the automotive industry – has struggled to keep up with phenomenal demand over the past few years. And whilst we have recently emerged from the worldwide shortages that led to plant shutdowns and which has delayed estimates of a 100 million car production year by around a decade, we’ll also examine why the relief might be short-lived.
So, what are semiconductors, and how are they created?
1. Semiconductors: a closer look at the basics
Imagine semiconductors as the backbone of our modern electronic world – they’re materials that have an ability to act as both conductors and insulators (i.e., they are semiconductors of electricity). Silicon is the most common semiconductor material, and it’s the foundation upon which tiny integrated circuits (ICs), that power everything from our cars to the latest large language model, are built.
As you might have guessed, manufacturing semiconductors is no easy feat. It’s a complex, precision-driven process that requires a lot of attention to detail and, of course, money. Let’s break down the key stages of the process:
- First, you create a silicon wafer, which is a thin slice of ultra-pure silicon. This wafer acts as the foundation for the IC.
- Next, you use a process called photolithography, which involves shining ultraviolet light through a patterned mask onto the wafer. This creates the desired circuit pattern on the silicon wafer.
- After this, it’s time for etching, where unwanted silicon is removed to create the specific structures needed for the IC.
- Doping then comes into play, where impurities (think of them as “performance-enhancing substances”) are intentionally added to the silicon to modify its electrical properties.
This entire process must happen in a super clean environment called a cleanroom – even the tiniest speck of dust or contamination can wreak havoc on the final product. It’s like performing surgery, everything must be spotless to ensure the best possible outcome.
2. Semiconductors and cars: a powerful connection
That’s the basics of semiconductors, so let’s talk about how they have become an essential component of modern cars. Long gone are the days when cars were just mechanical machines – today’s vehicles are packed with electronic systems that rely on semiconductors to function.
Think about engine control units, transmission control modules, and infotainment systems – all of these rely on semiconductors to work their magic. In fact, estimates suggest that a single modern car can contain up to around 3,000 semiconductors, depending on its complexity and features.
With more and more battery electric vehicles (BEVs) hitting the roads, the demand for semiconductors is only increasing. BEVs require even more advanced electronics to manage their battery systems, electric motors, and charging infrastructure. And guess what? All of these systems need semiconductors to function properly. So, as we continue to embrace alternative transportation options, the demand for semiconductors in the automotive industry will only continue to grow.
We’ve not even mentioned self-driving cars yet, which could fairly be described as computers-on-wheels. The computing power provided by semiconductors is fundamental to a self-driving car’s ability to perceive, understand, and communicate with the environment around it. Even the progress made in advanced driver-assistance systems, such as adaptive cruise control and lane assistance, is enabled by semiconductors.
With all in this mind, it’s reasonable suggest that at least one of the keys to unlocking the challenges confronting the industry has been, and will continue to be, made of silicon.
3. Semiconductor nodes: a journey through the world of miniaturization
But how can semiconductor chip technology keep up with rapid advances in electronics, whether in the automotive industry or otherwise? The answer is the ever-shrinking world of semiconductor nodes. Nodes play a huge role in shaping the performance and capabilities of the semiconductor chips that power our electronics.
A semiconductor node refers to the size of the transistors used in the fabrication of semiconductor chips. The smaller the node, the more advanced the technology – it’s like packing more and more power into a smaller and smaller package. This miniaturization allows us to cram more transistors onto a single chip, leading to increased performance and improved energy efficiency.
The size of semiconductors and the nodes that they belong to are measured in nanometres (nm). Think about the first-generation computers that took up a whole room. These early computers didn’t use semiconductors, but instead relied upon vacuum tubes as switches to carry out each computation. To give you a sense of scale, 1 centimetre is 10,000,000 (ten million) nanometers. So, if you were to measure vacuum tubes in terms of nanometers in the way that we measure modern transistors, you’d be looking at measurements in the billions of nanometers.
Over the years, we’ve seen a remarkable evolution in semiconductor node sizes. Back in the early 2000s, we had 130nm nodes, which were already quite impressive. Fast forward to today and we’re talking about the 2nm nodes – an impressive amount of progress in just a couple of decades! Progress that was predicted by the late Gordon Moore – one of the godfathers of semiconductor chips – as far back as 1965 (yes, that’s what ‘Moore’s law’ is referring to).
But why does this matter? Well, smaller nodes enable us to create more powerful and efficient electronic devices. As technology continues to advance at a breakneck pace, the demand for smaller and better nodes keeps growing. This relentless pursuit of miniaturization has pushed the limits of engineering and science, driving innovation across various industries. It’s also meant that the ability to manufacture these chips has been concentrated into the hands of just a few companies.
4. Automotive needs: why reliability has traditionally trumped cutting-edge speed
At this point, you might be wondering: if smaller nodes are so amazing, why doesn’t the automotive industry just jump on the bandwagon and go all-in on the latest and greatest nodes? The answer lies in the fact that cars have very different priorities than, say, your smartphone or computer.
The automotive industry is all about reliability and longevity. Cars need to be able to withstand a wide range of conditions – from freezing temperatures to scorching heat, from bumpy off-road adventures to high-speed highway cruises. Because of this, carmakers often prefer to use slightly older, more established nodes that have proven their ability to perform consistently under various conditions whilst meeting the strict safety requirements that cars are subject to.
Think about the semiconductors used in your computer, which could house chips with transistors as small as 5nm. These chips are designed to deliver lightning-fast performance and handle complex tasks, but they usually operate in a relatively stable environment. Now, picture a semiconductor inside a car’s engine control unit, dealing with high temperatures, vibrations, and even exposure to oil and other fluids. It’s a completely different wheelhouse (apologies)!
In this context, it’s easy to see why the automotive industry leans towards more mature nodes. They need semiconductors that can stand up to the unique challenges of a car’s environment while still delivering consistent performance. So, while cutting-edge speed might be all the rage in consumer electronics, when it comes to cars, it’s more about stability and reliability. This picture does however look set to shift slightly, as the industry moves toward electrification and self-driving it will need to consume more advanced chips.
5. Fabrication investment: a glimpse into the world of semiconductor production
Now that we’ve covered the ins and outs of semiconductor nodes and why the automotive industry prefers mature technology (for now), the next pit stop on our journey is semiconductor fabrication (or manufacture). The difficulty of fabrication explains why the automotive sector in particular struggled with recent shortages.
Setting up and running a fab (or fabrication facility) is a costly endeavour. We’re talking billions of pounds just to get the ball rolling. With that kind of investment on the line, companies must make strategic decisions about which equipment and technology they’ll invest in to keep pace with Moore’s law. As technology is always evolving, fabs often focus on what they think will be the most popular and profitable nodes five years down the line.
Here’s the thing, though: fabs tend to prioritise cutting-edge technology, chasing after the highest demand and greatest profit margins they can find. This means that while they’re busy investing in the newest and most advanced nodes, there’s often a relative underinvestment in the mature nodes that the automotive industry relies on.
To make matters worse, the recent surge in demand for semiconductors across various industries (for example the AI, data storage and web connectivity industries) has put even more pressure on fabs, stretching their resources thin and exacerbating the shortage of mature nodes. It’s like a game of musical chairs where the automotive industry is often left without a seat when the music stops.
6. Geopolitics, subsidies, and partnerships: looking toward the future
So how is the automotive industry responding to these problems? Supply chains are being restructured to give original equipment manufacturers more control. Some original equipment manufacturers are now purchasing their chips directly from fabs. Others are partnering with established manufacturers to set up their own, automotive focussed fabs. Governments and the chip industry itself are also helping, with chip manufacturers taking advantage of generous subsidies to set up fabs that aren’t focused on the cutting edge.
But threats to chip supply still loom. The US and China are locked in a chip war and the automotive industry could become one of the biggest victims. As the US restricts China’s access to the most advanced semiconductor chips, fearing nefarious use-cases in AI and military applications, China has responded by restricting the export of key raw materials involved in the semiconductor manufacturing process. China also has other cards to play, and did just that when it restricted the export of graphite, an absolutely key ingredient in BEV batteries. This tit-for-tat is expected to continue and the automotive industry is already investing in finding alternative raw material supply to China.
7. Navigating semiconductor supply chains in the automotive world
So, there you have it – a whirlwind tour of the semiconductor landscape and how it ties into the automotive industry. As we’ve seen, it’s not a simple case of supply and demand; it’s a complex web of factors involving the unique needs of carmakers, the relentless pursuit of cutting-edge technology, the strategic investment decisions made by semiconductor fabs, and the geopolitical arena.
What’s clear is that as our world becomes more interconnected and dependent on electronics, the demand for semiconductors will only continue to grow. The automotive industry should be commended for emerging from the COVID pandemic with clear remediation plans in place when it comes to chip supply, but it isn’t in the clear just yet.