FlexICs versus silicon chips: what’s the difference?

Flexible and silicon chips will both play a part in powering devices at the edge, but where will each technology excel? Technology Editor Hellen Norman investigates.

For more than half a century, silicon has been the undisputed ruler of the electronics world, powering everything from the smartphones in our pockets to the servers managing global data. But as we move towards a world of pervasive intelligence and agentic AI systems served by item-level data, new device classes are emerging that prioritise form factor, cost and sustainability over raw performance.

This shift towards a hyper-connected world demands more than just smart devices. It requires an explosion of intelligence in everyday objects: billions of items seamlessly sensing, sharing, and responding to their environment.

That’s where FlexICs – flexible integrated circuits – come in.

FlexICs versus silicon

Although it’s the title of this blog (I know, I know…) pitting FlexICs against silicon is actually the wrong way to look at their respective merits.

Ultra-thin, physically flexible, low cost and low carbon, FlexICs aren’t trying to replace silicon. What they bring is something complementary: a radically different way of delivering intelligence to the physical world – where traditional chips are too costly, too rigid, or too resource-intensive to be viable.

The long lead times and high costs associated with silicon chip design, not to mention the supply chain issues in recent years, have led to the creation of high-specification chips that are ready for any task you throw at them. If chips are likely to take months, or even years, to reach production, they have to be suitable for a wide range of tasks – whatever it is you need them to do right now. But this means that sometimes you’re using a Ferrari where a tricycle would do, because silicon chips deliver high performance – but at high cost and with a significant carbon impact.

Flexible chips, however, can deliver ‘just enough’ performance for the specific task at hand, at a fraction of the time, cost and carbon footprint. This, along with their flexible form factor, means they can be used where it would not be cost-effective – or even possible – to use conventional electronics. It also makes them a compelling alternative to legacy silicon.

While they’re wildly different in terms of fabrication, form factor and even application, FlexICs and silicon are partners in building the next generation of electronics. Operating at opposite ends of the power envelope, together they form a more efficient, sustainable, and future-ready semiconductor ecosystem that maximises the value of every chip.

But you still want to know what’s different? OK, then. Here we go…

Material and process differences

Composition-wise, the most fundamental difference is that silicon chips are built on rigid monocrystalline wafers. FlexICs use indium gallium zinc oxide (IGZO) thin-film transistor technology (TFT) and an ultra-thin polyimide substrate on a reusable carrier. They’re less than 40 microns thick and can be flexed to a 5mm radius without losing functionality.

This core difference is obviously reflected in the manufacturing process.

For silicon, this is complex, energy intensive and requires dozens of lithographic and chemical steps, many at temperatures over 1000°C. That complexity means that silicon fabs are massive and cost billions of dollars.

By contrast, FlexIC fabrication uses a simple substrate of flexible polyimide prepared in batches on a reusable carrier by spin coating. g. The process takes less than a day, and is carried out at much lower temperatures – under 400°C.

The rest of the chip fabrication process is also much simpler than for silicon, uses around 30 steps, and is largely performed at room temperature. No process exceeds 300°C. This reduced complexity also means that the entire process is significantly faster, and can take place in a single, compact location.

Sustainability

And talking of carbon, while complexity and temperature of fabrication naturally influence a chip’s carbon footprint, there are other factors to consider.

Silicon itself is massively energy-intensive to produce. In its natural state, it tends to be bound to oxygen and so needs to be heated with carbon at around 1500°C to form carbon dioxide and silicon that’s 99.99% pure. From this, an ingot of pure silicon is produced, ready to be sliced into the wafers we know and love.

Silicon chip fabrication is also a water-intensive process. And despite the industry’s adoption of water stewardship strategies to mitigate overall withdrawal, increased demand for chips has, in many cases, seen an overall increase in water consumption.

Significantly, much of this is in the form of ultrapure water (UPW). This is a fundamental material necessary for surface cleaning, typically generated by purifying tap, surface, or reclaimed water, but with high water loss and considerable energy input in its complex production process.

In addition, silicon chip production uses up to 30 different process gases, including perfluorocarbons¹ – potent greenhouse gases with a global warming potential up to 25,000 times² higher than CO₂.

Add to this the transportation of chips and it’s fair to say that silicon isn’t winning any sustainability prizes anytime soon.

By nature of its shorter, simplified process, FlexIC production uses less energy and less water, including UPW. It also uses far fewer gases, with the majority less potent or inert, such as argon and nitrogen. It does, however, use small quantities of just one greenhouse gas – tetrafluoromethane, also known as carbon tetrafluoride (CF₄). The amount used is equivalent to just 0.000001g CO₂ per FlexIC, and 10kg per year, per fab.

When you consider this in combination with the single site fabrication, it’s not hard to see why FlexICs’ carbon footprint is significantly lower.

Silicon chips vs FlexICs

Form factor

We’ve already established that silicon chips are rigid and FlexICs … flex. But silicon is also brittle, and so will shatter if subjected to even minor mechanical stress or bending. Its thickness also creates a detectable ‘bump’.

This has implications for chips on consumer goods, which are particularly vulnerable on busy production lines or during transit. A flat label distributes pressure evenly. A ‘bump’ concentrates force onto a single, microscopic point – in this case, the silicon die.

FlexICs, by contrast, are totally flat, distributing pressure evenly. They also flex with the packaging, making them less susceptible to failure.

Performance

It goes without saying that silicon is a powerhouse. When it comes to raw processing power, it remains the gold standard. If you need to render 4K video or run complex AI models, silicon is the only choice.

But FlexICs aren’t designed to compete with high-end processors. They’re optimised for functional density and cost efficiency. They excel at specific connect, sense and compute tasks, such as:

• Near field communication (NFC) e.g. customer engagement, authentication
• Basic sensing, e.g. temperature, humidity

FlexICs aim to add ‘just enough’ intelligence to everyday objects – like a milk carton or a medicine bottle – at a price point that makes sense at scale. This, in combination with their low carbon footprint, makes them ideal for embedding imperceptibly in packaging, labels – or even directly into products – at scale.

They also lend a hand with system miniaturisation. Used as substrates with high-density interconnects or integrated passive devices, they can reduce layout complexity – and the bill of materials – to deliver high-density circuitry in a cost-effective, ultra-compact form.

FlexICs and silicon: perfect partners

Silicon remains the foundation of high-performance electronics. With their low carbon footprint and low price point, FlexICs unlock item-level intelligence, extend electronics into high volume, and mechanically constrained environments where silicon is impractical.

As intelligence moves from the cloud to the edge and ultimately to individual items, both device classes will coexist within a more heterogeneous semiconductor ecosystem that spans performance, scale and sustainability requirements.

Together, FlexICs and silicon will deliver a future where innovation and sustainability are perfectly aligned – enabling pervasive intelligence to unlock the full promise of a hyperconnected world.

Find out more on FlexIC technology