Integrated Photonics: When It Makes Commercial Sense — and When It Doesn’t
- ankelohmann
- Jan 12
- 5 min read
Integrated photonics has advanced beyond initial research environments and telecommunications. Today, data centres and artificial intelligence are primary catalysts for growth. Additionally, photonic integrated circuits (PICs) are increasingly utilised in sensing, healthcare, and quantum technologies, offering more compact, stable, and scalable solutions compared to traditional bulk optics.
Yet integrated photonics is not a universal replacement for lenses, mirrors, and discrete sources. For many products, miniature bulk-optic systems remain the better business choice. The real question is not whether you can integrate, but whether you should.
This article looks at integrated photonics through a commercial lens: volume, development time, cost, and manufacturability.
What Is integrated photonics and why it matters
Integrated photonics involves fabricating optical circuits on semiconductor chips, similar to how electronic circuits are made. Instead of using discrete lenses, mirrors, and fibres, integrated photonics combines these functions into waveguides and components etched on a chip. The promise is the same one that drove microelectronics:
Smaller size and weight
Higher stability and alignment-free operation
Mass manufacturing at a low per-unit cost
Easier system integration
For applications such as portable sensors, point-of-care diagnostics, or scalable quantum processors, these benefits are compelling. A single photonic chip can replace an optical bench full of components.
But these benefits only materialise when certain commercial conditions are met.
Where integrated photonics shines
1. High-volume sensing
Gas sensors, environmental monitors, LiDAR, biosensors, and industrial spectroscopy increasingly benefit from photonic integration.
When volume is high (tens of thousands to millions of units), the economics change dramatically. The cost of designing a photonic chip and setting up fabrication can be high, but the cost per chip drops steeply once production ramps up. Integrated photonics also enables:
Easy alignment is possible since most components are fabricated on the chip
Smaller, sealed systems with better long-term stability
Easier embedding into consumer or industrial products
For a company targeting large-volume markets, photonic integration is often the only path to a competitive product.
2. Healthcare and diagnostics
Healthcare devices in clinical and point-of-care settings must be compact, stable, reproducible, and compliant with strict regulations. PICs enable functions like optical sensing, interferometry, and spectroscopy on a single chip, reducing alignment issues and variability, ensuring consistent large-scale performance. Validated PIC designs can be produced at scale with tight tolerances, making them ideal for diagnostic platforms that incorporate chips with microfluidics and electronics into compact systems. Although development and validation are costly and time-consuming, approved photonic hardware can be reused across multiple assays and variants, supporting long-term and widespread deployment.
3. Quantum technologies
Quantum photonics is perhaps the clearest case where bulk optics reach a limit. Quantum systems need extreme phase stability, precise interferometry, and scalable replication.
Even at low volumes, quantum companies often accept high development costs because scalability and system complexity cannot realistically be achieved with tabletop optics.
Photonic Integration Consideration: When integrated photonics is commercially viable
Integrated photonics makes sense when most of the following are true:
You need scale
If you expect to ship thousands or millions of units, integration wins. Even if each chip costs a few euros to fabricate, it will be cheaper than assembling and aligning dozens of optical parts by hand or with robots.
You need compactness and robustness
Portable, wearable, or embedded systems almost always favour integrated photonics. Mechanical alignment, vibration sensitivity, and thermal drift make miniature bulk optics expensive to support over time.
Your product lifetime is long
If your product will be sold and iterated over many years, the upfront investment in chip design is amortised over large revenue. Telecom, medical devices, and sensors all fall into this category.
You can tolerate long development cycles
Designing a photonic integrated circuit typically takes 6–18 months for a first working version, followed by multiple fabrication cycles. If your market allows this, integration is viable.
When bulk optics are still the better business choice
Integrated photonics is not always the right answer.
Low-volume or bespoke systems
If you are building 10, 50, or even a few hundred systems, bulk optics is almost always cheaper. You can:
Buy off-the-shelf components
Assemble and test quickly
Modify designs without refabricating chips
The non-recurring engineering cost of photonic chip design can easily reach hundreds of thousands of euros. That cannot be justified for small volumes.
Rapidly evolving or uncertain designs
Early-stage R&D and fast-moving markets benefit from flexibility. Changing a lens, a wavelength, or a beam path in bulk optics is a quick process. Changing a photonic chip means months of delay and another fabrication run.
If your product definition is still shifting, integration is risky.
Extreme performance at low volume
Some applications need ultra-high power, ultra-low noise, or unusual wavelengths. Bulk optics can handle regimes that are still difficult or expensive on chips. If performance is the main differentiator and volume is low, traditional optics often win.
The real cost equation
Integrated photonics follows the same economics as electronics:
Factor | Integrated Photonics | Bulk / Miniature Optics |
Development cost | High upfront | Low to moderate |
Time to first product | Long (months to years) | Short (weeks to months) |
Cost per unit at low volume | High | Lower |
Cost per unit at high volume | Very low | High |
Reproducibility | Excellent | Variable |
Scalability | Excellent | Poor |
The mistake people often make is focusing solely on the per-unit chip cost. The real decision hinges on whether you can sell enough units to recover the development investment.
Balancing design time, development cost, and fabrication expense
Developing integrated photonics products involves several cost and time factors:
Design Time: Creating photonic circuit layouts and performing optical simulations can take months, especially for complex, multi-function chips that require iteration and validation.
Development Cost: Upfront expenses include mask sets for photonic fabrication (typically ranging from tens to hundreds of thousands of dollars in multi-project runs), design tools, and engineering resources.
Fabrication Cost: Foundry runs have minimum fees, and even shared wafer slots can cost between roughly $10,000 and $100,000 per batch, depending on the platform technology and features.
Testing and Packaging: Precision optical and electrical packaging, plus detailed performance testing, significantly adds to development costs and schedules.
Companies must evaluate these factors against expected production volume and product price. A device with a relatively high selling price and large volume can absorb high upfront costs over its lifetime, whereas very low-volume niche products may not be economically viable given these development and manufacturing investments.
Practical tips for businesses considering integrated photonics
Start with a clear understanding of the target volume and price point.
Use prototyping with bulk optics or hybrid solutions before committing to photonic chip fabrication.
Collaborate with foundries offering multi-project wafer runs to reduce initial costs.
Plan for packaging and testing early in the design process.
Consider modular designs that allow for upgrades without requiring a full chip redesign.
Evaluate the total cost of ownership, including maintenance and scalability.
Integrated photonics offers powerful advantages for sensing, healthcare, and quantum technologies, especially when miniaturisation, performance, and integration matter. It becomes commercially viable when production volumes are high, optical functions are complex, and product lifecycles justify upfront investments. Conversely, low volumes, rapid design changes, or simple functions favour traditional bulk optics.
If you want an introduction to PICs and guidance on when to use them, join the PIC Bootcamp Explorer module. The next session will take place on 15 January. It is free and includes a panel where you can pose your questions. This Explorer session is organised by Optica in collaboration with our PIC Bootcamp partner, Cornerstone (University of Southampton). Explorer
Comments