Silicon photonics and photonic integrated circuits (PICs) have moved decisively from a promising technology to a structural necessity of modern computing. The driver is artificial intelligence. AI training and inference require enormous volumes of data to move between accelerators, servers and racks at very low latency, and the copper interconnects that served the industry for decades have reached their physical limits ? an "interconnect bottleneck" in which expensive, power-hungry accelerators sit idle waiting for data. Photonics is the industry's answer: photons travel faster, lose less signal over distance, and carry more information per channel. PICs bring those advantages onto silicon chips manufactured with the established CMOS infrastructure of the semiconductor industry.
Optical transceivers remain the engine of the market. The data rate has doubled every few years, and 2026 has seen the commercialisation of 1.6 terabit-per-second transceivers, with 3.2T expected to sample around 2027 and ramp toward 2028. As rates climb, even the short copper trace between an optical engine and a switching or accelerator ASIC limits performance, which is why co-packaged optics (CPO) ? relocating the optics onto the ASIC substrate ? has become the central packaging story of the decade. Industry forecasts suggest CPO could reach roughly 35% of AI-data-centre optical modules by 2030.
The competitive landscape reflects this momentum. Foundries are central: TSMC's COUPE platform, developed alongside NVIDIA for the Quantum-X and Spectrum-X photonic switches, has become a reference point, while Samsung Foundry has formally entered silicon photonics with a completed process design kit, a 300mm platform, a major optical-module order, and a turnkey CPO roadmap targeted for 2029. Consolidation has been intense. Marvell acquired plasmonics-based modulator developer Polariton Technologies to extend its optical roadmap to 3.2T and beyond; Credo agreed to acquire DustPhotonics for approximately $750 million to bring silicon-photonic PICs in-house; and Ciena acquired Nubis Communications for co-packaged optical engines. Independent design houses remain well funded ? OpenLight extended its Series A with an additional $50 million for standards-based 1.6T and 3.2T reference PICs.
Material diversity distinguishes PICs from logic chips. Silicon dominates on CMOS compatibility and scale, but as an indirect-bandgap semiconductor it cannot emit light efficiently, so it is paired with indium phosphide for lasers and detectors. Thin-film lithium niobate, with its low loss and strong electro-optic effect, is emerging for high-performance modulation and quantum systems; barium titanate and silicon nitride add further options. Beyond datacom, telecommunications, sensing and LiDAR, and an increasingly well-funded quantum-photonics segment broaden the demand base.
The supply chain is shifting too: optical-module assembly has concentrated in Southeast Asia, high-value lasers remain with US and Japanese suppliers, and indium-phosphide raw material is concentrated in China, making opportunity and strategic risk tightly coupled. According to industry projections, the silicon photonics and PIC market for transceivers and quantum technologies is set to grow strongly through 2036, led overwhelmingly by AI-driven optical interconnect.
The Global Silicon Photonics and Photonic Integrated Circuits Market 2026-2036 is a comprehensive market and technology assessment of one of the fastest-growing segments of the semiconductor industry. As artificial intelligence and high-performance computing push copper interconnect past its physical limits, silicon photonics has become the structural solution to the data-centre interconnect bottleneck. This report provides an in-depth, independent analysis of the technologies, materials, supply chains, applications and market trajectory of photonic integrated circuits over the coming decade.
The report opens with the fundamentals ? what PICs are, how they differ from electronic integrated circuits, their advantages and challenges, and the key components including modulators, lasers, waveguides and detectors. It examines every major material platform, benchmarking silicon and silicon-on-insulator, indium phosphide, silicon nitride, thin-film lithium niobate, barium titanate and electro-optic polymers, and assesses manufacturing, integration and packaging, including a detailed treatment of co-packaged optics and the TSMC COUPE and Samsung Foundry platforms.
A dedicated analysis covers optical transceivers ? the industry's killer application ? tracing the roadmap from 800G through the 1.6T transceivers commercialised in 2026 to 3.2T and beyond. The report addresses the shift from pluggable optics to co-packaged optics, the divergent NVIDIA and Broadcom CPO ecosystems, and the emerging "wide-and-slow" MicroLED optical interconnect architecture as a response to the chip-edge "beachfront" density crisis. Further chapters examine photonic engines for AI and neuromorphic computing, and a substantial assessment of photonic integrated circuits for quantum computing, quantum communications and quantum sensing.
The report delivers a deep supply-chain analysis from EDA and foundries to OSAT, covering the shift of optical-module assembly to Southeast Asia, indium-phosphide wafer supply, the EML laser shortage, and silicon photonics in Greater China. It includes extensive ten-year market forecasts in units, value and wafers ? covering the total PIC market, datacom transceivers, cost-per-gigabit, AI accelerator shipments, co-packaged optics, MicroLED interconnect, the quantum PIC market, and a breakdown by material platform.
Based on extensive research and interviews with industry experts, the report also profiles the leading and emerging companies across the value chain, capturing the wave of consolidation reshaping the industry ? including the Marvell-Polariton, Credo-DustPhotonics and Ciena-Nubis acquisitions and major fundraising rounds. It offers analyst insight, technology readiness assessments and clear forecasts, providing essential intelligence for component suppliers, foundries, system integrators, hyperscalers, investors and anyone seeking to understand the future of photonic integrated circuits.
Contents include:
- Executive summary: major deals, definitions, market opportunity, the copper wall, roadmap for photonics in data centres, analyst opinion
- Introduction and key concepts: integrated circuits, photonics versus electronics, advantages and challenges of PICs
- Key components of a photonic integrated circuit: component requirements, transceiver component breakdown, TSMC COUPE PDK
- Light sources and detectors: compound semiconductor lasers, EELs, VCSELs, CPO ultra-high-power laser requirements, EML shortages, photodetectors
- Modulators: Mach-Zehnder, micro-ring and electro-absorption modulators, SiGe EAMs, EO-polymer modulators
- Passive devices: PIC architecture, waveguides, optical I/O, coupling and component density
- Materials and manufacturing: wafers, integration schemes, SOI, silicon nitride, indium phosphide, organic polymer, thin-film lithium niobate, barium titanate, materials benchmarking
- Supply chain and market analysis: photonics and InP supply chains, foundries, optical modules, Southeast Asia shift, NVIDIA and Broadcom CPO ecosystems, Greater China, regulatory considerations
- Photonics for data centres: scale-up and scale-out networks, the bottleneck gap, pluggables to co-packaged optics, CPO applications, roadmap
- MicroLED optical interconnect: the beachfront crisis, wide-and-slow architecture, GaN-on-silicon, application analysis
- Photonic engines and accelerators for AI and neuromorphic compute, programmable photonics
- Photonic integrated circuits for quantum computing, quantum networks and quantum sensing
- Market forecasts: total PIC market, datacom transceivers, cost per gigabit, AI accelerator shipments, co-packaged optics, MicroLED interconnect, quantum PIC market, market by material
- Company profiles including ACCRETECH, AEPONYX, Aledia, ALLOS Semiconductors, Amkor, Analog Photonics, ASE, Avicena, Ayar Labs, Black Semiconductor, Broadcom, Broadex, Cambridge Industries Group, CEA-Leti, Celestial AI, Centera Photonics, Ciena, Cisco, Coherent, CompoundTek, Credo, CyberRidge, DustPhotonics, EFFECT Photonics, EVG, GlobalFoundries, HD Microsystems, Henkel, HyperLight, Infineon, Infleqtion, Intel, iPronics, JCET Group, JSR Corporation, Lightelligence, Lightium, Lightmatter, Lightsynq Technologies, Lightwave Logic, LioniX, LIPAC, LPKF, Lumentum, Lumiphase, MACOM, Marvell and more.....
Table of Contents
1 PURPOSE AND SCOPE OF THIS REVISION
2 EXECUTIVE SUMMARY
- 2.1 Market Overview
- 2.2 Electronic and Photonic Integration Compared
- 2.3 Silicon Photonic Transceiver Evolution
- 2.4 Market Map
- 2.5 Global Market Trends in Silicon Photonics
- 2.6 Competing and Complementary Photonics Technologies
- 2.6.1 Metaphotonics
- 2.6.2 III-V Photonics
- 2.6.3 Lithium Niobate Photonics
- 2.6.4 Polymer Photonics
- 2.6.5 Plasmonic Photonics
- 2.7 Potential of Photonic AI Acceleration
- 2.8 The Copper Wall and the Beachfront-Density Crisis
- 2.9 Manufacturing Capacity Shifts to Southeast Asia
- 2.10 Commercial deployment of silicon photonics
- 2.11 Co-Packaged Optics
- 2.11.1 Divergent CPO Ecosystems: NVIDIA and Broadcom
- 2.11.2 The TSMC COUPE Packaging Platform
- 2.12 Manufacturing challenges
- 2.13 The Market Opportunity
- 2.14 Regional Strengths & Research Focus
3 INTRODUCTION TO SILICON PHOTONICS
- 3.1 What is Silicon Photonics?
- 3.1.1 Definition and Principles of Silicon Photonics
- 3.1.2 Comparison with traditional technologies
- 3.1.3 Silicon and Photonic Integrated Circuits
- 3.1.4 Optical IO, Coupling and Couplers
- 3.1.5 Emission and Photon Sources/Lasers
- 3.1.6 Detection and Photodetectors
- 3.1.7 Compound Semiconductor Lasers and Photodetectors (III-V)
- 3.1.8 Modulation, Modulators, and Mach-Zehnder Interferometers
- 3.1.8.1 New modulator technologies
- 3.1.9 Light Propagation and Waveguides
- 3.1.10 Optical Component Density
- 3.2 Advantages of Silicon Photonics
- 3.3 Applications of Silicon Photonics
- 3.4 Comparison with Other Photonic Integration Technologies
- 3.5 Evolution from Electronic to Photonic Integration
- 3.6 Silicon Photonics vs Traditional Electronics
- 3.7 Modern high-performance AI data centers
- 3.8 Core Technology Components
- 3.8.1 Optical IO, Coupling and Couplers
- 3.8.2 Emission and Photon Sources/Lasers
- 3.8.2.1 III-V Integration Challenges
- 3.8.2.2 Laser Integration Approaches
- 3.8.3 Detection and Photodetectors
- 3.8.4 Modulation Technologies
- 3.8.4.1 Mach-Zehnder Interferometers
- 3.8.4.2 Ring Modulators
- 3.8.4.3 Micro-Ring Modulators as a Competitive Differentiator
- 3.8.5 Light Propagation and Waveguides
- 3.8.6 Optical Component Density
- 3.9 Basic Optical Data Transmission
- 3.10 Silicon Photonic Circuit Architecture
4 MATERIALS AND COMPONENTS
- 4.1 Silicon
- 4.1.1 Silicon as a Photonic Material
- 4.1.1.1 Optical Properties of Silicon
- 4.1.1.2 Fabrication Processes for Silicon Photonics
- 4.1.2 Silicon-on-insulator (SOI)
- 4.1.2.1 SOI Manufacturing Process
- 4.1.2.2 Key SOI Players
- 4.2 Germanium
- 4.2.1 Germanium Integration in Silicon Photonics
- 4.2.2 Germanium Photodetectors
- 4.2.3 Germanium-on-Silicon Modulators
- 4.3 Silicon Nitride
- 4.3.1 Silicon Nitride (SiN) in Photonics Integrated Circuits
- 4.3.2 Optical Properties and Fabrication of SiN
- 4.3.3 SiN Modulator Technologies
- 4.3.4 SiN Applications in Photonics Integrated Circuits
- 4.3.5 Advances in SiN Modulator Technologies
- 4.3.6 SiN-based Waveguides and Devices
- 4.3.7 SiN Performance Analysis
- 4.3.8 Applications of SiN in Photonics
- 4.3.9 SiN PIC Players
- 4.3.10 SiN Key Foundries
- 4.4 Thin Film Lithium Niobate (TFLN)
- 4.4.1 Overview
- 4.4.2 Lithium Niobate on Insulator (LNOI)
- 4.4.2.1 Overview of LNOI Technology
- 4.4.2.2 Characteristics and Properties of LNOI
- 4.4.2.3 LNOI Fabrication Processes
- 4.4.2.4 LNOI-based Modulator and Switch Technologies
- 4.4.2.5 Trends Toward Higher Speed and Improved Power Efficiency
- 4.4.2.6 High-Speed LNOI Modulators
- 4.4.2.6.1 Energy-Efficient LNOI Devices
- 4.4.2.6.2 Emerging LNOI Device Technologies
- 4.5 Indium Phosphide
- 4.5.1 Indium Phosphide (InP) Integration
- 4.5.1.1 InP as a Direct Bandgap Semiconductor
- 4.5.1.2 InP-based Active Components
- 4.5.1.3 Hybrid Integration of InP with Silicon Photonics
- 4.5.2 InP PIC Players
- 4.6 Barium Titanite and Rare Earth metals
- 4.6.1 Barium Titanate (BTO) Modulators
- 4.7 Organic Polymer on Silicon
- 4.7.1 Polymer-based Modulators
- 4.8 Wafer Processing
- 4.8.1 Wafer Sizes by Platform
- 4.8.2 Processing Challenges
- 4.8.3 Yield Management
- 4.9 Hybrid and Heterogeneous Integration
- 4.9.1 Monolithic Integration
- 4.9.2 Hybrid Integration
- 4.9.3 Heterogeneous Integration
- 4.9.4 III-V-on-Silicon
- 4.9.5 Bonding and Die-Attachment Techniques
- 4.9.6 Monolithic versus Hybrid Integration
5 ADVANCED PACKAGING TECHNOLOGIES
- 5.1 Evolution of Packaging Technologies
- 5.1.1 Traditional Packaging Approaches
- 5.1.2 Advanced Packaging Roadmap
- 5.1.3 Key Performance Metrics
- 5.2 2.5D Integration Technologies
- 5.2.1 Silicon Interposer Technology
- 5.2.2 Glass Interposer Solutions
- 5.2.3 Organic Substrate Options
- 5.3 3D Integration Approaches
- 5.3.1 Through-Silicon Via (TSV)
- 5.3.1.1 TSV Manufacturing Process
- 5.3.1.2 TSV Challenges and Solutions
- 5.3.2 Hybrid Bonding Technologies
- 5.3.2.1 Cu-Cu Bonding
- 5.3.2.2 Direct Bonding
- 5.4 Co-Packaged Optics (CPO)
- 5.4.1 CPO Architecture Overview
- 5.4.2 Benefits and Challenges
- 5.4.3 Integration Approaches
- 5.4.3.1 2D Integration
- 5.4.3.2 2.5D Integration
- 5.4.3.3 3D Integration
- 5.4.4 Thermal Management
- 5.4.5 Optical Coupling Solutions
- 5.5 Optical Alignment
- 5.5.1 Active vs Passive Alignment
- 5.5.2 Coupling Efficiency
- 5.6 Manufacturing Challenges
6 MARKETS AND APPLICATIONS
- 6.1 Datacom Applications
- 6.1.1 Data Center Architecture Evolution
- 6.1.2 Transceivers
- 6.1.3 Artificial intelligence (AI) and machine learning (ML)
- 6.1.4 Pluggable optics
- 6.1.5 Linear drive and linear pluggable optics (LPO)
- 6.1.6 Interconnects
- 6.1.6.1 PIC-based on-device interconnects
- 6.1.6.2 Advanced Packaging and Co-Packaged Optics
- 6.1.6.2.1 Glass materials
- 6.1.6.2.2 Co-Packaged Optics
- 6.1.6.3 Photonic Engines and Accelerators
- 6.1.6.3.1 Photonic processing for AI
- 6.1.6.3.2 Convergence with software
- 6.1.6.3.3 Photonic field-programmable gate arrays (FPGAs)
- 6.1.6.4 Photonic Integrated Circuits for Quantum Computing
- 6.1.6.4.1 Photonic qubits
- 6.1.7 Optical Transceivers
- 6.1.7.1 Architecture and Operation
- 6.1.7.2 Market Players
- 6.1.7.3 Technology Roadmap
- 6.1.8 Co-Packaged Optics for Switches
- 6.1.8.1 CPO vs Pluggable Solutions
- 6.1.8.2 Power and Performance Benefits
- 6.1.8.3 Implementation Challenges
- 6.1.9 Data Center Networks
- 6.1.10 High-Performance Computing
- 6.1.10.1 On-Device Interconnects
- 6.1.10.2 Chip-to-Chip Communication
- 6.1.10.3 System Architecture Impact
- 6.1.11 Chip-to-Chip and Board-to-Board Interconnects
- 6.1.12 Ethernet Networking
- 6.2 Telecommunications
- 6.2.1 5G/6G Infrastructure
- 6.2.2 Bandwidth Requirements
- 6.2.3 Long-Haul and Metro Networks
- 6.2.4 5G and Fiber-to-the-X (FTTx) Applications
- 6.2.5 Optical Transceivers and Transponders
- 6.3 Sensing Applications
- 6.3.1 Lidar and Automotive Sensing
- 6.3.1.1 Photonic Integrated Circuit-based LiDAR
- 6.3.2 Chemical and Biological Sensing
- 6.3.3 Optical Coherence Tomography
- 6.4 Artificial Intelligence and Machine Learning
- 6.4.1 AI Data Traffic Requirements
- 6.4.2 Silicon Photonics for AI Accelerators
- 6.4.3 Photonic Processors
- 6.4.4 Photonic Processing for AI
- 6.4.5 Programmable Photonics
- 6.4.6 Neural Network Applications
- 6.4.7 Future AI Architecture Requirements
- 6.5 Quantum Computing and Communication
- 6.5.1 Quantum Photonic Requirements
- 6.5.2 Integration Challenges
- 6.5.3 Photonic Platform Quantum Computing
- 6.5.4 PICs for Quantum systems
- 6.5.5 Operational cycle of photonic quantum computers
- 6.5.6 Market Players and Development
- 6.6 Biophotonics and Medical Diagnostics
- 6.7 Future Applications
7 MICROLED OPTICAL INTERCONNECT
- 7.1 Introduction and the Beachfront Crisis
- 7.1.1 Why density, not speed, is the new constraint
- 7.1.2 The link dilemma
- 7.2 The MicroLED Interconnect Architecture
- 7.2.1 Wide-and-slow versus narrow-and-fast
- 7.2.2 Operational mechanism and link architecture
- 7.2.3 Challenges of the MicroLED approach
- 7.3 MicroLEDs and the GaN-on-Silicon Materials Question
- 7.4 Application Analysis
- 7.5 MicroLED Interconnect Market Forecast
8 GLOBAL MARKET SIZE
- 8.1 Global Silicon Photonics and Photonic Integrated Circuits Market Overview
- 8.1.1 Market Size and Growth Trends
- 8.1.2 Market Segmentation by Application
- 8.1.3 Server Boards, CPUs and Accelerators
- 8.1.4 Modules & PICs (Dies) Market Forecast 2023-2035
- 8.1.5 SOI Wafers for Silicon Photonics
- 8.1.6 LPO & New Modulator Materials Market Forecast 2023-2035
- 8.2 Datacom Applications
- 8.2.1 Market Forecast
- 8.2.1.1 Datacom and Telecom Modules and PICs
- 8.2.1.2 PIC Transceivers for AI
- 8.2.1.3 PIC Transceiver Pricing
- 8.2.2 PIC Transceiver Cost per Gigabit
- 8.2.3 PIC Datacom Transceiver Market
- 8.2.4 Datacom Transceiver Revenue by Customer Type
- 8.2.5 Key Drivers and Restraints
- 8.3 Co-Packaged Optics
- 8.4 Telecom Applications
- 8.4.1 Market Forecast
- 8.4.1.1 PIC-based Transceivers for 5G and 6G
- 8.4.2 Key Drivers and Restraints
- 8.5 Sensing Applications
- 8.5.1 Market Forecast
- 8.5.2 Key Drivers and Restraints
- 8.6 Photonic Integrated Circuit Market, by Material
9 SUPPLY CHAIN ANALYSIS
- 9.1 Foundries and Wafer Suppliers
- 9.1.1 CMOS Foundries
- 9.1.2 Specialty Photonics Foundries
- 9.1.3 Indium Phosphide Wafer Supply
- 9.2 Integrated Device Manufacturers (IDMs)
- 9.2.1 Fabless Companies
- 9.2.2 Fully Integrated Photonics Companies
- 9.3 Foundries and Wafer Suppliers
- 9.4 Packaging and Testing
- 9.4.1 Chip-Scale Packaging
- 9.4.2 Module-Level Packaging
- 9.4.3 Testing and Characterization
- 9.4.4 Optical Module Assembly: The Shift to Southeast Asia
- 9.4.5 The EML Laser Shortage
- 9.5 System Integrators and End-Users
- 9.5.1 CPO Partner Ecosystems: NVIDIA and Broadco
10 TECHNOLOGY TRENDS
- 10.1 Laser Integration Techniques
- 10.1.1 Direct Epitaxial Growth
- 10.1.2 Flip-Chip Bonding
- 10.1.3 Hybrid Integration
- 10.1.4 Advances and Challenges
- 10.2 Modulator Technologies
- 10.2.1 Silicon Modulators
- 10.2.2 Germanium Modulators
- 10.2.3 Lithium Niobate Modulators
- 10.2.4 Polymer Modulators
- 10.2.4.1 Tower Semiconductor and Lightwave Logic EO-Polymer
- 10.3 Photodetector Technologies
- 10.3.1 Silicon Photodetectors
- 10.3.2 Germanium Photodetectors
- 10.3.3 III-V Photodetectors
- 10.4 Waveguide and Coupling Innovations
- 10.4.1 Silicon Waveguides
- 10.4.2 Silicon Nitride Waveguides
- 10.4.3 Coupling Techniques
- 10.5 Packaging and Integration Advancements
- 10.5.1 Chip-Scale Packaging
- 10.5.2 Wafer-Scale Integration
- 10.5.3 3D Integration and Interposer Technologies
11 CHALLENGES AND FUTURE TRENDS
- 11.1 CMOS-Foundry-Compatible Devices and Integration
- 11.1.1 Scaling and Miniaturization
- 11.1.2 Process Complexity and Yield Improvement
- 11.2 Power Consumption and Thermal Management
- 11.2.1 Energy-Efficient Photonic Devices
- 11.2.2 Thermal Optimization Techniques
- 11.3 Packaging and Testing
- 11.3.1 Advanced Packaging Solutions
- 11.3.2 Automated Testing and Characterization
- 11.4 Scalability and Cost-Effectiveness
- 11.4.1 Wafer-Scale Integration
- 11.4.2 Outsourced Semiconductor Assembly and Test (OSAT)
- 11.5 Emerging Materials and Hybrid Integration
- 11.5.1 Novel Semiconductor Materials
- 11.5.2 Heterogeneous Integration Approaches
- 11.6 Technology Readiness Assessment
12 COMPANY PROFILES (192 company profiles)
13 APPENDICES
- 13.1 Glossary of Terms
- 13.2 List of Abbreviations
- 13.3 Research Methodology
14 REFERENCES