Huatai Securities Co., Ltd. has released an analysis focusing on the development opportunities in core upstream materials for optical modules. The report highlights that with the rapid increase in demand for 800G and 1.6T optical modules in recent years and the approaching era of 3.2T, significant growth potential is seen for key upstream materials. The analysis systematically examines the growth logic of two major industries: indium phosphide (InP) substrates and thin-film lithium niobate (TFLN). InP substrates, as a core raw material upstream of optical chips, are benefiting from strong demand pull from optical chip manufacturers, leading to a supply shortage trend. TFLN modulators, with advantages such as low power consumption and high bandwidth, are expected to see an adoption window in future 3.2T pluggable solutions, indicating broad growth potential for the industry chain.
The demand for optical modules is experiencing high growth driven by AI computing power, leading to a rapid release of demand for optical chips. As global tech giants continue to increase investments in AI computing, the optical module industry chain is expected to maintain high growth. According to a forecast by LightCounting in January 2026, the global market size for high-speed data center optical modules is projected to expand from $16.4 billion in 2025 to $52.1 billion by 2031. Optical chips, as a core upstream material, are also seeing strong demand growth. Data from a Hong Kong listing document shows the global optical chip market is expected to grow from $2.6 billion in 2024 to $22.9 billion by 2030, representing a compound annual growth rate of 44%.
InP substrates are a fundamental material for optical chip production. With demand for optical chips driven by 800G and 1.6T modules expected to accelerate around 2026-2027, leading global optical chip manufacturers are actively expanding capacity. Yole predicts global InP substrate sales will increase from 500,000 units in 2019 to 1.28 million units by 2026. The market is highly concentrated, with Sumitomo, Beijing Tongmei, and JX holding over 90% market share. New entrants may find opportunities amid this high growth. The InP substrate supply chain involves high-purity raw materials, polycrystalline synthesis, single crystal growth, substrate preparation, and epitaxial wafers/chips.
As the 3.2T era approaches, requiring single-channel modulation rates of 400G, TFLN modulators are anticipated to gain adoption due to advantages like ultra-high bandwidth and low power consumption compared to pure silicon modulators. Estimates suggest the TFLN modulator market space driven solely by 3.2T modules could reach nearly $382 million by 2031, with a CAGR of 271% from 2029 to 2031. The TFLN industry chain includes lithium niobate crystal materials, TFLN wafers, and TFLN modulators, with high technical barriers at each stage. Chinese manufacturers have made significant progress across these segments and are well-positioned for growth in the 3.2T era.
Key companies involved in the core segments of the industry chain include those focused on high-purity indium, polycrystalline synthesis, and InP substrates for the InP segment. For the TFLN segment, companies are engaged in lithium niobate crystal materials, TFLN wafers, and TFLN modulators. Specific company details can be found in the full report.
Risks include potential underperformance in capital expenditure by cloud providers, intensified industry competition, and the fact that mentions of unlisted or uncovered companies are based on public information and do not constitute recommendations.
Comments