With the global RF devices industry valued at $51.3 billion in 2024 and projected to grow to $69.7 billion by 2030, demand for advanced wireless technologies is accelerating across consumer electronics, telecom infrastructure, and emerging applications, as recently stated in Yole Group’s Status of RF Industry 2025 report. Within this landscape, the rollout of 5G and the first whispers of 6G are driving the need for highly integrated RF front-end (RFFE) solutions in base stations and mobile devices and the shift to wide-bandgap (WBG) semiconductors, including gallium nitride (GaN) and silicon carbide (SiC). These front ends combine power amplifiers (PAs), filters, and switches to support both sub-6 GHz and millimeter-wave (mmWave) operations, addressing the coexistence of multiple frequency bands with diverse RF requirements.

The rising role of GaN-on-Si in 5G/6G

The GaN RF device market, valued at $1.2 billion in 2024 and projected to reach $2.0 billion by 2030 with a compound annual growth rate (CAGR) of 8.4%, reflects the growing adoption of GaN across wireless infrastructure and devices, as shown in Figure 1. In telecom networks, 5G massive MIMO antennas are increasingly turning to GaN technology, gradually displacing LDMOS in PAs. At the heart of this shift, WBG semiconductors, particularly GaN-on-Si and GaN-on-SiC, are emerging as key enablers. By balancing performance, cost, and scalability, GaN promises to redefine efficiency, integration, and frequency reach while challenging incumbent technologies across the RFFE landscape.

Figure 1: The GaN RF device market is projected to reach $2 billion by 2030, up from $1.2 billion in 2024. (Source: Yole Group)

WBG materials such as GaN bring several electrical advantages over traditional, silicon-based devices. GaN offers higher breakdown voltages and greater electron mobility in HEMT devices thanks to the 2D electron gas present in GaN/AlGaN heterostructures. These properties enable higher operating voltages, greater power densities, and higher operating frequencies, well beyond the capabilities of legacy LDMOS and gallium arsenide (GaAs) technologies.

As 5G networks expand worldwide, operators face mounting challenges: exponential data growth, rising power consumption, and the need for smaller, more compact base stations with higher power density. GaN technologies can directly address these demands by delivering higher operating voltages, greater power density, and superior frequency performance, positioning GaN as a cornerstone for both 5G deployments and the transition to future network generations.

Looking ahead to 6G, the technology landscape becomes increasingly complex. Antenna architectures, including MIMO, are expected to scale to 256T/256R, 512T/512R, or even 1024T/1024R as they extend into FR3 frequency bands, shifting the balance between performance, cost, and integration. This opens a new opportunity window for GaN-on-Si to compete alongside established technologies, particularly in scenarios where large numbers of PAs operate at lower individual output power.

In contrast, GaN-on-SiC, traditionally strong in high-power applications, faces integration and cost challenges at these levels, while silicon germanium (SiGe) remains a strong contender in higher-frequency domains where integration efficiency is critical.

From a frequency perspective, GaN shows relevance both in sub-6 GHz (FR1), where it competes with LDMOS, and in FR3 (7–20 GHz), where integration density and low parasitics are vital and which are central to ongoing 6G research. As 6G concepts take shape, emphasizing terabit-per-second speeds, sub-millisecond latencies, and AI-driven network optimization, GaN-on-Si’s combination of high-frequency performance, scalability, and improving economics positions it as a competitor for the next leap in wireless infrastructure.

Integration, size, and the economics of scaling

In wireless infrastructure, size isn’t just a design consideration; it’s a cost driver. Smaller, lighter base stations are easier to install, cheaper to maintain, and more flexible in placement, especially in dense urban environments.

GaN-on-Si supports these goals by enabling high-power devices on cost-effective silicon substrates. Using standard 6-inch and increasingly 8-inch silicon wafers reduces material costs and enables manufacturers to leverage existing CMOS-compatible processes, scaling production without the high capital expenses associated with GaN-on-SiC fabrication.

Integration is also advancing beyond the die level. Packaging innovations, such as copper-pressed CuMo copper, ceramic air cavity, and overmolded (plastic) designs, are paired with GaN-on-Si to minimize parasitics, improve thermal performance, and shrink footprints. This is especially important in massive MIMO antenna arrays, where space is limited and heat dissipation is challenging.

Economically, GaN-on-Si still competes with mature LDMOS in lower-power applications in which the cost-per-watt advantage is narrower. However, adoption is rising. Market forecasts suggest GaN-on-Si’s share in base station PAs could grow from low single digits today to over 10% by 2029 (Figure 2), with a CAGR of ~45% from 2025 to 2030, outpacing GaN-on-SiC’s 6% growth, driven by cost reductions from larger wafer processing, yield improvements, and broader OEM acceptance.

Figure 2: GaN technologies are gradually replacing LDMOS in telecom infrastructure applications. (Source: Yole Group)

Beyond telecom: GaN-on-Si in satellite and mobile

While GaN-on-Si is often discussed in the context of 5G and emerging 6G infrastructure, its potential extends into satellite communications and mobile devices. In satellite communications, particularly in satellite-to-cell services, today’s deployments rely largely on GaAs-based PAs to enable messaging and voice with limited data rates.

To support higher throughput and efficiency, particularly as operators like AST SpaceMobile and Lynk push toward direct-to-cell broadband, GaN-on-Si could offer advantages in bandwidth and high-frequency performance. However, supply chain maturity, cost competitiveness, and the absence of commercial, GaN-based solutions at the handset level remain barriers, leaving GaAs as the dominant short-term choice.

In parallel, mobile handsets represent another frontier. GaN-on-Si could, in theory, bring benefits for sub-7 GHz and FR3 (7–20 GHz) bands, enhancing both satellite communication capabilities and future high-bandwidth mobile applications. However, challenges remain significant: The technology must achieve scalability to 8- and 12-inch wafers, resolve reliability issues, and adapt to new system-on-chip integration demands. GaAs, RF-SOI, and SiGe continue to provide established alternatives, keeping GaN-on-Si’s entry into mobile markets on a longer horizon, likely closer to the late 2020s and early 2030s, aligned with 6G development.

Manufacturing momentum and deployment pathways

The GaN-on-Si ecosystem is accelerating as major RF players leverage existing silicon fabs to fast-track adoption, with most strategies moving directly to 8-inch platforms to support cost reduction. Infineon Technologies AG entered the telecom market in 2023 with GaN-on-Si PA modules on 8-inch wafers, aiming to challenge GaN-on-SiC in upcoming base station designs. Macom, bolstered by acquisitions such as Ommic and Wolfspeed’s RF GaN business, combines long-standing 6-inch capabilities with inherited 4-inch lines.

GlobalFoundries is advancing GaN-on-Si production on 8-inch platforms through its partnership with Finwave Semiconductor Inc. and the recent acquisition of Tagore. United Microelectronics Inc., which operates one of the largest 8-inch fab facilities, has also commercialized GaN-on-Si. Win Semiconductors Corp. has a 6-inch line. Meanwhile, Sony Semiconductor Solutions does not yet have commercial GaN-on-Si offerings for handsets but is reportedly exploring opportunities in the telecom space.

At the cutting edge, Intel Corp. is developing GaN-on-Si on 12-inch wafers, targeting cost-competitive mmWave solutions for future 5G and early 6G applications. This shift to larger wafers reduces cost-per-device and scales capacity without the need for greenfield fab investments. These manufacturing advances align closely with network evolution: As antenna systems expand from 32T/32R and 64T/64R to 128T/128R and beyond, eventually reaching hundreds of streams, the number of PAs per site will increase as the individual PA output power declines.

Figure 3: Chipmakers and foundries are set to fast-track the adoption of GaN-on-Si to leverage its benefits in integration, size, and scaling for wireless infrastructure applications. (Source: Yole Group)

GaN-on-Si’s strategic place in the wireless future

As the telecom industry moves from late-stage 5G deployment to 6G, the need for efficient, scalable, high-frequency-capable devices is greater than ever. GaN-on-Si’s unique balance of performance, integration potential, and economic scalability positions it as a strong contender for the next decade of wireless innovation.

Its future success will hinge on continued cost reductions, sustained improvements in reliability, and robust supply chains to support high-volume production. If these conditions are met, GaN-on-Si could become a foundational semiconductor technology for the expansion of 5G and emergence of 6G, delivering faster, more reliable connectivity while keeping energy consumption in check.

About the author

Ahmad Abbas is a technology and market analyst for compound semiconductors at Yole Group. His core expertise is GaN RF technologies and related markets, including telecommunications and defense applications.

Prior to joining Yole Group, Abbas spent four years developing and optimizing power semiconductor devices. Following his internship at CEA-Leti, where he focused on developing GaN HEMTs for consumer applications, Abbas pursued his Ph.D., studying and developing SiC power components for EV applications, including testing SiC devices on engineered substrates provided by industry partners. At Yole Group, he focuses on analyzing GaN RF technologies and the evolution of the market and supply chain.

Abbas holds a master’s degree in nanosciences and nanotechnologies from Université Grenoble Alpes (France), earned in 2021, and completed his Ph.D. in 2025. He has authored and co-authored several scientific publications and presented at international conferences in the field of power semiconductors.

The post GaN-on-Si in the 5G era: Paving the road to 6G appeared first on Electronic Products.