The SiC market has been very active lately, with constant news coming from device suppliers and car makers. And there seems to be an ongoing tug-of-war between supply and demand.

Toshiba announced in April the groundbreaking of its power semiconductor fab for SiC in Ishikawa Prefecture, with the first stage beginning in the 2024 fiscal year. This news echoes earlier reports from Japanese media that Toshiba is strengthening the vertical integration throughout SiC equipment, wafers, and devices, and planning to increase the production by three times in 2024 and 10 times by 2026.

Meanwhile, over the past two years, leading companies in the Europe and the US such as Infineon and ST have also accelerated M&A as well as internal expansion for SiC production devices at an unprecedented pace, aiming to expand their SiC-related businesses and maintain their core competitiveness in the market.

Despite aggressive demand-driven expansion plans, the unexpected announcement from Tesla in mid-March that it plans to reduce overall SiC usage by 75% in the next generation of electric vehicle platforms has sparked various speculations in the industry. This move was made without compromising the performance and efficiency of the cars and represents one of the few specific details that Tesla has revealed about its new car plans.

Now here is the question – will the popularity of SiC be a genuine trend, or merely a passing fad that could lead to a potential bubble in the market?

SiC or Si-based solutions?

Compared to IGBT and MOSFET, the dominant technologies in power semiconductor, SiC offers stronger advantages such as low resistance, high temperature and high voltage tolerance that can overcome the technical bottlenecks of EVs by improving battery efficiency and solving component heat dissipation issues. SiC can also make chip design sizes smaller, which means more flexibility in vehicle design.

These advantages have made SiC the most sought-after technology. According to TrendForce, the SiC power device market is expected to grow at a CAGR of 35% to reach $5.33 billion annually from 2022 to 2026, driven by mainstream applications such as electric vehicles and renewable energy.

There is a long-standing debate among the industry about whether SiC will replace IGBTs entirely. What we believe is that SiC may not completely replace IGBTs considering their distinct targeted use scenarios.

In terms of use cases, SiC is particularly suitable for high-frequency, high-voltage applications, especially in the field of new energy vehicles. Traditional Si-based IGBT chips have reached the physical limit in high-voltage fast charging models, making SiC more favorable for new energy vehicles.

However, SiC transistors are expensive due to complex production processes, slow crystal growth, and difficult cutting. Unlike silicon, which can be pulled quickly, SiC crystals grow at a slow rate of 0.2-1mm/hour and are prone to cracking during the cutting process due to their high hardness and brittleness, leading to hundreds of hours of cutting time.

Additionally, SiC transistors also have some drawbacks such as vulnerability to damage and temperature sensitivity, which makes them unsuitable for low-cost and low-power applications.

IGBT, on the contrary, is preferred over SiC in such a field because it is more cost-effective, reliable, and has better capacitance and surge capability for high-power and high-current applications. In certain scenarios, such as DC-DC charging piles, IGBT is irreplaceable due to its cost advantage and suitability.

Could a Hybrid Solution be the Answer?

The premise above can help to explain Tesla’s conflicting decision to cut back on SiC usage.

Tesla’s reluctance to fully adopt SiC technology is mainly due to concerns about reliability and supply chain stability, as evidenced by a mass recall of Model 3 due to issues with SiC components in the rear electric motor inverter.

In addition, the shortage of substrate materials is another challenge facing the SiC industry as a whole, with major manufacturers such as Wolfspeed, Infineon, and ST ramping up production capacity to address the issue. As a result, Tesla is considering alternative ways to mitigate the risks associated with supply chain constraints.

Despite these challenges, SiC remains a promising trend for the EV industry. Even Tesla recognizes its enormous potential commercial value.

In terms of technological innovations, Tesla’s next-generation EVs may feature a novel packaging design for the primary inverter, utilizing a hybrid SiC/Si IGBT packaging approach that leverages the unique strengths of both technologies while avoiding potential pitfalls. This technological advancement poses certain difficulties, but the groundbreaking innovation at the engineering design level is definitely something to look forward to.

Read more:

• Chinese Players Rally for Demand Surge in MOSFET, IGBT, and SiC
• The M&A battle for SiC: Who’s the Top Acquirers?

(Photo credit: Tesla)

Amid a prolonged market downturn and persistent weakness in end demand, the world’s top three memory chipmakers – Samsung, SK Hynix, and Micron – have implemented production cuts in an effort to control the continuing decline in memory prices through supply management. Recently, news emerged in the memory channel market that Micron had notified its customers that starting in May, it will not accept inquiries for DRAM and NAND Flash below current market prices.

According to TrendForce, the situation is not widespread at the moment, but is limited to low-priced memory chips. As for other product categories with high inventory levels, they still cannot avoid the situation of falling prices.

Contract market:

Although DRAM suppliers have actively reduced production, the output bit volume has not yet reached an effective convergence in 2Q23, so the quarterly contract price decline will be greater than originally expected, with an expected drop of more than 15%. TrendForce has observed that there is a strong wait-and-see atmosphere on the OEM side. While the willingness to purchase DRAM has increased, the premise of the deal is that low-priced quotes are attractive enough to OEMs. Due to poor demand prospects, the purchasing behavior of buyers still appears to be passive.

Spot market:

TrendForce pointed out that Micron’s subsidiary brand, Spectek, has slightly raised prices for its products this week, especially in the low-priced chip segment, indicating a reluctance to further reduce prices. Therefore, trading in the spot market appears stagnant, similar to the strong wait-and-see attitude mentioned in the contract market.

As suppliers have already entered a stage of significant losses, it is necessary to continue to expand production cuts to avoid prices from collapsing again. Among them, DDR4 still has a price decline due to high inventory levels and weak demand, while the supply of DDR5 is limited by the PMIC compatibility issue, resulting in an upward trend in spot prices.

In the post-Moore’s Law era, chiplet design has been burgeoning as the mainstream architecture.

With the widespread adoption of EUV technology by foundries on process nodes of 5nm and below, the cost of semiconductor fabrication has skyrocketed. The cost of the 5nm process has grown by almost 1x compared to the 7nm process, and the 3nm process is expected to increase by almost 1x compared to the 5nm process.

To address this issue, IC design companies have started to split chip components or connect multiple chips and adopt advanced packaging such as 2.5D/3D IC to integrate multiple chips together.

Compared to traditional chip design methods, chiplet design has superior characteristics such as shorter upgrade cycles, lower costs, and higher yields, which is one of the reasons why chiplet technology is gaining popularity.

AMD’s chiplet design is a representative example. Through close collaboration with TSMC, AMD has fully transitioned its CPUs to chiplets since the 7nm process, with the Ryzen 7000 series CPU and Radeon RX 7000 series graphics cards released in 2022. The latter uses the RDNA 3 architecture and integrates the GCD and MCD produced by the 5nm and 6nm processes respectively, as a result improving overall performance, with a 54% increase in RDNA 3’s Performance per Watt.

Under the leadership of industry leaders such as AMD and Intel, chiplet design has had a significant impact on the entire semiconductor industry – substrates manufacturers in particular.

ABF Substrates Set to Soar

Aside from CPUs, developments in AMD and Intel’s server platforms indicate that the trend towards higher-layer-count and larger-area ABF substrates is expected to continue.

Given the server shipment volume is expected to remain stable and grow steadily in the mid to low single digits for the next 3-5 years, the growth momentum of ABF substrates mainly comes from the increase in layer count and area brought by 2.5D/3D packaging adoption in servers.

Starting in 2020, ABF substrates saw a surge in demand due to the pandemic. The supply-demand gap peaked in 2021, and in the first half of 2022, ABF substrate prices increased while volume increased and gross profit margins hit new highs.

Due to the impact of shortage in ABF substrates in 2020-2021, major substrate manufacturers have initiated large-scale expansion plans, with the expectation that demand for ABF substrates would continue to grow with the upcoming releases of new server platforms and the integration of 2.5D packaging for PC CPUs.

Growing demands with Some Hiccups

However, the moves have been put on hold for now. Since the second half of 2022, due to inventory correction in the overall semiconductor industry and the delayed production time of Intel’s new server platform, there’s been a supply glut in ABF substrates.

Therefore, Unimicron has taken the lead in adjusting its capital expenditure plans, reducing its planned capacity increase for 2023 from about 20% to only 3.5%. AT&S has also tentatively postponed the significant increase in capacity planned for the end of 2024. It is unclear when the expansion will resume or whether the expansion will be scaled back.

This indicates that current substrate manufacturers have not only lowered their demand projections for 2023, but also for 2025-2026. Further adjustments to the expansion plans of other manufacturers will also affect the future market supply-demands dynamics.

Back on Track for Major Growth in 2024

Looking into the future, things are looking up for the ABF substrate industry. In the second quarter of 2023, we can expect the release of new server platforms from AMD and Intel, as well as the completion of PC inventory adjustments.

With expansion plans in place, it’s predicted that global ABF substrate production capacity will only increase by 15-20% in the latter half of 2023, continuing to put pressure on substrate manufacturers, according to TrendForce.

Things are expected to pick up in 2024 with the release of AMD and Intel’s next-generation server platforms, Zen 5 and Birch Stream. Plus, the anticipated introduction of 2.5D packaging for PC CPUs will drive a new wave of demand for ABF substrates. All in all, we can expect a significant rebound for the ABF substrate industry in 2024.

From the Entity List in 2020 to the Chips and Science Act of 2022, the US government has been tightening its grip on China’s semiconductor industry by blocking the export of advanced semiconductor manufacturing equipment. The pressing question on everyone’s mind is: Will China’s semiconductor industry crumble under this pressure?

The answer, based on recent market reactions, is a resounding no.

Riding the Waves through Headwinds

Despite international semiconductor equipment manufacturers facing production cutbacks, China’s semiconductor equipment industry is thriving. In the first quarter, Naura, the leading semiconductor equipment manufacturer, reported a whopping 68.56%-87.29% increase in revenue, with a 171.24% to 200.3% increase in net profit. This has spurred growth across the entire Chinese A-share market for semiconductor equipment concept stocks such as Piotech, PNC process System, Advanced Micro, ACM Research and Hwatsing Technology.

This growth highlights a great leap forward in semiconductor process technology. Despite the adverse effects of the US’s broad-based restrictions, they have nonetheless created a favorable environment for testing and substitution opportunities. This, in turn, has enabled Chinese manufacturers of semiconductor equipment to increase their market share in the area of established semiconductor processes.

Full Speed Ahead: Aiming High for 5nm

In key semiconductor manufacturing processes such as thin film deposition, etching, ion implantation, CMP, and cleaning, Chinese manufacturers have already moved beyond traditional equipment development cycles and are progressing towards advanced process technology at full speed.

According to TrendForce, Chinese semiconductor equipment companies such as Naura and Advanced Micro(AMEC) are capable of supporting 28/14 nm in some process steps, and have even tentatively established their presence in 5 nm process technology.

Our summary identifies the main players to watch in thin film deposition, etching, and EUV:

• Thin film deposition: Naura

Naura has achieved full coverage of PVD, CVD, and ALD product lines, with product lines matching international leaders such as Applied Materials, Lam, and Tokyo Electron. Naura has unique competitive advantages in the PVD field, with over 20% of its PVD equipment being supplied to Chinese 12-inch production lines such as YMTC(Yangtze Memory Technologies Co., Ltd), making it the second-largest PVD equipment supplier after Applied Materials.

Additionally, since 2012, Naura has sold over 200 PVD equipment, gradually achieving their goals for equipment substitution.

• Etching: AMEC and Naura

As the leading CCP etching machine, AMEC has successfully penetrated TSMC’s 5nm production line, becoming the first domestic etching equipment to break through in the advanced process area. AMEC has also achieved large-scale adoption in 64-layer, 128-layer 3D NAND process, and 1x DRAM process. These main product portfolios contributed to the company’s 47.3% YoY revenue growth rate in the first half of 2022. In addition, AMEC’s etching equipment also enjoys a high gross profit margin of 46.02%.

On the other hand, Naura is at the forefront of ICP silicon etching equipment. Its first-generation 12-inch etching equipment underwent certification for 90-65nm at the SMIC’s fab in Beijing in 2008. In addition, with the support of national research projects, Naura’s ICP etching machine has also broken through 14nm barriers and been adopted by mainstream foundries.

• Photolithography: Shanghai MicroElectronics Equipment(SMEE)

Photolithography is a critical process that China is strategically including in their semiconductor industry plans. They’re aiming to develop 28nm immersion exposure machines and core components through collaborative efforts: SMEE will lead the overall design and integration, with five or more companies providing key components.

Although SMEE has preliminary DUV exposure machine technology, it’s limited to more mature processes on 8-inch and 12-inch wafers at 90nm, 110nm, and 280nm, leaving a significant gap with international leaders.

From Toddler to Major Player

Although China’s equipment manufacturers are still at their toddler stage, the increasing momentum suggests that they will continue to make significant progress. Assuming that China’s policy support towards the development of 14nm and below semiconductor processes remains unchanged in the coming years, it is highly likely that the country’s market will fundamentally experience a transformation.

At this point, China’s semiconductor industry will enter a new era of high-speed growth, paving the way for the country to become a major player at global level. As China’s domestic market grasps the technology and commercial logic along the way, it will potentially have more influence over the global supply chain, as a result triggering a shift in the worldwide semiconductor industry in the long run.

The compound semiconductor market has been flourishing in recent years thanks to the strong demand from markets such as electric vehicles and renewable energy. This has led to an increase in M&A as companies race to establish their position in the industry.

The market has seen a significant surge in M&A deals over the last few years: from 2006 to 2017, there was only one deal every two years, but since 2018, there have been six deals annually, surpassing historical data.

While SiC and GaN are the top categories for M&A, 21 of the transactions are directly related to SiC. This is because after its development over 20 years, SiC has been able to be mass-produced for market demands, particularly in the automotive industry where SiC has become the mainstream technology.

Vertical Integration driven by Industry Titans

Industry leaders in the US and Europe, such as Wolfspeed, On Semi, II-VI, ST, and Infineon, have started accelerating vertical integration in recent years, as reflected in the frequency of M&A.

The United States has led 12 M&A deals, with only four of them occurring before 2018, and Wolfspeed contributed to three of them. Over the past three years, On-Semi, II-VI, and Macom have led several deals with a focus on SiC’s vertical integration to meet market demands.

In Europe, there were eight M&A deals in total, all of which took place in 2018 and beyond, with ST and Infineon being the major players. Both companies have been accumulating technical strength through strategic acquisition to maintain their leading ground in the SiC power device market.

In 2019 and 2020, ST acquired Norstel AB to bolster its SiC wafer manufacturing capabilities and Exgan to improve the GaN power device design expertise. Similarly, Infineon acquired Siltectra GMbH in 2018 to gain control of the crucial SiC wafer cold split process technology and recently acquired GaN Systems to reinforce its presence in the GaN market.

It’s evident from the cases that the high frequency of M&As in the US and Europe is mainly driven by leading companies in the industry, gradually defining the landscape of the market.

Wolfspeed, which has grown into a leading company after a long period of time, has accumulated enough capital for M&A and gradually been transforming into a platform-type company. Meanwhile, Onsemi, ST, and Infineon, which have traditionally been platform-type companies with established expertise in the field of compound semiconductors, are now ramping up their M&A activities to expand market presence and generate strong growth momentum.

Market Landscape Continues to Change

M&A deals among semiconductor equipment companies are also receiving attention. Recently, ASM and Veeco have successively acquired LPE and Epiluvac, indicating that equipment manufacturers have also realized the huge potential of the SiC market and are accelerating their investment.

Given the rapid technology breakthroughs, the overall SiC power device market is predicted to grow at an annual rate of 41.4% to reach $2.28 billion by 2023 and $5.33 billion by 2026 at 35% annual growth, according to TrendForce’s latest report.

However, with the current market boom comes a new challenge – the supply shortage. One of the biggest obstacles to industry growth is the scarcity of SiC substrate material, despite efforts from companies like STM and Onsemi to ramp up their production.

Manufacturers are now on the hunt for both internal and external sources to keep the supply flowing. While most of the SiC substrate suppliers are expanding, only a few, like Wolfspeed, are controlling the manufacturing capacity for high-end SiC substrates used in automotive main inverters, which worsens the bottleneck in SiC devices’ production for cars.

With that being said, major players must quickly address technology hurdles and supply issues to bridge the market gap. This will inevitably drive intense competition and industry consolidation, and only the ones that can adapt quickly will be thriving in the long run.

Read more:

• SiC vs. Silicon Debate: Will the Winner Take All?
• Chinese Players Rally for Demand Surge in MOSFET, IGBT, and SiC