ASMPT and IBM Join Forces to Conquer Chiplet Bonding Challenges

By streamlining the packaging process, this partnership aims to expedite the transition of chiplet technology from research to mass production.

Introduction

In a significant move to push the boundaries of semiconductor technology, ASMPT and IBM have renewed their collaboration to advance bonding methods for AI Chiplet Bonding Technology.

This partnership aims to enhance thermocompression and hybrid bonding techniques, utilizing ASMPT’s state-of-the-art Firebird TCB and Lithobold hybrid bonding tools.

The renewed agreement underscores the importance of innovative packaging technologies in meeting the rapidly growing demands of AI applications.

Advanced Bonding Techniques: The collaboration will leverage ASMPT’s cutting-edge Firebird TCB and Lithobolt tools to develop superior bonding methods.

Accelerated Chiplet Production: By streamlining the packaging process, this partnership aims to expedite the transition of chiplet technology from research to mass production.

Building on Success: The current agreement expands upon a previous collaboration that resulted in a groundbreaking hybrid bonding approach, significantly enhancing bonding quality between chiplets.

The ultimate goal is to create chiplet packages that are smaller, faster, more power-efficient, and more reliable than traditional SoCs. This will have a profound impact on various industries, including AI, high-performance computing, and automotive.

Follow us on Twitter: https://x.com/TechoVedas

What are Chiplets?

Before delving deeper into the partnership, it’s essential to grasp the concept of chiplets. Traditionally, a System on Chip (SoC) integrated all components onto a single die. Chiplets break down this monolithic structure into smaller, specialized chips that can be combined to form a complete system. This approach offers several advantages:

  • Increased design flexibility: Allows for mixing and matching different chiplets to create customized systems.
  • Faster time-to-market: Enables parallel development of individual chiplets.
  • Reduced risk: If one chiplet fails, the entire system isn’t compromised.
  • Potential cost reduction: By leveraging existing chip designs and manufacturing processes.

The Role of Bonding

To realize the potential of chiplets, advanced bonding techniques are crucial. This is where ASMPT and IBM’s collaboration shines.

  • Thermocompression Bonding: This method involves applying pressure and heat to fuse two materials together without the use of an adhesive. It’s often used for connecting copper interconnects between chiplets.
  • Hybrid Bonding: This technique combines different bonding methods, such as thermocompression and adhesive bonding, to achieve specific properties. The partnership’s breakthrough involved using a combination of copper and oxide layers to create incredibly thin and strong bonds.

Advancing Chiplet Packaging Technologies

Chiplets represent a significant shift in semiconductor design, deconstructing system-on-chips (SOCs) into smaller, individual components that can be packaged together to function as a single system.

This approach offers numerous benefits, including improved energy efficiency, faster system development cycles, and reduced costs.

However, the transition from research to mass production of chiplets requires advanced packaging techniques to keep pace with the rapid innovations in AI computing.

67% superior Performance: Micron Unveils World’s Fastest Data Center SSD – techovedas

Building on Previous Success

The latest agreement between ASMPT and IBM builds on an existing collaboration that saw the debut of a new hybrid bonding approach last year.

This method optimized bonding quality between chiplets, setting the stage for further advancements.

Now, the two companies will continue to develop cutting-edge bonding technologies for chiplet packages, ensuring they meet the demands of modern AI applications.

Why Amazon is Developing its own AI Chips like Nvidia? – techovedas

Statements from Industry Leaders

Huiming Bu, Vice President of IBM Semiconductors Global R&D and Albany Operations, IBM Research, stated:

IBM has been at the forefront of developing advanced packaging technology for the age of AI. We are proud to continue our work with ASMPT to advance chiplet packaging technology to pave the way for smaller, more powerful, and more energy-efficient chips.”

Lim Choon Khoon, Senior Vice President of ASMPT, added:

We are excited to build on our strong relationship with IBM to drive the frontiers of Advanced Packaging in tandem with accelerating innovations in artificial intelligence. “We are excited to collaborate with IBM to push the boundaries of next-generation packaging and Heterogeneous Integration solutions for the AI era.”

Importance of Advanced Packaging in AI

The collaboration between ASMPT and IBM highlights the critical role of advanced packaging in the semiconductor industry, particularly for AI applications.

As AI systems demand more power and efficiency, innovations in packaging technology become essential.

Advanced packaging methods like thermocompression and hybrid bonding enable the integration of multiple chiplets into a single package, enhancing performance and reducing energy consumption.

Meta’s LLama 3 Training Hampered by Faulty Nvidia H100 GPUs: A Glitch Every Three Hours – techovedas

Technical Details: Thermocompression and Hybrid Bonding

Thermocompression Bonding (TCB): This method uses heat and pressure to bond chiplets together, creating a reliable and durable connection.

“ASMPT’s Firebird TCB tool optimizes the bonding process, delivering high-quality bonds that meet the stringent requirements of AI applications.”

Hybrid Bonding: This technique combines elements of TCB with other bonding methods to enhance the quality and reliability of the connections between chiplets.

The Lithobolt hybrid bonding tool from ASMPT offers advanced capabilities, making it possible to achieve superior bonding quality.

The Role of Chiplets in AI

Chiplets are revolutionizing the semiconductor industry by breaking down complex SOCs into manageable components.

This modular approach allows for greater flexibility in design and manufacturing, enabling faster development cycles and cost savings.

For AI applications, chiplets offer the potential for significant improvements in performance and energy efficiency, making them a critical component of future AI systems.

Potential Applications

The advancements in chiplet bonding technology enabled by this partnership could lead to:

  • More powerful AI systems: By allowing for the integration of a larger number of specialized AI accelerators.
  • Energy-efficient data centers: Through the development of optimized chiplet-based processors.
  • Advanced autonomous vehicles: By enabling the integration of complex sensor and computing systems.

Intel Introduces Free AI Playground App for Arc GPUs – techovedas

IBM’s Commitment to AI and Advanced Packaging

IBM has long been a leader in semiconductor technology, with a strong focus on AI and advanced packaging.

The company’s expertise in these areas is instrumental in driving innovations that meet the demands of modern AI systems.

By collaborating with ASMPT, IBM is leveraging its strengths to push the boundaries of what is possible in chiplet packaging.

IBM is a global leader in hybrid cloud and AI, providing consulting expertise to clients in over 175 countries. The company’s innovations in AI, quantum computing, and industry-specific cloud solutions help organizations transform their business processes efficiently and securely. IBM is committed to trust, transparency, responsibility, inclusivity, and service.

ASMPT’s Role in Semiconductor Manufacturing

ASMPT is a global leader in semiconductor assembly and packaging solutions.

The company’s advanced tools and technologies are used to manufacture a wide range of electronic components, from mobile devices to automotive systems.

With a strong commitment to research and development, ASMPT is at the forefront of innovations that shape the semiconductor industry.

ASMPT is a leading supplier of hardware and software solutions for semiconductor and electronics manufacturing. Headquartered in Singapore, ASMPT’s offerings span semiconductor assembly and packaging, as well as surface mount technology. The company’s continuous investment in R&D ensures it provides industry-shaping solutions that achieve higher productivity, reliability, and quality. ASMPT is listed on the Stock Exchange of Hong Kong.

Industry Impact and Future Prospects

The collaboration between ASMPT and IBM is set to have a significant impact on the semiconductor industry.

By advancing bonding methods for chiplet packages, the partnership will help accelerate the transition from research to mass production.

This, in turn, will enable the development of smaller, more powerful, and more energy-efficient AI systems.

Meta’s LLama 3 Training Hampered by Faulty Nvidia H100 GPUs: A Glitch Every Three Hours – techovedas

Conclusion

The renewed collaboration between ASMPT and IBM marks an important milestone in the advancement of AI Chiplet Bonding Technology.

By leveraging their combined expertise, the two companies are poised to drive significant innovations in thermocompression and hybrid bonding methods, paving the way for the next generation of AI systems.

As the semiconductor industry continues to evolve, partnerships like this will be essential in meeting the growing demands of modern technology.

Kumar Priyadarshi
Kumar Priyadarshi

Kumar Priyadarshi is a prominent figure in the world of technology and semiconductors. With a deep passion for innovation and a keen understanding of the intricacies of the semiconductor industry, Kumar has established himself as a thought leader and expert in the field. He is the founder of Techovedas, India’s first semiconductor and AI tech media company, where he shares insights, analysis, and trends related to the semiconductor and AI industries.

Kumar Joined IISER Pune after qualifying IIT-JEE in 2012. In his 5th year, he travelled to Singapore for his master’s thesis which yielded a Research Paper in ACS Nano. Kumar Joined Global Foundries as a process Engineer in Singapore working at 40 nm Process node. He couldn’t find joy working in the fab and moved to India. Working as a scientist at IIT Bombay as Senior Scientist, Kumar Led the team which built India’s 1st Memory Chip with Semiconductor Lab (SCL)

Articles: 2237