Perstorp and Intel collaborate on advanced liquid cooling solutions

Perstorp Holding AB, a global leader in specialty chemicals and a subsidiary of PETRONAS Chemicals Group Berhad (PCG), has announced a collaboration with Intel’s Open IP Advanced Liquid Cooling team. Together, they have developed a high-performance synthetic thermal management fluid specifically designed for immersion cooling in data centres.

Since initiating their partnership in 2022, Perstorp and Intel have been working on creating sustainable and efficient cooling solutions to support the rising demand for AI applications and high-density computing. Their efforts were recently showcased at COMPUTEX 2024 in Taipei, Taiwan, where Intel presented its Gaudi 3 AI accelerator solution. This solution utilises Intel’s SuperFluid Cooling Technology, which leverages the advanced properties of Perstorp’s thermal management fluid.

Valentina Serra-Holm, vice president for Engineered Fluids at Perstorp, highlighted the importance of this collaboration in addressing the challenges of modern data centres. “The recent surge in the AI race has brought a lot of attention to liquid cooling. Our innovative capabilities in engineered fluids have been widely recognised. In collaboration with Intel, we have developed an advanced synthetic thermal management fluid suitable for their SuperFluid technology, achieving the critical goals of high thermal performance and low viscosity,” Serra-Holm stated.

Sustainability and performance

The new thermal management fluid is not only efficient but also environmentally friendly. It boasts a low global warming potential (GWP) and zero ozone depletion potential. Moreover, it is PFAS-free, biodegradable, and has an improved toxicity profile compared to existing solutions. The fluid is designed with future sourcing options in mind, allowing for a potential shift to 100% renewable or recycled raw materials, aligning with Perstorp’s long-term sustainability goals.

Intel’s SuperFluid technology, enhanced by Perstorp’s fluid, can significantly increase cooling capacity. Traditional single-phase immersion cooling systems typically manage around 500 watts per chip, but with this advanced solution, cooling capacity can reach up to 1,500 watts per chip, making it ideal for next-generation high-end AI chips.