Why conventional cooling is running out of road
Liquid metal cooling is gaining renewed urgency because power densities in electronics, electric vehicles, and concentrated solar power are pushing conventional water and dielectric cooling toward their physical limits. Gallium-based alloys such as Galinstan and EGaIn, along with alkali metals including liquid sodium and lithium, deliver thermal conductivities roughly 10–100 times higher than water—enabling significantly lower thermal resistance at equivalent channel geometries.
The technology operates across two distinct temperature regimes. At the high-temperature end, alkali metals such as sodium and lithium serve as primary heat transfer fluids in concentrated solar power (CSP) receivers and nuclear systems—an approach with validated receiver operation and more than 30 years of system handling knowledge documented by the German Aerospace Center (DLR). At the low-to-moderate temperature end, gallium-based alloys are applied in microchannel heat sinks and micro-circulation systems for electronics.
Liquid metals such as gallium alloys (Galinstan, EGaIn) and alkali metals (sodium, lithium) achieve thermal conductivities roughly 10–100 times higher than water, enabling significantly lower thermal resistance in microchannel heat sink geometries compared to water-cooled equivalents.
The broader competitive landscape includes cold-plate water cooling, dielectric immersion cooling, microchannel cooling, and vapor chamber/heat pipe hybrids. Each of these is represented in the patent dataset as the incumbent technology base that liquid metal approaches must displace or augment. According to WIPO, thermal management is among the fastest-growing technology fields in global patent filings, reflecting the structural pressure that rising device power densities are placing on cooling innovation worldwide.
This landscape is derived from a targeted set of patent and literature records spanning 2007–2026. It represents a snapshot of innovation signals within this dataset only and should not be interpreted as a comprehensive view of the full industry.
From foundational patents to 2026 filings: the innovation arc
The liquid metal cooling patent record follows a clear four-phase arc from baseline water-loop architectures to active infrastructure-scale deployments—with liquid metal research accelerating sharply from 2015 onward.
The foundational period (pre-2010) is defined by cold-plate and water-cooling patents that establish the performance benchmarks liquid metals must surpass. IBM’s microjet array cooler, demonstrating more than 2.5 MW/m² cooling capability with water (2007), set the performance bar for electronics thermal management. Early cold plate designs from Amulaire Thermal Technology (US, 2007) and Sharkoon Technologies (DE, 2004) represent the baseline water-loop paradigm.
The developmental period (2015–2020) saw liquid metal-specific research accelerate. DLR’s 2017 literature on liquid metals for solar power systems marked a revival of high-temperature liquid metal cooling. Tianjin Polytechnic University’s liquid metal microcirculation system study (2019) and the Global Energy Interconnection Research Institute comparative analysis of liquid metal versus water cooling in IGBT power electronic heat sinks (2020) represent key applied investigations. The Global Energy Interconnection work explicitly concluded that liquid metal outperforms water as a cooling medium in power electronic devices at elevated power densities.
The growth and diversification period (2020–2023) shows a surge in liquid cooling patent filings addressing electric vehicles, data centres, and high-power electronics. The Yangzhou liquid metal microchannel study (2022) introduced multi-parameter optimisation of channel cross-section and working fluid type, framing liquid metal as the preferred solution for extreme heat flux conditions—rocket nozzles, miniature nuclear reactors, and solar thermal generation—where water cooling is insufficient.
The most recent signals (2024–2026) come from Google LLC (EP, 2024–2025), Canaan Creative (EP, 2026), and Xiamen Hithium Energy Storage Technology (EP, 2025–2026). These reflect active infrastructure-scale liquid cooling deployments for AI computing and battery energy storage using modular and immersion architectures. These filings do not yet deploy liquid metal per se, but they establish the infrastructure context into which liquid metal cooling is being inserted.
The Global Energy Interconnection Research Institute (2020) concluded that liquid metal outperforms deionised water as a cooling medium in IGBT power electronic heat sinks at elevated power densities, directly addressing the failure mode of conventional water cooling as converter valve power densities increase.
Four technology clusters shaping the competitive landscape
The patent and literature dataset resolves into four distinct technology clusters, each representing a different position on the performance-versus-maturity spectrum for liquid and advanced cooling.
Cluster 1: Liquid metal microchannel heat sinks for extreme heat flux
This cluster focuses on replacing water or conventional fluids in microchannel geometries with liquid metals to handle heat flux densities beyond water’s capability. Research from Yangzhou Collaborative Innovation Research Institute (2022) identifies lithium as the optimal working fluid and circular cross-section as the optimal channel geometry, numerically demonstrating superiority over water at high temperature and high inlet velocity. Tianjin Polytechnic University’s electromagnetic pump-driven system (2019) achieved an overall structure thickness of just 0.9 mm with heat dissipation capacity more than double that of an equivalent ultra-thin heat pipe.
“Liquid metal microcirculation systems can achieve an overall structure thickness of just 0.9 mm with heat dissipation capacity more than double that of an equivalent ultra-thin heat pipe—without any mechanical pump.”
Cluster 2: High-temperature liquid metal systems for energy generation
Alkali metals—sodium and lithium—deployed as primary heat transfer fluids in CSP receivers and nuclear thermal systems represent the most technically mature application of liquid metal cooling. DLR’s 2017 literature reviews sodium receiver experience at Plataforma Solar de Almería and documents Helmholtz Alliance activities on system simulation and experimental validation. Safety—specifically spray fire mitigation—is identified as the primary barrier to commercial adoption, not thermal performance.
Cluster 3: Conventional cold-plate and microchannel water cooling (competitive baseline)
A large cluster of patents covering cold plate designs, serpentine/parallel/tree-shaped channel configurations, and hybrid vapor chamber systems defines the performance benchmark and engineering architecture that liquid metal cooling seeks to surpass. ABB Schweiz AG’s 2023 cold plate patent describes a porous/hollow spacer architecture maximising cooling channel volume to at least 80% of plate space for high heat flux power electronics. IMEC VZW’s 2020 patent introduces vertically oriented inlet/outlet cooling channels with direct chip surface impingement architecture.
Explore the full patent landscape for liquid metal cooling and microchannel heat sink technologies in PatSnap Eureka.
Explore Patent Data in PatSnap Eureka →Cluster 4: Immersion and submersion liquid cooling for data centre and AI infrastructure
Dielectric fluid and oil immersion cooling represents a competing high-performance approach that partially overlaps liquid metal’s addressable market in high-density computing. In this dataset, it is the most actively patented area in 2024–2026. Google LLC’s 2025 EP patent describes an infrastructure module combining cold-plate and convective air cooling with a payload module immersion system governed by a programmable logic controller. Canaan Creative’s 2026 EP patent covers an oil-path circulation immersion system for cryptocurrency and AI compute devices with flow-equalising plates and dual device slot tanks.
Tianjin Polytechnic University (2019) demonstrated an electromagnetic pump-driven gallium-alloy liquid metal microcirculation system with an overall structure thickness of just 0.9 mm and heat dissipation capacity more than double that of an equivalent ultra-thin heat pipe, with no mechanical pump required.
Application domains: where liquid metal cooling wins
Liquid metal cooling’s performance advantage is not uniform across all applications—it is most decisive in a specific set of high-heat-flux domains where water cooling approaches or exceeds its physical limits.
Power electronics and IGBT thermal management
The most direct and technically validated liquid metal cooling application in this dataset is power electronics. The Global Energy Interconnection Research Institute study (2020) demonstrates that liquid metal cooling media outperform water in IGBT heat sinks under high power density conditions, directly addressing the failure mode of conventional water cooling as converter valve power densities increase. This represents a near-term commercial opportunity as HVDC and industrial converter systems scale. Standards bodies such as IEEE are actively developing thermal management standards for next-generation power electronics that will further drive demand for higher-conductivity cooling media.
Concentrated solar power receivers
DLR (2017) documents the maturity of liquid sodium as a CSP heat transfer fluid, with validated receiver operation at Plataforma Solar de Almería and more than 30 years of system handling knowledge. The principal barrier is not thermal performance but safety engineering—specifically spray fire mitigation. Renewed interest in this sector is driven by the need for higher operating temperatures to improve thermodynamic efficiency in next-generation CSP plants operating above 600°C, where molten salt approaches lose efficiency.
High heat flux electronics: CPU, GPU, and AI accelerators
IBM’s microjet cooler achieving more than 2.5 MW/m² (2007) established that extreme heat fluxes require advanced liquid approaches. Liquid metal is the logical next step for chips exceeding water’s thermal removal capability. The Tianjin Polytechnic liquid metal microcirculation system (2019) and the IMEC direct chip cooling patent (EP, 2020) illustrate the trajectory toward direct-contact liquid metal or high-performance dielectric cooling for processor-class devices. As AI accelerator clusters scale, the heat flux density trajectory creates structural demand for higher-conductivity fluids. Research published through bodies such as Nature has documented the thermal challenges of next-generation GPU packaging that make advanced cooling essential.
Shanghai Jiao Tong University’s hollow metallic microlattice active cooling system for microsatellites (2022) achieved 301.7 K surface temperatures under stringent mass constraints—illustrating the aerospace niche where liquid metal’s high thermal conductivity and electromagnetic pumpability (no moving parts) are particularly attractive.
Electric vehicle battery thermal management
While no retrieved record explicitly applies liquid metal to EV battery cooling, this dataset contains the densest cluster of innovation on cold-plate optimisation for lithium-ion batteries—spanning research from Seoul National University of Science & Technology, Harbin Institute of Technology, Hangzhou Dianzi University, University of Exeter, and Jiangsu University of Technology. This defines both the competitive landscape and the unmet need that liquid metal cooling could address at extreme fast-charging rates above 3C. Xiamen Hithium Energy Storage Technology’s two recent EP patents (2025–2026) on structured liquid cooling plates surrounding battery cell receiving spaces signal that battery energy storage manufacturers are actively patenting novel cooling architectures.
Map the full competitive patent landscape for EV battery thermal management and AI cooling with PatSnap Eureka.
Analyse Patents with PatSnap Eureka →Geographic and assignee patterns: who holds the IP
The dataset reveals a bifurcated innovation landscape: commercial product development is concentrated among a small number of Western and East Asian companies, while fundamental liquid metal cooling research is distributed across Chinese universities and European national laboratories.
Among patent records with jurisdiction data, EP-filed patents are the most numerous in recent years—covering filings from Google LLC, IMEC, ABB Schweiz, LiquidCool Solutions, Canaan Creative, Xiamen Hithium, Asetek, and INPRO Technologies. US-filed patents include Heatscape, Auras Technology, Amulaire Thermal Technology, and Asetek. Two DE patents from Sharkoon Technologies (2004) represent early European water-cooling filings.
Google LLC (US/EP) holds the most recent and largest-scale liquid cooling infrastructure patents (2024–2025), reflecting hyperscaler investment in advanced thermal management for AI. Xiamen Hithium Energy Storage Technology (CN/EP) filed two active EP patents (2025–2026) on liquid cooling plates for battery modules, among the most recent battery-sector filers in this dataset. IMEC VZW (BE/EP) represents European semiconductor research institute leadership in advanced electronics cooling, with its direct chip liquid cooling patent (EP, 2020). ABB Schweiz AG (CH/EP) addresses high heat flux power electronics cooling from a major industrial automation incumbent (2023).
Liquid metal-specific fundamental research in the 2007–2026 patent and literature dataset is concentrated in Chinese academic and research institutions—including the Global Energy Interconnection Research Institute, Yangzhou Collaborative Innovation Research Institute, Tianjin Polytechnic University, and Guangdong University of Technology—while European research is led by the German Aerospace Center (DLR) for high-temperature solar applications.
This bifurcation has strategic implications. Western industrial IP strategists should monitor Chinese patent families deriving from this academic research base, as the trajectory from academic publication to commercial patent filing is well-established in Chinese innovation systems. The EPO‘s patent information services provide one mechanism for tracking these emerging family extensions into European jurisdictions.
Strategic implications and white-space opportunities
Liquid metal cooling occupies a defensible high-performance niche above water cooling’s practical heat flux ceiling, making it strategically relevant for the top 1–5% of heat flux applications: AI accelerator chips, IGBT power modules in HVDC systems, rocket nozzles, and CSP receivers. R&D teams should position liquid metal as a premium-tier technology, not a universal replacement for water cooling.
Electromagnetic pumpability as a key IP differentiator
The absence of mechanical pump wear, vibration, and seal failure risk makes electromagnetic pump-driven liquid metal architectures highly attractive for space, defence, and implantable or wearable electronics. IP in electromagnetic pump design for liquid metals represents an underdense filing area in this dataset and a potential white-space opportunity for organisations with relevant competencies in electromagnetic actuation.
Battery thermal management: the highest-volume adjacent market
This dataset contains the largest cluster of innovation around EV battery liquid cooling. If liquid metal or liquid metal-enhanced nanofluid cooling plates can be demonstrated at cost parity with water-based systems at extreme fast-charging rates, this market represents a significant commercial opportunity—and a major competitive threat to incumbents such as Xiamen Hithium. The transition from water to higher-conductivity fluids for batteries undergoing fast charging at 4C+ rates is a logical but not yet patent-dense direction in this dataset.
Safety and materials compatibility as IP moats
Safety and materials compatibility remain the primary commercialisation barriers for both alkali metal (reactivity, spray fire) and gallium-alloy (gallium embrittlement of aluminium alloys, toxicity profile, cost) liquid metals. IP protection of passivation layers, corrosion-resistant channel coatings, and containment architectures for liquid metal systems will be as strategically important as thermal performance patents. Organisations filing in these areas early can establish durable IP moats that protect market entry regardless of which specific liquid metal formulation prevails. Guidance from bodies such as ISO on materials compatibility and safety standards for advanced thermal fluids will shape the regulatory environment for commercial deployment.
“IP protection of passivation layers, corrosion-resistant channel coatings, and containment architectures for liquid metal systems will be as strategically important as thermal performance patents.”
Modular immersion infrastructure as the insertion point
Google’s 2025 EP patent on modular liquid cooling architecture combining infrastructure cold-plate modules with payload immersion modules, and Canaan Creative’s 2026 EP immersion system, signal that AI compute infrastructure is adopting immersion as standard. Liquid metal cold plates could serve as high-conductivity heat spreaders within such hybrid architectures—a positioning that allows liquid metal to enter the market through a component role rather than requiring full system replacement.
For R&D and IP strategy teams seeking to map white spaces, emerging assignees, and claim landscapes across the full liquid metal cooling patent corpus, PatSnap’s innovation intelligence platform provides AI-native analysis across more than 2 billion data points spanning 120+ countries.