Liquid Cooling Technologies Driving Data Centre Efficiency

Aggreko invests in liquid cooling test equipment

Aggreko, a British multinational temporary power generation company, has invested in a fleet of liquid-cooled load banks to support the testing and commissioning of AI data centre cooling systems across Europe. The units are designed to simulate the thermal and electrical loads of IT equipment, allowing operators to test cooling and electrical infrastructure before a facility goes live or during refurbishment. Liquid cooling is increasingly used in high-density AI and high-performance computing environments. During commissioning, the load banks can help operators validate equipment including chiller lines, pipework, pumps, heat exchangers, and coolant distribution units (CDUs), reducing the risk of equipment failures during operation. Load banks support testing at different power levels The units can be configured to accept 200kW, 300kW, or 600kW across single or multiple connections, accommodating different electrical supply arrangements on site. Constructed from 316L stainless steel, the equipment is designed to reduce material degradation and the risk of contamination in liquid cooling systems. Each unit and its accessories undergo specialist cleaning, flushing, sealing, and testing before deployment. The load banks also offer adjustable operating modes, allowing engineers to control power loads, temperature differentials, and leaving water temperatures within specified parameters. Multiple units can be networked to manage larger loads and support testing under different operating conditions. Torben Hiller, Lead Development Engineer at Aggreko, comments, “The rapid growth of AI has ushered in a new era of data centre cooling. However, with even a single outage presenting six- or seven-figure ramifications, ensuring that this cooling infrastructure is fit to handle the demands of day-to-day operation is paramount. “Our new LCLBs are the first of their kind on the market, and feature bespoke control systems that allow them to be set up for precise, scenario-specific testing. "Combined with the knowledge and sector experience of our data centre team, this new investment will allow us to offer unrivalled testing and commissioning support for European AI data centres and their cooling equipment, providing operators with the best possible assurance that their infrastructure will be able to handle real load conditions.” Torben also says testing under different scenarios is important as rack densities increase, warning that inadequate testing can lead to overheating, poor heat rejection, coolant contamination, and reduced equipment lifespan. For more from Aggreko, click here.

BAC introduces Trillium adiabatic cooler for data centres

Baltimore Aircoil Company (BAC), a provider of data centre cooling equipment, has introduced the Trillium Adiabatic Cooler TRFS, a double-stacked adiabatic cooler designed for hyperscale and colocation data centres. It combines BAC’s adiabatic cooling design with intelligent controls for applications where dry cooling alone may not provide sufficient performance. The TRFS is designed to help data centre operators lower fluid temperatures, reduce peak power demand and power usage effectiveness (PUE), limit water consumption, and increase free-cooling opportunities. The company says it also supports high-density chip cooling, reduces air recirculation in large deployments, and can help sites meet design conditions without oversizing chillers or other infrastructure. Cooling capacity and water management With greater capacity per unit than dry coolers, the TRFS can reduce equipment numbers, connections, and installation time. Its adiabatic design maintains capacity at high ambient temperatures and supports water-cooled chiller, chiller-less, and direct-to-CDU configurations. Depending on the design, it can also allow chillers to be downsized or eliminated. The cooler is designed to limit annual water consumption and can be configured to meet site requirements, including water constraints, permitting requirements, and sustainability goals. Factory-installed adiabatic pads and intelligent controls activate cooling when needed, balancing water and energy use. Dustin Cohick, Market Leader - Americas Data Center at BAC, says, "Data centre operators are being asked to meet more aggressive temperature, power, and water targets on increasingly compressed schedules. "The Trillium Adiabatic Cooler TRFS gives them a way to use water only when it matters most and gain meaningful capacity and efficiency benefits in return. It's already operating at multiple hyperscale and colocation data centres." The TRFS is manufactured globally and features a pre-configured design intended to shorten engineering cycles. Options and features include Whisper Quiet fans (which can reportedly reduce sound levels by up to 13 dBA); automatic night-quiet operation; reduced steel support requirements; redundancy across fans, motors, and drives; accessible service platforms; and pre-programmed controls housed in a UL enclosure to support start-up and troubleshooting. A dry cooler version, the Trillium Dry Cooler, is also available for water-free operation. For more from BAC, click here.

Supporting decommissioning through refrigerant recovery

As demand for AI, cloud-based services, and high-performance computing continues to grow, the UK data centre landscape is changing rapidly. Larger, more energy-intensive facilities are being developed to meet future digital infrastructure requirements, while some older and smaller data centres are being decommissioned, repurposed, or replaced. As part of a wider demolition project linked to the closure of a data centre, A‑Gas Rapid Recovery was appointed to safely recover refrigerant from cooling equipment before the site was stripped out, preventing the release of high global warming potential (GWP) refrigerants into the atmosphere and keeping valuable product in circulation. The challenge Data centres rely on cooling systems to maintain critical operating conditions. When these systems are decommissioned, refrigerant must be recovered safely, efficiently, and in line with F-Gas requirements. Releasing refrigerant into the atmosphere is not only harmful to the environment but also represents the loss of a valuable resource that can be reclaimed and reused. In this case, the project formed part of a broader demolition programme following the planned closure of the data centre. The customer required a specialist recovery partner capable of working safely within a complex site environment whilst helping ensure that high GWP refrigerants were captured and managed responsibly. The solution A-Gas Rapid Recovery provided an on-site refrigerant recovery solution, using specialist equipment and experienced engineers to recover refrigerant from the data centre’s cooling systems. By carrying out the recovery work ahead of demolition, A-Gas helped ensure that the refrigerant was removed safely and that the customer could progress the wider site closure project with confidence. The recovered refrigerant was managed in line with A-Gas’s Lifecycle Refrigerant Management (LRM) philosophy, which focuses on recovering, reclaiming, and repurposing refrigerant wherever possible. This approach helps reduce the need for virgin product, supports circularity, and ensures refrigerants are kept in use for as long as possible. The results • Recovered over one tonne of high-GWP refrigerants from the data centre site• Prevented refrigerant from being released into the atmosphere• Supported the wider demolition programme by enabling the safe decommissioning of cooling equipment• Kept valuable refrigerant product in circulation through responsible recovery and onward lifecycle management• Demonstrated how data centre closure projects can support both compliance and circular economy goals Jake Matthews, Head of Gas Acquisition at A-Gas, says, “Recovering refrigerant is critical to reducing emissions and protecting valuable resources. Every kilogram we recover is a kilogram prevented from being released into the atmosphere, helping our customers meet their environmental responsibilities while supporting a more circular approach to refrigerant management.” Why it matters The growth of AI and cloud-based services is increasing demand for modern data centre infrastructure. As the market evolves, older facilities may be closed, replaced, or upgraded to meet new requirements around capacity, power density, and efficiency. These transition points create an important opportunity to recover refrigerants responsibly and reduce the environmental impact of decommissioning activity. By partnering with A-Gas Rapid Recovery, customers can ensure that refrigerants are captured efficiently, handled compliantly, and directed into appropriate recovery, reclamation, or reuse pathways. For more from A-Gas, click here.

Why water hygiene is key to data centre efficiency

AI is driving higher rack densities and increasing thermal loads, placing unprecedented demands on cooling infrastructure. In this exclusive article for DCNN, Gary Nicholls (pictured above), Managing Director of Swiftclean, discusses how proactive water quality management can help data centres maximise the efficient life of critical assets whilst improving operational resilience: According to the Uptime Institute, average rack power density increased by 38% between 2022 and 2024, largely down to the launch of generative AI to a mainstream audience. We are now on an irreversible trajectory that will continue to compound thermal and energy pressures on data centres and the systems that keep them cool. As the margin for error shrinks, cooling infrastructure must be managed with the same focus on reliability and resilience as other mission-critical systems. Many data centres are moving to liquid-based cooling systems that can manage rising thermal loads, so water quality plays a fundamental role in maintaining performance and reliability. With 40% of power usage in a data centre going towards cooling on average, and with water increasingly becoming the primary means of achieving appropriate temperatures, water hygiene is essential and can have a significant impact on operational efficiency and cost. Water quality: The hidden performance factor Closed-loop water systems are a critical part of many data centre cooling infrastructures, circulating treated water through chillers, heat exchangers, cooling distribution units, and associated pipework. Although these systems are designed to operate continuously for long periods, they remain vulnerable to gradual deterioration. Oxygen ingress, depleted corrosion inhibitors, microbiological activity, and the accumulation of suspended solids can all affect water quality over time. The result is a gradual decline in system performance, with corrosion, scale, and fouling reducing heat transfer efficiency and increasing the workload placed on pumps and other components. Equally as important, poor water hygiene can also affect system reliability. Corrosion and contamination can contribute to component degradation, increasing the potential for leaks, equipment damage, and unplanned downtime. Investing in advanced cooling technology is one part of the solution, but it does not guarantee long-term performance alone. Protecting both efficiency and reliability requires ongoing water testing and treatment to maintain the condition of the water circulating through the system. Why water sampling matters Because closed-loop cooling systems often operate quietly in the background, changes in water condition can go unnoticed until cooling performance begins to decline. By this stage, deterioration may already be affecting heat transfer efficiency or the condition of critical components. Regular water testing for microbial activity provides operators with visibility into system health, allowing changes in water quality to be identified before they become operational issues. Samples can be analysed for key indicators including poor water chemistry and a range of different bacteria. Water treatment is a lifecycle process Water Treatment for Closed Heating and Cooling Systems (BG 50/2021) provides best practice guidance for monitoring and maintaining closed circuits, including chilled water systems and closed condenser water systems. The guidance highlights the importance of corrosion control, pH management, dissolved gases, suspended solids, filtration, and microbiological control in maintaining system performance. BG 50 also reinforces that water treatment is not a one-off activity but a lifecycle process involving commissioning, testing, monitoring, and adjustment throughout the operational life of the system. For mission-critical environments such as data centres, this proactive approach supports equipment protection, energy efficiency, and operational resilience. Specialist expertise in mission-critical environments The increasing complexity of data centre infrastructure also places greater demands on those responsible for maintaining cooling systems. In high-security facilities, access to critical infrastructure is tightly controlled and maintenance windows are often limited. Specialist contractors must, therefore, combine technical expertise with robust security procedures and careful planning to ensure essential water treatment and testing activities can be completed efficiently without affecting site security or operational continuity. As AI continues to increase the demands placed on data centre cooling systems, water hygiene should be treated as an operational risk control measure rather than a routine maintenance activity. Through regular water testing, effective treatment strategies, and proactive maintenance, operators can preserve thermal performance, extend equipment life, and reduce the likelihood of cooling-related downtime.

A-Gas at Chillventa: Discover the LRM journey for data centres

A-Gas is a company specialising in the lifecycle management of refrigerants, which is increasingly relevant as the industry looks to reduce its carbon emissions and move away from high global warming potential (GWP) gases. For over 30 years, A-Gas has supported clients and partners, including those in the data centre sector, on their environmental journey as an enabler of lifecycle refrigerant management (LRM) and circularity in the industry, with world-class processes for the safe recovery, reclamation, and destruction of used refrigerants. This LRM circular approach is the opposite of the 'take, make, and dispose' model that has dominated the refrigerant industry for decades. By recovering used refrigerant gases and reclaiming them to AHRI 700 standards, A-Gas makes better use of existing resources and reduces the need for virgin production. Re-supplying these products to the market helps the wider heating, ventilation, air conditioning, and refrigeration (HVAC-R) industries embrace circular economy principles. Where no future reuse is possible, A-Gas uses United Nations-approved destruction technology to eliminate the gases in a safe and environmentally responsible way. During the three days of Chillventa, A-Gas will take visitors on a journey through the different phases of Lifecycle Refrigerant Management (LRM). A-Gas will highlight how each phase impacts the refrigerant lifecycle and demonstrate how its products and services support customers at every stage. • Recovery — featuring globally renowned services such as Rapid Recovery and Rapid Exchange• Supply — showcasing its expertise in multiple refrigerants, including low and ultra-low GWP refrigerants, as well as the new blends• Reclaim — highlighting world-class reclamation capabilities that restore refrigerants to AHRI 700 standards for re-supply Why Rapid Recovery is important for data centres For data centre operators, Rapid Recovery is a key part of this approach. The sector has specific requirements in terms of speed and efficiency, which A-Gas is able to address promptly, keeping server downtime to a minimum. In this context, one aspect that is often underestimated is the logistical complexity involved in challenging environments such as hyperscale data centres and edge infrastructure. In these cases, decommissioning or retrofit operations may involve large quantities of refrigerant and require rapid intervention to minimise disruption to operations, which could otherwise result in significant inconvenience and substantial financial losses. Rapid Recovery is a high-speed, on-site refrigerant recovery programme delivered by highly specialised technicians using purpose-built proprietary technologies, lorries equipped with 120-metre flexible recovery hoses and other company assets, including 1,000-litre drums and high-performance vacuum pumps. The service removes refrigerant at speeds of up to 10 times those of traditional recovery processes, while operating in full compliance with local and international regulations. It also includes comprehensive support with all required documentation. The service can be adapted to applications of any size and is available 24 hours a day, seven days a week. A-Gas meets visitors at Chillventa Discover A-Gas solutions and more by visiting A-Gas in Hall 1 - Booth 1-332. The team looks forward to welcoming you and guiding you through the LRM journey. For more from A-Gas, click here.

Datacenter United signs multi-megawatt capacity agreement

Datacenter United, an independent Belgian data centre operator, has signed a binding agreement to provide multi-megawatt data centre capacity for AI workloads at DC Machelen in Belgium. The environment will use liquid cooling to support the power densities and heat loads associated with high-density AI systems. The customer's identity and commercial terms of the agreement will remain confidential. AI systems concentrate increasing amounts of computing power within a limited footprint, raising both the power required per rack and the amount of heat that must be removed. Power supply, cooling, connectivity, redundancy, and operations must, therefore, be engineered as an integrated environment. Liquid cooling removes heat closer to its source and can enable high-density configurations that are increasingly difficult to support with air cooling alone. The system at DC Machelen will be engineered around the customer's confirmed architecture and operational requirements. The agreement reflects demand for specialised AI infrastructure in Belgium, with the project moving from planning to deployment. Liquid cooling for high-density AI Friso Haringsma, CEO of Datacenter United, comments, “Demand for specialised AI infrastructure in Belgium is moving from ambition to actual deployment. "This binding agreement translates a specific customer requirement into a concrete, multi-megawatt project in Machelen. It demonstrates that a Belgian operator can support high-density AI workloads at this scale.” The liquid-cooling design is intended to support higher computing densities whilst improving cooling and energy efficiency. Datacenter United says the project also demonstrates the importance of access to power, connectivity, suitable locations, and predictable permitting for AI infrastructure development in Belgium. DC Machelen is designed for organisations with high power and capacity requirements. The site combines power, space, carrier-neutral connectivity, and the flexibility required for large-scale, high-density infrastructure. Technical implementation will proceed according to the agreed system configuration and deployment plan. Further information on the commissioning schedule is set to be provided at a later stage. For more from Datacenter United, click here.

Circular thinking: A new imperative for data centre cooling

In this article for DCNN, Todd Scheving, Business Development Director – Global Data Center at A-Gas, outlines why a circular approach to refrigerant management is becoming increasingly important as data centre operators balance cooling requirements with sustainability and supply considerations: The AI revolution has sparked heated debate about data centre cooling. Refrigerant decisions are emerging earlier in the value chain, spurring pressing procurement and sustainability conversations. This is for good reason: the choices operators make today will shape emission footprints for the next decade. Maintaining a circular approach to data centre cooling is more important than ever. The industry is feeling the effects. Firms are engaging with refrigerant suppliers far earlier in the planning process, integrating their sustainability teams. Questions about supply security and reclaim options are now a fixture in project conversations. In short, sustainability targets are turning refrigerants into strategic assets that must be managed. This reclassification from commodity to asset matters. Managing refrigerants holistically reveals hidden problems. Often, the solutions involve maintaining a circular approach to cooling: recover, reclaim, and destroy only what cannot be restored to spec. Reclaimed refrigerants have a significantly lower carbon footprint than virgin alternatives because the molecules already exist. Rather than venting (which is illegal across North America and the European Union) or leaving them in storage, the goal is to keep refrigerants in circulation. Circular thinking is the environmentally responsible path and, increasingly, the economically rational one. End-of-life handling is responsible for much of the industry's emission problem. Unfortunately, despite it being illegal, venting is often a by-product of strict time constraints during decommissioning or servicing. The solution is to engage dedicated, high-speed refrigerant recovery teams to recover refrigerants and deliver them to reclaimers for reprocessing for future reuse. Recovery is the future of cooling. There are enough HFCs (hydrofluorocarbons) already installed today to meet industry needs for the foreseeable future, provided they are managed responsibly. The infrastructure exists; operators must now partner with recovery teams to treat refrigerants like the strategic assets they were always meant to be. For more from A-Gas, click here.

aquatherm blue powers cooling at Novva’s data centre

At the hyperscale Novva data centre near Salt Lake City, USA, the operator’s cooling strategy is designed for maximum reliability and future orientation. The core of the waterless cooling concept is the PP‑RCT piping system aquatherm blue, which circulates the cooling medium in a closed loop and makes traditional evaporation cooling – and its high water consumption – unnecessary. Corrosion- and incrustation-free pipes ensure stable flow rates over the long term and significantly reduce leakage risks – a key factor for protecting critical IT infrastructure. Thanks to their low weight and welded joints, aquatherm blue could be integrated into the extensive underfloor construction. To learn more about the project at Novva data centre, click here. For more from aquatherm, click here.

When power is a constraint, cooling is an opportunity

In this article for DCNN, ExxonMobil’s Glen Sharkowicz (Director Strategic Market Development, Immersion Cooling) and Betsy Cossette (Immersion Cooling Market Development Manager) assess how immersion cooling could help data centre operators overcome power constraints, improve thermal efficiency, and extract more compute capacity from existing infrastructure: Immersion cooling offers relief from the growing energy bottleneck AI is accelerating demand for one of the industry’s most constrained resources: power. Across the United States, major players are aggressively pursuing the energy needed for ambitious new projects. If accessing enough power is difficult now, forecasts do not show much relief ahead. The International Energy Agency projects that global data centre electricity consumption will more than double by 2030, reaching roughly 945TWh, with the United States accounting for nearly half of the electricity demand growth over that period.[1] When power becomes a bottleneck, timelines slip, expansion costs rise, and access to compute is constrained. In the AI race, those delays can set companies back years. CBRE has reported that power availability can extend data centre construction timelines by two to four years and, in some cases, by as many as six.[2] The logic is simple: when power is the primary constraint, every megawatt counts. As processor power rises and AI deployments scale, operators are re-evaluating long-held assumptions about how data centres should be cooled. Direct-to-chip moved the market forward; AI is now changing that equation. Modern systems are generating more heat in less space, and thermal pressure now extends beyond CPUs and GPUs to memory, power delivery, and other system elements. Heat affects how much supporting infrastructure must be built around the IT load. It influences how much of a site’s total power can be directed to compute rather than cooling overhead. Direct-to-chip cooling has been an important bridge technology for high-density environments. McKinsey notes that compared with air cooling, operators can run the same tasks on the same equipment with a 31% lower energy bill using direct-to-chip cooling.[3] But, as densities continue to rise, the limitations of cooling only certain hot components become more apparent. For one, liquid cooling systems can introduce significant plumbing complexity. In 2024, Uptime Institute data cited by industry observers showed that liquid cooling system failures contributed to 13% of significant outages. Immersion extends the opportunity Instead of targeting only the hottest chips, immersion cooling places the system in a thermally stable liquid environment designed to cool the full assembly more comprehensively, which itself can reduce temperature variation, lower dependence on fans and heavy air handling, and simplify the environment needed to support high-density compute. With immersion becoming more mainstream, the most likely path forward is a hybrid approach in which cooling strategies are tailored to specific business and application needs. What’s more, immersion cooling can reduce TCO compared with direct-to-chip and air cooling. When AI is driving revenue, the ability to shift energy from cooling to compute can have outsized business impact. A recent survey by KPMG found that 55% of hyperscalers and data centre developers would be willing to pay up to 50% more on electricity costs to expand capacity.[4] When power is the limiting factor in data centre growth, the central question becomes how to get the most compute from the power already available. Immersion cooling is an answer. Immersion cooling can offer meaningful savings relative to cold plate cooling in AI-focused environments. If less power must be devoted to cooling overhead, more can be directed to compute. Power is scarce and time is expensive, so these effects can materially change the return profile of an AI infrastructure investment. Reliability is part of the ROI equation Operators need confidence that the environment will be stable, maintainable, and supportable over time. Downtime, service complexity, and unplanned maintenance can quickly erode the value of any savings. Immersion cooling can help address those concerns by creating a more thermally stable operating environment and reducing some of the infrastructure complexity that contributes to service burden. At the 2025 OCP EMEA Summit, findings presented showed that immersion cooling offered the following advantages:[5] • Higher server reliability• Reduced downtime• Lower maintenance requirements The case extends beyond the rack Water use, physical footprint, and local operating impact of data centres are becoming more prominent considerations in siting, permitting, and long-term planning. Immersion cooling can potentially lower water consumption compared to other cooling strategies, alongside improved thermal management and lower cost. A 2025 Nature study found that advanced cooling methods such as cold plates and immersion cooling can reduce blue water consumption by 31–52% in data centres, depending on conditions.[6] Traditional fan-heavy data centre environments can create significant noise pollution.[7] Immersion cooling can help reduce that burden by lowering reliance on fan-driven airflow. In some configurations, immersion systems have been measured below 50dB, which is materially quieter than conventional data centre cooling environments.[8] Immersion cooling can help maximise power to revenue It’s important to consider how much value a cooling strategy can unlock from the power, space, water, and capital already available. Implementing immersion cooling can help address thermal management at the system level and ties that improvement to core business outcomes: lower operating costs, less infrastructure burden, improved reliability, and the potential for more compute capacity within the same power envelope. Immersion cooling offers a way to turn constrained power into scalable growth, and it belongs in today’s infrastructure planning discussion. For organisations exploring how to maximise performance from constrained power, it is a conversation worth having now. For more from ExxonMobil, click here. [1] International Energy Agency, Energy and AI: Executive Summary [2] CBRE, High Demand, Power Availability Delays Lead to Record Data Center Construction [3] McKinsey & Company, Keeping Cool in the Data Age [4] KPMG, Fueling the Future: Bridging the Energy Demand Gap in the AI Era [5] Open Compute Project, Immersion Server Demonstrated Reliability at Datacenter Scale Deployment [6] Nature, Using Life Cycle Assessment to Drive Innovation for Sustainable Cool Clouds [7] WSP, Beyond Power and Cooling: Air Quality’s Role in Data Center Resilience [8] Science and Technology for Energy Transition, Immersion cooling technology development status of data centers

KROHNE introduces data centre liquid cooling flowmeters

KROHNE, a German manufacturer of process instrumentation and flow measurement technology, is targeting data centre cooling efficiency with its flow measurement technology, including chilled water and liquid-to-chip applications, as operators manage increasing power and rack densities. The company’s electromagnetic and ultrasonic flowmeters measure liquid flow in cooling water loops, secondary circuits, and heat recovery systems. This data can then be used to monitor coolant distribution, manage pump and valve operation, and assess thermal performance. In liquid-to-chip cooling systems, flow measurement can help operators monitor glycol-water and other coolant circuits supplying individual racks and processors. Accurate flow data can also support automated control systems that adjust pump speeds and valve positions according to cooling requirements. Flow measurement supports heat recovery According to KROHNE, one data centre project using hybrid liquid cooling and waste heat recovery deployed approximately 40 KROHNE OPTIFLUX 4100 electromagnetic flowmeters to measure glycol-water flow through cooling circuits ranging from DN100 to DN250. The measurements are used to monitor heat dissipation and system efficiency, as well as support a thermal management system that captures waste heat for use in district heating networks. KROHNE also supplies flowmeters and temperature sensors for heating circuits, heat exchangers, and district heating connections. Measuring heat and cold transfer can help operators assess system performance and monitor the results of efficiency measures. The company cites other applications where flow and temperature measurement has been used to track hot and chilled water between geothermal probes, heat pumps, and concrete core conditioning systems. In one building-scale installation, the monitoring contributed to reduced natural gas consumption. KROHNE says its flowmeters are designed for long-term operation in demanding industrial environments, with applications including chemical processing and wastewater treatment. The technology can be integrated with existing monitoring and control systems used for data centre energy management.