GRC and Grundfos Nano-Oil Validation
Long-term thermal performance and material compatibility testing of Synano nano-oil in a single-phase immersion cooling system
By introducing Synano’s nano-oil into a single-phase immersion cooling system, Green Revolution Cooling and Grundfos recorded a consistent 10 to 15% reduction in thermal resistance over 12 weeks of continuous operation. Testing ran on a live Dell server with two 270W CPUs fully submerged in a GRC Pico tank, with Grundfos pumps running in parallel under the same fluid. No material compatibility issues were detected on the server, tank, or pump components. Performance held stable from week one to week twelve, confirming long-term reliability alongside the thermal gain.
[ Project ]
COOLING CHALLENGE
Immersion cooling is gaining ground in high-density data centers, but operators and infrastructure partners face a fundamental unanswered question before committing to a new fluid class: does it hold up over time, and what does it do to the equipment it touches?
Nanoparticle additives have shown clear thermal benefits in laboratory settings, but no long-term study had been conducted within the OCP community to evaluate their real-world behavior: how thermal performance evolves over weeks of operation, whether nano-additives remain stable in suspension, and whether they cause wear or degradation to server hardware, tank walls, or pump components. Without that data, adoption stays theoretical.
Synano, GRC, and Grundfos set out to fill that gap with a structured, multi-partner study covering all three dimensions: thermal performance, long-term stability, and material compatibility.
Industry
Data Centers / Immersion
Product
Synano Nano-Oil
Location
Austin, Texas (USA) / Denmark
Year
2025/2026
SOLUTION
Synano formulated and supplied the nano-oil used throughout the study. The fluid, designated Nano-oil E, was produced at a 75:25 ratio of base hydrocarbon oil to nano-additive, using 200 nm aluminum oxide nanoparticles. Compared to the base oil, the nano-oil delivered 3% higher thermal conductivity, 15% lower kinematic viscosity, and a 90% increase in dielectric breakdown voltage, from 30 kV to 57 kV.
GRC ran the thermal performance testing at their facility using a single-phase Pico tank (50 gallons) with a Dell R650 server running two CPUs at 270W each. Synano supplied 250L of nano-oil for this stage. Testing ran weekly for 12 consecutive weeks across three fixed conditions: radiator fan at 100% with 30 L/min flow, and fan at 30% with 37.5 and 45 L/min respectively. CPU temperature, heatsink surface temperature, and fluid inlet temperature were logged at each weekly interval. A baseline test using the original base oil was run both before and after the nano-oil phase to confirm repeatability and eliminate the possibility of residual fluid effects.
Grundfos ran concurrent pump compatibility testing using three Grundfos CME3-3 AQQV horizontal multistage pumps, each operated at identical duty points (4 m³/h at 28 m head) for different durations: 1.5 months, 3 months, and 5 months. Shaft seal topography was measured before and after each test. Viton A elastomers were exposed separately to the fluid for 4 weeks at 40°C and assessed for compression set, volume swell, modulus drop, and tensile strength change.
RESULTS
The nano-oil delivered consistent thermal resistance reduction across every test condition, every week, for the full 12-week period.
Initial testing showed 10 to 15% lower thermal resistance compared to base oil across both CPUs, at all fan power and flow rate combinations tested. The improvement was most pronounced at lower flow rates and reduced fan power, confirming that the nano-oil performs best precisely where conventional fluids struggle most: under constrained cooling conditions.
Long-term results held steady. Weekly measurements for CPU temperature, heatsink average temperature, and fluid inlet temperature remained stable from week one through week twelve, with no degradation trend visible across either CPU. Thermal resistance reduction stayed between 8 and 14% throughout the full test period.
The final baseline test with pure base oil confirmed repeatability. Heatsink temperatures returned to levels consistent with the initial baseline, confirming that the nano-oil left no adverse residual effect on the system. Heat transfer in the final base oil run was slightly higher than the initial baseline, indicating that the nano-oil had not degraded the system’s thermal performance in any measurable way.
On the pump side, neither Pump 1 nor Pump 2 showed leakage at any point during operation. Disassembly revealed no wear on any stainless-steel hydraulic components. Shaft seal profilometry showed only minor surface polishing within acceptable tolerances. The Viton A elastomers passed all four compatibility tests with no notable changes after four weeks of fluid exposure.
TECHNICAL AND PERFORMANCE IMPLICATIONS
- 10 to 15% lower thermal resistance versus base oil, measured at the heatsink-to-fluid interface
- Results consistent across all fan power and flow rate conditions tested
- 8 to 14% reduction maintained steadily over 12 weeks of continuous operation
- 90% higher dielectric breakdown voltage, from 30 kV to 57 kV
- 15% lower kinematic viscosity, reducing pumping resistance
- No leakage detected across two Grundfos pumps running for up to 3 months continuously
- No wear on stainless-steel hydraulic components after disassembly
- Shaft seal surfaces showed polishing only, remaining within acceptable tolerances
- Viton A elastomers passed all compatibility tests after 4 weeks of immersion at 40°C
- Final baseline test confirmed no adverse residual effect on system thermal performance
- Drop-in formulation with no hardware changes nor system modifications
CONCLUSION
The GRC and Grundfos study produced what the industry has been missing: a controlled, multi-partner, long-term dataset on nano-oil performance in a real immersion cooling environment. The result is clear. Thermal resistance dropped by 10 to 15% from day one and stayed there across 12 weeks of weekly testing. The fluid was stable, the pumps showed no wear, and the hardware came out clean.
For data center operators evaluating immersion cooling at higher power densities, the implication is direct. Lower thermal resistance at the heatsink-to-fluid interface means lower CPU operating temperatures, more headroom for next-generation processors, and potential reductions in cooling infrastructure CAPEX. A 90% improvement in dielectric breakdown voltage also increases system safety margins without requiring a change to tank or pump specifications.
This is the first study of its kind conducted within the OCP community. The data is available on the OCP website with performance validated. The next step is scaling it.
[ Contact ]
Interested in similar results?
Synano offers collaborative testing and design support for data center operators, immersion cooling infrastructure partners, and OEMs seeking validated performance improvements through nano-enhanced thermal fluids.