Above 200 °C, heat pumps stop and thermal fluids take over

Where the heat-pump story stops

Industrial process heat is being electrified. Heat pumps, once confined to space heating and low-temperature drying, are now being deployed for processes that need steam, hot water and thermal oil at temperatures that a decade ago would have been thought out of reach. Authoritative public assessments describe commercially available high-temperature heat pumps reaching up to about 200 °C, and research programmes are targeting higher.

The reporting has focused, understandably, on the technology that is changing. That has left a gap in the discussion that matters for a buyer whose process runs at 280 °C, or 320 °C, or 340 °C.

The gap is this: the heat-pump story stops at about 200 °C, but industrial processes do not. Polymer production, carbon-fibre manufacturing, chemical synthesis, oil and gas processing, and concentrated solar-thermal plants all run above that threshold. Above it, the fluid that carries the heat is still a thermal oil or a silicone fluid, circulated through a heater or a heat exchanger rather than through a heat-pump cycle. This article is about choosing that fluid.

What the catalogue prints above 200 °C

The Glacier catalogue describes two families for high-temperature heat transfer.

The LM-16 series is described as modified hydrocarbons, insoluble in water, with flash points above 120 °C. The catalogue states the products have good thermal stability and the advantages of homogeneous heating, accurate temperature control and good heat transfer. Three models span the high-temperature range:

The important observation in that table is not the headline 340 °C figure. It is that the three models differ in viscosity by a factor of more than fourteen, even though all are described as modified hydrocarbons for high-temperature duty. A fluid that reaches 340 °C may also need to be pumpable at a cold start of −30 °C. LM-16C is the model that attempts both. LM-16A is the one that sacrifices the low end for stability at the high end.

The LM-15 series takes a different approach. It is described as modified polysiloxane — a silicone fluid — and the catalogue lists physiological inertness, good chemical stability, a low freezing point and long service life among its properties. Its three models sit lower in temperature than the thermal oils:

The silicone route is chosen where the process temperature is moderate by thermal-oil standards but the system demands chemical stability over very long service intervals, or where the low-temperature end of the window matters as much as the high end.

Open or closed — it changes the ceiling

A published observation that gets less attention than the temperature range on the datasheet is the role of the system atmosphere. Silicone fluids are typically described as stable in open systems to about 200 °C, and in closed, inerted systems to 300 °C or higher. The same principle applies to thermal oils: contact with oxygen at elevated temperature accelerates oxidation, and the oxidation products — acids, sludge, carbon deposits — change the fluid's properties and attack the system.

The practical consequence is that a fluid's published temperature ceiling is not the fluid's temperature ceiling in your system. It is the fluid's ceiling under the conditions the test was run at, which is normally an inert, closed loop. If your system is vented, or carries moisture, or sees periodic air ingress during maintenance, the derated limit is lower. Ask for stability data at your actual atmosphere, not just a temperature in a table.

Long life or wide window — you trade one for the other

Published experience with industrial thermal-fluid systems shows enormous variation in service life: the same fluid chemistry can last months in a demanding extrusion process and over a decade in a large, tightly closed, well-maintained loop. The difference is not the fluid. It is the system.

The variables that shorten life are well known: operating at or near the fluid's maximum rated temperature continuously, thermal cycling that repeatedly expands and contracts the fluid, oxygen ingress through expansion tanks or pump seals, and contamination from process-side leaks. A fluid that is technically within its temperature range but running 10 °C below its decomposition threshold every day of the year will degrade far faster than the same fluid running 50 °C below it.

For the buyer, this means that the temperature range on the datasheet is a condition at which the fluid is usable, not a condition at which it is indefinitely stable. A process at 310 °C may be within the range of a fluid rated to 340 °C, and the question worth asking is how many hours at 310 °C the fluid is expected to last before its acid number, viscosity or flash point cross a condemning limit. If the answer is "we do not have that data," the next question is whether a fluid with a wider margin above the operating temperature — and the cost and property trade-offs that come with it — is the safer route.

What to confirm before you commit

• The actual process temperature, not the setpoint. A process described as 280 °C may have hot spots 20 °C or 30 °C higher at the heat-transfer surface. Size the fluid to the film temperature, not the bulk temperature.
• The system atmosphere. Open, closed and inerted are three different environments. Confirm which applies.
• The cold-start requirement. A fluid rated to +340 °C that solidifies at +20 °C needs a heat-tracing plan before it can be pumped. Check the low-end viscosity and pour point against the site's lowest ambient.
• Expected fluid life under your conditions. Request degradation data or field references for similar processes, not just the temperature range.
• Monitoring plan. High-temperature fluids are monitored by acid number, viscosity, flash point and insolubles. Set the condemning limits and sampling interval before the fluid goes in.
• The disposal route. Spent thermal oil is a waste stream. Confirm the disposal requirements with the supplier and the site permit.

What this article does not decide

It does not recommend a fluid for a specific process, and it does not give a service life. Service life depends on the system atmosphere, the operating temperature relative to the fluid's limit, the cycling regime and the maintenance programme — four variables that a catalogue cannot know. The catalogue gives the temperature range and the reference properties; the process and the system give the rest.

It also does not predict when heat pumps will reach 300 °C. Research programmes are underway, and the date at which a laboratory demonstration becomes a commercially available system is not a number this article can supply. What it can supply is a clear statement of where the commercially available technology stops today, and a description of the fluid families that the catalogue carries for the temperature band above it. Purchasing decisions taken this year for processes above 200 °C should be taken on today's evidence, and the industrial process guide and the technical document request page are the starting points for that evidence.