Heat Pipe Material Compatibility – Fluid/Envelope/Wick

A heat pipe material system includes the envelope material, the wick material, the working fluid, and any braze, solder or weld filler materials used in sealing the heat pipe. To ensure heat pipe material compatibility, a designer needs to consider a number of variables.

The process starts once the operating temperature range of the heat pipe is known, allowing the working fluid to be chosen.   Next, the heat pipe designer must choose envelope and wick materials that are suitable for the application, and are compatible with the working fluid.  Two major results of material incompatibility are corrosion and generation of non-condensable gas (NCG). If the wall or wick material is soluble in the working fluid, mass transfer is likely to occur between the condenser and evaporator, with solid material being deposited in the latter. This will result in either local hot spots or blocking of the pores of the wick. NCG generation is the most common indication of a heat pipe failure. As the NCG accumulates in the heat pipe condenser section, it gradually blocks the heat transfer area, consequently degrading the heat pipe performance.

Another consideration in envelope material selection is that the heat pipe is a pressure vessel.  It must resist both vacuum (external pressure) for temperatures below the normal boiling point (100ºC for water), and internal pressure at higher temperatures.  Ideally, the heat pipe envelope will have a high yield and tensile strength, and a low creep rate (creep is mostly important for the higher temperature heat pipes, with alkali metal working fluids). Most working fluids have a number of compatible envelope/wick materials.

The heat pipe material is selected based on a trade-off

• Thermal conductivity
• Strength/mass (high tensile strength allows a thinner wall)
• Cost

For example, copper is chosen for electronics cooling applications at temperatures below 150 ºC, since it has very high thermal conductivity, and adequate strength.  At higher temperatures, copper is not suitable due to the higher vapor pressure, so titanium or Monel envelopes are used instead.

Heat pipes can operate for 15-20 years or more, so fluid/material compatibility is very important. Problems with incompatible fluids/material pairs include:

• Non-Condensable Gas Generation, which blocks a portion of the condenser, and reduces maximum heat pipe power at a given temperature
• Corrosion, which can cause leaks that stop heat pipe operation
• Materials Transport, which can plug the heat pipe wick

The compatibility of working fluids and wall/wick materials is determined by long-term Life Tests.  In the past, life tests have been conducted on hundreds of different working fluid/envelope/wick combinations. The most commonly used fluid/envelope combinations are discussed below.   Other compatible combinations, and suggested operating temperature ranges are given below in the Compatible Fluids and Materials section.

Most Commonly Used Envelope/Fluid Pairs for Common Applications

• Electronics Cooling: copper envelope and wick with water working fluid is the most commonly used envelope/wick/fluid system for electronics cooling for temperatures between roughly 25 and 150ºC.  Water has the highest Merit number in this temperature range.  Copper/methanol is used when the heat pipes must operate at temperatures below 25ºC.  Titanium and Monel envelopes allow water to be used for temperatures up to 300ºC (short term) and 280ºC (long term).

• Energy Recovery: Energy recovery heat exchangers typically use R134a refrigerant as the working fluid, with stainless steel or copper envelopes. The refrigerant R134a is used as the working fluid with copper or steel envelopes in these products since it has no ozone depletion, low global warming potential, reasonable vapor pressure, and is non-flammable.  In addition, the HVAC community is comfortable with and experienced with this working fluid.
• Spacecraft Thermal Control: Grooved aluminum heat pipes with ammonia working fluid, commonly known as  Aluminum/ammonia CCHPs (Constant Conductance Heat Pipes)and aluminum/stainless/ammonia LHPs (Loop Heat Pipes) are the default combinations used in spacecraft thermal control.  Ammonia has the highest merit number in this temperature range For applications where the condenser temperature can fall below ~ -60ºC (ammonia freezes ~ -77ºC), ethane is used in CCHPs, and propylene is used in LHPs.
• 

  The envelope for this heat pipe is TZM (Titanium-Zirconium-Molybdenum alloy, which has 0.5% titanium and 0.08% zirconium.).

  High-Temperature Heat Pipes: Superalloy envelope & wick/alkali metal working fluids (cesium, potassium and sodium) are used for high-temperature heat pipes in the range 450ºC to 1100ºC. The merit number for the alkali metals are much higher than the merit numbers for working fluids that are suitable at lower temperatures.

  • High-temperature isothermal furnace liners have a superalloy envelope. The working fluid is cesium, potassium, or sodium, depending on the operating temperature range.
  • Ultra-high-temperature heat pipes have a refractory metal wick and envelope, such as tungsten or molybdenum,  and use lithium as the working fluid.

Compatible Fluids and Materials

Since heat pipes were rediscovered by George Grover in 1963, extensive life tests have been conducted to determine compatible envelope/fluid pairs, and a large number have been found.  Some of these life tests have been conducted for decades.  As mentioned above, most heat pipes are fabricated for electronics cooling, and are either copper/water or copper/methanol.  Most spacecraft heat pipes are aluminum/ammonia, most heat pipes for HVAC applications are copper/R134a or steel/R134a, and most high-temperature heat pipes are superalloy/alkali metal.  These envelope/fluid pairs cover the vast majority of heat pipes used today.

This section discusses the compatibility for these envelope/fluid pairs, and other pairs used in special circumstances.  The table below lists most envelope/fluid pairs used today, as well as some envelopes that are known to be incompatible.

An estimated upper and lower fluid temperature range is also shown.  In most cases, the lower limit is set by the sonic limit, while the upper limit is set by the maximum vapor pressure that can be contained with a reasonable envelope wall thickness.  Heat pipes can be built that operate at lower temperatures with a large diameter to maximize the sonic limit, theoretically down to the triple point).  Heat pipes can also be built that operate at higher temperatures, theoretically up to the critical point, but will require a thicker envelope to withstand the vapor pressure, and will have a reduced capillary limit.

The practical upper temperature limit for copper/water heat pipes is set by the vapor pressure at around 150°C; Monel or titanium are used at higher temperatures.   It is very important to note that the table lists the generic type of material, such as Monel or Superalloy.  In many cases, only some alloys are known to be compatible, while others have not been tested.  Only some stainless steels are suitable for cryogenic heat pipes, since other steels become very brittle at low temperature.

For additional help in selecting fluids and materials, please email one of our capable engineers at solutions@1-ACT.com or call us at (717)-295-6061.

Working Fluid and Envelope Compatibility Table
(includes Practical Temperature Limits)

*scroll right for additional columns

Figure 1. Sonic and Wicking Limits for Cesium and Potassium Heat Pipes. For these specific designs, the sonic limit controls the power below 400ºC for cesium, and below 500ºC for potassium.

The upper temperature limits for cesium, potassium, and NaK are set by ranking the properties of suitable alkali metals at a given temperature.  For example, cesium is not normally used at higher temperatures than 600º, since potassium is a superior working fluid.   This can be seen graphically in Figure 1, which compares heat pipe power versus temperature for identical heat pipes using either cesium or potassium as the working fluid.  On the left side of the graph, the maximum heat pipe power is set by the sonic limit (the roughly parabolic part of the curve), while on the right side of the graph, the maximum power is set by the capillary limit (the roughly flat part of the curve).  At lower temperatures, more power can be carried with cesium, since it has a higher vapor density (and higher sonic limit) at any given temperature.  Once the temperature is increased above roughly 500°C, the potassium heat pipe carries more power (for this particular design).  This is the reason that the table above states that cesium is not normally used above 600°C.

 Incompatible Fluids/Envelope Pairs

Demonstrating problems that can occur when exploring potential fluid/envelope pairs.

As discussed above, there are a large number of fluid/envelope pairs that are compatible, with some life tests conducted for decades.  There is no longer any question that such envelope/fluid pairs as aluminum/ammonia, copper/water, copper/methanol, and superalloy/alkali metals are compatible when fabricated and processed properly.  Most Heat Pipe Life Tests today are conducted as a Quality Control measure to confirm that the heat pipe fabrication processes are under control.  For example, samples of each of ACT’s extrusions for aluminum/ammonia Constant Conductance Heat Pipes (CCHPs) are on life test at elevated temperature, to demonstrate the long-term life required in heat pipes for satellites. One area for active research on compatibility is the intermediate temperature range, from roughly 250 C to 400ºC.  As discussed in Intermediate Temperature Fluids Life Tests ACT has been examining different envelope/fluid pairs for the past decade, and demonstrated several new compatible fluid pairs.  Other pairs were found to be incompatible.

Possible problems with incompatible fluid/envelope pairs include:

• Non-Condensable Gas Generation (most Common)
• Corrosion
• Materials Transport

This section provides examples of these problems.  Please remember that these problems were observed during research on new envelope/fluid pairs, or when life tests are conducted at higher temperatures than previous tests, and are not representative of standard envelope/fluid pairs.

Non-Condensable Gas Generation

The most common symptom of incompatibility is Non-Condensable Gas (NCG) generation in the heat pipe.  During operation, the NCG collects at the end of the condenser, reducing its effective length.  In turn, this reduces the power that the heat pipe can carry at a given temperature.   NCG generation is generally caused by impurities in the working fluid or contaminants on the wick or wall.

Figure 2. Therminol Life Tests at 400 and 450ºC. Gas is generated quickly at 450ºC, and more slowly at 400ºC. No gas was observed after 1000 hours at 350°C.

Gas generation can also be caused in organic working fluids when the temperature is too high.  Eutectic Diphenyl/Diphenyl Oxide is an intermediate-temperature fluid that is sold under the trade names of Dowtherm A and Therminol VP-1.  Therminol VP-1 heat pipes were life tested at three different operating temperatures, 350, 400, and 450°C, and monitored for gas generation.

As shown in Figure 2, the life test pipes operating at 450°C showed a significant increase in ΔT after only 90 hours, most likely caused by excess non-condensable gas.  These pipes were taken off the life test and purged, then put back on the life test.  Non-condensable gas generation continued at a high rate.

Charring of Therminol during life tests at 450 ºC.

The 450 °C pipes were taken off the life test after 300 hours, sectioned, and analyzed.  As shown to the left, a portion of the Therminol charred during the life test.

The gas generation rate is dependent on the operating temperature.  Figure 1 shows that the gas generation rate was much lower when the heat pipe operation temperature was reduced to 400°C.  No gas generation was observed in the 350°C life tests after 1,000 hours of operation.

Corrosion

GaCl3 is incompatible with superalloys. A leak developed at the pinch-off tubes within one week after the life test was started.

Corrosion occurs when the working fluid is not compatible with the envelope or wick (NCG can also be generated, as discussed above).  Corrosion products can block a portion of the wick, reducing the heat pipe maximum power.  In more extreme cases, a leak can develop, with the heat pipe ceasing operation.

ACT has conducted a series of life tests in the 250 – 400°C with superalloy envelopes and halide working fluids.  While several of the halides have been shown to be compatible with superalloys through life tests, gallium trichloride is not compatible.  The GaCl₃/superalloy pipes all leaked at the pinch-off weld after roughly one week of operation at 360°C (633K); as shown at left.

Materials Transport

Material transport after 1000 hour life test with cesium at 475°C. Monel and Copper/Nickel 70/30 (a) Monel. (b) Copper/Nickel 70/30. Note the copper particles that were transported from the condenser and deposited in the evaporator.

Material transport of the wick/envelope can occur when the working fluid has a high solubility for one of the wick/envelope components.  During heat pipe operation, working fluid is vaporized in the evaporator, and then condensed in the condenser, transferring heat.  The working fluid vapor is very pure with no impurities, so the working fluid in the condenser is also very pure.  If the working fluid has a high solubility for one or more of the components in the wick or envelope, these components will dissolve in the working fluid, and be carried along to the evaporator. When the working fluid vaporizes, the dissolved components are left behind.

A beautiful example of material transport with an incompatible envelope/fluid pair is shown in the photo to the right.  ACT (Passive Thermal Management for a Fuel Cell Reforming Process) examined a Variable Conductance Heat Pipe (VCHP) Heat Exchanger with cesium as the working fluid.  Cold seawater was a potential coolant for the heat pipes (with a large ΔT between the heat pipe interior and the water).  Monel and copper-nickel are known to resist seawater corrosion, so they were potential heat pipe envelope materials.

Since there was no known life test data at this time between cesium and Monel or copper-nickel, heat pipes were fabricated with Monel 400 and 70/30 copper/nickel envelopes, and life tested at 475°C for 1000 hours with cesium.  The heat pipes were then sectioned and examined; see Figure 4.  The copper grains in the evaporator show that copper was transported in the cesium from the condenser to the evaporator, hence the Monel and copper/nickel is not compatible with cesium.