ASTM D6743 Standard Test Method for Thermal Stability of Organic Heat Transfer Fluids
8. Procedure
8.1 Determine the initial boiling point (IBP) and final boiling point (FBP) ofthe unstressed heat transfer fluid by GC, in accordance with Test Method D2887 with the following requirements: the column shall be wall-coated open tubular type of 7.5 m to 10 m length with a 100 % polydimethylsiloxane film thickness of 0.88 µm, the detector shall be flame ionization type, the initial oven temperature shall be set to 35 °C (95 °F) eliminating cryogenic cooling, the calibration mixture shall cover the boiling range from n-C5 to n-C60. The following GC parameters are recommended: oven temperature rate 10 °C (18 °F) per minute, oven final temperature 375 °C (707 °F), time at oven final temperature 3 min, injector initial temperature 100 °C (212 °F), injector temperature rate 10 °C (18 °F) per minute, injector final temperature 375 °C (707 °F), detector temperature 375 °C (707 °F).

8.2 Measure the mass of a clean, dry test cell including compression fittings to the nearest 0.01 g. Pour the unstressed heat transfer fluid into the clean, dry test cell in a vertical position. The quantity of heat transfer fluid transferred to the test cell shall be 27 g +/- 0.2 g. Invert the test cell in a vertical position and allow it to drain until all free-flowing material has been removed. More viscous fluids may require as long as 15 min to drain completely. At the end of the draining period, tap the test cell to remove a drop clinging to the open end of the test cell - do not wipe away any fluid. Measure the mass of the test cell and its remaining contents including compression fittings to the nearest 0.01 g.

NOTE 2 - The intent is to perform this step only once for each heat transfer fluid being tested at this time.

8.3 Measure the mass of a clean, dry test cell including compression fittings to the nearest 0.01 g. Introduce high purity nitrogen using tubing at the bottom of the clean, dry test cell for 2 min at 60 mL/min to 70 mL/min.

NOTE 3 - To ensure accurate results, at least three test cells containing samples of the same heat transfer fluid should be heated simultaneously.

8.4 Pour the thermally unstressed heat transfer fluid into the clean, dry test cell. The quantity of heat transfer fluid transferred to the test cell shall be 27 g +/- 0.2 g.

8.5 Completely displace the air remaining in the gas space in the test cell by introducing high purity nitrogen using tubing just above the liquid surface of fluid inside the test cell at 30 mL/min to 35 mL/min for 12 min at ambient temperature.

8.6 Carefully seal the test cell and measure its mass to the nearest 0.01 g.

8.7 Insert the test cell vertically in the oven. Adjust the heating oven to the proper test temperature. The time to achieve proper test temperature should be approximately 3 h. The test temperature shall be maintained throughout the entire test duration and controlled in such a way that the temperature of the test liquid does not deviate by more than +/- 1 °C (+/- 1.8 °F) at any location, including the heated wall. Temperature shall be measured and recorded throughout the test at least once per day. If test cells containing different fluids are tested at the same time, the test cells shall be distributed symmetrically inside the oven to minimize the effect of oven temperature variation on the results. The test duration shall be the time from attaining the test temperature to the time the heat supply is cut off. The test duration at the specified test temperature shall be a minimum of 500 h. The preferred test duration is 500 h +/- 1 h, however, a longer test duration may be used. Thermal degradation cannot be assumed to be linear with time. Therefore, the stability of two fluids can only be compared at the same test temperature and test duration.

8.8 Protect the oven from heat transfer fluid that may spill in case of damage by placing a collecting pan under the test cell. (Warning - If fluid leaks out due to improper sealing of the test cell, there may be the potential of flammable vapors inside the oven. The oven design and installation should consider this possibility.)

8.9 At the conclusion of the heating period, shut off the oven. Do not immediately remove the test cell. Leave the oven closed and allow the oven and the test cell to cool to ambient temperature to reduce the internal pressure. (Warning - Pressure inside the test cell may reach several thousand kPa during the test.)

8.10 Remove the test cell from the oven. (Warning - Use adequate safety precautions when removing the test cells from the oven in case some portion of the equipment is still hot.)

8.11 Carefully measure the mass of the test cell to the nearest 0.01 g. If the evaporation loss of gaseous decomposition products is calculated at greater than 0.5 mass %, the test should be repeated since this would indicate tube leakage.

8.12 Place the test cell in a Dewar flask containing a cooling mixture of acetone or isopropanol and dry ice. Allow the test cell to cool to at least -55 °C (-67 °F). The duration of cooling is approximately 5 min to 10 min. Stand the test cell in a vertical position and allow it to reach ambient temperature, then exercise care to remove any condensed water on the exterior of the test cell. Stand the test cell in a vertical position and open the top of the test cell. Then measure the mass of the test cell including compression fittings and its contents to the nearest 0.01 g. Put a portion of the fluid into sample bottles for analytical evaluation and store the remainder for additional measurement in a glass bottle that is hermetically sealed. Invert the test cell and allow it to drain until all free-flowing material has been removed. More viscous fluids may require as long as 15 min to drain completely. At the end of the draining period, tap the test cell to remove a drop clinging to the open end ofthe test cell - do not wipe away any fluid. Measure the mass of the test cell and its remaining contents including compression fittings to the nearest 0.01 g.

8.13 Visually observe the appearance ofthe fluid sample for any insolubles, or other changes in the fluid. Examples include high pressure upon opening the test cell, appearance of fouling in the head space of the test cell and evidence of a leak from the test cell. Observations shall be noted in the report.

8.14 Determine the mass percentage oflow and high boiling components in the thermally stressed sample, in accordance with Test Method D2887 using the same equipment and requirements as specified in 8.1.

8.15 The decomposition products that cannot be vaporized are determined separately in a bulb tube distillation apparatus. Measure approximately 4 g of the thermally stressed heat transfer fluid into the distillation flask. Record the mass to the nearest 0.01 g. Apply vacuum slowly by means of a vacuum pump. Pressure shall be 0.1 mm Hg +/- 0.01 mm Hg at the end of distillation. Heat the bulb tube distillation apparatus slowly to 250 °C (482 °F). Avoid any delays in boiling. Continue distillation for at least 30 min after constant mass of distillation residue is achieved. Measure the mass of the residue in the distillation flask to the nearest 0.01 g.

NOTE 4 - The heat transfer fluid is not further thermally damaged by the distillation process.

8.16 Compare the GC test results from the thermally stressed sample to those of the unstressed heat transfer fluid.

9. Calculation
9.1 The distillation curves of the heated samples and of the original heat transfer fluid are determined by way of simulated distillation by gas chromatography, in accordance with Test Method D2887 (with the exceptions noted in 8.1). Determine the initial boiling point and the final boiling point of the thermally stressed and unstressed heat transfer fluid.

9.2 The components ofthe heated samples are subdivided as follows:
9.2.1 Gaseous decomposition products (G).

9.2.2 Low boiling components (LB).

9.2.3 Fluid within the unstressed fluid boiling range (F).

9.2.4 High boiling components (HB).

9.2.5 Decomposition products that cannot be vaporized (R).

9.2.6 Unstressed fluid remaining in the test cell (FR).

9.2.7 Material remaining in the test cell after heating (MR).

9.2.8 Decomposition products remaining in the test cell (DR).

9.3 The mass percentage m(G) is determined by subtracting the mass ofthe opened test cell measured in 8.12 from the mass of the sealed test cell in 8.6, dividing by the mass of fluid measured into the test cell in 8.4, and then multiplying by 100.

9.4 The mass percentage m(R) is determined by dividing the mass of the residue measured in 8.15 by the mass of thermally stressed heat transfer fluid measured into the distillation flask, and then multiplying by 100.

9.5 The mass percentage m(FR) is determined in 8.2 by subtracting the mass of the clean, dry test cell from the mass of the test cell and its remaining contents, dividing by the mass of fluid measured into the test cell, and then multiplying by 100.

9.6 The mass percentage m(MR) is determined by subtracting the mass ofthe clean, dry test cell measured in 8.3 from the mass of the test cell and its remaining contents measured in 8.12, dividing by the mass of fluid measured into the test cell in 8.4, and then multiplying by 100.

9.7 The mass percentage m(DR) is determined by subtracting m(FR) from m( MR). The material represented by m(DR) is considered to consist of decomposition products.

9.8 Mass percentages m(LB) and m(HB ) can be obtained from the boiling graphs, directly from the gas chromatogram, or by using simulated distillation software.

9.9 To take into account the mass percentages m(G), m(DR) and m(R), the mass percentages m(LB) and m(HB) must be corrected in accordance with Eq 1 and 2.
mcorr(LB) = m(LB)·[{100 - m(G) - m(DR) - m(R)}/100], in %
where:
mcorr(LB) = corrected mass percentage of low boiling components.

mcorr(HB) = m(HB)·[{100 - m(G) - m(DR) - m(R)}/100], in %
where:
mcorr(HB) = corrected mass percentage of high boiling components.

9.10 The degree to which secondary products are formed is equated with the degree of decomposition, as mass percentage of heat transfer fluid, in accordance with Eq 3 and 4. Mass percentage m(DL) represents the total low boiling decomposition products. Mass percentage m(DH ) represents the total high boiling decomposition products.
m(DL) = m(G) + mcorr(LB), in %
m(DH) = mcorr(HB) + m(R) + m(DR), in %

9.11 Test temperature, test temperature variation and test duration shall be recorded for each test. The smaller the degree of decomposition (at the same test temperature and test duration) of a heat transfer fluid, the higher is the product's thermal stability.