Non-linear dependency of wax appearance temperature on cooling rate

Abstract

The Wax Appearance Temperature (WAT) is usually the first parameter to consider, when assessing whether paraffin wax contained in the crude oil can become an issue during production. As such, correct determination of the WAT is crucial. Changes in cooling rate can cause large differences in the measured WAT. This article therefore investigates the relationship between WAT and cooling rate in detail. A new algorithm is devised to determine the WAT from experimental data more accurately. This algorithm is tested with waxy model solutions and crude oil, and the results are compared between established techniques such as differential scanning calorimetry (DSC), cross-polarized microscopy (CPM), and viscometry. The data is in good agreement with modelling from homogeneous nucleation theory. Contrary to the general belief, the results show a predominantly non-linear dependency of WAT on cooling rate within the tested range of 20 °C/min to 0.5 °C/min. The assumption of linearity is therefore an oversimplification. Linear extrapolation to a zero cooling rate should only be done for cooling rates between 0.5 °C/min and 2 °C/min, as this range showed to be approximately linear. However, large gaps can exist between the theoretical solubility limit and the measured or extrapolated WAT. Moreover, for some samples one technique could be more sensitive at high cooling rates, whereas another technique could be more sensitive at lower cooling rates. It is therefore concluded that when assessing WAT, testing a representative range of cooling rates in addition to testing multiple instruments could be helpful in determining the worst-case-scenario.