Experimental Investigation of Time-Dependent Thickness and Composition of Multicomponent Wax Deposits on Cold Surfaces

Abstract

The time-dependent rates of growth of thickness of wax deposits, as well as the spatial distributions of wax components within the deposits, formed on a cold finger from stirred baths of n-dodecane (n-C12) and mineral oils containing multiple, high-carbon-number, n-alkane wax components, were measured for multiple stirring rates of the bath and multiple oil viscosities. While the deposit thickness was found to stop growing at long time, its composition continued to become more enriched in higher molecular weight compounds. The growth of the deposit thickness was explained largely by heat transfer, but the slower enrichment of the deposit required consideration of convective and diffusive mass transfer. Both the heat and mass transfer coefficients for transport through a boundary layer at the interface between the bath and the deposit were measured as functions of stirring speed and were found to follow expected heat and mass transport correlations involving Reynolds, Prandtl, and Schmidt numbers. The findings provide a set of data that should help in the development of refined models for wax deposition on cold surfaces, including the interior surfaces of oil pipelines.

Modeling multicomponent wax deposition and aging on cold surfaces with application to pipeline fouling

“…which correlates the Sherwood number, Sh ¼ kcDcfþdeposit D w=o,i , with the Reynolds number, Re ¼ obtained for our group's previous experimental cold-finger data, 16 giving 0:038 and 0:78, which are close to the values of C = 0.0365 and m = 0.8, respectively, for turbulent flow over a flat plate. 23 From this, a formula for k c,i is obtained;…”

“…Figures 8 and 9 compare the theoretical and experimental wax component distributions 16 in the inner and outer edges of the deposit, respectively, for the Clarus wax mixture. Here, is maintained at as the WAT predicted by Coutinho model for the Clarus wax mixture is approximately equal to the measured cloud point.…”

“…Building on the single‐component wax deposition model of Mahir et al, 5 we here develop a multicomponent wax deposition model that simulates the time evolution of the deposit thickness along with the spatial and temporal compositional variation (aging) of a multicomponent deposit. The model is then applied to the cold‐finger experiment using the experimental parameters from the work of Lee et al 16 We run the multi‐component model for a single wax component in oil mixture and compare its predictions with those of the single‐component model and data of Mahir et al, 5 which shows that the multi‐component model gives predictions very similar to those of the single‐component model in this limit. We then run it for multi‐component mixtures and compare the results with the experimental work of Lee et al 16 for both the six‐component and Clarus wax mixtures.…”

“…The cold‐finger simulation is continued for up to 156 h of deposition time to obtain the wax components distribution in the deposit and the thickness trajectory. The transport parameters, drawn from the experiments of Lee et al, 16 are shown in Table 2.…”

“…The model predictions for the aging and thickness of the deposit agreed well with the experimental results for varying conditions. Lee et al 16 increased the complexity of these cold-finger experiments by using multicomponent waxes. These are a six-component…”