Methods for Predicting Wax Precipitation and Deposition

Abstract

Summary. Removal of wax from wells and flowlines can account for significant additional operating costs. To evaluate these potential costs, the operating conditions that allow waxes to precipitate in the wellbore must be identified, and deposition rates must be estimated to determine the costs associated with removal of wax deposits. Presented in this paper are laboratory and analytic methods that can be used to estimate both the critical operating conditions and the deposition rates. The laboratory tests and analysis presented may be used to characterize any type of oil. Introduction Deposition of wax in production tubing and pipelines reduces operating efficiency. The costs of removing wax deposits can be substantial. The degree of wax deposition in wells, however, cannot always be predicted by previous experience in a particular reservoir. The operating conditions in the well are as important as the oil composition in predicting wax problems. Wax deposition is particularly problematic in low-rate wells. Low flow rates in wells affect wax deposition mainly because of the greater residence time of the oil in the wellbore. The increased flow time permits more heat loss and leads to lower oil temperatures, which in can lead to wax precipitation and deposition. Wellbore studies have shown that the temperature profile in the wellbore is a strong function of the flow rate. The operating pressure in a wellbore also affects wax deposition through its effects on solution gas. As oil flows up the wellbore, its pressure drops, and if the oil is saturated with gas, this pressure drop causes gas to be liberated. Because solution gas acts to some degree as a solvent for waxes, the loss of dissolved gas as the crude flows up the wellbore makes precipitation more likely. Solution gas also affects the rate of wax accumulation through its effects on oil viscosity.A program of laboratory experiments was undertaken to resolve differences between previous predictions and field experience and to quantify the effects of varying the oil flow rate. Three types of experiments were undertaken:wax crystallization experiments on live oil to define the conditions under which wax would precipitate,diffusion deposition experiments on dead oil to determine the contribution of wax diffusion to deposition rates under various conditions, andshear deposition experiments, also on dead oil, to determine the rate of transport of precipitated wax particles to the wall. Wax Crystallization Experiments Background. The purpose of performing the wax crystallization experiments was to determine the temperature at which wax precipitation first occurs in the wellbore (the crystallization temperature). This critical waxing temperature depends on the composition of the wax-containing fluid and therefore on the amount of gas dissolved in the oil. The oil used in this study is saturated with gas under reservoir conditions; as it flows up the wellbore, its pressure drops and gas is liberated. The wax-containing fluid therefore has a continually changing composition. Because gas in solution in the oil acts to some degree as a solvent for wax, the loss of solution gas raises the wax crystallization temperature. When the temperature at some point in the wellbore is lower than the crystallization temperature of the fluid at that location, wax will begin to come out of solution and become available for deposition on the wellbore walls. Ideal-solution theory provides a basis for a theoretical understanding of the crystallization phenomenon. According to the theory, the crystallization temperature is a function of the number of moles of solute (wax) in solution, the number of moles of solvent (live oil) dissolving it, and the melting point and latent heat of fusion of the solute. The nature of the solvent is considered to be irrelevant. The theory states that .......................................(1) Note that when the temperature drops to a point where fw, the mole fraction of wax that the oil can hold in solution, drops below the amount of wax previously in solution, wax will precipitate. The liberation of gas will contribute to precipitation by essentially concentrating the wax in solution: a given amount of wax will be dissolved in fewer and fewer moles of total solvent as gas is evolved. The model described by Eq.1 is simplistic for several reasons. First, it ignores the less-than-ideal solvation power of the oil and assumes, for example, that a mole of methane can dissolve as much wax as a mole of decane; in reality, the solubility of a higher-molecular-weight paraffin in a paraffin-based oil increases with increasing molecular weight of the solvent. Second, the model treats the wax as a single quantity when it is actually a complex mixture of components. Thus, the model is a lumped-parameter model and, strictly speaking, the latent heat of fusion and melting point have no meaning. Finally, the model does not address other phenomena that can affect precipitation of waxes. In particular, it ignores nucleation and flocculation phenomena, for which volumetric effects can be important, as well as the existence of micellar structures, which can hold asphaltic waxes in suspension in some instances, yielding an apparent solubility. Nevertheless, idealsolution theory provides a good starting point, especially because the oils in this study contained predominantly paraffinic waxes in paraffin-based oils. Equipment and Procedures. Crystallization temperatures of live oil were measured as functions of bubblepoint pressure in the filtration apparatus shown in Fig. 1. Live oil of a predetermined bubblepoint pressure was stored in an insulated piston cylinder, which was heated to reservoir temperature (191 degrees F 6 degrees [88 degrees C 3 degrees]). The oil was pumped from the cylinder through two filters in series at a pressure maintained above the bubblepoint by a backpressure regulator downstream of the filters. The first filter was of 1.2- m mesh and was maintained at about 165 degrees F [74 degrees C] by a water bath heated with an immersion heater controlled by an on/off temperature controller to 2 degrees F [1 degrees C). This first filter was used to remove any non-wax solids from the oil. The second filter was also of 1.2 m mesh and was immersed in a variable-temperature glycol bath, which could be controlled between about -20 and 160 degrees F [-29 and 71 degrees C]. SPEPE P. 121^