Cloud-Point Determination for Crude Oils

Abstract

Summary The cloud point represents the temperature at which wax or paraffin begins to precipitate from a hydrocarbon solution. Conventional American Soc. for Testing & Materials (ASTM) procedures for cloud-point determination are not applicable to dark crude oils and also do not account for potential subcooling of the wax. A review of possible methods and testing with several crude oils indicate that a reliable method consists of determining the temperature at which wax deposits begin to form on a cooled surface exposed to warm, flowing oil. A concurrent thermal analysis of the waxy hydrocarbon can indicate the presence of possible multiple wax-precipitation temperature regions in the solution. Introduction The oil industry has experienced a series of wax-precipitation problems in the Gulf of Mexico evidenced by wax deposition in well tubulars, flowlines, and gathering pipelines as well as dehydration difficulties with slightly waxy crude oils. The effort described here on cloud-point determination arose from an attempt to resolve actual or potential difficulties in the field. The questions that needed to be answered consisted of which crude oil or condensate is subject to deposition, under what conditions would such deposition occur, at what rate would the deposits grow, and how can the deposits be eliminated or reduced. The cloud point, the temperature at which wax begins to precipitate, is essential in determining which crudes will form deposits under field conditions as well as for establishing the minimum required dehydration temperature for avoiding wax/oil/water rag layers detrimental to water settling. During evaluation of a series of crude oils, it became apparent that the accuracy of previously used methods for cloud-point determination was questionable for the majority of samples tested. The methods consisted of noticing deviations from exponential temperature decay during sealed sample cooling as well as visual observations of cloudiness in thin-sections of the oil. These methods appeared satisfactory (5 F) for samples that precipitated large quantities of wax suddenly (on the order of 0.2 wt%/ F). The large quantities of wax available for precipitation make the cloud-point determination easy and apparently relatively insensitive to the method used for its determination. However, most of the Gulf of Mexico oils tested to date have wax-precipitation rates an order of magnitude smaller than indicated above, and the cloud-point determination for these crudes was difficult and strongly dependent on the method used. Various methods were considered and tested over the past few years. This paper describes the results of these tests. Review of Possible Methods ASTM Standard Test Method for Cloud Point (ASTM Standard D-2500) and Method Modifications. The ASTM standard cloud-point determination method is strictly applicable to relatively transparent oils because it determines the cloud point by visual observation of cloudiness in the bottom of a 1 1/2-in.-diameter glass jar. The glass jar is placed in a somewhat larger metal cylinder that is immersed in a cooling bath. The temperature of the sample oil is read by a glass thermometer resting against the bottom of the jar. The oil sample is not stirred. The cooling bath is 60 to 80 F cooler than the initial sample temperature. The procedure contains obvious inaccuracies. Because of thermal inertia, the relatively large mercury-bulb glass thermometer will lag the dynamically cooling sample temperature. In addition, because the sample is not stirred, the coldest sample is at the jar bottom and the cloudiness appears there, but the mercury-bulb thermometer extends well above this layer. Some of the perceived inaccuracies inherent in the ASTM procedure can be removed by an apparatus like that in Fig. 1, which was designed specifically for determining cloud-point variations with pressure in condensates. In this case, the bulky glass thermometer has been replaced by a small, fast-response thermocouple and provision has been made for sample stirring by three glass spheres in the sample container. Stirring is achieved by gentle tipping of the container, which rolls the spheres along the container wall and agitates the sample. The sample agitation serves to mix the sample to provide a relatively homogeneous, measurable temperature as well as to provide easier detection of presence of solid wax. Moving discontinuities are easier to detect by eye than static ones. The cloud-point apparatus in Fig. 1 can handle pressures up to 2,000 psig. To demonstrate the method, a sample of condensate after the primary separator from a Gulf of Mexico Vermilion well was captured in a sample bomb for cloud-point analysis as a function of degassing pressure. The condensate is very clear with only a slight green tint and has a wax content of 25.3 wt%. Fig. 2 shows the results for both the ASTM-recommended cooling method (appearance of cloudiness) and for a heating procedure (disappearance of cloudiness). An atmospheric pressure sample was also tested by the ASTM Standard D-2500 procedure; Table 1 summarizes the results. Stirring and use of a rapidly responding thermocouple appears to give a slightly higher cloud point than the standard ASTM method. However, more significantly, a substantial disagreement exists in indicated cloud points between the heating and cooling cycles of the sample. The precipitated wax disappears at a 14 F-higher temperature than the temperature at which it appears during the cooling cycle. Clearly, the wax can exist in a subcooled but dissolved state as well as in a superheated solid state. The cooling and heating rates used in the tests were designed to give an 1 F/min temperature change, which is comparable with rates for the ASTM Standard D-2500 procedure. Existence of subcooling and superheating implies that dynamic tests will always be in error. Only stepwise, static tests can establish the true cloud point. We speculated that the degree of subcooling is higher than that of superheating. Precipitation of wax not only requires a sufficiently cool temperature but also a nucleation site on which it can come out of solution. The cooled wax molecule therefore must meet a nucleation site to precipitate. In the case of the Vermilion well condensate, such sites appear too limited in number. On the other hand, the wax molecule must be at the proper temperature and surrounded by the solvent to dissolve. Essentially, an infinite number of solvent molecules are available; therefore, the superheating should be less extensive than the subcooling. The rate of dissolution will, of course, be determined by the molecular diffusivity and under saturation of the solvent around the wax particle; but both of these parameters also determine the rate of growth of a wax crystal and should not justify variable degrees of subcooling and superheating. In summary, dynamic tests are subject to error as a result of subcooling and superheating of the wax. The subcooling is thought to be more severe. The actual cloud point is thought to be somewhere between the average of the heating- and cooling-cycle cloud points and the heating-cycle value. The observed subcooling can possibly be reduced by addition of nucleation sites to the oil. Such an approach was not pursued in this paper because appropriate nucleation sites were unknown and a risk of altering the oil/wax behavior existed. Lower rates of temperature change should bring the cooling- and heating-cycle results into closer proximity. P. 681