Comparison of Two SEM-EDX Methods for the Analysis of Produced Water Scale Filter Papers

Abstract Low-vacuum scanning electron microscopy (SEM) / energy dispersive x-ray (EDX) analysis can be used to characterize the nature of inorganic scale from produced water (Method 1); routinely used to visually determine the degree, form and composition of scale particulates. Quantitative data on scale coverage can be extracted through image analysis, and morphology can indicate origins of particulates (transported scale, active scale…). Recent trends demand more detailed quantitative analysis, believed to produce more accurate / reproducible results. Such a method is automated SEM-EDX particle analysis (Method 2). This has the advantage of full automation and delivers quantitative data on scale coverage, composition, shape and size. Neither method is perfect, the first relies on experienced SEM users, is a manual method, susceptible to bias, and is often perceived as producing qualitative data, while the second method although producing large quantitative data sets, depends upon the criteria used to classify particles, and can be time consuming. Both methods were used to examine a number of filtered produced water samples. The traditional manual method provides good representative results on scale coverage, details on particulate morphology and composition, and can be undertaken in about thirty minutes per sample; it is also a simple matter to differentiate between particulate and blanket scale deposits. The second method generates superior levels of quantitative data, but results are dependent on image thresholding (for particle selection), erroneous misleading results are all too easily obtained (unless rigorously tested particle classification schemes are used), and the method can take in excess of an hour per sample. In general Method 1 should be adequate to track scale issues from produced water, which can be supplemented where desired by automated particle analysis (APA). Where APA is to be used it is recommended that an industry standard classification criterion be developed, which will increase the degree of confidence that can be applied to results, and allow direct comparison of results between laboratories.