DeepStar Comparisons of Cloud Point Measurement & Paraffin Prediction Methods

Abstract

Abstract The performance and costs of deepwater developments can be compromised if paraffin deposition occurs in a subsea production system. Given the desirability of accurate cloud point measurement and paraffin prediction methods, and the consensus that available capabilities did not satisfy normal design level confidence, DeepStar undertook an evaluation process. The primary goal was to use field experience 1) to devise guidelines for obtaining and using measured cloud points, and 2) to "ground truth" the various paraffin thermodynamic and deposition models in use within the industry. Cloud point measurement studies showed that under preferred operating conditions, cloud points by each of four methods agreed with the average value of all methods within 3 F. The reproducibility for duplicate field samples run by the same method was about 5 F. When very slow cooling rates were not possible, an approximate supercooling correction in F of 2 times the cooling rate in F/min was determined. A 5-component cloud point standard showed promise for method evaluation. A definite advantage was found to using more than one cloud point method for a given sample because wax detection sensitivity was method and crude oil dependent. There was good agreement between measured stock-tank oil cloud points and the highest temperature at which wax buildup was detected in the field. Results for ten paraffin thermodynamic models indicated several available models provided useful cloud point forecasts. A representative fluid sample was required to obtain data input for these models, including quantitative n-paraffin analysis using high-temperature gas chromatography. Comparisons of cloud point results for stock-tank versus live oil suggested that the hotter (more conservative) stock-tank oil predictions and measurements more closely approximated field paraffin deposition experience. Six paraffin deposition models correctly identified that wax would deposit and which wellbore had the higher buildup rate. However, predicted wax thicknesses and patterns of paraffin deposition varied widely. and could not be judged quantitatively using available field data. Overall, predicted maximum depths of paraffin deposition showed poor agreement with field observations. Good field data for evaluating cloud point measurement and paraffin prediction methods were difficult to obtain, and additional work to field validate paraffin deposition know-how is warranted. Introduction The DeepStar projects described here were motivated in June 1994 by the need to make early decisions regarding deepwater developments where paraffin deposition can compromise project costs and risks. Paraffin predictive methods which were both effective and available were considered essential to allow identification of problematic fluid types and to guide selection of management options. Given the consensus that paraffin predictive methods could not be used with normal design level confidence, an evaluation process was undertaken to "ground truth" the various models in use. Paraffin predictive methods currently used by the industry were screened via a questionnaire distributed to DeepStar members and suggested vendors. Responses revealed interest in two types of paraffin models: thermodynamic and deposition. Thermodynamic models predict the amount of wax out of solution. A common output is the cloud point, the saturation temperature where the heaviest n-paraffins crystallize for a given oil composition and pressure. As temperature cools below the cloud point, thermodynamic models predict that increasing amounts of wax with a wider carbon number distribution precipitate. Deposition models predict whether temperature conditions cause wax to deposit on a surface. Common outputs are the location and rate of wax buildup. Deposition models can also assess prevention/remediation strategies. P. 257^