Performance Tests of Gas Characterization Methods for Predicting Freeze-out in LNG Production

Hassan A. Attalla, Nour A. El-Emam, Tarek M. Aboul-Fotouh, Eric F. May

Research output: Contribution to journalArticlepeer-review

1 Citation (Scopus)

Abstract

The formation and deposition of solids in plant equipment is a perennial risk to the cryogenic processing of natural gas. While several tools are available to predict the temperatures at which heavy hydrocarbon solids (HHC) will form, the accuracy of the gas mixture's compositional characterization can significantly impact the reliability of those predictions. A complete characterization of the mixture is the most desirable scenario but is challenging and expensive to obtain. More typically, C-6 hydrocarbons and heavier compounds are lumped into pseudocomponents based on their boiling point to represent the HHC composition of the mixture. Recently, Miethe et al. (Hydrocarb Process, 2015) detailed a new approach based on splitting each pseudocomponent further according to its paraffinic, isoparaffinic, naphthenic and aromatic (PINA) composition. An associated defined component is used to represent each of these sub-fractions to improve the melting temperature prediction accuracy. However, this "Lump + PINA(API) " approach has not been validated for mixtures that contain HHCs beyond C-10. This work compares freeze-out predictions based on a complete compositional characterization of a gas mixture with HHCs up to C-14 with results obtained using (1) the new Lump + PINA(API) approach and (2) several other freeze-out prediction methods described in the literature. For two gas samples, the fully characterized mixtures were predicted to have melting temperatures of 263.2 K (14.1 & DEG;F) and 260.1 K (8.5 & DEG;F), respectively. At the same time, the Lump + PINA(API) predictions were 153.4 K (- 183.6 & DEG;F) and 157.4 K (- 176.3 & DEG;F), respectively. The large discrepancy between melting temperature predictions highlights the need to either (1) obtain full characterization of natural gas mixture compositions where possible, and/or (2) to develop an improved set of correlations pseudocomponent correlations for predicting freeze-out in natural gas mixtures.

Original languageEnglish
Article number25
Number of pages19
JournalInternational Journal of Thermophysics
Volume44
Issue number2
DOIs
Publication statusPublished - Feb 2023

Fingerprint

Dive into the research topics of 'Performance Tests of Gas Characterization Methods for Predicting Freeze-out in LNG Production'. Together they form a unique fingerprint.

Cite this