Skip to main navigation Skip to search Skip to main content

Natural Gas Sweetening By Membrane Cascade Using a Standardized Model

  • Lauren Ward
  • , Michael Sees
  • , Sheima Khatib
  • , Chau-Chyun Chen

Research output: Contribution to conferenceAbstractpeer-review

Abstract

Natural gas prices are at a low due to high availability which has driven several companies to consider using C1 chemistry to generate petroleum products and petrochemicals, particularly from stranded gas reserves that are typically flared. [1] Processes which aim to take advantage of these resources must be modular, but have similar requirements to centralized processing plants in terms of required feedstock and product purities. Membrane cascades, which use a series of interconnected, permi-selective membrane modules to separate fluid mixtures, have been proven for removing contaminants from natural gas, mainly CO2. [2] However, most commercial process simulators lack the models needed to simulate gas separation membranes. As part of the RAPID Center for Process Modeling’s effort to provide models and recommendations to the engineering community, a “membrane separation toolkit” has been developed to standardize simple membrane modeling. A test case involving the purification of 25 MMSCFD of CH4 laden with CO2 as the primary impurity using a standard cellulose acetate membrane was developed to demonstrate the use of the toolkit. The cascade was optimized with respect to the stage operating pressures, temperatures, and cuts to produce pipeline quality natural gas containing ≤ 3 % CO2 by minimizing OPEX and maximizing the produced volume of salable gas.
Original languageUndefined/Unknown
StatePublished - 2019
Externally publishedYes
Event2019 AIChE Annual Meeting - Orlando, United States
Duration: Nov 10 2019Nov 15 2019

Conference

Conference2019 AIChE Annual Meeting
Country/TerritoryUnited States
CityOrlando
Period11/10/1911/15/19

Fingerprint

Dive into the research topics of 'Natural Gas Sweetening By Membrane Cascade Using a Standardized Model'. Together they form a unique fingerprint.

Cite this