Elsevier

Journal of Membrane Science

Volume 539, 1 October 2017, Pages 432-440
Journal of Membrane Science

The impact of ethylene glycol and hydrogen sulphide on the performance of cellulose triacetate membranes in natural gas sweetening

https://doi.org/10.1016/j.memsci.2017.06.023Get rights and content

Highlights

  • Glycol carryover from natural gas dehydration can impact downstream CO2 removal.

  • Ethylene and triethylene glycol plasticise cellulose triacetate membranes used for CO2 removal.

  • The permeability of methane increases dramatically and selectivity falls.

  • A methanol wash can be used to reverse most, but not all, of these effects.

  • H2S at partial pressures of up to 0.75 kPa has no impact on cellulose triacetate performance.

Abstract

In natural gas sweetening, gas dehydration with glycols is typically carried out upstream of membrane separation of carbon dioxide. This means that when process upsets occur, these glycols can reach the membrane unit. In this work, we study the impact of two common glycols (monoethylene glycol and triethylene glycol) on the gas transport performance of cellulose triacetate membranes. We find that the glycol absorbed into the membrane initially obstructs the permeation of CH4 and CO2, due to pore filling or antiplasticisation effects, but the permeability then increases again, indicative of polymer relaxation and a loss of crystallinity in the polymer. The smaller helium molecule is significantly less affected by the presence of the glycols, possibly because its lower solubility within glycol limits its movement through the swollen structure. However, after removing the glycols with a methanol wash, the membrane performance recovers with only a slight residual plasticisation observed. In addition, the permeation of H2S, a common contaminant within natural gas streams, was studied across a range of temperatures. At the partial pressures studied (up to 0.75 kPa), H2S had very little effect on the membrane performance even in long-term exposure for up to 300 days.

Introduction

Natural gas is a primary energy resource that will occupy over 25% of the global electricity market in the next decades, as well as acting as a transport fuel and direct heating resource [1]. The composition of raw natural gas varies widely but typically contains impurities such as nitrogen (N2), carbon dioxide (CO2), water (H2O) and hydrogen sulphide (H2S) that require removal to meet pipeline specifications. Membrane separation has been used for many decades for acid gas removal, known as natural gas sweetening, with advantages in energy efficiency, land footprint and a lack of chemical consumption [2]. Although many new membrane materials have been developed, cellulose triacetate (CTA) membranes still retain the bulk of this separation market because of their high CO2 – methane (CH4) selectivity, commercial readiness and acceptance as a low risk option by the industry [3], [4].

Raw natural gas is usually saturated with water which is generally removed upstream of the membrane unit to avoid pipeline corrosion and hydrate formation [5], [6], [7]. Glycols such as monoethylene glycol (MEG), diethylene glycol (DEG) and triethylene glycol (TEG) are the most common solvents utilised for this purpose [8], [9]. Due to the extremely low vapour pressure of glycols (28 Pa for monoethylene glycol [10] and < 1 Pa for triethylene glycol [11] at 35 °C), carryover of these solvents in the vapour state is usually limited. However, carryover of entrained glycol droplets can occur [7]. This is a significant issue, because glycol is known to plasticise polymers [12], and the entrainment of the glycol solution into the membrane unit can thus alter the permselectivity of the membrane [13]. A study on the effect of MEG and TEG vapours on CO2/CH4 separation across a facilitated transport membrane has been reported [14]. However, to the best knowledge of the authors, there is no study on the effect of glycol solutions on the gas separation performance of CTA membranes.

Hydrogen sulphide is a common species in natural gas with concentration varying from 4 to 10,000 ppm [15] that will enter the membrane unit with the natural gas. Many studies on CO2 removal from natural gas by cellulose acetate membranes have observed the co-permeation of H2S with CO2 in the membrane unit [15], [16], [17]. Li et al. reports the performance of cellulose acetate membranes in the presence of H2S – H2O mixtures [18]. However, the impact of temperature on H2S permeation through the CTA membrane has not been well studied. Heilman et al. [19] reported the permeability and sorption of H2S into several polymer films including a cellulose acetate film manufactured by Polaroid but the results presented were limited. Data on the long term effect of H2S on CTA gas separation performance is also limited.

In this investigation, the effect of two standard glycols, MEG and TEG, on the gas separation of CTA membranes over a 2000 h period is investigated. The permeation of H2S through CTA membrane at different partial pressures (0.2–0.75 kPa) and temperatures (22–80 °C) is also reported, as is the long term impact of H2S on the membrane performance over a 7200 h period.

Section snippets

Membrane fabrication

The polymer utilised in the investigation was a commercial cellulose triacetate powder supplied by Cellulose Company – Daicel Corporation, Japan. The degree of substitution of acetyl groups on the polymer is 2.85, corresponding to a degree of acetylation of 61.6%.

Dense CTA membranes were fabricated by a solvent casting method. The polymer powder was dried under vacuum at 100 °C overnight prior to membrane fabrication. The dried powder was then dissolved in dicholoromethane (ChemSupply,

Sorption uptake of glycol liquids in cellulose triacetate

The sorption uptake of both MEG and TEG (Fig. 1) shows a sigmoidal shape which indicates that while the glycols initially swell the CTA membrane by Fickian diffusion, polymer relaxation or plasticisation occurs as the solute concentration increases [27], [28], [29], [30]. The transition from Fickian diffusion to non-Fickian polymer relaxation occurs at roughly 30 h of exposure of these 70 µm membranes to the solution. It should be noted that the transition time would be of the order of seconds

Conclusions

The impact of glycols on dense CTA membranes has been studied for up to 2000 h. The absorption of ethylene glycol and triethylene glycol into the CTA membrane enhanced the permeation of CH4 and CO2 through the wet membranes but resulted in a gradual decline in the permeation of He. This reflected relaxation of the polymer structure. In particular, WAXD analysis confirmed that a significant loss of crystallinity occurred during exposure to these glycols, providing more accessible free volume in

Acknowledgements

The authors would like to acknowledge the funding support for this research project from The University of Melbourne, Particulate and Fluid Processing Centre (PFPC), the Peter Cook Centre for Carbon Capture and Storage Research at the University of Melbourne and Brown Coal Innovation Australia (BCIA). The X-ray diffraction analysis was performed within the Materials Characterisation and Fabrication Platform (MCFP) at the University of Melbourne and the Victorian Node of the Australian National

References (70)

  • J. Wijmans et al.

    The solution-diffusion model: a review

    J. Membr. Sci.

    (1995)
  • L.M. Robeson

    Contributions of diffusion and solubility selectivity to the upper bound analysis for glassy gas separation membranes

    J. Membr. Sci.

    (2014)
  • J.S. Yang

    Analysis of pervaporation of methanol-MTBE mixtures through cellulose acetate and cellulose triacetate membranes

    Polymer

    (1998)
  • M.D. Donohue et al.

    Permeation behavior of carbon dioxide-methane mixtures in cellulose acetate membranes

    J. Membr. Sci.

    (1989)
  • A.M. Das et al.

    Synthesis and characterization of cellulose acetate from rice husk: eco-friendly condition

    Carbohydr. Polym.

    (2014)
  • J. Yang

    Application of cellulose acetate to the selective adsorption and recovery of Au(III)

    Carbohydr. Polym.

    (2014)
  • A.C. Puleo et al.

    The effect of degree of acetylation on gas sorption and transport behavior in cellulose acetate

    J. Membr. Sci.

    (1989)
  • B.D. Bhide et al.

    Membrane processes for the removal of acid gases from natural gas. II. Effects of operating conditions, economic parameters, and membrane properties

    J. Membr. Sci.

    (1993)
  • W.-H. Lin et al.

    Gas permeability, diffusivity, solubility, and aging characteristics of 6FDA-durene polyimide membranes

    J. Membr. Sci.

    (2001)
  • C.J.T. Landry et al.

    Physical aging of blends of cellulose acetate polymers with dyes and plasticizers

    Polymer

    (2001)
  • C.A. Scholes

    Plasticization of ultra-thin polysulfone membranes by carbon dioxide

    J. Membr. Sci.

    (2010)
  • A.C. Galvão et al.

    Solubility of methane and carbon dioxide in ethylene glycol at pressures up to 14 MPa and temperatures ranging from (303 to 423) K

    J. Chem. Thermodyn.

    (2010)
  • R. Fernández-Prini et al.

    Solubilities of inert gases in ethylene glycol

    J. Chem. Thermodyn.

    (1987)
  • F.Y. Jou

    Vapor-liquid equilibria for acid gases and lower alkanes in triethylene glycol

    Fluid Phase Equilibria

    (1987)
  • G. Chatterjee et al.

    Poly(ether urethane) and poly(ether urethane urea) membranes with high H2S/CH4 selectivity

    J. Membr. Sci.

    (1997)
  • L.M. Robeson

    An empirical correlation of gas permeability and permselectivity in polymers and its theoretical basis

    J. Membr. Sci.

    (2009)
  • M.M. Dal-Cin et al.

    Revisiting the experimental and theoretical upper bounds of light pure gas selectivity-permeability for polymeric membranes

    J. Membr. Sci.

    (2008)
  • W.J. Cole, K.B. Medlock Iii, A. Jani, A view to the future of natural gas and electricity: an integrated modeling...
  • R.W. Baker

    Membrane Technology and Applications

    (2004)
  • J.M.S. Henis et al.

    A novel approach to gas separation using composite hollow fiber membranes

    Sep. Sci. Technol.

    (1980)
  • T. Løkken, et al., Water content of high pressure natural gas: data, prediction and experience from field, in:...
  • A.L. Kohl, R.B. Nielsen, Chapter 11 – absorption of water vapor by dehydrating solutions, in: Gas Purification (Fifth...
  • K.A. Lokhandwala et al.

    Sour Gas Treatment Process

    (1995)
  • G. Gavlin, B. Goltsin, Gas Dehydration Process, Gas Research Institute, Chicago. Illinois,...
  • MEGlobal, Ethylene Glycol – Product Guide, 2008 [cited 9 Oct 2014]. Available from:...
  • Cited by (26)

    • Bio-based nonporous membranes: Evolution and benchmarking review

      2023, Journal of Industrial and Engineering Chemistry
    • The influence of propane and n-butane on the structure and separation performance of cellulose acetate membranes

      2021, Journal of Membrane Science
      Citation Excerpt :

      This agrees well with the sorption results in Fig. 1. As a comparison, Lu et al. reported that the CTA membrane density decreased from 1.299 g/cm3 to 1.270 g/cm3 after the membrane was soaked in triethylene glycol [20]. XRD analysis was used to investigate the influence of propane or n-butane on the polymer crystallinity.

    View all citing articles on Scopus
    View full text