Elsevier

Journal of Membrane Science

Volume 566, 15 November 2018, Pages 336-345
Journal of Membrane Science

Ultrathin gutter layer for high-performance thin-film composite membranes for CO2 separation

https://doi.org/10.1016/j.memsci.2018.09.017Get rights and content

Highlights

  • A gutter layer with over sixfold enhanced CO2 permeance was prepared.

  • Inductively coupled plasma was used for hydrophilic treatment with minimized damage.

  • Ultrathin (< 100 nm) selective layer was successfully coated on the gutter layer.

  • The as-prepared membrane exhibited 1455 GPU of CO2 permeance with CO2/N2 68.

  • Long-term stability was confirmed upon 10,000 min operation.

Abstract

Ultrathin film composite (TFC) membranes with selective layers less than 100 nm thick are highly desired to maximize the permeance of gas separation membranes for high energy efficiency. For membranes with ultrathin selective layers, a gutter layer is usually required to prevent pore penetration in the selective layers. Also, according to a recent model, the introduction of a gutter layer strongly improves TFC membrane performance by increasing the overall membrane performance up to an order of magnitude. Unfortunately, this improvement comes with an undesired decrease in selectivity unless the gutter layer is properly designed. This study found that the gutter layer permeability should be five- to tenfold that of the selective layer to minimize the decrease in selectivity. However, the most commonly used material for the gutter layer, polydimethylsiloxane, does not meet the requirements for high-performance membrane materials. Thus, here we report a gutter material with CO2 permeance sixfold that of polydimethylsiloxane, and prepare a TFC membrane with ultrathin gutter (75 nm) and selective layers (70 nm) for CO2/N2 separation. This work represents a guideline for developing next-generation TFC membranes and provides a comprehensive understanding of the impact of the gutter layer.

Introduction

Research on membrane science and technology is being actively pursued in both academic and industrial contexts because of a broad range of potential membrane applications for simple, energy-efficient chemical separation [1], [2], [3]. For high energy efficiency in separation, a multilayer composite membrane is usually used, consisting of an ultrathin selective layer mechanically supported by a microporous substrate. High energy efficiency requires a high-permeance gas membrane and minimum thickness of the selective layer, of which have been continual challenges for membrane technologies. These challenges arise because of the low concentration solution needed for the thin selective layers, which usually causes pore penetration in a porous substrate, thus resulting in surface defects.

Many thin-film composite membranes use a gutter layer to prevent such pore penetration [4], [5], [6], [7], [8], [9]. Gutter layer materials must have high permeability to minimize transport resistance. Therefore, high-permeability materials such as poly(1-trimethylsilyl-1-propyne) and polydimethylsiloxane (PDMS) are usually preferred for gutter layer materials, but because of the rapid physical aging of the former, the latter has been the most widely used gutter layer material so far [10].

Although PDMS is widely used in gas-separation membranes, there is still significant room to enhance ultrathin film composite (TFC) membranes just by increasing the gas transport properties of the gutter layer. Teflon AF, especially AF2400, is more attractive as a gutter layer material than PDMS because it offers higher gas permeance, no additional crosslinking, harmless dissolution, a high tendency to form uniform thin films, etc. For these reasons, the gas transport properties of AF2400 have been widely studied for membrane applications [11], [12], [13], [14]. However, the superhydrophobic nature of AF2400 and its low wettability by most solvents have conspired to hinder the use of a selective layer on top of an AF2400 gutter layer, so that, despite the significant potential of AF2400, no reports exist of its use as gutter layer material.

Although oxygen-plasma treatment is an ideal non-chemical method to overcome the superhydrophobicity of AF2400, the ultrathinness (< 100 nm) of the AF2400 gutter layer makes it very susceptible to plasma-induced damage. Minimizing such plasma damage is essential to apply AF2400 as a gutter layer material in TFC membranes. Capacitively coupled plasma (CCP) sources can generate plasma by generating a strong static electric field between the electrode and the ground. This type of plasma delivers high energy via ion bombardment, which damages membrane surfaces, especially those of thin membranes. In general, ion bombardment damage decreases with increasing pressure, so dielectric barrier discharge plasma has become popular owing to its less severe damage under atmospheric pressure. However, the plasma and the membrane must be placed sufficiently close together to allow proper plasma treatment, which inevitably increases the ion-bombardment damage [15]. Contrastingly, unlike CCP sources, inductively coupled plasma (ICP) sources generate plasma via an electric field induced by a time-varying magnetic field. ICP is characterized by weaker ion bombardment and higher plasma density than CCP [16].

We used ICP to minimize the damage of plasma treatment while increasing its impact, producing appropriate gutter layer surface properties without damage, even for gutter layers thinner than 100 nm. By using this method, we fabricated ultrathin film composite CO2/N2 separation membranes thinner than 100 nm for use as both gutter layer and selective layer, thereby attaining high-performance TFC membranes.

This study compares the gas permeances of PDMS and AF2400 membranes as gutter layers and discusses how the gas transport properties of the gutter layer affect the TFC membrane permeance and selectivity as functions of gutter layer permeance and selective layer thickness. We used the floating harmonic method [17], [18] to measure the electron temperature and ion flux, which are two parameters related to membrane damage. The AF2400 membrane fabricated for use as a gutter layer was tested under mixed gas conditions, and the experimental gas transport properties were compared with calculations based on a series resistance model.

Section snippets

Materials

Teflon AF2400 was procured from DuPont (USA), and Sylgard184 base and curing agent were procured from Dow Corning (USA). Galden HT110 as a solvent for AF2400 was procured from Solvay (Brussels, Belgium), and n-hexane (95%) as a solvent for Sylgard184 was procured from Daejung Chemicals (Gyeonggi-do, South Korea). The PAN350 membrane for the substrate was procured from Nanostone (Waltham, USA). Mixed gas (CO2 14 mol%, N2 86 mol%) was procured from AirKorea (Gyeonggi-do, South Korea). PEA-GO for

Gutter layer function in the thin-film composite membrane

Membranes used industrially usually consist of multilayer composite structures, as shown in Fig. 2. To obtain high efficiency, the selective layer must be thin (< 1 µm), so this layer is usually too fragile for direct use. Accordingly, a porous substrate with the requisite mechanical strength is used to support the membrane. The mass transport resistance of the porous substrate should be minimized so as to maximize the pore size and porosity of the substrate. However, such large pores allow

Conclusions

A gutter layer is usually used in TFC membranes with ultrathin selective layers, and PDMS has been the most widely used material for gutter layers. However, the theory developed herein shows that a PDMS gutter layer is not the best option in terms of TFC membrane performance due to its low permeance and relatively complex preparation process. Teflon is much more attractive as gutter layer material than PDMS because it offers up to sixfold the gas permeance of PDMS and exhibits easy

Acknowledgements

This work was supported by the Korea CCS R&D Center (Korea CCS 2020 Project) grant funded by the Korea government (Ministry of Science and ICT) in 2016 (2014M1A8A1049307).

References (48)

  • J.G. Wijmans et al.

    The solution-diffusion model – a review

    J. Membr. Sci.

    (1995)
  • M.A. Islam et al.

    Non-equilibrium surface reactions – a factor in determining steady state diffusion flux

    J. Membr. Sci.

    (2002)
  • W.S. Cho et al.

    Thickness dependence of incremental permeability in Mumetal(t)/CoCr bilayers

    J. Magn. Magn. Mater.

    (2002)
  • S.J. Song et al.

    Thickness dependence of hydrogen permeability for Ni-BaCe0.8Y0.2O3-delta

    Solid State Ion.

    (2008)
  • G. Firpo et al.

    Permeability thickness dependence of polydimethylsiloxane (PDMS) membranes

    J. Membr. Sci.

    (2015)
  • C.Z. Liang et al.

    High-performance composite hollow fiber membrane for flue gas and air separations

    J. Membr. Sci.

    (2017)
  • Y. Huang et al.

    Pressure ratio and its impact on membrane gas separation processes

    J. Membr. Sci.

    (2014)
  • R.P. Lively et al.

    From water to organics in membrane separations

    Nat. Mater.

    (2017)
  • P. Bernardo et al.

    Membrane gas separation: a review/state of the art

    Ind. Eng. Chem. Res.

    (2009)
  • Q. Fu et al.

    Highly permeable membrane materials for CO2 capture

    J. Mater. Chem. A

    (2013)
  • A. Halim et al.

    Soft polymeric nanoparticle additives for next generation gas separation membranes

    J. Mater. Chem. A

    (2014)
  • Q. Fu et al.

    The effect of soft nanoparticle morphologies on thin film composite membrane performance

    J. Mater. Chem. A

    (2014)
  • P.Y. Li et al.

    High-performance multi layer composite membranes with mussel-inspired polydopamine as a versatile molecular bridge for CO2 separation

    ACS Appl. Mater. Interfaces

    (2015)
  • C.Z. Liang et al.

    Ultrahigh flux composite hollow fiber membrane via highly crosslinked PDMS for recovery of hydrocarbons: propane and propene

    Macromol. Rapid Commun.

    (2018)
  • Cited by (47)

    View all citing articles on Scopus
    1

    These authors contributed equally to this work.

    View full text