Influence of organic solvents and operating conditions on the performance of polyphenylsulfone (PPSU)/copper-1,3,5-benzenetricarboxylate (Cu-BTC) solvent resistant nanofiltration (SRNF) membranes

https://doi.org/10.1016/j.cherd.2016.09.021Get rights and content

Highlights

  • Separation performance of SRNF membranes could be affected by solvent pretreatment.

  • Membrane showed high separation rate against dye at low concentration of solution.

  • Membrane flux was improved at high pressure, but permeate quality was affected.

  • Incorporation of nanofillers could improve flux and rejection of SRNF membrane.

Abstract

This article reports the influence of organic solvents (methanol, ethanol, isopropanol, acetonitrile, ethyl acetate, n-hexane and n-heptane) on the filtration performance of polyphenylsulfone (PPSU) membrane incorporated with and without copper-1,3,5-benzenetricarboxylate (Cu-BTC) nanoparticles. The pure methanol flux of the membranes and their rejection against Reactive Orange 16 (RO16, Mw = 616 g/mol) were studied and compared before and after 48 h pretreatment process using seven different kinds of solvents. The results showed that the solvent pretreatment could alter the hydrophilicity/hydrophobicity of membrane surfaces, leading to different methanol flux and dye rejection. The variation in membrane performance is mainly due to the reorganization of polymeric chains upon contact with organic solvent. On the other hand, the membrane methanol flux and dye rejection was found to vary by increasing dye concentration and operating pressure. Results showed that increasing dye concentration generally affected dye rejection rate without significantly affecting methanol flux while higher operating pressure tended to improve methanol flux by producing permeate of lower quality. In all cases, the PPSU membrane incorporated with 0.8 wt% Cu-BTC nanoparticles (PPSU/0.8Cu-BTC) showed greater separation performance than that of pristine PPSU membrane. This newly prepared composite membrane also exhibited acceptable durability and performance stability when it was tested for 24 h using dye/methanol solution.

Introduction

The performances of nanofiltration (NF) membranes have been widely evaluated in the area of water purification and other water-related treatments since the 1980s (Hilal et al., 2004, Misdan et al., 2012, Ong et al., 2012, Lau et al., 2013, Lau et al., 2015a, Lau et al., 2015b). At the beginning of this century, the application of water-based NF has been broadened to solvent filtration and purification area. This relatively new technology so-called solvent resistant nanofiltration (SRNF) or organic solvent nanofiltration (OSN) holds enormous potential as it allows separation of small compounds with molecular weight (Mw) ranging from 200 to 1000 g/mol from organic solvents (Vandezande et al., 2008, Cheng et al., 2014, Sani et al., 2014a, Sani et al., 2014b). Since then, an increasing number of potential applications have been reported. These include recovery of solvents from dewaxed lube oil filtrates, organometallic complexes recovery from various organic solvents, separation of phase transfer catalyst from toluene, deacidification of vegetable oils and concentration of active pharmaceutical ingredients (Raman et al., 1996, Subramanian et al., 1998, White and Nitsch, 2000, Luthra et al., 2002, Scarpello et al., 2002, Sheth et al., 2003, Geens et al., 2007, Tylkowski et al., 2011).

Most commonly used membranes for SRNF applications are integrally skinned asymmetric membranes and composite membranes where a top layer of a different material is adhered to a porous support. A number of methods such as interfacial polymerization, slow-fast phase separation and layer-by-layer assembly have been employed to develop high performance and chemically stable SRNF membranes (Joseph et al., 2014, Xiang et al., 2014, Zhou et al., 2014, Sun et al., 2015a). These developed membranes in general exhibit promising solvent stability and separation characteristics, depending on the types of organic solvents used. Currently, the advent of inorganic-organic hybrid membrane which combines the processability of polymers and the superior properties of inorganic materials has captured the attention of researchers, owing to the unique advantages of this novel membrane in comparison to the conventionally-made polymeric membrane. Such composite materials known as mixed matrix membranes (MMMs), which were originally developed for gas separation processes, have also been explored in SRNF field.

Previous works have shown that the polymeric-based membranes incorporated with inorganic materials could potentially improve membrane stability and compaction without affecting their separation characteristics (Gevers et al., 2005, Campbell et al., 2014, Siddique et al., 2014, Sun et al., 2015b). In this study, copper-1,3,5-benzenetricarboxylate (herein referred to as Cu-BTC) was selected as inorganic filler due to its better affinity with polymeric phase compared to other fillers such as zeolites, carbon and titanium dioxide. Therefore, formation of non-selective voids during membrane preparation can be avoided (Sorribas et al., 2013). Furthermore, Cu-BTC contains nanoscale pore size of around 0.9 nm in diameter, making it suitable to transport most solvents used in SRNF whilst capable of rejecting solute of bigger size (Küsgens et al., 2009, Li et al., 2009).

In aqueous-based NF system, water is the main solvent and common to all applications. The interactions between solute and water however are less complicated in comparison to non-aqueous NF system. Very often, SRNF shows an unpredictable performance as a result of complex interaction between solvents and membranes. To address the issues, studies have been carried out to understand the interactions by investigating the effects of solvent properties on the SRNF membrane performances.

Generally, the physical properties and separation performance of the membranes are studied by pretreating the membranes with different types of solvents for certain periods of time. For example, Van der Bruggen et al. (2002) investigated the pure water flux and maltose rejection of hydrophilic and hydrophobic commercial NF membranes (N30F, NF-PES-10, MPF 44, and MPF 50) before and after 10-day pretreatment with different types of organic solvents such as methylene chloride, acetone, hexane, ethyl acetate and ethanol. The researchers found that the surface hydrophilicity/hydrophobicity of a membrane has strong effects on membrane performance, i.e. hydrophilic membranes tended to have lower rejection rate while hydrophobic membrane to have greater water flux after solvent exposure. This behavior is likely caused by the reorganization of the membrane structure resulted from clustering of hydrophobic and hydrophilic zones within the active layer. Jansen et al. (2013) on the other hand, used in-house made PPSU/polyimide membranes to evaluate the membrane performance with respect to pure methanol flux and dye (Sudan II, Mw: 276.3 g/mol) rejection before and after 10-day solvent treatment with acetone and methyl ethyl ketone. After the pretreatment process, it was reported that the hydrophobicity of the membrane was affected following a decrease in surface contact angle which in turn increased significantly membrane methanol flux. They attributed the flux improvement to the pore wetting mechanism. It is because the solvent molecules are able to enter and wet the membrane pores during pretreatment process, causing the pore walls to become slightly hydrophilic. Darvishmanesh et al. (2011) evaluated the effect of the membrane pretreatment against several types of solvents on methanol flux and Rose Bengal (Mw: 973.7 g/mol) rejection. They found that the degree of membrane swelling and its plasticization varied depending on type of solvent used. Therefore, the selection of solvent for membrane pretreatment process has important role in the separation performance of an SRNF membrane. Previously, Van der Bruggen et al. (2002), Darvishmanesh et al. (2011) and Jansen et al. (2013) have studied how the changes in solvent pretreatment condition could affect the intrinsic properties of same membrane and further its performance during water and/or organic solvent filtration process.

In addition to the solvent pretreatment, the variation in operating condition during separation process could also play a role in affecting the performance of SRNF membranes. Some of the important operating parameters that are generally known to affect membrane solvent permeation and rejection are pressure, temperature, solute concentration and type of solvent and solute involved. Scarpello et al. (2002) and Whu et al. (2000) reported that increasing operating pressure can be beneficial to improve both solvent flux and solute rejection. They attributed the performance improvement to the partial reversible in which increasing compression on the active layer of asymmetric membrane would result in tightening or sealing of the pores (cylinder pore-based permeability) and further improve separation performance. However, they explained that membrane fouling might become more severe at high operating pressures owing to greater permeate volume produced. With respect to solute concentration, SRNF membrane behaves very similar as commonly used NF membrane, i.e. increasing solute concentration tends to negatively affect solvent flux and vice versa (He et al., 2008). This reduced solvent flux and better rejection is mainly due to increased osmotic pressure and pore fouling as a result of increasing solute concentration (Silva and Livingston, 2006).

In view of the importance of membrane solvent pretreatment and operating conditions on the overall SRNF membrane performance, the main objective of this work is to study the influence of seven different solvents, i.e. methanol, ethanol, isopropanol, acetonitrile, ethyl acetate, n-hexane and n-heptane on the properties and separation performance of two types of self-fabricated SRNF membranes, i.e. polyphenylsulfone (PPSU) and PPSU incorporated with 0.8 wt% of Cu-BTC in methanol/dye solutions. These solvents were chosen in order to cover a wide range of different chemical structures: an alcohol, an ester, an aliphatic hydrocarbon and a nitrile. Changes in the membrane performances, either in the methanol flux or in the dye rejection, indicate the effect of the organic solvents on the studied membranes. With respect to process conditions, the effects of operating pressure and solute concentration on membrane performance were studied.

Section snippets

Materials

PPSU (Radel R-5000 NT) with specific gravity of 1.29 was procured from Solvay Advanced Polymers, United States. N-Methyl-2-pyrrolidinone (purity >99.5%) purchased from Merck, Malaysia was used as solvent without further purification. Methanol, ethanol, isopropanol, acetonitrile, ethyl acetate, n-hexane and n-heptane supplied from Merck, Malaysia were used in the membrane pretreatment process and their most important properties are shown in Table 1. Copper nitrate trihydrate (Cu(NO3)2·3H2O),

Characteristics of PPSU and PPSU/0.8Cu-BTC membranes

In our previous work, we have studied the impacts of nanofiller loadings (0.5, 0.8, 1.0 and 3.0 wt% Cu-BTC) on the physicochemical and separation characteristics of SRNF membranes made of PPSU (Sani et al., 2015). The present work will further investigate the influence of solvent pretreatment and process conditions on the properties and performance of two selected membranes from our earlier work, i.e. control PPSU membrane and PPSU membrane incorporated with optimized Cu-BTC loading

Conclusions

The pretreatment of PPSU and PPSU/0.8Cu-BTC membrane in organic solvents has a significant effect on solvent permeation and dye separation. These demonstrated that the membranes have strong interactions with different solvents, resulting in the changes of membrane hydrophilicity/hydrophobicity and filtration properties. The structure of the membrane might change due to the impact of organic solvents on the membrane, resulting in variation of membrane morphology and further its rejection

Acknowledgements

The authors are grateful for research financial support by the Ministry of Higher Education under Long-term Research Grant Scheme (Grant no. 4L803) and AMTEC-HICOE project (Grant no. R.J090301.7846.4J175). Author, N.A.A. Sani thanks the sponsorship given by the Ministry of Higher Education under MyBrain15 (MyPhD) scheme during her PhD. studies.

References (45)

  • R.W. Lencki et al.

    Effect of nonaqueous solvents on the flux behavior of ultrafiltration membranes

    J. Membr. Sci.

    (1995)
  • Z.-Q. Li et al.

    Ultrasonic synthesis of the microporous metal–organic framework Cu3(BTC)2 at ambient temperature and pressure: an efficient and environmentally friendly method

    Mater. Lett.

    (2009)
  • S.S. Luthra et al.

    Homogeneous phase transfer catalyst recovery and re-use using solvent resistant membranes

    J. Membr. Sci.

    (2002)
  • N. Misdan et al.

    Seawater Reverse Osmosis (SWRO) desalination by thin-film composite membrane—current development, challenges and future prospects

    Desalination

    (2012)
  • L.G. Peeva et al.

    Effect of concentration polarisation and osmotic pressure on flux in organic solvent nanofiltration

    J. Membr. Sci.

    (2004)
  • A.R. Roudman et al.

    Surface energy of experimental and commercial nanofiltration membranes: effects of wetting and natural organic matter fouling

    J. Membr. Sci.

    (2000)
  • J.T. Scarpello et al.

    The separation of homogeneous organometallic catalysts using solvent resistant nanofiltration

    J. Membr. Sci.

    (2002)
  • J.P. Sheth et al.

    Nanofiltration-based diafiltration process for solvent exchange in pharmaceutical manufacturing

    J. Membr. Sci.

    (2003)
  • H. Siddique et al.

    Mixed matrix membranes for organic solvent nanofiltration

    J. Membr. Sci.

    (2014)
  • P. Silva et al.

    Effect of solute concentration and mass transfer limitations on transport in organic solvent nanofiltration—partially rejected solute

    J. Membr. Sci.

    (2006)
  • I. Soroko et al.

    Impact of TiO2 nanoparticles on morphology and performance of crosslinked polyimide organic solvent nanofiltration (OSN) membranes

    J. Membr. Sci.

    (2009)
  • R. Subramanian et al.

    Processing of vegetable oils using polymeric composite membranes

    J. Food Eng.

    (1998)
  • Cited by (18)

    • High performance in-situ tuned self-doped polyaniline (PANI) membranes for organic solvent (nano)filtration

      2022, Polymer
      Citation Excerpt :

      The use of dye, for instance, might not reflect the actual MWCO by size exclusion mechanism due to the potential involvement of other mechanisms such as adsorption and Donnan Exclusion [91]. Nonetheless, dyes of different MW are commonly used as a marker to estimate the MWCO of the OSN membranes as reported elsewhere [13,71,92,93]. At this work, membrane MWCO was measured using wide range of dyes with molecular weight ranging from 327 to 1470 g mol−1.

    • Internal arcing and lightning strike damage in short carbon fiber reinforced thermoplastic composites

      2021, Composites Science and Technology
      Citation Excerpt :

      PPSU is also a high-temperature thermoplastic that has superior impact and chemical resistance [27]. PPSU is a member of the polysulfone family and with a chemical structure comprised of sulfone moieties, ether linkage, and a biphyenyl group, as shown in Fig. 2(b) [28]. PPSU is an amorphous material with a heat deflection temperature of 207 °C, making it suitable for electrical circuitry in aircraft interiors.

    View all citing articles on Scopus
    View full text