Polydopamine-modified halloysite nanotube-incorporated polyvinyl alcohol membrane for pervaporation of water-isopropanol mixture

https://doi.org/10.1016/j.jiec.2021.09.016Get rights and content

Highlights

  • Hydrophilic surface modification of HNT performed bydopamine.

  • TGA, FE-SEM, TEM and XPS analysis conducted to confirmed the PDA modified HNT surfaces.

  • 5wt.% PDA/HNT incorporated PVA/PVAm membrane showed optimized pervaporation output.

  • PDA modified HNTs show effective filler for separation of alcohols in pervaporation.

Abstract

In reference, the mussel-encouraged surface adhesion chemistry, dopamine oxidative auto-polymerization on the surface of halloysite nanotubes (HNTs) was performed to confer hydrophilicity to the HNTs. The additional hydrophilic moieties with surface coating by polydopamine (PDA) was assured with transmission electron microscopy (TEM), thermogravimetric analysis (TGA), and X-ray photoelectron spectroscopy (XPS) techniques. A PDA/HNT-incorporated polyvinyl alcohol/polyvinyl amine (PVA-PVAm) membrane was fabricated for the pervaporation-dehydration of isopropanol/water (IPA/water), and the filler dispersion, crystallinity, and hydrophilicity of the membrane were confirmed by field-emission scanning electron microscopy (FE-SEM), X-ray diffraction (XRD), and contact angle measurements, respectively. The pervaporation test was performed with membranes having different PDA/HNT contents, where the flux increased from 0.19 to 0.39 kg/m2h, and the separation factor decreased from 479 to 63 with the 80/20 (w/w, IPA/water) feed mixture at 40 °C. Compared to the unmodified HNT-incorporated membrane, the membrane with 5 wt.% PDA/HNT exhibited superior performance. Upon increasing the feed (85/15 IPA/water w/w) temperature from 40 to 70 °C, the flux of water and IPA increased from 0.062 to 0.13 kg/m2h and 0.000063 to 0.00144 kg/m2h, respectively. The apparent energy of activation calculated using the Arrhenius equation was positive for both IPA and water. The higher energy required for permeation of IPA (99.65 kJ/mol) compared to water (22.69 kJ/mol) is attributed to facile permeation of water compared to IPA through the hydrophilic channel created by PDA-HNT. In long-term operation (120 h), the IPA and water flux of the 5 wt.% PDA/HNT-incorporated PVA-PVAm membrane was maintained using a feed composition of 80/20 (w/w %) IPA/water at 40 °C, indicating remarkable reusability of the membrane.

Introduction

Most energy-intensive industrial chemical processes have two main drawbacks: consuming a large amount of energy and global warming. Therefore, it is essential to explore environmentally friendly and energy-efficient processes for chemical industries. Separation in industries accounts for 60–80% of costs in most chemical processes [1]. Traditionally, various alcohols and water have been separated by distillation, gas stripping, adsorption on molecular sieves, or azeotropic distillation. However, these processes often require large installation areas and intensive energy consumption [2].

Additionally, it is costly to purify alcohols, such as isopropyl alcohol (IPA), by distillation only because alcohols-water form an azeotropic mixture (87.7 wt.% IPA composition, IPA/water mixture exhibit constant boiling at 80.4 °C). In contrast, pervaporation relies on a membrane-based liquid-liquid separation method consisting of permeation plus evaporation that utilizes either inorganic or polymeric or polymeric-inorganic combined nonporous membranes [3], [4]. Furthermore, it is an economical and eco-friendly process since separation components do not need to evaporate, which leads to pollution and spends tremendous energy in the conventional distillation process. It has been reported that energy consumption can be reduced by up to 87% by replacing distillation with the pervaporation technique [5]. In addition, cross-contamination due to the addition of an entrainer during distillation can be effectively avoided.

Generally, polymer materials with abundant hydrophilic functionalities (COOH, OH, or NH2) such as sodium alginate, chitosan, etc. [6], [7] and synthetic polymer polyvinyl alcohol (PVA), have been utilized to fabricate pervaporation membranes. Additionally, with the benefit of hindsight regarding PVA, the German company GFT achieved the industrial application of PVA in the pervaporation field by launching PVA/poly (acrylonitrile) composite membranes [8], [9]. It is well known that untreated PVA membranes exhibit higher membrane swelling due to the formation of defects in the membrane, which leads to diminished separation ability [10]. Chemical crosslinking or heat treatment of the membrane leads to a trade-off between flux and separation. This trade-off can be solved by introducing a filler into the polymer matrix. Recently, there have been several reports on the fabrication of nanocomposites or mixed matrix membrane aim to upgrade the physical, chemical, and morphological cretic of polymeric membranes. To this end, several fillers such as inorganic zeolites, organics such as graphene oxide, and organic–inorganic species such as metal organic frameworks [7], [11], [12], [13] have been explored for pervaporation applications. The use of these fillers results in a hybrid heterogeneous polymer matrix exhibiting the synergetic characteristics of the organic and inorganic phases. In addition, the hybrids may provide novel membrane properties for environmentally friendly and energy-intensive separation processes like pervaporation, and by controlling membrane swelling, the trade-off hurdle between flux and separation can be avoided. In that way, increased efficiency from the hybrid membrane can be useful for reducing energy consumption in distillation and, in fact, it can replace with pervaporation which is a clean and environment-friendly process.

While fabricating mixed-matrix membranes, the most important consideration is the uniform dispersion of the filler amidst the polymer network. Thus, the polymer must have specific functional moieties that can undergo interfacial interactions with the filler (good host) to afford uniform dispersion. The same has been achieved with the above filler materials; however, these materials are not readily available. Moreover, they require high cost and time-consuming manufacturing process for their synthesis; therefore, these fillers are commercially unpopular for membrane applications. On the other hand, a biocompatible aluminosilicate material with a relatively inexpensive, called halloysite, is abundant in nature. The stunning nanotube with a high aspect ratio (ca.20.) type variant has numerous uses in various fields [14], [15], [16].

Halloysite, like kaolinite, has the molecular formula Al2SiO5(OH)4.nH2O and has a multilayer structure with spacing 1 nm in hydrous form and that of 0.7 nm while anhydrous form. The outer surface is protected with siloxane (Si-O-Si) functionalities and Al-OH moieties lining between layers. Due to the intrinsic properties of HNTs, its application has been extended to various filed like gas separation [17], air filtration [18] adsorbents for wastewater treatments [19], and additionally, for improvement of thermal and mechanical strength of polymers such as polybutylene succinate and polyamide HNTs as a filler [20], [21]. Nonetheless, the application of HNTs in pervaporation has to study more since Al-OH lining inside the enclosed layers can provide a smooth passage for the transport of water during the IPA/water pervaporation process. The transportation of molecules via pervaporation takes place from one hydrophilic site to another [22]. However, HNTs with a hydrophobic external surface (due to the Si-O-Si groups) can exhibit resistance to a uniform distribution in hydrophilic polymer matrices such as PVA. It is well known that the dispersion of the filler is strongly controlled by its interaction with the polymer matrix [23].

Therefore, in our previous study, we modified the surface of HNTs by generating hydrophilic functionality (hydroxyl) at the surface via Piranha solution etching [24]. However, because of the strongly acidic environment, even though the hydrophilic functionality could be generated easily, the modified HNTs were fragile and mechanically weak and did not retain their original shape. However, due to the hydrophilicity enhancement in mixed-matrix membrane resulted from acid-treated HNTs, polyvinyl alcohol, and polyvinyl amine showed remarkable pervaporation performance, where the best performance was achieved with the 5 wt.% modified HNT loaded membrane, affording a flux in the range of 0.031–0.13 kg/m2h and selectivity of 427-93313 from 10-20 wt.% of water in feed composition, at 40 °C. Nevertheless, as mentioned above, HNTs have an open tubular multilayer structure and 0.7 to 1.0 nm distance inside layers of hydrophilic aluminum hydroxide. Therefore, molecular sieving during the pervaporation IPA/water molecule separation will be more profound if hydrophilic modification at the surface of HNTs is done with maintaining the actual shape and size of HNTs.

The strong adhesion of marine mussels to wet rocks has inspired many scientists to investigate this interesting phenomenon. In their molecular level investigations, it was found that the catecholic amino acid is an important component of the adhesive protein responsible for the strong adhesion of mussels [25]. Dopamine possesses a similar catechol functionality and therefore undergoes self-oxidative polymerization under weakly alkaline conditions [26]. The pioneering work of Smith and co-workers achieved facile surface modification of various substrates with polydopamine [27]. In addition, recently, several researchers have investigated successful HNTs modification with polydopamine (PDA) through a single-step process for enzyme immobilization [28], [29], [30], wastewater treatments [31], drug delivery [32], catalysis [33], oil-water separation [34]. Moreover, PDA possesses a characteristic hydrophilic functionality as catechol, amine, and quinone carboxylic in both open-chain and cyclized indoles units, etc. In that way, the dispersion ability of HNT can also be improved in polymer matrix by PDA modification.

Therefore, in present study, prompted by the results of the studies as mentioned above, we have successfully attempted to generate a hydrophilic functionality (amine and hydroxyl) at the HNT surface via oxidative dopamine polymerization without compromising the original shape and size of the HNTs for increasing IPA dehydration efficiency of membrane. Since PVA and PVAm have numerous (hydroxyl and amine) functionalities in their polymer chains; thus, the PVA-PVAm, polymer matrix was selected for fabricating the PDA–HNTs-incorporated mixed matrix membrane. The modification of the HNTs with dopamine is confirmed herein using diverse analytical tools, like, XPS, TEM, and TGA. The solid-state morphology and dispersion of the PDA/HNTs-incorporated membrane are evaluated by field-emission scanning electron microscopy (FE-SEM). The pervaporation output is optimized with different contents of PDA/HNTs by varying the feed composition and feed temperature. Based on literature surveys, PDA-modified HNTs incorporated PVA-PVAm membranes for pervaporation water/IPA feed systems have not yet been explored.

Section snippets

Materials

Poly(vinyl alcohol) (88,000–97,000 g/mol) with a hydrolysis degree of 98–99% and glutaraldehyde (GA) (25 wt.%) were obtained from Alfa Aesar, USA. Lupamin-9095 (Mw = 340,000) (polyvinyl amine) was supplied by BASF, Indonesia. Halloysite nanotubes (length: 1–3 µm, diameter: 30–70 nm), dopamine hydrochloride, Trizma® base, and Trizma® hydrochloride buffers were procured from Sigma Aldrich Co. (USA). All other chemicals and IPA (99.5 wt.%) and hydrochloric acid (HCl; 36%) were purchased from

Characterization of PDA-modified HNTs

Fig. 2a to 2d show the FE-SEM and TEM images of the untreated HNTs and PDA-modified HNTs (PDA/HNTs). The FE-SEM micrograph shows that by the supplier’s specifications, the HNTs consist of a hollow tube structure with a length of 500 nm to 1100 nm. The TEM images show that both the untreated and PDA/HNTs have clear open tubular structures. The inner diameter for both the HNTs and PDA/HNTs remained unchanged (10–17 nm); however, the PDA/HNTs have a larger outer diameter than the untreated HNTs

Conclusion

Surface modification of HNTs was successfully achieved by the oxidative self-polymerization of dopamine to obtain a hydrophilic surface. Non-defective PVA-PVAm membranes with PDA/HNTs incorporation (5 wt.%) were developed for the pervaporation of IPA/water mixtures within the azeotropic range. It was concluded that the PV output from the PVA-PVAm membrane was significantly increased by PDA/HNTs incorporation. Among the different PDA/HNTs-incorporated PVA-PVAm membranes, the 5 wt.%

Declaration of Competing Interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: 1) Korea Institute of Energy Technology Evaluation and Planning(KETEP) and the Ministry of Trade, Industry & Energy (MOTIE) of the Republic of Korea (No. 20194010201840) 2) Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea government (MOTIE) (20202020800330, Development and demonstration of energy efficient

Acknowledgements

This work was supported by the 1) Korea Institute of Energy Technology Evaluation and Planning (KETEP) and the Ministry of Trade, Industry & Energy (MOTIE) of the Republic of Korea (No. 20194010201840) 2) Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea government (MOTIE) (20202020800330, Development and demonstration of energy-efficient reaction-separation·purification process for fine chemical industry)

References (60)

  • P.D. Chapman et al.

    J. Membr. Sci.

    (2008)
  • S. Doguparthy

    J. Membr. Sci.

    (2001)
  • Z. Huang et al.

    Chem. Eng. Res. Des.

    (2019)
  • X. Cheng et al.

    J. Membr. Sci.

    (2018)
  • Z. Xu et al.

    J. Membr. Sci.

    (2018)
  • A. Philip et al.

    Appl. Clay Sci.

    (2017)
  • L. Ge et al.

    Sep. Purif. Technol.

    (2017)
  • G. Zeng et al.

    Chem. Engi. J.

    (2017)
  • P. Shao et al.

    J. Membr. Sci.

    (2007)
  • X.L. Xie et al.

    Mater. Sci. Eng. R-49

    (2005)
  • S. Chaudhari et al.

    Appl. Surf. Sci.

    (2019)
  • Y. Liu et al.

    Int. J. Hydrogen Energy

    (2018)
  • M. Massaro et al.

    J. Coll. Inter. Sci.

    (2019)
  • S. Sadjadi et al.

    J. Catalysis

    (2018)
  • Y. Liu et al.

    Chem. Engi. J.

    (2018)
  • K. Feng et al.

    J. Membr. Sci.

    (2015)
  • S. Hu et al.

    J. Membr. Sci.

    (2012)
  • B. Smitha et al.

    J. Membr. Sci.

    (2004)
  • L. Liu et al.

    J. Membr. Sci.

    (2018)
  • D. Hua et al.

    J. Membr. Sci.

    (2014)
  • F. Kursun et al.

    J. Ind. Eng. Chem.

    (2016)
  • A. Sajjan et al.

    J. Ind. Eng. Chem.

    (2015)
  • Y. Han et al.

    J. Membr. Sci.

    (2014)
  • A. Ragauskas et al.

    Science

    (2006)
  • L. Vane

    Biofuels Bioprod. Bioref.

    (2008)
  • K. Sawamura et al.

    Appl. Mater. Interfaces

    (2015)
  • O. Bakhtiari et al.

    Desalin. Water Treat.

    (2012)
  • F. Nigiz, 485 (2020) 114465,...
  • J. Wu et al.

    J. Membr. Sci.

    (2019)
  • W. Cai et al.

    ACS Omega

    (2020)
  • Cited by (19)

    View all citing articles on Scopus
    1

    Note: Represents sharing 1st authorship.

    View full text