Desktop version

Home arrow Mathematics

  • Increase font
  • Decrease font

<<   CONTENTS   >>


Recently, carbon nanotubes (CNTs) have attracted attention for the synthesis of novel membranes with attractive properties for water treatment. CNTs have unique properties of high mechanical stability, high thermal stability and conductivity, good chemical stability, and high electrical conductivity (Saifuddin et al., 2012; Ihsanullah. 2019). Many researchers studied the effect of CNTs on membrane properties, either as surface modifiers or as additives.


The most common synthesis techniques of CNTs are laser ablation, arc- discharge evaporation, and chemical vapor deposition (CVD) (Saifuddin et al.,

2012). Arc discharge and laser ablation were used for preparation of carbon nanotubes at high temperature. Currently, low temperature CVD is applied (Pandey and Dahiya, 2016). By CVD. CNTs are grown on metal catalyst surfaces such as iron, nickel, and cobalt (Ahn et al., 2012).

Effect on Morphology

Membrane roughness increases in most cases when adding CNTs, while in some cases, membrane roughness decreases. Polyamide membrane surface modification using acidified (H2SO4+HNO3) MWNTs increased membrane roughness, Also, it was observed that addition of multiwall carbon nanotubes (MWCNT) with concentrations of 0.05% and 0.1% (w/v) packs the membrane surface together (Zhang et al., 2011). It has been reported that adding interlayers of acidified (H2SO4+HNO3) MWCNTs to NF membranes also increased membrane roughness (Wu et al., 2016). Typical SEM images are presented in Figure 2.1. Hydrophilic surface is more suitable for MF, UF, NF. and RO membranes, while a hydrophobic surface is required for MD membranes. Generally, CNTs have hydrophobic effects on membrane surfaces; however, it can by modified with acids to produce hydrophilic effects. Zhang studied the effect of acidified (H2S04+ HNO3) MWCNT surface modification for RO polyamide membranes and found that membranes became more hydrophilic (Zhang et al., 2011). Also, in nanofiltration membranes, when

SEM cross-section images of CNT/ PES composite membranes with different magnifications.(a. d)

FIGURE 2.1 SEM cross-section images of CNT/ PES composite membranes with different magnifications.(a. d): vertically aligned-CNT/PES membrane; (b, e): randomly distributed CNTs/PES membrane; and (c, f): PES pure membrane (Li et al., 2014). Reprinted from (Li et al., 2014), with permission from Royal Society of Chemistry, Order License Id: 4612430890809 (FIG 1 - “Enhanced water flux in vertically aligned carbon nanotube arrays and polyethersulfone composite membranes" Li, Shaoyun, et al / Journal of Materials Chemistry A2 31 (2014) 12171-12176.

adding an acidified (H2S04+HN03) MWNTs layer between the microfiltration membrane support and interfacial skin, the polymerization membrane became more hydrophilic (Wu et al., 2016).

For DCMD PVDF-co-HFP membranes, porosity and pore size increase when adding CNTs as a dope additive, as a result of fiber structure overlapping (Tijing et al., 2016).

Effect on Performance

CNTs affect membrane performance (flux and rejection) in different manners, based on permeation mechanism for each type of membrane. In the case of UF and MF, membrane rejection depends mainly on pore size (Baghbanzadeh et al., 2016). In the case of membrane distillation, carbon nanotubes enhance both flux and rejection, owing to higher vapor flux and higher hydrophobicity. For RO membranes, (Zhang et al., 2011) found that acidified MWCNT enhanced membrane flux significantly, while NaCl rejection decreased dramatically. It has been reported that addition of CNTs enhances membrane flux, Also, the additive concentration of optimum CNTs enhances membrane rejection (El- Din et al., 2015). The effect of surface modification of PSF membranes using CNTs (0.12-0.32wt. %) indicated that increases in acidified MWCNTs increase membrane flux until a concentration of 0.19 wt. % is achieved; after that,

Effect of MWCNT content in dope solution on PWP. and rejection of ultrafiltration PSF /CNT composite membrane (Qiu et al., 2009) with permission from Elsevier Order Number

FIGURE 2.2 Effect of MWCNT content in dope solution on PWP. and rejection of ultrafiltration PSF /CNT composite membrane (Qiu et al., 2009) with permission from Elsevier Order Number: 4612470410778 (FIG 10 - “'Preparation and properties of functionalized carbon nanotube/PSF blend ultrafiltration membranes”’ S. Qiu et al. / Journal of Membrane Science 342 (2009) 165-172).

membrane flux decreased, which indicated that there was a threshold for adding CNTs to enhance membrane flux, while membrane rejection decreased (Qiu et al., 2009). A typical performance curve is presented in Figure 2.2.

Effect on Mechanical Properties

Several studies indicated that CNTs enhance membrane mechanical properties. A study reported that insertion of MWCNTs interlayer between membrane support and coating layer improved mechanical stability and allow operating at higher pressure for composite nanofiltration PES membranes (Wu et al., 2016). Also, adding CNTs as dope additives to direct contact membrane distillation (DCMD) membranes enhances mechanical stability (Tijing et al., 2016).It is worth mentioning that addition of CNTs decreases membranes’ biofouling tendency because it deactivates bacteria (Baghbanzadeh et al., 2016). CNTs enhance membrane thermal stability as well (Gethard et al., 2010).

Table 2.2 summarizes the effect of CNTs on membrane morphology and performance from previous studies.

Mechanism of Action

Gethard et al. (2010) studied vapor permeation enhancement mechanisms in the case of CNT-enhanced MD. This study illustrated that addition of CNTs results in more hydrophobic membranes that decrease pore-wetting tendency, enhancing pure vapor transport through pores. Also, CNTs allow vapor transport via diffusion because CNTs have high adsorption and desorption capacities. Moreover, CNTs allow fast vapor diffusion along their surfaces, as well as a direct vapor transport through CNTs inner pores (Gethard et al., 2010). CNTs provide a pore network between pores inside membrane matrices, and provide additional pathways within membrane matrices that improves membrane permeability (Li et al., 2014; Baghbanzadeh et al., 2016).


Zeolite Preparation

Zeolites are microporous crystalline aluminosilicate materials with uniform pore and channel size; they are used in various applications owing to their unique properties. Incorporation of zeolites into polymer matrices has attracted great attention in membrane technology, because of such excellent advantages as permeability improvement of the selective component, in addition to the enhancement of molecular sieving property, thermal resistance, and chemical stability (Maghami and Abdelrasoul, 2018).

Zeolites are commonly prepared by using hydrothermal crystallization, hydrothermal synthesis, vapor phase transport, sol - gel method, chemical growth, seeding method (embedding zeolite into a support), and microwave synthesis. Typically, hydrothermal synthesis is used to prepare zeolite on a porous

Effect of CNTs additives on membrane morphology and performance



IP % (w/v)





PS/NM P/PEG/LiCl/Triton



PIP (3%)TMC (0.5%)

#CA: 41 decreased

#thickness: increase #Ra: (18.9 nm)


#zeta potential increase negative charge

#LMH: 34.5 #R%: 19.9 for NaCI

(Abolfazli and Rahim- pour, 2017)

PS, NMP. PEG, LiCI, Triton X-100, glycerol/TETA (4%) + SiOi (0.1%)

PIP (3%) TMC (0.5%)

#CA: 30 decreased

#thickness: increase

#Ra: (11.9 nm) decrease

#zeta potential: increase negative charge

#LMH:31 #R% 26 for NaCI


PSU, DMF/porous MCM-41 silica NPs (100 nm) (OtoO.lwt.. %) MCM-41

MPD 2.0wt..% (3min) TMC 0.15wt..% (2min)

#Thickness: (300-500nm) #CA: decrease from 57 to 27.9

#Ra: increase from 135 to 159nm #zeta potential decrease from -5.71 to -9.54m V

#LMH increase from 28 to 46 #R% no change

97.5, 98.5% for (NaCI, Na2S04)

(Yin et al., 2012)

PSU, DMF/spherical silica (NPs)

#Thickness: (300- 500nm) #CA decrease from 57 to 30

#Ra: increase from 135 to 159 nm #zeta potential decrease from -5.71 to -9.54mV

#LMH increase from 28 to35 # R% no change 98.5% for (NaCI, Na2S04)


PSF/silica NPs (wt„ 15 nm) (1-3%)

РАМАМ 0.5% (wt..)/SDSSi02 (10 min) TMC 0.3%wt..(60s)

#R increase from 8.72 to 36.5nm #zcta potential increase negativity from -16.78 to-20.56 mV

#LMI1: increase to 10 # R%: 91.23(MgS04), 48.80 (MgCl2), 46.01 (NaCI), 92.62 (Na,S04), atl.5%SI02

(Jin ct al., 2012)




IP % (w/v)





PVDF/silica NPs

#Pore size increased to 140 nm #CA:94

#Porosity decreased to 50% #Ra: 27 nm

LMH: increase from 0.7 to 2.9

(Efome el al., 2015)


PVDF/silica NPs

#Viscosity: increase

#CA: decrease from 82 to 53' #porosity: increase from 5 to 84 % #pore size: increase #Ra: increase

LMH: 301 increase

  • (Yuetal.,
  • 2009b)
  • 1 dispersed in mixed acid (H2S04/HN03 = 3/1 by volume), then treated by ultrasonic agitation at 80 °C for 6 h.
  • 2 MWCNTs (3 g) were treated by 100 mL H2S04/HN03 (3/1, v/v) solution under I20°C for 2 h.
  • 3 CA:Acetone 20:80.
  • 4 CA:Acelone 25:75.
  • 5 Functionalized by oxidation/ purification in a strong acidic medium to enhance their dispersion within the polymer matrix.
  • 6 Multiwallcd carbon nanotubes (M WNTs) functionalized by isocyanate and isophthaloyl chloride groups were synthesized via the reaction between carboxylated carbon nanotubes and 5-isocyanato-isophlhaloyl chloride (ICIC).

support for membrane-based separation. This approach produces a thin dense layer of zeolite membranes that enables high water permeability (Feng et al., 2015; Makhtar et al., 2020). They are generally prepared hydrothermally by mixing solutions of aluminates and silicates, often with the formation of a gel, and by maintaining the various mixing temperatures for selected periods (Barrer, 1982; Cundy and Paul, 2005; Hani et al., 2009; Dahe et al., 2012). A typical hydrothermal zeolite synthesis can be explained as follows:

  • • Reactants containing silica and alumina are mixed together with a cation source, usually in a basic medium.
  • • The aqueous reaction mixture is heated in a sealed autoclave.
  • • For some time after rising to synthesis temperature, an amorphous phase is formed.
  • • After the aforementioned induction period, crystalline zeolite products can be detected and gradually, all amorphous material is replaced by an approximately equal mass of zeolite crystals.
  • • Zeolite crystals are recovered by filtration, washing, and drying.

Zeolites are commonly categorized into three groups in terms of the Si/Al ratio in their framework: low silica (= 1), intermediate silica (between 1 and 10), and high silica (> 10); the higher the silica in zeolite, the higher surface hydrophobicity (Rezaei, Dasht Arzhandi et al., 2016). It was observed that zeolite inclusion into the membranes was conducted by two ways: as a coat for the membrane (by interfacial polymerization), and as an additive in the dope of the prepared membrane. In both, the most common effects of zeolite incorporation observed on the membrane performance and characteristics are illustrated in Table 2.3 and are summarized in the following four subsections.

Effect on Water Permeability and Hydrophilicity

Fathizadeh et al. (2011) showed that A and X zeolites in sodium form [incorporated in polyamide membranes (PA)] have pore diameters of approximately

  • 4.2 and 7.4A, respectively that is between water diameter (2.7A) and the hydrated sodium and chloride ion diameters (8-9A).Therefore, these particles provide preferential flow paths only for water and can increase the hydrophilicity of the polyamide membrane surface. This concept was validated by Lind
  • (2009) and Yin et al. (2012). The latter added that embedding the inorganic MCM-41 nanoparticles in the PA polymer could simultaneously reduce the cross-linking in the thin-film layer and provide additional short paths for water flow through the hydrophilic ordered porous structure. An alternative explanation is that the increased negative charge in thin-film nanocomposite (TFN) membranes caused by MCM-41 NPs has contributed to salt rejection by the electrical repulsion or Donnan exclusion; accordingly, the level of salt rejection can be maintained. Zeolitic imidazolate framework-8 (ZIF-8) (with large cavity 11.16 A) revealed that the PA/Z1F-8 membrane surface has an enhanced water uptake, and a more negative charge density with increased ZIF-8 content, which increases water uptake values (Li et al., 2018).

Lind (2009) showed that the AgA-TFN membranes have higher pure water flux than NaA-TFN membranes, while maintaining similar rejections of NaCl. For perspective, when similarly tested, commercially fabricated seawater, brackish water, and high-flux RO membranes produced pure water permeabilities of 2.6, 16.8, and 27.7 pm MPa'1 s'1 and NaCl rejections of 92.0, 93.0, and 75.2%, respectively. Because the characteristic pore dimension of AgA nanocrystals (3.5 A) is smaller than that of NaA nanocrystals, the permeability of AgA-TFN membranes was smaller.

On the other hand, Dahe et al. (2012) used Lucidot Nanozeolite in the PS dope preparation and showed that increase in zeolite-added ratio resulted in an initial increase, followed by a subsequent decrease in flux of the hollow fiber membrane (HFM). The increase in flux of HFMs at initial loading may be attributed to zeolite suspension at nanoscale, leading to enhancement in nucleation and nodule formation. Further decrease in flux could result from reduction in pore number and increase in pore size, because of agglomeration of zeolite at higher loading. It also explains the increase in pore size, which resulted in increase in NMWCO. Rezaei et al. (2016) tested ZSM5 zeolite-filled PVDF mixed matrix membranes that were used for C02 absorption in contactor systems. The hydrophobicity, porosity, and fingerlike macrovoids of membranes were increased. Typical performance data is presented in Figure 2.3.


Lind (2009) demonstrated the strong antimicrobial reactivity of silver-exchanged LTA nanocrystals and suggested the possibility of developing more fouling-

Permeate water flux, NaCl rejection, RMS roughness, contact angle, and interfacial free energy versus zeolite weight % in zeolite membranes (Fathizadeh et al., 2011) (License Number 4615280311102)

FIGURE 2.3 Permeate water flux, NaCl rejection, RMS roughness, contact angle, and interfacial free energy versus zeolite weight % in zeolite membranes (Fathizadeh et al., 2011) (License Number 4615280311102).

resistant membrane materials. In addition, Yin et al. (2012) concluded that the high hydrophilicity and negative charge of TFN membranes introduced by MCM-41 nanoparticles could improve the membranes’ resistance to fouling.

Effect on Morphology

NaX zeolite is very hydrophilic and improves the interfacial free energy and contact angle in the TFN membrane, which decreases the surface roughness in the membrane. The variation of the RMS roughness confirmed that the surface of the membrane turns into a smooth layer after adding nano-NaX in IP polymerization. These improvements in surface properties lead to good physical and chemical stability properties (Fathizadeh et al., 2011).

Rezaei et al. (2016) tested ZSM5 zeolite-filled PVDF mixed matrix membranes and showed that the surface roughness, wettability resistance, and mechanical stability of membranes were also considerably improved.

Li et al. (2018) used zeolitic imidazolate framework-8 (Z1F-8) to prepare high- performance, thin-film-nanocomposite cation exchange membranes and showed that by raising the initial concentration of ZIF-8 to 0.08%, the surface roughness increased to 70.8 nm, with the contact angles decreasing from 78° to 71°.

Lind (2009) deduced that the addition of Nano zeolite in the polyamide thin films alter the interfacial polymerization kinetics and polyamide thin film structure. In addition, silver-exchanged LTA zeolites produced membranes with higher flux and rejection, along with smoother and more hydrophilic interfaces than those formed from as-synthesized LTA in the sodium form. Dahe et al.

(2012) showed that zeolite nanoparticles participated in, and directed, the nucle- ation process during phase separation and the mixed matrix membrane skin consisted of nodules containing more zeolite nanoparticles. Rezaei et al. (2016) tested ZSM5 zeolite-filled PVDF mixed-matrix membranes, and fingerlike macrovoids of membranes were increased. Considering the opposing effects of the increase in thermodynamic instability (thermodynamic effect) and the increase in viscosity (kinetic effect), the enhancement of fingerlike microvoid formation by the incorporation of ZSM5 nanoparticles indicates that the thermodynamic effect dominates the formation of the cross-sectional structure.

Shi et al. (2016) proved that the use of silver-loaded zeolite Y as a carrier in dual layer PVDF HF membranes had not only effectively avoided potential aggregation of Ag nanoparticles, but had also facilitated their long term stability in the membrane outer surfaces for longer-term antifouling effects in seawater, providing additional routes to extend the service life of the current desalination units.

Effect on Mechanical Properties

Dahe et al. (2012), among others, have reported the effects of nanocomposite PSF/zeolite HFMs on mechanical properties. Their work is depicted in Figure

2.4 and is included in Table 2.3.

Effect of zeolite Nanoadditives on membrane morphology and performance.





Morphology Mechanical



Flat sheet: zeolite added in the membrane coat


NaX zeolite (40-150 nm) (0.004 0.2 wt./vol)


15/5/80(wt..%) MPD/ TMC 2/0.1

#CA decreased from 70 to 42

//thickness decrease from 0.282 to 0.15pm

#LMII increase from 8.32 to 14.6 #R% decrease from 98 to 95 (NaCl)

Fathizadch ct al. (2011)

MPD/TMC 3/0.15

#CA decreased from 65 to 45 //thickness decrease

from 0.253 to0.173pm

#LMH increase from 8 to 29, #R% decrease from 98 to 60 (NaCl)


silver-exchanged LTA zeolites (prepared using hydrothermal method) 0.4% (w/v)


2.3:6.6:0.02% (15s) TMC 0.1%( Imin)

#zeta potential mV(-ve) //surface area difference % (SAD)(30-59).

# Ra increase from 60 to 70


increased from 5.9 to 7 urn/ Mpa/s

#R%: no change 93-95 (NaCl)

Lind et al. (2009)


(AgA) Nanocrystals with silver 0.4% (w/v): added in coating

#zcta potential (-vc) //surface area difference % (SAD) (30-19).

# Ra decrease from 60 to 40


increased from 5.9 to 9.5 um/Mpa/s

#R%: no change 93-95 (NaCl)

Hollow fiber: zeolite added in the membrane dope


Lucidot NZL 40: 0.01-1%

PSf/TPGS/NMP/NZL: 25/1/74/0.01-0.1-1 wt„


#MWCO: increase from 9500 to 5400 //ID: decrease (823-783) # wall thickness: increase (128-132) then decrease to 121

#Young's Modulus: decrease from 182 to 168 Mpa //Break strength: decrease from 9.44 to 8.87 Mpa //Elongation: decrease from 80.1 to 63.9%

# LMH: initial increase to 21.31, then decrease to 11.79

Dahe et al. (2012)


ZSM-5: 0-5 wt..%


18/79.5/2.5/0-5 wt..%

# fully asymmetric structure with increase in of fingerlike macrovoids #Porc size: decrease from 126 to 59 nm #CA increase from 89 to 104 #Ra increase from 19 to 48

# CO, removal via membrane contactor



Arzhandi, et al. (2016)


Ag-loaded zeolite Y



#Avoided potential aggregation of Ag nanoparticles, but also had facilitated their long term stability in the membrane outer surfaces

Shi et al. (2012)

Mechanical properties plot of nanocomposite PSF/zeolite HFMs. (Dahe et al., 2012) (License Number4615281150855)

FIGURE 2.4 Mechanical properties plot of nanocomposite PSF/zeolite HFMs. (Dahe et al., 2012) (License Number4615281150855).

<<   CONTENTS   >>

Related topics