Introduction

One of the most common sources of pollution is phenol-contaminated effluents generated during the manufacture of various chemical compounds and refinery industries. The elimination of phenol is of major importance in wastewater treatment because phenolics, even at low concentrations, have the potential to damage people, wildlife, and the environment. The Environmental Protection Agency's (EPA) water purity standards include a regulation level of one part per billion (ppb) for phenol content. The toxicity tolerance for both humans and marine species has been observed to be between 9 and 25 mg L−1 (Wibawa Hendra Saputera 2021).

Wastewater separation techniques are critical to producing high-quality products with acceptable yields at a cheap cost. Conventional methods of separating mixtures, such as distillation, extraction, or absorption, as well as crystallization processes and combinations of these methods, are well known; however, many of them are costly energy, so alternative methods, such as membrane-based separations, are being researched. Membrane methods are reliable and cost-effective for separation. It consumes less energy than other methods such as distillation because mixture separation may be done at considerably lower temperatures than the boiling temperatures of the components (Zeitoun 2020). The development of appropriate membrane materials and modules is important for effective competition with current separation technologies. The most important membrane technologies used to remove phenols from wastewater are extractive membrane bioreactors and hollow fiber membranes; photocatalytic membrane reactors; high-pressure membrane processes such as nanofiltration, reverse osmosis, and pervaporation; and membrane distillation (Loh et al. 2016).

In this work, a single-step pervaporation technique is integrated with liquid–liquid phase separation. Pervaporation is recognized as an efficient procedure for separating azeotropic mixtures, chemicals with low boiling points, and mixes containing heat-sensitive compounds. In recent years, various studies have been conducted using polymeric membranes (Mohammadi et al. 2015). A good pervaporation membrane material will have a high permeation flux and separation factor. The success of the pervaporation process is dependent on membranes that have high permeability, selectivity, and mechanical strength (Satyanarayana 2012). Table 1 lists current investigations on the recovery of phenol and organic chemicals from wastewater streams using the pervaporation method.

Table 1 Pervaporation data for various membranes for removal of organic pollutants
Full size table

There is little information about the removal of phenol using cellulose acetate till now. Most of the studies using cellulose acetate incorporated different polymeric materials to enhance the separation efficiency. The present research aims to study the effect of phenol separation using CA and compare it with different polymeric membranes and study the effect of surface nature; hydrophilicity/hydrophobicity on the phenol separation. The experimental work explores the effect of varying phenol concentrations in the feed mixture on membrane flux, selectivity, permeate, and separation factors.

Materials and method

Materials

The material specifications and suppliers are given in (Table 2).

Table 2 Specifications, and supplier of the raw materials used in membrane preparation
Full size table

Membrane preparation

PVA/SA membrane

The 2.5 wt% PVA solutions and 2wt% SA were prepared by dissolving separately 2.5 g PVA and 2 g SA in 100 ml deionized water. then 50:50 proportion of the two solutions were mixed with continuous stirring for 2 hr. to ensure the complete mixing of the two polymers followed by drying the polymers at room temperature. For the crosslinking, the prepared PVA/SA was immersed in a solution containing 10% GA (as a crosslinking agent) and 0.05 vol% HCl in acetone (as a catalyst). The prepared PVA/SA membrane was immersed in methanol for 24 hr. to remove any unreacted GA, rinsed with water, and then dried at room temperature (Dave and Nath 2017a).

Cellulose acetate membrane

The manufacturing of the cellulose acetate membrane involved the implementation of the phase inversion method for the CA membrane. A solvent blend containing equal parts of acetone and dimethylformamide was employed to dissolve a 25% weight fraction of cellulose acetate (CA). The mixture was stirred periodically for a duration of eight hours. Following a 24-hour waiting period, any air bubbles that could have been retained in the polymeric solution were effectively removed. The polyester support functioned as the operative substrate during the casting procedure. Subsequently, the process of coagulation occurs within a cold-water bath maintained at a temperature of 5 °C. Subsequently, the fabricated membrane underwent annealing in a water bath maintained at a temperature of 85 ºC for a duration of five minutes (Algieri and Chakraborty 2022).

Method

The pervaporation apparatus consisted of a permeation cell made of stainless steel, a constant temperature bath, and glass tubes for condensing and collecting the permeate vapor. The membrane area in contact with liquid was about 16 cm2. The downstream pressure was maintained from 200 mbar. Pervaporation experiments were carried out at a feed temperature from 25 to 55 °C, and feed concentration ranged from 1000 to 9000 ppm. The experiments were carried out on a Pervaporation apparatus as shown in Fig. 1. The heated feed mixture at the desired temperature was continuously circulated over the membrane using a circulating peristaltic pump from the feed tank, while a vacuum is applied on the downstream side of the membrane. The liquid permeant vaporizes between the upstream and downstream sides of the membrane where it is obtained as vapor and condensed (Ong et al. 2016). The permeate was collected and analyzed using an Agilent UV–VIS spectrophotometer. The analysis of permeate was carried out according to ASTM 5530C (Cristian perla Martinez 2010). The separation factor (α) and flux (J) were determined using the following equations

$$\alpha = \frac{{{\text{Yphenol }}/{\text{Ywater}}}}{{{\text{Xphenol }}/{\text{Xwater}}}}$$

where X and Y are the weight ratio of each component respectively in feed and permeate.

$${\text{J = }}\frac{{\text{W}}}{{{\text{A}} \times {\text{ time}}}}$$

where W is the wt. of the permeate (kg), A is the effective membrane area (m2).

Fig. 1
figure 1

Scheme of the Laboratory Scale setup of the Pervaporation Process

Full size image

Characterization

The investigation of the three membranes' properties was discussed using a combination of techniques, including scanning electron microscopy (SEM; S4800, Hitachi, Japan) at an accelerating voltage of 20 kV, contact angle, Fourier-transform infrared spectroscope (FTIR; Tensor 37, Bruker, Germany) with a wave number range of 500–4000 cm−1, a scan number of 32, and a resolution of 4 cm−1, and X-ray diffractometer (D8 ADVANCE, Bruker, Germany) using Cu-Ka radiation (k¼1.5438 A°).

Result and discussion

Fourier-transform infrared spectroscopy (FTIR)

It was used to characterize the functional groups of the three developed membranes. Figure 2a–c show the visible characteristic peaks of PVA, PVDF, and CA membranes, respectively. The peaks of the PVA/SA membrane are indicated in Fig. 2a at 3422.23, 1639.45, and 1100.43 cm−1. The peak at 3422.23 cm−1 represents the peak of intramolecular hydrogen bonding (O–H) (Wang et al. 2021). FTIR spectra of PVDF thin films are also shown in Fig. 2b. Peaks at 435, 502, 631, 724, 792, 1120, 1384, and 1446 cm−1 are used for α -phase identification, while peaks at 874 cm−1 are β-phase fingerprints. α–phase molecular vibration peaks are very close to β -phase. As a result, separating two phases by using FTIR spectroscopy is extremely difficult (Zeitoun 2020).

Fig. 2
figure 2

FTIR Spectra of: (a) PVA/SA membrane film, (b) PVDF membrane film, and (c) CA membrane film

Full size image

The spectrum of FTIR for the CA membrane is shown in Fig. 2c, with a peak at 3420 cm−1 representing the carboxylic acid group. The vibration of the C=O group was assigned a strong peak around 2112 cm−1, and the stretching mode of the C=C was assigned at 1640 cm−1. The distinctive peak at around 724 cm-1 reflected the bending of the C–H group (Zafar and Ali 2012b).

X-ray diffraction analysis (XRD)

The X-ray of PVA/SA, PVDF, and CA membranes are illustrated in Fig. 3a–c. In Fig. 3a, the XRD of the PVA/SA nanofiber membranes showed three peaks at 17.54°, 22.6°, and 25.94°. Crosslinks typically reduce the chain mobility of a polymer as the crosslinking reaction progresses. As a result, PVA showed a sharp peak with high intensity at 2 θ = 17.54°, and the resulting semi-crystal peak (SA) was found at 2 θ = 22.6 (Hezma et al. 2023). Because of the presence of an amorphous structure, the wide range from 17.54° to 25.94° can provide information on the presence of a component. The X-ray diffraction pattern of PVDF thin films is shown in Fig. 3b. The peaks in the XRD pattern at 2 θ = 17.9 and 26.26 confirm the presence of -phase (Satapathy et al. 2011).

Fig. 3
figure 3

XRD Pattern of: (a) PVA/SA membrane film, (b) PVDF membrane film, and c) CA membrane film

Full size image

Furthermore, the XRD of the CA nanofiber is described in Fig. 3c, which shows two peaks with nearly identical intensities at 2 θ = 23.34 and 26.6, which are also typical cellulose peaks (Boccaccio 2002).

Scanning electron microscopy (SEM)

Figure 4a, b illustrates the cross-sectional surface structure of the cellulose acetate (CA) membrane, indicating its composition of two layers exhibiting different pore sizes. The membrane that is synthesized exhibits anisotropic characteristics, wherein the outer layer (1) consists of smaller pores, while the inner layer (2) consists of slightly bigger pores. In comparison to the total thickness found in the inner supporting layer, the skin's outermost layer exhibits a relatively reduced thickness. Having a dense arrangement of pores near the skin, reinforced by a spongy microporous structure, is considered advantageous due to its capacity to enhance membrane permeability (Hoshi et al. 1997).

Fig. 4
figure 4

SEM Images of: (A, B) CA membrane, (C, D) PVA/SA membrane, and (E, F) PVDF Membranes

Full size image
Fig. 5
figure 5

Mechanical Properties of CA, PVA/SA, & PVDF Membranes

Full size image

The surface structure of the crosslinked PVA/SA membranes was depicted in Fig. 4 (c-d), revealing a heterogeneous arrangement of the nanofibers characterized by a smooth surface and the presence of many spaces between the fibers. The formation of a dense network structure with comparatively smaller pore size was attributed to the increased involvement of molecular chains in the crosslinking reaction (Chung et al. 1998).

Figure 4e–f demonstrates the PVDF membrane's properties. The PVDF membrane has a smaller mean pore size, resulting in a significantly greater liquid entry pressure and excellent anti-wetting properties. In addition to its high porosity, the PVDF membrane has a dense and smooth top surface, as shown in Fig. 4f (Liu et al. 2018).

Contact angle

The hydrophilicity of membranes was determined using a contact angle test. (Table 3) shows that the test was performed on three different membrane samples: PVA/SA, PVDF, and CA. The PVDF membrane's water contact angle was 74.8°, indicating low wettability on the surface (highest contact angle value) since hydrophobicity increases with contact angle (Vetrivel et al. 2021). Furthermore, the hydrophilicity of SA increased when 2.5% PVA was added, and the contact angle decreased to 64.8°, increasing PVA/SA permeability (Mallakpour and Behranvand 2017). Furthermore, the figure shows that CA has a lower contact angle of 48.3°. This indicates that the CA membrane is more hydrophilic than the other membranes.

Table 3 Contact angle measurement
Full size table

Mechanical properties

The mechanical properties of the CA, PVA/SA, and PVDF membranes are illustrated in Fig. 5. The maximum tensile strength at break followed the same pattern as the elongation at break of three membranes. CA has a maximum tensile strength of 65.5 MPa and a % elongation of 48% compared to PVDF and PVA/SA tensile strengths of 55.1, 47.8 MPa, and % elongation of 36.3, and 35.2%, respectively. This could be because CA and PVDF have a more porous membrane structure than PVA/SA, which has numerous voids between the fibers, reducing its stability (Zeitoun et al. 2022).

The impact of feed temperature on various membranes

Figures 6, 7, 8 illustrates the impact of differing feeding temperatures on the separation factor of phenol, as well as the water and phenol fluxes, utilizing membranes made of PVA/SA, PVDF, and CA. The range of temperature from 25 to 65 °C, while the phenol concentration remains constant at 1000 ppm. At an elevated temperature of 65 °C, the separation factor of the CA membrane exhibits a significant rise, rising from 2.1 to 6 In the interim, the separation factor of the PVA/SA and PVDF membranes exhibits a little increase and remains rather constant. Similar results were obtained in the water flux curve illustrated in The water flux of the CA membrane raised from 125 to 295 kg m−2 h−1 with increasing temperature, whereas the water flux of the PVA/SA and PVDF membranes decreased.

Fig. 6
figure 6

The Effect of Temperature on The Phenol Separation Factor While Employing (CA), (PVDF), & (PVA/SA) Membranes

Full size image
Fig. 7
figure 7

The Effect of Temperature on Water Flux While Employing (CA), (PVDF), & (PVA/SA) Membranes

Full size image
Fig. 8
figure 8

The Effect of Temperature on Phenol Flux While Employing (CA), (PVDF), & (PVA/SA) Membranes

Full size image

These findings are consistent with other studies that show that increasing the temperature makes permeating molecules stronger, thereby increasing permeation flux and the ease of swelling in the membrane structure (Dave and Nath 2017b). The hydrophilic nature of the membranes provides a strong driving force for water affinity, the CA membrane facilitates permeate transport at higher temperatures and increases water flux. As a result of increasing the vapor pressure at higher temperatures, water is preferably dissolved and transported in hydrophilic membranes, increasing water flux.

Meanwhile, Fig. 8 demonstrated that the PVDF membrane has a higher phenol flux of 1 kg m−2 h−1 at 25 °C than CA and PVA//SA because PVDF is a hydrophobic membrane, and thus phenol flux decreases with increasing temperature to 0.79 kg m−2 h−1 at 35 °C and nearly remains constant until 65 °C due to the partial pressure difference created by the temperature gradient in the membrane allows only water to pass through it. This was attributed to an increase in water vapor pressure as the temperature gradient across the membrane increased (Yadav and Yadav 2021). Similar results were obtained in the PVA/SA membrane, SA has a higher tendency to absorb water and causes significant swelling, which affects the PVA pores and slightly increases the flux rate. The increase in the feed temperature affects the thermal mobility of the polymeric chain, increasing the diffusion rate of the permeating molecules (Rhimi et al. 2021).

The impact of feed concentration on various membranes

Figures 9, 10, 11 illustrate the effect of feed concentration on the separation factor of the three membranes. The operating conditions were at 65 °C and the feed concentrations ranged from 1000 to 9000 ppm. Figures 9, 10 showed that the separation factor and water flux of the CA membrane increased from 5.2 to 38.2 and 233.5 to 274.7 kg m−2 h−1 respectively with increasing phenol concentration. Pervaporation is a solution-diffusion process, but it appears clear that the membrane's high selectivity is primarily due to its hydrophilicity nature. This is consistent with the general observation that when organic compounds are separated from dilute aqueous solutions by pervaporation using organophilic membranes, their preferential sorption in the membrane often results in selective permeation (Hao et al. 2009).

Fig. 9
figure 9

The Impact of Feed Concentration on The Phenol Separation Factor While Employing (CA), (PVDF), & (PVA/SA) Membranes

Full size image
Fig. 10
figure 10

The Impact of Feed Concentration on Water Flux While Employing (CA), (PVDF), & (PVA/SA) Membranes

Full size image
Fig. 11
figure 11

The Impact of Feed Concentration on Phenol Flux While Employing (CA), (PVDF), & (PVA/SA) Membranes

Full size image

In dilute solutions, it is observed that the higher phenol concentration in the feed, increases the driving force and the permeability of the phenol membrane as well as decreases for water permeation, but the decrease is relatively low because phenol is in less concentration. As a result of pervaporation theory and the selectivity of CA, a hydrophilic membrane, the transport of molecules is controlled not only by the sorption step but also by diffusion through the membrane. Water molecules are smaller than phenol molecules and are expected to diffuse more freely through the membrane (Zafar and Ali 2012a).

Figures 10, 11 show separation factor water flux of PVA/SA increased from 0.15 to 10.2 and 81.1 to 128.1 kg m−2 h−1 but at a lower rate compared to the CA membrane. This can be explained due to the higher selectivity of the SA to water which causes the membrane to swell during the pervaporation process and decreases the membrane mobility and diffusivity of phenol permeating through it and the flux will be less increased. However, Fig. 11 demonstrated that the highest phenol flux was for the PVDF membrane due to the PVDF membrane's higher selectivity to phenol (Vatani et al. 2018).

Effect of flow rate on different membrane

Figures 12, 13, 14 explain the behavior of increasing flow rate of aqueous solution through the CA, PVDF, and PVA/SA membranes at a phenol concentration of 9000 ppm and temperature of 65 °C, and flow rate ranged from 2 to 6 L h−1. Figure 12 shows that the separation factor of the CA membrane increases with flow rate due to the velocity increases therefore Reynold‘s number and the rate of water transfer from the bulk solution to the membrane surface increased and subsequently, the water flux and separation factor enhanced (Khatinzadeh et al. 2016).

Fig. 12
figure 12

Effect of Flow Rate on Phenol Separation Factor Using CA, PVDF, and PVA Membranes

Full size image
Fig. 13
figure 13

Effect of Flow Rate on Water Flux Using CA, PVDF, and PVA Membranes

Full size image
Fig. 14
figure 14

The Impact of Flow Rate on Phenol Flux While Employing (CA), (PVDF), & (PVA/SA) Membranes

Full size image

As observed from Figs. 12 and 13, by increasing the flow rate of feed from 2 to 6 L h−1, the PVA/SA membrane selectivity and separation factor slightly increased and almost constant. Increasing flow rate leads to the increased penetration of water molecules However, as the results show, the variations in flow rate in the PVA/SA membrane are less pronounced because the swelling effect of SA inhibits the penetration and mobility of water molecules, thereby affecting the flow rate (Alanezi et al. 2016).

The impact of the feed flow rate on the phenol flux for PVDF was examined in Fig. 14. As the feed flow rate increases, so does the PVDF membrane's phenol flux. This rise is the result of a higher Reynolds number, which increases turbulence and the flow's greater mixing in the channels. This suggests that more turbulent flow increases the driving force for evaporation by decreasing the thickness of the boundary layers for both temperature and concentration (i.e., the boundary layer resistance). Turbulence further speeds the phenol from the bulk feed to the membrane surface by raising the convective mass transfer coefficient at the feed boundary layer, which is directly related to Reynold number (Xu et al. 2016).

Membrane performance

The performance of the CA, PVA/SA, and PVDF membranes for the removal of phenol from industrial wastewater at a concentration of 9000 ppm and flow rate of 6 L h−1 at 65 °C is investigated in Fig. 15 and (Table 4). The CA was found to have a higher separation factor of 52 due to its hydrophilic properties with spongy micropores, which increase the separation factor and increase water flux in permeate. Due to its hydrophobicity and phenol selectivity, the PVDF membrane has a higher phenol flux than the CA membrane, reaching 3.3 kg m−2 h−1.

Fig. 15
figure 15

Membrane Performance at the Optimum Condition

Full size image
Table 4 Membranes performance
Full size table

Conclusion

The present study investigated the removal of phenol from an aqueous environment utilizing membranes made of PVA/SA, PVDF, and CA through the pervaporation process.

  • The three membranes were characterized using scanning electron microscopy (SEM), X-ray diffraction Analysis (XRD), Fourier transforms infrared spectroscopy (FTIR), and contact angles which indicate the hydrophilicity structure of membranes. Also, the optimum fluxes and separation factor measurements were conducted using synthetic solutions under certain operating conditions, including a phenol concentration of 9000 ppm, temperature of 65 °C, and a flow rate of 6 L h−1.

  • SEM results showed that the CA membrane has a spongy structure which enhances permeability compared to PVA/SA and PVDF membranes which have dense structures.

  • The CA membrane has a maximum tensile strength compared to PVA/SA and PVDF due to its high pore size.

  • The CA membrane possessed a higher water flux of 348.25 kg m−2 h−1 and a separation factor of 49 compared to PVDF, and PVA/SA membranes at 65 °C and a flow rate of 6 L h−1. This can be attributed to the higher porosity and hydrophilicity structure of the CA membrane. Meanwhile, the PVDF membrane indicates a higher phenol flux of 3.3 kg m−2 h−1 compared to the other membranes due to the hydrophobicity structure of the membrane.

  • CA membrane is highly recommended for phenol removal industrial wastewater than PVA/SA and PVDF membrane.

Future work

  • Use industrial wastewater for this application.

  • Examine the membrane fouling test and evaluate how vacuum pressure affects permeation and separation.

  • Compare the performance of cellulose acetate membranes made from organic materials like sugarcane bagasse.