The Removal of Reactive Red 141 From Wastewater: A Study of Dye Adsorption Capability of Water-Stable Electrospun Polyvinyl Alcohol Nanofibers

2.1 Materials

PVA with an average molecular weight of 125,000 g/mol, BTCA, sodium hypophosphite monohydrate: (NaPO₂H₂·H₂O), sodium sulfate (Na₂SO₄) and sodium carbonate (Na₂CO₃) were purchased from Sigma Aldrich Chemical Company. The dye, C.I. Reactive Red 141, was supplied from DyStar and used as received. The chemical structure of the dye is illustrated in Figure 1.

Figure 1

Chemical structure of Reactive Red 141 [8].

2.2 The preparation of electrospinning solutions

A weighed amount of PVA was dissolved in distilled water at 100°C to prepare an aqueous PVA stock solution at a fixed concentration of 10% w/w. BTCA as a crosslinking agent was selected to be 5% (w/w_polymer) and was directly added into the spinning solution with sodium hypophosphite monohydrate as a catalyst in the ratio of 2:1 (w/w) followed by stirring for 15 min. As a control, neat-PVA nanofibers were also prepared by electrospinning.

2.3 Electrospinning

Electrospinning of the polymer solutions was carried out by a setup consisting of a syringe (10 ml) with a stainless steel needle (22 gauges and flat tip), a ground electrode and a high-voltage supply (Simco, MP Series CM5 30 P, Charging Generator Output 30 kV DC). A grounded rotating metal drum collector covered by a 90 mesh of plain weave monofilament of 100% polyester screen mesh fabric was used as deposition material. Neat-PVA and PVA−BTCA solutions were electrospun for 3, 5 and 10 h. After electrospinning, the nanofibers were heat fixated at different temperatures such as 130, 150 and 170°C for 10 min and coded as PVA-130, PVA-150, PVA-170, PVA−BTCA-130, PVA−BTCA-150 and PVA−BTCA-170. All electrospinning experiments were performed at room temperature (22 ± 2°C), where the relative humidity was 22–40%.

2.4 Thermogravimetric analysis (TGA) of nanofibrous mats

TGA (Perkin Elmer TGA-4000) of PVA and PVA−BTCA cross-linked nanofibers were carried out by heating samples from 25 to 600°C at a rate of 10°C/min under continuous nitrogen purge at the rate of 20 ml/min.

2.5 Scanning electron microscope (SEM) analysis

The electrospun nanofibers were sputtered by Quorum Q150R S ion sputtering device with a thin layer of gold prior to SEM observation. The morphology of the nanofibers was analyzed by an SEM (Phenom G2 pro).

2.6 Measurement of thickness, weight per square meter and fiber diameters

The thicknesses of the nanofiber mats were measured by Mitutoyo digital micrometer at 0.01 mm accuracy. Weights of nanofiber mats were calculated from the weights of small rectangular pieces. A total of eight and three measurements were carried out from the different parts of each sample for thickness and weight, respectively. The mean diameter of the resultant fibers was calculated from measurements on SEM images of ×10,000 magnification by using ImageJ program. Approximately 50 measurements were carried out from the different parts of PVA nanofiber samples. Measurements of all thicknesses, weights and fiber diameters were expressed as mean ± SD.

2.7 Swelling and weight loss of nanofibrous mats

Swelling of the neat-PVA and PVA−BTCA electrospun nanofiber mats was measured upon exposure in distilled water for 24 h. The samples were removed from the water and carefully blotted with tissue paper to remove the excess from the surface. The degree of swelling due to the water uptake was determined, and weight loss was calculated at the end after drying for 4 h at 70°C in an oven. The percentages of degree of swelling and weight loss were calculated as follows [32]:

and

where M was the weight of each nanofiber mat sample upon exposure in distilled water for 24 h, M_d was the weight of the PVA nanofibers after swelling and subsequent drying and M_i was the initial weight of the sample.

2.8 Pore size measurements

The pore size measurements of the PVA nanofiber mats were measured by capillary flow porometry (Porolux 1000, Germany). All samples were wetted by a wetting liquid with a low surface tension of 16 dyne/cm before the measurements. Mean flow pore size (MFP), first bubble point (FBP) and PSD were measured by the wet-up/dry-up method.

2.9 Adsorption experiments

Since the reactive dyes are the most used dye class in the textile industry, the neat-PVA and PVA−BTCA nanofiber mats were used for the adsorption of a commercial reactive dye Reactive Red 141. Na₂SO₄ and Na₂CO₃ were added to the adsorption solution in order to simulate the dyeing condition. The kinetic studies were carried out in a 15 ml test tube containing 10 mg adsorbent and 10 ml of the adsorption solution with 45 g/l Na₂SO₄, 10 g/l Na₂CO₃ and 85 mg/l dye concentrations at pH 10.5–11. These tubes were stirred on an orbital shaker (Nüve, ST 402) at 25°C over a time interval of 0–48 h. The residual dye concentration in the liquid was analyzed using a UV–Vis spectrophotometer (Perkin Elmer Lambda 25) at 518 nm. The data were used to calculate the adsorption capacity (mg/g), q_e, of the adsorbent. The dye concentration on the adsorbent surface at equilibrium was calculated by

where C₀ is the initial dye concentration in the liquid phase (mg/l), C_e the equilibrium dye concentration (mg/l), V the total volume of dye solution used (l) and m the mass of adsorbent used (g).

2.10 Adsorption kinetics

Two of the most frequently used kinetic models that are the pseudo-first-order and the pseudo-second-order kinetic models were used for the characterization of the adsorption studies.

The pseudo-first-order kinetic model is given by the following equation:

By taking q_t=0 at t=0 and q_t=q_t at time t, the integrated form of Equation 4 becomes

where q_t is the amount of adsorbed dye at time t and k₁ is the rate constant of first-order adsorption. The parameters of the equation can be obtained from the linear plot of ln(q_e-q_t) versus t.

The pseudo-second-order kinetic model is given by the following equation:

where k₂ is the pseudo-second-order rate constant of adsorption. As in Equation 4, by taking q_t=0 at t=0 and q_t=q_t at time t, the integrated form of Equation 6 becomes

The plot of t/q versus t gives a straight line with a slope of 1/k2qe2 and an intercept of 1/q_e [14].

3.1 TGA of heat-fixated neat and BTCA cross-linked electrospun PVA nanofibers

In Figures 2–4, the TGA thermograms of 10 h produced neat-PVA and heat-fixated PVA nanofibers with and without BTCA are given according to fixation temperature. In these curves, the electrospun neat-PVA nanofibers lost ~7% mass at temperatures up to 100°C due to the evaporation of absorbed moisture. The decomposition of neat-PVA began around 270°C and lost half of its mass at 330°C, leaving merely ~2% residue at 600°C. Neat-PVA nanofibers had two degradation steps: the elimination reactions started at ~225°C and the breakdown of the polymer backbone (mostly chain-scission reactions) started at ~374°C as a second stage (residue weight: ~83% and ~28%, respectively) (Table 1) [33]. The temperatures of elimination reactions were approximately at 281, 269 and 262°C and the chain-scission reactions were approximately at 477, 468 and 461°C for PVA-130, PVA-150 and PVA-170 nanofibers, respectively. The starting point of elimination reaction almost stayed same, but heat fixation significantly increased the starting temperature of chain-scission reaction due to self-crosslinking with heat. This showed that heat crosslinking had a significant effect on chain scission. After the second stage, these nanofibers showed residue weight of approximately 6–9%. However, the final residue weight at 600°C was not affected from sole heat fixation. For PVA–BTCA-130, PVA–BTCA-150 and PVA–BTCA-170 nanofibers, elimination reactions started at approximately 225, 237 and 223°C and chain-scission reactions started at approximately 352, 352 and 325°C, leaving 46, 46 and 51% residue weight, respectively. These results confirm that an increase in thermal stability is a direct consequence of crosslinking with BTCA, and when compared to BTCA crosslinking is more effective than heat fixation self-crosslinking.

Figure 2

TGA thermograms of 10 h produced neat-PVA, PVA-130 and PVA–BTCA-130 nanofibers.

Figure 3

TGA thermograms of 10 h produced neat-PVA, PVA-150 and PVA–BTCA-150 nanofibers.

Figure 4

TGA thermograms of 10 h collected neat-PVA, PVA-170 and PVA–BTCA-170 nanofibers.

3.2 SEM images of PVA and PVA−BTCA nanofibers

The surface morphology of electrospun nanofibers of neat-PVA and PVA−BTCA was investigated by SEM imaging. Bead free, uniform PVA and PVA−BTCA nanofibers were produced (Figure 5). Mean fiber diameters of the neat-PVA and PVA− BTCA nanofibers were approximately 328.9 and 272.36 nm, respectively, and as expected, any change in nanofiber diameters was not observed depending on the collection time. Differences in the fiber diameters were analyzed by one-way ANOVA. It was observed that the difference between the neat-PVA and PVA−BTCA nanofiber diameters was statistically significant (p < 0.05). PVA−BTCA nanofibers were thinner than PVA nanofibers due to the conductivity of the BTCA-added solutions. The addition of BTCA significantly increases the solution conductivity because of the acidic nature [34].

Figure 5

SEM images of (a) neat-PVA and (b) PVA−BTCA nanofibers.

3.3 Thickness and weight per square meter of the neat-PVA and PVA−BTCA nanofibers

Table 2 lists the mean thickness and weight of the neat-PVA and PVA−BTCA nanofibers. Heat fixation did not have an effect on these data; thus, they were given according to the collection periods. As expected, the thickness and weight of the nanofiber mats were significantly affected by the collection period, and the thickness of the neat-PVA nanofiber mats was more than that of the PVA−BTCA nanofibers due to their thicker nanofiber diameter.

3.4 Swelling and weight loss of nanofibrous mats

Swelling and weight loss of the neat−PVA and PVA−BTCA nanofibers are given in Figure 6 according to the heat fixation temperatures. Without BTCA, the degree of swelling of PVA nanofibers was : was ~4000% for the heat fixation temperatures of 130 and 150°C, while the percentage of weight loss was as much as ~82–96%. However, when the heat fixation temperature was increased to 170°C, swelling of the nanofibers significantly decreased to 690% and weight loss to 15% due to the self-crosslinking of PVA with heat [32]. BTCA crosslinking led to a significant decrease in the degree of swelling at 150°C. It was ~650%, and weight loss of the PVA–BTCA-150 nanofibers was 8%. Slight swelling of the nanofibrous structure occurred with negligible weight loss. However, for PVA–BTCA-130 nanofibers, swelling of the mats was ~2500%, while weight loss was ~41%, which means 130°C was not enough for crosslinking. Swelling and weight loss of PVA–BTCA-170 nanofibers were close to those of PVA-150 nanofibers and showed that they were cross-linked properly.

Figure 6

Swelling and weight loss of 10 h collected neat-PVA and PVA−BTCA nanofibers after heat fixation at 130, 150 and 170°C.

3.5 Pore size analysis

To compare the pore size of the neat-PVA and PVA−BTCA nanofibers mats according to their collection period, the vertical capillary method was performed by capillary flow porometry. It provides a better understanding of water penetration on the surface since this method takes the whole structure of the nanofibers into account [35].

The FBP, the MFP and the smallest pore size are given in Table 3. A decrease in the mean nanofiber diameter of PVA−BTCA nanofibers did not cause a significant effect on the pore sizes; they were close to that of the neat−PVA nanofibers. The FBP was in the range of 2.01–2.45 and 2.05–2.34, and the MFP was in the range of 1.05–1.33 and 1.26–1.32 for the neat-PVA and PVA−BTCA nanofibers, respectively. The effect of the electrospun membrane thickness on the pore size can be explained by depositing more fibers on the substrate, which results in smaller pores when other variables (such as fiber diameter and fiber length) are kept constant. However, the mean pore size reached an asymptotic level with the increase in mat thickness [36]. In our case, the membrane thicknesses of both the neat−PVA and PVA−BTCA nanofibers were too high and did not play an important role in reducing the pore size. There was not any statistical difference (p < 0.05). According to these results, to achieve smaller pore sizes, there is no need to extent the collection period to 5 or 10 h. In all, 3 h of collections with thickness of 0.029–0.079 mm is enough for approximately 1 μm MFP.

3.6 Adsorption results

In this study, 3, 5 and 10 h collected neat-PVA and PVA− BTCA nanofiber mats and heat was fixated at 130, 150 and 170°C. Then, these nanofibrous mats were evaluated for the color removal of Reactive Red 141. The dye adsorption onto the neat-PVA and PVA−BTCA nanofiber mats was studied as a function of contact time, and the adsorption results are presented in Figure 7.

Figure 7

Amounts of the dye adsorbed versus contact time for PVA and PVA–BTCA nanofibers produced in different collection times and fixation temperatures.

The adsorption capacity (q) increased in time but reached the plateau at the same time (approximately 4–6 h) for different collection times and fixation temperatures. The maximum equilibrium adsorption capacities of the nanofiber membranes after 48 h were equal to 84.36 and 87.40 mg/g for 150°C heat-fixated neat-PVA and PVA−BTCA nanofibers, respectively.

It was observed that there was no significant difference between the neat-PVA and PVA−BTCA nanofiber mats in mean values of adsorption properties. The results of adsorption experiments also showed that collection periods did not significantly affected the adsorbed amount of the dye, they were close to each other. Thus, adsorption capabilities of PVA nanofibers did not rely on the thickness of the PVA mats, and 3 -h collection period was enough for the dye adsorption. However, it was seen that fixation temperature had a significant effect on the dye adsorption. The adsorbed amount of the dye reached the highest and the lowest adsorption rates at 150°C and 130°C, respectively. According to these results, 130°C was not enough and 170°C was too much for the higher adsorption amounts. In all, 150°C of fixation temperature was the most suitable temperature for both the neat-PVA and PVA−BTCA nanofiber mats. These results were also noticeable in the color of the dye solutions after the adsorption process (Figure 8).

Figure 8

Images of the dye baths before and after the adsorption process. *Untreated dye solution (a). The dye solutions after the adsorption with 3 h (b), 5 h (c) and 10 h (d) collected and heat-fixated neat-PVA nanofibers; The dye solutions after the adsorption with 3 h (e), 5 h (f) and 10 h (g) collected and heat-fixated PVA−BTCA nanofibers.

In order to evaluate the morphology of the nanofibers after the adsorption process, SEM images of the nanofibers were taken. In Figure 9, PVA-130 nanofibers were not stable in water; they almost lost their fibrous structure. With the increasing temperature, PVA nanofibers had better preserved their fibrous structure. Although in case of 150°C, PVA-150 nanofiber mats shrunk and nanofibers were entangled. PVA-170 nanofibers were still straight and in a nanofiber form when compared to PVA-150. Adsorption of dye aggregates and particles of Na₂SO₄ and Na₂CO₃ could be also seen for both PVA-150 and PVA-170 nanofibers. On the other hand, BTCA crosslinking significantly improved the water stability of the PVA nanofibers when compared to that of heat-fixated neat-PVA nanofibers. Even at 130°C, nanofibers preserved their fibrous structure. In all three cases, adsorption of dye aggregates and particles of Na₂SO₄ and Na₂CO₃ could be seen on the surfaces of nanofibers. However, it could be clearly seen from the SEM images that dye aggregates and particles of Na₂SO₄ and Na₂CO₃ were also penetrated inside the fibrous structure, so it could be said that preserving the nanofibrous structure has an impact on adsorption of these aggregates and particles.

Figure 9

SEM images of the heat-fixated neat−PVA and PVA−BTCA nanofibers after adsorption studies.

It was clear from the SEM images of the nanofibers after the adsorption process (Figure 9) and weight loss test results (Figure 6) that PVA nanofibers were not stable to water at 130 and 150°C. Thus, to remove the color from the wastewater, PVA nanofibers could be used but PVA nanofibers should be chemically cross-linked properly. According to these results, 150°C heat-fixated PVA–BTCA nanofibers mats are convenient for this application.

The pseudo-first-order kinetic equation (Equation 3) and pseudo-second-order kinetic equation (Equation 5) were also applied to the adsorption data. The parameters of the kinetic models and correlation coefficients are given in Tables 4 and 5. The values of the correlation coefficient implied that the adsorption of the dye on the neat-PVA and PVA−BTCA nanofibers was described by a pseudo-second-order equation.