PP/TiO2 Melt-Blown Membranes for Oil/Water Separation and Photocatalysis: Manufacturing Techniques and Property Evaluations

This study aims to produce polypropylene (PP)/titanium dioxide (TiO2) melt-blown membranes for oil/water separation and photocatalysis. PP and different contents of TiO2 are melt-blended to prepare master batches using a single screw extruder. The master batches are then fabricated into PP/TiO2 melt-blown membranes. The thermal properties of the master batches are analyzed using differential scanning calorimetry and thermogravimetric analysis, and their particle dispersion and melt-blown membrane morphology are evaluated by scanning electron microscopy. TiO2 loaded on melt-blown membranes is confirmed by X-ray diffraction (XRD). The oil/water separation ability of the melt-blown membranes is evaluated to examine the influence of TiO2 content. Results show that the thermal stability and photocatalytic effect of the membranes increase with TiO2 content. TiO2 shows a good dispersion in the PP membranes. After 3 wt.% TiO2 addition, crystallinity increases by 6.4%, thermal decomposition temperature increases by 25 °C compared with pure PP membranes. The resultant PP/TiO2 melt-blown membrane has a good morphology, and better hydrophobicity even in acetone solution or 6 h ultraviolet irradiation, and a high oil flux of about 15,000 L·m−2·h−1. Moreover, the membranes have stabilized oil/water separation efficiency after being repeatedly used. The proposed melt-blown membranes are suitable for mass production for separating oil from water in massively industrial dyeing wastewater.


Introduction
Currently, water pollution problems have given rise to strong repercussions in society, particularly for oily wastewater pollution. This considerable amount of oily wastewater usually comes from textile, petrochemical, and steel factories. Sometimes, in the marine transportation, oil spills happen by leakage, which pollutes the ocean and causes oily wastewater pollution [1,2]. Therefore, how to deal with these oily wastewater pollutions and to achieve oil/water separation have attracted the attention of researchers.
In the industry, oil/water mixtures can be purified by some mechanical devices, such as skimmers, air flotation, centrifugation, booms, and chemical coagulation. However, these purifying methods need to input energy and high pressure to operate [3,4]. Porous materials such as foam [5,6] and textiles [7,8] are often used to deal with oil leakage by adsorption. However, water and oil are both adsorbed and the separation efficiency is poor. Moreover, recovery is difficult and time consuming and even leads to secondary environmental pollution [9]. To address these issues, many scholars coated thermal- [10] or pH- [11,12] responsive materials on metallic mesh to improve oil/water separation ability and to provide different applications. Similarly, titanium meshes [13] and porous polyethylene (PE) meshes [14] that are modified with poly(3,4-ethylenedioxythiophene) -poly(styrenesulfonate) have high resistance to the harshest chemical conditions. Metallic meshes have shown promising results for the selective separation of oil/water mixtures. However, such meshes have disadvantages such as heavy weight, high material cost, and poor corrosion resistance impeding practical applications. By contrast, textiles/fabric based substrates are low cost, light weight, flexible, and corrosion resistant. Several studies focused on nanofiber membranes and their application in the separation of oil/water systems. For example, electrospun mats composed of polydimethylsiloxane (PDMS)-block-poly (4-vinylpyridine) polymers show excellent pH switchability for oil-water separation [15]. Electrospun membranes composed of a fluorine-containing polymer exhibit self-healing ability by self-cleaning, which is ascribed to the low-surface-energy of fluorine-containing polymer [16]. However, this fluorine-containing polymer has environmental damage. Preferably, TiO 2 not only can decompose the organic pollutants on the membranes, but also demonstrate light-induced oil resistance and self-cleaning function [17,18]. As a photocatalyst, TiO 2 decomposes organic pollutants via light-induced oxidative holes and hydroxyl radicals [19,20]. Comparatively, TiO 2 is commonly used in the photocatalytic degradation of organic pollutants because of its superior advantages, such as low production cost, high stability, and good photocatalysis [21]. TiO 2 also commonly used in oil/water separation. For example, depositing TiO 2 on stainless steel meshes [22,23] or titanium meshes [18] can produce oil/water-separating membranes. A good prospect in separating oil from water effectively is achieved by adding TiO 2 with meshes or membranes. However, these methods still have some limitations. Even under low water resistance, TiO 2 and meshes present a low adhesion between them. Therefore, ideal photocatalyst-immobilized substrate materials should be selected for stable anchoring to prevent catalyst leaching, maintain reactive oxygen stability, and increase selective affinity toward target contaminants [24].
To overcome the aforementioned shortcomings of metal mesh and electrospinning, previous studies employed an effective melt-blown method to form microfiber membranes and improve TiO 2 adhesion [25][26][27]. As a commonly available low-cost melt-blown substrate, polypropylene (PP) exhibits efficient air filtration performance and water purification [28][29][30][31]. Liu et al. [28] found that melt-blown PP membrane significantly decreased the amount of chemical oxygen demand (COD), suspended solids (SS), and NH 3 -N in the reclaimed water. Moreover, some particles, such as nanoclay [29], silicon dioxide [30], and tourmaline particles [31] can be added in the melt-blowing process in order to create additional new function in a greater diversity application. In addition, melt-blending process is an efficient method to produce massive functional matrix, featuring high production efficiency, less pollution, and ease of processing. Moreover, some particles, such as nanoclay [29], silicon dioxide [30], and tourmaline particles [31], can be added in the melt-blowing process to create new functions for different applications. Melt-blending is an efficient method to produce massive functional matrices featuring high production efficiency, less pollution, and ease of processing. Therefore, in this study, a single-screw melting blending process is first used to prepare PP/TiO 2 blends, and then photocatalytic Co., Ltd.), and organic solvents (petroleum ether, n-hexane, acetone) were purchased from Aladdin Chemistry Co. Ltd., Shanghai, China. All chemicals were of analytical and used as received without further purification.

Preparation of Melt-Blown Membranes
PP and TiO 2 are mixed through several physical processes at mass ratios of 100/0, 100/1, 100/3, and 100/5, melted, extruded, and then pelletized for seven times using a single-screw extruder (Qingdao Keshengda Plastic Machinery, Qingdao, China) forming a PP/TiO 2 master batch. PP/TiO 2 master batches are prepared after seven cycles, and the screw and pelletizing speed ratio is 1:1.52. The processing parameters of the melted blends are shown in Table 1, and the prepared master batches are named PP/TiO 2 -1, PP/TiO 2 -3, and PP/TiO 2 -5 depending on the different amounts of TiO 2 . Some master batches are then poured into a twin-screw mixer at 180 • C for 3 min at a rotary speed of 130 rpm. The obtained blends are then hot-pressed at 160 • C for 2 min under a pressure of 20 MPa to obtain PP/TiO 2 hot-pressed films. The hot-pressed films are named F-PP/TiO 2 -0, F-PP/TiO 2 -1, F-PP/TiO 2 -3, and F-PP/TiO 2 -5 depending on the different amounts of TiO 2 .
The master batches are dried at 80 • C for 10 h and then poured into a melt-blowing machine (Tianjin Shengruiyuan Machinery Technology, Tianjin, China) with a die temperature of 180-230 • C (Table 1), a hot air temperature of 180 • C, and a pressure of 0.04 MPa. The melt-blowing process is shown in Figure 1

Measurements and Characterizations
The thermal properties of the PP/TiO2 master batches are analyzed using differential scanning calorimetry (DSC 209F3, NETZSCH, Bavaria, Germany). All samples are heated from 25 °C to 200 °C at increments of 10 °C/min and then stored at 200 °C for 3 min to eliminate the thermal

Measurements and Characterizations
The thermal properties of the PP/TiO 2 master batches are analyzed using differential scanning calorimetry (DSC 209F3, NETZSCH, Bavaria, Germany). All samples are heated from 25 • C to 200 • C at increments of 10 • C/min and then stored at 200 • C for 3 min to eliminate the thermal history. Afterward, the samples are cooled to 25 • C at 10 • C/min, during which the cooling curves are recorded. Samples are again heated to 200 • C at increments of 10 • C/min, during which the heating curves are recorded. All experiments are performed under nitrogen atmosphere. Thermogravimetric is conducted with a thermogravimetric analyzer (TG209F3, NETZSCH, Bavaria, Germany) with dry nitrogen gas at a flow rate of 60 mL/min. The relative mass losses of the PP/TiO 2 master batches are recorded from 50 • C to 700 • C at a heating rate of 20 • C/min. The fractured surface morphology of the F-PP/TiO 2 membranes and the surface morphology of the M-PP/TiO 2 membranes are analyzed by scanning electron microscopy (SEM, TM3030, HITACHI, Tokyo, Japan) at an accelerating voltage of 3 KV. Prior to the test, the F-PP/TiO 2 hot-press film samples are brittle fractured with liquid nitrogen, and then all samples are sputtered with gold to provide enhanced conductivity. The SEM images are analyzed using Image-Pro Plus 6.0 software. A bundle of fibers per image are used to measure the diameter of the nanofibers, and Origin is used to plot the diameter distribution and compute the standard deviations. The contact angles of the M-PP/TiO 2 melt-blown membranes are measured using a tester JC2000DM (Powereach, Shanghai Zhongchen Digital Technology Apparatus, Shanghai, China). The contact angles at five sites of the sample are measured and recorded.

Oil/Water Separation and Photocatalytic Activity Analysis
The M-PP/TiO 2 melt-blown membrane is affixed between two glass vessels. To distinguish, water is dyed with methylene blue, and organic solvents (i.e., kerosene, hexane, petroleum ether, and toluene) are dyed with oil red O. Water (30 mL) and oil (30 mL) are blended, and the mixture is gently poured into the container. Oil and water are separated under gravity, and the water/oil separation efficiency is computed using Equation (1) [32]: where V 0 and V 1 correspond to the volume of oil before and after the separation, respectively. Oil that adheres to the vessel is not included in the calculation. The photocatalytic activities of the M-PP/TiO 2 melt-blown membranes are evaluated using rhodamine B and kerosene. Rhodamine B has an initial concentration of 10 mg/L. Moreover, 5 mg/30 mL of TiO 2 is used as the photocatalyst. Photocatalytic degradation of Rhodamine B is conducted using a photochemical reaction apparatus (XPA-VII photochemical reactor, Xujiang electromechanical, Nanjing, China). Ultraviolet light irradiation is sourced from a 300 W Hg lamp that resides in a double glazing jacket where circulating cool water flows. Then, a specified volume of solution is extracted at different times. The solution is first centrifuged, after which the supernatant liquor is taken and the absorbance is recorded at 554 nm using a UV-Vis spectrophotometer (Mapada, UV-1800PC, Shanghai, China). Rhodamine B has the maximum absorbance wavelength of 554 nm.
The decomposition of kerosene is analyzed using Fourier transform infrared spectroscopy (FTIR, Nicolet iS10, Waltham, MA, USA). PP/TiO 2 melt-blown membranes that are immersed in kerosene are exposed to a 300 W radiation from a high-pressure Hg lamp for 3 h. A specified volume of kerosene is collected and then scanned in the range of 400-4000 cm −1 using a FTIR instrument. Differences in the plotted spectrum are compared to evaluate the photocatalytic capabilities of the membranes.  Figure 2a,b shows the DSC cooling and heating curves of the PP/TiO 2 master batches after eliminating the thermal history. Table 2 shows the melt crystallization temperature (T mc ), the glass transition temperature (T g ), the melt temperature (T m ), and the degree of crystallinity (X c ) of the PP/TiO 2 master batches with different TiO 2 contents. The T g of pure PP is 41 • C, which is heightened with the presence of TiO 2 . However, the content of TiO 2 only has a marginal positive influence. During the cooling process, the master batches exhibit a cold crystallization peak. The cold crystallization peak increases with TiO 2 content, indicating that the crystallization rate of the PP/TiO 2 master batch. This result suggests that using TiO 2 accelerates the crystallinity of PP. The reason is that TiO 2 inorganic particles can serve as a heterogeneous nucleation agent. Thus, polymer crystals have changed from homogeneous nucleation to heterogeneous nucleation, facilitating the crystallization. The crystallinity of the PP/TiO 2 master batches is proportional to the content of TiO 2 . The degree of crystallinity (X c ) of the samples is calculated using the following equation:

Thermal Behaviors and Thermal Stability of the PP/TiO 2 Master Batch
where ∆H m is the melting enthalpy, ∆H cc is the cold crystallization enthalpy, and ∆H 0 is the fusion enthalpy of the completely crystalline PP [33]. When the TiO 2 content is 3 wt. %, the PP/TiO 2 -3 master batch has a crystallinity of 42.69%, which is higher than that of PP/TiO 2 -0 (36.29%). Nevertheless, greater TiO 2 content is not beneficial to the crystallinity of PP, which suggests that increasing TiO 2 to a certain extent decreases crystallization. As a result, the crystallinity remains the same and the heterogeneous nucleation agent is saturated. The Thermogravimetric analysis (TGA) and differential thermogravimetry analysis (DTG) curves of the PP/TiO 2 master batches are related to the content of TiO 2 as shown in Figure 2c,d. The results of thermogravimetric show that the TGA curves of the PP/TiO 2 master batches shift toward high temperature as the TiO 2 content is increased. Table 3 shows the initial decomposition temperature (T 0.05 ), a weight loss 50% temperature (T 0.5 ), remains mass, and "T max " is the decomposition temperature corresponding to the weight loss rate obtained from the DTG curve in Figure 2d. The PP/TiO 2 master batches decomposition temperature increases with increasing TiO 2 content. The T 0.5 of PP/TiO 2 -0 is 350 • C, and that of the PP/TiO 2 master batch containing 5 wt. % TiO 2 is 400 • C. This finding demonstrates that the presence of TiO 2 increases the decomposition temperature and improves the thermal stability of PP/TiO 2 . This result is due to the fact that TiO 2 can enhance the interaction between polymer molecular chains, which consequently increases the degree of crosslinking and restricts the segmental motion. The thermal decomposition of PP molecular chains is terminated, while the distribution of TiO 2 in the PP polymer limits the transmission of heat energy [34]. In addition, the PP/TiO 2 master batches have high remnant mass than PP/TiO 2 -0, and the remaining amount is almost equivalent to the content of TiO 2 . The greater TiO 2 content generates the higher remnant mass. degree of crosslinking and restricts the segmental motion. The thermal decomposition of PP molecular chains is terminated, while the distribution of TiO2 in the PP polymer limits the transmission of heat energy [34]. In addition, the PP/TiO2 master batches have high remnant mass than PP/TiO2-0, and the remaining amount is almost equivalent to the content of TiO2. The greater TiO2 content generates the higher remnant mass.

Morphology of PP/TiO2 Hot-Press Film and Melt-Blown Membranes
The fractured surface of the PP/TiO2 hot-press film and the surface morphology of the PP/TiO2 melt-blown membranes are shown in Figure 3a-d. The fractured surface of PP/TiO2-0 is continual and homogeneous. By contrast, the PP/TiO2 hot-press film has a rugged fractured surface where the white particles are TiO2. During melt-blending, some agglomerated particles can be separated by a strong shear force effectively [35]. However, increasing the content of TiO2 leads to agglomeration in the polymer matrix, and the agglomeration becomes more apparent with increasing TiO2 content, suggesting that the adhesion between PP and TiO2 is low. Some of the TiO2 particles fall apart from the membranes because of the brittle fracture caused by liquid nitrogen, leaving some empty holes. This phenomenon is more exaggerated as the TiO2 content is increased Figure 3e-h shows the fiber morphology of the PP/TiO2 melt-blown membranes with the corresponding fiber diameter distribution (Figure 3i-l). Table 4 shows the average diameter and standard deviation of the membranes. M-PP/TiO2-0 is composed of fibers with an average diameter of about 2 µm and has a smooth surface. With the addition of TiO2, M-PP/TiO2 is composed of

Morphology of PP/TiO 2 Hot-Press Film and Melt-Blown Membranes
The fractured surface of the PP/TiO 2 hot-press film and the surface morphology of the PP/TiO 2 melt-blown membranes are shown in Figure 3a-d. The fractured surface of PP/TiO 2 -0 is continual and homogeneous. By contrast, the PP/TiO 2 hot-press film has a rugged fractured surface where the white particles are TiO 2 . During melt-blending, some agglomerated particles can be separated by a strong shear force effectively [35]. However, increasing the content of TiO 2 leads to agglomeration in the polymer matrix, and the agglomeration becomes more apparent with increasing TiO 2 content, suggesting that the adhesion between PP and TiO 2 is low. Some of the TiO 2 particles fall apart from the membranes because of the brittle fracture caused by liquid nitrogen, leaving some empty holes. This phenomenon is more exaggerated as the TiO 2 content is increased Figure 3e-h shows the fiber morphology of the PP/TiO 2 melt-blown membranes with the corresponding fiber diameter distribution (Figure 3i-l). Table 4 shows the average diameter and standard deviation of the membranes. M-PP/TiO 2 -0 is composed of fibers with an average diameter of about 2 µm and has a smooth surface. With the addition of TiO 2 , M-PP/TiO 2 is composed of fibers with different diameters, and some fibers are coalesced obviously, which is ascribed to the presence of TiO 2 . During melt-blowing, TiO 2 of a small particle size is enwrapped in the fibers, whereas some agglomerated TiO 2 particles protrude the fibers and are responsible for the rough fiber surface. fibers with different diameters, and some fibers are coalesced obviously, which is ascribed to the presence of TiO2. During melt-blowing, TiO2 of a small particle size is enwrapped in the fibers, whereas some agglomerated TiO2 particles protrude the fibers and are responsible for the rough fiber surface.   vibrations, respectively; and those at 1459 and 1377 cm −1 correspond to -CH 2 -bending vibration and -CH 3 symmetric deformation vibration, respectively. No additional XRD and FTIR characteristic peaks are found, which shows that the incorporation of TiO 2 cannot change the crystal structure and chemical groups of the membrane compared with M-PP/TiO 2 -0. vibrations, respectively; those at 2873 and 2840 cm −1 correspond to the -CH3 and -CH2-symmetric stretching vibrations, respectively; and those at 1459 and 1377 cm −1 correspond to -CH2-bending vibration and -CH3 symmetric deformation vibration, respectively. No additional XRD and FTIR characteristic peaks are found, which shows that the incorporation of TiO2 cannot change the crystal structure and chemical groups of the membrane compared with M-PP/TiO2-0.

Oil/Water Separation of PP/TiO2 Melt-Blown Membranes
A high selective wettability for oil and water is another crucial factor for oil/water separation materials [36]. The wettability of M-PP/TiO2 is determined by the water contact angle measurements. Figure 5a shows the contact angle observation of the membranes, and the accurate contact angle is shown in Table 5. M-PP/TiO2-0 has a contact angle of 140°, representing good hydrophobicity. Although high TiO2 contents decrease the water contact angle, it remains beyond 130°. Therefore, the membranes possess hydrophobicity, and the presence of TiO2 does not impact their wettability. Moreover, the oil (i.e., kerosene) and organic solvent (i.e., petroleum ether, n-hexane, and acetone) contact angle of the membranes is 0° (Figure 5b), suggesting super lipophilicity. The contact angle of the acetone drops on the filtration membrane gradually increases with increasing TiO2 content. When the ratio of acetone and water is 3:2, the contact angle of acetone solution in M-PP/TiO2-0 is 20°, whereas that in M-PP/TiO2-3 is 98°. The surface tension of the acetone-simulated organic pollutant solution is measured using a liquid surface tension tester (BZY-3B, Shanghai Hengping Instrument and Meter Factory) to confirm the interaction of TiO2 in the membrane and operating environment at organic pollutants. Figure 5c shows the surface tension of different acetone solutions. The surface tension of the solution decreases gradually with increasing acetone concentration, whereas the surface tension of the membranes decreases gradually with increasing TiO2 content to meet the different surface tensions of liquid filtration.
According to Young equation, the wetting angle of a droplet is related to the surface tension of the solid-gas, liquid-gas, and liquid-solid interfaces. For the same droplet, the liquid surface tension is constant. The greater the contact angle, the greater the difference between the surface tension of the solid-gas and solid-liquid interfaces. As a result, the introduction of TiO2 changes the surface tension of the membranes.
where is the surface tension of the solid-gas interface, is the surface tension of the solidliquid interface, is the surface tension of the liquid-gas interface, and is the contact angle.

Oil/Water Separation of PP/TiO 2 Melt-Blown Membranes
A high selective wettability for oil and water is another crucial factor for oil/water separation materials [36]. The wettability of M-PP/TiO 2 is determined by the water contact angle measurements. Figure 5a shows the contact angle observation of the membranes, and the accurate contact angle is shown in Table 5. M-PP/TiO 2 -0 has a contact angle of 140 • , representing good hydrophobicity. Although high TiO 2 contents decrease the water contact angle, it remains beyond 130 • . Therefore, the membranes possess hydrophobicity, and the presence of TiO 2 does not impact their wettability. Moreover, the oil (i.e., kerosene) and organic solvent (i.e., petroleum ether, n-hexane, and acetone) contact angle of the membranes is 0 • (Figure 5b), suggesting super lipophilicity. The contact angle of the acetone drops on the filtration membrane gradually increases with increasing TiO 2 content. When the ratio of acetone and water is 3:2, the contact angle of acetone solution in M-PP/TiO 2 -0 is 20 • , whereas that in M-PP/TiO 2 -3 is 98 • . The surface tension of the acetone-simulated organic pollutant solution is measured using a liquid surface tension tester (BZY-3B, Shanghai Hengping Instrument and Meter Factory) to confirm the interaction of TiO 2 in the membrane and operating environment at organic pollutants. Figure 5c shows the surface tension of different acetone solutions. The surface tension of the solution decreases gradually with increasing acetone concentration, whereas the surface tension of the membranes decreases gradually with increasing TiO 2 content to meet the different surface tensions of liquid filtration.
According to Young equation, the wetting angle of a droplet is related to the surface tension of the solid-gas, liquid-gas, and liquid-solid interfaces. For the same droplet, the liquid surface tension is constant. The greater the contact angle, the greater the difference between the surface tension of the solid-gas and solid-liquid interfaces. As a result, the introduction of TiO 2 changes the surface tension of the membranes.
γ sg − γ sl = γ lg cosθ, where γ sg is the surface tension of the solid-gas interface, γ sl is the surface tension of the solid-liquid interface, γ lg is the surface tension of the liquid-gas interface, and θ is the contact angle. M-PP/TiO2 exhibits both hydrophobicity and super lipophilicity and thus can be applied to oil/water separation. A simple oil/water separation equipment is assembled, and its separation process is shown in Figure 6. The membrane is affixed between two glass containers, after which the water/kerosene mixture is poured into the upper container slowly. With the help of gravity, kerosene can permeate through the membrane and reach the lower glass container, during which water is intercepted and stays over the melt-blown membrane. No other external forces are applied. As a result, oil is separated from water effectively, which proves that the separation process can be easily conducted with a low energy cost. M-PP/TiO 2 exhibits both hydrophobicity and super lipophilicity and thus can be applied to oil/water separation. A simple oil/water separation equipment is assembled, and its separation process is shown in Figure 6. The membrane is affixed between two glass containers, after which the water/kerosene mixture is poured into the upper container slowly. With the help of gravity, kerosene can permeate through the membrane and reach the lower glass container, during which water is intercepted and stays over the melt-blown membrane. No other external forces are applied. As a result, oil is separated from water effectively, which proves that the separation process can be easily conducted with a low energy cost. Polymers 2018, 10, x FOR PEER REVIEW 10 of 13 Afterward, the oil/water separation performance of M-PP/TiO2 is computed to obtain the oilwater separation efficiency (ηs). Figure 7a shows that the separation efficiency of the membranes is influenced by TiO2 content. The separation efficiency is between 95%-98%, suggesting effective oil/water separation. Noticeably, the membranes retain their oil/water separation capability even after 100 repeated tests. The test result proves that the proposed membranes have a high stability (Figure 7b). Importantly, the separation efficiency is not dependent on the TiO2 content when the test is conducted under gravity. The oil flux (F) can be yielded using the following Equation (3): where V is the volume of kerosene (i.e., V = 50 mL), S is the effective geometric area (1256 mm 2 ), and t is the time required. The oil flux of the PP/TiO2 membranes is 14,789-15,410 L·m −2 ·h −1 as shown in Figure 7c, which is higher than those of the stainless filtering net-based membranes and electrospinning membranes [37].  Afterward, the oil/water separation performance of M-PP/TiO 2 is computed to obtain the oil-water separation efficiency (η s ). Figure 7a shows that the separation efficiency of the membranes is influenced by TiO 2 content. The separation efficiency is between 95-98%, suggesting effective oil/water separation. Noticeably, the membranes retain their oil/water separation capability even after 100 repeated tests. The test result proves that the proposed membranes have a high stability (Figure 7b). Importantly, the separation efficiency is not dependent on the TiO 2 content when the test is conducted under gravity. The oil flux (F) can be yielded using the following Equation (3): where V is the volume of kerosene (i.e., V = 50 mL), S is the effective geometric area (1256 mm 2 ), and t is the time required. The oil flux of the PP/TiO 2 membranes is 14,789-15,410 L·m −2 ·h −1 as shown in Figure 7c, which is higher than those of the stainless filtering net-based membranes and electrospinning membranes [37]. Afterward, the oil/water separation performance of M-PP/TiO2 is computed to obtain the oilwater separation efficiency (ηs). Figure 7a shows that the separation efficiency of the membranes is influenced by TiO2 content. The separation efficiency is between 95%-98%, suggesting effective oil/water separation. Noticeably, the membranes retain their oil/water separation capability even after 100 repeated tests. The test result proves that the proposed membranes have a high stability ( Figure 7b). Importantly, the separation efficiency is not dependent on the TiO2 content when the test is conducted under gravity. The oil flux (F) can be yielded using the following Equation (3): where V is the volume of kerosene (i.e., V = 50 mL), S is the effective geometric area (1256 mm 2 ), and t is the time required. The oil flux of the PP/TiO2 membranes is 14,789-15,410 L·m −2 ·h −1 as shown in Figure 7c, which is higher than those of the stainless filtering net-based membranes and electrospinning membranes [37].   Figure 8 shows the photocatalysis of the F-PP/TiO 2 melt-blown membranes. Rhodamine B solution almost does not decompose under UV light irradiation [38]. Therefore, rhodamine B is selected as a model organic pollutant to study the photocatalytic performance of the F-PP/TiO 2 membranes. Figure 8a shows the photocatalysis of the melt-blown membrane-decomposed rhodamine B solution. The melt-blown membranes composed of high TiO 2 amounts demonstrate greater photocatalysis, which is lower than that of pure TiO 2 . This phenomenon can be attributed to two reasons. For one thing, the melt-blown membranes are hydrophobic, and the polluted sample and photocatalyst may not have sufficient contact during the photocatalysis. For another, some TiO 2 particles are enwrapped in the PP melt-blown membranes or fibers, which deprives TiO 2 of its possibility to directly contact with light sources. Therefore, the membranes demonstrate a low photocatalysis.

Photocatalysis of PP/TiO 2 Melt-Blown Membranes
Kerosene is used as the contaminant to testify the major factor affecting the photocatalysis of the membranes. The membranes possess a good lipophilic property and thus have good wettability. Figure 8b shows the spectrum of kerosene before/after photocatalysis. Kerosene demonstrates C-H stretching at 3000-2850 cm −1 and C-H vibration at 1465-1340 cm −1 , which indicates strong characteristic peak. When the TiO 2 content is increased, the intensity of the characteristic peak gradually declines. The photocatalysis of the PP/TiO 2 melt-blown membranes is dependent on the content of TiO 2 distributed over the fibers. When the TiO 2 content is 3% or 5%, the photocatalytic effect is similar, which proves that the TiO 2 content in the membranes is similar. Combined with the SEM image (Figure 3), when the TiO 2 content is 3%, the membranes show higher photocatalytic effect and better morphology.  Figure 8 shows the photocatalysis of the F-PP/TiO2 melt-blown membranes. Rhodamine B solution almost does not decompose under UV light irradiation [38]. Therefore, rhodamine B is selected as a model organic pollutant to study the photocatalytic performance of the F-PP/TiO2 membranes. Figure 8a shows the photocatalysis of the melt-blown membrane-decomposed rhodamine B solution. The melt-blown membranes composed of high TiO2 amounts demonstrate greater photocatalysis, which is lower than that of pure TiO2. This phenomenon can be attributed to two reasons. For one thing, the melt-blown membranes are hydrophobic, and the polluted sample and photocatalyst may not have sufficient contact during the photocatalysis. For another, some TiO2 particles are enwrapped in the PP melt-blown membranes or fibers, which deprives TiO2 of its possibility to directly contact with light sources. Therefore, the membranes demonstrate a low photocatalysis.

Photocatalysis of PP/TiO2 Melt-Blown Membranes
Kerosene is used as the contaminant to testify the major factor affecting the photocatalysis of the membranes. The membranes possess a good lipophilic property and thus have good wettability. Figure 8b shows the spectrum of kerosene before/after photocatalysis. Kerosene demonstrates C-H stretching at 3000-2850 cm −1 and C-H vibration at 1465-1340 cm −1 , which indicates strong characteristic peak. When the TiO2 content is increased, the intensity of the characteristic peak gradually declines. The photocatalysis of the PP/TiO2 melt-blown membranes is dependent on the content of TiO2 distributed over the fibers. When the TiO2 content is 3% or 5%, the photocatalytic effect is similar, which proves that the TiO2 content in the membranes is similar. Combined with the SEM image (Figure 3), when the TiO2 content is 3%, the membranes show higher photocatalytic effect and better morphology.

Conclusion
In this study, TiO2 and PP are melt-blended to form PP/TiO2 master batches using a single-screw extruder, after which the master batches are prepared into PP/TiO2 melt-blown membranes. The thermal properties of PP/TiO2 master batches and the melt-blown membrane morphology are characterized. DSC shows that the presence of TiO2 accelerates crystallization rate, providing the master batch with high crystallinity. Similarly, TGA results show that the presence of TiO2 enhances the thermostability of PP, and the thermal decomposition temperature is proportional to the content of TiO2 and reaches the required temperature of 180-230 °C. SEM images show that TiO2 is evenly dispersed in the PP matrices due to multiple melt-blending processes. In addition, TiO2 provides the fibers with a large diameter and a rough surface, which makes the fiber web fluffy. TiO2 also contributes photocatalytic activity to the membranes and a great amount of kerosene flux, leading to high water/oil separation efficiency. Moreover, the membranes can be used repeatedly and maintain stabilized filtering efficacy. The membranes remain stable after 6 h of ultraviolet irradiation, indicating that TiO2 does not degrade PP. The proposed melt-blown membranes are suitable for mass production for separating oil from water in massively industrial dyeing wastewater.

Conclusions
In this study, TiO 2 and PP are melt-blended to form PP/TiO 2 master batches using a single-screw extruder, after which the master batches are prepared into PP/TiO 2 melt-blown membranes. The thermal properties of PP/TiO 2 master batches and the melt-blown membrane morphology are characterized. DSC shows that the presence of TiO 2 accelerates crystallization rate, providing the master batch with high crystallinity. Similarly, TGA results show that the presence of TiO 2 enhances the thermostability of PP, and the thermal decomposition temperature is proportional to the content of TiO 2 and reaches the required temperature of 180-230 • C. SEM images show that TiO 2 is evenly dispersed in the PP matrices due to multiple melt-blending processes. In addition, TiO 2 provides the fibers with a large diameter and a rough surface, which makes the fiber web fluffy. TiO 2 also contributes photocatalytic activity to the membranes and a great amount of kerosene flux, leading to high water/oil separation efficiency. Moreover, the membranes can be used repeatedly and maintain stabilized filtering efficacy. The membranes remain stable after 6 h of ultraviolet irradiation, indicating that TiO 2 does not degrade PP. The proposed melt-blown membranes are suitable for mass production for separating oil from water in massively industrial dyeing wastewater.