Photocatalytic Degradation of Humic Acids with Titanium Dioxide Embedded into Polyethylene Pellets to Enhance the Postrecovery of Catalyst

Abstract

Heterogeneous photocatalysis with titanium dioxide (TiO₂) has been shown to be a sustainable treatment technology to remove natural organic matter (NOM). This is of interest to the drinking water treatment industry. However, heterogeneous photocatalysis with TiO₂ continues to suffer from several limitations such as the postrecovery of TiO₂ particles, which impede its use in the drinking water treatment industry. In this study, the repeated use of the same photoactive TiO₂ embedded into polyethylene pellets (PE-TiO₂) made by the controlled-temperature embedding method was studied in the photocatalytic degradation of commercial humic acid (HA) to solve the postrecovery of TiO₂ in slurry. Photocatalytic degradation percentage of dissolved organic carbon (DOC) removal with PE-TiO₂ was stable after the second use of the same PE-TiO₂ pellets, which was evaluated during five tests. PE-TiO₂ removed 64.6% of the initial DOC during 270 min of photocatalytic degradation, and TiO₂ in slurry removed 64.5% of the initial DOC during 180 min of photocatalytic degradation. Moreover, PE-TiO₂ led to a considerable reduction in specific ultraviolet absorbances (SUVA = UV/DOC): SUVA₂₅₄, SUVA₂₈₀, SUVA₃₆₅, and SCOA₄₃₆ (specific color absorbance) and trihalomethane formation potential (THMFP). Therefore, PE-TiO₂ is a promising material to remove NOM and solve the problem of postrecovery of TiO₂ particles after water treatment.

Introduction

Heterogeneous photocatalysis with titanium dioxide (TiO₂) has been shown to be a sustainable treatment technology to remove natural organic matter (NOM), humic acid (HA), fulvic acid (FA), and microorganisms of interest to the drinking water treatment industry (Valencia et al., 2014; Dong et al., 2015; Sillanpää and Matilainen, 2015; Fagan et al., 2016). The principle of heterogeneous photocatalysis with TiO₂ is based on the alteration of redox properties of the TiO₂ surface by the absorption of photons with energy equal to or greater than the TiO₂ band gap. This alteration promotes charge transfer reactions through the TiO₂-reaction media interface, leading to the formation of a free electron (e_f⁻) in the conduction band and highly oxidative holes (h_f⁺) in the valence band. This results in a TiO₂ with reactive character toward organic contaminants (Hoffmann et al., 1995; Herrmann, 1999; Valencia et al., 2011a).

Photocatalytic degradation of NOM by TiO₂ occurs in two steps: a preferential adsorption of large aromatic NOM compounds with high UV₂₅₄ absorption onto TiO₂ in the dark, followed by a preferential photocatalytic degradation of the adsorbed NOM into smaller compounds with low UV₂₅₄ absorption upon irradiation with ultraviolet (UV) light (Tercero et al., 2009; Gora and Andrews, 2017).

NOM is a problem for drinking water production. NOM causes taste, odor and color, fouls membranes, and promotes biological growth in water distribution (Bhatnagar and Sillanpää, 2017). Indeed, NOM is well known as a precursor of carcinogenic and mutagenic disinfection byproducts (DBPs) in water treatment, such as trihalomethanes (THMs), haloacetic acids (HAAs), and so on. (DeMarini, 2011; Koumaki et al., 2015).

NOM is formed by the decomposition of plants and animals in the environment (Uyguner-Demirel and Bekbolet, 2011). Moreover, different activities can increase organic matter concentration and potentially affect organic matter quantity and chemistry such as wildfires in forested watersheds (Tsai and Chow, 2016; Wang et al., 2016). NOM is a complex mixture of organic compounds, ranging from largely aliphatic compounds to highly colored aromatic molecules. Fractionation of NOM based on the adsorption/desorption of the compounds onto/from XAD-8 resin at a specific pH value produces hydrophobic and hydrophilic compounds (Valencia et al., 2013b; Pan et al., 2016).

Hydrophobic compounds form the major fraction of aquatic NOM, constituting more than half of dissolved organic carbon (DOC) in water. Hydrophobic compounds mainly contain aromatic carbon, phenolic structures, and conjugated double bonds. The main components of the hydrophobic fraction are humic substances (HS). HS can be portioned in HA, FA, and humin. On the other hand, hydrophilic compounds contain aliphatic carbon and nitrogenous compounds such as carbohydrates and proteins, sugar, and amino acid (Thurman, 1985).

Heterogeneous photocatalysis with TiO₂ continues to suffer from several limitations that impede the drinking water industry from using it. The main technical barriers that obstruct the commercialization of heterogeneous photocatalysis with TiO₂ in drinking water treatment are the inefficient exploitation of visible light, the uneven distribution of TiO₂ in aqueous suspension, and the postrecovery of TiO₂ particles after treatment (Chong et al., 2010; Dong et al., 2015). Postrecovery of TiO₂ slurry can be achieved through charge neutralization, coagulation–flocculation and a sedimentation process (Fernandez-Ibánez et al., 2003; Santos et al., 2013), cross-flow filtration (Doll and Frimmel, 2005), or several membrane filtrations (Zhang et al., 2008). Nevertheless, these processes of postrecovery are very expensive.

Several alternatives have been proposed to solve the postrecovery of TiO₂ by immobilizing TiO₂ particles on various substrates to provide a cost-effective solid–liquid separation. A variety of coating techniques have been used: the sol-gel process, chemical vapor deposition, electrodeposition, thermal treatment methods, grafting, reactive magnetron sputtering, spray pyrolysis, dip coating, spread coating, and so on. (Chang et al., 2009; Augugliaro et al., 2010; Shan et al., 2010; Lugo-Vera et al., 2016). However, these coating techniques are expensive and the required equipment is complex.

Velásquez et al. (2012) reported a controlled-temperature embedding method for immobilizing TiO₂ particles on polymer pellets, which is a low-cost technique. In this method, Evonik-P25 TiO₂ was dispersed in a polyalcohol medium (glycerine); then the suspension was heated to the melting point of a polymer; next, polymer pellets were added; and finally, the system was cooled. This results in TiO₂-embedded polymer pellets. This method has led to a photoactive polymeric material with strongly adherent TiO₂ on the surface. In addition, polymer pellets are abundant, easily available, and inexpensive compared with other immobilizer substrates.

In this study, the repeated use of the same photoactive Evonik-P25 TiO₂ embedded into polyethylene pellets (PE-TiO₂) made by the controlled-temperature embedding method was researched in the photocatalytic degradation of a commercial HA from Sigma-Aldrich (Aldrich-HA), as an aquatic HA representative, to solve the postrecovery of TiO₂ in slurry after treatment. The changes on Aldrich-HA after the photocatalytic degradation with PE-TiO₂ were monitored by measuring the DOC, and UV absorbance, specific ultraviolet absorbance (SUVA = UV/DOC), and the THM formation potential (THMFP) analysis.

Materials and Methods

Materials

Glycerine (USP), 4-chlorophenol (p.a.; Merck), isopropanol (99.8%; JT Backer), P25 TiO₂ (Evonik), and linear low-density polyethylene (PE) pellets (DNDA 8320; specific gravity [sg] = 0.926) were used in this study. Sodium hydroxide (0.1 N), sodium sulfite, and hydrochloric acid (0.1 M) were obtained from Merck, and sodium salt of HA was obtained from Sigma-Aldrich (Aldrich-HA). Water obtained from a Milli-Q system (Millipore) was used for all experiments.

Aldrich-HA was chosen as a model organic matter for being good analog for aquatic HA. The results obtained from photocatalytic degradation of Aldrich-HA have been consistent with those obtained from aquatic HA and NOM from different sources (Huang et al., 2008; Liu et al., 2008; Valencia et al., 2013a). Aldrich-HA has an organic content of 69.98%, carboxylic acid groups equal to 4.38 ± 0.03 meq⁻¹, phenol groups equal to 2.71 ± 0.034 meq⁻¹, and an elemental composition of 5.26% H, 0.74% N, 4.24% S, and 43.45% O (Malcolm and MacCarthy, 1986; Shin et al., 1999).

Controlled-temperature embedding method for TiO₂ coating on polyethylene pellets (PE-TiO₂)

A TiO₂ suspension was prepared by suspending 26 g of Evonik-P25 TiO₂ in water at 40°C using a heating magnetic stirrer (Arex) with a digital thermoregulator (VTF; VELP Scientifica). The suspension was mixed using an overhead stirrer (RW 20 digital; Fisher Scientific) and was heated to the melting point of the PE (106°C). Then, 90 g of PE pellets was added. The system was held at the melting temperature for 20 min with mechanical agitation. After that, it was cooled down to room temperature. Finally, the coated pellets (PE-TiO₂) were washed with water and dried at 25°C for 24 h (Velásquez et al., 2012).

Characterization, erosion, and reuse tests of PE-TiO₂

Thermogravimetric analysis (TGA) of PE-TiO₂ samples was recorded on a Linseis STA Platinum series 1600 instrument. The temperature was ramped from 25°C to up to 1,000°C at 10°C/min. The experiments were conducted in air atmosphere. Scanning electron micrographs (SEM) were obtained in a JEOL JSM-6490LV-SEM.

The PE-TiO₂ erosion test was performed under extreme conditions to determine the resistance of the coating adhesion of TiO₂. Pellets of PE-TiO₂ were subjected to the erosion test by sonication and mechanical agitation to assess the erosion behavior of PE-TiO₂, where the coated pellets (PE-TiO₂) collided with each other. The sonication-erosion test was carried out using 60 mL of isopropanol 1:1 in a sonicating water bath (Elma Ultrasonic LC-30H) at 35 KHz for 1 h. The mechanical agitation-erosion test was done using 150 mL of Milli-Q water with an overhead stirrer (RW 20 digital; Fisher Scientific) at 500 rpm for 70 h. Then, PE pellets were dried at 25°C for 24 h and weighed. A detailed explanation of these methods has been reported elsewhere (Velásquez et al., 2012).

The PE-TiO₂ reuse test was studied by repetitive use of the same PE-TiO₂ (five tests) on photocatalytic degradation of the Aldrich-HA solution of ρ₀(DOC) = 10 mg/L. The changes in the efficiency of photocatalytic degradation of TiO₂ were monitored by DOC analysis. For each test of photocatalytic degradation, a new solution of ρ₀(DOC) = 10 mg/L of Aldrich-HA was used.

Photocatalytic reactor

Photocatalytic degradation experiments of Aldrich-HA were performed with a batch photocatalytic reactor of Plexiglas with a volume of 4.5 L (external diameter = 13.3 cm, height = 33.0 cm, and thickness = 8.0 mm) and a cooling jacket using water (external diameter = 20.0 cm, height = 33.0 cm, and thickness = 8.0 mm) (Fig. 1a). The photocatalytic reactor was equipped with a stirrer system and an air injector system that pumps air through the suspension to maintain a saturated concentration of dissolved oxygen.

FIG. 1.

Schematic of photocatalysis system (a) aluminum housing and (b) reactor.

The UV source was a set of eight UV lamps, each of them supplying 25 W of nominal power (Black Light Bulb; Sylvania) with an output between 310 and 410 nm and a maximum emission at 350 nm (Sylvania). UV lamps were placed vertically around the photocatalytic reactor (external UV lamps) in aluminum housing (outer reflective cylinder; external diameter = 35 cm, height = 46.8 cm, and thickness = 2.0 mm). Highly reflective polished raw aluminum (percent reflected = 56.5) was used as a reflector (Fig. 1b).

Radiometric measurements were performed with a UV-A radiometer, Solar Light (Model PMA 2200, with UV detector), with the detector placed at the center of the aluminum housing, the center of the reactor, and the center of the jacket-reactor at different positions (at the top, in the middle, and at the bottom). Before each measurement was taken, lamps were warmed up for a period of 30 min to stabilize lamp emission.

For each photocatalytic degradation experiment carried out, 3 L of Aldrich-HA with ρ₀(DOC) = 10 mg/L and initial pH = 7 was used. PE-TiO₂ pellets or TiO₂ were loaded and the stirrer system was used. In the TiO₂ slurry experiments, after photocatalytic degradation, TiO₂ was immediately removed from the reaction medium by 0.4 μm HTTP membrane filters (Isopore; Millipore) before analysis. In this study, all photocatalytic experiments were performed in triplicate.

Aldrich-HA solution preparation

The Aldrich-HA solution was prepared by dissolving 1.0 g of Aldrich-HA in 5 L of water and pH was adjusted to 9 with NaOH, 0.1 N. The sample was stirred overnight; after that, pH was adjusted to 7 with HCl, 0.1 M. The sample was filtered through a 0.45 μm cellulose nitrate membrane filter to remove the undissolved HA. The Aldrich-HA solution obtained was diluted to ρ₀(DOC) = 10 mg/L for experiments of photocatalytic degradation.

DOC analysis, UV-visible spectroscopy, and THMFP analysis

DOC was measured with a TOC/TN-analyzer (Torch; Teledyne Tekmar) following Standard Method 5310B (2012). Ultraviolet-visible (UV/Vis) absorbance of the Aldrich-HA was estimated with a spectrophotometer (Thermo Evolution 600). SUVA₂₅₄, SUVA₂₈₀, SUVA₃₆₅ (the specific UV absorbance), and SCOA₄₃₆ (specific color absorbance) parameters were calculated as indicative of the photocatalytic degradation efficiency of Aldrich-HA. SUVA is defined as the UV absorbance per milligram of DOC (L/[mg·m]), and SCOA₄₃₆ is defined as Color₄₃₆ per milligram of DOC (L/[mg·m]).

In the THMFP analysis, all samples were diluted to ρ(DOC) = 3 mg/L and buffered to pH = 7 using a 0.1 M phosphate buffer. Chlorination of the samples was performed using sodium hypochlorite at ρ(Cl₂) = 10 mg/L in a ratio of ρ(DOC):ρ(Cl₂) = 3:10. After 48 h of reaction in the dark, chlorination was stopped by the addition of sodium sulfite. THM (chloroform, bromodichloromethane, dibromochloromethane, and bromoform) measurements were carried out using gas chromatography on an Agilent Technologies 7890 A GC system gas chromatograph coupled to an Agilent Technologies 5975C VI mass detector (GC/MSD). For the extraction of THMs, headspace solid-phase microextraction (HS-SPME) was performed manually using 75 μm carboxen/polydimethylsiloxane-coated fiber (75-CAR.PDMS). The precision of the method was evaluated by calculating the relative standard deviation (%RSD) of five replicates using THM standards at 100 μg/L. %RSD for precision ranged between 1.4% and 3.3%. The reproducibility of the method was evaluated by peak area using 100 μg/L THM standard in five replicate samples. %RSD for reproducibility ranged between 3.5% and 6.7%. The accuracy was evaluated by the percent relative recovery (%REC) of five replicates. %REC for accuracy ranged between 94.1% and 108.6%. Thus, the method was precise enough to analyze THMs. In this study, all THM analyses were performed in triplicate. A detailed description of THM measurements has been reported elsewhere (Valencia et al., 2013a).

Results and Discussion

Material characterization

SEM of PE-pellet and PE-TiO₂ (Fig. 2) show that the TiO₂ particles were distributed homogeneously on the surface of pellets, although there were small agglomerated TiO₂ particles on the surface and some areas of the pellet without TiO₂ particles.

FIG. 2.

SEM images of PE pellets (a) and PE-TiO₂ pellets (b). PE, polyethylene; TiO₂, titanium dioxide.

TGA reveals that the total weight loss of PE samples heated to 1,000°C was 100% (data not shown). The total weight loss of PE-TiO₂ was 95.3% (data not shown). Consequently, the residual weight of the samples calcined at 1,000°C represented the weight of TiO₂ coating. The analysis was acquired in triplicate, and the RSD was 4.3%. The weight percentage of TiO₂ coating was 4.7% for PE-TiO₂. Each pellet of PE-TiO₂ contained 1.2–1.5 mg of TiO₂, and 100 pellets of PE-TiO₂ weighed 3.6 g. Therefore, 3.6 g of PE-TiO₂ contained 0.12–0.15 g of TiO₂. A detailed characterization of PE-TiO₂ material by the controlled-temperature encapsulation method has been reported elsewhere (Velásquez et al., 2012).

Erosion test analysis

Results obtained in the erosion test with sonication and mechanical agitation of PE-TiO₂ demonstrated that TiO₂ adhered to PE pellets was sufficiently resistant to erosion. There were weight losses of 0.8% and 0.5% under sonication and mechanical-agitation erosion test, respectively. The loss in weight of PE-TiO₂ was due to the loss of TiO₂ embedded into PE.

Photolysis experiments of Aldrich-HA

Table 1 shows intensities of incident radiation in the photocatalytic reactor (Fig. 1). Plexiglas absorbed 61% of incident radiation, while the water jacket did not absorb radiation.

Table 1.

Intensities of the Incident Radiation of the System

	Radiation intensity (W/m²)
Position	Aluminum housing lamps	Aluminum housing lamps + reactor	Aluminum housing-lamps + reactor + water jacket
Upper part	9.71	3.35	3.02
Medium part	10.30	4.22	4.02
Lower part	7.25	2.95	2.57

Set of eight UV lamps, supplying 25 W of nominal power (Black Light Bulb; Sylvania), with output between 310 and 410 nm and a maximum at 350 nm.

UV, ultraviolet.

The photolysis experiment (in the absence of TiO₂ or PE-TiO₂) of Aldrich-HA with ρ₀(DOC) = 10 mg/L at initial pH = 7 was performed using the photocatalytic reactor (Fig. 1) during 480 min of irradiation. There was a removal of DOC of 9% during the first 120 min of exposition; after that, no change in the DOC was observed (Fig. 3). In addition, the photolysis experiment did not lead to a significant decrease in the UV/Vis spectra absorption of Aldrich-HA (Fig. 4). The experiment was performed in triplicate. The repeatability was below 5% of RSD.

FIG. 3.

Effect of irradiation time on DOC during photolysis experiment of HA-Aldrich after 480 min of irradiation. ρ(DOC) = 10 mg/L at initial pH = 7. DOC, dissolved organic carbon; HA, humic acid; HA-Aldrich, humic acid from Sigma-Aldrich.

FIG. 4.

Effect of irradiation time on UV/Vis absorption spectral during photolysis experiment of HA-Aldrich after 120 min of irradiation. ρ(DOC) = 10 mg/L at initial pH = 7. UV/Vis, ultraviolet/visible.

Adsorption experiments of Aldrich-HA onto PE-TiO₂ and TiO₂

Adsorption experiments in the dark with 3 L of Aldrich-HA solution with ρ₀(DOC) = 10 mg/L at initial pH = 7 and ρ(TiO₂) = 0.6 g/L in slurry or ρ(PE-TiO₂) = 18.1 g/L [approximately ρ(TiO₂) = 0.6 g/L] showed that equilibrium of Aldrich-HA on TiO₂ or PE-TiO₂ was reached in <5 min (data not shown). An adsorption time of 60 min was chosen to ensure saturation for all subsequent experiments.

Photocatalytic degradation experiments on Aldrich-HA with TiO₂ in slurry and PE-TiO₂

Figure 5 shows the remaining DOC as a function of TiO₂ dosage in slurry after adsorption in the dark to reach the equilibrium state, and after photocatalytic degradation with 180 min of UV irradiation time. The experiment was performed in triplicate. The repeatability was 4.5% of RSD. DOC removal during photocatalytic degradation in slurry was larger with increasing ρ(TiO₂) between 0.1 and 0.6 g/L due to bigger number of active sites. At higher ρ(TiO₂), DOC removal remained constant because the turbidity of TiO₂ suspension reduced the effective UV irradiation. Therefore, the optimum ρ(TiO₂) was 0.6 g/L in slurry.

FIG. 5.

Effect of TiO₂ dose in slurry on Aldrich-HA adsorption and photocatalytic degradation. ρ(DOC) = 10 mg/L, at initial pH = 7, adsorption time = 60 min, photocatalytic degradation time = 180 min.

Photocatalytic degradation kinetics of NOM generally can be approximated by first-order kinetics, and the first-order rate constant is the most accurate measurement of photoactivity of the catalyst. For a degradation that follows a kinetic first order the following equation was used: ln(C/C₀) = −k_at (Huang et al., 2008), where C₀ is the initial concentration of DOC in the liquid phase, C is the actual DOC concentration in the liquid phase, and k_a (min⁻¹) denotes the pseudo-first-order reaction rate constant.

It was found that the photocatalytic degradation of Aldrich-HA with ρ(DOC) = 10 mg/L at initial pH = 7 and ρ(TiO₂) = 0.6 g/L with TiO₂ in slurry, followed a pseudo-first-order rate with k_a = 0.011 min⁻¹ (Fig. 6). The photocatalytic degradation of Aldrich-HA with PE-TiO₂ with ρ(PE-TiO₂) = 18.1 g/L, which is equivalent to ρ(TiO₂) = 0.6 g/L in slurry (∼54.0 g of PE-TiO₂ in 3 L of Aldrich-HA) also followed a pseudo-first-order rate with k_a = 0.009 min⁻¹ (Fig. 7). There were good fit of the data to a linear response. The correlation coefficients (R²) were 0.992 and 0.982, for TiO₂ in slurry and PE-TiO₂, respectively. The experiment was performed in triplicate. The repeatability was below 5% of RSD.

FIG. 6.

Photocatalytic degradation kinetics of Aldrich-HA with P25 TiO₂ in slurry. ρ(DOC) = 10 mg/L, at initial pH = 7, ρ(TiO₂) = 0.6 g/L, adsorption time = 60 min.

FIG. 7.

Photocatalytic degradation kinetics of Aldrich-HA with PE-TiO₂. ρ(DOC) = 10 mg/L, at initial pH = 7, ρ(PE-TiO₂) = 6.0 g/L [approximately ρ(TiO₂) = 0.6 g/L], adsorption time = 60 min, photocatalytic degradation time = 270 min.

The k_a of Aldrich-HA with PE-TiO₂ was lower than that with TiO₂ in slurry. This could be due to the reduction in the exposed surface area of PE-TiO₂ and the reduction in the kinetic processes of the interfacial charge transfer. A certain amount of TiO₂ was embedded into PE-pellets, where UV irradiation and Aldrich-HA could not reach TiO₂. This led to a decrease in kinetics of degradation. k_a increased as the exposed surface area augmented; it occurred because the number of reactive sites increases and the photocatalytic degradation of HA-Aldrich proceeds on the TiO₂ surface by oxidation of the adsorbed Aldrich-HA, where h_f⁺ reacts directly with Aldrich-HA (Valencia et al., 2011a, 2012).

In contrast, the increase in PE-TiO₂ pellets in the solution led to a higher decrease in UV light penetration (screening effect), in comparison with TiO₂ in slurry. As a result, photocatalytic degradation with PE-TiO₂ required more UV irradiation time to achieve a similar DOC removal as the one obtained with TiO₂ in slurry.

PE-TiO₂ reuse test analysis for Aldrich-HA photocatalytic degradation

Figure 8 shows the photocatalytic degradation of ρ(DOC) = 10 mg/L of Aldrich-HA at initial pH = 7 with ρ(PE-TiO₂) = 18.1 g/L¹ [approximately ρ(TiO₂) = 0.6 g/L] by the repeated use of the same PE-TiO₂ pellets (five tests). Aldrich-HA photocatalytic degradation percentages achieved with the same PE-TiO₂ pellets were relatively stable after repeated use in the five runs (61.1% of DOC) during 270 min of UV irradiation in comparison with TiO₂ in slurry with only one use (95.7% of DOC). The loss in efficiency in the first two repeated uses of the same PE-TiO₂ pellets might be due to the weight loss of TiO₂ coating. After the second use of PE-TiO₂ pellets, TiO₂ embedded into PE pellets was stable and it was not detached. Therefore, PE-TiO₂ can be used to provide a viable way to solve the problem of the postrecovery of TiO₂ particles after water treatment.

FIG. 8.

Catalyst-deactivation test using PE-TiO₂ with ρ(PE-TiO₂) = 18.1 g/L [approximately ρ(TiO₂) = 0.6 g/L] (reuse of PE-TiO₂), and TiO₂ in slurry with ρ(TiO₂) = 6.0 g/L in the photocatalytic degradation of Aldrich-HA, ρ(DOC) = 10 mg/L, at initial pH = 7. Adsorption time = 60 min, photocatalytic degradation time = 270 min.

Formation of DBPs

Chlorination was studied to determine the change in the total THMFP and specific THM formation potential (THMFP/DOC) of Aldrich-HA samples at initial pH = 7 as a function of irradiation time during photocatalytic degradation of Aldrich-HA with PE-TiO₂ and TiO₂ in slurry using the batch photocatalytic reactor of Plexiglas. Chlorination was performed using sodium hypochlorite at ρ(Cl₂) = 10 mg/L in a ratio of ρ(DOC):ρ(Cl₂) = 3:10. After 48 h of reaction in the dark, chlorination was stopped by the addition of sodium sulfite. After that, THMs were measured. The THM analysis only showed the formation of CHCl₃ (chloroform). The brominated THMs were not detected.

For chlorination of the samples with PE-TiO₂, ρ(PE-TiO₂) = 6.0 g/L [approximately ρ(TiO₂) = 0.2 g/L], using the batch photocatalytic reactor of Plexiglas, four samples (M1–M4) were analyzed: M1 represents the original sample of Aldrich-HA with ρ₀(DOC) = 9.90 mg/L, at initial pH = 7; M2 is the sample after adsorption onto PE-TiO₂ in the dark with ρ(DOC) = 8.55 mg/L; M3 is the sample after PE-TiO₂-photocatalytic degradation with an irradiation time of 120 min, reaching ρ(DOC) = 6.21 mg/L; and M4 is the sample after PE-TiO₂-photocatalytic degradation with a longer irradiation time of 270 min, reaching ρ(DOC) = 3.20 mg/L.

Table 2 shows specific absorbance parameters SUVA₂₅₄, SUVA₂₈₀, SUVA₃₆₅, and SCOA₄₃₆ of samples M1–M4 with ρ(PE-TiO₂) = 6.0 g/L [approximately ρ(TiO₂) = 0.2 g/L]. SUVA₂₅₄ and SUVA₂₈₀ (L/[mg·m]) represent the UV-absorbing aromatic structures and double bonds of NOM, respectively (Sarathy and Mohseni, 2007). SUVA₂₅₄ also can be used to describe the composition of water in terms of hydrophobicity and hydrophilicity. A value of SUVA₂₅₄ > 4 indicates mainly hydrophobic and especially aromatic material, whereas 2 < SUVA₂₅₄ < 4 represents a mixture of hydrophobic and hydrophilic NOM, and SUVA₂₅₄ < 2 represents hydrophilic material (Uyguner and Bekbolet, 2005). It has been shown that SUVA₃₆₅ increases with increasing molecular weight (Peuravuori and Pihlaja, 2004). SCOA₄₃₆ represents the functional groups that give the samples a yellow to brown color and it is used to estimate the content of quinones (Kumke et al., 2001).

Table 2.

Dissolved Organic Carbon and Specific Ultraviolet Absorbance (L/[mg·m]) Parameters of the Remaining Humic Acid from Sigma-Aldrich, ρ(DOC) = 10 mg/L at Initial pH = 7, with ρ(PE-TiO₂) = 6.0 g/L [Approximately ρ(TiO₂) = 0.2 g/L], Adsorption Time = 60 min, Using the Batch Photocatalytic Reactor of Plexiglas

Parameter	M1	M2	M3	M4	% Removal (M2 vs. M4)
Irradiation time (min)	0	0	120	270	—
DOC (mg/L)	9.90	9.05	6.21	3.20	64.6
SUVA₂₅₄	11.16	9.11	5.13	1.50	83.5
SUVA₂₈₀	9.43	8.27	3.61	0.95	88.5
SUVA₃₆₅	3.77	3.14	1.12	0.21	93.3
SCOA₄₃₆	1.75	1.42	0.41	0.13	90.8

Aldrich-HA, humic acid from Sigma-Aldrich; DOC, dissolved organic carbon; HA, humic acid; M1, the original sample of Aldrich-HA; M2, the sample after adsorption onto PE-TiO₂ in the dark; M3, the sample after photocatalytic degradation time of 120 min; M4, the sample after photocatalytic degradation time of 270 min; SCOA, specific color absorbance; SUVA, specific ultraviolet absorbance; TiO₂, titanium dioxide.

There was a considerable decrease in SUVA₂₅₄, SUVA₂₈₀, SUVA₃₆₅, and SCOA₄₃₆ from M1 to M4 with PE-TiO₂ (Table 2) implying a reduction of aromatic compounds and double bonds, a decrease in molecular size, and a reduction of quinones for Aldrich-HA, respectively. Furthermore, SUVA₂₅₄, SUVA₂₈₀, SUVA₃₆₅, and SCOA₄₃₆ removals were significantly higher than DOC removal with PE-TiO₂, showing that photocatalytic degradation with PE-TiO₂ led to the loss of aromaticity and the decrease of the molecular weight of Aldrich-HA at a faster rate than the rate of mineralization.

These results are due to the fact that the hydrophobic compounds (larger molecular weight fractions) are preferably adsorbed onto the TiO₂ surface. It favored the photocatalytic degradation of hydrophobic compounds and more hydrophobic intermediate oxidation products (smaller molecular weight fractions) that were then released (Huang et al., 2008; Tercero et al., 2009). These results are consistent with Liu et al. (2014), who worked with HA from Sigma-Aldrich and TiO₂ in slurry, Valencia et al. (2012) who studied the photocatalytic degradation of NOM from Lake Hohloh (Germany) with TiO₂ in slurry, and Valencia et al. (2013b) who worked with HA and FA from XAD fractions of NOM from a bog lake with TiO₂ in slurry.

Table 3 shows THMFP and specific THMFP (STHMFP) evolution for Aldrich-HA at initial pH = 7 after photocatalytic degradation with ρ(PE-TiO₂) = 6.0 g/L [approximately ρ(TiO₂) = 0.2 g/L; samples M1–M4]. There was a high removal of THMFP after Aldrich-HA adsorption onto PE-TiO₂ in the dark (M2; 34.7%), with a DOC removal of 8.6% (M2). Therefore, PE-TiO₂ adsorption is a viable way to remove THM precursors from drinking water. Moreover, the photocatalytic degradation with PE-TiO₂ led to a substantial reduction in THMFP and STHMFP. The reduction of THMFP was 86.4% (in 270 min), exceeding the percentage removal of DOC, 64.5% (in 270 min). Therefore, the organic fractions of Aldrich-HA that form DBPs were specially removed during photocatalytic degradation with PE-TiO₂. These results agree with those of the specific absorbance parameters, where all of them decreased. Moreover, these results are in agreement with Liu et al. (2010) who worked with TiO₂ in slurry.

Table 3.

Trihalomethane Formation Potential and Specific Trihalomethane Formation Potential Evolution During the Photocatalytic Degradation of Humic Acid from Sigma-Aldrich at Initial pH = 7, with ρ(PE-TiO₂) = 6.0 g/L [Approximately ρ(TiO₂) = 0.2 g/L], Using the Batch Photocatalytic Reactor of Plexiglas

Parameter	M1	M2	M3	M4	% Removal (M2 vs. M4)
Irradiation time (min)	0	0	120	270	—
ρ(DOC) (mg/L)	9.90	9.05	6.21	3.21	64.5
THMFP (μg/L)	312.5	204.1	154.8	27.7	86.4
STHMFP (μg/mg DOC)	31.6	23.9	24.9	8.8	63.2

M3, the sample after photocatalytic degradation with an UV irradiation time of 120 min; M4, the sample after photocatalytic degradation with an UV irradiation time of 270 min; STHMFP, specific THMFP; THMFP, trihalomethane formation potential.

On the other hand, there was an increase of STHMP in M3 with respect to M2. This is because, during photocatalytic degradation of Aldrich-HA with PE-TiO₂, the compounds of large molecular weight are rather adsorbed onto PE-TiO₂ and are preferably degraded, leading to increasing compounds of small molecular weight. These small compounds were hydrophobic intermediate products that were released from the PE-TiO₂ surface (Tercero et al., 2009; Valencia et al., 2011b; Gora and Andrew, 2017).

For chlorination of the samples with TiO₂ in slurry with ρ(TiO₂) = 0.2 g/L, using the batch photocatalytic reactor of Plexiglas, four samples (M5–M8) were also analyzed: M5 represents the original sample of Aldrich-HA with ρ₀(DOC) = 9.90 mg/L at initial pH = 7; M6 is the sample after adsorption onto TiO₂ in the dark with ρ(DOC) = 7.78 mg/L; M7 is a sample after TiO₂-photocatalytic degradation with an irradiation time of 105 min, reaching ρ(DOC) = 5.90 mg/L; and M8 is a sample after TiO₂-photocatalytic degradation with a longer irradiation time of 180 min, reaching ρ(DOC) = 3.10 mg/L. The optimum dosage [ρ(TiO₂) = 0.6 g/L] was not used because the DOC transformation and UV₂₅₄ absorption occurred faster upon irradiation.

Table 4 shows specific absorbance parameters SUVA₂₅₄, SUVA₂₈₀, SUVA₃₆₅, and SCOA₄₃₆ of samples M5–M8 with ρ(TiO₂) = 0.2 g/L in slurry, using the batch photocatalytic reactor of Plexiglas. A considerable decrease of SUVA₂₅₄, SUVA₂₈₀, SUVA₃₆₅, and SCOA₄₃₆ parameters from M5 to M8 with P25 TiO₂ in slurry was found (Table 2), implying a reduction of aromatic compounds and double bonds, a decrease in molecular size, and a reduction of quinones for Aldrich-HA, respectively. Moreover, SUVA₂₅₄ decreased from 11.21 (M5) to 1.34 (M8) implying that Aldrich-HA composition changed from hydrophobic to hydrophilic material with less aromatic compounds. On the other hand, SUVA₂₅₄, SUVA₂₈₀, SUVA₃₆₅, and SCOA₄₃₆ removals were significantly higher than the DOC removal with TiO₂ in slurry (Table 2).

Table 4.

Dissolved Organic Carbon and Specific Absorbance (L/[mg·m]) Parameters of the Remaining Humic Acid from Sigma-Aldrich, ρ(DOC) = 10 mg/L at Initial pH = 7, with TiO₂ in Slurry, ρ(TiO₂) = 0.2 g/L Using the Batch Photocatalytic Reactor of Plexiglas

Parameter	M5	M6	M7	M8	% Removal (M6 vs. M8)
Irradiation time (min)	0	0	105	180	—
DOC (mg/L)	9.90	8.55	5.90	3.10	63.7
SUVA₂₅₄	11.21	10.12	5.06	1.34	86.8
SUVA₂₈₀	9.49	9.13	3.49	0.90	90.1
SUVA₃₆₅	3.74	3.40	1.02	0.19	94.4
SCOA₄₃₆	1.72	1.55	0.37	0.11	92.9

HA, humic acid; M5, the original sample of Aldrich-HA; M6, the sample after adsorption onto TiO₂ in the dark; M7, the sample after photocatalytic degradation time = 105 min; M8, the sample after photocatalytic degradation time of 180 min.

Table 5 shows THMFP and STHMFP (STHMFP = THMFP/DOC) evolution for Aldrich-HA at initial pH = 7 before and after photocatalytic degradation with ρ(TiO₂) = 0.2 g/L in slurry (samples M5–M8). There was a high removal of THMFP after Aldrich-HA adsorption onto TiO₂ in slurry in the dark (M6; 39.3%), with a DOC removal of 13.6% (M6). Moreover, the photocatalytic process with TiO₂ in slurry led to a substantial reduction in THMFP from M5 to M8 due to the overall reduction of DOC. The removal of THMFP was 85.8% (in 180 min), exceeding the percentage removal of DOC, 68.7% (in 180 min). In addition, an increase of STHMP in M7 with respect to M8 was found.

Table 5.

Trihalomethane Formation Potential and Specific Trihalomethane Formation Potential Evolution During the Photocatalytic Degradation of Humic Acid from Sigma-Aldrich, ρ(DOC) = 10 mg/L at Initial pH = 7, with TiO₂ in Slurry, ρ(TiO₂) = 0.2 g/L, Using the Batch Photocatalytic Reactor of Plexiglas

Parameter	M5	M6	M7	M8	% Removal (M6 vs. M8)
Irradiation time (min)	0	0	105	180	—
ρ(DOC) (mg/L)	9.90	8.55	5.91	3.13	63.4
THMFP (μg/L)	312.5	189.7	151.7	26.9	85.8
STHMFP (μg THM mg⁻¹ DOC)	31.6	24.4	25.7	9.3	61.9

THM, trihalomethane.

Results obtained in this study with PE-TiO₂ were similar to those with TiO₂ in slurry. Therefore, PE-TiO₂ has shown to be a sustainable treatment technology to remove HA, which is of interest to the drinking water treatment industry. Moreover, embedding of TiO₂ onto PE pellets by the controlled-temperature embedding method did not change the properties of TiO₂ for HA photocatalytic degradation. However, the photocatalytic degradation with PE-TiO₂ requires a longer UV irradiation time than the required time with TiO₂ in slurry to achieve a similar removal of parameters (SUVA and THMFP) (Tables 3 and 5). This result is explained by the incidence of PE-TiO₂ pellets in the solution that led to a higher decrease in UV light penetration (screening effect) in comparison with TiO₂ in slurry. On the other hand, the adsorption of DOC onto PE-TiO₂ in the dark (with a removal of DOC of 8.6% in M3) was lower than that obtained with TiO₂ in slurry (with a removal of DOC of 13.6% in M6). DOC adsorption is higher with an increasing superficial area of TiO₂, and PE-TiO₂ showed a lower area than TiO₂ in slurry.

Results with PE-TiO₂ pellets may imply a higher UV irradiation cost for photocatalytic degradation of HA in comparison to the cost associated with the use of TiO₂ slurry. However, postrecovery of TiO₂ slurry with coagulants can be more expensive than the longer UV irradiation time with PE-TiO₂. The repeated use of PE-TiO₂ pellets reduces the cost for the longer UV irradiation time and does not generate substantial waste. On the other hand, PE pellets are abundant, easily available, and inexpensive as an immobilizer substrate. PE and PE-TiO₂ exhibit UV absorption and TiO₂/UV degradation resistance, and UV irradiation does not lead to considerable surface damage of PE (Velásquez et al., 2012). Therefore, PE-TiO₂ is a promising material to solve the problem of postrecovery of TiO₂ particles after water treatment.

Conclusions

Polyethylene pellets coated with P25 TiO₂ (PE-TiO₂) using the controlled-temperature embedding method allowed the reuse of the same TiO₂ in Aldrich-HA photocatalytic degradation with a relatively stable efficiency. PE-TiO₂ led to a high photocatalytic degradation of Aldrich-HA and a large reduction in specific absorbance parameters: SUVA₂₅₄, SUVA₂₈₀, SUVA₃₆₅, and SCOA₄₃₆, and THMFP. Therefore, embedding of TiO₂ onto PE pellets by the controlled-temperature embedding method did not change the properties of TiO₂ for HA photocatalytic degradation, and PE-TiO₂ is a promising material to remove NOM and to solve the problem of postrecovery of TiO₂ particles after water treatment.

Footnotes

Acknowledgments

The authors thank Universidad de Antioquia and COLCIENCIAS for financial support.

Author Disclosure Statement

No competing financial interests exist.

References

Augugliaro

, Loddo

, Pagliaro

, Palmisano

, and Palmisano

(2010). Clean by Light Irradiation, Practical Applications of Supported TiO₂. Cambridge: The Royal Society of Chemistry.

Bhatnagar

, and Sillanpää

(2017). Removal of natural organic matter (NOM) and its constituents from matter by adsorption-a review. Chemosphere, 166, 497.

Chang

, Hsin

, and Doong

(2009). Characterization and photocatalytic activity of wanadium-doped titanium dioxide nanocatalyst. Water Sci. Technol. 59, 523.

Chong

, Jin

, Chow

, and Saint

(2010). Recent development in photocatalytic water treatment technology: A review. Water Res. 44, 2997.

DeMarini

D.M.

(2011). Genotoxicity of disinfection by-products: Comparison to carcinogenicity. In Nriagu

, Ed., Encyclopedia of Environmental Health. New York, Michigan: Elsevier Science & Technology, p. 920.

Doll

, and Frimmel

F.H.

(2005). Removal of selected persistent organic pollutants by heterogeneous photocatalysis in water. Catal. Today, 101, 195.

Dong

, Zeng

, Tang

, Fan

, Zhang

, He

, and He

(2015). An overview on limitations of TiO₂-based particles for photocatalytic degradation of organic pollutants and the corresponding countermeasures. Water Res. 79, 128.

Fagan

, McCormack

, Dionysiou

, and Pillai

(2016). A review of solar and visible light active TiO₂ photocatalysis for treating bacteria, cyanotoxins and contaminants of emegins concern. Mat. Sci. Semicon. Proc., 42, 2.

Fernandez-Ibánez

, Blanco

, Malato

, and de las Nieves

F.J.

(2003). Application of the colloidal stability of TiO₂ particles for recovery and reuse in solar photocatalysis. Water Res. 37, 3180.

10.

Gora

, and Andrews

(2017). Adsorption of natural organic matter and disinfection by product precursors from surface water onto TiO₂ nanoparticles: pH effect, isotherm modelling and implications using TiO₂. For drinking water treatment. Chemophosfere, 174, 363.

11.

Herrmann

J.M.

(1999). Heterogeneous photocatalysis: Fundamentals and applications to the removal of various types of aqueous pollutants. Catal. Today, 53, 115.

12.

Hoffmann

M.R.

, Martin

S.T.

, Choi

, and Bahnemann

D.W.

(1995). Environmental applications of semiconductor photocatalysis. Chem. Rev., 95, 69.

13.

Huang

, Leal

, and Li

(2008). Degradation of natural organic matter by TiO₂ photocatalytic oxidation and its effect on fouling of low-pressure membranes. Water Res. 42, 1142.

14.

Koumaki

, Mamais

, Noutsopoulos

, Nika

, Bletsou

, Thomaidis

, Eftaxias

, and Stratogiani

(2015). Degradation of emerging contaminants from water under natural sunlight: The effect of season, pH, humic acids and nitrate and identification of photodegradation by products. Chemosphere, 138, 675.

15.

Kumke

, Specht

, Brinkmann

, and Frimmel

F.H.

(2001). Alkaline hydrolysis of humic sustances-spectroscopic and chromatographic investigation. Chemosphere, 45, 102.

16.

Liu

, Lim

, Fabris

, Chow

, Drikas

, and Amal

(2010). Comparison of photocatalytic degradation of natural organic matter in two Australian surface water using multiple analytical techniques. Org. Geochem., 41, 124.

17.

Liu

, Lim

, and Amal

(2014). TiO₂-coated natural zeolites: Raped humic acid adsorption and effective photocatalytic regeneration. Chem. Eng. Sci., 105, 46.

18.

Liu

, Lim

, Fabris

, Chow

, Drikas

, and Amal

(2008). Removal of humic acid using TiO₂ photocatalytic process-formation and molecular weight characterization studies. Chemosphere, 72, 263.

19.

Lugo-Vera

, Serrano-Rosales

, and Lasa

(2016). Immobilized particle coating for optimum photon and TiO₂ utilization in scaled air treatment photoreactor. Appl. Catal. B Environ., 198, 211.

20.

Malcolm

, and MacCarthy

(1986). Limitation in the use of commercial humic acids in water and soil research. Environ. Sci. Technol., 20, 904.

21.

Pan

, Li

, Zhang

, and Li

(2016). Characterization of natural organic matter in drinking water: Sample preparation and analytical approaches. Trends Environ. Anal. Chem., 12, 23.

22.

Peuravuori

, and Pihlaja

(2004). Preliminary study of lake dissolved organic matter in light of nanoscale supramolecular assembly. Environ. Sci. Technol., 38, 5958.

23.

Santos

, Valente

, Jorge

, Padilha

, Saeki

, Castro

, and Florentino

(2013). Coagulation-flocculation of TiO₂ in suspension used in heterogeneous photocatalysis. Orbital Electron. J. Chem., 5, 233.

24.

Sarathy

, and Mohseni

(2007). The impact of UV/H₂O₂ advanced oxidation on molecular size distribution of chromophoric natural organic matter. Environ. Sci. Technol., 41, 8315.

25.

Shan

, Ghazi

, and Rashid

(2010). Immobilization of titanium dioxide onto supporting materials in heterogeneous photocatalysis: A review. Appl. Catal. A General, 389, 1.

26.

Shin

, Monsaller

, and Chopping

(1999). Spectroscopy and chemical characterization of molecular size fractioned humic acid. Talanta, 50, 641.

27.

Sillanpää

, and Matilainen

(2015). NOM removal by advanced oxidation processes. In Sillanpää

, Ed., Natural Organic Matter in Water: Characterization and Treatment Methods. Oxford: Elsevier, p. 159.

28.

Standard Method 5310B. (2012). Total organic carbon (TOC)—High-temperature combustion method. In Rice

, Baird

, Eaton

, and Closceri

Eds., Standard Methods for the Examination of Water and Wastewater, 21st ed. Washington DC: American Public Health Association, p. 5.

29.

Tercero

L.A.

, Haseborg

E.T.

, Weber

, and Frimmel

F.H.

(2009). Investigation of photocatalytic degradation of brown water natural organic matter by size exclusion chromatography. Appl. Catal. B Environ., 87, 56.

30.

Thurman

E.M.

(1985). Organic Geochemistry of Natural Waters. Dordrecht: Martinus Nijhoff-Dr. Junk Publisher.

31.

Tsai

K.P.

, and Chow

(2016). Growing algae alter spectroscopic characteristics and chlorine reactivity of dissolved organic matter from thermally-altered forest litters. Environ. Sci. Tech., 50, 7991.

32.

Uyguner

, and Bekbolet

(2005). Implementation of spectroscopic parameters for practical monitoring of natural organic matter. Desalination, 176, 47.

33.

Uyguner-Demirel

, and Bekbolet

(2011). Significance of analytical parameters for the understanding of natural organic matter in relation to photocatalytic oxidation. Chemosphere, 84, 1009.

34.

Valencia

, Cataño

, Rios

, Restrepo

, and Marín

(2011a). A new kinetic model for heterogeneous photocatalysis with titanium dioxide: Case of non-specific adsorption considering back reaction. Appl. Catal. B Environ., 104, 300.

35.

Valencia

, Marín

, Restrepo

, and Frimmel

F.H.

(2011b). Evaluation of photocatalytic degradation of a commercial humic acid in water using a simulated solar UV irradiation and monitoring the changes by size exclusion chromatography. Water Sci. Technol., 11, 692.

36.

Valencia

, Marín

, and Restrepo

(2013a). Method for trihalomethane analysis in drinking water by solid-phase microextration with gas chromatography and mass spectrometry detection. Water Sci. Technol., 13, 499.

37.

Valencia

, Marin

, Restrepo

, and Frimmel

F.H.

(2013b). Evaluation of the TiO₂/simulated solar UV degradation of XAD fractions of natural organic matter from a bog lake using size-exclusion chromatography. Water Res. 47, 5130.

38.

Valencia

, Marin

, Restrepo

, and Frimmel

F.H.

(2014). Evaluation of natural organic matter changes from Lake Hohloh by three-dimensional excitation/emission matrix fluorescence spectroscopy during TiO₂/UV process. Water Res. 51, 124.

39.

Valencia

, Marín

, Velásquez

, Restrepo

, and Frimmel

F.H.

(2012). Study of pH effects on the evolution of properties of Brown-water natural organic matter as revealed by size-exclusion chromatography during photocatalytic degradation. Water Res. 46, 1198.

40.

Velásquez

, Valencia

, Rios

, Restrepo

, and Marín

(2012). Characterization and photocatalytic evaluation of polypropylene and polyethylene pellets coated with P25 TiO₂ using the controlled temperature embedding method. J. Chem. Eng., 203, 398.

41.

Wang

J. J.

, Dahlgren

, Ersan

, Karanfil

and Chow

(2016). Temporal variations of disinfections by products precursors in wildfire detritus. Water Res. 99, 66.

42.

Zhang

, Du

A.J.

, Lee

, Sun

D.D.

, and Leckie

J.O.

(2008). TiO₂ nanowire membrane for concurrent filtration and photocatalytic oxidation of humic acids in water. J. Memb. Sci., 313, 44.

Photocatalytic Degradation of Humic Acids with Titanium Dioxide Embedded into Polyethylene Pellets to Enhance the Postrecovery of Catalyst

Abstract

Abstract

Introduction

Materials and Methods

Materials

Controlled-temperature embedding method for TiO2 coating on polyethylene pellets (PE-TiO2)

Characterization, erosion, and reuse tests of PE-TiO2

Photocatalytic reactor

Aldrich-HA solution preparation

DOC analysis, UV-visible spectroscopy, and THMFP analysis

Results and Discussion

Material characterization

Erosion test analysis

Photolysis experiments of Aldrich-HA

Adsorption experiments of Aldrich-HA onto PE-TiO2 and TiO2

Photocatalytic degradation experiments on Aldrich-HA with TiO2 in slurry and PE-TiO2

PE-TiO2 reuse test analysis for Aldrich-HA photocatalytic degradation

Formation of DBPs

Conclusions

Footnotes

Acknowledgments

Author Disclosure Statement

References

Controlled-temperature embedding method for TiO₂ coating on polyethylene pellets (PE-TiO₂)

Characterization, erosion, and reuse tests of PE-TiO₂

Adsorption experiments of Aldrich-HA onto PE-TiO₂ and TiO₂

Photocatalytic degradation experiments on Aldrich-HA with TiO₂ in slurry and PE-TiO₂

PE-TiO₂ reuse test analysis for Aldrich-HA photocatalytic degradation