Supported Pt–Cu bimetallic catalysts: preparation and synergic effects in their catalytic oxidative degradation of aniline

Abstract

Catalytic Fenton oxidation is an effective way to remove organic pollutants in water, and the performance of the catalyst is a key issue for the competiveness of this method. In this work, various supported bimetallic Pt–Cu catalysts were prepared by different impregnation methods and their performances for catalytic Fenton oxidation of aniline in water were investigated. In the different impregnation methods employed, factors including the reduction method of the metal precursor, type of catalytic support, and loading of metal were investigated. The effect of different reduction methods on actual loadings of the active components on the supported Pt–Cu catalysts showed the order of (i) H₂ reduction > (ii) liquid phase methanal reduction. Meanwhile, compared with the monometallic catalysts, the Pt–Cu alloy phase (mainly in the form of PtCu₃) was generated and the specific surface area was significantly reduced for the bimetallic catalysts. In the process of Fenton catalytic oxidation of aniline, it was found that most of the prepared catalysts had a certain catalytic activity for H₂O₂ accompanied with aniline degradation. It was found that Pt_0.5Cu_1.5/AC (where AC denotes activated carbon) exhibited superb catalytic activity compared with all other prepared catalysts. In particular, aniline was almost completely mineralized in a neutral solution (500 mg L⁻¹ aniline, 0.098 mol L⁻¹ H₂O₂) after 60 min at 50 °C using Pt–Cu/AC (Pt: 0.5%, Cu: 1.5%). The characterization results showed that the Pt and Cu components were rather evenly distributed on the AC support for this catalyst. More importantly, there was an obvious synergic effect on the supported bimetallic catalyst between the Pt and Cu components for the catalytic oxidation of aniline.

---

1. Introduction

In recent years, environmental accidents caused by aniline pollution¹ have frequently occurred, and thus the treatment of aniline wastewater has become a hot topic in the research field of water treatment.^2,3 The development of economical, environmentally friendly and efficient treatment technologies are undoubtedly always desirable for disposal of waste water in general and of that containing aniline in particular.⁴ In general, the treatment technologies for wastewater containing organic pollutants have a long history of research and practice, and they can be roughly classified into physical, chemical and biochemical methods.^5,6 The traditional physical and biochemical treatment methods always require a rather long time to degrade organic compounds in wastewater. Owing to its fast speed and ability to achieve rather complete degradation of organic pollutants, chemical treatment method casts more and more attentions,⁷ like incineration,⁸ chemical oxidation,⁹ advanced oxidation processes (AOPs).¹⁰ AOPs including Fenton oxidation, catalytic-ozone oxidation and photocatalytic oxidation,^11–14 share a general principle that highly-active oxygenated intermediate like ˙OH radicals are generated to oxidize the organic pollutants efficiently and thoroughly.^15–17 In particular, Fenton oxidation method is suitable for degrading a variety of non-biodegradable organic substances.^18,19 However, the traditional Fenton oxidation method suffers from the following problems, such as requiring a low aqueous pH value (typically around 3), the presence of a large amount of metal ions in the treated solution leading to a second-time contamination, insufficient amount of ˙OH produced, and poor catalytic activity, etc.^20–22

To overcome the above-mentioned problems, various types of new supported metal catalysts were developed, in particular, transition metals components like Fe and Cu immobilized on porous catalytic supports. For example, Stair et al.²³ loaded Cu onto carbon microspheres using a spray-drying method, and the obtained catalysts presented excellent performance for the Fenton oxidative degradation of methyl orange, methyl blue, and rhodamine B in aqueous solutions. Yao et al.²⁴ reported that FeO_x/SiO₂ catalyst could degrade aniline to a low concentration (1 mmol L⁻¹). In addition, the conversion of aniline was 79% at pH = 3 and at the temperature of 30 °C after 160 min of reaction. In whole, novel supported metal catalysts have good potentials for degradation as well as mineralization of refractory organic pollutants. Researches on supported bimetallic Fenton-like systems are being carried out by using one metal catalyst after another, and currently there are only a few researches focusing on bi-transition metallic catalysts. For example, Choi et al.²⁵ used Fe/Al as a heterogeneous Fenton catalyst for the oxidation of acetone, achieving an acetone conversion rate of up to 78.5%. The reason for the high conversion is that different metal ions may show a synergic catalytic effect in the bimetallic multi-type Fenton system.²⁶ At the same time, it has a certain effect on suppressing the dissolution of catalytic metal ions.²⁷ However, the researches focusing on the supported catalysts containing one noble metal component and another transition metal component are sparse. As is known, supported bimetallic catalysts containing precious metals often show good synergistic effects for reactions other than Fenton oxidation even though the metal loadings were extremely low.²⁸ For instance, synergy between Pt and Re was found in carbon-supported Pt, Re, and Pt–Re catalysts, which were used in aqueous phase reforming (APR) of glycerol and water gas shift (WGS) reactions.²⁹

In order to degrade high concentration of aniline in water, various strategies have been adopted to improve the activity of heterogeneous Fenton catalysts. For example, Liu et al.³⁰ reported that Ni–Fe oxalic acid complex catalyst (the total loadings of Ni + Fe being about 31%) was used for the degradation of 20 mg L⁻¹ aniline. The removal efficiency of aniline was ∼100% at pH = 5.4 after a 35 minute reaction, and the total organic carbon (TOC) removal rate was 88%. According to report in the literature,^31–34 other metal components can be added to modify the supported precious metal catalysts. For example, Xie et al. found that the interaction of Pt and other metals was beneficial for improving the exposure rate of Pt nanoparticles on catalytic supports. The low-temperature reduction method could also prevent the high-temperature agglomeration of Pt particles.³⁴

This work aims at providing new supported bimetallic catalyst for improved efficiency for Fenton oxidation of high-concentration aniline in water. Since both of Pt and Cu are often used in catalytic oxidation of organic compounds,^35–38 these two metal elements were selected in this work to prepared supported bimetallic catalysts, partly for understanding whether they can present a synergic effect for Fenton oxidation of aniline. From the aspect of catalyst preparation method, the equal volume impregnation method^39,40 under different conditions (including different types of catalytic support, amounts of metal loading and reduction methods) were examined. By developing new reduction method other than hydrogen reduction, highly dispersed metal components can be obtained on the catalyst surface, which can be beneficial to aniline oxidation. To the best of our knowledge, the preparation of supported Pt catalyst using imidazolidinyl urea as reductant, and the Pt-based supported bimetallic catalyst used for the oxidative degradation of aniline were both reported for the first time. The catalytic results reported in this work provide a potentially hopeful method for efficient removal of aniline pollutant in water.

2. Experimental section

2.1 Materials

Chloroplatinic acid (H₂PtCl₆·6H₂O, AR, Pt > 37.5%), copper nitrate (Cu(NO₃)₂·3H₂O, AR, 99%), iron nitrate (Fe(NO₃)₃·9H₂O, AR), methanal (CH₂O, AR), potassium permanganate (KMnO₄, ≥99.5%), sodium hydroxide (NaOH, ≥96%), and aqueous hydrogen peroxide solution (H₂O₂, AR, H₂O₂ being 30 wt%) were provided by Sinopharm Chemical Reagent Co, Ltd. Imidazolidinyl urea (C₁₁H₁₆N₈O₈, AR, 98%) was provided by Aladdin Reagent (Shanghai) Co, Ltd. Alumina (Al₂O₃) particles were provided by Yantai Henghui Chemical Co, Ltd. SiO₂ (20–40 mesh) particles were provided by Qingdao Ocean Chemical. Activated carbon (20–40 mesh) was provided by FuJian Xinsen Carbon Co, Ltd. All chemicals were of analytical purity and used without further treatment. Deionized (DI) water was applied for the whole experiment.

2.2 Catalyst preparation

Supported Pt–Cu catalysts were synthesized using an equal volume impregnation method with the following steps, and the routine procedures can be found elsewhere in the literature.³⁵ As a typical example, the preparation procedure of a Pt_1.5Cu_0.5/Al₂O₃ (the subscripts of 1.5 and 0.5 representing roughly the weight percents of Pt and Cu, respectively) catalyst containing 5 steps is described as follows.

First, 4.0 g Al₂O₃ (being 20–40 mesh particles) was placed in a crucible, transferred into a muffle furnace and heated to 500 °C in air for 4 h to remove moisture and organic impurities. The calcined Al₂O₃ particles were placed in a clean plastic ziplock bag after cooling down. This step can be named as a pre-calcination step.

The second step was to impregnate the Cu component into Al₂O₃. For each gram of precalcined Al₂O₃, Cu(NO₃)₂·3H₂O of 0.0193 g was dissolved in 0.90 g DI water, and the obtained Cu(II) solution was mixed with the Al₂O₃ particles. It should be noted that the volume of water in above the Cu(II) solution was equal to the water absorbed by amount of the calcined Al₂O₃ (0.90 g water can be absorbed by per gram of Al₂O₃).

In the third step of metal component reduction, Al₂O₃ particles impregnated with the Cu components were kept at room temperature for 12 hours and then were treated in a H₂ flow at 300 °C for 4 h in order to reduce the Cu components. At this stage, the obtained sample can be named as Cu_0.5/Al₂O₃.

The fourth step was to impregnate the Pt component to Cu_0.5/Al₂O₃, which was similar to the second step, with the only differences in that 0.0406 g H₂PtCl₆·6H₂O took the place of 0.0193 g Cu(NO₃)₂·3H₂O.

The fifth step was to reduce the Pt component, and the treatment process was almost the same to the third step of Cu-component reduction. Finally, the catalyst obtained by these steps was represented as Pt_1.5Cu_0.5/Al₂O₃, where the percentages, indicated in the subscripts in the catalyst notation, were calculated from the mass of the Pt or Cu element relative to that of the Al₂O₃ support.

Different from the above-mentioned procedure (called standard preparation procedure), one or more of the preparation conditions including impregnation amount, bimetallic type, impregnation ratio of two metals, reduction temperature, reduction method, and type of catalytic support were changed to obtain different catalysts. These preparation methods are described as follows by comparing with the above-described standard preparation procedures.

For Pt_nCu_m/Al₂O₃ catalyst: change the amount of Cu and Pt sources in steps 2 and 4, respectively, to obtain the target weight percent of n% and m%, where (n + m) was always kept at 2.0. The detailed n and n value will be specified in the text hereafter.

For Pt_nCu_m/Al₂O₃-600 catalysts: compared to the case of Pt_nCu_m/Al₂O₃, the H₂ treatment temperature was 600 °C instead of 300 °C in steps 3 & 5.

For Pt_nCu_m/SiO₂ catalysts: use SiO₂ instead of Al₂O₃ particle as catalytic support.

For Pt_nFe_m/Al₂O₃ catalysts: use Fe(NO₃)₃·9H₂O instead of Cu(NO₃)₃·3H₂O in the second step in order to impregnate the Fe component into the catalytic support.

For Pt_nCu_m/AC catalysts: use activated carbon instead of Al₂O₃ particle as the catalytic support, and the pre-calcination temperature was 300 °C in step 1.

For Pt_nCu_m/AC-MR catalysts: compared to the case of Pt_nCu_m/AC, in the fifth step, methanal reduction (MR) method was developed in this work. The Pt component impregnated AC particles was reduced at 80 °C in a solution containing imidazolidinyl urea having the same number of mole as the Pt source, excess amount of methanal, and having a pH value of 11. Then the particles were washed with a large amount of water and then dried at 120 °C over night. The reduction method will be named as methanal reduction method hereafter.

2.3 Catalyst characterization

The crystal phase structure of supported metal catalysts was measured by a powder X-ray diffractometer (XRD, D8 Advance, Germany) using a Cu K_α radiation (λ = 0.15418 nm) having a power of 2.2 kW with a scanning step of 0.02° and a 2θ range of 2–75°.

In a temperature-programmed reduction by H₂ (H₂-TPR) measurement of a certain catalyst, the H₂ signal was monitored by a thermal conductivity detector (TCD). First, the catalyst was pre-treated for a period of time (∼30 min) in an Ar flow, then switched to a 20 sccm flow of 10% H₂–He for 20 min after cooling down to room temperature, and then heated up to 800 °C at a rate of 10 °C min⁻¹. The TCD pool was kept at 60 °C & 30 mA.

Scanning electron microscope (SEM) and energy spectrometer (SEM-EDX) were used to analyze the morphology of the sample, and the element distribution by mapping. The working voltage of the instrument was 10.0 kV, the solid sample was more than 10 mg and needed to be dried and processed. Transmission electron microscopy (TEM) projected an accelerated and concentrated electron beam onto a very thin sample to collide, resulting in solid angle scattering, and get an image of light and dark. It was used to observe the morphology and dispersion of nanoparticles and measure the particle size of nanoparticles.

The N₂ adsorption–desorption curves of samples were obtained by specific surface area and microporous physical adsorption analysis. Brunauer–Emmett–Teller (BET), Barret–Joyner–Halenda (BJH) were used to calculate the specific surface area and pore size distribution.

Thermogravimetric (TG) analyzed the decomposition temperature of samples. TG experiments were carried out in N₂ flow (60 mL min⁻¹), and heated from 40 °C to 800 °C at a heating rate of 10 °C min⁻¹.

The prepared catalysts were ground to >200 mesh particles and then X-ray fluorescence (XRF, PANalytical, Netherlands) spectroscopy was used to measure the Si, Al, Pt or Cu contents of the catalysts.

2.4 Catalytic Fenton oxidation of aniline

The Fenton oxidation of aniline was performed in a 250 mL flask equipped with a water bath thermostat as the reactor. In a typical oxidation experiment, a 200 mL aniline aqueous solution (500 mg L⁻¹) was added to the reactor. The temperature of the reaction system was heated to 50 °C by a water bath, and then a 2.0 mL H₂O₂ solution (30 wt%) was added to the above aniline solution (thus the concentration of H₂O₂ being 0.087 mol L⁻¹ at the beginning of the reaction). At the absence of a catalyst, it was confirmed that there was no observable reaction between H₂O₂ and aniline at the reaction temperature. Then 1.0 g selected catalyst (or 5.0 g_cat L_water⁻¹) was added to the reactor in order to initiate the catalytic oxidation reaction.

A 2.0 mL liquid was sampled from the reaction mixture at certain time intervals, and then centrifuged to separate solid residue. The liquid sample was then analyzed by a HACH DR 1010 chemical oxygen demand (COD) analyzer or a TOC-LCPH/CPN total organic carbon content (TOC) analyzer. The catalytic performance of a certain catalyst was evaluated by calculating the COD or TOC removal rate of the reaction mixture where the catalyst was used.

2.5 Catalytic decomposition of H₂O₂

The catalytic H₂O₂ decomposition was performed with the same reaction conditions mentioned above (Section 2.4), with the only difference in that the 500 mg L⁻¹ aniline solution was replaced by DI water of a same volume.

At a certain reaction time, a 2 mL sampled solution was mixed with a 2 mL 3 mol L⁻¹ sulfuric acid. The concentration of H₂O₂ was determined by acidic KMnO₄ (5 mol L⁻¹) titration, according to the volume of KMnO₄ solution consumed. The titration reaction principle⁴¹ can be expressed as follows:

2MnO₄⁻ + 5H₂O₂ + 6H⁺ → 2Mn²⁺ + 5O₂ + 8H₂O

3. Results and discussion

3.1 Results about the novel preparation method of the Pt–Cu/AC bimetal catalysts

It is noticeable that the impregnation and metal component reduction methods for preparing the Pt–Cu/AC catalysts (see Section 2.2 for detail) is reported for the first time. The main motivation for us to develop a new liquid phase reduction method came from the case that the activated carbon (AC) support may tend to be damaged at high temperatures. In addition, treatment with imidazolium urea solution before reduction can decrease the risk of agglomeration of metal nanoparticles at high temperatures, which could make the Pt and Cu active components more uniformly dispersed on the catalytic supports. This hypothesis was further verified by the TEM images of catalyst as shown later in this paper.

Table 1 shows that the metal content losses of the Pt–Cu/AC catalysts prepared with the methanal reduction (MR) method were higher than those prepared with the H₂ reduction method. The Pt and Cu loading rates (loading rate = actual loading/theoretical loading calculated from the experimental conditions) in the former case were about 88% and 77%, respectively, which are close to 100% for the latter case. Possible reasons accounted for this phenomenon could be as follows. (i) The times for ultrasonication during impregnation and liquid phase reduction were insufficient. (ii) The pores of catalytic supports became narrower after the first impregnation and reduction steps, which hindered the loading of the second metal component.

Table 1 The actual Pt and Cu loadings of the Pt–Cu/AC catalysts measured by XRF spectroscopy. See Section 2.2 for catalyst preparation procedure as well

Catalyst	Pt wt%	Cu wt%
Pt1.5Cu0.5/AC-MR	1.329	0.386
Pt1.0Cu1.0/AC-MR	0.894	0.775
Pt0.5Cu1.5/AC-MR	0.445	1.275
Pt1.5Cu0.5/AC	1.529	0.446
Pt1.0Cu1.0/AC	0.994	0.965
Pt0.5Cu1.5/AC	0.505	1.475

---

3.2 Catalyst characterization results for physiochemical properties of catalysts

The H₂-TPR profiles presented in Fig. 1a shows that the reduction temperature of the unreduced Cu_2.0/Al₂O₃ catalyst started from ∼380 °C with the peak around ∼600 °C. The curves in Fig. 1b and c indicate that the metal components on the prepared supported Cu and Pt/Al₂O₃ catalysts was reduced to their metallic states. Xi et al. had reported that the reduction conditions of the 2% Pt/Al₂O₃ catalyst³⁵ reduced in presence of H₂ at 300 °C for 4 h, which is consistent in the results in Fig. 1c.

Fig. 1 The H₂-TPR profiles of the (a) Cu_2.0/Al₂O₃ (before H₂ reduction) (b) Cu_2.0/Al₂O₃ (after H₂ reduction) and (c) Pt_2.0/Al₂O₃ (after H₂ reduction) catalysts.

Fig. 2 shows the XRD patterns of some selected catalysts. Inspection of Fig. 2a shows that, compared with the data of the reported standard XRD pattern in Table 2,^42–47 both curves (i) and (ii) were correspond to the characteristic peaks of γ-Al₂O₃. Therefore the main crystal phase of the catalytic supports after calcination was γ-Al₂O₃. Compared with Fig. 2c, it was found that the peaks of catalyst were changed from sharp to broad, showing that Al₂O₃ in the catalyst was amorphous alumina. Comparing between pictorial panels Fig. 2a and b, it can be seen that the catalysts with different reduction temperatures had different peak area and width, but the other characteristic peak positions were basically the same. Combined with the H₂-TPR results in Fig. 1, it can be further illustrated that the reduction temperature had an effect on the complete reduction of Cu. For curves (ii), (iii), and (iv) in Fig. 2a, the matching characteristic diffraction peaks of Pt and Cu were not obvious. It also indirectly reflects that these components had low crystallinity. In combination with Table 2 which contains the standard peak positions of selected materials, a peak of the Pt–Cu compound indicating that a Pt–Cu alloy was formed, indicating that there was an interaction between the Pt and Cu components. Although the overall metal loadings for the cases of (ii), (iii) and (iv) in Fig. 2a were the same, the catalytic supports were different. Thus the width and area of the PtCu alloy peaks were different. This shows that the degree of the interactions between the Pt and Cu components were different, following the order of Pt–Cu/AC > Pt–Cu/SiO₂ > Pt–Cu/Al₂O₃. Fig. 2b shows that the peak width and intensity of the AC supports in the cases curve (i) and (ii) were basically unchanged before and after pre-calcination, indicating that the AC crystal phase did not have a noticeable change during calcination.

Fig. 2 XRD patterns of the prepared catalysts. In panel (a), curve (i): Pt_1.0Cu_1.0/Al₂O₃, (ii): Pt_1.0Cu_1.0/Al₂O₃-600, (iii): Pt_1.0Cu_1.0/SiO₂ and (iv): Pt_1.0-Cu_1.0/AC. In panel (b), curve (i): pure AC calcined at 300 °C for 4 h, (ii): uncalcined AC, (iii): Pt_1.5Cu_0.5/AC, (iv): Pt_1.0Cu_1.0/AC and (v): Pt_0.5Cu_1.5/AC. In panel (c), curve (i): Pt_1.5Cu_0.5/Al₂O₃, (ii): Pt_1.0Cu_1.0/Al₂O₃, (iii): Pt_0.5Cu_1.5/Al₂O₃, (iv): Al₂O₃ (calcined at 500 °Cfor 4 h), and (v): uncalcined Al₂O₃.

Table 2 The typical X-ray diffraction peak position^38–43 of AC, γ-Al₂O₃, Pt, Cu, Pt₁Cu₁, and Pt₁Cu₃

Peak position		1	2	3	4
γ-Al2O3	2θ/(°)	37.71	39.45	45.87	66.89
AC	2θ/(°)			43.92	51.28
Pt	2θ/(°)		39.79	46.28	67.53
Pt1Cu3	2θ/(°)			42.24
Pt1Cu1	2θ/(°)			41.01
Cu	2θ/(°)			43.30

---

The N₂ adsorption–desorption isotherms, calculated BET surface area, average pore diameter and pore volumes of the prepared catalysts are shown in Fig. 3 and in Table 3. It can be noticed that the specific surface area of supports become larger after calcination, indicating that it was necessary to be pre-calcinated for the catalytic supports. As shown in Fig. 3, the nitrogen adsorption/desorption isotherms of the Pt–Cu catalysts on AC/Al₂O₃ and the calcined AC/Al₂O₃ supports belonged to the IUPAC IV and H1 type hysteresis loops, respectively. This shows that the catalytic materials were mainly mesoporous in structure and the pore size range was uniformly narrow.^48,49 The N₂ adsorption of the material becomes smaller after impregnation and metal reduction. The pore size of Pt–Cu/AC obtained from MR and the calcined supports was mainly between 2-4 nm, while the pore size of Pt–Cu/Al₂O₃ obtained from the H₂ reduction and the calcined Al₂O₃ was between 7–10 nm, as shown in Panels b, e and f in Fig. 3. This was due to AC has a greater specific surface area. The higher loading of Cu the catalyst had, the larger maximum pore diameter of the catalysts was, which may be related to the difference of particle size caused by incomplete reduction of Cu.

Fig. 3 N₂ adsorption–desorption isotherms and pore size distributions of selected catalysts as indicated in the figure. (a) Isotherms for pure AC and AC supported Pt–Cu catalyst. (b) Pore size distributions analyzed from (a). (c) and (d) Isotherms for pure Al₂O₃ and Al₂O₃ supported Pt–Cu catalysts. (e) and (f) Pore size distributions analyzed from (c) and (d), respectively.

Table 3 The specific surface area and pore size of the selected prepared catalysts in this work from their N₂ adsorption–desorption isotherms

Sample	Specific surface area SBET (m^2 g^−1)	Total pore volume (cm^3 g^−1)	Average pore size (nm)
AC (uncalcined)	1137	0.29	2.74
AC (calcined at 300 °C, 4 h)	1415	0.30	2.71
Pt1.5Cu0.5/AC-MR	1030	0.25	2.72
Pt1.0Cu1.0/AC-MR	1029	0.24	2.74
Pt0.5Cu1.5/AC-MR	1028	0.23	2.76
Pt0.5Cu1.5/Al2O3	192	0.60	8.26
Pt0.5Cu1.5/SiO2	437	0.14	2.56
Al2O3 (uncalcined)	195	0.57	7.50
Al2O3 (calcined at 500 °C, 4 h)	228	0.64	7.76
Pt2.0/Al2O3	216	0.62	7.85
Pt1.5Cu0.5/Al2O3	205	0.62	7.83
Pt1.0Cu1.0/Al2O3	202	0.62	7.99
Pt0.5Cu1.5/Al2O3	202	0.61	8.05

---

In general, the specific surface area of AC was increased from 1137 m² g⁻¹ to 1415 m² g⁻¹ after calcination at 300 °C for 4 h, while the pore volume and average diameter were unchanged. However the specific surface area of Al₂O₃ increased from 195 m² g⁻¹ to 228 m² g⁻¹ after calcination at 500 °C for 4 h, and both of the pore volume and average diameter had a mild increase.

The pore volume and pore diameter of the catalysts with different supports obtained by the two-stage reduction had mild changes, reflecting that the platinum and copper entered the surface and the interior of the catalyst and changed the pore structure and specific surface area of the catalyst. While the difference between the specific surface area of the AC and Pt–Cu/AC was larger than the Al₂O₃ cases, indicating that AC was somehow destroyed at high temperatures. The specific surface area, pore volume and diameter of the supported Pt and Cu catalysts obtained from the methanal reduction changed little. The reason was that the specific surface area of AC was much larger than that of Al₂O₃ and SiO₂, and the pore volume and pore size are smaller than those of Al₂O₃ and this reduction method done little damage to the structural. Therefore, the Pt and Cu components were dispersed uniformly. For these catalysts, the order of specific surface area was: AC ≫ SiO₂ > Al₂O₃, and pore volume and pore size was: Al₂O₃ > AC > SiO₂. For Pt–Cu/Al₂O₃, when reduced at 300 °C and 600 °C in the H₂ atmosphere, the specific surface area was reduced from 202 m² g⁻¹ to 192 m² g⁻¹. The pore volume was basically unchanged, but the pore diameter was increased. This may be related to the reduction at high temperatures.

The TEM images of the selected catalysts are shown in Fig. 4. For the catalyst obtained from MR, the Pt particle size follows the order of Pt_1.0Cu_1.0/AC-MR > Pt_0.5Cu_1.5/AC-MR > Pt_1.5Cu_0.5/AC-MR > Pt_2.0/AC-MR. In all four cases the Pt or PtCu particles were small and uniformly dispersed. The particle size of the Pt particles was in the range of 1–4 nm. In particular, the metal particle size of catalyst Pt_2.0/AC-MR mainly distributed in the 1–2 nm range, while those of Pt–Cu/AC catalysts presented a larger range of particle size mainly within 1–4 nm. This may be related to the formation of the Pt–Cu alloy, which made the metal particle size larger. In addition, the amount of Pt–Cu alloy with different loadings could be different, making the Pt particle size different.

Fig. 4 The particle size charts and TEM images of (a) Pt_2.0/AC-MR, (b) Pt_0.5Cu_1.5/AC-MR, (c) Pt_1.0Cu_1.0/AC-MR, (d) Pt_1.5Cu_0.5/AC-MR, (e) Pt_0.5Cu_1.5/SiO₂ and (f) Pt_0.5Cu_1.5/Al₂O₃.

The choice of catalytic support had a significant influence on the particle size of the active metal component for the catalysts with a same loading. The Pt particle sizes were in the range of 10–50 nm for Pt_0.5Cu_1.5/Al₂O₃, being much larger than those for the Pt_0.5Cu_1.5/SiO₂ case (2–8 nm). The Pt particle sizes were uneven for both cases, which was dramatically different to the Pt–Cu/AC cases. The Pt particle size distribution for the Al₂O₃ and SiO₂ cases was wide, which may be related to the nature of the support as well as the reduction method.

As can be seen in the following section about the results of catalytic reaction test, the Pt_0.5Cu_1.5/AC-MR catalyst presented desirably good catalytic performance. So, more characterization results focusing on the Pt_0.5Cu_1.5 catalysts were performed, with the results shown in the ESI† including the TG-DSC curves, SEM images and TEM-EDX elemental mappings.

3.3 Catalytic Fenton oxidation of aniline by H₂O₂

3.3.1 Effect of hydrogen peroxide concentration. Hydrogen peroxide is the direct source of reactive intermediates like ˙OH⁵⁰ in the catalytic Fenton oxidation system in this paper. From Fig. 5a, it can be seen that the H₂O₂ concentration in the reaction system had an obvious influence on the catalytic degradation of aniline. For a same Pt–Cu/AC-MR catalyst, the COD removal rate follows the order 0.049 mol L⁻¹ H₂O₂ (1 mL 30 wt% H₂O₂ added to the reaction system, see Section 2.4) ≪ 0.098 mol L⁻¹ (2 mL) ≈ 0.196 mol L⁻¹ (4 mL). For different Pt and Cu supported catalysts, the content of Pt, Cu and theirs alloys were different, resulting in different catalytic degradation effects of aniline (Fig. 5b–d).

Fig. 5 Effects of different catalysts and H₂O₂ concentration on catalytic Fenton oxidation of aniline. (a) Effect of different Pt_nCu_m/AC-MR catalyst and H₂O₂ concentration on the removal of COD in the catalyzed oxidation of aniline after 3 h. (b) Effect of added volume of 30 wt% H₂O₂ to the reaction system on the removal of COD using the Pt_0.5Cu_1.5/AC-MR catalyst at different reaction times. (c) Effect of added volume of 30 wt% H₂O₂ to the reaction system on the removal of COD using the Pt_1.0Cu_1.0/AC-MR catalyst. (d) Effect of added volume of 30 wt% H₂O₂ to the reaction system on the removal of COD using the Pt_1.5Cu_0.5/AC-MR catalyst. *Reaction conditions: at the initial time of reaction, H₂O₂ concentration was 0.196, 0.098 and 0.049 mol L⁻¹ with the added volume of 1.0, 2.0 and 4.0 mL, aniline concentration = 500 mg L⁻¹, catalyst dosage = 5 g L⁻¹ temperature = 50 °C.

When the amount of H₂O₂ was small, the amount of ˙OH generated would be also small. On the other hand, the catalyst catalyzed the pure decomposition of H₂O₂ (see in Fig. 6), making the amount of ˙OH effective for the catalytic oxidation of aniline further smaller. However, when the H₂O₂ concentration was too high, it also inhibited the production of ˙OH.⁵¹ Therefore, there should be a most suitable H₂O₂ concentration for aniline degradation reaction system, which was 0.098 mol L⁻¹ in this work.

Fig. 6 H₂O₂ conversion in terms of volume of KMnO₄ consuming at the absence of aniline, at 50 °C and using 5 g L⁻¹ of catalyst as indicated in the figure. See the experimental conditions in Section 2.5 in detail as well. (a) The Pt_nCu_m/Al₂O₃ catalysts, where n% + m% = 1.0%, (b) the Pt_nCu_m/Al₂O₃ catalysts, where n% + m% = 2.0%, (c) the Pt_nCu_m/AC-MR catalysts, where n% + m% = 2.0%.

The concentration of H₂O₂ had a great influence on the COD value, especially in the initial stage of reaction, and thus it also affected the apparent COD removal rate. It can be anticipated that H₂O₂ will also be decomposed catalytically in the presence of Pt–Cu catalysts, which intrigued us to test the performance of our prepared catalysts for pure decomposition of H₂O₂ at the absence of aniline in water.

It can be seen from Fig. 6 that for the Pt–Cu/Al₂O₃ catalysts the influence of the loading amounts of Pt and Cu components prepared under a same reduction method on the catalytic degradation of H₂O₂ was huge (panel a vs. b). When the loadings of Pt and Cu were fixed, the Pt–Cu bimetal catalysts with different supports carriers also had a great influence on the catalytic degradation of H₂O₂ (panel b vs. c). For all the Pt–Cu catalysts, the rate of catalytic decomposition of H₂O₂ always increased with the content of Pt. The decomposition rate of H₂O₂ was quite slow and basically unchanged after 30 minutes of reaction for all cases. This may be related to the rapid catalytic reaction caused by the high activity of Pt, which may also explain the rapid decrease of COD value during the degradation of aniline for 1 h in Fig. 5.

3.3.2 Effect of type of support. Since the value of COD cannot completely represent the conversion rate of aniline, the TOC value was also used for better evaluation. As can be seen from Fig. 7a, all catalysts prepared from three different catalytic supports had good catalytic performances. However, there was a certain deviation between the TOC removal rate of the catalysts prepared by Al₂O₃ and SiO₂ as supports with the order of SiO₂ > Al₂O₃. The TOC removal rate of Pt–Cu/AC-MR was the highest in three types of support, reaching above 97.5% (Fig. 7a). It was also found that the type of support and the reduction method had a great influence on removal rate of COD (Fig. 7c–e), with the order of COD removal rate of AC > SiO₂ > Al₂O₃ observed in this work. For three different Pt/Cu ratios on the Pt–Cu/AC catalysts, the COD removal rate was also different, with Pt_0.5Cu_1.5/AC-MR catalyst showing the best performance.

Fig. 7 The catalytic test results for different catalysts in terms of TOC or COD value of aniline at the presence of H₂O₂. (a) The TOC removal rate of aniline catalyzed by different catalysts as indicated, (b) the TOC value of catalytic degradation of aniline using different catalysts as indicated, (c) the COD value of catalytic degradation of aniline using Pt_0.5Cu_1.5/AC-MR, Pt_0.5Cu_1.5/Al₂O₃ and Pt_0.5Cu_1.5/SiO₂, respectively, as the catalysts (d) the COD value of catalytic degradation of aniline using Pt_1.0Cu_1.0/AC-MR, Pt_1.0Cu_1.0/Al₂O₃ and Pt_1.0Cu_1.0/SiO₂, respectively, as the catalysts (e) the COD value of catalytic degradation of aniline using Pt_1.5Cu_0.5/AC-MR, Pt_1.5Cu_0.5/Al₂O₃ and Pt_1.5Cu_0.5/SiO₂, respectively, as the catalysts. *Reaction conditions: H₂O₂ concentration = 0.098 mol L⁻¹, aniline concentration = 500 mg L⁻¹, dosage of catalyst = 5 g L⁻¹, temperature = 50 °C, and reaction time = 3 h.

In general, the results shown in Fig. 7 shows that the performance of the Pt–Cu/AC-MR catalysts for aniline degradation was the best one among three cases of catalytic support. The good TOC removal ability was the same as good capability of mineralization rate of aniline (the efficiency of complete degradation to H₂O and CO_x) for the Pt–Cu/AC-MR catalysts. In contrast, the degradation of aniline by using the Pt–Cu/SiO₂ and Pt–Cu/Al₂O₃ catalysts was often not complete, and aniline may be degraded into other organic molecules which stayed in water. This is further supported by an interesting phenomenon described below.

As can be observed with naked eyes, the aniline solution was transparent almost during the entire degradation process when AC-supported catalysts were used. While SiO₂ was used as the catalytic support, the solution became brown quickly. Suspended black materials can be observed in the solution when Al₂O₃ was used. According to ref. 52, the oxidative degradation of aniline affords compounds in brown at the first stage, and then affords compounds in black, until it was completely degraded. This phenomenon further demonstrated that the choice of catalytic support was important, and the performances for degradation of aniline showed the order of AC > Al₂O₃ > SiO₂. This may be associable with different pore structures in different supports (Table 3), and the reduction method (see in Section 2.2) used affecting the dispersion of Pt and Cu components and generation of Pt, Cu and their alloys (see in Section 3.2).

3.3.3 Effect of metal loading and metal type. The results in Fig. 8 shows that the performance of different catalysts with a same support and a same overall metal loading followed the order of Pt–Cu/AC-MR > Pt–Fe/AC-MR > Cu/AC-MR > Pt/AC-MR. This order may be related to the possible synergistic effect between different metal components. At the same time, the performance of catalysts with different metal loadings was also different. However, during the experiment, the dissolution of the Fe component into water seemed to be serious, and therefore the subsequent studies using Fe catalysts did not continued. For the cases of Pt/AC-MR and Cu/AC-MR, the COD value was reduced to the lowest value within 1 h, and gradually increased in the following times, being still lower than the initial value, indicating that aniline was degraded to a certain degree.

Fig. 8 The values of COD and TOC of the solution of Fenton oxidation of aniline using Pt–Cu/AC-MR, Cu/AC-MR, Pt/AC-MR, Pt–Fe/AC-MR as catalysts, respectively. (a) Effect of different catalysts on the COD value, (b) effect of different catalysts on TOC value, and (c) comparison of COD and TOC removals of each catalyst. *Reaction conditions: H₂O₂ concentration = 0.098 mol L⁻¹, aniline concentration = 500 mg L⁻¹, temperature = 50 °C, catalyst dosage = 5 g L⁻¹, degradation time = 3 h, overall metal loading = 2%.

The TOC removal rate was lower than the COD case, indicating that the degradation of aniline was incomplete and some other organic molecules were produced. For the supported monometallic catalysts, COD and TOC removal rate of the Cu/AC catalyst was higher than Pt/AC (Fig. 8b and c). First, the aniline molecules ma be physically adsorbed (vide infra) onto the outer surface of the catalyst and part of them may enter the inner pores, and then were degraded by the catalysis of the active metal components, so the COD value decreased within 1 h. Then, the partly degraded products from aniline returned to the solution after a long time of stirring, leading to the COD value increased slightly again (Fig. 8). Comparison of the results shown in Fig. 8 with those in Fig. 7, it can be seen that the bimetallic Pt–Cu/AC-MR catalysts had a better performance for the catalytic removal of aniline than the monometallic Pt or Cu catalysts, although the overall metal loading were the same.

According to the definition of synergy effect in catalytic chemistry, the formula of synergy index (SI) can be expressed as follows.

Here x is n/(m + n) for the Pt_nCu_m/AC-MR catalyst. According to the data of TOC removal rates in Fig. 7 and 8, the SI_(Pt–Cu) was calculated to be 309–312%, which indicates that synergic effect on the Pt–Cu/AC-MR catalyst could improve the catalytic efficiency by up to 212% for the case of Pt_0.5Cu_1.5/AC-MR. So the synergistic effect of catalytic degradation of aniline was excellent. The possible reasons accounting for this effect may be associable with the dispersion of active components (Fig. 4) as well as the formation of alloy in the metal nanoparticles (see Fig. 2 and Table 2).

3.3.4 Preliminary discussion about the catalytic reaction mechanism of aniline degradation. In order to gain a deep understanding of the above catalytic/adsorption test results for Fenton oxidation of aniline in water, a simple catalytic reaction mechanism (see Scheme 1) was proposed according to the above results.

Scheme 1 Possible reaction mechanism of the reaction between aniline and H₂O₂ at the presence of supported bimetallic Pt–Cu catalysts.

The pathways indicted with k₁–k₃ shown in Scheme 1 result in the oxidative degradation of aniline by reacting with H₂O₂. However, not all aniline molecules can be directly oxidized to mineralized product of gaseous CO/CO₂, and part of the aniline intermediates remained in water, which is accounted for the difference of COD and TOC removal rates. From the catalytic test results shown in Fig. 7 and 8, it can be seen that the catalysts using AC as the support favors k₂ and/or k₃ over k₁ compared to the ones using Al₂O₃ or SiO₂ as supports. The contribution of the physical adsorption processes (k₅) should be further noted during the catalytic degradation of aniline with the involvement of AC, as reflected by the results shown in Fig. 8a. Competing with pathways k₁–k₃, pure catalytic decomposition of H₂O₂ also occurred. As is known, decomposition of H₂O₂ to produce reactive intermediate of HO˙radical is key to oxidize aniline, however, not all reactive intermediate can attack aniline or its intermediate, leading to a pure decomposition process. An increasing amount of Pt was advantageous to the H₂O₂ decomposition (Fig. 6), however, too much Pt component may lead to a larger portion of H₂O₂ decomposition. A good catalyst should be able to give large k₄ and at the same time give large (k₁ + k₃)/k₄. From this point of view, the significant synergic effect of the Pt_0.5Cu_1.5/AC-MR catalyst may be accounted for its ability to accelerate the rate of H₂O₂ decomposition (k₄), and the rate of aniline oxidation (k₁–k₃, in particular, k₁ and k₃) at the same time, with the preference of the latter pathways. A detailed mechanism study of the related catalytic systems are undergoing in our group.

3.3.5 Comparison of the result of this work with other works in the literature. When the reaction time was shortened from 3 h (Fig. 7) to 1 h (Fig. 9), it can been seen that the COD value did not change after 1 h, and the decomposition of H₂O₂ almost stopped at 30 min. These phenomena showed that aniline could be almost completely degraded in 1 h at 50 °C. And Pt_0.5Cu_1.5/AC-MR had best performance on degradation and mineralization, and removal of TOC was more than 99%. The reason for this phenomenon was that the slightly higher loading of Pt has high catalytic activity, and the hydrogen peroxide is rapidly decomposed on the catalyst surface, which affects the further oxidative degradation of the migration of aniline molecules to the inner surface of PtCu/AC-MR. A summary of comparison of the results of our work with other similar works can be seen in Table 4, which showed the superiority of our catalytic systems for oxidative degradation of aniline in water. It is interesting to notice that in our catalytic systems, the pH value of the solution did not required to adjust to a typical case of 3–4 in many Fenton reaction researches, and instead, the solutions were almost neutral in this work. This adds to the appealing advantage of our catalysts for potential application in the Fenton oxidation of aniline in particular and of organic pollution in general.

Fig. 9 The effect of different metal loadings on catalytic degradation of aniline by using Pt–Cu/AC-MR. (a) The COD value of the solution, (b) the TOC value of the solution, (c) the COD and TOC removal rate after 1 h. *Reaction conditions: H₂O₂ concentration = 0.098 mol L⁻¹, aniline concentration = 500 mg L⁻¹, the catalyst dosage = 5 g L⁻¹, degradation reaction time = 1 h, overall metal loading capacity = 2% and reaction temperature = 50 °C.

Table 4 Comparison of the catalytic performance of our catalysts to those in the similar works in the literature for oxidative degradation of aniline in water

Catalyst	TOC or COD removal	Conditions	References
Pt0.5Cu1.5/AC-MR	Up to 99% TOC removal	50 °C, 5 g L^−1 catalyst, 500 mg L^−1 aniline, 0.098 mol L^−1 H2O2, 60 min	This work
5% Ru/SiO2	90% COD	200–220 °C, 0.69 MPa O2, 1.33 g L^−1 catalyst, 500 ppm aniline, 120 min	53
Ru/Ti0.9Zr0.1O2	84.6% COD	180 °C, 1.5 MPa O2, 4 g L^−1 catalyst, 2 g L^−1 aniline, 300 min	54
PAC@Fe^IIFe2^IIIO4	71.2% aniline	25 °C, pH = 6, UV + 5 mmol L^−1H2O2, 0.3 g L^−1 catalyst, 120 mg L^−1 aniline, 180 min	55
10 mg L^−1 Fe^2+	78% COD	25 °C, pH = 8, 2.08 mg min^−1 L^−1O3, 10 mg L^−1 catalyst, 75 mg L^−1 AAF, 180 min	56
Co/Fe(3 : 1)-LDH	88% COD	70 °C, pH = 12, 400 mL min^− 1 O3, 12.5 g L^−1 catalyst, 960 mg L^−1 initial COD, 180 min	57

---

4. Conclusion

This paper reports a systematic investigation of the supported Pt–Cu bimetallic catalysts prepared with various impregnation methods, and their performances for catalytic Fenton oxidation of aniline in water as well. Various preparation conditions including impregnation amount, bimetallic type, impregnation ratio of two metals, reduction temperature, reduction method, and types of catalytic support were changed to obtain different supported catalysts. In particular, the methanal reduction (MR) method developed in this work was reported for the first time to the best of our knowledge.

The XRF analysis results revealed that the MR method led to more metal lost on Pt–Cu/AC than the H₂ reduction method, but the particle diameter was rather small (concentrated in the <3 nm range, by TEM) for the former case. The XRD analysis results disclosed that Pt–Cu alloy can be formed in the bimetallic catalysts. However, the crystal phase of AC was destroyed the specific surface area and was decreased significantly, and the pore volume was slightly increased. The results of BET show that the specific surface area of the catalyst supported on AC was significantly larger than that of Al₂O₃ and SiO₂.

Pt_0.5Cu_1.5/AC-MR presented the highest activity for catalytic degradation of aniline (500 mg L⁻¹) in the neutral aqueous solution among all catalysts examined. The catalytic test results showed that the liquid pollutant was almost completely degraded within 60 minutes at the presence of 0.098 mol L⁻¹ H₂O₂ using this catalyst. The performance of Pt_0.5Cu_1.5/AC-MR was much better than Pt/AC and Cu/AC under the same preparation and reaction conditions, which showed that there was a significant synergic effect between the Pt and Cu components. The synergic effect may lead to an improvement of catalytic efficiency by 212% in this case in terms of TOC removal rate. This effect could play a pivotal role for the degradation of aniline and also provide potential hopefulness for future industrial application.

Author contributions

Qiuyue Ding: catalyst preparation, characterization, reaction test, writing. Wumin Zhang: catalyst preparation, reaction test, analysis, writing. Yuanyuan Zhu: catalyst preparation, characterization. Lu Wang: catalyst preparation, reaction test. Yanyan Xi: catalyst preparation, characterization, discussion. Xufeng Lin: idea development, research funding provision, writing and organization. Xiyuan Feng: catalyst preparation, reaction setup construction.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

Support from the National Natural Science Foundation of China (21576291), and Shandong Province Natural Science Foundation (ZR2014BM002) is gratefully acknowledged.

References

1. S. Shi, J. Cao, L. Feng, W. Liang and L. Zhang, Chin. J. Environ. Eng., 2015, 9, 4871–4876.
2. S. Jiang, J. Zhu, S. Bai, Y. Guan and J. Yao, Desalin. Water Treat., 2016, 57, 791–798.
3. J. Xiang, Z. Zhang, H. Fan and S. Lv, Environ. Sci. Technol., 2010, 33, 81–85.
4. S. Vimalnath, H. Ravishankar, C. Schwandt, R. V. Kumar and S. Subramanian, Water Sci. Technol., 2018, 78, 290–300.
5. W. W. Eckenfelder, Stud. Environ. Sci., 1986, 29, 43–50.
6. B. Wang, X. Wu, Y. Luo and Y. Liu, Ind. Water Treat., 2015, 35, 93–95.
7. M. Kiel, D. Dobslaw and K. H. Engesser, Water Res., 2014, 66, 1–11.
8. C. Comninellis, Electrochim. Acta, 1994, 39, 1857–1862.
9. X. Nie, T. Huang, J. Zhang, S. Zhang, H. Chen, L. Ding and L. Liu, Huanjing Kexue, 2012, 32, 274–278.
10. J. Yoon, Y. Lee and S. Kim, Water Sci. Technol., 2001, 44, 15–21.
11. W. Z. Tang and C. P. Huang, Environ. Technol., 1996, 17, 1371–1378.
12. W. Y. Zhao, X. F. Zheng, L. W. Xu and G. Y. Yang, Adv. Mat. Res., 2013, 726, 1710–1714.
13. Z. Y. Li, W. Song and L. Zhang, Adv. Mat. Res., 2013, 781, 2184–2188.
14. Y. H. Huang, Y. F. Huang, P. S. Chang and C. Y. Chen, J. Hazard. Mater., 2008, 154, 655–662.
15. Y. Jing and B. P. Chaplin, Environ. Sci. Technol., 2017, 51, 2355–2365.
16. C. Lee, J. Yoon and U. V. Gunten, Water Res., 2007, 41, 581–590.
17. N. A. Landsman, K. L. Swancutt, C. N. Bradford, C. R. Cox, J. J. Kiddle and S. P. Mezyk, Environ. Sci. Technol., 2007, 41, 5818–5823.
18. S. A. Wang, Dyes Pigm., 2008, 76, 714–720.
19. F. Velichkova, H. Delmas, C. Julcour and B. Koumanova, AIChE J., 2017, 63, 669–679.
20. J. J. Pignatello, Environ. Sci. Technol., 1992, 26, 944–951.
21. J. Sun, X. Li, J. Feng and X. Tian, Water Res., 2009, 43, 4363–4369.
22. S. T. Yang, W. Zhang, J. Xie, R. Liao, X. Zhang, B. Yu, R. Wu, X. Liu, H. Li and Z. Guo, RSC Adv., 2015, 5, 5458–5463.
23. S. T. Christensen, H. Feng, J. L. Libera, N. Guo, J. T. Miller, P. C. Stair and J. W. Elam, Nano Lett., 2010, 10, 3047–3051.
24. Y. H. Huang, C. C. Su, Y. P. Yang and M. C. Lu, Environ. Prog. Sustain. Energy, 2013, 32, 187–192.
25. J. Choi, J. H. Jeong and J. Chung, Chem. Eng. J., 2013, 218, 260–266.
26. Z. Huang, Z. Chen, Y. Chen and Y. Hu, Chemosphere, 2018, 203, 442–449.
27. Z. Han, Y. Dong and S. Dong, J. Hazard. Mater., 2011, 189, 241–248.
28. I. Yamanaka and Y. Nabae, Adv. Sci. Technol., 2006, 45, 2067–2076.
29. A. Ciftci, M. D. A. J. Ligthart, O. A. Sen, V. A. J. F. Hoof, H. Friedrich and E. J. M. Hensen, J. Catal., 2014, 311, 88–101.
30. Y. Liu, G. Zhang, S. Fang, S. Chong and J. Zhu, J. Environ. Manage., 2016, 182, 367–373.
31. C. Vidya, C. Manjunatha, M. N. Chandraprabha, M. Rajshekar and M. A. L. A. Raj, J. Environ. Chem. Eng., 2017, 5, 3172–3180.
32. Y. S. González, C. Costa, M. C. Márquez and P. Ramos, J. Hazard. Mater., 2011, 187, 101–112.
33. J. D. Zeeuw, D. R. C. M. Nijs and L. T. Henrich, J. Chromatogr. Sci., 1987, 25, 71–83.
34. Y. Xie, Y. Zhu, B. Zhao and Y. Tang, Stud. Surf. Sci. Catal., 1998, 118, 441–449.
35. M. Ciobanu, G. Petcu, E. M. Anghel, F. Papa, N. G. Apostol, D. C. Culita, I. Atkinson, S. Todorova, M. Shopska, A. Naydenov, R. Velinova and V. Parvulescu, Appl. Catal., A, 2021, 619, 118123.
36. M. Zhang, S. Cai, J. Li, E. A. Elimian, J. Chen and H. Jia, J. Hazard. Mater., 2021, 412, 125266.
37. Y. Zhou, D. Liu, W. Qiao, Z. Liu, J. Yang and L. Feng, Materials Today Physics, 2021, 17, 100357.
38. Z. Lan, Y. Yu, H. Li, L. Zhang and Y. Cao, Mater. Sci. Semicond. Process., 2021, 123, 105547.
39. Y. Xi, J. Xiao, X. Lin, W. Yan, C. Wang and C. Liu, Ind. Eng. Chem. Res., 2018, 57, 10137–10147.
40. X. Li, X. You, P. Ying, J. Xiao and C. Li, Top. Catal., 2003, 25, 63–70.
41. C. E. Huckaba and F. G. Keyes, J. Am. Chem. Soc., 1948, 70, 1640–1644.
42. P. Shafiee, S. M. Alavi and M. Rezaei, J. Energy Inst., 2021, 96, 1–10.
43. N. Ji, Z. Liu, X. Diao, J. Bao, Z. Yu, C. Song, Q. Liu, D. Ma and X. Lu, Mol. Catal., 2020, 495, 111155–111163.
44. A. H. Valekar, K. R. Oh, S. K. Lee and Y. K. Hwang, J. Ind. Eng. Chem., 2021, 101, 66–77.
45. H. Lee, W. I. Kim, K. D. Jung and H. L. Koh, Korean J. Chem. Eng., 2017, 34, 1337–1345.
46. X. Zhang, N. He, C. Liu and H. Guo, Catal. Letters, 2019, 149, 974–984.
47. W. Wang, K. Nakagawa, T. Yoshikawa, T. Masuda, E. Fumoto, Y. Koyama, T. Tago and H. Fujitsuka, Appl. Catal. A Gen., 2021, 619, 118152–118160.
48. M. H. Stacey, Stud. Surf. Sci. Catal., 1988, 39, 55–65.
49. A. V. Neimark, Stud. Surf. Sci. Catal., 1991, 62, 67–74.
50. S. S. Lin and M. D. Gurol, Water Sci. Technol., 1996, 34, 57–64.
51. G. T. Chi and K. D. Huddersman, J. Adv. Oxid. Technol., 2011, 14, 235–243.
52. M. Martinez, A. Irabien, A. R. Arnaiz and C. Santiago, J. Therm. Anal. Calorim., 1984, 29, 251–255.
53. G. R. Reddy and V. V. Mahajani, Ind. Eng. Chem. Res., 2005, 44, 7320–7328.
54. M. Song, Y. Wang, Y. Guo, L. Wang, W. Zhan, Y. Guo and G. Lu, Chin. J. Catal., 2017, 38, 1155–1165.
55. M. Ahmadi, B. Kakavandi, S. Jorfi and M. Azizi, J. Photochem. Photobiol., 2017, 336, 42–53.
56. P. Fu, L. Wang, G. Li, Z. Hou and Y. Ma, J. Environ. Chem. Eng., 2020, 8, 103714–103721.
57. Y. Qi, C. Guo, X. Xu, B. Gao, Q. Yue, B. Jiang, Z. Qian, C. Wang and Y. Zhang, Sci. Total Environ., 2020, 715, 136982–136994.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1ra05762f

‡ These authors contributed equally to this paper.

---