The development of new catalysts from abundant raw materials, generating attractive photocatalytic activity, constitutes a real challenge in the context of sustainable development concerns. In this setting, a dolomite was treated at 800 °C (D800) and then chemically modified by Ca(NO3)2 (CaD800) using a simple procedure. The resulting materials were characterized by X-ray diffraction, scanning electron microscopy, energy dispersive X-ray spectroscopy (EDS), solid state UV spectroscopy, and used as catalysts of pentachlorophenol (PCP) degradation in aqueous solutions under UV light irradiation. The treatment of dolomite at 800 °C enabled a full decarbonation of CaMg(CO3)2, with formation of CaO, Ca(OH)2, and MgO. Additional CaO was generated after chemical treatment as revealed by EDS analysis; the Ca/Mg ratio increased from 1.29 (D800) to 1.44 for CaD800. This CaO in aqueous medium hydrates by giving Ca(OH)2. CaD800 was found to be the best photocatalyst with a PCP degradation rate of 95% after only 1 h of treatment, for a CaD800/D800 degradation rate constant ratio of 1.58. In this regard, we investigated the Fourier transform infrared spectra of CaD800, PCP, and CaD800 loaded with PCP after degradation. We thus evidenced the involvement of Ca(OH)2 in the PCP degradation process. Catalytic activity was discussed through the contribution of OH radicals and electrodonation.
Emerging pollutants are compounds that were not originally considered as such, but can be found in the worldwide environment. They generally come from industrial, agricultural, and municipal wastewater sources, and are divided into micropollutants and nanopollutants. These include pesticides, pharmaceuticals, surfactants, plasticizers, heavy metals, and industrial compounds/by-products (Leite et al. 2018; Onsy & Paleologos 2018). One of the most effective means of removing these contaminants is photodegradation in the presence of a catalyst. Although TiO2 has been widely used for decades, recent work is increasingly oriented towards new materials. The latter are not based on TiO2 modified by doping or mixing with other oxides. In this connection, photocatalytic properties of calcium oxide have been investigated either in combination with other oxides (Zhou et al. 2015; Shtare et al. 2016) or as a single phase (Madhusudhana et al. 2012; Veeranna et al. 2014; Ruiz Peralta et al. 2018). In the latter form, synthetic CaO nanoparticles were purchased from chemical companies or synthesized in the laboratory in several time-consuming steps. Also, the results are somewhat poor with band gap energies around 5 eV (Madhusudhana et al. 2012; Veeranna et al. 2014; Ruiz Peralta et al. 2018).
Dolomite represents an important carbonated material class. It is a natural mineral that contains alternating planes of Ca2+ and Mg2+ ions, with an ideal formula of CaMg(CO3)2. Calcined at optimum temperature, it releases carbon dioxide and provides a CaO.MgO type compound. Pozan & Kambur (2013) found that the impregnation of TiO2 by MgO decreases the band gap of TiO2 and improves the photocatalytic degradation efficiency of chlorophenol under UV light, so a CaO.MgO combination could have interesting properties.
Chlorophenols (especially pentachlorophenol (PCP)) are a soil and water contaminant and are highly harmful to humans and aquatic organisms. The use of PCP is currently limited or banned in many countries; however, many parts of the world are still contaminated (León-Santiesteban et al. 2016). PCP has been employed as fungicide, insecticide, herbicide, and most importantly as a paint antifouling agent. Among its main harmful effects is endocrine disruption (Nollet 2012). PCP was degraded by Pd/Fe nanoparticles (Shih et al. 2016), alkali metal halides (Khuzwayo & Chirwa 2017), activated magnesium (Garbou et al. 2017), and graphene oxide (Marinescu et al. 2018) catalysts.
The originality of this work is that, to date, no thermally and chemically treated dolomite has been used as a catalyst for the photodegradation of any contaminant. So this paper examines the possibility of using an Algerian dolomite in the photocatalytic degradation of PCP from aqueous solutions and under UV light irradiation. Dolomite was processed at 800 °C (D800) and then chemically modified by Ca(NO3)2 (CaD800) using a simple procedure. The resulting materials were characterized by X-ray diffraction (XRD), scanning electron microscopy (SEM), energy dispersive X-ray spectroscopy (EDS), and solid state UV spectroscopy. The degradation process was monitored as a function of processing time. The parameters investigated were the evolution of the UV spectra of the PCP solution under UV irradiation, chemical oxygen demand (COD), the titration of the chlorides released in solution, and the concentration of PCP remaining in solution. The Fourier transform infrared (FTIR) spectra of the best photocatalyst before and after photodegradation were interpreted and finally the catalytic activity discussed.
MATERIALS AND METHODS
The catalyst used in this study is a dolomite coming from a deposit situated at Djebel Teiouelet, Ain M'lila (eastern Algeria). Its chemical composition consists primarily of MgO and CaO. In a first approach, the dolomite was treated at 600, 800, 900, and 1,000 °C to determine the optimum temperature. Calcination showed that the dolomite sample treated at 800 °C (designated D800) possessed the largest specific surface. In a second step, D800 was chemically modified by Ca(NO3)2 by mixing 2 mmol of Ca(NO3)2 with 1 g of D800 at 60 °C for 8 h. The suspension was subsequently filtered and the recovered solid dried at 120 °C. The final product was calcined at 800 °C for 2 h and named CaD800. We followed the experimental protocol of Ngamcharussrivichai et al. (2007), changing the contact time: 8 h instead of 4 h.
Chemical composition was determined by a Cameca SX-50 electronic microprobe. XRD diffractograms were obtained on a Rigaku D/Max 2200 powder X-ray diffractometer using Cu-Kα radiation and exploring the 2θ range 10–90°. Morphology and surface composition were evidenced by a Jeol JSM-6060 LV scanning electron microscope coupled to an EDS device. The diffuse reflectance spectra were registered on a Bruker UV-Visible Solid spectrophotometer equipped with an integrating sphere in BaSO4, taking into account the amount of diffusely reflected photons on the surface of a sample. The sample is irradiated by a UV lamp with a wavelength of 254 nm during various exposure times until absorption saturation. Infrared spectra were recorded through a Shimadzu Prestige 21 spectrophotometer employing KBr pellets composed of 0.5% of the sample.
A 100 mg L−1 solution of PCP was chosen to test the performance of the catalyst in the degradation of high-concentration pollutants. Although this concentration cannot be found in wastewater, authors have examined concentrations of 50 (Matta & Chiron 2018), 100 (Karn et al. 2010) and even 2,000 mg L−1 (Quan et al. 2007). We added NaOH to dissolve 100 mg L−1 of PCP, so the pH used for the subsequent experiments was 11.5. For each experiment, 50 mL of solution containing a PCP (Aldrich) concentration of 100 mg L−1 in 0.001 mol L−1 NaOH was mixed with an optimized catalyst concentration of 1.5 g L−1. Before testing the photocatalytic activity of D800 and CaD800, the reaction mixture was maintained in the dark under continuous shaking for 1 h to reach the adsorption–desorption equilibrium and eliminate the contribution of adsorption to degradation. Subsequently, the suspension was irradiated using a lamp with a 60 W power and absorbing at λmax of 365 nm. The solution–UV lamp distance was 8 cm. The degradation process was followed as a function of the processing time. All experiments were conducted in triplicate with errors below 8% and the average values reported. The parameters monitored were the evolution of the UV spectra of the PCP solution under irradiation, the titration of chlorides released in solution using the Mohr's method (Rodier 2009), the evaluation of COD according to AFNOR NFT 90-101 (AFNOR 1994), and the determination of PCP concentration remaining in solution. The latter was measured at 318 nm using a Shimadzu 1240 UV-Vis spectrophotometer. The photodegradation device is given in Figure 1.
RESULTS AND DISCUSSION
The powder XRD patterns of raw dolomite (D) and products calcined in air at 800 °C (D800) and modified with Ca(NO3)2 (CaD800) are shown in Figure 2. The room temperature XRD pattern (D) displays sharp peaks that can be attributed to dolomite (JCPDS Files card 11-78; 1999), which correspond to the reflections (from left to right) (012), (104), (006), (113), (021), (018) and (116), respectively. The pattern of the untreated dolomite also highlights two peaks corresponding to lime (CaO). The diffractogram of D800 highlights peaks for MgO, CaO, and Ca(OH)2. The latter is due to the ability of CaO to easily hydrate. Treatment of dolomite at 800 °C thus enabled full decarbonation of CaMg(CO3)2, a result previously demonstrated by thermogravimetric/differential thermal analysis (Khalfa et al. 2018). After reaction with Ca(NO3)2, CaD800 shows a new peak due to CaCO3. It concerns the reflection (104) of the calcite phase whose relative intensity is 100%. Other calcite reflections of lesser intensity were not represented on the CaD800 diffractogram. The emergence of the calcite peak could be explained by the excess CaO, due to the Ca(NO3)2–CaMg(CO3)2 combination, which would react with CO2 after exposure to air even after heat processing (Ji et al. 2009).
SEM images of D, D800 and CaD800 are displayed in Figure 3. The SEM image of the raw dolomite shows a characteristic sharpened morphology. In this regard, another image of the same material indicated cleavages and an apparent preferential orientation of dolomite crystals along the c-axis (Ziane et al. 2018). The appearance of clearer areas (b) is due to CaO, in accordance with XRD data. The micrograph of D800 shows a less compact and more airy structure, a consequence of full decarbonation at 800 °C. As a result, new pores and slots have arisen. After chemical modification by Ca(NO3)2, the surface topography of CaD800 changes slightly with the presence also of brighter areas (a) synonymous with dense matter, probably CaO introduced in excess in comparison with MgO. Atoms with high atomic number possess more electrons around their nucleus, and thus more incident electrons will be scattered. This leads to brighter areas for such species. The reduction in resolution expressed by a lack of sharpness of the CaD800 pattern is linked to a loss of crystallinity due to the introduction of a new compound (Ca(NO3)2) into a previously formed structure (Joy 1975).
An EDS analysis (Figure 3) gave Ca/Mg ratios of 1.64, 1.29 and 1.44 for D, D800 and CaD800, respectively. The ratio found for untreated dolomite is within the 1.3–1.8 range reported in the literature (Quitete & Souza 2017). A molar ratio of 1.64 is larger than the stoichiometric value for dolomite. This phenomenon is frequently encountered in raw dolomite because it occurs, in our case, in conjunction with CaO as revealed from XRD and SEM analyses. The heat treatment at 800 °C reduces the Ca/Mg ratio to 1.29. After calcination at 800 °C, a Mg-rich surface was produced, a consequence of a faster diffusion rate of Mg towards the surface owing to its smaller ionic radius (0.66 Å) in comparison with Ca (0.99 Å). Since the chemical composition regarding Ca and Mg does not change significantly after calcination, it is assumed that the bulk phase should be Ca-rich (Sasaki et al. 2013). As some superficial atomic layers can be detected by EDS, the Ca/Mg ratio of a calcined dolomite could only be low. In this regard, the ratio of 1.44 of CaD800 (compared to 1.29 for D800) would be explained by the excess Ca chemically combined with D800.
Solid UV analysis
Figure 4 illustrates the UV-Vis absorption pattern of D800 and CaD800. There is an obvious and strong absorption in the UV light range for both materials, which extends into the short wavelength range of visible light, for CaD800. The spectrum is similar in appearance to that obtained by Zhang (2014) for a nanosized calcium hydroxide (Ca(OH)2) synthesized by precipitation.
By drawing versus , we evaluated the energy band gap of each photocatalyst from the intersection of the tangent, at the inflection point of the curve, with the x-axis. We found 3.7 and 3.28 eV for D800 and CaD800, respectively.
It has been shown that the dolomite band gap is 5 eV (Hossain et al. 2011). Heat treatment at 800 °C of the Teiouelet dolomite led to complete decarbonation with CaO-MgO formation, resulting in a 3.7 eV band gap (D800). This decrease from 5 to 3.7 eV is explained by the presence of CaO, which has a beneficial effect on this optical property (Lakshminarayana et al. 2017). As the chemical treatment of D800 with Ca(NO3)2 results in an excess of CaO, the CaD800 band gap further decreases to reach 3.28 eV. Pozan & Kambur (2013) also achieved a band gap reduction from 3.2 to 2.8 eV for TiO2 doped with CaO and MgO. In parallel, the presence of calcite identified by XRD could also be responsible for this decrease (Stone 1990).
Photodegradation of PCP
The PCP spectrum before UV treatment (Figure S1; 0 min) displays three bands at 220, 250, and 318 nm (Figure S1 is available with the online version of this paper). The first two are due to π → π* aromatic transitions while the last one belongs to n → π* band. The exposure to UV light leads to the abatement of these bands until their disappearance after 4 hours, particularly for bands at 250 and 318 nm.
The extent of degradation was also quantified by determining the PCP concentration remaining in solution as a function of treatment time (Figure 5). The concentration of PCP in solution decreases gradually with increasing time. After 4 h, the PCP amount non-degraded was 15.68 mg L−1, corresponding to an elimination rate of 84.3%. The UV treatment was performed at an initial pH of 11.5, which is favorable for PCP degradation. It has been shown that the photolysis of PCP in homogeneous systems under UV light increases with increasing solution pH (Lan et al. 2011). At basic pH, additional OH radicals can be generated, which accelerate the photolysis of PCP molecules. In addition, we monitored the chloride ions released in solution. The degradation of PCP is accompanied by a continuous release of chloride (Antonopoulou et al. 2015). It reaches a maximum value of 64.4 mg L−1 after 4 h, representing a rate of 97%. Xue et al. (2009) showed that the first step of PCP degradation is dechlorination.
To reduce the processing time, we used D800 and CaD800 as catalysts. The evolution of PCP concentration remaining in solution after photocatalytic degradation is shown in Figure S2 (available online). The decrease in concentration is fast, resulting in a degradation of 83 and 95.0% after an exposure time of 1 h, in the presence of D800 and CaD800, respectively. Photodegradation of PCP required 4 h in the presence of the α-Fe2O3/ZnO catalyst (Xie et al. 2015).
The reaction kinetics were studied by plotting ln([PCP]0/[PCP]) against time (Figure S3, available online). Linear relationships were obtained with determination coefficients greater than 0.980, indicating that PCP removal follows pseudo-first order kinetics. Several authors have reached such kinetics for different catalysts (Zhang et al. 2014; Antonopoulou et al. 2015). By calculating the rate constants, k, we found a k value of 0.007 min−1 in absence of catalyst (data not shown) and 0.031 and 0.049 min−1 in presence of D800 and CaD800, i.e. a ratio of 7.1 and 1.58 in favor of CaD800, respectively.
The degradation of PCP is accompanied by the release of chlorides in solution. Figure 6 highlights this characteristic. The decrease is caused by the adsorption of Cl− ions by D800 and CaD800. Note that this reduction is more significant for CaD800. A previous study showed that chlorides adsorb on portlandite (Ca(OH)2) (Mainguy & Coussy 2000).
The evolution of the degradation rate against the exposure time was also examined (Figure 7). The experimental COD values initially found were 68.08 and 65.2 mg O2 L−1, for D800 and CaD800, respectively. Figure 7 shows a continuous increase in percentage degradation as the experiment progresses. After 1 h of treatment, these values were 83 and 95% for D800 and CaD800, respectively. This proves the efficiency of the alkaline-earth oxides when refractory organic compounds such as PCP are subjected to degradation.
Photocatalytic degradation of PCP showed that CaD800 is the most effective catalyst with an elimination rate of 95%. So we investigated the FTIR spectra of CaD800, PCP, and CaD800 loaded with PCP after degradation (PCP/CaD800). The CaD800 spectrum (Figure 8) shows an intense band at 3,640 cm−1 corresponding to Ca(OH)2 present in excess (Bodénan & Deniard 2003), in correlation with XRD and EDS analyses. The 3,422 cm−1 broad band belongs to the superimposition of the asymmetric and symmetric stretches of hydroxyl pertaining to water molecules (Mehdi et al. 2019). The water bending vibration takes place at 1,592 cm−1, while the peak at 1,379 cm−1 results from the recovered CO2. The 1,124 cm−1 peak is indicative of dypingite (4MgCO3·Mg(OH)2·5H2O) (Surface et al. 2013; Chukanov & Chervonnyi 2016), a hydrated species composed of Mg(OH)2 and CO2 recovered by MgO. Both Mg(OH)2 and MgO were identified by XRD. The band at 870 cm−1 is the consequence of the in-plane bending vibration of calcite carbonates.
The PCP spectrum shows a wide band centered at 3,419 cm−1 corresponding to the hydrogen-bonded O−H stretching. The aromatic aspect occurs at 1,544 and 1,413 cm−1, corresponding to C=C stretches. The 1,379 and 1,206 cm−1 peaks results from in-plane OH deformation and C−OH stretch, respectively. The intermediate peak (1,304 cm−1) is due to a coupling of C−OH stretch and OH deformation. The sharp peak at 764 cm−1 derives from C−Cl elongation. Out-of-plane ring C=C bend and hydrogen-bonded out-of-plane O−H bend appear at 670 and 645 cm−1, respectively. The allocation of PCP bands was based on different sources (Silverstein et al. 1998; Mistry 2009).
The spectrum of PCP/CaD800 shows significant changes. The intense broad band at 3,447 cm−1 could be attributed to a carboxylic acid salt (Sadtler 2005), arising from the combination of Na+ and the carboxylate ion generated by the degradation of PCP. The occurrence of Na+ is due to the fact that PCP has been dissolved in a NaOH solution. The appearance of weak bands at 2,918 and 2,843 cm−1 is induced by asymmetric and symmetric stretches of methylene groups. The 1,636 and 1,442 cm−1 bands are attributed to asymmetric and symmetric stretching of carboxylate ion (O=C–O) (Mistry 2009), respectively. The presence of such bands indicates cycle opening and PCP degradation. In this vein, the disappearance of the bands at 1,379 (in-plane OH deformation), 1,206 (C−OH stretch), 670 (out-of-plane ring C=C bend), and 645 cm−1 (out-of-plane O−H bend) reinforces the hypothesis of the cycle opening. The presence of Ca(OH)2 and Cl− in the same medium leads to the formation of CaOHCl species (Mainguy & Coussy 2000), as follows: Ca(OH)2 + Cl− → CaOHCl + OH−. In parallel, the association of electron-attracting elements such as Cl− with CaOH+ species decreases the band intensity and increases the frequency (Colthup et al. 1990). Knowing that Ca(OH)2 and chlorides appear at 3,640 and 764 cm−1, respectively, the 3,687 and 865 cm−1 bands could be assigned to CaOHCl. This would explain why the concentration of chlorides released in solution during treatment decreased (Figure 6) and the solution pH increased to 12.
Catalytic activity is related to the band gap value of a catalyst. For CaD800, it is 3.28 eV, which is less than the irradiation energy of the lamp used (3.40 eV). In this context, its catalytic activity is due to the presence of CaO in excess, as revealed from EDS analysis, which in aqueous medium hydrates by yielding Ca(OH)2. Several authors reported that Ca(OH)2 has high catalytic activity and was used efficiently in the photodegradation of organic dyes and as an alternative in advanced oxidation processes (Zhang 2014; Sánchez-Cantú et al. 2017). For example, Ruiz Peralta et al. (2018) reported that CaO was not the real active phase in the Rhodamine 6G photodegradation but Ca(OH)2.
In alkaline media, which is the case in this study, a greater hydroxide ion concentration susceptible to reacting with photocreated holes is present near the catalyst surface, which should lead to more effective generation of hydroxyl radicals. As these are strong oxidants, photodegradation of PCP in the presence of CaD800 can be enhanced in alkaline solution.
Knowing that this superoxide is an excellent oxidant, it also contributes to the degradation of PCP. A literature survey showed that 53% of PCP degradation is induced by electron transfer and 47% by hydroxyl radical addition (Fang et al. 2000).
A new photocatalyst was developed from dolomite treated at 800 °C and chemically modified with Ca(NO3)2. CaD800 is thus composed of Ca(OH)2, CaO, and MgO with a Ca/Mg ratio of 1.44 (against 1.29 for D800) and a band gap energy of 3.28 eV. It led to a PCP degradation rate of 95% after only 1 h of treatment, for a CaD800/D800 degradation rate ratio of 1.61. The infrared study showed the involvement of Ca(OH)2 with formation of CaOHCl species, generation of carboxylate ions and, thus, cycle opening. A surplus Ca(OH)2 resulted in an excess of OH− anions and therefore hydroxyl radicals, under UV light. The catalytic activity was explained by the simultaneous action of hydroxyl (OH•) and superoxide (O2−•) radicals. Our material prepared from an abundant and cheap raw material has proven to be a good catalyst and is expected to improve the degradation reactions of other emerging pollutants.