*Article* **Photocatalytic Study of Cyanide Oxidation Using Titanium Dioxide (TiO2)-Activated Carbon Composites in a Continuous Flow Photo-Reactor**

**Stalin Coronel, Diana Endara \*, Ana Belén Lozada, Lucía E. Manangón-Perugachi and Ernesto de la Torre**

Department of Extractive Metallurgy, Escuela Politécnica Nacional, Ladrón de Guevara E11-253, Quito 170517, Ecuador; stalin.coronel@epn.edu.ec (S.C.); ana.lozada@epn.edu.ec (A.B.L.); lucia.manangon@epn.edu.ec (L.E.M.-P.); ernesto.delatorre@epn.edu.ec (E.d.l.T.) **\*** Correspondence: diana.endara@epn.edu.ec; Tel.: +593-(9)9854-9231

**Abstract:** The photocatalytic oxidation of cyanide by titanium dioxide (TiO<sup>2</sup> ) supported on activated carbon (AC) was evaluated in a continuous flow UV photo-reactor. The continuous photo-reactor was made of glass and covered with a wood box to isolate the fluid of external conditions. The TiO<sup>2</sup> -AC synthesized by the impregnation of TiO<sup>2</sup> on granular AC composites was characterized by inductively coupled plasma optical emission spectrometry (ICP-OES), Scanning Electron Microscopy (SEM), and nitrogen adsorption-desorption isotherms. Photocatalytic and adsorption tests were conducted separately and simultaneously. The results showed that 97% of CN− was degraded within 24 h due to combined photocatalytic oxidation and adsorption. To estimate the contribution of only adsorption, two-stage tests were performed. First, 74% cyanide ion degradation was reached in 24 h under dark conditions. This result was attributed to CN− adsorption and oxidation due to the generation of H2O<sup>2</sup> on the surface of AC. Then, 99% degradation of cyanide ion was obtained through photocatalysis during 24 h. These results showed that photocatalysis and the continuous photo-reactor's design enhanced the photocatalytic cyanide oxidation performance compared to an agitated batch system. Therefore, the use of TiO<sup>2</sup> -AC composites in a continuous flow photo-reactor is a promising process for the photocatalytic degradation of cyanide in aqueous solutions.

**Keywords:** cyanide; activated carbon; titanium dioxide; composites; continuous flow; photocatalysis; adsorption

#### **1. Introduction**

Cyanide is a highly toxic pollutant, which, even at low concentration, may cause human health and environmental problems [1]. Cyanide is rapidly and extensively absorbed by the human body through the oral inhalation and dermal routes. It prevents the transport of oxygen, which affects the cellular respiration process, leading to suffocation in the worst case and eventually to death [2].

Cyanide is present in industrial wastewaters such as coal gasification, electroplating, plastics, pharmaceuticals, and the mining industry. These wastewaters are discharged in the water bodies causing serious threats to the environment [1,3,4].

In Ecuador, the artisanal and small-scale gold mining activities make significant contributions to mineral production [5,6]. Large-scale gold mining projects as "Fruta del Norte" and "Cascabel" have been developed in the last years. Both small- and large-scale gold mining industries use cyanidation to recover gold from ores.

In Ecuador, cyanidation and carbon in pulp (CIP) processes use an aqueous 500 mg/L NaCN solution. The Ecuadorian legislation TULSMA establishes a discharge limit of 1 mg/L total cyanide into sewers and 0.1 mg/L into surface fresh waters [7]. Cyanides exist in the form of free ions (CN−) and also can form complexes, which makes its treatment more difficult. For example, a study with gold mining wastewater that contained copper

**Citation:** Coronel, S.; Endara, D.; Lozada, A.B.; Manangón-Perugachi, L.E.; de la Torre, E. Photocatalytic Study of Cyanide Oxidation Using Titanium Dioxide (TiO2)-Activated Carbon Composites in a Continuous Flow Photo-Reactor. *Catalysts* **2021**, *11*, 924. https://doi.org/10.3390/ catal11080924

Academic Editors: José Ignacio Lombraña, Héctor Valdés, Cristian Ferreiro and Giuseppina Iervolino

Received: 16 May 2021 Accepted: 27 July 2021 Published: 30 July 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

and zinc was treated with zeolite, achieving a 93.97% degradation efficiency of total cyanides [8]. Due to the hazard of cyanide, its treatment is very important. Several physical, biological, and chemical treatment processes have been applied to remove cyanide. The most commonly used processes include chemical oxidation, alkaline chlorination, hydrogen peroxide oxidation, INCO process (purification with SO<sup>2</sup> and air), oxidation with Caro acid, and ozonation. Chlorination is the most used process, and it is very efficient in the elimination of cyanide; however, it has certain disadvantages, such as the generation of toxic intermediate compounds which must be treated, making the process expensive [9,10].

Advanced oxidation processes (AOPs) have been studied extensively in the removal of different contaminants. Hydroxyl radicals are considered the most reactive oxygenated species within AOPs due to their high oxidation potential and their non-selective nature. Photocatalysis is an effective technique for treating toxic substances, including cyanide. [11,12].

Several semiconductor materials have been tested as photocatalysts for the removal of aqueous pollutants. However, difficulties related to the stability of the photocatalyst under irradiation in water have been evidenced. It is accepted that titanium dioxide TiO<sup>2</sup> in anatase phase is the most reliable photocatalyst for aqueous pollutant removal [13].

In heterogeneous photocatalysis based on semiconductors, the photocatalyst TiO<sup>2</sup> is excited by absorbing incident UV radiation. Momentarily, the electron of the valence band is transferred to the conduction band, and the electron/hole pair (e−/h<sup>+</sup> ) is formed. Then, the electron/hole pair reacts with water and dissolved molecular oxygen to generate hydroxyl and superoxide radicals, which are responsible for the photocatalytic oxidation of free cyanide [14].

Direct and indirect mechanisms of photocatalytic oxidation of free cyanide with TiO<sup>2</sup> have been proposed. In the direct mechanism, free cyanide present in the solution is oxidized directly through the transfer of electrons from the holes of the valence band. The indirect mechanism occurs through the adsorbed OH• radicals, implying that the contaminants are first adsorbed on the photocatalyst surface and then react with the excited superficial e−/h<sup>+</sup> pairs or the OH• radicals [15].

The photocatalytic oxidation of cyanide starts with the formation of cyanide radical, which dimerizes to form cyanogen. Then, the cyanogen undergoes transformation in an alkaline medium to give cyanide and cyanate. Finally, cyanate oxidizes to form nitrite (NO<sup>2</sup> <sup>−</sup>), nitrate (NO<sup>3</sup> <sup>−</sup>), carbonate (CO<sup>3</sup> <sup>2</sup>−), carbon dioxide (CO2), and nitrogen (N2). These reactions are shown in the Equations (1)–(5) [16].

$$\text{CN}^{-} \stackrel{\text{h}^{+} / \text{OH}^{\bullet}}{\stackrel{\bullet}{\rightarrow}} \text{CN}^{\bullet} \tag{1}$$

$$\text{2CN}^{\bullet} \to (\text{CN})\_2 \tag{2}$$

$$\text{(CN)}\_{2} + 2\text{OH}^{-} \rightarrow \text{CN}^{-} + \text{CNO}^{-} + \text{H}\_{2}\text{O} \tag{3}$$

$$\rm{CNO^{-}} + 8\rm{OH^{-}} + 8\rm{h^{+}} \rightarrow \rm{NO\_{3}^{-}} + \rm{CO\_{2}} + 4\rm{H\_{2}O} \tag{4}$$

$$\mathrm{CNO^{-}} + 2\mathrm{H\_2O} + 3\mathrm{h^{+}} \rightarrow \mathrm{CO\_3}^{2-} + \frac{1}{2}\mathrm{N\_2} + 4\mathrm{H^{+}} \tag{5}$$

TiO<sup>2</sup> has been widely used as photocatalyst due to its non-toxicity, low cost, chemical stability, and its favorable chemical and physical properties. It can be reused several times without reducing its catalytic activity. TiO<sup>2</sup> has been tested in the photocatalytic degradation of wastewater under ultraviolet (UV) light. However, low efficiencies in photodegradation have been achieved due to its rapid unfavorable charge carrier recombination reaction in TiO<sup>2</sup> and the high band gap energy of 3.2 eV, which limits its application from using solar energy [4,15]. In addition, filtration and separation processes are required at the end of the treatment, which increases the costs due to the granulometric size of TiO<sup>2</sup> (74 microns) [17].

Some strategies have been proposed in order to overcome these drawbacks. These include the use of supports such as silica [18], zeolites [19], and activated carbon. For example, the use of activated carbon (AC) as support achieves minimal losses of TiO2. Studies show that AC improves the efficiency of the photocatalytic process thanks to its high adsorption capacity given by its porous structure. AC is a good support, due to its granulometry, hardness, and high surface area [17].

AC develops a synergistic adsorption-degradation effect according to its surface chemistry [17]. AC can oxidize cyanide through the adsorption of molecular oxygen on the AC surface. Adsorbed oxygen reacts with functional groups characteristic of the AC surface to form hydrogen peroxide (H2O2) that finally reacts with cyanide ion (CN−) to form cyanate, according to Equation (6) [20]. The cyanide oxidation process with AC reached oxidation percentages between 60 and 70% after 8 h [21]. Although AC could adsorb cyanide, preliminary tests have shown that the adsorption percentage of cyanide is less than 5% [22].

$$\text{CN}^-\text{(aq)} + \text{H}\_2\text{O}\_{2(aq)} \rightleftharpoons \text{CNO}^-\text{(aq)} + \text{H}\_2\text{O}\_{(aq)}\tag{6}$$

The optimization of the photocatalytic material and the study of the influence of factors such as pH, hydroxyl radicals concentration, or the organic compounds concentration on the cyanide photodegradation have been analyzed to improve the photocatalytic activity [10,14]. Nevertheless, the design of the photocatalytic reactor has been less studied, and there is more information available about the optimization of photocatalytic catalyst. For this reason, new alternatives referring to the photo-reactors configuration are necessary to improve the photocatalytic processes [23].

Batch photo-reactors require long times to achieve significant cyanide removal percentages (>90%); for this reason, the design of photocatalytic reactors in a different configuration than batch is essential. In a batch reactor, the UV light has contact mainly at the surface level, without considerable entry and dissipation within the fluid. The thickness of liquid formed from the base of the reactors decreases the activation of the photocatalyst and consequently the degradation of cyanide [17,24,25].

Another disadvantage of a batch reactor is the difficult separation of catalyst after the degradation process. This could be tackled by the implementation of a continuous reactor in the treatment of pollutants [26]. In a previous study, a multi-phase continuous flow reactor was tested in the photocatalytic oxidation of cyanide using TiO<sup>2</sup> as a photocatalyst. Then, the reactor was scaled up to degrade cyanide on an industrial level [27].

This investigation is oriented to the use of composites of titanium dioxide impregnated on active carbon (TiO2-AC) as photocatalyst for the degradation of cyanide in a continuous flow photo-reactor. The use of the TiO2-AC composite and the design of the continuous photo-reactor could enhance the photocatalytic oxidation performance. In addition, this strategic photo-reactor configuration can replace the conventional agitated batch system and reaches high cyanide degradation percentages.

#### **2. Results**

*2.1. Physical and Chemical Characterization of GCR-20 Activated Carbon and TiO2-AC Composite*

The granular composite used as photocatalyst was obtained by the wet impregnation of TiO<sup>2</sup> over AC. A porous network was observed in SEM micrographs of AC and TiO2-AC composite (Figure S1, Supplementary Materials). The content of TiO<sup>2</sup> on the AC was analyzed by ICP-OES, and the impregnation was 0.27% *w*/*w*.

The textural properties determined by N<sup>2</sup> physisorption and BET (Brunauer-Emmett\_Teller) modeling listed in Table 1 show that AC support and TiO2-AC catalyst had more than 900 m2/g of specific surface area. It indicates that there exists available AC porosity for the adsorption of cyanide ion. ASTM (American Society for Testing and Materials) analysis results for d<sup>80</sup> particle size, humidity, volatile, ashes, and fixed carbon listed in Table 2 show that the support material is resistant to thermal and mechanical environments, which are conditions that make the AC a good support and synergic effect material to oxidize cyanide, since it enables work with a clear fluid process. Meanwhile, the granular catalyst (3.10 mm) was immobilized avoiding following recovery operations [28].


**Table 1.** Textural properties determined by N<sup>2</sup> physisorption and BET modeling of AC and TiO<sup>2</sup> - AC catalyst. **(m2/g) (cm3/g) (Å)** AC 1336 0.618 58.92

**Pore Volume**

**Φ**

more than 900 m2/g of specific surface area. It indicates that there exists available AC po‐ rosity for the adsorption of cyanide ion. ASTM (American Society for Testing and Materi‐ als) analysis results for d80 particle size, humidity, volatile, ashes, and fixed carbon listed in Table 2 show that the support material is resistant to thermal and mechanical environ‐ ments, which are conditions that make the AC a good support and synergic effect material to oxidize cyanide, since it enables work with a clear fluid process. Meanwhile, the gran‐ ular catalyst (3.10 mm) was immobilized avoiding following recovery operations [28].

**Table 1.** Textural properties determined by N2 physisorption and BET modeling of AC and TiO2‐

**Table 2.** Physical and chemical properties of AC. Particle size d80 (mm) 3.10

*Catalysts* **2021**, *11*, x FOR PEER REVIEW 4 of 13


#### *2.2. Photo-Reactor Construction* The investigation was performed in a continuous flow glass reactor with irradiation

**Material. SBET** 

The investigation was performed in a continuous flow glass reactor with irradiation of UV lamps. TiO2-AC catalysts were added into the reactor using nylon material nets to support the composite in several beds. We selected a reactor design with the maximum proximity between UV lamps and a stable and consistent flow into the reactor. The configuration of the used material is summarized in Scheme 1. of UV lamps. TiO2‐AC catalysts were added into the reactor using nylon material nets to support the composite in several beds. We selected a reactor design with the maximum proximity between UV lamps and a stable and consistent flow into the reactor. The con‐ figuration of the used material is summarized in Scheme 1.

**Scheme 1.** Distribution of immobilized beds of AC or TiO2‐AC during the adsorption process and photocatalytic degradation of the cyanide ion in the continuous flow photo‐reactor. **Scheme 1.** Distribution of immobilized beds of AC or TiO<sup>2</sup> -AC during the adsorption process and photocatalytic degradation of the cyanide ion in the continuous flow photo-reactor.

Then, the main configuration conditions were a continuous flow of 6.60 mL/s of cyanide solution generated by a 120 rpm peristaltic geopump, 8 mm height layer liquid, nine immobilized beds of AC or TiO2-AC, and a total recirculation volume of 5 L.

Since the adsorption and photocatalytic degradation of cyanide can occur simultaneously, two stage-tests were performed in order to study adsorption and photocatalytic degradation individually. In addition, a simultaneous process was carried out. We found that the cyanide adsorption study had a better fit to the linearized mathe‐ matical model of the Langmuir isotherm for both AC and TiO2‐AC (results summarized in Table 3). The results indicated that the TiO2‐AC maximum cyanide adsorption capacity

Then, the main configuration conditions were a continuous flow of 6.60 mL/s of cya‐

Since the adsorption and photocatalytic degradation of cyanide can occur simultane‐

Through tests of adsorption of CN<sup>−</sup> under dark conditions (no UV irradiation) on the

AC and TiO2‐AC composites, a required adsorption equilibrium time of 24 h was deter‐

nide solution generated by a 120 rpm peristaltic geopump, 8 mm height layer liquid, nine

ously, two stage‐tests were performed in order to study adsorption and photocatalytic

immobilized beds of AC or TiO2‐AC, and a total recirculation volume of 5 L.

degradation individually. In addition, a simultaneous process was carried out.

*Catalysts* **2021**, *11*, x FOR PEER REVIEW 5 of 13

#### *2.3. Cyanide Adsorption Tests* was lower (1/3) than AC. This finding agreed with the textural properties, since the spe‐

*2.3. Cyanide Adsorption Tests*

mined.

Through tests of adsorption of CN− under dark conditions (no UV irradiation) on the AC and TiO2-AC composites, a required adsorption equilibrium time of 24 h was determined. cific surface area of AC decreased once TiO2 was impregnated. Nonetheless, the adsorp‐

We found that the cyanide adsorption study had a better fit to the linearized mathematical model of the Langmuir isotherm for both AC and TiO2-AC (results summarized in Table 3). The results indicated that the TiO2-AC maximum cyanide adsorption capacity was lower (1/3) than AC. This finding agreed with the textural properties, since the specific surface area of AC decreased once TiO<sup>2</sup> was impregnated. Nonetheless, the adsorption and the energy associated to the process had similar behavior, as shown in Figure 1. tion and the energy associated to the process had similar behavior, as shown in Figure 1. **Table 3.** Parameters calculated for Langmuir isotherm model. **Parameter AC TiO2‐AC** 

**Table 3.** Parameters calculated for Langmuir isotherm model. qmax (mg∙g−1) 155.17 52.33


**Figure 1.** Langmuir adsorption isotherms of cyanide (T = 20 ◦C; 1 g TiO<sup>2</sup> -AC/L).

**Figure 1.** Langmuir adsorption isotherms of cyanide (T = 20 °C; 1 g TiO2‐AC/L). Individually, the adsorption of cyanide ion tests were performed for three concentra‐ tions 30 g/L, 45 g/L, and 60 g/L of TiO2‐AC in 500 mg/L synthetic NaCN solutions, under dark conditions in the continuous flow reactor during 24 h at a pH of 10.5 and 20 °C. Another assay was carried out with 60 g/L of AC at the same conditions. The results listed Individually, the adsorption of cyanide ion tests were performed for three concentrations 30 g/L, 45 g/L, and 60 g/L of TiO2-AC in 500 mg/L synthetic NaCN solutions, under dark conditions in the continuous flow reactor during 24 h at a pH of 10.5 and 20 ◦C. Another assay was carried out with 60 g/L of AC at the same conditions. The results listed in Table 4 demonstrate that AC (60 g/L) reached 78% of cyanide degradation due to the adsorption process. It is assumed that adsorption and oxidation processes simultaneously occurred due to oxidant species formed during the process with the dissolved oxygen and functional groups of the AC surface.

in Table 4 demonstrate that AC (60 g/L) reached 78% of cyanide degradation due to the adsorption process. It is assumed that adsorption and oxidation processes simultaneously occurred due to oxidant species formed during the process with the dissolved oxygen and

functional groups of the AC surface.

ysis.


**Table 4.** Kinetic modeling of cyanide ion adsorption (pseudo-second order). **Parameter AC TiO2‐AC**

**Table 4.** Kinetic modeling of cyanide ion adsorption (pseudo‐second order).

*Catalysts* **2021**, *11*, x FOR PEER REVIEW 6 of 13

On the other hand, TiO2-AC catalyst in 60 g/L concentration reached 74% of CN<sup>−</sup> degradation. The cyanide adsorption in the performed tests show similar trends (Figure 2a) and kinetics show that adsorption fits as a pseudo-second-order model through higher correlation coefficients calculation (Table 4). These results indicate that a chemisorption is possible, as suggested by Eskandari [29]. radation. The cyanide adsorption in the performed tests show similar trends (Figure 2a) and kinetics show that adsorption fits as a pseudo‐second‐order model through higher correlation coefficients calculation (Table 4). These results indicate that a chemisorption is possible, as suggested by Eskandari [29].

On the other hand, TiO2‐AC catalyst in 60 g/L concentration reached 74% of CN<sup>−</sup> deg‐

**TiO2‐AC**

**TiO2‐AC**

**Figure 2.** Cyanide ion degradation by (**a**) adsorption, (**b**) photocatalysis, and (**c**) simultaneous adsorption and photocatal‐ **Figure 2.** Cyanide ion degradation by (**a**) adsorption, (**b**) photocatalysis, and (**c**) simultaneous adsorption and photocatalysis.

#### *2.4. Photocatalytic Cyanide Ion Degradation*

*2.4. Photocatalytic Cyanide Ion Degradation* Once the cyanide ion adsorption was performed during 24 h under dark conditions, the photocatalytic tests started by turning on the UV lamps. Tests were carried out at pH 10.5 with three concentrations 30, 45, and 60 g/L (tested in adsorption) during 24 h. More than 90% of cyanide ion was degraded in all assays using TiO2‐AC composites (Table 5). In order to compare the photocatalytic activity performance of materials used for compo‐ sites on CN<sup>−</sup> degradation, tests with TiO2 and AC were performed separately. Although Once the cyanide ion adsorption was performed during 24 h under dark conditions, the photocatalytic tests started by turning on the UV lamps. Tests were carried out at pH 10.5 with three concentrations 30, 45, and 60 g/L (tested in adsorption) during 24 h. More than 90% of cyanide ion was degraded in all assays using TiO2-AC composites (Table 5). In order to compare the photocatalytic activity performance of materials used for composites on CN<sup>−</sup> degradation, tests with TiO<sup>2</sup> and AC were performed separately. Although we focused on the TiO2-AC photocatalytic performance, AC yielded an unexpected 9.73% of cyanide ion degradation. Based on other investigations of AOPs with hydroxyl radicals,

we focused on the TiO2‐AC photocatalytic performance, AC yielded an unexpected 9.73%

this result was explained by photocatalysis promoted by oxygen peroxide. This mechanism considers that H2O<sup>2</sup> can be formed in an unstable and very fast chemical interaction of O2, H<sup>+</sup> , and e− in an aqueous medium [29,30]. Moreover, the kinetics obtained for the system was pseudo-first order for all tests performed. TiO2-AC 60 g/L composite assay aimed at 99.16% of cyanide ion degradation; meanwhile, 82.11% was obtained using TiO<sup>2</sup> 0.45 g/L (equivalent mass of TiO<sup>2</sup> in the TiO2-AC 60 g/L composite).


**Table 5.** Kinetic model results in cyanide ion photodegradation.

For comparison reasons, both processes adsorption and photocatalytic cyanide ion degradation were performed in simultaneous assays with the same TiO2-AC concentrations established in this study. As shown in Figure 2, more than 90% of CN- was removed from the solution. In 24 h, 60 g/L of TiO2-AC removed 97% of CN-under UV lights radiation in the continuous flow reactor. If this value is compared to the process that consisted of 24 h of adsorption (75%) and 24 h of photocatalysis (99%), a synergic effect between materials and phenomena appeared. The kinetics for simultaneous tests were pseudo-first order as well as an individual photocatalytic process, as shown in Table 5. Conversely, the kinetics of the adsorption process was not similar as simultaneous study kinetics.

#### **3. Discussion**

In order to test the adsorption and photocatalysis oxidation of cyanide ion in a continuous flow, a photo-reactor was fabricated based on previous investigations [19,30,31]. Scheme 1 summarizes the design of the photoreactor for CN− degradation, which contributed to enhance the catalyst-fluid contact. Thus, we used a stable continuous flow of 6.60 mL/s of NaCN synthetic solution with an 8 mm liquid layer, inert material, and a maximum proximity between the UV source and material-fluid. These conditions indicated the advantage of AC and TiO<sup>2</sup> performance as a composite, because light can penetrate better in the composite. Since the photoreactor design considered a hollow box with a flow generated by a peristaltic pump through nine separated composite beds, O<sup>2</sup> from air can be introduced during all the processes, contributing to enhance the adsorption and photocatalytic oxidation of CN−. Moreover, UV lamps radiation interacted directly with the fluid and the composite. On the other hand, granular AC (derived of coconut shell by steam activation) showed a high specific surface of 1336 m2/g, which would favor the adsorption of contaminants such as CN−. Indeed, the resistant properties of AC to thermal and mechanical processes and a 3.1 mm grain diameter allow building immobilized beds of the composite inside the photoreactor, avoiding post-recovery difficulties. The diameter of AC was smaller than TiO<sup>2</sup> grain size (170 nm), and SEM-EDX results showed that TiO<sup>2</sup> occupied the external surface of AC, allowing a major contact between UV light source and TiO<sup>2</sup> on the support [32]. The specific surface area of the TiO2-AC composite decreased to 902 m2/g due to the TiO<sup>2</sup> impregnation However, this result represents a high specific surface area available for the adsorption of cyanide ions. Based on the literature,

the semiconductor would occupy the meso and macro porosity of AC; meanwhile, the microporosity is not affected during the wet impregnation of titania [33,34].

Adsorption and photocatalytic oxidation occurred simultaneously during the treatment of cyanide solutions. Thus, we performed two-stage tests to evaluate each process separately. Both processes were studied at 20 ◦C and pH = 10.5. During adsorption batch tests, 24 h were obtained as an equilibrium time (when adsorption does not continue). In addition, the Langmuir isotherm model was the better fit to the adsorption for AC and TiO2-AC, where qmax was reduced from 155 to 52 mg CN−/g TiO2-AC once the impregnation technique was performed. The CN− adsorption in AC was drastically reduced once titania was supported, because the TiO<sup>2</sup> added to the external surface of AC occupies large holes in the support. In the continuous flow reactor tests, we could estimate that the CN− adsorption kinetic follows a pseudo-second-order reaction. This result showed that a chemisorption would take place as well as a physisorption of CN− in AC and TiO2-AC.

Results over 70% of CN− oxidation in AC and composites revealed that in the process, in light absence, not only adsorption takes place but also an oxidation of cyanide ion could be carried out due to the –OH bonds of the AC surface and existing O<sup>2</sup> in aqueous pumping media [29,35].

As expected, granular TiO2-AC composites showed a higher degradation of cyanide ion according to the amount introduced to the continuous flow system. Kinetics and CN− degradation indicated the following order: TiO2-AC 30 g/L < TiO2-AC 45 g/L < TiO2-AC 60 g/L < AC 60 g/L. Since pH 10.5 was performed in all tests, a negative charge of AC surface was expected, giving a repel effect with cyanide ion. Nonetheless, 75% and 78% of CN<sup>−</sup> degradation were determined in light absence using TiO2-AC 60 g/L and AC 60 g/L, respectively.

In photocatalytic tests, as an individual process study, more than 90% of CN− removal was determined using TiO2-AC composites. Thus, external TiO<sup>2</sup> on the AC surface contributed to degrade pollutants in a continuous flow system with composites immobilized with UV 15 W lamps irradiation. For comparison reasons, TiO<sup>2</sup> and AC were tested separately in weight amounts corresponding to 60 g/L TiO2-AC. The results listed in Table 5 showed that titania aimed at 82% of CN− removal; meanwhile, a surprising result of 10% CN− degradation was obtained with AC. Although AC was not considered as a photo-catalyst, it showed CN<sup>−</sup> removal due to the operational conditions and continuous flow system, and H2O<sup>2</sup> could be formed rapidly during UV irradiation once the pair e−/h<sup>+</sup> is formed [32,36]. Some investigations indicate that hydrogen peroxide can contribute to pollutant degradation in AOPs, since this compound can be formed in an electrochemical process associated to UV incidence and dissolved oxygen presence during the pumping process. These results ensure that TiO2-AC composites enhanced the photocatalysis, increasing the synergic effect between AC and TiO2. Kinetics and cyanide degradation were determined in order 60 g/L AC < 0.45 g/L TiO<sup>2</sup> < TiO2-AC 30 g/L < TiO2-AC 45 g/L < TiO2-AC 60 g/L.

Simultaneous adsorption and photocatalysis tests were performed on UV radiation. More than 90% of CN− was removed in all tests. However, in a comparison analysis of each concentration composite dosage, a numerical variation was detected. For 60 g/L of TiO2- AC, 96% of CN− removal was obtained during 24 h, where adsorption and photocatalysis took place at the same time. Whereas 74% and 99% of CN− degradation were obtained in 24 h adsorption and 24 h of photocatalysis, respectively. Thus, a simultaneous process is suitable for cyanide ion degradation under operational continuous flow photorector design. Both photocatalysis and the simultaneous process (photocatalysis + adsorption) correspond to a pseudo-first-order reaction. This result is in concordance with the literature, which attributed the oxidation of pollutants by hydroxyl radical activity. The success of CN<sup>−</sup> degradation was enhanced by TiO2-AC presence and the design parameters of the photoreactor, because a stable system could be formed under a near interaction between the UV source and materials of this study.

#### **4. Materials and Methods**

## *4.1. TiO2-AC Composites Preparation*

TiO2-AC composites were prepared by the wet impregnation of TiO<sup>2</sup> on activated carbon (AC). First, the AC was washed with distilled water under magnetic stirring, until a clarified washing solution was obtained. Then, 1 g of commercial TiO<sup>2</sup> (United States Pharmacopeia (USP) reference standard, 99% anatase) and 100 g of commercial activated carbon (CALGON GRC-20®) were added to 350 mL of distilled water under vigorous stirring for 2 h. The solvent was removed by evaporation, and composites were washed with distilled water several times. Finally, TiO2-AC composites were dried at 110 ◦C for 2 h [37,38].

## *4.2. Physical and Chemical Characterization of GCR-20 Activated Carbon and TiO2-AC Composite*

Standardized sieves were used to determine the particle size distribution of AC by the Standard Test Method for Particle Size Distribution of Granular Activated Carbon ASTM-D2862. Moisture, volatile material, ash, and fixed carbon content were measured for the AC support according to ASTM-D3173 (Standard Test Method for Moisture in the Analysis Sample of Coal and Coke), ASTM-D3175 (Standard Test Method for Volatile Matter in the Analysis Sample of Coal and Coke), and ASTM-D3174 (Standard Test Method for Ash in the Analysis Sample of Coal and Coke from Coal).

The textural properties of the GCR-20 AC and TiO2-AC composites were determined by N<sup>2</sup> physisorption isotherms in a Quantachrome Instruments Nova 4200e (Quantachrome Instruments, Boynton Beach, FL, USA). The BET model was used to determine the specific surface area.

A scanning electron microscopy analysis (SEM) was performed to analyze the TiO2- AC composite texture with a Vega TESCAN (TESCAN, Brno, Czech Republic) microscope equipped with secondary electron (SE) and backscattered electron (BSE) detectors. The chemical mapping was determined by an energy-dispersive analysis of the X-ray (EDX) using Vega TESCAN scanning electron microscopy attached with a Bruker XFlash 5010 Detector (Bruker, Billerica, MA, USA), with an accelerating voltage of 20 kV under vacuum.

In order to determine the impregnation of TiO<sup>2</sup> on the activated carbon, all composites underwent microwave acid digestion with HNO3, HF, and HCl. Titanium content in the acid solution was analyzed by inductively coupled plasma optical emission spectrometry (ICP-OES) (Perkin Elmer Optima 8000, Perkin Elmer Inc., Waltham, MA, USA).

#### *4.3. Photo-Reactor Construction*

A continuous photo-reactor was made of glass and covered with a wood box to isolate the fluid of the external conditions. Three 15 W UV lamps of 43.74 cm were attached to the lid of the reactor in parallel distribution in order to penetrate as far as possible the fluid and obtain the greatest proximity between the UV lamps, the photo-catalyst, and the fluid. The internal part of the photo-reactor made of glass is a rectangle of 135 cm × 50 cm with walls of 9 cm. Inside of this structure glass, plates of 46 cm × 4 cm were placed perpendicularly, as shown in Scheme 1. A 5-degree angle was adapted between the reactor and the horizontal level in order to maintain stable circulation of the fluid through the path formed inside the reactor.

Constant flow rate in the reactor, thickness of the liquid, residence time of the cyanide solution, and concentration of the catalyst/composite were measured to determine the optimal operating conditions.

#### *4.4. Cyanide Adsorption Study*

Sodium cyanide solutions with concentrations of 200, 400, 600, 800, and 1000 mg/L were prepared. The adsorption tests were carried out in a batch with 50 mL of sodium cyanide solution and 0.05 g of each AC and TiO2-AC composite under continuous stirring for 24 h. The data of cyanide ion adsorption were studied using the mathematical model of the Langmuir isotherm. After adsorption-desorption equilibrium was achieved, the solid was filtered, and the cyanide ion concentration (in solution) was determined by titration with a 0.2256 N AgNO<sup>3</sup> solution [28].

The data were studied and modeled by the Langmuir isotherm according to the mathematical relationship described in Equation (7).

$$\frac{\mathbf{C\_e}}{\mathbf{q\_e}} = \frac{1}{\mathbf{q\_{max}}} \times \mathbf{C\_e} + \frac{1}{\mathbf{q\_{max}} \times \mathbf{b}} \tag{7}$$

where C<sup>e</sup> is the cyanide ion equilibrium concentration in the solution (mg/L), q<sup>e</sup> is the equilibrium concentration of cyanide ion over the adsorbents (AC or TiO2-AC) (mg/g), qmax is the maximum mass of cyanide ion adsorbed per 1 g of adsorbent (AC or TiO2-AC) (mg/g), and b is the independent variable referring to the free energy of adsorption (L/mg).

#### *4.5. Adsorption Study*

#### 4.5.1. Adsorption with AC

Adsorption tests were carried out in a photo-reactor of continuous flow under dark conditions at room temperature. A sodium cyanide solution of 500 mg/L was recirculated through a peristaltic pump (Geopump Inc., Medina, NY, USA) for 24 h. The pH of solution was adjusted at 10.5 throughout the adsorption experiments with the addition of NaOH. Concentrations of 30, 45, and 60 g/L of activated carbon were used for each test. Samples of 5 mL were taken each 10 min during the first hour and then each 30 min for three hours. The solution was recirculated overnight, and the next day, samples of 5 mL were taken each hour for 2 h. The cyanide ion concentration was determined by titration with a 0.2256 N AgNO<sup>3</sup> solution.

## 4.5.2. Adsorption with TiO2-AC Composite and Photodegradation Process

Cyanide ion adsorption on AC and the photodegradation of cyanide ion occur simultaneously. Experiments of cyanide ion adsorption on the TiO2-AC composite were carried out under the same conditions as in the adsorption with AC. At the end of the adsorption experiments, the photodegradation of cyanide ion was evaluated for 24 h under UV light. Samples of 5 mL were taken (in the time intervals explained in Section 4.5.1) to determine the cyanide ion concentration by titration with a 0.2256 N AgNO<sup>3</sup> solution.

#### *4.6. Photocatalytic Cyanide Ion Degradation*

Photocatalytic activity was performed in a continuous photo-reactor using an initial concentration of sodium cyanide of 500 mg/L and a composite concentration of 30, 45, and 60 g/L at pH 10.5 and room temperature. The system (Scheme 1) was maintained in dark conditions during 24 h to ensure adsorption-desorption equilibrium. Then, UV lamps (Philips, Amsterdam, The Netherlands, T8 G13 UV-C 15 W) were turned on to initiate the photodegradation. Aliquots of 5 mL were taken (in the time intervals explained in Section 4.5.1) to analyze the residual cyanide ion by titration with a 0.2256 N AgNO<sup>3</sup> solution.

Cyanide ion photodegradation with AC (saturated with cyanide) and TiO<sup>2</sup> of 60 g/L and 0.45 g/L (corresponding to the impregnation percentage) respectively were carried out in order to compare with the photodegradation results with the composite.

After the photodegradation test, samples of the remaining solution were taken to determine the concentration of titanium and sodium dissolved by atomic absorption spectroscopy.

#### **5. Conclusions**

Within this study, we built and performed a continuous flow photo-reactor during cyanide ion degradation using TiO2-AC composites. Design operational conditions were established into the photoreactor where a minimum distance between UV source and fluid/composites and a stable continuous flow though immobilized composites were achieved. Therefore, inside of the reactor, adsorption and photocatalysis of CN− synthetic solutions were performed using TiO2-AC in concentrations of 30, 45, and 60 g/L. When

both processes, adsorption and photocatalysis, were studied separately, the adsorption of CN<sup>−</sup> on AC in light absence decreased when TiO2-AC was performed, achieving 75% of CN<sup>−</sup> removal. Conversely, TiO2-AC composites markedly enhance the photocatalytic process in comparison to individual TiO<sup>2</sup> and AC performances, achieving 99% of cyanide ion degradation. During 24 h of simultaneous process, 96% of CN− was aimed. In the success operational conditions of the photoreactor, we determined that AC could contribute to photocatalysis cyanide degradation, since AC showed 10% of CN− removal with UV irradiation. By other hand, in light absence, granular AC tested showed a considerable amount of cyanide degradation since –OH bonds on the AC surface and O<sup>2</sup> added to the system could contribute to cyanide removal. Thus, this reactor design and TiO2-AC would represent an encouraging alternative of cyanide degradation in a continuous flow system due to their synergic effect in wastewaters remediation.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/10 .3390/catal11080924/s1, Figure S1. (a) SEM micrograph of AC, 783x; (b) SEM micrograph of TiO<sup>2</sup> -AC composite, 927x; (c) EDX mapping performed on the micrograph of the TiO<sup>2</sup> -AC composite, 927x, Figure S2. EDX mapping performed on a micrograph of the TiO<sup>2</sup> -AC composite, 783x.

**Author Contributions:** Conceptualization, S.C. and D.E.; methodology, S.C., A.B.L. and D.E.; formal analysis, S.C., A.B.L.; investigation, S.C., A.B.L., E.d.l.T. and D.E.; resources, E.d.l.T. and D.E.; data curation, S.C., A.B.L.; writing—original draft preparation, S.C., A.B.L., L.E.M.-P.; writing—review and editing, S.C., A.B.L., L.E.M.-P. and D.E.; visualization, S.C., A.B.L., L.E.M.-P., E.d.l.T. and D.E.; supervision, L.E.M.-P., E.d.l.T. and D.E.; project administration, E.d.l.T. and D.E. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** The authors would like to thank the financial support provided by Escuela Politécnica Nacional through the project PII-DEMEX-20-01 belonging to the Master of Research in Metallurgy.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Zhiguo Sun \* , Yue Zhou, Shichao Jia, Yaru Wang, Dazhan Jiang and Li Zhang \***

School of Environmental and Material Engineering, Shanghai Polytechnic University, Shanghai 201209, China; zhouyueyue626@163.com (Y.Z.); jiashichao2021@163.com (S.J.); wangyaru8468@163.com (Y.W.); jiangdazhan2021@163.com (D.J.)

**\*** Correspondence: zgsun@sspu.edu.cn (Z.S.); zhangli@sspu.edu.cn (L.Z.); Tel.: +86-21-50211217 (Z.S.)

**Abstract:** A novel method of improving the SO<sup>2</sup> absorption performance of sodium citrate (Ci-Na) using sodium humate (HA–Na) as an additive was put forward. The influence of different Ci-Na concentration, inlet SO<sup>2</sup> concentration and gas flow rate on desulfurization performance were studied. The synergistic mechanism of SO<sup>2</sup> absorption by HA–Na and Ci-Na was also analyzed. The consequence shows that the efficiency of SO<sup>2</sup> absorption by Ci-Na is above 90% and the desulfurization time added with the Ci-Na concentration rising from 0.01 to 0.1 mol/L. Both the desulfurization efficiency and time may increase with the adding of HA–Na quality in Ci-Na solution. Due to adding HA–Na, the desulfurization efficiency of Ci-Na increased from 90% to 99% and the desulfurization time increased from 40 to 55 min. Under the optimum conditions, the desulfurization time of Ci-Na can exceed 70 min because of adding HA–Na, which is nearly doubled. The growth of inlet SO<sup>2</sup> concentration has little effect on the desulfurization efficiency. The SO<sup>2</sup> adsorption efficiency decreases with the increase of inlet flow gas. The presence of O<sup>2</sup> improves the SO<sup>2</sup> removal efficiency and prolongs the desulfurization time. Therefore, HA–Na plays a key role during SO<sup>2</sup> absorption and can dramatically enhance the SO<sup>2</sup> adsorption performance of Ci-Na solution.

**Keywords:** sodium citrate; sodium humate; SO<sup>2</sup> ; absorption

#### **1. Introduction**

It is well known that fossil fuels is mainly used to generate electrical energy in power plants and the combustion of fossil fuels generates SO2, which is the major source of acid rain and a major air pollutant, which severely influences the atmosphere environment and human health if not controlled [1,2]. Controlling SO<sup>2</sup> is critical to improve air quality and has always caught people's eye in recent years due to the environmental issues [3–5]. Therefore, improving the desulfurization performance economically and effectively of existing desulfurization technology has become a research hotspot at home and abroad [6–8].

There are plenty of desulfurization processes developed on the laboratory scale, some of which are applied at industrial standards around the world [9]. In the traditional methods, limestone, sodium hydroxide solutions, calcium hydroxide and magnesium hydroxide and a number of organic solvents have been used as an adsorbent [10]. There are other desulfurization processes such as the citrate method. During the citrate process, SO<sup>2</sup> in the flue gas is absorbed by the sodium citrate (Ci-Na) solution [11,12]. According to the physical characteristics of Ci-Na [13], adopting the desulfurization technology of Ci-Na can meet the advantages of environmental protection, flexible operation, recyclable absorbents and recyclable SO<sup>2</sup> for resource utilization, also meeting the increasingly strict requirements of environmental protection, desulfurization, etc. [14–16].

Humic acid (HA) is a type of amorphous organic molecular compound, most of these extensively exist in nature. It can be obtained from lignite and peat [17,18]. Due to its "sponge-like" structure, HA produces a large surface area (330–340 m2/g) and surface energy and has a strong adsorption capacity [19]. The adsorption capacity of HA is not

**Citation:** Sun, Z.; Zhou, Y.; Jia, S.; Wang, Y.; Jiang, D.; Zhang, L. Enhanced SO<sup>2</sup> Absorption Capacity of Sodium Citrate Using Sodium Humate. *Catalysts* **2021**, *11*, 865. https://doi.org/10.3390/ catal11070865

Academic Editors: José Ignacio Lombraña, Héctor Valdés and Cristian Ferreiro

Received: 9 June 2021 Accepted: 18 July 2021 Published: 20 July 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

only related to its surface area and surface energy, but also the swelling property of HA to water [20]. Sodium humate (HA–Na) is a water-soluble sodium salt of HA and a costeffective absorbent, which reacts with H<sup>+</sup> to produce HA precipitate thus promoting the dissolution of SO<sup>2</sup> in the water [21]. They have been studied more broadly for biological breeding and pollution control due to their characteristics of adsorption, chelation and ion exchange [22]. HA–Na has higher swelling property than HA itself [23]. With the enhancement of the swelling property of HA, the active groups of HA can be more fully exposed in the aqueous solution and the probability of contact between HA and adsorbed ions is increased and then it improves the adsorption effect [24–26]. However, there have been few reports with regard to the addition of HA–Na to modify the Ci-Na solution to improve the adsorption capacity [27].

Sun et al. [28] investigated the desulfurization activity of the HA–Na/α-Al2O<sup>3</sup> composite adsorbent on the fixed-bed quartz reactor. A series of characterization showed that coating α-Al2O<sup>3</sup> fibers after being immersed in HA–Na solution can enhance the flue gas desulfurization performance of the α-Al2O<sup>3</sup> carrier. The reason is that the HA–Na adsorbent has a stronger adsorption capacity for NH4OH. The longer the conversion rate of SO<sup>2</sup> is maintained, the more NH4OH will be adsorbed in the HA–Na/α-Al2O<sup>3</sup> adsorbent. According to the previous study, it revealed that the HA–Na solution has good SO<sup>2</sup> absorption characteristics. The desulfurization products can be made into the HA compound fertilizer, which provides an economical and effective way to reduce SO<sup>2</sup> from flue gas [29].

This paper studies the absorption performance of HA–Na/Ci-Na and the desulfurization mechanism, which will lay the foundation for further research and popularization in the future.

#### **2. The Enhancement Mechanism**

HA–Na may bring an enhancement effect on SO<sup>2</sup> capture by the Ci-Na method. Table 1 shows the relevant reactions and the enhancement mechanism was put forward as follows: (1) The SO<sup>2</sup> absorption by Ci-Na mainly depend on the buffering properties of its absorbing solution. (2) After adding HA–Na, the carboxyl (COO–) and hydroxyl (OH–) of HA–Na reacts with H<sup>+</sup> rapidly and HA–Na is transferred to the HA sediment (Equation (8)). Due to these reactions, the reaction equilibrium of Equations (1)–(7) moves to the right and the amount of SO<sup>2</sup> dissolved into solution is increased. (3) HA–Na may reduce the rate of pH decline of Ci-Na solution since the HA–Na solution is also a kind of acidic buffer solution, which also may enhance SO<sup>2</sup> absorption.


**Table 1.** The reaction equation of CO<sup>2</sup> capture.

R–(COONa)<sup>n</sup> is the structural formula of HA–Na and R–(COOH)<sup>n</sup> is the structural formula of HA.

## **3. Results and Discussion**

#### *3.1. Desulfurization Performance of Only Ci-Na*

The influence of different concentrations of Ci-Na on the removal rate of SO<sup>2</sup> was analyzed as can be seen in Figure 1. It shows the relationship of Ci-Na concentration and SO<sup>2</sup> desulfurization efficiency [30]. The SO<sup>2</sup> absorption by different concentrations of

Ci-Na all shows higher efficiency and the SO<sup>2</sup> absorption efficiency had no obvious change and basically maintained above 90% with the increase of Ci-Na concentration. The duration of high efficiency desulfurization also added with the increasing of Ci-Na concentration. When the concentration of Ci-Na added from 0.01 to 0.1 mol/L, the desulfurization time was extended from 20 to 80 min, which was increased by 4 times. Na all shows higher efficiency and the SO2 absorption efficiency had no obvious change and basically maintained above 90% with the increase of Ci-Na concentration. The duration of high efficiency desulfurization also added with the increasing of Ci-Na concentration. When the concentration of Ci-Na added from 0.01 to 0.1 mol/L, the desulfurization time was extended from 20 to 80 min, which was increased by 4 times. and basically maintained above 90% with the increase of Ci-Na concentration. The duration of high efficiency desulfurization also added with the increasing of Ci-Na concentration. When the concentration of Ci-Na added from 0.01 to 0.1 mol/L, the desulfurization time was extended from 20 to 80 min, which was increased by 4 times.

The influence of different concentrations of Ci-Na on the removal rate of SO2 was analyzed as can be seen in Figure 1. It shows the relationship of Ci-Na concentration and SO2 desulfurization efficiency [30]. The SO2 absorption by different concentrations of Ci-

The influence of different concentrations of Ci-Na on the removal rate of SO2 was analyzed as can be seen in Figure 1. It shows the relationship of Ci-Na concentration and SO2 desulfurization efficiency [30]. The SO2 absorption by different concentrations of Ci-Na all shows higher efficiency and the SO2 absorption efficiency had no obvious change

*Catalysts* **2021**, *11*, x FOR PEER REVIEW 3 of 10

*Catalysts* **2021**, *11*, x FOR PEER REVIEW 3 of 10

*3.1. Desulfurization Performance of Only Ci-Na* 

*3.1. Desulfurization Performance of Only Ci-Na* 

**Figure 1.** Effect of Ci-Na concentration on desulfurization efficiency. SO2 = 2300 ppm, gas flow = 1.68 L/min and absorption solution = 60 mL and 25 °C. **Figure 1.** Effect of Ci-Na concentration on desulfurization efficiency. SO<sup>2</sup> = 2300 ppm, gas flow = 1.68 L/min and absorption solution = 60 mL and 25 ◦C. 1.68 L/min and absorption solution = 60 mL and 25 °C.

Figure 2 shows the relationship between desulfurization time and Ci-Na concentra-

Figure 2 shows the relationship between desulfurization time and Ci-Na concentration. The desulfurization time experienced two rapid growth phases with the growth of Ci-Na concentration and it tended to be flat after 0.08 mol/L. The Ci-Na solution is weakly alkaline and the citrate ion has good buffering capacity. In this experiment, the concentration of 0.06 mol/L Ci-Na was selected as the optimum condition and the SO2 absorption Figure 2 shows the relationship between desulfurization time and Ci-Na concentration. The desulfurization time experienced two rapid growth phases with the growth of Ci-Na concentration and it tended to be flat after 0.08 mol/L. The Ci-Na solution is weakly alkaline and the citrate ion has good buffering capacity. In this experiment, the concentration of 0.06 mol/L Ci-Na was selected as the optimum condition and the SO<sup>2</sup> absorption efficiency was 96.4% and the duration was 40 min. tion. The desulfurization time experienced two rapid growth phases with the growth of Ci-Na concentration and it tended to be flat after 0.08 mol/L. The Ci-Na solution is weakly alkaline and the citrate ion has good buffering capacity. In this experiment, the concentration of 0.06 mol/L Ci-Na was selected as the optimum condition and the SO2 absorption efficiency was 96.4% and the duration was 40 min.

efficiency was 96.4% and the duration was 40 min.

**Figure 2.** The effect of Ci-Na concentration on desulfurization time. SO2 = 2300 ppm, gas flow = 1.68 L/min and absorption solution = 60 mL and 25 °C. **Figure 2.** The effect of Ci-Na concentration on desulfurization time. SO<sup>2</sup> = 2300 ppm, gas flow = 1.68 L/min and absorption solution = 60 mL and 25 ◦C.

#### **Figure 2.** The effect of Ci-Na concentration on desulfurization time. SO2 = 2300 ppm, gas flow = 1.68 L/min and absorption solution = 60 mL and 25 °C. *3.2. Effect of HA–Na Concentration*

The different quantity of HA–Na was a significant factor on the reduction of SO<sup>2</sup> concentration, hence a series of experiments were carried out to study the effect of quantity on desulfurization efficiency [31]. The desulfurization efficiency using only HA–Na solution was shown in Figure 3. With the quantity increasing of HA–Na, the SO<sup>2</sup> absorption efficiency increased from 82% to 98%, which also had a certain impact on the break-through time. In addition, as the amount of HA–Na quantity increased (from 0.05 to 2.4 g), the

desulfurization time also was enhanced and almost remained above 40 min when HA–Na mass was 2.4 g. The reason is as follows: the HA–Na solution is alkaline (generally the PH value is 10), and the hydroxide (OH−) in the solution is rapidly neutralized with the generated H<sup>+</sup> . Moreover, a large number of acid ions ionized by HA–Na (such as COO- and OH−), which will interact with a large number of H<sup>+</sup> . The H<sup>+</sup> combines with HA–Na to generate HA precipitation, which moves the dissolution balance to the right and promotes the dissolution of more SO<sup>2</sup> into the HA–Na solution. g), the desulfurization time also was enhanced and almost remained above 40 min when HA–Na mass was 2.4 g. The reason is as follows: the HA–Na solution is alkaline (generally the PH value is 10), and the hydroxide (OH−) in the solution is rapidly neutralized with the generated H+. Moreover, a large number of acid ions ionized by HA–Na (such as COOand OH−), which will interact with a large number of H+. The H+ combines with HA–Na to generate HA precipitation, which moves the dissolution balance to the right and promotes the dissolution of more SO2 into the HA–Na solution.

The different quantity of HA–Na was a significant factor on the reduction of SO2 concentration, hence a series of experiments were carried out to study the effect of quantity on desulfurization efficiency [31]. The desulfurization efficiency using only HA–Na solution was shown in Figure 3. With the quantity increasing of HA–Na, the SO2 absorption efficiency increased from 82% to 98%, which also had a certain impact on the breakthrough time. In addition, as the amount of HA–Na quantity increased (from 0.05 to 2.4

*Catalysts* **2021**, *11*, x FOR PEER REVIEW 4 of 10

*3.2. Effect of HA–Na Concentration* 

**Figure 3.** The effect of different quality HA–Na on desulfurization efficiency. SO2 = 2300 ppm, gas flow = 1.68 L/min and absorption solution = 60 mL and 25 °C. **Figure 3.** The effect of different quality HA–Na on desulfurization efficiency. SO<sup>2</sup> = 2300 ppm, gas flow = 1.68 L/min and absorption solution = 60 mL and 25 ◦C.

#### *3.3. Effect of the Additive Amount of HA–Na on the Desulfurization Performance of Ci-Na 3.3. Effect of the Additive Amount of HA–Na on the Desulfurization Performance of Ci-Na*

The addition of a different quantity of HA–Na may be one of the factors affecting the desulfurization efficiency of Ci-Na [32]. HA–Na was added into Ci-Na solution as an additive, such as 0.05 g, 0.1 g, 0.2 g, 0.4 g, 0.8 g, 1.2 g and 2.4 g, respectively, and the desulfurization effect was shown in Figure 4. The SO2 absorption efficiency increased as the adding amount of HA–Na, and the saturation time also was improved, from 40 to 70 min. The reason may be that the addition of HA–Na increases the hydroxide ion (OH−) in the solution, promoting more SO2 absorption. Ci-Na and HA–Na had a synergistic effect for SO2 absorption. This is more clearly confirmed in Figure 5. The addition of a different quantity of HA–Na may be one of the factors affecting the desulfurization efficiency of Ci-Na [32]. HA–Na was added into Ci-Na solution as an additive, such as 0.05 g, 0.1 g, 0.2 g, 0.4 g, 0.8 g, 1.2 g and 2.4 g, respectively, and the desulfurization effect was shown in Figure 4. The SO<sup>2</sup> absorption efficiency increased as the adding amount of HA–Na, and the saturation time also was improved, from 40 to 70 min. The reason may be that the addition of HA–Na increases the hydroxide ion (OH−) in the solution, promoting more SO<sup>2</sup> absorption. Ci-Na and HA–Na had a synergistic effect for SO<sup>2</sup> absorption. This is more clearly confirmed in Figure 5. *Catalysts* **2021**, *11*, x FOR PEER REVIEW 5 of 10

**Figure 4.** Effect of HA–Na additive on the desulfurization effect. SO2 = 2300 ppm, gas flow = 1.68 L/min, absorption solution = 60 mL and Ci-Na = 0.06 mol/L and 25 °C. **Figure 4.** Effect of HA–Na additive on the desulfurization effect. SO<sup>2</sup> = 2300 ppm, gas flow = 1.68 L/min, absorption solution = 60 mL and Ci-Na = 0.06 mol/L and 25 ◦C.

**Figure 5.** Comparison of HA–Na and Ci-Na solution. SO2 = 2300 ppm, gas flow = 1.68 L/min, ab-

It is evident from Figure 5 that the addition of HA–Na can remarkably enhance the desulfurization efficiency and saturation time of Ci-Na. It was also found that the SO2 absorption efficiency was close to 0% at 40 min when there was no HA–Na added, but it was still about 50% after adding HA–Na. Moreover, the desulfurization time increased by 15 min. The causes of this phenomenon are various [33]. In addition to the hydrolysis of HA–Na to generate hydroxide ions, it can also ionize the acid radical ions (carboxylate), thus consuming H+ to move the dissolution balance to the right and cooperating with Ci-

The concentration of SO2 is different in the actual industrial flue gas. Hence, it might be necessary to research the influence of SO2 concentration on SO2 absorption efficiency. The influence of different SO2 concentrations on the SO2 removal efficiency as illustrated in Figure 6. Simulated flue gas with the SO2 concentrations of 1000 ppm, 2300 ppm and 3000 ppm were used for the desulfurization experiment. The results are represented in

sorption solution = 60 mL and Ci-Na = 0.06 mol/L and 25 °C.

Na to absorb more SO2.

*3.4. Effect of the Inlet SO2 Concentration* 

L/min, absorption solution = 60 mL and Ci-Na = 0.06 mol/L and 25 °C.

**Figure 5.** Comparison of HA–Na and Ci-Na solution. SO2 = 2300 ppm, gas flow = 1.68 L/min, absorption solution = 60 mL and Ci-Na = 0.06 mol/L and 25 °C. **Figure 5.** Comparison of HA–Na and Ci-Na solution. SO<sup>2</sup> = 2300 ppm, gas flow = 1.68 L/min, absorption solution = 60 mL and Ci-Na = 0.06 mol/L and 25 ◦C.

**Figure 4.** Effect of HA–Na additive on the desulfurization effect. SO2 = 2300 ppm, gas flow = 1.68

It is evident from Figure 5 that the addition of HA–Na can remarkably enhance the desulfurization efficiency and saturation time of Ci-Na. It was also found that the SO2 absorption efficiency was close to 0% at 40 min when there was no HA–Na added, but it was still about 50% after adding HA–Na. Moreover, the desulfurization time increased by 15 min. The causes of this phenomenon are various [33]. In addition to the hydrolysis of HA–Na to generate hydroxide ions, it can also ionize the acid radical ions (carboxylate), thus consuming H+ to move the dissolution balance to the right and cooperating with Ci-Na to absorb more SO2. It is evident from Figure 5 that the addition of HA–Na can remarkably enhance the desulfurization efficiency and saturation time of Ci-Na. It was also found that the SO<sup>2</sup> absorption efficiency was close to 0% at 40 min when there was no HA–Na added, but it was still about 50% after adding HA–Na. Moreover, the desulfurization time increased by 15 min. The causes of this phenomenon are various [33]. In addition to the hydrolysis of HA–Na to generate hydroxide ions, it can also ionize the acid radical ions (carboxylate), thus consuming H<sup>+</sup> to move the dissolution balance to the right and cooperating with Ci-Na to absorb more SO2.

#### *3.4. Effect of the Inlet SO2 Concentration 3.4. Effect of the Inlet SO<sup>2</sup> Concentration*

The concentration of SO2 is different in the actual industrial flue gas. Hence, it might be necessary to research the influence of SO2 concentration on SO2 absorption efficiency. The influence of different SO2 concentrations on the SO2 removal efficiency as illustrated in Figure 6. Simulated flue gas with the SO2 concentrations of 1000 ppm, 2300 ppm and 3000 ppm were used for the desulfurization experiment. The results are represented in The concentration of SO<sup>2</sup> is different in the actual industrial flue gas. Hence, it might be necessary to research the influence of SO<sup>2</sup> concentration on SO<sup>2</sup> absorption efficiency. The influence of different SO<sup>2</sup> concentrations on the SO<sup>2</sup> removal efficiency as illustrated in Figure 6. Simulated flue gas with the SO<sup>2</sup> concentrations of 1000 ppm, 2300 ppm and 3000 ppm were used for the desulfurization experiment. The results are represented in Figure 6 that with the increase of SO<sup>2</sup> concentration, the desulfurization time of reaching saturation decreased from 92 to 40 min and diminished by 2.3 times. Moreover, the desulfurization time decreased significantly at 40 min, only about 5% under the high SO<sup>2</sup> concentration condition, while the desulfurization time was still close to 100% under the condition of low SO<sup>2</sup> concentration. The result shows that the inlet SO<sup>2</sup> concentration had a certain influence on the removal efficiency.

The main reason is that the driving force of mass transfer increased with the increasing of SO<sup>2</sup> concentration, which is beneficial to the absorption reaction [34]. However, the SO<sup>2</sup> capacity per unit volume of the solution was constant. As the inlet SO<sup>2</sup> concentration increased, the mass transfer rate was heightened while the time of SO<sup>2</sup> absorption saturation was shortened. So, the SO<sup>2</sup> absorption rate will be accelerated and the desulfurization time will be reduced.

certain influence on the removal efficiency.

**Figure 6.** Effect of SO2 concentration on desulfurization efficiency. Gas flow = 1.68 L/min, absorption **Figure 6.** Effect of SO<sup>2</sup> concentration on desulfurization efficiency. Gas flow = 1.68 L/min, absorption solution = 60 mL, Ci-Na = 0.06 mol/L and HA–Na = 0.2 g and 25 ◦C.

Figure 6 that with the increase of SO2 concentration, the desulfurization time of reaching saturation decreased from 92 to 40 min and diminished by 2.3 times. Moreover, the desulfurization time decreased significantly at 40 min, only about 5% under the high SO2 concentration condition, while the desulfurization time was still close to 100% under the condition of low SO2 concentration. The result shows that the inlet SO2 concentration had a

#### solution = 60 mL, Ci-Na = 0.06 mol/L and HA–Na = 0.2 g and 25 °C. *3.5. Effect of the Gas Flow Rate*

The main reason is that the driving force of mass transfer increased with the increasing of SO2 concentration, which is beneficial to the absorption reaction [34]. However, the SO2 capacity per unit volume of the solution was constant. As the inlet SO2 concentration increased, the mass transfer rate was heightened while the time of SO2 absorption saturation was shortened. So, the SO2 absorption rate will be accelerated and the desulfurization time will be reduced. *3.5. Effect of the Gas Flow Rate*  Most of the experiments were discussed as the influence of the gas flow rate on SO<sup>2</sup> removal efficiency. The initial inlet gas flow rate were respectively set as 1.0 L/min, 1.3 L/min and 1.6 L/min in the experiment. The results was presented in Figure 7. It is proved by the experiment that the removal efficiency of SO<sup>2</sup> increased as the initial inlet gas flow decreased. The increasing of the gas flow rate reduced the driving force of the absorption reaction, which is unfavorable for the desulfurization reaction. At the same time, the gas flow rate increased and the gas–liquid reaction time was reduced. A part of SO<sup>2</sup> was released before the reaction, which affected the absorption efficiency. In general, increasing gas flow had only some negative consequences. *Catalysts* **2021**, *11*, x FOR PEER REVIEW 7 of 10

**Figure 7.** Effect of the gas flow rate. SO2 = 2300 ppm, absorption solution = 60 mL, Ci-Na = 0.06 mol/L and HA–Na = 0.2 g and 25 °C. **Figure 7.** Effect of the gas flow rate. SO<sup>2</sup> = 2300 ppm, absorption solution = 60 mL, Ci-Na = 0.06 mol/L and HA–Na = 0.2 g and 25 ◦C.

## *3.6. Effect of O<sup>2</sup>*

*3.6. Effect of O2* The actual industrial flue gas contains a variety of ingredients, such as O2. For instance, the flue gas of coal-fired power plant typically contains about 5–15 vol% O2 [35]. Therefore, the existence of O2 in the simulated flue gas was also explored. Figure 8 shows the effect of the presence of O2 on the SO2 removal efficiency. The experimental results The actual industrial flue gas contains a variety of ingredients, such as O2. For instance, the flue gas of coal-fired power plant typically contains about 5–15 vol% O<sup>2</sup> [35]. Therefore, the existence of O<sup>2</sup> in the simulated flue gas was also explored. Figure 8 shows the effect of the presence of O<sup>2</sup> on the SO<sup>2</sup> removal efficiency. The experimental results indicated that SO<sup>2</sup> absorption efficiency at 50 min was improved significantly from 5% to 95% by the

furization efficiency noticeably and also prolonged the desulfurization time from 55 to 80 min. The possible reasons were as follows [36]. The presence of O2 could be more effective in improving O2 dissolving into water so that the centration of dissolved O2 into the solution was far higher than before, which is conducive to the oxidation of sulfite. According to Equation (9), it can infer that dissolved O2 could accelerate the oxidation of sulfate. This

**Figure 8.** Effect of O2. SO2 = 2300 ppm, gas flow = 1.68 L/min, absorption solution = 60 mL, Ci-Na =

ି in the liquid phase and makes Equation (10) shift to

the right. The liquid phase mass transfer coefficient was reduced.

reduces the concentration of HSOଷ

0.06 mol/L and HA–Na = 0.2 g and 25 °C.

addition of 15% O2. It can be seen that the existence of O<sup>2</sup> increased the desulfurization efficiency noticeably and also prolonged the desulfurization time from 55 to 80 min. The possible reasons were as follows [36]. The presence of O<sup>2</sup> could be more effective in improving O<sup>2</sup> dissolving into water so that the centration of dissolved O<sup>2</sup> into the solution was far higher than before, which is conducive to the oxidation of sulfite. According to Equation (9), it can infer that dissolved O<sup>2</sup> could accelerate the oxidation of sulfate. This reduces the concentration of HSO− 3 in the liquid phase and makes Equation (10) shift to the right. The liquid phase mass transfer coefficient was reduced. furization efficiency noticeably and also prolonged the desulfurization time from 55 to 80 min. The possible reasons were as follows [36]. The presence of O2 could be more effective in improving O2 dissolving into water so that the centration of dissolved O2 into the solution was far higher than before, which is conducive to the oxidation of sulfite. According to Equation (9), it can infer that dissolved O2 could accelerate the oxidation of sulfate. This reduces the concentration of HSOଷ ି in the liquid phase and makes Equation (10) shift to the right. The liquid phase mass transfer coefficient was reduced.

**Figure 7.** Effect of the gas flow rate. SO2 = 2300 ppm, absorption solution = 60 mL, Ci-Na = 0.06 mol/L

The actual industrial flue gas contains a variety of ingredients, such as O2. For instance, the flue gas of coal-fired power plant typically contains about 5–15 vol% O2 [35]. Therefore, the existence of O2 in the simulated flue gas was also explored. Figure 8 shows the effect of the presence of O2 on the SO2 removal efficiency. The experimental results indicated that SO2 absorption efficiency at 50 min was improved significantly from 5% to 95% by the addition of 15% O2. It can be seen that the existence of O2 increased the desul-

*Catalysts* **2021**, *11*, x FOR PEER REVIEW 7 of 10

**Figure 8.** Effect of O2. SO2 = 2300 ppm, gas flow = 1.68 L/min, absorption solution = 60 mL, Ci-Na = 0.06 mol/L and HA–Na = 0.2 g and 25 °C. **Figure 8.** Effect of O<sup>2</sup> . SO<sup>2</sup> = 2300 ppm, gas flow = 1.68 L/min, absorption solution = 60 mL, Ci-Na = 0.06 mol/L and HA–Na = 0.2 g and 25 ◦C.

After the addition of O2, the desulfurization process will be accompanied by the following reactions.

$$2\text{SO}\_3^{2-}\text{(aq)} + \text{O}\_2\text{(g)} \rightarrow 2\text{SO}\_4^{2-}\text{(aq)}\tag{9}$$

$$\mathrm{HSO\_3^-(aq)} \leftrightarrow \mathrm{H^+(aq)} + \mathrm{SO\_3^{2-}(aq)}\tag{10}$$

It could be deduced that more O<sup>2</sup> in the solution participated in the desulfurization reaction and both the amount of SO<sup>2</sup> absorbed and desulfurization time were increased.

#### **4. Materials and Methods**

and HA–Na = 0.2 g and 25 °C.

*3.6. Effect of O2*

#### *4.1. Sample Preparation*

Ci-Na, sodium hydroxide, sodium acetate, acetic acid solution, anhydrous ethanol and sodium carbonate were from Sino pharm Chemical Reagent Co., Ltd., in Shanghai, China. HA–Na was from Shanghai Jincheng Biochemical Co., Ltd, in Shanghai, China. Deionized water was made in the laboratory.

#### *4.2. Desulfurization Test*

A principle diagram of the experimental devices are represented in Figure 9 below. Absorption experiments of the SO<sup>2</sup> in the laboratory consisted of SO2, O<sup>2</sup> and balance N<sup>2</sup> as simulated flue gas. The SO2, O<sup>2</sup> and N<sup>2</sup> gases were provided by cylinders. The experiment adopted SO<sup>2</sup> with a concentration range of 1000–3000 ppm. The total flow rate of the simulated flue gas was controlled with a mass flow controller (MFC). The flue gas analyzer was used to monitor the change of SO<sup>2</sup> concentration at the inlet and outlet of the reactor (KANE-9506, UK) [29].

**Figure 9.** Schematic diagram of the experimental apparatus. **Figure 9.** Schematic diagram of the experimental apparatus.

**5. Conclusions**  The absorption efficiency SO<sup>2</sup> can be obtained by the following formula:

$$\eta = \frac{\left(\text{C}\_{\text{in}} - \text{C}\_{\text{out}}\right) \times 100\%}{\text{C}\_{\text{in}}} \tag{11}$$

elements on the desulfurization performance were studied. The mechanism of HA–Na as an addition agent to improve the desulfurization performance of Ci-Na was discussed. For the absorption process, the higher the Ci-Na concentration and the lower the inlet flue where *η* is the SO<sup>2</sup> absorption efficiency and Can and Coot are the inlet and outlet of the SO<sup>2</sup> concentration, respectively.

After the addition of O2, the desulfurization process will be accompanied by the fol-

It could be deduced that more O2 in the solution participated in the desulfurization reaction and both the amount of SO2 absorbed and desulfurization time were increased.

Ci-Na, sodium hydroxide, sodium acetate, acetic acid solution, anhydrous ethanol and sodium carbonate were from Sino pharm Chemical Reagent Co., Ltd., in Shanghai, China. HA–Na was from Shanghai Jincheng Biochemical Co., Ltd, in Shanghai, China.

A principle diagram of the experimental devices are represented in Figure 9 below. Absorption experiments of the SO2 in the laboratory consisted of SO2, O2 and balance N2 as simulated flue gas. The SO2, O2 and N2 gases were provided by cylinders. The experiment adopted SO2 with a concentration range of 1000–3000 ppm. The total flow rate of the simulated flue gas was controlled with a mass flow controller (MFC). The flue gas analyzer was used to monitor the change of SO2 concentration at the inlet and outlet of the

in

(11)

in out C <sup>C</sup> <sup>−</sup> <sup>C</sup> <sup>×</sup>100% <sup>=</sup> ( )

where *η* is the SO2 absorption efficiency and Can and Coot are the inlet and outlet of the SO2

The absorption efficiency SO2 can be obtained by the following formula:

η

2SO32−(aq) + O2(g) → 2SO42−(aq) (9) HSO3−(aq) ↔ H+(aq) + SO32−(aq) (10)

#### gas flow, the more conducive to the SO2 absorption. The presence of O2 had a slight **5. Conclusions**

lowing reactions.

**4. Materials and Methods**  *4.1. Sample Preparation* 

*4.2. Desulfurization Test* 

reactor (KANE-9506, UK) [29].

concentration, respectively.

Deionized water was made in the laboratory.

The new desulfurization method with Ci-Na/HA–Na solution was put forward. The influence of different Ci-Na concentration, inlet SO<sup>2</sup> concentration, flow rate and other elements on the desulfurization performance were studied. The mechanism of HA–Na as an addition agent to improve the desulfurization performance of Ci-Na was discussed. For the absorption process, the higher the Ci-Na concentration and the lower the inlet flue gas flow, the more conducive to the SO<sup>2</sup> absorption. The presence of O<sup>2</sup> had a slight influence on the desulfurization efficiency. HA–Na played a key role during SO<sup>2</sup> absorption by the Ci-Na solution and can improve obviously the desulfurization performance of the Ci-Na solution.

**Author Contributions:** Conceptualization, Z.S.; methodology, S.J.; validation, Z.S.; investigation, Y.W. and D.J.; resources, Z.S.; writing—original draft preparation, Y.Z.; writing—review and editing, Z.S. and L.Z.; project administration, Z.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** The authors gratefully acknowledge financial support by National Natural Science Foundation of China (No. 21806101), Natural Science Foundation of Shanghai (No.16ZR1412600), Research Center of Resource Recycling Science and Engineering, Shanghai Polytechnic University and Gaoyuan Discipline of Shanghai—Environmental Science and Engineering (Resource Recycling Science and Engineering), Cultivate discipline fund of Shanghai Polytechnic University (No.XXKPY1601).

**Data Availability Statement:** No data associated with this publication to be link. All the data associated in presented in this paper.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

