*Article* **Environment-Friendly Catalytic Mineralization of Phenol and Chlorophenols with Cu- and Fe- Tetrakis(4-aminophenyl) porphyrin—Silica Hybrid Aerogels**

**Enik˝o Gy˝ori <sup>1</sup> , Ádám Kecskeméti <sup>1</sup> , István Fábián 1,2 , Máté Szarka 3,4 and István Lázár 1,\***


**Abstract:** Fenton reactions with metal complexes of substituted porphyrins and hydrogen peroxide are useful tools for the mineralization of environmentally dangerous substances. In the homogeneous phase, autooxidation of the prophyrin ring may also occur. Covalent binding of porphyrins to a solid support may increase the lifetime of the catalysts and might change its activity. In this study, highly water-insoluble copper and iron complexes of 5,10,15,20-tetrakis(4-aminophenyl)porphyrin were synthesized and bonded covalently to a very hydrophilic silica aerogel matrix prepared by co-gelation of the propyl triethoxysilyl-functionalized porphyrin complex precursors with tetramethoxysilane, followed by a supercritical carbon dioxide drying. In contrast to the insoluble nature of the porphyrin complexes, the as-prepared aerogel catalysts were highly compatible with the aqueous phase. Their catalytic activities were tested in the mineralization reaction of phenol, 3-chlorophenol, and 2,4-dichlorophenol with hydrogen peroxide. The results show that both aerogels catalyzed the oxidation of phenol and chlorophenols to harmless short-chained carboxylic acids under neutral conditions. In batch experiments, and also in a miniature continuous-flow tubular reactor, the aerogel catalysts gradually reduced their activity, due to the slow oxidation of the porphyrin ring. However, the rate and extent of the degradation was moderate and did not exclude the possibility that the as-prepared catalysts, as well as their more stable derivatives, might find practical applications in environment protection.

**Keywords:** silica aerogel; aerogel hybrid; covalent immobilization; porphyrin complexes; heterogeneous catalyst; phenol mineralization; chlorophenol mineralization

#### **1. Introduction**

Phenol is an extensively used reagent in the production of phenolic resins, bisphenol A, caprolactam, and other chemicals. As an unfortunate consequence, similarly to many compounds used in the industry, phenol can be found in wastewater or in the soil. Due to its high toxicity, complete elimination of phenol is an important environmental goal.

Chlorophenols are produced by the chlorination of phenol, resulting in 19 different compounds including isomers. All of them show significant antiseptic activity and higher toxicity than that of phenol. That property is utilized when they are applied as disinfectants, preservative agents, herbicides, or insecticides [1]. However, these compounds are not biodegradable at all. By bioaccumulation in plant and animal species, they can affect the food chain. Due to their persistence in the environment, they are listed as first-priority pollutants, so the necessity of their effective mineralization is indisputable [2]. Since

**Citation:** Gy˝ori, E.; Kecskeméti, Á.; Fábián, I.; Szarka, M.; Lázár, I. Environment-Friendly Catalytic Mineralization of Phenol and Chlorophenols with Cu- and Fe-Tetrakis(4-aminophenyl) porphyrin—Silica Hybrid Aerogels. *Gels* **2022**, *8*, 202. https://doi.org/ 10.3390/gels8040202

Academic Editor: Miguel Sanchez-Soto

Received: 17 February 2022 Accepted: 21 March 2022 Published: 23 March 2022

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

**Copyright:** © 2022 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/).

chlorinated organic compounds are resistant to biodegradation, processes other than microbial processes are needed to eliminate them [3]. Thus far, many techniques have been developed for the removal of phenols from the environment [4,5]. The most efficient methods are the advanced oxidation processes (AOPs), in which hydroxyl or sulfate radicals act as active agents. In practice, several reactions and materials are used to generate them, such as UV light combined with either hydrogen peroxide or ozone, the Fenton reaction, or salts of the peroxy monosulfate ion, for example [6–9].

In the Fenton and Fenton-like reactions, hydroxyl radical (OH−) is generated from hydrogen peroxide, although the exact mechanism is still not clarified. H2O<sup>2</sup> can oxidize Fe2+ ion to produce Fe3+ ion, hydroxide ion and highly reactive radicals (see Equations (1) and (2)) and these free radicals can easily oxidize the organic pollutants [10–12].

$$\mathrm{Fe^{2+} + H\_2O\_2 \to Fe^{3+} + OH^- + OH^\bullet} \tag{1}$$

$$\rm OH^{\bullet} + H\_2O\_2 \rightarrow HO\_2^{\bullet} + H\_2O \tag{2}$$

The Fenton-like metal ion catalysts are used mostly in the homogeneous phase. Their main drawbacks are the difficulty of separation and regeneration of the catalyst, as well as the increasing concentration of metal ions in water and the soil.

The most common catalytically active materials are metals, oxides and sulfides. The efficiency of the heterogeneous catalysts may be increased by applying them on a solid support. Although the turnover rate is in general lower compared to the homogeneous phase reactions, heterogeneous catalysis has several advantages. Easy separation from the reaction mixture, better tolerance towards extreme reaction parameters, larger surface area and a higher number of active sites make them valuable materials [13,14].

Numerous heterogeneous catalysts have been studied, such as, the Cu- or Fe-containing zeolites due to their high catalytic activity and selectivity [15–17]. Additional examples are the porphyrins, which are tetradentate macrocyclic ligands and form complexes of extremely high stability. Various porphyrin derivatives containing zinc, copper, iron, manganese, palladium, vanadium or other transition metal ions are extensively used as catalysts [18–20]. In the homogeneous phase catalytic processes, porphyrins are susceptible to self-degradation and loss of activity [21,22]. Therefore, it is an important goal to develop heterogeneous phase porphyrin catalysts possessing high activity and increased stability.

Aerogels are extremely light solid materials exhibiting unique physical and surface properties. Some of the most versatile ones are silica aerogels, which are prepared by the sol-gel technique from an organic silane precursor, and dried to a solid under supercritical conditions. The process results in a substance with specific properties, such as high and open porosity, large specific surface area, extremely low bulk density, high insulating capacity, to name a few. Thanks to these features, aerogels can be applied in many different industrial fields, for example as insulating materials, Cherenkov radiators, biomaterials, or catalysts [23–27]. The siloxane network can be functionalized by covalently binding organic moieties to the skeleton, embedding guest particles in the structure, or adsorbing metal ions on the surface. Such aerogel-based materials are widely used as catalysts in hydrogen production [28,29], dye degradation in wastewaters [30], or methanol electrooxidation [31,32].

There are several methods for the immobilization of porphyrins on a carrier. The simplest technique is adsorption of the molecules on the surface [33]. Although the process is straightforward, the chance of leaching from the matrix is rather high. Another method is the "ship in a bottle" process [34], which embeds the molecules in narrow necked cavities. A major disadvantage of such a catalyst is the limited access of substrates to the catalytically active centers. The most sophisticated way is the covalent immobilization, which prevents leaching from the solid phase, and provides a good contact with the substrates. However, the technique may require special knowledge of synthetic chemistry [35–37].

In an earlier study it was demonstrated that highly water-soluble porphyrin complexes may undergo decomposition in Fenton reactions due to their autocatalytic oxidation [38].

It was supposed that immobilization of catalytically active porphyrin complexes may decrease the rate of autooxidation. Recently, silica aerogels covalently functionalized with tetraaza macrocyclic copper complexes were prepared and tested in our laboratory for catalytic oxidation of phenols with hydrogen peroxide [26]. In this paper we report the synthesis and characterization of silica aerogels which are covalently functionalized with water-insoluble 5,10,15,20-tetrakis(4-aminophenyl)porphyrin complexes. Their catalytic activities were tested in mineralization of environment-polluting materials, phenol and chlorophenols, in batch mode and in continuous-flow microreactor and the life expectances of the catalysts were determined at different temperatures and molar ratios. substrates. However, the technique may require special knowledge of synthetic chemistry [35–37]. In an earlier study it was demonstrated that highly water-soluble porphyrin complexes may undergo decomposition in Fenton reactions due to their autocatalytic oxidation. [38] It was supposed that immobilization of catalytically active porphyrin complexes may decrease the rate of autooxidation. Recently, silica aerogels covalently functionalized with tetraaza macrocyclic copper complexes were prepared and tested in our laboratory for catalytic oxidation of phenols with hydrogen peroxide. [26] In this paper we report the synthesis and characterization of silica aerogels which are covalently functionalized with water-insoluble 5,10,15,20-tetrakis(4-aminophenyl)porphyrin complexes. Their catalytic

activities were tested in mineralization of environment-polluting materials, phenol and

#### **2. Results and Discussion** chlorophenols, in batch mode and in continuous-flow microreactor and the life expec-

#### *2.1. Preparation of the Heterogeneous Catalysts* tances of the catalysts were determined at different temperatures and molar ratios.

Scheme 1 shows the synthetic steps of the preparation. First, the porphyrin ring was functionalized with 3-isocyanatopropyl triethoxysilane (3-IPTES), which acted as a bifunctional spacer and coupling agent. A 40.4 mg (5.99 <sup>×</sup> <sup>10</sup>−<sup>5</sup> mmol) portion of TAPP was dissolved in 8.00 mL anhydrous DMF under an argon atmosphere. To that 0.67 cm<sup>3</sup> (656 mg, 2.39 <sup>×</sup> <sup>10</sup>−<sup>3</sup> mmol) of 3-IPTES was added, and the solution was heated at 70 ◦<sup>C</sup> for 96 h. Metal complexes were prepared then by carefully reacting the functionalized porphyrin rings with substoichiometric portions of either copper(2+) acetate or crystalline iron(2+) sulfate heptahydrate under anhydrous conditions in DMF or DMSO until the UV fluorescence of the free porphyrine ring disappeared. The aerogel catalysts were obtained by the ammonia-catalyzed co-hydrolysis and co-condensation of the triethoxypropylsilylfunctionalized porphyrin complexes with tetramethoxysilane (TMOS) in a methanol-water mixture using the sol-gel technique, as described earlier. After the necessary ageing and solvent exchange process, the supercritical carbon dioxide drying process resulted in aerogel monoliths [39]. The photographs of the as-prepared catalysts are shown in Scheme 1. Although it might be difficult to see their true colour in the cylindrical form, microscopic images of thin fragments revealed the red colour of the copper-containing (denoted as CuPA), and greenish brown colour of the iron-containing (denoted as FePA) aerogels. **2. Results and Discussion**  *2.1. Preparation of the Heterogeneous Catalysts*  Scheme 1 shows the synthetic steps of the preparation. First, the porphyrin ring was functionalized with 3-isocyanatopropyl triethoxysilane (3-IPTES), which acted as a bifunctional spacer and coupling agent. A 40.4 mg (5.99 × 10−5 mmol) portion of TAPP was dissolved in 8.00 mL anhydrous DMF under an argon atmosphere. To that 0.67 cm3 (656 mg, 2.39 × 10−3 mmol) of 3-IPTES was added, and the solution was heated at 70 °C for 96 h. Metal complexes were prepared then by carefully reacting the functionalized porphyrin rings with substoichiometric portions of either copper(2+) acetate or crystalline iron(2+) sulfate heptahydrate under anhydrous conditions in DMF or DMSO until the UV fluorescence of the free porphyrine ring disappeared. The aerogel catalysts were obtained by the ammonia-catalyzed co-hydrolysis and co-condensation of the triethoxypropylsilyl-functionalized porphyrin complexes with tetramethoxysilane (TMOS) in a methanol-water mixture using the sol-gel technique, as described earlier. After the necessary ageing and solvent exchange process, the supercritical carbon dioxide drying process resulted in aerogel monoliths [39]. The photographs of the as-prepared catalysts are shown in Scheme 1. Although it might be difficult to see their true colour in the cylindrical form, microscopic images of thin fragments revealed the red colour of the copper-containing (denoted as CuPA), and greenish brown colour of the iron-containing (denoted as FePA) aerogels.

**Scheme 1.** Synthetic pathway and reaction conditions of the complex formation, covalent binding and co-gelation of the porphyrin complexes with tetramethoxysilane (TMOS) leading to hybrid metal complex–aerogel catalysts FePA and CuPA. On the right side: Photographs (left) and microscopic images (40×) (right) of monolithic and fragmented pieces of CuPA and FePA catalysts.

#### *2.2. Characterization of the Catalysts*

The functionalization of the porphyrin ring with 3-isocyanatopropyltriethoxysilane (3-IPTES) was monitored by NMR spectroscopy (See in Figures S1 and S2). The differences between the two spectra can be clearly seen. The significant change in the aromatic region of the spectra (Figure S2) indicates that the coupling was successful. The complexation of the functionalized porphyrin ring with Cu(2+) and Fe(2+) ion was monitored with a 366 nm UV lamp. The non-complexed porphyrin ring had a strong red fluorescence, which disappears when the complex is formed. We could observe the vanishing of the fluorescence in the case of both metal ions.

The FT-IR spectra of the complex solutions were recorded as well, they can be found in the Supplementary Materials (Figure S3). Figure 1 shows the fingerprint regions of the spectra compared with that of the non-complexed porphyrin ring. The differences prove the change in the structure of the porphyrin ring and thus the formation of the complexes. Figure 2 shows the FT-IR spectra of an aerogel functionalized with porphyrin complex. The following peaks can be assigned to silica aerogels: The O–H stretching vibration at ~3400 cm−<sup>1</sup> , the Si–O–Si asymmetric stretching vibration at around 1050 cm−<sup>1</sup> and the asymmetric stretching vibration at approximately 950 cm−<sup>1</sup> . The vibrations of C–H bonds appearing between 2800 and 3000 cm−<sup>1</sup> indicate the successful functionalization.

The Raman spectra of the porphyrin ring, the complexes and the functionalized aerogels were also recorded. Figure 3 shows the difference between the spectrum of the empty porphyrin ring and the complexes. In the high-frequency region the Raman bands are sensitive to the electron density, the axial ligation and to the core size of the central metal ion. The band at 1543 cm−<sup>1</sup> of the porphyrin ligand can be assigned to the CβC<sup>β</sup> stretch, which was upshifted to 1546 cm−<sup>1</sup> in the Fe and 1576 cm−<sup>1</sup> in the Cu complexes, respectively. This band appears as one of the most intense bands in the spectra for the Fe and Cu complexes. The band at 1487 cm−<sup>1</sup> of the porphyrin ligand could be assigned to the phenyl ring vibration, which was practically the same in the Fe complex but was shifted to 1497 cm−<sup>1</sup> in the Cu complex, indicating a higher effect of the Cu ion on the phenyl at the meso-positions. The bands between 1300 cm−<sup>1</sup> and 1450 cm−<sup>1</sup> are most likely the out-of-phase coupled CαCβ/CαN stretching modes. The 1323 and 1360 cm−<sup>1</sup> bands of the porphyrin ligand were most probably the pyrrole quarter ring stretching and the CαCβ/CαN stretching modes, respectively. The CαCβ/CαN stretching appeared at 1333 cm−<sup>1</sup> for both the Cu and the Fe complexes too. The pyrrole stretching was found at 1360 cm−<sup>1</sup> for the Cu and Fe complexes as well. The 1236 cm−<sup>1</sup> band of the 5,10,15,20-tetrakis(4-aminophenyl)-porphyrin (TAPP) and the 1230 and 1234 cm−<sup>1</sup> bands of the Cu and Fe complexes were most likely attributed to the Cm-ph stretching. The band at 1076 cm−<sup>1</sup> of the porphyrin ligand was most probably the vibration of the pyrrole Cβ–H stretching, which appeared at the same wavenumber for the Cu complex and shifted to 1080 cm−<sup>1</sup> for the Fe complex. The band at 997 cm−<sup>1</sup> of the porphyrin ligand was most likely the vibration of pyrrole breathing and phenyl stretching, which shifted to 1001 cm−<sup>1</sup> in the case of both the Cu and the Fe complexes as well. The band at 960 cm−<sup>1</sup> of the porphyrin ligand was most probably the pyrrole breathing, but it could not be found in the complexes due to the exchange of the hydrogen atom with the metal ion in the N-H bonding. The band at 710 cm−<sup>1</sup> was most likely the π3, phenyl mode, which was not seen in the Cu complex but shifted to 714 cm−<sup>1</sup> for the Fe complex. At 384 cm−<sup>1</sup> for the Cu and at 385 cm−<sup>1</sup> for the Fe complexes the bands are assigned to the M-N vibration, which cannot be seen in the porphyrin ligand [40]. For the porphyrin ligand, a weak Raman band can be seen at 330 cm−<sup>1</sup> , which was most likely the in-plane translational motion of the pyrrole [41]. The peaks at around 1000 cm−<sup>1</sup> , 1340 cm−<sup>1</sup> and 1580 cm−<sup>1</sup> indicate the presence of the aromatic hydrocarbons in which a hydrogen has been replaced by an amino group (benzene and pyrrole). These peaks appear in the spectra of the complexes as well as in that of the empty porphyrin ring but at a slightly different position (circled peaks in the spectra). For easier comparability, adifferent intensity scale was used in the case of the Cu-porphyrin complex since the intensities were too weak compared to the other two samples. Figure 4 shows the Raman spectra of the functionalized aerogels. Since the aerogels contain a little amount of the complexes, it was quite difficult to record acceptable quality spectra. Despite this hardship, the difference between the spectrum of the CuPA aerogel and that of the free complexes clearly indicate a change in the structure of the complex by which the formation of the covalent bond between the complex and silica framework is confirmed.

The results of the elemental analysis can be seen in Table 1. The non-zero carbon and nitrogen content of the blank aerogel sample indicates that after the supercritical drying adsorbed residues of the solvents and the catalyst ammonia may be present in the aerogels. Compared to that as a background, both the carbon and nitrogen contents of CuPA and FePA are higher in the obtained samples, indicating the successful incorporation of the porphyrin complexes.

Porosity of the aerogel samples was measured by nitrogen adsorption porosimetry at 77 K temperature, after degassing the samples at 100 ◦C for 24 h. Figure 5 shows the cumulative pore volumes and pore size distribution curves calculated by the BJH method. Nitrogen adsorption-desorption isotherms can be found in the Supplementary Materials (Figure S4). The specific surface area of the CuPA and FePA aerogels was 980 m2/g, and 1019 m2/g, respectively. The actual values are significantly higher than the 600–700 m2/g values characteristic of pristine silica aerogels prepared by the same method. Bulk densities of the functionalized aerogels were in the range of 0.076–0.085 g/cm<sup>3</sup> , which was a bit lower than 0.085–0.090 g/cm<sup>3</sup> obtained for the pristine silica aerogels. Measured and calculated parameters are given in Table 2. The values are characteristic of the silica-based aerogels in general, thus the incorporation of the porphyrin complexes did not alter the gel structure significantly. Most of the pores are in the 2–50 nm mesopore region, and only a negligible volume falls in the less than 2 nm diameter micropore region, as calculated by the t-plot method.

Scanning electron microscopy (SEM) pictures were in good agreement with the porosimetry results. The samples showed the homogeneous structure, and the visible pores were in the higher mesopore and lower macropore region (Figure 5). The size of the globules and the pores are also characteristic of the silica-based aerogels. Most importantly, no detectable agglomeration of the porphyrin complexes was observed. According to these results a uniform and molecular level distribution of the complexes was obtained in the silica matrix.

The metal content of the aerogels was determined by an inductively coupled plasmaoptical emission spectrometry (ICP-OES) method, and the results are summarized in Table 3. The measured values are a bit higher than the values calculated from the chemical composition of the reaction mixtures, due to the leaching of short-chained partially hydrolized siloxanes in the ageing and solvent exchange steps. However, the colourless nature of the ageing solutions proved that no intense-coloured porphyrin complex was lost in the process, and their entire amount was incorporated.

**Figure 1.** Fingerprint region of the FT-IR spectra of the free copper and iron complexes (blue curves) compared with that of the empty porphyrin ring (red curves). The full recorded spectra can be seen in the Supplementary Materials (Figure S3). Although accurate assignation of the peaks is not available, a detailed analysis of the IR spectra of tetraphenyl porphyrin complexes is available in the literature [42]. **Figure 1.** Fingerprint region of the FT-IR spectra of the free copper and iron complexes (blue curves) compared with that of the empty porphyrin ring (red curves). The full recorded spectra can be seen in the Supplementary Materials (Figure S3). Although accurate assignation of the peaks is not available, a detailed analysis of the IR spectra of tetraphenyl porphyrin complexes is available in the literature [42]. **Figure 1.** Fingerprint region of the FT-IR spectra of the free copper and iron complexes (blue curves) compared with that of the empty porphyrin ring (red curves). The full recorded spectra can be seen in the Supplementary Materials (Figure S3). Although accurate assignation of the peaks is not available, a detailed analysis of the IR spectra of tetraphenyl porphyrin complexes is available in the literature [42].

**Figure 2.** FT-IR spectra of the aerogels CuAP and FeAP functionalized with the copper and iron porphyrin complexes. They are in good agreement with the spectra published in the literature for silica aerogels [43] and silicas covalently coupled with porphyrins [44]. Due to the low concentration of the complexes in the silica aerogel matrix, the characteristic peaks of the complexes shown in Figure 1 are too weak to be observed.

Figure 1 are too weak to be observed.

*Gels* **2022**, *8*, x FOR PEER REVIEW 7 of 20

**Figure 2.** FT-IR spectra of the aerogels CuAP and FeAP functionalized with the copper and iron porphyrin complexes. They are in good agreement with the spectra published in the literature for silica aerogels [43] and silicas covalently coupled with porphyrins [44]. Due to the low concentration of the complexes in the silica aerogel matrix, the characteristic peaks of the complexes shown in

**Figure 2.** FT-IR spectra of the aerogels CuAP and FeAP functionalized with the copper and iron porphyrin complexes. They are in good agreement with the spectra published in the literature for silica aerogels [43] and silicas covalently coupled with porphyrins [44]. Due to the low concentration of the complexes in the silica aerogel matrix, the characteristic peaks of the complexes shown in

**Figure 3.** Raman-spectra of the empty 5,10,15,20-tetrakis(4-aminophenyl)porphyrin ring (TAPP) and its Cu- and Fe-complexes. In the case of Cu-complex, a different intensity scale was used for easier comparability of the spectra. The differences between the positions of the circled peaks which can be assigned to the aromatic hydrocarbons, in which a hydrogen has been replaced by amino group—indicates change in the structure and the formation of the complexes. **Figure 3.** Raman-spectra of the empty 5,10,15,20-tetrakis(4-aminophenyl)porphyrin ring (TAPP) and its Cu- and Fe-complexes. In the case of Cu-complex, a different intensity scale was used for easier comparability of the spectra. The differences between the positions of the circled peaks—which can be assigned to the aromatic hydrocarbons, in which a hydrogen has been replaced by amino group—indicates change in the structure and the formation of the complexes. **Figure 3.** Raman-spectra of the empty 5,10,15,20-tetrakis(4-aminophenyl)porphyrin ring (TAPP) and its Cu- and Fe-complexes. In the case of Cu-complex, a different intensity scale was used for easier comparability of the spectra. The differences between the positions of the circled peaks which can be assigned to the aromatic hydrocarbons, in which a hydrogen has been replaced by amino group—indicates change in the structure and the formation of the complexes.

the porphyrin complex, the spectrum of FePA is less informative than that of CuPA. The spectral differences and the direction of the intensity changes of the CuPA and FePA aerogels compared to the complexes (shown in Figure 3) clearly indicate the formation of covalent bonds between the functionalized porphyrin complexes and the silica aerogel matrix. (\*) CuPA, (\*\*) FePA. **Figure 4.** Raman-spectra of the aerogel materials CuPA and FePA. Due to the low concentration of the porphyrin complex, the spectrum of FePA is less informative than that of CuPA. The spectral differences and the direction of the intensity changes of the CuPA and FePA aerogels compared to the complexes (shown in Figure 3) clearly indicate the formation of covalent bonds between the functionalized porphyrin complexes and the silica aerogel matrix. (\*) CuPA, (\*\*) FePA. **Figure 4.** Raman-spectra of the aerogel materials CuPA and FePA. Due to the low concentration of the porphyrin complex, the spectrum of FePA is less informative than that of CuPA. The spectral differences and the direction of the intensity changes of the CuPA and FePA aerogels compared to the complexes (shown in Figure 3) clearly indicate the formation of covalent bonds between the functionalized porphyrin complexes and the silica aerogel matrix. (\*) CuPA, (\*\*) FePA.

**Table 1.** Elemental analysis of the hybrid aerogel samples. The increased carbon and nitrogen content indicate the presence of the porphyrin complexes in the silica matrix. The non-zero values of the blank silica sample are an indication of the presence of adsorbed organic residues and the catalyst ammonia after the supercritical CO<sup>2</sup> drying process. **Table 1.** Elemental analysis of the hybrid aerogel samples. The increased carbon and nitrogen content indicate the presence of the porphyrin complexes in the silica matrix. The non-zero values of the blank silica sample are an indication of the presence of adsorbed organic residues and the catalyst ammonia after the supercritical CO2 drying process.


**Figure 5.** Scanning electron micrographs of the aerogel catalysts (**a**) CuPA, and (**b**) FePA. Insert: Pore size distribution curves calculated by the BJH method for CuPA (red line) and FePA (blue line). Both of the curves show the characteristic pore diameter of around 12 nm, which is typical for the silica aerogels. **Figure 5.** Scanning electron micrographs of the aerogel catalysts (**a**) CuPA, and (**b**) FePA. Insert: Pore size distribution curves calculated by the BJH method for CuPA (red line) and FePA (blue line). Both of the curves show the characteristic pore diameter of around 12 nm, which is typical for the silica aerogels.

**Table 2.** Summary of the nitrogen adsorption/desorption porosimetry results. Specific surface area (SBET) was calculated by the multi-point BET method. Characteristic pore diameters (d), and total pore volumes (Vtotal) were calculated from the isotherms by the BJH method. Mesopore and macropore volumes (Vmeso, Vmacro) were calculated from the cumulative pore volume curves for the regions 2–50 nm and above 50 nm, respectively. Micropore contribution (Vmicr) was calculated by the t-plot method.  **CuPA FePA Table 2.** Summary of the nitrogen adsorption/desorption porosimetry results. Specific surface area (SBET) was calculated by the multi-point BET method. Characteristic pore diameters (d), and total pore volumes (Vtotal) were calculated from the isotherms by the BJH method. Mesopore and macropore volumes (Vmeso, Vmacro) were calculated from the cumulative pore volume curves for the regions 2–50 nm and above 50 nm, respectively. Micropore contribution (Vmicr) was calculated by the t-plot method.


**Table 3.** Metal ion content of the catalysts measured by the ICP-OES technique. **Sample Theoretical (***w***/***w* **%) Measured (***w***/***w* **%) Table 3.** Metal ion content of the catalysts measured by the ICP-OES technique.


#### The catalytic activity of the samples was evaluated through the oxidation of phenol *2.3. Catalytic Activity*

and chlorophenols with hydrogen peroxide in aqueous solutions at different temperatures. The pH of the reaction mixtures was left to change spontaneously in the course of the reactions in order to simulate the behaviour of real-life wastewaters. The main oxidation products of phenol were identified by HPLC measurements; further products and the The catalytic activity of the samples was evaluated through the oxidation of phenol and chlorophenols with hydrogen peroxide in aqueous solutions at different temperatures. The pH of the reaction mixtures was left to change spontaneously in the course of the reactions in order to simulate the behaviour of real-life wastewaters. The main oxidation

products of phenol were identified by HPLC measurements; further products and the proposed degradation pathways are published in a previous work [26]. In the case of chlorinated phenols, the oxidation products were undetectable by UV/HPLC, and mass spectrometry was used instead.

We made several attempts to determine the catalytic activities of the complexes in the homogeneous phase. Unfortunately, all of them failed, due to the insolubility of the porphyrin complexes in water, water–DMF and water–DMSO mixtures. The reaction mixture seemed to be homogeneous at the applied 90 ◦C, nevertheless when it started to cool down, solid particles appeared and the solution turned colourless (Figure S5). The reaction could have been tested in a homogeneous phase using DMF as the solvent, in which only the calculated volume of 30% (m/m) aqueous hydrogen peroxide was dissolved. However, none of the complexes showed any catalytic activity against the phenols. The reason for this can be either that a fast self-oxidation destroyed all the porphyrin complexes, or that the oxidation of the solvent DMF, which was present in large excess, consumed the oxidant.

#### 2.3.1. Phenol Oxidation

Phenol oxidation and conversion was monitored by a reversed-phase HPLC technique, the concentrations were determined by using a five-point calibration curve and UV detection. Several catalyst-to-substrate, and catalyst-to-hydrogen peroxide molar ratios were tested in batch experiments. Conversion curves of phenol are shown in Figure 6. The main feature is that 80% of the phenol was converted within three hours even when the catalyst was applied only in a 0.33 mol%. Obviously, the free copper(II) ions in the homogeneous phase were much more effective. Nevertheless, in a homogeneous phase the separation of the catalyst Cu2+ would be difficult and it is not favorable in industrial use. As expected, the free iron(2+) ions were more effective catalysts in the homogeneous phase than the complex in the heterogeneous phase. However, when the catalyst was applied in 1 mol% quantity, more than 90% of the phenol was eliminated within one hour (Figure 6b). The FePA catalyst applied in a smaller amount did not show as high catalytic activity as the catalyst CuPA did. Both catalysts showed slightly S-shaped conversion curves, which may be the indication of autocatalytic reactions or the consequence of hindered diffusion and materials transport in the pores.

Depending on the applied metal complexes, a different intermediate profile was obtained (Figure 7). In the case of the copper complex, two main intermediates were detected: catechol and hydrochinon. Only catechol was detected as an intermediate when the iron complex was applied, supposedly due to the different reaction mechanisms. It was confirmed by mass spectrometry that the intermediates continued to transform into further products, and finally into short-chained carboxylic acids [45].

In order to compare the efficiency of the catalysts more expressively, the turnover frequencies (TOFs) were calculated. We selected a set of points from the initial linear section of the kinetic curves and applied linear regression (Figure 8). The TOF values and the regression coefficients of the fittings are shown in Table 4. As it can be seen, the rate of the phenol degradation is higher in the case of catalyst CuPA but both of the TOF values are comparable to that of the industrial catalysts, since for the most relevant industrial applications the TOF values are in the range of 10−2–10<sup>2</sup> s −1 [46].

Unfortunately, the catalysts lost their activity after the first cycle, most likely because the porphyrin ring suffered self-oxidation. Due to the static conditions applied during the reactions, the high excess of hydrogen peroxide could cause the oxidation of the porphyrin ring.

The reaction was carried out with lower H2O<sup>2</sup> excess, the applied molar ratios were: catalyst:phenol:H2O<sup>2</sup> = 1:100:1400. The kinetic curve can be seen in Figure 6. The conversion was higher than 90% after one hour in this case as well, but the colour of the catalyst changed this time too, which indicated the degradation of it. Therefore, we tried to optimize the temperature too. The reaction was carried out at different temperatures

besides 90 ◦C: 30–70 ◦C using the catalyst:phenol:H2O<sup>2</sup> = 1:100:1400 molar ratio. The final phenol conversion as a function of the temperature can be seen in Figure 9. We found that the highest conversion was achieved at 90 ◦C. There is a breakpoint between 50 ◦C and 60 ◦C, the conversion was 50% compared to 12% at 50 ◦C. At 60 ◦C the degradation of the catalyst was minimal, although the rate of the reaction was much lower than at 90 ◦C, the calculated turnover frequency was 4.06 <sup>×</sup> <sup>10</sup>−<sup>4</sup> s −1 . conversion as a function of the temperature can be seen in Figure 9. We found that the highest conversion was achieved at 90 °C. There is a breakpoint between 50 °C and 60 °C, the conversion was 50 % compared to 12 % at 50 °C. At 60 °C the degradation of the catalyst was minimal, although the rate of the reaction was much lower than at 90 °C, the calculated turnover frequency was 4.06 × 10−4 s−1. conversion as a function of the temperature can be seen in Figure 9. We found that the highest conversion was achieved at 90 °C. There is a breakpoint between 50 °C and 60 °C, the conversion was 50 % compared to 12 % at 50 °C. At 60 °C the degradation of the catalyst was minimal, although the rate of the reaction was much lower than at 90 °C, the calculated turnover frequency was 4.06 × 10−4 s−1. *Gels* **2022**, *8*, x FOR PEER REVIEW 10 of 20 conversion as a function of the temperature can be seen in Figure 9. We found that the highest conversion was achieved at 90 °C. There is a breakpoint between 50 °C and 60 °C,

*Gels* **2022**, *8*, x FOR PEER REVIEW 10 of 20

**Figure 6.** Kinetic curves of phenol conversion over time using the CuPA catalyst (**a**) and FePA catalyst (**b**), and the free metal ions using different catalyst-to-phenol (1:100, 1:200, 1:300), and phenolto-H2O2 molar ratios (1:100 and 1:14). **Figure 6.** Kinetic curves of phenol conversion over time using the CuPA catalyst (**a**) and FePA catalyst (**b**), and the free metal ions using different catalyst-to-phenol (1:100, 1:200, 1:300), and phenol-to-H2O<sup>2</sup> molar ratios (1:100 and 1:14). **Figure 6.** Kinetic curves of phenol conversion over time using the CuPA catalyst (**a**) and FePA catalyst (**b**), and the free metal ions using different catalyst-to-phenol (1:100, 1:200, 1:300), and phenolto-H2O2 molar ratios (1:100 and 1:14).

**Figure 6.** Kinetic curves of phenol conversion over time using the CuPA catalyst (**a**) and FePA cata-

**Figure 7.** Intermediates profile of phenol oxidation by CuPA (**a**) and FePA (**b**) catalysts. In the case of the copper complex, catechol and hydrochinon were detected, and only catechol was observed when iron ions were applied. **Figure 7.** Intermediates profile of phenol oxidation by CuPA (**a**) and FePA (**b**) catalysts. In the case of the copper complex, catechol and hydrochinon were detected, and only catechol was observed when iron ions were applied. **Figure 7.** Intermediates profile of phenol oxidation by CuPA (**a**) and FePA (**b**) catalysts. In the case of the copper complex, catechol and hydrochinon were detected, and only catechol was observed when iron ions were applied.

**Figure 8.** Linear trend lines fitted to points selected from the initial linear phase of the kinetic curves of phenol oxidation using CuPA (●) and FePA (■) catalysts. **Figure 8.** Linear trend lines fitted to points selected from the initial linear phase of the kinetic curves of phenol oxidation using CuPA (●) and FePA (■) catalysts. **Figure 8.** Linear trend lines fitted to points selected from the initial linear phase of the kinetic curves of phenol oxidation using CuPA (•) and FePA () catalysts.


**Table 4.** Turnover frequencies (TOFs) of the immobilized complexes applied for oxidation of phenol. **Table 4.** Turnover frequencies (TOFs) of the immobilized complexes applied for oxidation of phe-

**Figure 9.** Phenol conversion after 180 min as a function of the reaction temperature. The highest conversion was achieved at 90 °C. Nevertheless, at 60 °C the degradation of the catalyst was minimal, so this value is found to be the most advantageous to carry out the reaction. **Figure 9.** Phenol conversion after 180 min as a function of the reaction temperature. The highest conversion was achieved at 90 ◦C. Nevertheless, at 60 ◦C the degradation of the catalyst was minimal, so this value is found to be the most advantageous to carry out the reaction.

The quantity of the free metal ions in the reaction mixture after the decomposition of phenol was determined by an ICP-OES method. After three hours reaction time, 36 % of the total copper ion, and 89 % of the total iron ion was free. In a control experiment, when only the catalysts were suspended in distilled water and heated for three hours, only 3.7 % of the total copper-ion was measured in the supernatant. This was in good agreement with the increasing solubility of the amorphous silica aerogel in water at elevated temperatures. The concentration of the free iron ions in the control experiment was under the limit of detection (LOD). Both experiments proved that the presence of the oxidizing agent hydrogen peroxide was the prerequisite for the degradation of the porphyrin rings, which led then to the release of metal ions. That is the reason why all of the regeneration attempts failed. The degradation of the porphyrin ring was discernible through the colour change of the catalysts. It gradually turned from red to brown in the case of CuPA, and the brown colour of the FePA catalyst slowly disappeared during the reaction. The quantity of the free metal ions in the reaction mixture after the decomposition of phenol was determined by an ICP-OES method. After three hours reaction time, 36% of the total copper ion, and 89% of the total iron ion was free. In a control experiment, when only the catalysts were suspended in distilled water and heated for three hours, only 3.7% of the total copper-ion was measured in the supernatant. This was in good agreement with the increasing solubility of the amorphous silica aerogel in water at elevated temperatures. The concentration of the free iron ions in the control experiment was under the limit of detection (LOD). Both experiments proved that the presence of the oxidizing agent hydrogen peroxide was the prerequisite for the degradation of the porphyrin rings, which led then to the release of metal ions. That is the reason why all of the regeneration attempts failed. The degradation of the porphyrin ring was discernible through the colour change of the catalysts. It gradually turned from red to brown in the case of CuPA, and the brown colour of the FePA catalyst slowly disappeared during the reaction.

Beyond the batch experiments, a custom made continuous-flow reactor (see in Supplementary Materials, Figure S6) was tested as well. Despite of the short contact time (9 min) we observed good catalytic activity, although the efficiency gradually decreased as the reaction proceeded (Figure 10). However, it might be approaching a constant value in longer times when the neighboring catalytic centers became so distant that they could not oxidize each other. The results proved that our aerogel-based materials can be used as catalyst beds in continuous flow processes, which is more advantageous and manageable for the industry. Beyond the batch experiments, a custom made continuous-flow reactor (see in Supplementary Materials, Figure S6) was tested as well. Despite of the short contact time (9 min) we observed good catalytic activity, although the efficiency gradually decreased as the reaction proceeded (Figure 10). However, it might be approaching a constant value in longer times when the neighboring catalytic centers became so distant that they could not oxidize each other. The results proved that our aerogel-based materials can be used as catalyst beds in continuous flow processes, which is more advantageous and manageable for the industry.

**Figure 10.** Conversion of phenol in a short-bed continuous-flow tubular reactor determined for 30 min time segments. (temperature 90 °C). **Figure 10.** Conversion of phenol in a short-bed continuous-flow tubular reactor determined for 30 min time segments. (temperature 90 ◦C).

#### 2.3.2. Oxidation of 3-Chloro- and 2,4-Dichlorophenol 2.3.2. Oxidation of 3-Chloro- and 2,4-Dichlorophenol

The kinetic curves of the conversion of 3-chlorophenol (3-CP) can be seen in Figure 11. In contrast to the phenol oxidation, the catalytic activity of the CuPA catalyst was smaller than that of the FePA catalyst. Furthermore, the immobilized complex showed higher activity than the free iron ions even when it was applied in Fe:3-CP=1:200 molar ratio. The decreased activity of the free iron ions is most likely due to the hydrolysis of Fe3+ ions forming catalytically inactive hydroxo-iron precipitates under the solution pH. The "S" shape of the curve in that case may also indicate either autocatalytic or diffusion controlled processes. The oxidation products were identified by high-resolution mass spectrometry (HRMS), both in the positive and in the negative ion mode. Based on the results, we suggested a reaction pathway, which is given in Scheme 2. The aromatic ring was hydroxylated first, then dechlorinated, split open and fragmented into short-chained carboxylic acids. In each case, fragmentation occurred through the loss of carbon dioxide. In the case of 2,4-dichlorophenol (2,4-DCP) (Figure 12) a significant decrease in the catalytic activity of the CuPA catalyst was observed compared to the 3-chlorophenol. Although the activity of the FePA decreased as well, it was almost as effective as in the case of the monochlorophenol. The free iron ions showed poor efficiency due to their hydrolysis under the reaction conditions, as mentioned above. Considering the complexity of the reaction, as well as the lack of appropriate analytical standards, the details of the mecha-The kinetic curves of the conversion of 3-chlorophenol (3-CP) can be seen in Figure 11. In contrast to the phenol oxidation, the catalytic activity of the CuPA catalyst was smaller than that of the FePA catalyst. Furthermore, the immobilized complex showed higher activity than the free iron ions even when it was applied in Fe:3-CP = 1:200 molar ratio. The decreased activity of the free iron ions is most likely due to the hydrolysis of Fe3+ ions forming catalytically inactive hydroxo-iron precipitates under the solution pH. The "S" shape of the curve in that case may also indicate either autocatalytic or diffusion controlled processes. The oxidation products were identified by high-resolution mass spectrometry (HRMS), both in the positive and in the negative ion mode. Based on the results, we suggested a reaction pathway, which is given in Scheme 2. The aromatic ring was hydroxylated first, then dechlorinated, split open and fragmented into short-chained carboxylic acids. In each case, fragmentation occurred through the loss of carbon dioxide. In the case of 2,4-dichlorophenol (2,4-DCP) (Figure 12) a significant decrease in the catalytic activity of the CuPA catalyst was observed compared to the 3-chlorophenol. Although the activity of the FePA decreased as well, it was almost as effective as in the case of the monochlorophenol. The free iron ions showed poor efficiency due to their hydrolysis under the reaction conditions, as mentioned above. Considering the complexity of the reaction, as well as the lack of appropriate analytical standards, the details of the mechanism has not been explored.

nism has not been explored. The turnover frequencies were also calculated to compare the efficiency of the catalysts (Figure 13), the results are summarized in Table 5. According to the results, the efficiency of the CuPA catalyst did not reach that of the catalysts applied in the industry if we used it for oxidation of a dichloro derivative of the phenol. In contrast to that, the TOF values of the FePA catalyst are still comparable with the industrial catalysts' TOF values.

The chlorophenols (mono- and dichlorinated) were identified by capillary electrophoresis coupled to a mass spectrometer (CE-MS). The advantage of using this technique over HPLC is the higher sensitivity of the MS detector, which enables detection of chlorophenols in a concentration range more relevant to environmental regulations [47]. A simple method has developed for the CE separation of the chlorophenols, which also made it possible to set the limit of quantitation (LOQ) below 1 ppm, which is the regulatory limit for phenol in wastewaters. Chlorophenols are ionized more readily in the negative ion mode, therefore, the negative ion mode was applied along with separation in basic background electrolyte (40 mM ammonium formate/ammonia, pH = 9.5).

ions. The 3-chlorophenol-to-hydrogen peroxide ratio was at a constant value of 1:100.

**Figure 11.** Kinetic curves of 3-chlorophenol conversion over time using the CuPA catalyst (**a**) and FePA catalyst (**b**) in different catalyst-to-phenol molar ratios (1:100 and 1:200), as well as free metal ions. The 3-chlorophenol-to-hydrogen peroxide ratio was at a constant value of 1:100. **Figure 12.** Kinetic curves of 2,4-dichlorophenol (2,4-DCP) conversion over time using the CuPA catalyst (**a**) and FePA catalyst (**b**), as well as free metal ions. The 2,4-dichlorophenol-to-hydrogen peroxide ratio was kept at a constant value of 1:100.

**Figure 11.** Kinetic curves of 3-chlorophenol conversion over time using the CuPA catalyst (**a**) and FePA catalyst (**b**) in different catalyst-to-phenol molar ratios (1:100 and 1:200), as well as free metal

**Figure 10.** Conversion of phenol in a short-bed continuous-flow tubular reactor determined for 30

The kinetic curves of the conversion of 3-chlorophenol (3-CP) can be seen in Figure 11. In contrast to the phenol oxidation, the catalytic activity of the CuPA catalyst was smaller than that of the FePA catalyst. Furthermore, the immobilized complex showed higher activity than the free iron ions even when it was applied in Fe:3-CP=1:200 molar ratio. The decreased activity of the free iron ions is most likely due to the hydrolysis of Fe3+ ions forming catalytically inactive hydroxo-iron precipitates under the solution pH. The "S" shape of the curve in that case may also indicate either autocatalytic or diffusion controlled processes. The oxidation products were identified by high-resolution mass spectrometry (HRMS), both in the positive and in the negative ion mode. Based on the results, we suggested a reaction pathway, which is given in Scheme 2. The aromatic ring was hydroxylated first, then dechlorinated, split open and fragmented into short-chained carboxylic acids. In each case, fragmentation occurred through the loss of carbon dioxide. In the case of 2,4-dichlorophenol (2,4-DCP) (Figure 12) a significant decrease in the catalytic activity of the CuPA catalyst was observed compared to the 3-chlorophenol. Although the activity of the FePA decreased as well, it was almost as effective as in the case of the monochlorophenol. The free iron ions showed poor efficiency due to their hydrolysis under the reaction conditions, as mentioned above. Considering the complexity of the reaction, as well as the lack of appropriate analytical standards, the details of the mecha-

min time segments. (temperature 90 °C).

nism has not been explored.

2.3.2. Oxidation of 3-Chloro- and 2,4-Dichlorophenol

*Gels* **2022**, *8*, x FOR PEER REVIEW 13 of 20

**Scheme 2.** Proposed reaction pathway, formal stoichiometry, and reaction mechanisms of Fentonoxidation of 3-chlorophenol. Through catalytic hydroxylation and dechlorination, chlorophenol was transformed into non-toxic short-chained carboxylic acids. All of the intermediates shown were identified by high-resolution mass spectrometry (HRMS). **Scheme 2.** Proposed reaction pathway, formal stoichiometry, and reaction mechanisms of Fentonoxidation of 3-chlorophenol. Through catalytic hydroxylation and dechlorination, chlorophenol was transformed into non-toxic short-chained carboxylic acids. All of the intermediates shown were identified by high-resolution mass spectrometry (HRMS).

The turnover frequencies were also calculated to compare the efficiency of the catalysts (Figure 13), the results are summarized in Table 5. According to the results, the efficiency of the CuPA catalyst did not reach that of the catalysts applied in the industry if

ions. The 3-chlorophenol-to-hydrogen peroxide ratio was at a constant value of 1:100.

**Figure 11.** Kinetic curves of 3-chlorophenol conversion over time using the CuPA catalyst (**a**) and FePA catalyst (**b**) in different catalyst-to-phenol molar ratios (1:100 and 1:200), as well as free metal

**Figure 12.** Kinetic curves of 2,4-dichlorophenol (2,4-DCP) conversion over time using the CuPA catalyst (**a**) and FePA catalyst (**b**), as well as free metal ions. The 2,4-dichlorophenol-to-hydrogen peroxide ratio was kept at a constant value of 1:100. **Figure 12.** Kinetic curves of 2,4-dichlorophenol (2,4-DCP) conversion over time using the CuPA catalyst (**a**) and FePA catalyst (**b**), as well as free metal ions. The 2,4-dichlorophenol-to-hydrogen peroxide ratio was kept at a constant value of 1:100. we used it for oxidation of a dichloro derivative of the phenol. In contrast to that, the TOF values of the FePA catalyst are still comparable with the industrial catalysts' TOF values.

**Figure 13.** Linear trend lines fitted to points selected from the initial linear phase of the kinetic curves of 3-chlorophenol (**a**) and 2,4-dichlorophenol (**b**) oxidation using CuPA (●) and FePA (■) catalysts. **Figure 13.** Linear trend lines fitted to points selected from the initial linear phase of the kinetic curves of 3-chlorophenol (**a**) and 2,4-dichlorophenol (**b**) oxidation using CuPA (•) and FePA () catalysts.


**Scheme 2.** Proposed reaction pathway, formal stoichiometry, and reaction mechanisms of Fenton-

The chlorophenols (mono- and dichlorinated) were identified by capillary electro-

**Table 5.** Turnover frequencies (TOFs) of the immobilized complexes applied for oxidation of 3-chlorophenol and 2,4-dichlorophenol. **Table 5.** Turnover frequencies (TOFs) of the immobilized complexes applied for oxidation of 3-chlorophenol and 2,4-dichlorophenol.

#### oxidation of 3-chlorophenol. Through catalytic hydroxylation and dechlorination, chlorophenol was phoresis coupled to a mass spectrometer (CE-MS). The advantage of using this technique **3. Conclusions**

modes.

transformed into non-toxic short-chained carboxylic acids. All of the intermediates shown were identified by high-resolution mass spectrometry (HRMS). The turnover frequencies were also calculated to compare the efficiency of the catalysts (Figure 13), the results are summarized in Table 5. According to the results, the efficiency of the CuPA catalyst did not reach that of the catalysts applied in the industry if over HPLC is the higher sensitivity of the MS detector, which enables detection of chlorophenols in a concentration range more relevant to environmental regulations [47]. A simple method has developed for the CE separation of the chlorophenols, which also made it possible to set the limit of quantitation (LOQ) below 1 ppm, which is the regulatory limit for phenol in wastewaters. Chlorophenols are ionized more readily in the negative ion mode, therefore, the negative ion mode was applied along with separation in basic back-Copper and iron complexes of 4-aminophenylporphyrin have been immobilized successfully in silica aerogels using isocyanopropyl triehoxysilane as a bifunctional linker reagent. By functionalizing the porphyrin rings, followed by complexation with selected metal ions, we were able to bind the complexes to the silica aerogel matrix with strong covalent bonds. The as-obtained catalysts have a large specific surface area and an open mesoporous structure, which are important features of the heterogeneous catalysts.

ground electrolyte (40 mM ammonium formate/ammonia, pH=9.5). **3. Conclusions**  Copper and iron complexes of 4-aminophenylporphyrin have been immobilized suc-The catalytic activity of the samples was tested through the oxidation of phenol, 3-chlorophenol and 2,4-dichlorophenol by hydrogen peroxide. The oxidation products were identified by high-pressure liquid chromatography in the case of phenol and by high-resolution mass spectrometry in the case of 3-chlorophenol. All intermediates of the

cessfully in silica aerogels using isocyanopropyl triehoxysilane as a bifunctional linker reagent. By functionalizing the porphyrin rings, followed by complexation with selected

covalent bonds. The as-obtained catalysts have a large specific surface area and an open mesoporous structure, which are important features of the heterogeneous catalysts.

The catalytic activity of the samples was tested through the oxidation of phenol, 3 chlorophenol and 2,4-dichlorophenol by hydrogen peroxide. The oxidation products were identified by high-pressure liquid chromatography in the case of phenol and by high-resolution mass spectrometry in the case of 3-chlorophenol. All intermediates of the degradation process were identified and a reaction scheme was proposed to describe the entire process. The FePA and CuPA catalysts showed different selectivity towards the substrates. In the case of phenol, the copper complex proved to be more efficient, while for the chlorinated derivatives the iron complex showed significantly higher activity. We have demonstrated that they can be used both in the batch and the continuous-flow degradation process were identified and a reaction scheme was proposed to describe the entire process. The FePA and CuPA catalysts showed different selectivity towards the substrates. In the case of phenol, the copper complex proved to be more efficient, while for the chlorinated derivatives the iron complex showed significantly higher activity. We have demonstrated that they can be used both in the batch and the continuous-flow modes.

Both the CuPA and FePA catalyst oxidized the dangerous phenolic pollutants in water by the safe agent hydrogen peroxide. Phenol, 3-chlorophenol and 2,4-dichlorophenol were converted to non-toxic short-chained carboxylic acids, and then finally mineralized into carbon dioxide, water and hydrochloric acid in the process. A special advantage of the process is that it can be used directly with contaminated natural waters, as the process does not require any buffering or use of additives.

Our study clearly showed that the covalent incorporation of the ab ovo water-insoluble porphyrin complexes in a hydrophilic silica aerogel matrix made them compatible with the aqueous medium and allowed their active use in such an environment. By immobilization, the extent and the rate of the catalysts' self-oxidation was reduced, although it was still present at a lower level. Since the direct interaction of the catalytically active centers of the complexes was not possible due to their immobilization, the degradation was most likely the consequence of the attack of active hydroxyl radicals generated in the catalytic cycles. For future application, the chemical stability of the complexes should be improved for example by changing the nature of the connecting pendant arms [48]. In order to minimize the degradation of the catalysts but keep the reaction fast enough, the temperature was optimized to 60 ◦C and the phenol:hydrogen peroxide molar ratio was cut back to 1:14, which is close to the theoretical limit of 1:12–14 molar ratio required for complete mineralization of chlorinated phenols. However, under such conditions, the turnover frequency dropped from the 10−1–10−<sup>2</sup> s −1 range to as low as 4.1 <sup>×</sup> <sup>10</sup>−<sup>4</sup> s −1 . Our results show that the as-obtained catalysts CuPA and FePA may be considered as potential alternatives for the mineralization of phenols with the environmentally safe oxidizing agent hydrogen peroxide.

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

#### *4.1. Materials*

Methanol (technical grade), acetone (technical grade), 30% hydrogen peroxide solution (analytical reagent grade), N,N-dimethylformamide (DMF) (analytical reagent grade) and 25 m/m% ammonia solution (analytical reagent grade) were purchased from Molar Chemicals Kft. (Hungary). Tetramethyl orthosilicate (TMOS) (purum), 3-chlorophenol (98 %) and 2,4-dichlorophenol (99 %) were obtained from Sigma-Aldrich Ltd. (St. Louis, MO, USA). 5,10,15,20-Tetrakis(4-aminophenyl)porphyrin (TAPP) and 3-isocyanatopropyltriethoxysilane (3-IPTES) were purchased from ABCR GmbH (Germany). Phenol (analytical reagent grade), copper acetate (reagent grade) and iron(II) sulfate (reagent grade) were acquired from Reanal Finomvegyszergyár Zrt. (Hungary). Carbon dioxide cylinder was purchased from Linde Gáz Magyarország Zrt. (Debrecen, Hungary). All the reagents were used without any further purification.

#### *4.2. Synthesis of Porphyrin Complexes*

The synthesis of the complexes was carried out in two consecutive steps. First, the porphyrin ring was functionalized with 3-isocyanatopropyltriethoxysilane (3-IPTES) in a two neck flask under anhydrous conditions. A total of 20 mg of 5,10,15,20-tetrakis(4 aminophenyl)porphyrin (2.97 <sup>×</sup> <sup>10</sup>−<sup>2</sup> mmol) was dissolved in 4.0 mL anhydrous DMF and kept under a dry argon atmosphere. 0.33 mL (1.191 mmol) of 3-IPTES was added to the solution also under an argon atmosphere. The mixture was stirred at 70 ◦C for five days. Next, the complexes were obtained by mixing 1.0 mL of 3-IPTES-TAPP with either copper or iron salts. The reaction mixtures were diluted to 4.0 mL final volume with DMF and stirred at 116 ◦C in sealed vessels. The metal ions were dosed in small portions until the red fluorescence of the free base 3-IPTES-TAPP disappeared, indicating the complete formation

of the complexes. The reaction time depended on the metal-ions. The Cu2+-TAPP-3-IPTES complex formed typically in 10 min, while the formation of the complex with iron ion took up to 72 h. The progress of the reactions was monitored by TLC.

#### *4.3. Synthesis of Heterogeneous Catalysts*

The catalysts (denoted as CuPA and FePA) were obtained by covalently binding the complexes to the silica precursors and then co-gelated with TMOS to develop the hybrid aerogel matrix. Two monoliths were prepared via the following general recipe. Two solutions (labelled "A" and "B") were prepared. Solution "A" contained 12.0 mL MeOH (297 mmol) and 3.0 mL TMOS (20.33 mmol). Solution "B" was made from 12.0 mL MeOH (297 mmol), 2.00 mL of the complex solution (3.705 <sup>×</sup> <sup>10</sup>−<sup>3</sup> mmol), 1.0 mL distilled water (5.54 mmol) and 1.00 mL of diluted (1:1) NH<sup>3</sup> solution (7.34 mmol). Solutions "A" and "B" were mixed together, then poured into cylindrical plastic molds (66 × 28 mm) and sealed with parafilm. The gels were kept in the molds overnight then they were transferred into perforated aluminum frames. The frames provided mechanical support and allowed for quick and efficient exchange of the solvents before supercritical drying. All gels were soaked in the following solvents in order to purify them and to remove the water: methanol-cc NH<sup>3</sup> (8:1), pure methanol; each for a day. Then methanol was gradually replaced by acetone, changed in 25% steps. There was no indication of leaching of the porphyrin complexes during the process. Finally, the gels were stored in a copious volume (2 L) of freshly distilled dry acetone for three days. After the change of the solvents, the aerogels were obtained by supercritical drying, which was carried out in a custom-made high-pressure reactor according to a general procedure published in a previous work [39].

#### *4.4. Characterization of the Catalysts*

Nitrogen gas porosimetry measurements were performed on a Quantachrome Nova 2200e surface area and porosity analyzer (Quantachrome Instruments, Boynton Beach, FL, USA). Pieces of the samples were ground in a mortar and outgassed under vacuum at 100 ◦C for 24 h before the measurements.

1H-NMR measurements were performed on a Bruker 360 spectrometer (Bruker Billerica, MA, USA).

FT-IR spectra were recorded on a Jasco FT/IR-4100 instrument (Easton, MD, USA).

Raman measurements were performed on a Renishaw InVia Raman microscope (Renishaw, Wotton-under-Edge, United Kingdom). It was used for the characterization of the aerogel samples in the range 100–5000 Raman shift/cm−<sup>1</sup> . The laser used for the measurements was a 532 nm, 50 mW diode laser. All spectra were recorded at a 10 s exposure time each, utilizing 2400 L/mm grating. Beam centering and Raman spectra calibration were performed daily before spectral acquisition using the inbuilt Si standard.

Elemental analysis was performed in a varioMICRO element analyzer (Elementar Analysensysteme GmbH, Hanau, Germany).

Scanning electron micrographs (SEM) were recorded on a Hitachi S-4300 instrument (Hitachi Ltd., Tokyo, Japan) equipped with a Bruker energy dispersive X-ray spectroscope (Bruker Corporation, Billerica, MA, USA). The surfaces were covered by a sputtered gold conductive layer and a 5–15 kV accelerating voltage was used for taking high resolution pictures.

The analysis of the metal content of the aerogels was carried out by an Agilent ICP-OES 5100 SVDV device (Agilent Technologies, Santa Clara, CA, USA) after nitric acid and hydrogen peroxide microwave digestion.

#### *4.5. Study of Catalytic Activity*

The catalytic activity of the aerogels was tested through the decomposition of phenol and chlorophenols in an aqueous solution.

In the case of phenol, typically the phenol solution was placed into a flask along with the appropriate amount of catalyst; the molar ratios were catalyst to phenol 1:100, 1:200 or 1:300. The required amounts of the aerogels were calculated according to the ICP results. The mixture was thermally equilibrated for several minutes before hydrogen peroxide (molar ratios phenol to hydrogen peroxide 1:100 or 1:14) was added to start the reaction. The total volume was 15.0 mL. The initial concentration of phenol was 250 ppm. The reaction mixtures were stirred in closed vessels between 30–70 ◦C and at 90 ◦C for three hours. A total of nine samples were taken (1.0 mL each) and centrifuged. The supernatants were analyzed with HPLC, applying the following parameters: the column was a Phenomenex Phenyl Hexyl column (150 × 4.6 mm, particle size: 5 µm); the composition of the mobile phase was 50% H2O, 50% MeOH; the flow rate was 1.0 mL/min and the analysis time was 5 min. The components were detected by a UV detector at 270 nm.

Beyond the batch experiments, the oxidation of phenol was carried out in a continuousflow reactor as well. The applied catalyst to phenol and phenol to hydrogen peroxide molar ratios were both 1:100. The phenol concentration was 500 ppm. The reaction mixture including the phenol solution, water and hydrogen peroxide solution—was filled in a syringe and was dosed by a syringe pump with 2 mL/h rate. The catalyst was filled in a U-shaped glass tube of 0.3 cm inner diameter. The bed length was 4.3 cm, the contact time was set to 9 min. The reactor was immersed in a 90 ◦C water bath.

In the case of the chlorophenols, the chlorophenol solution was placed into a flask along with the appropriate amount of catalyst, the molar ratios were catalyst to chlorophenol 1:100. The mixture was thermally equilibrated at 70 ◦C for several minutes before hydrogen peroxide (molar ratio: chlorophenol: hydrogen peroxide = 1:100) was added to start the reaction. The total volume was 15.0 mL. The initial concentration of the chlorophenols was 500 ppm. The reaction mixtures were stirred in closed vessels at 70 ◦C for three hours. In total, eight samples were taken (1.0 mL each) and centrifuged. The supernatants were analyzed with HPLC, applying the following parameters. The column was a Supelcosil LC-18 (250 mm × 4.6 mm, particle size: 5 µm) column; the composition of the mobile phase was mixture of 50% ammonium acetate (50 mM)—50% MeOH in the case of 3 chlorophenol; and ammonium acetate:methanol = 20:80 in the case of 2,4-DCP; the flow rate was 1.0 mL/min, the analysis time was 8 min in both cases. 3-CP and 2,4-DCP was detected by a UV detector at 280 nm and 287 nm, respectively.

Oxidation products of chlorophenols were detected by high resolution mass spectrometry (maXis II UHR ESI-QTOF MS instrument, Bruker, Karlsruhe, Germany), both in the positive and in the negative ion mode. For positive mode ESI, the following parameters were used: capillary voltage: 3.5 kV, nebulizer pressure: 0.5 bar, dry gas flow rate: 4.5 L/min, temperature: 200 ◦C. For negative mode ESI, the following parameters were used: capillary voltage: 2.5 kV, nebulizer pressure: 0.5 bar, dry gas flow rate: 4 L/min, temperature: 200 ◦C. MS tuning parameters were optimized in both cases to measure the relevant m/z range for chlorophenols (and their oxidation products), which was 50–600 m/z, to generally detect any possible products.

A more sensitive (CE-)MS method was developed to separate and detect chlorophenols in low concentrations (~1 ppm). The abovementioned MS instrument was coupled to a capillary electrophoresis (7100 CE System, Agilent, Waldbronn, Germany) instrument via a coaxial CE-ESI sprayer interface (G1607B, Agilent). Sheath liquid was transferred with a 1260 Infinity II isocratic pump (Agilent). CE instrument was operated by OpenLAB CDS Chemstation software.

Parameters for the capillary zone electrophoretic separation: capillary: 90 cm × 50 µm fused silica; background electrolyte: 40 mM ammonium formate/ammonia (pH 9.5); applied voltage: 20 kV; hydrodynamic injection: 500 mbar·s; sheath liquid: iso-propanol: water = 1:1 containing 5 mM ammonia; sheath liquid flow rate: 10 µL/min. After each injection, a small amount of background electrolyte was also injected from a distinct vial (150 mbar) to eliminate carry-over effects. During electrophoresis, 35 mbar pressure was applied to the inlet buffer reservoir to decrease migration times significantly. The MS method in the negative mode was tuned according to the desired mass range (80–250 *m*/*z*), a much narrower one compared to the general detection of products, to obtain better sensitivity, although parameters for the ESI source were the same. Applied spectra rate was 3 Hz. Electropherograms were recorded by otofControl version 4.1 (build: 3.5, Bruker). Spectral background correction and internal calibration were performed on each electropherogram and peaks were integrated by Compass DataAnalysis version 4.4 (build: 200.55.2969).

Leaching of metal ion from the catalysts was studied by the ICP-OES technique after the following procedure: the remaining reaction mixture was centrifuged and the supernatant was filtered through a PTFE membrane filter (pore size: 0.45 µm), after which the solution was filled up in a volumetric flask. The concentrations were determined against standard solutions.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/10.3 390/gels8040202/s1, Figure S1: <sup>1</sup>H NMR spectra of the 5,10,15,20-tetrakis(4-aminophenyl)porphyrin and the coupling agent: 3-isocyanatopropyltriethoxysilane, Figure S2: <sup>1</sup>H NMR spectra of the mixture of 5,10,15,20-tetrakis(4-aminophenyl)porphyrin and the coupling agent 3-isocyanatopropyltriethoxysilane. The spectra of the reaction mixture were recorded directly after mixing and after 96 hours reaction time. The changes—especially in the aromatic region—in the spectra indicate change in the chemical structure of the porphyrin ring, meaning that the functionalization was successful, Figure S3: FT-IR spectra of the empty porphyrin ring (a), and the complexes with Cu(II) (b), and Fe(II) ions (c), Figure S4: Nitrogen adsorption-desorption isotherms of the catalysts denoted as CuPA (left) and the iron-containing one, denoted as FePA (right). The shapes, as expected, are almost perfectly alike, Figure S5: Initially the reaction mixture seemed to be homogeneous (left), but after a while without heating solid particles of the porphyrin complex appeared (center), settled down and the solution became colorless (right), Figure S6: Drawing of the continuous-flow tubular reactor for phenol oxidation.

**Author Contributions:** Conceptualization, I.L.; methodology, E.G. and I.L.; formal analysis, E.G., Á.K. and I.L.; investigation, E.G., M.S., Á.K. and I.L.; resources, I.L. and I.F.; data evaluation, E.G. and Á.K.; writing—original draft preparation, E.G.; writing—review and editing, I.L. and I.F.; visualization, E.G.; supervision, I.L.; project administration, E.G., I.L. and I.F.; funding acquisition, I.F. and I.L. All authors have read and agreed to the published version of the manuscript.

**Funding:** The research was supported by the EU and co-financed by the European Regional Development Fund under the project GINOP-2.3.2-15-2016-00041, GINOP-2.2.1-15-2017-00068, GINOP-2.3.2- 15-2016-00008, GINOP-2.3.3-15-2016-00004 and COST Action CA18125. The authors also acknowledge the financial support provided to this project by the Hungarian Science Foundation (OTKA: 17-124983) and the National Research, Development and Innovation Office, Hungary (K127931). The RAMAN measurements were financed by the GINOP-2.3.3-15-2016-00029 'HSLab' project.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** The authors would like to express their thanks to Edina Baranyai, Sándor Harangi and Petra Herman for the ICP-OES measurements, Lajos Daróczi for the SEM measurements and László Tóth for the FT/IR measurements.

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

#### **References**


## *Article* **Features of Luminescent Properties of Alginate Aerogels with Rare Earth Elements as Photoactive Cross-Linking Agents**

**Vladislav Kaplin 1,\* , Aleksandr Kopylov 1,2, Anastasiia Koryakovtseva <sup>1</sup> , Nikita Minaev <sup>3</sup> , Evgenii Epifanov <sup>3</sup> , Aleksandr Gulin <sup>1</sup> , Nadejda Aksenova 1,4 , Peter Timashev 1,4,5 , Anastasiia Kuryanova <sup>1</sup> , Ilya Shershnev <sup>1</sup> and Anna Solovieva <sup>1</sup>**


**Abstract:** Luminescent aerogels based on sodium alginate cross-linked with ions of rare earth elements (Eu3+, Tb3+, Sm3+) and containing phenanthroline, thenoyltrifluoroacetone, dibenzoylmethane, and acetylacetone as ligands introduced into the matrix during the impregnation of alginate aerogels (AEG), were obtained for the first time in a supercritical carbon dioxide medium. The impregnation method used made it possible to introduce organically soluble sensitizing ligands into polysaccharide matrices over the entire thickness of the sample while maintaining the porous structure of the aerogel. It is shown that the pore size and their specific area are 150 nm and 270 m2/g, respectively. Moreover, metal ions with content of about 23 wt.%, acting as cross-linking agents, are uniformly distributed over the thickness of the sample. In addition, the effect of sensitizing ligands on the luminescence intensity of cross-linked aerogel matrices is considered. The interaction in the resulting metal/ligand systems is unique for each pair, which is confirmed by the detection of broad bands with individual positions in the luminescence excitation spectra of photoactive aerogels.

**Keywords:** aerogels; lanthanide luminescence; supercritical carbon dioxide; sodium alginate; luminescent sensor

#### **1. Introduction**

The prospect of using rare earth elements (REE) in the creation of luminophores for analytical purposes, in particular for sensors, is usually associated not only with high quantum yields, long luminescence lifetimes, and a wide spectral range (from UV to IR), in which narrow-band luminescence of REE compounds is observed [1], but is determined by the possibility of regulating the functional characteristics of such systems when organic ligands of different natures are introduced (usually, the introduction of several ligands out of 4–6 possible options is conformationally acceptable) with the formation of specific photoactive centers "REE ion–ligands" [2–4]. Moreover, ligands also act as "antennas" [5–7] for radiation-initiating luminescence. This is even more important because of the low intensity of the luminescence of rare earth ions, caused by the forbidden electronic parity transitions [8]. The presence of third-party molecules or ions in the medium can affect the response of "antennas". Thus, cellulose aerogels cross-linked with terbium and europium ions and exhibiting luminescence sensitive to K<sup>+</sup> , Ni2+, Co2+, Cu2+, and Fe2+ ions are described in [9]. Zhang et al. demonstrate the quenching of the luminescence of Eu3+ ions in a complex with YVO<sup>4</sup> introduced into an alginate aerogel under the vapors of various

**Citation:** Kaplin, V.; Kopylov, A.; Koryakovtseva, A.; Minaev, N.; Epifanov, E.; Gulin, A.; Aksenova, N.; Timashev, P.; Kuryanova, A.; Shershnev, I.; et al. Features of Luminescent Properties of Alginate Aerogels with Rare Earth Elements as Photoactive Cross-Linking Agents. *Gels* **2022**, *8*, 617. https://doi.org/ 10.3390/gels8100617

Academic Editors: István Lázár and Melita Menelaou

Received: 17 August 2022 Accepted: 21 September 2022 Published: 27 September 2022

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

**Copyright:** © 2022 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/).

organic solvents, including acetone, benzene, toluene, etc. [10]. Hai et al. show the binding of terbium ions with cellulose macromolecules using bridge ligands: 4-aminopyridine-2,6 dicarboxylic acid and 2-(2-aminobenzamido)benzoic acid. The resulting material exhibits reversible quenching/buildup of the luminescence of terbium ions in the presence of ClO– and SCN– ions, respectively [11].

The next range of problems that arise when creating analytical, in particular sensor, systems using luminescent complexes based on REE elements is associated with the formation of these complexes in polymer aerogels with a superporous cross-linked structure, bearing in mind the conformational possibilities for placement in such matrices of ions of organic ligands in the vicinity of REE. It is known [12] that for such purposes, the sodium salt of alginic acid can be chosen as the polymer base. Indeed, lanthanide ions, similar to calcium ions, are capable of binding with carboxyl groups of alginate polyanion macromolecules [13] to form three-dimensionally cross-linked gels, which acquire the structure of aerogels after treatment in a supercritical carbon dioxide (SC-CO2) medium [14]. The formation of three-dimensionally cross-linked structures prevents the extraction of ionically bound luminophore centers from the matrix during SC-CO<sup>2</sup> drying [15], which occurs, for example, in lanthanide oxide aerogels [16]. At the same time, it should be noted that, although the methods for obtaining alginate aerogels lack such stages as hydrolysis, sol–gel formation, and sintering, which are characteristic of the synthesis of lanthanide-containing silicon and aluminum oxide aerogels [17], a certain problem associated with the search for common solvent for photoactive dopants and polymer matrix arises in the preparation of such aerogels. This problem is solved in this work based on a result previously obtained by the authors of the article. Thus, it is shown that organic sensitizing ligands, in particular phenanthroline, are easily introduced into cross-linked polymer matrices in a SC-CO<sup>2</sup> medium [18]. By this method, for the first time, this work demonstrates the sensitization of the luminescence of rare earth ions in the composition of alginate aerogel matrices, performed in a supercritical CO<sup>2</sup> medium. No similar researches were found in the literature. This made it possible to obtain luminescent aerogels based on sodium alginate cross-linked with lanthanide ions (Eu3+, Tb3+, Sm3+), and to establish the effects of the exposure of the formed systems of introduced sensitizing ligands (phenanthroline, dibenzoylmethane, thenoyltrifluoroacetone, acetylacetonate) on the luminescence intensity, with a possible perspective of using such systems as optical sensors for volatile organic substances.

#### **2. Results and Discussion**

#### *2.1. Some Physicochemical Characteristics of Aerogel Matrices*

#### 2.1.1. Specific Surface Area

Superporous aerogel structures were obtained by drying in a supercritical CO<sup>2</sup> medium (Figure 1). The average pore diameter is 149 nm ± 61 nm.

The data presented in Table 1 were obtained by the method of low-temperature adsorption of argon. The specific surface area (SSA) values of AEGs cross-linked by REE ions are comparable with similar values for some inorganic aerogels [19–22]. This makes it possible to use the obtained luminescent alginate aerogels as matrices in the development of sensors for the identification of gases and volatile substances.

**Table 1.** Specific surface area of alginate aerogel films cross-linked with REE ions.


**Figure 1.** SEM images of the surface of an aerogel film cross-linked with Eu3+ ions: (**A**) magnification 100×, (**B**) 810×, (**C**) 15000×, (**D**) 65000×. **Figure 1.** SEM images of the surface of an aerogel film cross-linked with Eu3+ ions: (**A**) magnification 100×, (**B**) 810×, (**C**) 15,000×, (**D**) 65,000×.

#### The data presented in Table 1 were obtained by the method of low-temperature 2.1.2. The Content of Rare Earth Metals in Cross-Linked Aerogel Matrices

adsorption of argon. The specific surface area (SSA) values of AEGs cross-linked by REE ions are comparable with similar values for some inorganic aerogels [19–22]. This makes it possible to use the obtained luminescent alginate aerogels as matrices in the development of sensors for the identification of gases and volatile substances. **Table 1.** Specific surface area of alginate aerogel films cross-linked with REE ions. **Aerogel Matrix SSA Average Value, m2/g** Eu AEG 255 ± 22 Tb AEG 290 ± 43 Sm AEG 270 ± 9 Inorganic AEG ≈ 100–2300 Even after prolonged washing of the hydrogels, the metal content in the obtained aerogels remains constant, which indicates the fixation of all REE ions in the cross-linking sites of the alginate matrix. Therefore, the content of rare earth metals can provide information on the degree of cross-linking of the three-dimensional structure of the alginate aerogel. The elemental analysis data presented in Figure 2 as six data rows, correspond to six radial straight lines emerging from the center of the cylinder (200 µm) to its edges (6400 µm). The analysis was performed along each straight line at seven points. It can be seen that the mass content of europium ions is approximately the same over the entire cross-section of the Eu AEG sample (cylinder with a diameter of 13 mm), and is about 26%, which is close to the theoretical maximum content of europium (22.4 wt.%), corresponding to one trivalent europium ion per three carboxyl units. However, local measurements at the cut points lead to a high measurement error, about 23%. Therefore, the thermogravimetric method was used to determine the exact metal content in cross-linked aerogels.

2.1.2. The Content of Rare Earth Metals in Cross-Linked Aerogel Matrices Even after prolonged washing of the hydrogels, the metal content in the obtained aerogels remains constant, which indicates the fixation of all REE ions in the crosslinking sites of the alginate matrix. Therefore, the content of rare earth metals can provide information on the degree of cross-linking of the three-dimensional structure of the alginate aerogel. The elemental analysis data presented in Figure 2 as six data rows, correspond to six radial straight lines emerging from the center of the cylinder (200 µm) The metal content was estimated by the thermogravimetric method, based on the fact that after thermal oxidation in air (at 1000 ◦C), the residue contains only metal oxide. The calculated maximum possible metal content (theoretical) and the content estimated using the thermogravimetric method (experimental) are presented in Table 2, and is also close to the EDS analysis data. The values that are lower than the theoretical ones are due to the incomplete reaction of the substitution of sodium ions by REE ions (about 85%), caused by steric hindrances in the formation of a cross-linked structure.

to its edges (6400 µm). The analysis was performed along each straight line at seven points. It can be seen that the mass content of europium ions is approximately the same over the entire cross-section of the Eu AEG sample (cylinder with a diameter of 13 mm), **Table 2.** Experimental and theoretical mass content of metal in alginate aerogel films cross-linked with Eu3+, Tb3+, and Sm3+ ions.


**Figure 2.** Mass content of Eu in a cross-linked matrix of sodium alginate. **Figure 2.** Mass content of Eu in a cross-linked matrix of sodium alginate.

The metal content was estimated by the thermogravimetric method, based on the fact that after thermal oxidation in air (at 1000 °C), the residue contains only metal oxide. The calculated maximum possible metal content (theoretical) and the content estimated using the thermogravimetric method (experimental) are presented in Table 2, and is also close to the EDS analysis data. The values that are lower than the theoretical ones are due to the incomplete reaction of the substitution of sodium ions by REE ions (about 85%), caused by steric hindrances in the formation of a cross-linked structure. **Table 2.** Experimental and theoretical mass content of metal in alginate aerogel films cross-linked with Eu3+, Tb3+, and Sm3+ ions. **Sample Metal Content (Experimental), wt.% Metal Content (Theoretical), wt.%** Eu AEG 19.4 ± 0.3 22.4 Tb AEG 20.9 ± 0.2 23.2 Sm AEG 17.9 ± 0.2 22.2 For additional characterization of the complexes formed upon SC impregnation of cross-linked aerogels with organic ligands, the materials were analyzed using FTIR. The FTIR spectra of the initial films and ligands, as well as the resulting systems, are shown in Figure 3a–c. As can be seen from the spectra, during the cross-linking of sodium alginate, the bands at 1403 cm−<sup>1</sup> and 1591 cm−<sup>1</sup> (characteristic bands of symmetric and antisymmetric C = O vibrations for salts of carboxylic acids) shift towards each other up to 1415 cm−<sup>1</sup> and 1585 cm−1, respectively, when the Na+ ion is replaced by the lanthanide For additional characterization of the complexes formed upon SC impregnation of cross-linked aerogels with organic ligands, the materials were analyzed using FTIR. The FTIR spectra of the initial films and ligands, as well as the resulting systems, are shown in Figure 3a–c. As can be seen from the spectra, during the cross-linking of sodium alginate, the bands at 1403 cm−<sup>1</sup> and 1591 cm−<sup>1</sup> (characteristic bands of symmetric and antisymmetric C=O vibrations for salts of carboxylic acids) shift towards each other up to 1415 cm−<sup>1</sup> and 1585 cm−<sup>1</sup> , respectively, when the Na<sup>+</sup> ion is replaced by the lanthanide ion. The 1024 cm−<sup>1</sup> band (C-O hydroxyl groups) also shifts to 1032 cm−<sup>1</sup> under the influence of a more electronegative ion. The IR spectra of the impregnated films are a superposition of the most intense bands of the ligand on the spectrum of cross-linked alginate, with the exception of the Acac ligand, whose bands are not detected due to the extremely low concentration (Figure S1). According to other publications, the absorption bands of Dbm carbonyl groups do not undergo significant shifts relative to the absorption of the free ligand upon coordination with lanthanides [23]. For the films SC -impregnated with Dbm, shifts of the C=O vibration bands from 1525 cm−<sup>1</sup> to 1517 cm−<sup>1</sup> and C-H from 1461 cm−<sup>1</sup> to 1479 cm−<sup>1</sup> are observed (Figure 3b). In this case, a band at 516 cm–1 appears, which is related to the new Ln–O bond, and is absent in the initial ligand and film. Due to the low concentration of the ligand, this band is be detected in SC-impregnated films with the Tta ligand. However, it is known that the position of the C=O band of the Tta ligand shift down by about 40 cm−<sup>1</sup> when combined with a rare earth ion [24,25]. For Ln AEG + Tta films, a shift of this band from 1638 cm–1 to 1596 cm–1 is observed, which confirms the coordination. On the spectra of films SC-impregnated with Phen, only a few of the most intense bands belonging to phenanthroline are detected. However, their position is also shifted relative to the bands of free phenanthroline, which is typical for Phen complexes with metal: from 623 cm−<sup>1</sup> to 635 cm−<sup>1</sup> , from 737 cm−<sup>1</sup> to 730 cm−<sup>1</sup> , from 852 cm−<sup>1</sup> to 841 cm−<sup>1</sup> , and from 1504 cm−<sup>1</sup> to 1518 cm−<sup>1</sup> (Figure 3c).

impregnated with Dbm, shifts of the C=O vibration bands from 1525 cm−<sup>1</sup> to 1517 cm−<sup>1</sup> and C-H from 1461 cm−<sup>1</sup> to 1479 cm−<sup>1</sup> are observed (Figure 3b). In this case, a band at 516

ion. The 1024 cm−<sup>1</sup> band (C-O hydroxyl groups) also shifts to 1032 cm−<sup>1</sup> under the influence of a more electronegative ion. The IR spectra of the impregnated films are a superposition of the most intense bands of the ligand on the spectrum of cross-linked alginate, with the exception of the Acac ligand, whose bands are not detected due to the extremely low concentration (Figure S1). According to other publications, the absorption bands of Dbm carbonyl groups do not undergo significant shifts relative to the cm−<sup>1</sup> (Figure 3c).

cm–1 appears, which is related to the new Ln–O bond, and is absent in the initial ligand and film. Due to the low concentration of the ligand, this band is be detected in SCimpregnated films with the Tta ligand. However, it is known that the position of the C=O band of the Tta ligand shift down by about 40 cm−<sup>1</sup> when combined with a rare earth ion [24,25]. For Ln AEG + Tta films, a shift of this band from 1638 cm–1 to 1596 cm–1 is observed, which confirms the coordination. On the spectra of films SC-impregnated with Phen, only a few of the most intense bands belonging to phenanthroline are detected. However, their position is also shifted relative to the bands of free phenanthroline, which is typical for Phen complexes with metal: from 623 cm−<sup>1</sup> to 635 cm−1, from 737 cm−<sup>1</sup> to 730 cm−1, from 852 cm−<sup>1</sup> to 841 cm−1, and from 1504 cm−<sup>1</sup> to 1518

**Figure 3.** (**a**) FTIR spectra of sodium alginate, dibenzoylmethane, initial cross-linked films Ln AEG, and SC-impregnated films Ln AEG + Dbm. (**b**) FTIR spectra of thenoyltrifluoroacetone, initial cross-linked films Ln AEG, and SC-impregnated films Ln AEG + Tta. (**c**) FTIR spectra of phenanthroline, initial cross-linked films Ln AEG, and SC-impregnated films Ln AEG + Phen. **Figure 3.** (**a**) FTIR spectra of sodium alginate, dibenzoylmethane, initial cross-linked films Ln AEG, and SC-impregnated films Ln AEG + Dbm. (**b**) FTIR spectra of thenoyltrifluoroacetone, initial crosslinked films Ln AEG, and SC-impregnated films Ln AEG + Tta. (**c**) FTIR spectra of phenanthroline, initial cross-linked films Ln AEG, and SC-impregnated films Ln AEG + Phen.

Well-studied ligands with known triplet energy levels close to the radiative levels of a

thenoyltrifluoroacetone. Based on the difference between the energies of the triplet level of the ligand and the radiative level of the metal, which should be in the range of 1000 cm–1– 5500 cm–1 [26], one can predict in advance the efficiency of energy transfer from the ligand to the metal [27–30]. Table 3 lists the energies of the triplet levels of the abovementioned ligands and the radiative levels of Eu3+, Tb3+, and Sm3+ ions, as well as the difference between these energies (ΔE) for each metal/ligand pair. Green color indicates the pairs for which an effective sensitization process is expected and, as a result, an increase in the intensity of the luminescence of rare earth ions. Highlighted in red are ΔE, at which the energy transfer from the ligand to the metal either does not occur (ΔE > 5500 cm−1), or at which the reverse energy

**Table 3.** The energies of the triplet levels of the Tta, Phen, Acac, and Dbm ligands, the energies of the excited radiative levels of the Eu3+, Tb3+, and Sm3+ ions, and their difference ΔE for each

(17267 cm−1) 3233 cm−<sup>1</sup> 4808 cm−<sup>1</sup> 8043 cm−<sup>1</sup> 3033 cm−<sup>1</sup>

(20394 cm−1) 106 cm−<sup>1</sup> 1681 cm−<sup>1</sup> 4916 cm−<sup>1</sup> <sup>−</sup>94 cm−<sup>1</sup>

(17825 cm−1) 2675 cm−<sup>1</sup> 4250 cm−<sup>1</sup> 7485 cm−<sup>1</sup> 2475 cm−<sup>1</sup>

Thus, the effective REE–ligand interaction should be observed for Eu and Sm systems with Tta, Phen, and Dbm, and for Tb with Phen and Acac. The efficiency of the sensitization process after the SC introduction of ligands was evaluated by the increase in the luminescence intensity of aerogel matrices cross-linked with rare earth ions, and the occurrence of metal–ligand interaction by changes in the luminescence excitation

**Tta: 20500 cm−<sup>1</sup> Phen: 22075 cm−<sup>1</sup> Acac: 25310 cm−<sup>1</sup> Dbm: 20300 cm−<sup>1</sup>**

*2.2. Effect of Organic Sensitizing Ligands on the Luminescent Properties of Aerogel* 

*Polysaccharide Matrices Cross-linked with REE Ions*

2.2.1. Luminescence of Aerogel Films

transfer dominates (ΔE < 1500 cm−1).

metal/ligand pair.

Eu (5D0)

Tb (5D4)

Sm (4G5/2)

#### *2.2. Effect of Organic Sensitizing Ligands on the Luminescent Properties of Aerogel Polysaccharide Matrices Cross-Linked with REE Ions*

2.2.1. Luminescence of Aerogel Films

Well-studied ligands with known triplet energy levels close to the radiative levels of a given series of metals, as well as well-soluble in SC-CO<sup>2</sup> medium, were used for sensitization of luminescence: acetylacetone, phenanthroline, dibenzoylmethane, and thenoyltrifluoroacetone. Based on the difference between the energies of the triplet level of the ligand and the radiative level of the metal, which should be in the range of 1000 cm–1–5500 cm–1 [26], one can predict in advance the efficiency of energy transfer from the ligand to the metal [27–30]. Table 3 lists the energies of the triplet levels of the abovementioned ligands and the radiative levels of Eu3+, Tb3+, and Sm3+ ions, as well as the difference between these energies (∆E) for each metal/ligand pair. Green color indicates the pairs for which an effective sensitization process is expected and, as a result, an increase in the intensity of the luminescence of rare earth ions. Highlighted in red are ∆E, at which the energy transfer from the ligand to the metal either does not occur (∆E > 5500 cm−<sup>1</sup> ), or at which the reverse energy transfer dominates (∆E < 1500 cm−<sup>1</sup> ).

**Table 3.** The energies of the triplet levels of the Tta, Phen, Acac, and Dbm ligands, the energies of the excited radiative levels of the Eu3+, Tb3+, and Sm3+ ions, and their difference ∆E for each metal/ligand pair.


Thus, the effective REE–ligand interaction should be observed for Eu and Sm systems with Tta, Phen, and Dbm, and for Tb with Phen and Acac. The efficiency of the sensitization process after the SC introduction of ligands was evaluated by the increase in the luminescence intensity of aerogel matrices cross-linked with rare earth ions, and the occurrence of metal–ligand interaction by changes in the luminescence excitation spectra. Indeed, the change in the luminescence intensity for all AEGs occurs in accordance with the expected results, with the exception of the Eu AEG + Dbm pair. Dbm molecules show a weaker sensitizing ability compared to other ligands, and no sensitization is observed in the Eu AEG matrix. This can be explained by the features of the keto–enol equilibrium of the Dbm tautomers in the nonpolar SC-CO<sup>2</sup> medium [31,32]. In the luminescence spectra of aerogel matrices cross-linked with Eu3+ and Tb3+ ions, characteristic narrow bands of low intensity metal-centered luminescence are observed. Moreover, the luminescence excitation spectra are also represented by a set of narrow bands, the positions of which are given in Table 4. At the same time, the characteristic luminescence (bands at 563 nm, 598 nm, and 644 nm) are not detected in the samples cross-linked with Sm3+ ions. The excitation and luminescence spectra of cross-linked aerogels are presented in Figures S2 and S3. Before SC impregnation, the samples are transparent white or slightly yellow. After the introduction of ligands in the SC-CO<sup>2</sup> medium, the transparency is preserved, and the films acquire a pink (for Phen and Dbm) or yellow (for Acac and Tta) tint. Figure 4 shows the films of the original Eu AEG (1) and of the Eu AEG SC-impregnated with Phen (2) and Tta (3) ligands. In the first row (A), the films are placed on a light monitor. Transparency is also confirmed by the absorption spectra of the original Eu AEG film and of the Eu AEG impregnated with Phen ligands (Figure S4). The second row (B) shows the films in daylight, the third row (C) shows the films exposed to 365 nm UV light. The remaining samples have a similar appearance, except that their glow under the ultraviolet light is not apparent to the naked eye.


**Table 4.** The position of the maxima of the original alginate aerogels cross-linked with REE ions

**Table 4.** The position of the maxima of the original alginate aerogels cross-linked with REE ions and aerogels after SC impregnation with organic ligands, as well as changes in the luminescence

**Table 4.** The position of the maxima of the original alginate aerogels cross-linked with REE ions and aerogels after SC impregnation with organic ligands, as well as changes in the luminescence intensity. The letters "S", "M", and "W" denote strong, medium, and weak luminescence bands,

**Table 4.** The position of the maxima of the original alginate aerogels cross-linked with REE ions and aerogels after SC impregnation with organic ligands, as well as changes in the luminescence intensity. The letters "S", "M", and "W" denote strong, medium, and weak luminescence bands,

*Gels* **2022**, *8*, x FOR PEER REVIEW 8 of 14

*Gels* **2022**, *8*, x FOR PEER REVIEW 8 of 14

*Gels* **2022**, *8*, x FOR PEER REVIEW 8 of 14

*Gels* **2022**, *8*, x FOR PEER REVIEW 8 of 14

*Gels* **2022**, *8*, x FOR PEER REVIEW 8 of 14


molecules (Table 4). Correspondingly, ligands dissolved in SC fluid are adsorbed on the surface and in the volume of aerogels, and coordinate near REE ions, forming luminescent systems with them. All recorded spectra of the SC-impregnated aerogels are shown in Figures S5–S15. In Table 4, green indicates an increase in luminescence intensity after impregnation with organic ligands, while red indicates a decrease in intensity. It is known that the standard procedure for obtaining luminescent organic REE complexes includes mixing solutions of metal salts and ligands, adjusting pH to a certain value, and isolating and purifying the precipitated product [33]. The obtained complexes surface and in the volume of aerogels, and coordinate near REE ions, forming luminescent systems with them. All recorded spectra of the SC-impregnated aerogels are shown in Figures S5–S15. In Table 4, green indicates an increase in luminescence intensity after impregnation with organic ligands, while red indicates a decrease in intensity. It is known that the standard procedure for obtaining luminescent organic REE complexes includes mixing solutions of metal salts and ligands, adjusting pH to a certain value, and isolating and purifying the precipitated product [33]. The obtained complexes Sm AEG +Dbm 395 nm Luminescence (W) In all cases, after SC impregnation of the matrices, the excitation spectra are represented by broad bands (from 140 to 240 nm wide) that are characteristic for organic molecules (Table 4). Correspondingly, ligands dissolved in SC fluid are adsorbed on the surface and in the volume of aerogels, and coordinate near REE ions, forming luminescent systems with them. All recorded spectra of the SC-impregnated aerogels are shown in Figures S5–S15. luminescent systems with them. All recorded spectra of the SC-impregnated aerogels are shown in Figures S5–S15. In Table 4, green indicates an increase in luminescence intensity after impregnation with organic ligands, while red indicates a decrease in intensity. It is known that the standard procedure for obtaining luminescent organic REE complexes includes mixing solutions of metal salts and ligands, adjusting pH to a certain value, and isolating and purifying the precipitated product [33]. The obtained complexes Eu AEG +Dbm 386 nm (W) Tb AEG +Dbm 295 nm (W) Sm AEG +Dbm 395 nm Luminescence (W) In all cases, after SC impregnation of the matrices, the excitation spectra are represented by broad bands (from 140 to 240 nm wide) that are characteristic for organic molecules (Table 4). Correspondingly, ligands dissolved in SC fluid are adsorbed on the shown in Figures S5–S15. In Table 4, green indicates an increase in luminescence intensity after impregnation with organic ligands, while red indicates a decrease in intensity. It is known that the standard procedure for obtaining luminescent organic REE complexes includes mixing solutions of metal salts and ligands, adjusting pH to a certain +Dbm 395 nm Luminescence (W) In all cases, after SC impregnation of the matrices, the excitation spectra are represented by broad bands (from 140 to 240 nm wide) that are characteristic for organic molecules (Table 4). Correspondingly, ligands dissolved in SC fluid are adsorbed on the surface and in the volume of aerogels, and coordinate near REE ions, forming luminescent systems with them. All recorded spectra of the SC-impregnated aerogels are shown in Figures S5–S15. In Table 4, green indicates an increase in luminescence intensity after impregnation with organic ligands, while red indicates a decrease in intensity. In all cases, after SC impregnation of the matrices, the excitation spectra are represented by broad bands (from 140 to 240 nm wide) that are characteristic for organic molecules (Table 4). Correspondingly, ligands dissolved in SC fluid are adsorbed on the surface and in the volume of aerogels, and coordinate near REE ions, forming luminescent systems with them. All recorded spectra of the SC-impregnated aerogels are shown in Figures S5–S15. In Table 4, green indicates an increase in luminescence intensity after impregnation with organic ligands, while red indicates a decrease in intensity. It is known that the standard procedure for obtaining luminescent organic REE **Matrices impregnated with acetylacetone (Acac)** Eu AEG +Acac 338 nm (M) Tb AEG +Acac 304 nm (S) Sm AEG +Acac **--** No luminescence **Matrices impregnated with dibenzoylmethane (Dbm)** Eu AEG +Dbm 386 nm (W) It is known that the standard procedure for obtaining luminescent organic REE complexes includes mixing solutions of metal salts and ligands, adjusting pH to a certain value, and isolating and purifying the precipitated product [33]. The obtained complexes are no longer able to act as cross-linking agents for water-soluble polyanions, since they become insoluble in aqueous media. On the other hand, the introduction of a ready-made luminescent REE complex into cross-linked aerogels (for example, by impregnation in SC-CO2) is limited by solubility in SC-CO2, concentration quenching, and aggregation. For alginate aerogels cross-linked with REE, each luminescent center is located in the cross-link site and is shielded from neighboring centers by fragments of polymer molecules. Therefore, the sensitization of such distributed luminescent centers in alginate matrices can be consid-

In Table 4, green indicates an increase in luminescence intensity after impregnation

It is known that the standard procedure for obtaining luminescent organic REE complexes includes mixing solutions of metal salts and ligands, adjusting pH to a certain

value, and isolating and purifying the precipitated product [33]. The obtained complexes

+Dbm 395 nm Luminescence (W)

In Table 4, green indicates an increase in luminescence intensity after impregnation

It is known that the standard procedure for obtaining luminescent organic REE complexes includes mixing solutions of metal salts and ligands, adjusting pH to a certain value, and isolating and purifying the precipitated product [33]. The obtained complexes

In all cases, after SC impregnation of the matrices, the excitation spectra are represented by broad bands (from 140 to 240 nm wide) that are characteristic for organic molecules (Table 4). Correspondingly, ligands dissolved in SC fluid are adsorbed on the surface and in the volume of aerogels, and coordinate near REE ions, forming luminescent systems with them. All recorded spectra of the SC-impregnated aerogels are

In Table 4, green indicates an increase in luminescence intensity after impregnation

It is known that the standard procedure for obtaining luminescent organic REE complexes includes mixing solutions of metal salts and ligands, adjusting pH to a certain value, and isolating and purifying the precipitated product [33]. The obtained complexes

surface and in the volume of aerogels, and coordinate near REE ions, forming luminescent systems with them. All recorded spectra of the SC-impregnated aerogels are

complexes includes mixing solutions of metal salts and ligands, adjusting pH to a certain value, and isolating and purifying the precipitated product [33]. The obtained complexes

shown in Figures S5–S15.

shown in Figures S5–S15.

Tb AEG

Sm AEG

with organic ligands, while red indicates a decrease in intensity.

with organic ligands, while red indicates a decrease in intensity.

with organic ligands, while red indicates a decrease in intensity.

respectively.

ered as a way to bypass the problem of concentration quenching. However, luminescence quenching is observed in aerogels with an ion content of about 20%. Thus, the maximum luminescence intensity is achieved at a REE content of about 10 wt.%. Also, preliminary tests of the luminescence sensitivity of some aerogel matrices to the presence of organic and inorganic vapors (acetone, ammonia) were carried out. The original non-impregnated aerogels cross-linked with REE are not sensitive to the tested volatile compounds. It is interesting to note that only one matrix (AEG Eu, containing Tta) shows the reaction to acetone after SC impregnation: the intensity of characteristic luminescence bands in the presence of acetone vapor increase by 32% (Figure S16). Luminescence quenching by ammonia vapor is observed in AEG Eu and AEG Tb aerogels SC-impregnated with Phen. The luminescence intensity of the matrices drops by 18% and 60%, respectively (Figures S17 and S18). Thus, not only the structure of the organic ligand, but also the metal, has a significant effect on the nature of the response of luminescent aerogels. characteristic luminescence (bands at 563 nm, 598 nm, and 644 nm) are not detected in the samples cross-linked with Sm3+ ions. The excitation and luminescence spectra of cross-linked aerogels are presented in Figures S2 and S3. Before SC impregnation, the samples are transparent white or slightly yellow. After the introduction of ligands in the SC-CO2 medium, the transparency is preserved, and the films acquire a pink (for Phen and Dbm) or yellow (for Acac and Tta) tint. Figure 4 shows the films of the original Eu AEG (1) and of the Eu AEG SC-impregnated with Phen (2) and Tta (3) ligands. In the first row (A), the films are placed on a light monitor. Transparency is also confirmed by the absorption spectra of the original Eu AEG film and of the Eu AEG impregnated with Phen ligands (Figure S4). The second row (B) shows the films in daylight, the third row (C) shows the films exposed to 365 nm UV light. The remaining samples have a similar appearance, except that their glow under the ultraviolet light is not apparent to the naked eye.

spectra. Indeed, the change in the luminescence intensity for all AEGs occurs in accordance with the expected results, with the exception of the Eu AEG + Dbm pair. Dbm molecules show a weaker sensitizing ability compared to other ligands, and no sensitization is observed in the Eu AEG matrix. This can be explained by the features of the keto–enol equilibrium of the Dbm tautomers in the nonpolar SC-CO2 medium [31,32]. In the luminescence spectra of aerogel matrices cross-linked with Eu3+ and Tb3+ ions, characteristic narrow bands of low intensity metal-centered luminescence are observed. Moreover, the luminescence excitation spectra are also represented by a set of narrow bands, the positions of which are given in Table 4. At the same time, the

*Gels* **2022**, *8*, x FOR PEER REVIEW 7 of 14

**Figure 4.** Eu AEG (**1**), Eu AEG + Phen (**2**), and Eu AEG + Tta (**3**) aerogels in transmitted light (row **A**), daylight (row **B**), and under 365 nm UV light (row **C**). **Figure 4.** Eu AEG (**1**), Eu AEG + Phen (**2**), and Eu AEG + Tta (**3**) aerogels in transmitted light (row **A**), daylight (row **B**), and under 365 nm UV light (row **C**).

2.2.2. Features of the Distribution of Impregnated Ligands in the Volume of Aerogels

It is important to note that when mentioning the most characteristic properties of aerogel materials, such as porosity, mechanical properties, and refractive index, threedimensional structure properties are implied. This also applies to luminescent properties: up to a certain thickness, aerogels are optically transparent for the UV–NIR range; accordingly, radiation should occur not only from rare earth ions localized on the surface of the matrix, but also in its volume. Therefore, it is necessary to make sure that the interaction of luminescent ions with the introduced sensitizing ligands occurs through the volume of the matrix. For example, Zhang et al. demonstrate the penetration of a sensitizing ligand into a sample layer no thicker than 130 µm, after impregnating an aerogel matrix in solution [10].

This work shows that the impregnation of alginate aerogels in the SC-CO<sup>2</sup> medium ensures the impregnation of the matrix to a depth of at least 3.3 mm (cylinder with a radius of 6.6 mm) (Figure 5). At the same time, this value is still limited only by the difficulties in obtaining thicker aerogel blocks, but not by the capabilities of the SC-fluid. Figure 5 shows the intensity distribution of the luminescence signal at 613 nm over the thickness of the Eu AEG (curve 1) and Eu AEG–Phen (after SC impregnation) matrices (curve 2). It can be seen that the increase in signal intensity occurs throughout the entire volume of the matrix, which indicates the penetration of the ligand dissolved in the SC-medium to a given depth. The increased values from the edges of the matrix (0–100 µm, 6000–6500 µm) are associated with a more intense diffusion of the solution into the near-surface layers. *Gels* **2022**, *8*, x FOR PEER REVIEW 10 of 14

**Figure 5.** Distribution of luminescence intensity at 613 nm over a section of the Eu AEG–Phen sample: (1) luminescence intensity in the initial matrix, (2) intensity after impregnation with phenanthroline. **Figure 5.** Distribution of luminescence intensity at 613 nm over a section of the Eu AEG–Phen sample: (1) luminescence intensity in the initial matrix, (2) intensity after impregnation with phenanthroline.

#### **3. Conclusions**

aerogels.

REE compounds).

**3. Conclusions** The luminescent aerogels based on sodium alginate, cross-linked with ions of rare earth elements (Eu3+, Tb3+, Sm3+) and containing phenanthroline, thenoyltrifluoroacetone, dibenzoylmethane, and acetylacetonate as ligands, introduced into the matrix during SC impregnation of alginate aerogels, were obtained in a supercritical carbon dioxide medium for the first time. It is shown that the intensity of the luminescence bands change after impregnation. Moreover, the nature of the influence of organic additives (ligands) on the luminescent properties of REE ions depends on the nature of both the ion and the ligand. It is demonstrated that upon SC impregnation, ligands can penetrate and act as luminescence sensitizers of rare earth ions throughout the entire thickness of The luminescent aerogels based on sodium alginate, cross-linked with ions of rare earth elements (Eu3+, Tb3+, Sm3+) and containing phenanthroline, thenoyltrifluoroacetone, dibenzoylmethane, and acetylacetonate as ligands, introduced into the matrix during SC impregnation of alginate aerogels, were obtained in a supercritical carbon dioxide medium for the first time. It is shown that the intensity of the luminescence bands change after impregnation. Moreover, the nature of the influence of organic additives (ligands) on the luminescent properties of REE ions depends on the nature of both the ion and the ligand. It is demonstrated that upon SC impregnation, ligands can penetrate and act as luminescence sensitizers of rare earth ions throughout the entire thickness of aerogels.

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

#### **4. Materials and Methods** *4.1. Preparation of Alginate Aerogels Cross-Linked with REE Ions*

*4.1. Preparation of Alginate Aerogels Cross-linked with REE Ions* The following substances were used without additional preparation and purification: REE chloride hexahydrates: ХCl3x6H2O, where Х is Eu, Tb, Sm (Aldrich, St. Louis, MO, USA, 99,9%); gadolinium (III) acetylacetonate hydrate (Gd(Acac)3хН2О) (Aldrich, 99,9%); europium (III) theonyltrifluoroacetonate trihydrate (Eu(Tta)3x3H2O) (Acros Organics, Geel, Belgium 95%); sodium alginate (Rushim, Moscow, Russia); sensitizing ligands: 1,10-phenanthroline (Acros Organics, 99+%); thenoyltrifluoroacetone (Aldrich, 99+%); dibenzoylmethane (Aldrich, 99+%); isopropanol (HIMFARM, Moscow, Russia ,TU 2632-181-44493179-2014) (hereinafter, coordination water is not indicated for The following substances were used without additional preparation and purification: REE chloride hexahydrates: XCl3x6H2O, where X is Eu, Tb, Sm (Aldrich, St. Louis, MO, USA, 99.9%); gadolinium (III) acetylacetonate hydrate (Gd(Acac)3xH2O) (Aldrich, 99.9%); europium (III) theonyltrifluoroacetonate trihydrate (Eu(Tta)3x3H2O) (Acros Organics, Geel, Belgium 95%); sodium alginate (Rushim, Moscow, Russia); sensitizing ligands: 1,10-phenanthroline (Acros Organics, 99+%); thenoyltrifluoroacetone (Aldrich, 99+%); dibenzoylmethane (Aldrich, 99+%); isopropanol (HIMFARM, Moscow, Russia, TU 2632- 181-44493179-2014) (hereinafter, coordination water is not indicated for REE compounds).

To create supercritical conditions for impregnation and drying, dry carbon dioxide,

Alginate aerogels in the form of films and cylinders were obtained by the following method. First, hydrogel films were obtained by pouring 40 mL of an aqueous solution of

To create supercritical conditions for impregnation and drying, dry carbon dioxide, with the volume content of water vapor not exceeding 0.001%, according to the quality certificate, was utilized (OOO "NII KM" 99.8% All-Union State Standard 8050-85). *Gels* **2022**, *8*, x FOR PEER REVIEW 11 of 14

Alginate aerogels in the form of films and cylinders were obtained by the following method. First, hydrogel films were obtained by pouring 40 mL of an aqueous solution of REE chloride (5 wt.%) into 30 mL of an aqueous solution of sodium alginate (2 wt.%) in a plastic Petri dish (d = 85 mm). The thickness of the formed film varied from 1 mm along the edges to 3 mm in the center. Hydrogel cylinders of 2 cm in diameter were obtained by squeezing 10 mL of a 2% aqueous solution of sodium alginate into a 10-fold excess of a 5% aqueous solution of REE chloride from a 10 mL syringe. The hydrogels were kept in distilled water for 72 h, changing the water three times to remove unreacted REE chloride. Then, the water in the hydrogels was replaced with isopropanol: the hydrogels were kept in a mixture of isopropanol/water (25/75) for 24 h, and then the proportion of isopropanol was increased by 25% once a day, bringing it to 100%. REE chloride (5 wt.%) into 30 mL of an aqueous solution of sodium alginate (2 wt.%) in a plastic Petri dish (d = 85 mm). The thickness of the formed film varied from 1 mm along the edges to 3 mm in the center. Hydrogel cylinders of 2 cm in diameter were obtained by squeezing 10 mL of a 2% aqueous solution of sodium alginate into a 10-fold excess of a 5% aqueous solution of REE chloride from a 10 mL syringe. The hydrogels were kept in distilled water for 72 h, changing the water three times to remove unreacted REE chloride. Then, the water in the hydrogels was replaced with isopropanol: the hydrogels were kept in a mixture of isopropanol/water (25/75) for 24 h, and then the proportion of isopropanol was increased by 25% once a day, bringing it to 100%. The cross-linked alginate hydrogels were dried in a high-pressure flow reactor in *Gels* **2022**, *8*, x FOR PEER REVIEW 11 of 14 REE chloride (5 wt.%) into 30 mL of an aqueous solution of sodium alginate (2 wt.%) in a plastic Petri dish (d = 85 mm). The thickness of the formed film varied from 1 mm along the edges to 3 mm in the center. Hydrogel cylinders of 2 cm in diameter were obtained by squeezing 10 mL of a 2% aqueous solution of sodium alginate into a 10-fold excess of a 5% aqueous solution of REE chloride from a 10 mL syringe. The hydrogels were kept in distilled water for 72 h, changing the water three times to remove unreacted REE chloride. Then, the water in the hydrogels was replaced with isopropanol: the hydrogels were kept in a mixture of isopropanol/water (25/75) for 24 h, and then the proportion of

The cross-linked alginate hydrogels were dried in a high-pressure flow reactor in supercritical carbon dioxide at a temperature of 40 ◦C and a pressure of 115 bar. The diagram of the process is shown in Figure 6. supercritical carbon dioxide at a temperature of 40 °C and a pressure of 115 bar. The diagram of the process is shown in Figure 6. isopropanol was increased by 25% once a day, bringing it to 100%. The cross-linked alginate hydrogels were dried in a high-pressure flow reactor in supercritical carbon dioxide at a temperature of 40 °C and a pressure of 115 bar. The

**Figure 6.** Diagram for obtaining aerogel films cross-linked with REE ions. **Figure 6.** Diagram for obtaining aerogel films cross-linked with REE ions. **Figure 6.** Diagram for obtaining aerogel films cross-linked with REE ions.

*4.2. Impregnation of Aerogels Cross-linked with REE Ions by the Organic Ligands 4.2. Impregnation of Aerogels Cross-Linked with REE Ions by the Organic Ligands 4.2. Impregnation of Aerogels Cross-linked with REE Ions by the Organic Ligands*

Aerogels were impregnated with the organic ligands in SC-CO2 medium. The concentration of ligands in the supercritical solution was 0.25 mg/mL. The impregnation was carried out for 1 h at a pressure of 180 bar and a temperature of 90 °C. Previously, in our work it is shown that, under these conditions, it is possible to achieve a uniform distribution of impregnated compounds in various polymer matrices in a SC-CO2 medium [34]. The reactor was then cooled to room temperature and depressurized to atmospheric pressure for 30 min (Figure 7). Aerogels were impregnated with the organic ligands in SC-CO<sup>2</sup> medium. The concentration of ligands in the supercritical solution was 0.25 mg/mL. The impregnation was carried out for 1 h at a pressure of 180 bar and a temperature of 90 ◦C. Previously, in our work it is shown that, under these conditions, it is possible to achieve a uniform distribution of impregnated compounds in various polymer matrices in a SC-CO<sup>2</sup> medium [34]. The reactor was then cooled to room temperature and depressurized to atmospheric pressure for 30 min (Figure 7). Aerogels were impregnated with the organic ligands in SC-CO2 medium. The concentration of ligands in the supercritical solution was 0.25 mg/mL. The impregnation was carried out for 1 h at a pressure of 180 bar and a temperature of 90 °C. Previously, in our work it is shown that, under these conditions, it is possible to achieve a uniform distribution of impregnated compounds in various polymer matrices in a SC-CO2 medium [34]. The reactor was then cooled to room temperature and depressurized to atmospheric pressure for 30 min (Figure 7).

**Figure 7.** Diagram of complex formation of new systems of REE-containing alginate and sensitizing ligands Tta, Phen, Acac, and Dbm. **Figure 7.** Diagram of complex formation of new systems of REE-containing alginate and sensitizing ligands Tta, Phen, Acac, and Dbm.

**Figure 7.** Diagram of complex formation of new systems of REE-containing alginate and

sensitizing ligands Tta, Phen, Acac, and Dbm.

#### *4.3. Determination of Luminescent and Physicochemical Characteristics of Cross-Linked Aerogel Matrices*

The luminescence and luminescence excitation spectra of the aerogel films were recorded using a Horiba Fluoromax Plus (Horiba-Jobin-Yvon, Palaiseau, France) spectrofluorometer at room temperature. The distribution of the luminescence intensity over the thickness of the aerogel cylinders was determined using a flexible optical fiber with a diameter of 0.8 mm directed at the cross-section of the sample, and a QE Pro 65000 spectrometer (Ocean Insight, Orlando, FL, USA). The displacement was provided by a movable stage with a positioning accuracy of 10 ± 1 µm. The values were recorded from the surface of the cross-section of the cylinder along a straight line from the periphery to the center with a step of 100 µm.

The specific surface area (SSA) of polysaccharide aerogels was determined by the low-temperature argon adsorption method (BET method). The analysis was carried out at the V.V. Voevodsky Laboratory of Kinetics of Mechanochemical and Free-Radical Processes (N.N. Semenov Federal Research Center for Chemical Physics, RAS, Moscow, Russia).

SEM images of the porous structure of aerogels were obtained using a scanning electron microscope Prisma E (Thermo Fisher Scientific, Scheepsbouwersweg, The Netherlands) after deposition of a layer of gold (10 nm). Data on the metal content in aerogel matrices were obtained from the surface of a cross-section of a cylindrical sample using a Phenom ProX scanning electron microscope (Thermo Fisher Scientific, Scheepsbouwersweg, The Netherlands) equipped with an energy-dispersive spectroscopy (EDS) silicon drift detector, which allows the performance of elemental analysis. Also, the metal content in the matrices was determined by the gravimetric method, based on the residue after burning the samples in a Saturn 1 high-temperature furnace at a temperature of 1000 ◦C.

FTIR analysis of the initial components and the synthesized system was carried out using a spectrum two FT-IR spectrometer (PerkinElmer, Waltham, MA, USA) in attenuated total reflectance (ATR) mode. The spectrometer features were as follows: highperformance, room-temperature LiTaO<sup>3</sup> MIR detector, standard optical system with KBr windows for data collection over a spectral range of 4000–350 cm−<sup>1</sup> at a resolution of 0.5 cm−<sup>1</sup> . All spectra were initially collected in ATR mode and converted into IR transmittance mode.

**Supplementary Materials:** The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/gels8100617/s1, Figure S1: FTIR spectra of initial cross-linked films Ln AEG and SC-impregnated films Ln AEG + Acac; Figures S2 and S3: Luminescence spectra (orange curve) and luminescence excitation spectra (green curve) of alginate aerogels cross-linked with Eu3+ and Tb3+ ions, respectively; Figure S4: Absorption spectra of Eu AEG (curve 1) and Eu AEG film after SC impregnation of Phen (curve 2). Figures S5–S15: Luminescence (orange curve) and luminescence excitation (green curve) spectra of alginate aerogels cross-linked with Eu3+, Tb3+, and Sm3+ ions, SC-impregnated with Tta, Phen, Acac, and Dbm ligands; Figure S16: Luminescence spectra: 1.AEG Eu SC-impregnated with Tta before and 2. after exposure to acetone vapor; Figure S17: Luminescence spectra: 1. AEG Eu SC-impregnated with Phen before and 2. after exposure to ammonia vapor, Figure S18: Luminescence spectra: 1. AEG Tb SC-impregnated with Phen before and 2. after exposure to ammonia vapor.

**Author Contributions:** Writing—original draft preparation, V.K.; conceptualization and supervision, A.K. (Aleksandr Kopylov); methodology and investigation, V.K., A.K. (Anastasiia Koryakovtseva), E.E. and N.A.; resources, N.M. and A.G.; project administration, P.T.; visualization, A.K. (Anastasiia Kuryanova); data curation, I.S.; writing—review and editing and funding acquisition, A.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This study was conducted in the framework of the Russian Government assignment № 122040400099-5. The work was supported by the Ministry of Science and Higher Education as part of the work under the state task of the Federal Research Center "Crystallography and Photonics" of the Russian Academy of Sciences in terms of using the equipment of the Center for Collective Use "Structural Diagnostics of Materials" when characterizing samples using the energy-dispersive spectroscopy (EDX) method and measuring local luminescence spectra of aerogel samples. SEM

images and luminescence spectra were obtained with the equipment of the FRCCP RAS shared research facilities (No. 506694).

**Data Availability Statement:** Data are contained within the article or supplementary material.

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

#### **References**

