Synthesis of New Cyclodextrin-Based Adsorbents to Remove Direct Red 83:1

Two cyclodextrins (CDs), γ– and hydroxypropyl (HP)–γ–CDs were used to synthesize new adsorbents by using epichlorohydrin (EPI) as cross-linking agent in order to remove Direct Red 83:1 (DR) from water. Both polymers were characterized in terms of Fourier spectroscopy, nuclear magnetic resonance, particle size distribution and thermogravimetric analysis. Experimental data for both polymers were well fitted to the pseudo-second order and intraparticle diffusion model, indicating that in the adsorption both chemical and physical interactions are essential in the removal of DR. Three different isotherm models were analyzed, concluding that γ–CDs–EPI followed the Temkin isotherm and HP–γ–CDs-EPI the Freundlich isotherm, these results suggested that the adsorption was happening onto heterogeneous surfaces. The results of the Gibbs free energy showed that the adsorption was spontaneous at room temperature. In order to eliminate the remaining dye after the polymer treatment, and advanced oxidation process (AOP) was considered, achieving more than 90% of removal combining both mechanisms.


Introduction
There is a huge concern related to the release of emerging contaminants in the environment, this issue is considered nowadays as a worldwide risk. Emerging pollutants encompass a wide range of man-made chemicals (such as pesticides, cosmetics, personal and household care products, pharmaceuticals, textile dyes, etc.), which are in use worldwide and which are indispensable for modern society. Human activities have led to the contamination of water resources with micropollutants. Such agents have generated renewed awareness due to their potential pathogenicity.
Textile industries use a wide range of direct dyes to colour their goods. Apart from that, the consumption of water gives rise to an average of 200 L/kg of fibre, producing a large volume of wastewater [1]. The most common dyes used in the textile industry are sulfur, indigoid, anthraquinone, triphenylmethyl and azo derivatives, being azo dyes the most employed because of their high solubility in aqueous solution [2]. The discharge of dyes into water effluents is extremely dangerous due to the undesirable characteristics that will transfer to the ambient [3]. Therefore, due to the high restrictions related to the discharge of dyes in the effluents, it is mandatory to treat water before discharging it into the environment [4].
Conventional methods to remove dyes from wastewater are not effective due to their resistant to aerobic biodegradation, heat, light and oxidizing agents. One remarkable alternative is the use of adsorption techniques for the removal of dyes. Adsorption processes are considered the most useful

Adsorption Experiments
The experiments were conducted at room temperature using a set of Direct Red (DR) concentrations (from 25 to 300 mg/L). Flasks containing 50 mL of each concentration were mixed with 1 g of adsorbent. This mixture was agitated at 500 rpm and at specific intervals, aliquots were extracted from the samples in order to analyse the remaining dye in the liquid solution. Prior to measure the absorbance at 526 nm, the samples were centrifuged to discard impurities.
The dye entrapped on the polymeric adsorbents was analysed by using the following equation (1) [24]:

Adsorption Experiments
The experiments were conducted at room temperature using a set of Direct Red (DR) concentrations (from 25 to 300 mg/L). Flasks containing 50 mL of each concentration were mixed with 1 g of adsorbent. This mixture was agitated at 500 rpm and at specific intervals, aliquots were extracted from the samples in order to analyse the remaining dye in the liquid solution. Prior to measure the absorbance at 526 nm, the samples were centrifuged to discard impurities. The dye entrapped on the polymeric adsorbents was analysed by using the following Equation (1) [24]: where C o is the initial dye concentration (mg/L), C e the liquid phase dye concentration at equilibrium (mg/L), V the volume of dye used (L) and m is the mass of polymer utilized (g). Three replicates were performed and plotted.

Polymer Characterization
Fourier transform infrared spectroscopy (FTIR) spectra, nuclear magnetic resonance (NMR) spectra, particle size distribution, thermogravimetric analysis (TGA) and surface morphologies were carried out in order to characterize the polymer as reported before [25].
FTIR spectra of the samples were collected in a VARIAN FT-IR 670 in the range 4000-400 cm −1 in attenuated total reflection (ATR) mode with a spectral resolution of 0.1 cm −1 .
Nuclear magnetic resonance (NMR) analysis was performed in a Bruker Advance AV300 MHz spectrometer. Briefly 40 mg of the sample were dissolved or suspended in 1ml deuterium oxide (D 2 O), filtered through 0.45 µm and analyzed for 1 H NMR spectra in the chemical shift range of 14 to −1 ppm.
Particle size distribution was studied by laser diffraction using acetone as solvent in a Mastersizer 3000E (Malvern Panalytical).
The porosity and apparent density of the polymers was calculated using the following equations: where V t (cm 3 ) is the total volume of polymers, V a (cm 3 ) is the actual volume of the material, M a (g) is the mass of the polymers and ρ (g/cm 3 ) is the density of the material. The experiments were repeated three times. The swelling capacity of the polymers was measured by using a gravimetric method: 1 g of dry samples was immersed in 200 mL of distilled water at room temperature for 3 h to reach swelling equilibrium. After this time, both polymers were filtered to remove unabsorbed water. The swelling equilibrium (Q eq g/g) was determined according to the following equation: where W d is the mass of dried polymers (g) and W s is the mass of swollen polymers (g). The experiments were repeated three times [26]. For thermogravimetric analysis (TGA) between 1-3 mg of each sample in aluminium pans were introduced in a Discovery TGA55 (TA Instruments) thermogravimetric instrument. After stabilization, the samples were subjected to heat scanning from 25 to 350 • C at 10 • C·min −1 under nitrogen atmosphere and the loss of weight recorded and compared.

Advanced Oxidation Process (AOP)
The highest dye concentration remaining after the adsorption processes was 80-90 mg/L, which corresponded to the tests with initial concentration of 300 mg/L for both dyes. In order to degrade that concentration of dye, 20 mL of a mixture of dye and hydrogen peroxide at final concentrations of 86 mg/L (87 µM) and 1480 mg/L (43 mM), respectively, was prepared in a Petri dish and subjected to pulsed light treatment. A high ratio hydrogen peroxide/dye was used in order to avoid make the first the limiting reagent of the reaction. The Petri dish without cover was placed in a pulsed light device (XeMaticA-Basic-1L, Steribeam, Germany). The system was operated at 2.5 kV and produced a light fluence of 2.14 J/cm 2 at the surface of the liquid. Under these conditions, the emission of the lamp has a spectrum similar to the reported Cudemos et al., [27]. Increasing fluences up to 193 J/cm 2 were reached by applying multiple pulses. Tests were carried on in triplicate.
Pseudo-first order degradation constants (k, cm 2 /J) were calculated as function of fluence (H o , J/cm 2 ) according to the following equation: where C is the concentration of dye at fluence H o and C o is the initial dye concentration.

Polymer Characterization
In the infrared (IR) spectra of γ-CD, HP-γ-CD and their corresponding polymers ( Figure 2) a wide band was observed between 3400-3200 cm −1 . This band is attributed to the stretching vibrations of either the O-H bonds of the primary and secondary -OH groups of the glucopyranose units and the isopropyl moieties in the parent cyclodextrins, and to the remaining not linked C-OH groups in the polymers. An absorption band is also observed, belonging to the stretching vibrations of the C-H bonds in the CH and CH 2 groups with a maximum around 2920 cm −1 . In polymeric compounds an additional band appears at ca. 2880 cm −1 responding to the same sort of vibration from the new carbon backbone bonded to the cyclodextrin rings. The Petri dish without cover was placed in a pulsed light device (XeMaticA-Basic-1L, Steribeam, Germany). The system was operated at 2.5 kV and produced a light fluence of 2.14 J/cm 2 at the surface of the liquid. Under these conditions, the emission of the lamp has a spectrum similar to the reported Cudemos et al., [27]. Increasing fluences up to 193 J/cm 2 were reached by applying multiple pulses. Tests were carried on in triplicate.
Pseudo-first order degradation constants (k, cm 2 /J) were calculated as function of fluence (Ho, J/cm 2 ) according to the following equation: where C is the concentration of dye at fluence Ho and Co is the initial dye concentration.

Polymer Characterization
In the infrared (IR) spectra of γ-CD, HP-γ-CD and their corresponding polymers ( Figure 2) a wide band was observed between 3400-3200 cm −1 . This band is attributed to the stretching vibrations of either the O-H bonds of the primary and secondary -OH groups of the glucopyranose units and the isopropyl moieties in the parent cyclodextrins, and to the remaining not linked C-OH groups in the polymers. An absorption band is also observed, belonging to the stretching vibrations of the C-H bonds in the CH and CH2 groups with a maximum around 2920 cm −1 . In polymeric compounds an additional band appears at ca. 2880 cm −1 responding to the same sort of vibration from the new carbon backbone bonded to the cyclodextrin rings.  The 1 H NMR spectrum ( Figure S1, Supplementary Material) of the two polymeric structures and their parent cyclodextrins was performed in D2O to record protons bonded other than OH groups. Compared to the cyclodextrins the appearance of two novel signals at medium and low chemical shift together with their high intensity related to the signal of the parent compounds suggests a high level of crosslinking. Nevertheless, no additional information could be drawn from the spectrum of these polymers due to their low solubility and additional experiments will be undertaken to unravel the internal organization of the polymers.

-CDs
Particle size distribution ( Figure 3) was analysed in terms of the volumetric size (D [4:3]). Comparing both adsorbents, the result was higher for HP-γ-CDs-EPI (555 μm) than γ-CDs-EPI (54 μm). Span values were calculated by using the following equation (6): The results obtained were 1.8 for γ-CDs, 1.8 for HP-γ-CDs, 6.7 for γ-polymer and 1.6 for HPγ-polymer. In this case, lower values are in accordance with more homogeneous particle distribution. The 1 H NMR spectrum ( Figure S1, Supplementary Material) of the two polymeric structures and their parent cyclodextrins was performed in D 2 O to record protons bonded other than OH groups. Compared to the cyclodextrins the appearance of two novel signals at medium and low chemical shift together with their high intensity related to the signal of the parent compounds suggests a high level of crosslinking. Nevertheless, no additional information could be drawn from the spectrum of these polymers due to their low solubility and additional experiments will be undertaken to unravel the internal organization of the polymers.

HP--CDs-EPI
Particle size distribution ( Figure 3) was analysed in terms of the volumetric size (D [4:3]). Comparing both adsorbents, the result was higher for HP-γ-CDs-EPI (555 µm) than γ-CDs-EPI (54 µm). Span values were calculated by using the following Equation (6):  The characterization of both adsorbents involved the measure of the swelling capacity, porosity, density and the particle size distribution. The results obtained could be observed in Table 1. According to the results, both CD polymers showed similar properties in terms of swelling capacity, porosity and density. The most remarkable differences are related to the size of the particles and the homogeneity of them as explained previously.  It could be observed in all the thermograms ( Figure 4) a first weight loss around 100 °C that can be explained by removal and evaporation of surface-adsorbed water molecules. In the case of polymeric structures this loss is observed at higher temperatures, probably due to an increase in the surface area that holds the molecules more efficiently. The second loss of weigh in the curves corresponds to the thermal decomposition of cyclodextrin oligosaccharides in the case of the parent compounds. In the EPI-crosslinked compounds the second process normally starts at lower temperatures with a gradually degradation of the alkyl backbones followed by the subsequent decomposition of the cyclodextrin structures. The results obtained were 1.8 for γ-CDs, 1.8 for HP-γ-CDs, 6.7 for γ-polymer and 1.6 for HP-γ-polymer. In this case, lower values are in accordance with more homogeneous particle distribution.

Adsorbent Properties γ-CDs-EPI HP-γ-CDs-EPI
The characterization of both adsorbents involved the measure of the swelling capacity, porosity, density and the particle size distribution. The results obtained could be observed in Table 1. According to the results, both CD polymers showed similar properties in terms of swelling capacity, porosity and density. The most remarkable differences are related to the size of the particles and the homogeneity of them as explained previously. It could be observed in all the thermograms ( Figure 4) a first weight loss around 100 • C that can be explained by removal and evaporation of surface-adsorbed water molecules. In the case of polymeric structures this loss is observed at higher temperatures, probably due to an increase in the surface area that holds the molecules more efficiently. The second loss of weigh in the curves corresponds to the thermal decomposition of cyclodextrin oligosaccharides in the case of the parent compounds. In the EPI-crosslinked compounds the second process normally starts at lower temperatures with a gradually degradation of the alkyl backbones followed by the subsequent decomposition of the cyclodextrin structures.

Effect of Contact Time
The first step in the uptake of dyes from water is to determine the best adsorption conditions for the adsorbents. In order to obtain this information, it is essential to evaluate different factors, such as: adsorbent dosage, agitation speed or pH as the most important conditions. The main aim of these experiments is to measure the capability of the polymers synthesized to remove direct dyes from water by using the optimal adsorption conditions.
According to the results obtained, the following conditions were fixed to perform the adsorption experiments: 1 g of polymer, 500 rpm and pH 7. After evaluating the best adsorption conditions, the next step is to analyse the effect of contact time between the adsorbents and DR by using eight different concentrations.
The results of contact time for γ-and HP-γ-CDs-EPI could be seen in Figure 5. Analysing the results for γ-CDs-EPI, increasing the concentration of Direct Red gave rise to increasing qt values for the whole range of concentration used. However, from 25 to 150 mg/L, the adsorption was very fast, reaching adsorption equilibrium after 40 min. In the case of the highest dye concentrations, the trend was clearly different, the equilibrium time increased from 40 to 80 min, this is due to the high capability of this adsorbent to entrap more dye molecules at high concentrations of DR.

Effect of Contact Time
The first step in the uptake of dyes from water is to determine the best adsorption conditions for the adsorbents. In order to obtain this information, it is essential to evaluate different factors, such as: adsorbent dosage, agitation speed or pH as the most important conditions. The main aim of these experiments is to measure the capability of the polymers synthesized to remove direct dyes from water by using the optimal adsorption conditions.
According to the results obtained, the following conditions were fixed to perform the adsorption experiments: 1 g of polymer, 500 rpm and pH 7. After evaluating the best adsorption conditions, the next step is to analyse the effect of contact time between the adsorbents and DR by using eight different concentrations.
The results of contact time for γand HP-γ-CDs-EPI could be seen in Figure 5. Analysing the results for γ-CDs-EPI, increasing the concentration of Direct Red gave rise to increasing q t values for the whole range of concentration used. However, from 25 to 150 mg/L, the adsorption was very fast, reaching adsorption equilibrium after 40 min. In the case of the highest dye concentrations, the trend was clearly different, the equilibrium time increased from 40 to 80 min, this is due to the high capability of this adsorbent to entrap more dye molecules at high concentrations of DR. Polymers 2020, 12, x FOR PEER REVIEW 10 of 21 For HP-γ-CDs-EPI the results were similar for the different concentrations, the adsorption was rapid, independently of dye concentration, according to Figure 4, the equilibrium time was reached after 30-40 min of adsorption.

Adsorption Kinetics
Adsorption kinetics were determined by adjusting the experimental data to three different models (pseudo-first (PFOM), pseudo-second (PSOM) and intraparticle diffusion (IDM) models, respectively). The linearized equations for these models are listed as follows [28][29][30]: qe and qt are the quantity of dye adsorbed (mg/g), k1 (min −1 ) is the constant related to the pseudo-first model, k2 (g/mg min) is the constant related to the pseudo-second model, ki (mg/g min ½ ) is the constant related to the intraparticle diffusion model, t is the time and C is the intercept (mg/g). The results observed for the adjustment to the PFOM for both adsorbents could be seen in Figure  6 and Table 2. With the objective to obtain the best representation possible, half of the values of contact time were not taken into account, so for the PFOM plot only the first 50 min of contact were considered. Using the whole range of measure (120 min), the straight lines obtained for each concentration showed a very poor determination coefficient (R 2 ). In the case of 50 min of contact time, the experimental values obtained ( Table 2) were similar in some cases to the calculated values using the PFOM. The R 2 values ranged from 0.8 to 0.98, showing a high deviation between values. When this trend occurs, it is more likely that the adsorption of DR on both adsorbents might take place through the PSOM. For HP-γ-CDs-EPI the results were similar for the different concentrations, the adsorption was rapid, independently of dye concentration, according to Figure 4, the equilibrium time was reached after 30-40 min of adsorption.

Adsorption Kinetics
Adsorption kinetics were determined by adjusting the experimental data to three different models (pseudo-first (PFOM), pseudo-second (PSOM) and intraparticle diffusion (IDM) models, respectively). The linearized equations for these models are listed as follows [28][29][30]: q e and q t are the quantity of dye adsorbed (mg/g), k 1 (min −1 ) is the constant related to the pseudo-first model, k 2 (g/mg min) is the constant related to the pseudo-second model, k i (mg/g min 1 2 ) is the constant related to the intraparticle diffusion model, t is the time and C is the intercept (mg/g).
The results observed for the adjustment to the PFOM for both adsorbents could be seen in Figure 6 and Table 2. With the objective to obtain the best representation possible, half of the values of contact time were not taken into account, so for the PFOM plot only the first 50 min of contact were considered. Using the whole range of measure (120 min), the straight lines obtained for each concentration showed a very poor determination coefficient (R 2 ). In the case of 50 min of contact time, the experimental values obtained (Table 2) were similar in some cases to the calculated values using the PFOM. The R 2 values ranged from 0.8 to 0.98, showing a high deviation between values. When this trend occurs, it is more likely that the adsorption of DR on both adsorbents might take place through the PSOM. Polymers 2020, 12, x FOR PEER REVIEW 11 of 21      The results for the PSOM adjustment demonstrated the perfect fit between the experimental data and this model. The determination coefficients confirmed these results for both adsorbents (Figure 7, Table 2). Due to these values, it is possible to confirm that chemical forces were playing an essential role in the adsorption of DR. The comparison of these results with previously published papers showed that similar kinetics were obtained using different dyes and adsorbents such as the removal of methylene blue on fly ash [31], the removal of acidic dyes on silica [32], the adsorption of DR on CDs [16,17] or the elimination of Direct Blue on chitosan [7]. The results for the PSOM adjustment demonstrated the perfect fit between the experimental data and this model. The determination coefficients confirmed these results for both adsorbents (Figure 7, Table 2). Due to these values, it is possible to confirm that chemical forces were playing an essential role in the adsorption of DR. The comparison of these results with previously published papers showed that similar kinetics were obtained using different dyes and adsorbents such as the removal of methylene blue on fly ash [31], the removal of acidic dyes on silica [32], the adsorption of DR on CDs [16,17] or the elimination of Direct Blue on chitosan [7]. To analyse the effect of the intraparticle diffusion on the adsorption of DR is necessary to consider the presence of different steps in this plot. Two different straight lines indicated us that two different forces are controlling the adsorption, in our case both chemical and intraparticle diffusion were playing a key role in this adsorption [33]. The first part represents the chemical adsorption and the second one the intraparticle diffusion. According to our results ( Figure 8 and Table 2), both adsorbents showed this multi-step adsorption process. The first part of the plot is a straight line, indicating the surface adsorption for both polymers, the second part of the representation is a flat line, especially relevant at high DR concentrations, indicating that the IDM is important in the adsorption using cyclodextrin polymers. In this adsorption both chemical and intraparticle interactions are involved in the removal of this azo dye from water. Due to the presence of two different steps in the IDM it was not possible to obtain high R 2 values. This values ranged from 0.575 to 0.960 for γ-CDs-EPI and from 0.649 to 0.865 in the case of HP-γ-CDs-EPI. To analyse the effect of the intraparticle diffusion on the adsorption of DR is necessary to consider the presence of different steps in this plot. Two different straight lines indicated us that two different forces are controlling the adsorption, in our case both chemical and intraparticle diffusion were playing a key role in this adsorption [33]. The first part represents the chemical adsorption and the second one the intraparticle diffusion. According to our results ( Figure 8 and Table 2), both adsorbents showed this multi-step adsorption process. The first part of the plot is a straight line, indicating the surface adsorption for both polymers, the second part of the representation is a flat line, especially relevant at high DR concentrations, indicating that the IDM is important in the adsorption using cyclodextrin polymers. In this adsorption both chemical and intraparticle interactions are involved in the removal of this azo dye from water. Due to the presence of two different steps in the IDM it was not possible to obtain high R 2 values. This values ranged from 0.575 to 0.960 for γ-CDs-EPI and from 0.649 to 0.865 in the case of HP-γ-CDs-EPI. Polymers 2020, 12, x FOR PEER REVIEW 13 of 21

Adsorption Equilibrium
The analysis of the equilibrium adsorption state is very important as it allows understanding the adsorption mechanism. The experimental data were fitted using Freundlich, Langmuir and Temkin isotherm models that are the most common mechanisms to explain the aqueous phase adsorption and are expressed by using Equations (10)-(12) [34][35][36]: ln q e = ln K F + 1 n F ln C e (10) KF is the constant associated to the Freundlich isotherm (mg/g) (L/mg 1/n ), nF is the heterogeneity factor, KL (L/g) and aL (L/mg) are the constants related to the Langmuir model, qmax is defined by KL/aL. bT is the constant related to the adsorption heat (kJ/mol), aT is the constant related to the Temkin isotherm (L/g), R is the universal constant (8.314 J/mol K) and T is the temperature (K). The Freundlich isotherm model presented the best fit for HP-γ-CDs-EPI, taking into account the R 2 value (0.982). This means that heterogeneous surfaces are playing a key role in the adsorption of DR (Figure 9 and Table 3). The most useful parameter related to this isotherm is nF, which is the heterogeneity factor, when nF is ranged from 1 to 10, the adsorption process is favoured, this trend was accomplished for both adsorbents (Table 3).

Adsorption Equilibrium
The analysis of the equilibrium adsorption state is very important as it allows understanding the adsorption mechanism. The experimental data were fitted using Freundlich, Langmuir and Temkin isotherm models that are the most common mechanisms to explain the aqueous phase adsorption and are expressed by using Equations (10)-(12) [34][35][36]: K F is the constant associated to the Freundlich isotherm (mg/g) (L/mg 1/n ), n F is the heterogeneity factor, K L (L/g) and a L (L/mg) are the constants related to the Langmuir model, q max is defined by K L /a L . b T is the constant related to the adsorption heat (kJ/mol), a T is the constant related to the Temkin isotherm (L/g), R is the universal constant (8.314 J/mol K) and T is the temperature (K). The Freundlich isotherm model presented the best fit for HP-γ-CDs-EPI, taking into account the R 2 value (0.982). This means that heterogeneous surfaces are playing a key role in the adsorption of DR (Figure 9 and Table 3). The most useful parameter related to this isotherm is n F , which is the heterogeneity factor, when n F is ranged from 1 to 10, the adsorption process is favoured, this trend was accomplished for both adsorbents (Table 3). Polymers 2020, 12, x FOR PEER REVIEW 14 of 21  The lowest R 2 values (0.942 for γ-CDs-EPI and 0.977 for HP-γ-CDs-EPI) were observed by adjusting the experimental data to the Langmuir isotherm. For this isotherm, it is interesting to analyse the value of qmax (maximum adsorption for the adsorbent under specific experimental  The lowest R 2 values (0.942 for γ-CDs-EPI and 0.977 for HP-γ-CDs-EPI) were observed by adjusting the experimental data to the Langmuir isotherm. For this isotherm, it is interesting to analyse the value of q max (maximum adsorption for the adsorbent under specific experimental conditions), both polymers showed similar ability to entrap the dye (11.92 mg/g for γ-CDs-EPI and 14.60 mg/g for HP-γ-CDs-EPI). Table 4 shows an exhaustive comparison and analysis of different q max values for different CDs adsorbents and dyes. Apart from q max , the R L parameter must be considered for this isotherm. It is a dimensionless constant and is described by the following Equation (13): Finally, the experimental data were adjusted to the Temkin isotherm. The best adjustment for γ-CDs-EPI was achieved by using this model (0.946). This could be attributed to the adsorption happening onto heterogeneous surfaces [37]. The Temkin model demonstrated that the heat of adsorption decreased during the adsorption process [38]. The results for the b T parameter were 1.053 kJ/mol and 0.890 kJ/mol for γand HP-γ-CDs-EPI polymers, according to the range of binding energies stated on the literature, these results suggested that physicochemical forces were influencing the adsorption of DR on to the CD adsorbents [39].  When the results obtained for the separation factor ranges from 0 to 1, the adsorption is a favourable process and, according to our results, in both adsorbents the R L value was within this range, indicating and confirming that the adsorption process was favorable for both polymers as it was showed by the n F parameter of Freundlich isotherm.
Gibbs free energy value (∆G • ) indicates the spontaneity of the process being an essential tool to predict the development of chemical reactions. To calculate this value, the equation employed was: K p is the equilibrium constant (L/g), M adsorbate is the MW of DR and 55.5 is the constant related to the mole concentration of water (mol/L) [54,55]. The result obtained from Equation (14) was used in Equation (15) to elucidate the spontaneity of this adsorption process.
According to the results observed in Table 3, the standard free energy (∆G • ) was −25,175.97 for γ-CDs-EPI and −24,691.24 J/mol for HP-γ-CDs-EPI at 25 • C. The exergonic values obtained indicated that the adsorption process is spontaneous at the temperature tested.
The main advantages of the new synthesized polymers containing CDs are listed in Table 5. Table 5. Advantages of CDs polymers.

Advantages of CDs Polymers
Limited use of reagents employed in the synthesis Cyclodextrins are cheap compounds No toxic or harmful components remains in the polymeric adsorbents Cyclodextrin polymers can be stored for long periods These polymers can be reused many times effectively The desorption is easily achieved

Advanced Oxidation Process (AOP)
The AOP was able to eliminate 78% of the dye from the solution after application of a PL treatment regime of 193 J/cm 2 ( Figure 10). The treatment involved the application of 90 light pulses. Taking into account that some common PL systems available in the market work with a pulse repetition rate of three pulses per second [56], the treatment could achieve that level of dye degradation in just 30 s, which makes it very fast. The pseudo-first kinetic constant of this process was 0.0079 ± 0.0004 cm 2 /J. The CD adsorption process followed by the AOP was able to eliminate more than 90% of the dye from the solution.
Polymers 2020, 12, x FOR PEER REVIEW 17 of 21 According to the results observed in Table 3, the standard free energy (ΔG°) was −25,175.97 for γ-CDs-EPI and −24,691.24 J/mol for HP-γ-CDs-EPI at 25 °C. The exergonic values obtained indicated that the adsorption process is spontaneous at the temperature tested.
The main advantages of the new synthesized polymers containing CDs are listed in Table 5.

Advantages of CDs Polymers
Limited use of reagents employed in the synthesis Cyclodextrins are cheap compounds No toxic or harmful components remains in the polymeric adsorbents Cyclodextrin polymers can be stored for long periods These polymers can be reused many times effectively The desorption is easily achieved

Advanced Oxidation Process (AOP)
The AOP was able to eliminate 78% of the dye from the solution after application of a PL treatment regime of 193 J/cm 2 ( Figure 10). The treatment involved the application of 90 light pulses. Taking into account that some common PL systems available in the market work with a pulse repetition rate of three pulses per second [56], the treatment could achieve that level of dye degradation in just 30 s, which makes it very fast. The pseudo-first kinetic constant of this process was 0.0079 ± 0.0004 cm 2 /J. The CD adsorption process followed by the AOP was able to eliminate more than 90% of the dye from the solution.

Conclusions
The synthesis of two new cyclodextrin adsorbents was achieved successfully in order to remove DR 83:1 from wastewater. γ-and HP-γ-CDs-EPI showed similar behaviour in the adsorption process. The adjustment of the experimental results to PFOM, PSOM and IDM indicated that in both cases the results were well fitted by using the PSOM, indicating that the adsorption depends on chemical forces, apart from the fact that the IDM plays a key role in the process, showing that in the adsorption of DR both chemical and physical forces are involved.
According to the determination coefficient, γ-CDs-EPI followed the Temkin isotherm, whereas in the case of HP-γ-CDs-EPI, the Freundlich isotherm presented the best adjustment to the experimental data. The maximum adsorption ability of both adsorbents (qmax) was similar. In order

Conclusions
The synthesis of two new cyclodextrin adsorbents was achieved successfully in order to remove DR 83:1 from wastewater. γand HP-γ-CDs-EPI showed similar behaviour in the adsorption process. The adjustment of the experimental results to PFOM, PSOM and IDM indicated that in both cases the results were well fitted by using the PSOM, indicating that the adsorption depends on chemical forces, apart from the fact that the IDM plays a key role in the process, showing that in the adsorption of DR both chemical and physical forces are involved.
According to the determination coefficient, γ-CDs-EPI followed the Temkin isotherm, whereas in the case of HP-γ-CDs-EPI, the Freundlich isotherm presented the best adjustment to the experimental data. The maximum adsorption ability of both adsorbents (q max ) was similar. In order to remove the residual concentration of DR in water, an AOP was considered after using the CDs polymers. More than 95% of DR was removed from water by combining CDs polymers and the AOP at the highest concentrations of dye.